National Cancer Institute
                        PROCEEDINGS OF THE



            FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:



               PROGRESS ON 3OINT ENVIRONMENTAL AND



                   OCCUPATIONAL CANCER STUDIES
                            MAY 6-8, 1980
              SHERATON/POTOMAC, ROCKVILLE, MARYLAND

-------
The  papers  included in these Proceedings were printed as they
were submitted to this office.

Appropriate  portions  of the discussions,  working groups  and
plenary session were sent to the participants for editing.   The
style of editing varied, as could be expected.  To the extent
possible, we have attempted to arrive at a consistent format.

-------
                       PROCEEDINGS OF THE

         FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

             PROGRESS ON JOINT ENVIRONMENTAL AND

                 OCCUPATIONAL CANCER STUDIES
                           MAY 6-8, 1980
            SHERATON/POTOMAC, ROCKVILLE, MARYLAND
Proceedings  were developed  from  a  workshop on  the  National Cancer
Institute's, the Environmental Protection Agency's and the National Institute
for  Occupational Safety  and  Health's  Collaborative Programs on Environ-
mental and Occupational Carcinogenesis.

-------
              PROCEEDINGS OF THE

FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

    PROGRESS ON JOINT ENVIRONMENTAL AND

        OCCUPATIONAL CANCER STUDIES
                      Editors

               H. F. Kraybill, Ph. D.
               Ingeborg C. Black wood
                  Nancy B. Freas

              National Cancer Institute
                Editorial Committee

            Thomas P. Cameron, D.V.M.
               Morris I. Kelsey, Ph. D.
              National Cancer Institute

              Wayne Galbraith, Ph. D.
                 C. C. Lee, Ph. D.
          Environmental Protection Agency

              Kenneth Bridbord, M. D.
 National Institute for Occupational Safety and Health
                Technical Assistance

                    Sara DeLiso
                   Donna Young
              National Cancer Institute

-------
      FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:
          PROGRESS ON JOINT ENVIRONMENTAL AND
              OCCUPATIONAL CANCER STUDIES
         Program Organizing and Coordinating Committee
Kenneth Bridbord, M. D.
National Institute for Occupational
  Safety and Health

George Burton, Ph. D.
National Cancer Institute

Ms. Ingeborg Blackwood
National Cancer Institute

Thomas Cameron, D.V.M.
National Cancer Institute

Wayne Galbraith, Ph. D.
Environmental Protection Agency

Kay Kennedy
National Cancer Institute

H. F. Kraybill, Ph. D.
National Cancer Institute

C. C.  Lee, Ph. D.
Environmental Protection Agency

Nelson Leidel, Ph. D.
National Institute for Occupational
  Safety and Health

Carl Morris, Ph. D.
Environmental Protection Agency
Chairman
Coordinator
Coordinator and Logistics
Coordinator
Coordinator
Project Officer
Coordinator and Organizer
Coordinator
Program Organizer
Coordinator

-------
                        PROCEEDINGS OF THE
          FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:
              PROGRESS ON JOINT ENVIRONMENTAL AND
                  OCCUPATIONAL CANCER STUDIES

TABLE OF CONTENTS


                      TUESDAY MORNING. MAY 6

I.    INTRODUCTORY REMARKS AND OVERVIEW                           1-17

          Vincent DeVita                                                    3
          Director
          National Cancer Institute

          Vilma Hunt                                                        6
          Deputy Assistant Administrator
          Office of Health Research
          Environmental Protection Agency

          Anthony Robbins                                                   7
          Director
          National Institute for Occupational Safety and Health

          H. F. Kraybill                                                    16
          Scientific Coordinator for Environmental Cancer
          National Cancer Institute

II.    EPIDEMIOLOGICAL STATISTICAL SESSION                            18-97
                  Session Chairperson - Roger S. Cortesi

      Cancer Mortality in an Industrial Area of Baltimore - Genevieve M.          19
      Matanoski, Emanuel Landau, James Tonascia, Christina Lazar, Elizabeth
      A. Elliott, William McEnroe and Katherine King

      Epidemiologic Study of  a Population Previously  Exposed to Hexachloro-       46
      benzene - A. Gocmen, D. Cripps,  H. Peters, G.  T. Bryan and C. R. Morris

      Cancer in Southern Louisiana:  Progress Report of a Case-Control Study      53
      of Lung, Stomach and Pancreas Cancer - Linda  W. Pickle

      Support Services for Study of Bladder Cancer in New Hampshire,             55
      Vermont, and Maine (New England) - Robert Hoover

      A Case-Control Study of Lung Cancer Near A Zinc Smjjter - Linda M.       56
      Pottern, William J. Blot and Joseph F. Fraumeni,  Jr.

      General Discussion                                                     63

-------
     EPIDEMIOLOGICAL STATISTICAL SESSION (CONTINUED)
                   Session Chairperson - George Burton

     Industrial Emissions and Cancer Incidence in Contra Costa County:            68
     Progress  on the Epidemiological Study - Wilson B.  Riggan, Donald  E.
     Austin and William Mandel

     An Etiologic Study of Respiratory Cancer in Coastal Texas - Thomas 3.       88
     Mason and Linda W. Pickle

     Environmental Health Data Base for New Jersey - Thomas Mason             94
                     TUESDAY AFTERNOON, MAY 6


III.   EPIDEMIOLOGICAL STATISTICAL SESSION                            98-296
                   Session Chairman - Kenneth Bridbord

     Identification of High Risk Occupational Groups Using NOHS/RTECS -        99
     David H. Pedersen

     Industrial Hygiene Study of Workers Exposed to Nitrosamines - John M.      121
     Fa jen

     Mortality and Industrial Hygiene Study of Workers Exposed to Polychlori-     140
     nated Biphenyls - David P. Brown and Mark Jones

     Painting Trades Study - Dennis Zaebst, Marie Haring and Shiro Tanaka       173

     General Discussion                                                     205

     EPIDEMIOLOGICAL STATISTICAL SESSION (CONTINUED)
                  Session Chairperson - Joseph Fraumeni

     Mortality Study of Workers Employed at Organochlorine Pesticide Manu-     207
     facturing Plants  - David  P. Brown, David Ditraglia, Tsukasa Namekata
     and Norman Iverson

     Preliminary Findings of an Epidemiologic Study of Talc Workers (Indus-       229
     trial Hygiene Portion) - Alice Greif e

     Preliminary Findings of An Epidemiologic Study of Talc Workers - John       241
     Gamble, Alice Greif e and  John Hancock

     Lung Cancer in The National Coal Workers' Autopsy Study - V. Vallya-       276
     than, M. Attfield, R. Althouse, N. Rodman, C.  Boyd and F. H. Y. Green
                                   VI

-------
                    WEDNESDAY MORNING, MAY 7


IV.   METHODOLOGY/EXPERIMENTAL MODELS SESSION                 297-445
                   Session Chairperson - Carl R. Morris

     Evaluation of the Transformation Assay Using C3H 10T & Cells for Use      298
     in  Screening  Chemicals  for  Carcinogenic Potential  -  Thomas  P.
     Cameron

     Human Epithelial Cell Metabolic Activation Systems for Use with Human    306
     Cell  Mutagenesis  -  3.  Justin McCormick and Veronica  M.  Maher
     (presented by Michael Waters)

     Studies of Organ-Associated Antigens and Other Markers in Human         327
     Tumors  Which May be Useful for the Diagnosis of Malignant Diseases -
     Ronald Herberman

     Effects of Carcinogens, Mutagens and Teratogens on Non-Human Species    333
     (Aquatic Animals) - John A. Couch and Peter W. Schoor

     METHODOLOGY/EXPERIMENTAL MODELS SESSION (CONTINUED)
                    Session Chairperson - John Cooper

     Metabolism of Azo Dyes to Carcinogenic Amines - Larry K. Lowry,         378
     William  P. Tolos, Mark Boeniger, Charles Nony and Malcolm Bowman

     The Effect of Dietary Disulfiram Upon the Tissue Distribution and          399
     Excretion of   C-l,2-Dibromoethane in the Rat - Harry B. Plotnick,
     Walter W. Weigel, Donald E. Richards and Kenneth L. Cheever

     In Depth Biochemical, Pharmacological and Metabolic Studies on Tri-        412
     halomethanes  in Water -  R. J. Bull, M. Robinson,  T. J. Brown, F. L.
     Mink, R. D. Lingg and Carrie Whitmire


                   WEDNESDAY AFTERNOON,  MAY 7


V.   METHODOLOGY/EXPERIMENTAL MODELS SESSION                 446-526
                     Session Chairperson - C. C. Lee

     Chronic Animal Inhalation Study of Short (<5(jm) Asbestos Fibers -         447
     Stanley F. Platek and David H. Groth

     Mutagenicity Testing of Selected Industrial Chemicals - Tong-man Ong,     470
     Geoffrey Taylor, John Elliott, Carole A. Golden and Randy G.  Moon

     A Strategy to Validate Work Practices:  An Application to the Rein-         494
     forced Plastics Industry - R. J. Conard, B. L. Hopkins, H. Gordon Fitch,
     Michael J. Smith, W. Kent Anger and Richard R. Dangel
                                     vn

-------
                    WORKING GROUPS A, B AND C

VI.   SESSION A - WORKING GROUP ON PROBLEMS, NEEDS AND NEW         527
     DIRECTIONS FOR EPIDEMIOLOGY STUDIES

       Session Chairpersons - Kenneth Bridbord and Joseph Fraumeni


Vn.  SESSION B - WORKING GROUP ON TOXICOLOGY, METHODS             556
     DEVELOPMENT, PROBLEMS, AND NEW DIRECTIONS

         Session Chairpersons - Norbert Page and Gregory O'Conor


VHl. SESSION C - WORKING GROUP ON IMPORTANCE OF INTERAGENCY     579
     PROGRAMS:   DEVELOPMENT OF  FUTURE COLLABORATIVE PRO-
     GRAMS AND MEETING THE NEEDS OF REGULATORY AGENCIES

         Session Chairpersons - Richard Marland and Nelson Leidel


                     THURSDAY MORNING, MAY 8


IX.  CONCURRENT SESSION I:  RADIATION CARCINOGENESIS           603-684
                  Session Chairperson - Wayne Galbraith

     Laboratory and Field Trial Evaluations of the Cost/Effectiveness of Two     604
     Types  of Personal  Ultraviolet B. Dosimeters  - Arthur J. Sober and
     George Goldsmith (presented by Thomas W. Orme)

     Skin Cancer Epidemiological Studies - Joseph Scotto                      610

     CONCURRENT SESSION I  (CONTINUED)
                Session Chairperson - Thomas P. Cameron

     Hairless Mice for Carcinogenesis Studies - Thomas W. Orme                645

     Effects of Varying Doses of UV on Mammalian Skin:  Simulation of          664
     Decreasing  Stratospheric  Ozone  - I. Willis (presented by Thomas W.
     Orme)

     In Vitro Analysis of UV-B Induced Photooxidations - Norman I.  Krinsky      673
     (presented by Thomas W. Orme)

     Support Services for  Radiation Risk Estimation Following Radiotherapy      677
     for Cervical Cancer  - John D. Boice
                                    Vlll

-------
X.   CONCURRENT SESSION II: DATA BASES/MONITORING              685-828
                   Session Chairperson - George Simon

     Chemicals Found in Human Biological Media, A Data Base - Cindy          686
     Stroup

     Completion of 1972 Mortality Data, NCI/NCHS Cooperative Study -         707
     Thomas Mason

     National Mortality Data Bases for Environmental Epidemiology - Carol G.     709
     Graves, Jacob Thomas and Charles Poole

     Mapping Chemical Exposures - Kenneth D. Kreitel                        724

     CONCURRENT SESSION II (CONTINUED)
               Session Chairperson - Elizabeth K. Weisburger

     The National Occupational Hazard Survey (NOHS II) - David S. Sundin       740

     Extent of Industrial Exposure to Epichlorohydrin, Vinyl Fluoride, Vinyl       750
     Bromide and Ethylene Dibromide - 3ames L. Oser

     Environmental Levels and Urine Content of Workers Exposed to Azo         775
     Dyes  -  Mark F. Boeniger, Larry K. Lowry and William P.  Tolos,  Charles
     Nony  and Malcolm Bowman

     IH Characterization of Coal Liquefaction Facilities - Stephen               798
     Berardinelli

     Trade Name Ingredient Data Base - Herbert L. Venable                    823


                    THURSDAY AFTERNOON, MAY 8


XI.  PLENARY SESSION                                                   829
                 Session Chairperson - Gregory T. O'Conor


XII. LIST  OF PARTICIPANTS                                               855
                                      IX

-------

-------
FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

    PROGRESS ON 3OINT ENVIRONMENTAL AND

        OCCUPATIONAL CANCER STUDIES
              Tuesday Morning, May 6
            INTRODUCTORY REMARKS


              Vincent T. DeVita, M. D.
                     Director
              National Cancer Institute
                Vilma, Hunt, B. D. S.
           Deputy Assistant Administrator
             Office of Health Research
          Environmental Protection Agency
               Anthony Robbins, M. D.
                      Director
  National Institute for Occupational Safety and Health
                    OVERVIEW

                H. F. Kraybill, Ph. D.
    Scientific Coordinator for Environmental Cancer
              National Cancer Institute

-------
-2-

-------
                            INTRODUCTORY REMARKS

                              Vincent T. DeVita, M. D.

                                     Director
                              National Cancer Institute
                                Bethesda, Maryland


At the National Cancer Institute we are interested in knowing which chemicals exist in
the environment, how they interact to cause cancer, and how we can take advantage of
that knowledge to reduce the incidence of cancer.  I am an optimist.  I think that in our
lifetime we are going to be able to make a significant impact on the reduction  in the
incidence  and  mortality  from  cancer.   That is not  because  we expect to live  an
inordinate  time but  rather  because we have reached a critical mass,  in  terms  of
opportunities for prevention.   This accounts for the ferment, the controversy and the
excitement in the field.

From the  point of view of a person like myself, who grew up in the  treatment area,  I
recall the  same kinds of  sensations and ferment going on as  new advances, like the
development of anticancer drugs for treatment of cancer, proved to be successful.

There are opportunities in at least three  major areas for the prevention of cancer. One
is the  prevention of cancer causing chemicals from  entering the environment. Our
testing program to identify potential carcinogens is one of  those in a great state of
ferment.  Investigators are now discovering the variables in the in vitro tests, like the
Ames assay, which we thought were simple tests when they were first reported.  For
example, everyone's S-9 preparation is different from  everyone else's. Still there are
many opportunities in this area to exploit.

The second area is to reduce or eliminate chemicals in the environment that  might
cause cancer.  This meeting will devote much of its attention to the epidemiology of
identifying cancer  causing agents.  Our contribution  in this area, represented by Dr.
Fraumeni  and his staff, is very great. Considerable opportunity exists for us to interact
with NIOSH  and EPA, because of EPA's capabilities to analyze the environment and
NIOSH's authority in industrial settings, particularly in accessing records not normally
available to us.  These are examples of perfect collaborative opportunities.

The third area for the prevention of cancer involves lifestyle changes.  This is the most
controversial area.  I could do very well at this meeting if I said that most cancers are
caused by  exposure to chemicals  and  then, before  another group  with  a  different
orientation,  to  say most  are  caused by lifestyle.   Without  trying  to  be a  great
compromizer, I think the answer is obviously both chemicals and lifestyle.  It is best not
to think that a  certain fraction is caused by lifestyle and a certain fraction is caused by
chemicals, but  rather  to think that a certain  fraction is caused by both.  We are just
beginning  to  learn about the  multiplicity of cancer causation. Although the complexity
might be discouraging, it has an encouraging side, because we may not need to get  rid
of all influences to prevent cancer.  It may be quite possible to remove one influence,
an initiator, for example, and then  a lifestyle promoter may become unimportant, or
vice versa.  Although arguments about lifestyle versus chemicals have created a lot of
tension, this is a healthy phenomenon.

-------
As you may know, the National Cancer Institute is awaiting approval of a new division.
We  hope it will devote much of its activities to what we call applied prevention, using
knowledge  that may be generated from the kind of interactions that will be described
here in the next two or three days, to approach the problem of application of knowledge
to reduce the incidence of cancer. I am sure you will hear more about this activity.

I will conclude by showing a chart illustrating the area of carcinogen identification  and
regulation.  I was reading a memo describing the process and the only thing I could do to
save myself was to put it in the  form of a chart. It illustrates the complexity of the
problem.  All of us are represented on this chart.  Across the top is the process gone
through in terms of identifying and regulating chemical carcinogens.  Along the side are
the committees and organizations that have an  input into the process.  It is a  very
complex process.  We have a lot to do  to make the complexity work a little bit better.

I am delighted  to have the opportunity to meet those of you who work in this area and
whom I have not had the chance to meet before. I hope we have a chance to do this on
numberous occasions.

-------
                                                                             SCHEMATIC:  BASED ON  "REGULATION  OF  CHEMICAL CARCINOGENS"
 REGULATORY ACTIVITY SEQUENCE
OSHA
USDA ^^^
FDA • 	 '




OCE






\f;~ 	 	 	 . 	 _ 	 . 	 	 1
a? 	 	

AGENCY TEST/STUDY EPIDEM. AGENCY
CRITERIA POLICY I STUDIES 1 CRITERIA j
III 1 1
11 I
SET 1 DEFINE 	 ' ANIMAL 	 EVAL- STUDY OF RISK SET DEFINE ASSESS IMPLEM.
PRIORITIES) PROCED. — j TESTING ^_ ^[^ EXPOSURE "MEN? PRIORITIES ACTION [ OF REG. | ACTION \
	 ' "
1 OTHER _| 1 OUTSIDE 1
TESTS REVIEW
-r 1 (-1
n -CAncii | J-1
  1  OFFICE OF MANAGEMENT AND BUDGET (OMB)
  3  REGULATORY COUNCIL
                                                   ---  —  — — — — -- — — 1980 ZERO-BASED BUDGET EXERCISE FOR TOXICOLOGY RESEARCH -- -  — —  — — — —
                                                                                            - COORDINATES ACTIVITIES OF REGULATORY AGENCIES-
                                                                                                                                                                                                   ANALYSISOF
                                                                                                                                                                                                  REGUL. IMPACT
                                                                                                                                                         1  COORDINATING FEDERAL RESEARCH AND REGULATORY PROGRAMS ON TOXIC CHEMICALS

                                                                                                                                                         2  ELIMINATING OVER LAPS AND FILLING GAPS IN DATA ON TOXIC CHEMICALS
                                                                                                                                                                                     [	COORDINATES GOVERNMENT INTERVENTION	
         REGULATORY ANALYSIS REVIEW GROUP
                                                                                                                                                                                    r —  — — — —	ANAL'
                                                                                                                                                                                                                                   V ACTION	—
  4  CMTE TO COORD. TOXICOLOGY ETC. (CCTRP HEW)
  5  NATIONAL TOXICOLOGY PROGRAM (NTP)
  6   NATIONAL CANCER ADVISORY BOARD INCAB)
                                                                     ADVISES HEW ON TOXICOLOGICAL ACTIVITIES AND COORDINATES DATA AND RESOURCES IN HEW.


                                                          TEST CHEMICALS, DEVELOPS TEST SYSTEMS. SUPPORTS RESEARCH AND NEEDS OF REGULATORY AGENCIES
                                                                                                                                                         MAYBE
                                                                                                                                                       INVOLVED IN
                                                                                                                                                       RISK ASSESS.
                                                 f —	—	COORDINATION OF CANCER RESEARCH INSIDE AND OUTSIDE GOVERNMENT —  —	—  —	
  7  INTEHAGENCY REGULATORY LIAISON GRP [IRLG]



     a   RESEARCH PLANNING WORK GROUP



     b   TESTING STDS& GUIDELINES WK GHP.



     c   EPIDEMIOLOGY WORK GROUP



     d   RISK ASSESSMENT WORK GROUP



     e   REGULATORY DEVELOPMENT WK GRP.



     f    INFORMATION EXCHANGE WK GRP.



     9    COMPLIANCE & ENFORCEMENT WK. GRP



  B   NCI CLEARINGHOUSE FOR ENV CARCIN



  9   INTERAG. COLLAB GRP ON ENV. CARCIN.



 10.   INTERAGENCY TESTING CMTE



 11.   NCI CHEMICALSELECTION WKG GRP.



 12.   OFFICE OF SCIENCE & TECHNOLOGY POLICY
                                                  	ANALYSISOF GAPS AND OVERLAPS IN TOXICOLOGICAL RESEARCH	—	

                                                              DEV. OF UNIFORM
                                                              IRLG TEST STDS.&
                                                              GUIDELINES	

                                                                              DEFN. OFMIN.
                                                                              CHARACT. OF
                                                                             EPIDEM. STUDIES
                                                                                                                                                                                   IPRIMECOORD^I"
                                                                                                                                                                                   OF REG. ACTION I
                                                                                                                                                                                   [IRLG AGENCIES) |
PRIME COORD, I PRIME COORD.   I IRLG IMPACT
OF REGUL.    I OF REG. ACTION I COORD-ESP
PRIORITIES    I (IRLG AGENCIES) I  ECONOMIC
                                                      PLANNING a ADVICE
                                                      ON TEST METHODS
                                                       AND PRIORITIES	


                                                   —  — —  ~	—  — — 	  	 —  INFORMATION EXCHANGE WITH FEDERAL REGULATORY AND RESEARCH AGENCY STAFFS				

                                                RECOM.CHEM.
                                               TEST PRIORITIES
                                                   TO EPA

                                                 NOMINATES
                                                 CHEMICALS
                                                   TO N T P.


                                                t— 	 	  _ 	 	  	  	 	 	  	 	 	 	 	  — —  — 	  — — —  — ADVISORY TO THE PRESIDENT  	  — —		 _ 	 	 	  	 —. —  _
                                                          COORD. OF TOX
                                                            DATA FOR
                                                          IRLG AGENCIES

                                                          INTERAGENCY
                                                          COORD. OF
                                                          VIOLATIONS
                                                                                                                                                                                                                                              PR|ME COORD
                                                                                                                                                                                                                                              ENFORCEMENT
                                                                                                                                                                                                                                              POLICIES
13  CENTER FOR HEALTH STATISTICS


14  NAT. TOXIC SUBSTANCE DATA CMTE.


15.  COUNCIL ON ENV QUAL I/A TASK FORCE
                                                                                                                         - COORDINATION THROUGH COMMON DATA UTILIZATION .
                                                                                                                           - COORDINATES INFORMATION AND DATA SYSTEMS
                                                                                                                                                                                                                                 REVIEW OF
                                                                                                                                                                                                                                 POLLUTANT
                                                                                                                                                                                                                                MONITORING
16.  TASK FORCE ON ENVIRONMENTAL CANCER.
    HEART AND LUNG DISEASE

-------
                           INTRODUCTORY REMARKS

                                Vilma Hunt, B.D.S.
                          Deputy Assistant Administrator
                             Office of Health Research
                         Environmental Protection Agency


Since 1978 the relationship between the National Cancer Institute and the Environ-
mental Protection Agency on  this particular program of environmental carcinogenesis
has included the Office of  Pesticides and Toxic Substances and the Office of Research
and Development.  The interactions have been quite close during that period.  There has
been a gradual development of the relationship to the extent  that we now have ongoing
the projects that you will be hearing about over the next several days.

One thought I had  in terms of  Dr. DeVita's comments  is almost light years away from a
meeting I was at yesterday in Buffalo.  I thought you would be interested in hearing the
kind of presentation that was given there. It was an international conference on cancer
in blacks.  The first day -  today is the second day and deals  more with the American
scene - was virtually restricted to cancer in blacks in Africa. There was a  very heavy
representation of speakers  from medical schools or former medical schools in Africa.
The intent for  most of them  was to  present quite descriptive comments on the main
clinical experiences that they had had in Africa over the years. It became quite evident
that if there is ever a meeting of environment and lifestyle, that  is where we can see it.
The development of that meeting today, I am hoping,  will move  more to the American
reality in  which certainly  the  black population  living in this  country is somewhat
different to what it is in Africa.

To get back also to Dr. DeVita's chart, we are going to offer to him some additional
advice. I think the Task Force on Environmental Cancer and Heart and Lung Disease is
missing.  I  do not see it in casting my eye rapidly over it. That particular  report, which
I just signed off on this morning in fact, is now going  to the heads of agencies and will
reach Dr.  DeVita  before very long.  I am  hoping  that it will go through your office
expeditiously.

It is quite  important  that that Task  Force  Annual Report, which is the  third one, go
through its appointed rounds to  the agency  heads  and to the Secretary of Health  and
Human Services.  The Task Force was mandated by Congress and the annual reports are
directed  to them.  Again, it shows the  very specific concern  that  Congress  has
developed that we do just the  kind of things that Dr. DeVita mentioned earlier; that is,
there are very marked concerns that  the capabilities  of all the agencies be  brought to
bear  and focused on some of these areas of mutual  concern  rather  than disparate
efforts which  go in their  own directions.  Certainly  the program  that we  have been
having on environmental carcinogenesis between EPA and NCI fulfills the expectations
of that Task Force.

-------
                         INTRODUCTORY REMARKS

                            Anthony Robbins, VI. D.
                                   Director
              National Institute for Occupational Safety and Health

        It is a pleasure for me to be here today at the onset of this historic
collaborative  workshop.  The research conducted by our respective agencies
has contributed much to our understanding of  how factors in the working and
the general environment contribute to the etiology of cancer, this knowledge
ultimately contributing to the prevention of cancer.  MIOSH is very pleased
to be part of this cooperative effort and wishes to acknowledge the support
of the National Cancer Institute, without which much of this important work
would not have been possible.

        The decade of the 1970's was characterized by a major concern for the
problem of occupational and environmental cancer, which will surely continue
in future years. This decade was also characterized in large part by the recogni-
tion that chronic exposure to chemicals posed a greater risk to the health of
workers and the general  population than previously realized.  The decade of
the 1980's has begun with the promulgation of an historic general regulation
to deal with the problem of carcinogens  in  the workplace by OSHA, a regulation
which many of the scientists in this room contributed to, in terms of research,
testimony, and special analyses.  It is envisioned that future collaborative efforts
will continue to provide the often embattled regulatory agencies with the compo-
nent scientific input necessary to formulate defensive public health policies.

        If the Federal Government is to  adequately deal  with the problem of
occupational/environmental cancer, it will  only be through collaborative efforts
such as our  program in cancer research.  This  program cannot succeed  without
the collective strengths of our respective organizations working together,  ^y
working together, we build upon our strengths and complement our weaknesses.
It is our hope that this program will also create an atmosphere of cooperation
and friendship among individual scientists in our respective organizations.  One
of the best ways to foster such collaboration is to encourage  joint  program planning
efforts as well as an exchange of scientists from our respective organizations;
a proposal which we should consider  to further strengthen our program.

        As a result of this interagency program, NIOSH has initiated and/or
completed over 60 research projects dealing with a broad spectrum of topics
related to cancer in the workplace.  Tables 1-4 illustrate the magnitude of our
research program encompassing epidemiology, toxicology, and industrial hygiene
studies.  The results of these studies have potential applicability for protecting
not only workers but members of the general  population as well. This is because
risks defined in working populations may also apply to the general  population.
At this conference, the results of nearly one-third of these projects will be
summarized.

        Among the highlights of our  program are several rejects to be presented
at this workshop including:

Ethylene Dibromide/Disulfiram Interaction

        A carcinogenic interaction resulting from inhalation of ethylene
dibromide (BOB), an industrial solvent, and  oral doses of disulfiram, a
drug used for treatment of alcoholism, was identified. Rats experienced
inhalation exposures of 20 ppm  EDB, which is the U. S. Occupational Standard,


                                         1

-------
for 5 days per week, 7 hours per day.  The disulfiram dose was 0.05
percent by body weight.  Significant histopathologic findings in animals
receiving both treatments concurrently included:  hemangiosarcomas of
the liver, kidney and spleen; adenocarcinomas of the mammary gland in
females, and atrophy of the entire genital tract in males.  The incidence of
tumors was substantially increased in animals receiving both EDB and
disulfiram compared to animals receiving EDB alone or disulfiram alone.
Results from this study led to the issuance by NIOSH of a Current Intelligence
Bulletin (CIB) and notices in the scientific literature.  This study clearly
suggests the potential deliterious combination of workplace chemicals and
certain drugs.

     Benzidine-Based Dyes

     Work performed jointly (through Interagency Agreement) with the National
Center for Toxicologic Research showed that the benzidine-based azo dye,
Direct Black 38, is extensively metabolized in the hamster to yield metabolites
known to be carcinogenic.  Hamsters given purified Direct Black 38 were shown
to excrete benzidine, N-acetyl benzidine, and 4-aminobiphenyl - all  known
carcinogens in the urine.  Another azo dye, Direct Yellow 12, was not
metabolized to carcinogenic metabolites in the hamster.  As part of the same
study, urine from workers occupationally exposed to Direct Black 38 was also
found to contain benzidine.  NIOSH recently published a Special Hazard Alert
(SHA) on benzidine-based dyes which included r.ome of the results of this
study.  Industrial hygiene studies to determine extent-of-exposure among
workers to azo dyes were also used as input to the SHA.

     1'orkers Exposed to Polychlorinated Biphenyls

     Results of a retrospective cohort mortality study showed an. observed
excess mortality risk of cancer of the liver and cirrhosis of the liver
among exposed workers, although these observations were not confirmed by
latency of exposure analyses.  These data are among the first to investigate
a possible increased risk of death from certain causes among individuals
exposed to PCBs.

     Retrospective Cohort Mortality of Workers Exposed to Talc

     Two studies were published (Vermont and New York talc workers)  under
this project.  These reports will be used as input to the Talc Criteria
Document v.'hich is under preparation by NIOSH.  The Vermont Study is  unique
in that it showed an excess nonmalignant respiratory disease mortality
risk in a population exposed to nonasbestiform talc; while the New York
study shoved an excess mortality risk due to nonmalignant respiratory
disease and bronchogenic cancer.  NIOSH has also published an industrial
hygiene study documenting extent-of-exposure among workers to talc.

       Other highlights of our collaborative efforts which will  not  be
presented at this specific workshop include:

       Radon Daughters from Foundry Sands

       As  part of a larger project investigating the co-carcinogenetic  effects
of foundry particulates,  zirconium silicates are being  evaluated.  We  have


                                           8

-------
recently discovered that commercially available zirconium silicates, which
are used extensively in foundries as a replacement for silica sand, emit
large quantities of radiation through release of radon daughters.  This
has led NIOSH to initiate follow-up field investigations of foundry workers.
The U.S. EPA has been informed of our finding and is also investigating
the situation^

     Proportionate Mortality Study of Foundry Workers

     Results of these studies already published showed a significant
increase in mortality due to lung cancer and nonmalignant respiratory
disease.  These results are being evaluated in more detail through a case
control study.

     Carcinogenicity of 2-Nitropropane

     Inhalation exposure (200 ppm) of rats and rabbits to 2-nitropropane
(2-NP), a nitroparaffin once used widely in a variety of industrial
applications, induced liver neoplasms, indicating that 2-NP is a potent
carcinogen.  The exposure time was 6 months, an unusually short time to
demonstrate carcinogenicity in the rat.  A verification study performed
by the major producer of 2-NP yielded similar results as obtained by
NIOSH.  A Current Intelligence Bulletin (CIB) alerting the occupational
health community to this finding was issued by NIOSH, and a Health Hazard
Alert was also prepared by OSHA.

     This study contributed to the control of a commonly used substance
involving exposure to more than 100,000 workers.

     Retrospective Cohort Mortality of Workers Exposed to Beryllium

     Results of this study were published in Environmental Research,
suggesting an excess mortality due to lung cancer among workers exposed to
beryllium.  This study also contributed to the overall OSHA assessment of
beryllium during a public hearing on a proposed Beryllium Standard.

     Retrospective Cohort Mortality of U'orkers Exposed to Benzene

     Results of this study were used to support the OSHA promulgated
standard and were published in The Lancet.  This work, which represents
the pivotal epidemiology study of workers occupationally exposed to benzene,
shov/ed a five-fold excessive risk of all leukemias and a ten-fold excess
of deaths from myeloid and monocytic leukemias combined in the study
population compared with controls.

     Retrospective Cohort Mortality of Workers Exposed to Styrene-Butadiene
     Rubber

     The results of this study suggested an excess mortality (although not
statistically significant) from neoplasms of the lymphatic and,hematopoietic
tissues.

-------
     Our program this week includes not only an opportunity to share the
results of ongoing studies, but also an opportunity to discuss new
methodologies, future cooperative ventL.'es, and how our programs can be
more responsive to the needs of the regulatory agencies.  To have the
greatest impact, research conducted must meet certain criteria:  (1) it
must be of sound quality and able to meet the test of critical peer review;
(2) the results must be published in refereed journals; and (3) the results
must be summarized in language that is understandable to the workers and to
the general population who are to be protected.  In this regard, it is
important that the results from this workshop be published and widely
disseminated.  We may also desire to develop a summary of the proceedings
which would be useful to non-scientists who have an interest and concern
about the problem of occupational and environmental cancer.  We may also
desire to expand the audience at future workshops to include not only
government scientists but representatives from academia, labor, industry,
and public interest groups.

     In reviewing the overall accomplishments and future directions of our
program, it is clear that a reasonable balance needs to be established
between the amount of research characterized as problem identification (such
as toxicological and epidemiologic studies) in contrast with research devoted
to problem solution  (such as development of improved control technology or a
safer substitute material).  NIOSH is committed to expanding its own base
program efforts in the area of control technology to protect workers.
Hopefully in the future, we can consider control technology to be a candidate
for research under this collaborative program as well.

     As we look forward to the 1980's, it is important that our agencies
continue our vigorous efforts to identify cancer hazards in the general
and in the working environment.  It is also extremely important that
Federal agencies expend a greater effort finding solutions to these problems
as well.  I am confident that this workshop will mark an important step
in our overall efforts to effectively deal with the problem of environmentally-
related cancer.  Hopefully this will be the first of many such workshops.

     Thank you.
                                           10

-------
Table 1. - Cancer Projects Related to Epidemiology/Industrial Hygiene
                    Project (Performing NIQSH Division)
o Mortality and IH Study of Gold Miners (DSHEFS)
o Mortality Study of Pesticide Formulators (DSHEFS)
o Kepone Registry (DSHEFS)
o Mortality, Medical and IH Study of Talc Workers (DSHEFS)
o Mortality and IH Study of Styrene-Butadiene Rubber Workers (DSHEFS)
o Mortality, Medical and IH Study of PCBs (DSHEFS)
o Mortality and IH Study of Perchloroethylene Workers (DSHEFS)
o Mortality, Medical and IH Study of Chlorinated Hydrocarbons (DSHEFS)
o Mortality Study of Beryllium Workers (DSHEFS)
o Mortality and IH Study of Nitrosamines/Cutting Oils  (DSHEFS)
o Mortality, Medical and IH Study of Painting Trades (DSHEFS)
o Mortality and IH Study of Printing Trades (DSHEFS)
o Mortality and IH Study of Phosphate Workers (DSHEFS)
                                          11

-------
Table 1.- Cancer Projects Related to Epidemiologic/IH (Continued)
            Project (Performing NIOSH Division)
o Morbidity and IH Study of Fibrous Mineral Workers (DRDS)
o Mortality and IH Study of New Agents (DSHEFS)
o Mortality and IH Study of Workers Exposed to Styrene (DSHEFS)
o Mortality, Morbidity and IH Study of Brakeline Workers  (DSHEFS)
o Mortality and IH Study of Benzene Workers (DSHEFS)
o Mortality and IH Study of Foundry Industries (DSHEFS)
o Mortality Study of Workers in Plywood, Paper, and Pulp  Industries (DSHEFS)
o Mortality, Medical and IH Study of Workers Exposed to Ethylene Oxide (DSHEFS)
o Mortality Study of Crushed Stone Exposures (DRDS)
o Mortality and IH Study of Workers Exposed to Toluene (DSHEFS)
o Mortality and IH Study of Leather Industry Workers (DSHEFS)
o Mortality and IH Study of Workers Exposed to Dyes (DSHEFS)
o Mortality Survey of a Chemical Plant in  Kanawa Valley (DSHEFS)
o Cancer  Surveillance of Occupational Cohorts in a Western SMSA (DSHEFS)

-------
     Table 2. - Cancer Projects Related to Industrial Hyqiene
               Project  (Performing NIOSH Division)
o Occupational Cancer Surveillance (DSHEFS)
o IH and Medical Study of Workers Exposed to Azo Dyes (DSHEFS)
o IH Study of New Agents (DSHEFS)
o Utilization of Full File For Survey/Sampling Decisions (DSHEFS)
o Mapping Chemical Exposures (DSHEFS)

-------
Table 3. - Cancer Projects Related to Toxicology
           Project   (Performing NIOSH Division)
o Carcinogenicity of Inhaled 1,2-Dibromomethane (DBBS)
o Chronic Inhalation of Short Asbestos Fibers (DBBS)
o Perform Intratrachael Testing of Copper and Lead Smelter Dusts + Fractional
  Smelter Dust to Determine Their Carcinogenicity (DBBS)
o S02 Effect on Smelter Dust Carcinogenesis (DBBS)
o Metabolism of Azo Dyes to Carcinogenic Amines (DBBS)
o Foundry Mold and Sand Pryolysis Effluent-Carcinogenesis (DBBS)
o Roofing Asphalts and UVL Carcinogenesis (DBBS)
o Determine Whether Thallium Trioxide and Antimony Trioxide Are Carcinogenic
  in Animals (DBBS)
o Machine Oils and Nitrosamines (DBBS)
o Tumorigenicity of Nitroaliphatics (DBBS)
o Fibrogenic Responses and Pulmonary Carcinogenesis (DBBS)
o Workshop on the Role of Metals in Carcinogenesis (DBBS)
o Co-Carcinogenicity of Foundry Participates (DBBS)
o GI Absorption of Be & Pt Complexes (DBBS)
o Interaction Between Drugs and Industrial Chemicals (DBBS)
o Carcinogenicity of Dimethylformamide (DBBS)
                                         14

-------
Table 4. - Other Cancer Related Projects
                     Project (Performing NIOSH Division)
o Sampling and Analytical Methods for Four (4) Carcinogens (DPSE)
o Develop Behavioral Approaches For Improving Work Practices Directed Toward
  Reducing Occupational Exposure to Carcinogenicity (DBBS)
o Develop Protective Equipment and Determine Protective Clothing Permeability
  For Carcinogens  (DPSE)
o Development of Trade Name Ingredient Data Base (DSHEFS)
o Validation of NOHS and RTECS Risk Ranking Algorithm (DSHEFS)
o Physical and Chemical Properties of Foundry Particulates (DBBS)
                                         15

-------
                                   OVERVIEW

                              H. F. Kraybill, Ph. D.
                            National Cancer Institute

The  collaborative  program  on occupational cancer  between the  National  Cancer
Institute and the National Institute for Occupational Safety and Health for the conduct
of cooperative research  in this  area had  its inception  in  late  1976.   In  the  U. S.
Congress House Record of 1976, it was noted that "considerable wide public attention
has been  drawn to  the  number  of  known  and  potential environmental  carcinogenic
hazards, such as vinyl chloride, pesticides, water  pollutants and certain gene combina-
tions  that could result from certain  types of research."  This interest on the part of
the U. S. Congress gave  rise to  a Congressional mandate to set up a collaborative
program  between  the National  Cancer  Institute and  the  National Institute  for
Occupational Safety and Health.  This program had as its main thrust the identification
of cancer hazards in the general work environment.  The objective of this congres-
sional directive was to  encourage  interagency  collaboration with  a view towards
increased efficiency and  the  widest  possible  utilization  of staff  capabilities and
expertise.  During the earliest phase of this program  there  was simply a transfer  of
funds to NIOSH from the NCI annual budget allocation; the funds to be used essentially
at their discretion.   Since  then, however,  the program has evolved into a  truly
excellent collaboration, with NCI becoming a partner in  the decision-making process
on the type and scope of projects which would be of mutual interest to both agencies.
More recently, the  NCI  staff has begun to initiate  proposals in this general  area.
Clearly these  types  of  interfacing  reflect more  closely the initial intent  of the
Congress.

The  present agreement  allows that  proposed projects receive technical,  relevance,
need  and priority  reviews by senior  scientists in the Division of Cancer Cause and
Prevention  of  NCI  and  representative senior staff  from  NIOSH.   It is  ultimately
planned to have some of the contract project supervision by an equal  representation in
project officers from NCI and  NIOSH.  Of the total of 71  projects initiated to date, as
of March 1980, 19  have been either completed or satisfactorily terminated. For any
one year the dollar ceiling has been set at four million. This level of funding seems to
provide for the mutual needs for achievement of  goals and the  attainment of a quality
program which achieves  excellent collaboration.  The  first interagency agreement for
this  collaborative program  was signed by the sponsoring and accepting agencies  on
September 23,  1976.

The   National  Cancer  Institute/Environmental  Protection  Agency  Collaborative
Program on Environmental  Cancer had similar  beginnings to  the NCI/NIOSH agree-
ment previously described.   This program was implemented in early  1977 based on a
mandate from the Office of Management and Budget as a result of interest shown  by
the U. S. Congress.  The stipulation  was  that $*J million included in the NCI  budget  be
set aside for projects of  mutual interest to  both  EPA and NCI. The early planning of
this  program took place between the top staff of the Division of Cancer Cause and
Prevention of NCI and the Office of  Research and Development of the Environmental
Protection Agency.  Later the  Office of  Toxic Substances of  EPA joined in  these
planning meetings.
                                         16

-------
In November,  1977 a list  of  16 projects was approved conjointly by an NCI/EPA
coordinating and advisory group.  This initial list of projects soon expanded  to  20
projects.  Final approval of these  projects and plans  for a continuing program were
secured by  signatures of the Director of NCI and Assistant Administrators  of  the
Office of Toxic Substnces and the Office of Research and Development of EPA.  These
plans,  finalized in  late  1977, soon developed  into a Memorandum  of  Understanding
signed by Drs. Upton (NCI), Dr. Gage (EPA) and Mr. 3ellinek (EPA) on 3anuary 20,
1978.  In essence, the Environmental Protection Agency looked upon this collaborative
effort as supportive to their research in environmental programs which is responsive to
the requirements set forth in the Toxic Substances Control Act.

An interagency agrement for commitment to longterm continuance of such a program
was signed by representatives of NCI and EPA on 3une 22,  1978 for funding for  the
period 3une 1978 to 3une 21, 1984. For any one year the dollar ceiling has been set at
four million. This level of funding seems to provide the  mutual needs of each agency
for achievement of  goals  in effective  collaboration and the attainment  of  a high
quality program. The NCI/EPA collaborative program now has,  for fiscal year 1980,
thirty projects. Projects scheduled for completion with FY80 funding are  seventeen
unless a few are extended.

To monitor  the progress  in this program, biannual and annual reports are required from
each project officer.  The first edition of the annual report was recently mailed out to
all participating project  officers and program  staff of each agency.  If there are any
omissions on our distribution list then the respective NCI and EPA coordinators for the
program should be notified.

The program coordinators for  the NCI/NIOSH and NCI/EPA collaborative programs
considered it timely to sponsor a workshop which, of course, is why we are here today.

In  our view,  we considered that  the interfacing of all program  staff and project
officers, limited to federal representation, would serve the purpose of orienting  all
representatives in both programs of the activities and accomplishments in all technical
areas.  With reference to areas of emphasis in the NCI/EPA collaborative program,
and perhaps the same holds for the NCI/NIOSH program,  one is essentially dealing with
the following areas:  a)  Information/Monitoring and Data Resources, b)  Experimental
Studies/Mechanisms,  c)  Methodological Approaches and Developments, and d) Epi-
demiological Studies.

In order to provide adequate coverage within the timeframe of three days, we have
scheduled concurrent working sessions A, B and C. It is  hoped that all participants at
the workshop will identify the working session  they will attend so that we can achieve
active participation for each working session.

We look forward to a successful workshop and through your  participation and support
we believe that this objective will  be achieved. The Organizing Committee wishes to
express their thanks and appreciation for  your  efforts in this first workshop. We have
a compendium  of abstracts and,  in due course of  time, Proceedings for  the Workshop
will be made available.

-------
      FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:
          PROGRESS ON 3OINT ENVIRONMENTAL AND
              OCCUPATIONAL CANCER STUDIES
                    Tuesday Morning, May 6
           EPIDEMIOLOGICAL/STATISTICAL SESSION
                   SESSION CHAIRPERSON
                      Dr. Roger Cortesi,*
                Environmental Protection Agency
(Note: Dr. Vilma Hunt, EPA, substituted for Dr. Cortesi until his arrival.)
                                 18

-------
 CANCER MORTALITY IN AN INDUSTRIAL AREA OF BALTIMORE
              'Genevieve M.  Matanoski*
                   Emanuel  Landau0
                   James Tonascia*
                  Christina Lazar*
                Elizabeth A.  Elliott*
                  William McEnroe*
                   Katherine King*
              Johns Hopkins University*
         School of Hygiene and Public Health
               615 North Wolfe Street
              Baltimore, Maryland 21205
         American Public Health Association0
                  Washington, D.C.
         Supported by Contract # 68-01-3859
awarded to the American Public Health Association by
             Office of Toxic Substances
        U.S. Environmental Protection Agency
                  401 M Street, SW
               Washington, D.C. 20460
                                   10

-------
        Cancer Mortality in an Industrial Area of Baltimore
     Arsenic has long been known to be a poison when ingested in large
quantities by man, animals or plants.  It is known that continued ingestion
of high natural levels of arsenic in water or food will produce skin lesions
                 (Tseng et al_.,  1968;  Braun,  1958)
including cancer A    . "Consumption of arsenic as a therapeutic agent is also
known to cause skin lesions (Neubauer,  1947).
     The risk  of inhalation of arsenic has not been as extensively investigated.
Workers exposed to arsenic in the manufacture of pesticides have an  increased
                                   (Ott ejt a]_.,  1974;  Baetjer et al^,  1975)
risk of lung cancer and lymphomas  A   .  Less is known about the chronic health
effects in the general population exposed to arsenic in the air.  It is known
that children around smelters may have high arsenic levels in nails and hair
but it is not clear whether these observed indications of absorption of the
                                                          (1975)
agent also indicate long-term toxicity.  Blot and Fraumeni A  have suggested
that there may be an association between excessive lung cancer mortality and
the existence of non-ferrous smelting industries in several counties in the
U.S.  It  is  not known  whether some by-product of this industry such as arsenic
is associated with these carcinogenic effects.
     The  purpose of the current study is to determine whether there is an
excess mortality from  cancer in the population which resides near a chemical
plant in  the inner city of Baltimore and whether any observed excess can be
associated with previous exposure to arsenic.  The plant has produced
insecticides,  herbicides, and other arsenic products from 1897  until early
1976.  In 1952, the original plant was torn down and a new one  erected with
better hygienic conditions for the workers.  The plant produced arsenic acid,
calcium and  lead arsenate, Paris green  (a cupric acetoarsenite), and sodium
arsenite.   In  the  past, all products were dried and packaged except sodium
                                          9 f
0

-------
arsenite which was shipped as a liquid.  Paris green was not produced after
the early 1950's and no dry arsenicals after 1973.  Other pesticides such as
chlorinated hydrocarbons and organophosphates were not produced at this facility
but were made into formulations on-site since 1947.  There are several other
industries which are currently located in the area or have been manufacturing
in that vicinity in the past.
METHODS
     The census tracts which were selected as having had possible environmental
exposure to arsenic from the point source of the pesticide plant were defined
empirically as those for which at least 50 percent of their area lay within a
3/4 mile radius of the plant.  This distance was chosen so that large tracts
which lay across the river and in which the majority of the population did not
reside within a one mile radius of the plant would not be included.   The four
index census tracts which fit these criteria were 2303, 2302, 2404 and 2301.
The tract in which the plant was located was 2303.
     The comparison group of census tracts consisted of all tracts which matched
the index ones on age distribution, race, sex and socioeconomic factors.  Index
tracts 2303, 2302 and 2404 were similar in these matching characteristics and
were compared to the same set of comparison tracts designated as Match I.  Index
tract 2301 differed in age and race distribution from the others and was compared
to a second set of tracts, Match II.  Death rates for the index tracts in 1958
through 1962 were compared to tracts selected for matching through information
available from the 1960 census.  The matching criteria used to select control
tracts for comparison of death rates in all other years were derived from the
1970 census.  The age distribution differs in the tracts across time intervals,
so to compare rates between years, the figures have been age-adjusted.

-------
     The initial matching criteria for the 1970 census were:
          Age distribution + 10% for each age
          Race + 15%
          Sex ± 5%
          Median income + $1,000
          % below poverty level + 10%
          % head of household over 65 years + 20%
The matching criteria based on the 1960 census were the same  except  that  the
variation in median income was reduced to reflect current inflation, and
information on  the last two characteristics  was not available in the earlier
decade.
     The matching tracts in 1970 and  1960 are shown  in  figures  1  and  2.   A total
of 18 Match I control  tracts was  identified  from 1970  census data and  45 tracts
from 1960 data.   A total of five Match II tracts was  found in  both  census
periods.  The variation in the  numbers of Match I tracts  between  the  two  periods
is the result of changing racial  distribution especially  in middle income census
tracts over the past ten years.   The  three index tracts have  a  predominantly
white population and fewer census areas have  that racial  distribution in  the
later time period.
     The index area was stable  with  an increasing proportion  of individuals
living in the same household for five to seven years from the 1960 to the 1970
census.  This stability is also reflected in  the slight increase  in  age of the
population of the area.
     The scattered distribution of the control tracts has placed  them in  areas
which may also have had different risks.  The adjacent  tracts contiguous  to
the index ones may have had minimal  exposure  to the  same agents as in the major
area.   Southern tracts are in heavy industrial areas as are the central tracts
                                           2d

-------
but the characteristics of the populations and their stability are different.
The northern area consists of mainly residential dwellings with little industrial
exposure.  For tnese reasons the controls were divided into four groups for
comparison - adjacent, south, central, and north.
     Cancers were identified by examining all certificates of deaths which
occurred within the city for the years 1958-62 and 1968-74.  The death was
selected for study if cancer appeared as a cause listed anywhere on the death
certificate, with the exception of the years 1973-74 where only cancers listed
as underlying causes were chosen.  Deaths of city residents were selected from
the total cancer list.  This procedure would not include the deaths of city
residents which occurred outside the city.  In order to determine the extent of
these differences we abstracted information on out-of-city deaths of city
residents for the three years, 1970-72.  The proportional increase in deaths
for the  index tracts was five percent and for the control tracts 13 to 15
percent.  This difference is not large enough to account for the variation in
cancer rates observed.  Deaths were included only once using either underlying
cause or first cancer listed.  Adjustments were made in the changing codes in
the 7th  and 8th revision so that data by site of cancer were compatible for the
total period.
     The hospital records of a sample of cancer deaths were reviewed to verify
the accuracy of death certification of cancers  in Baltimore, to identify any
possible differences in diagnosis by area in the city, to determine any variation
in pathological characteristics of cancers in index and comparison areas and
to investigate differences in personal characteristics such as smoking as
described in hospital charts.  The review specifically focused on unusual
cell types of lung cancer and possible arsenic-associated  symptoms and diseases

-------
in cancer patients from the index and control areas.



     The soil  was sampled for the presence of arsenic in the areas near the



chemical plant.  The original selection of sampling sites was determined both by



distance from the plant and by direction from north through south coordinates.



We intended to collect about half the samples within the 1/4 mile radius and



40 percent at the next 1/4 mile distance with the remaining samples collected



further out on the radii.  Control samples would be taken from two parks nearby



but a distance greater than 1 mile from the plant.  The field survey team had



problems adhering to the sampling design since the sources of soil were limited



in the area.  We attempted to take samples near residences whenever possible



as long as there were no obvious problems of tree-cover, water run-off, or



redevelopment.  For those few samples taken at private housing, the residents



were interviewed concerning the use of herbicides or pesticides in the area



and the sample was avoided if the soil had been treated.  After collection



of the original 101 samples taken at 35 sites under these directives and



including additional samples in the park, a second set of samples was collected



in a north and north-west direction to determine how far distant the high



levels  could  be detected.  Special emphasis was placed on sampling from the



park which was adjacent  to the plant.  This park has a central grassed area



which had been recently  re-sodded.  Surrounding the park was a dirt-track



which had been undistrubed.  Part of this path was adjacent to the fence along



the plant boundary and near the areas where railroad cars were filled.  Another



portion bordered on the  water and the last was adjacent to railroad tracks.



     Samples  were collected at one, two and four inches at each location unless



otherwise noted.  A core sampler, with a 3/4 inch bore and marked at one inch



intervals, was driven  into the ground and samples were removed down to the
                                         24

-------
appropriate  depth marked.   A one foot circle was  marked  off around  a  selected
site  and  a  set of samples  was collected according to the described  technique
until  the 30 ml.  polyethylene sample bottle was filled with soil  from the
appropriate  depth but from different core samples.   Initially we  had  tested
four  sites  using  consecutive one inch samples down  to a  depth of  four inches.
We  found  that samples at three inches were usually  close to those at  four and
thus  it was  elected to take the extreme depth and discard the three-inch level.
Samples in  control areas were all obtained from two city parks, Riverside or
 Federal Hill.
      All  instruments used  in collecting samples were free of arsenic.  The
 analysis  was done using either conventional flame or flameless atomic absorption
 spectrophotometry depending on initial level of arsenic.

 RESULTS
Mortality  by Tracts
      The  crude rates for cancers at four specific sites, oral, pancreas, lung
 and prostate as well as for all cancers are presented for males in  tables 1 and
 2.   The first table includes data for the five-year period around the 1960
 census and  the second table for a seven-year period around the 1970 census.  As
 can be seen the risk for lung cancer and for all  cancers is excessive in the
 period around the 1970 census for tract 2303 compared to any of the control
 groups.  This is not consistently true in the earlier period. The  weighted
 relative  risk of lung cancer in white males from  index tract 2303 as  compared
 to  north  controls which had the highest control  rate is  2.5 as shown  in  table  2
 with a probability of .0005 as determined by the  chi square calculated  by the
 Woolf-Haldane method.  The black males in tract  2301 also have a  higher  rate  of
 lung cancer but this was not true for white males in the same tract.   In the  1960

-------
census period, although the lung cancer rate is higher in white males in tract
2303 than the north and south controls, there is very little difference between
the rates for all index tracts and for the adjacent and central controls.  There
are no differences in rates for males in tract 2301 and their control groups.
     If we examine the comparable crude mortality rates for females  in tables
3 and 4, we can find no excess risk of cancer at any site for census area 2303.
In fact, the overall cancer rate appears somewhat low especially in the 1958-62
period.  The mortality from breast cancer is slightly high in the early period
and there are no deaths from cervical cancers.  The lack of an observed increase
in lung cancer in women in 2303 might be the result of a small population size.
This will be discussed later in the report.
Adjusted Rates
     The age distribution of the index tract changed with time and these
differences were reflected in similar changes in the control group.  In order
to have appropriate comparisons the mortality rates for each cancer site and
for all cancers  have been adjusted using the method of standardized mortality
ratios.  The  average annual Baltimore City mortality rates were calculated
from  all deaths  in the 1968-74 period and these values were used as standards
to adjust the mortality in each time period.  As seen in table 5, the mortality
ratios  for white males in the tract 2303 were high for cancers of the lung,
pancreas, stomach, prostate, oral cavity and all sites.  The numbers of deaths
except  for lung  and all sites were small but the pancreas cancer rate was still
significantly higher than that for the city.  White females in 2303 had an
unremarkable  overall cancer rate with excesses noted only for oral and rectal
cancers of which only the latter ratio is significantly greater than unity.
      In figure 3, we examine the lung cancer mortality in two or three
 /e-ir  time intervals.  Using rates adjusted  by the direct method to the 1970

-------
Baltimore City population as a standard, we find that the death rate for this
cancer has always been higher in males from tract 2303 than from most controls
but that it has been rising rapidly.  The "all  cancer" rates have also shown
higher values than among controls.  A preliminary look at the lung cancer rates
for 1950-51 indicated that the adjusted rates for that period were high for tract
2303 with a rate of 253 per 100,000 population as compared to rates ranging
from 35.8 to 87.4 in other index tracts and controls.
     Employees of an industry may live in close proximity and could have
accounted for an increased mortality in the census tract due to occupational
exposure.  With the cooperation of the company and the investigators studying
the employees we reviewed lists of all employees to match with known deaths.
Four employees were found among the cancer deaths in tract 2303 but removing
these individuals did not change the significance of the rates.
     The geographic distribution of cases was plotted on spot maps as shown
in figures 4 and 5 for the two census periods.  For both periods, lung cancer
appears to be concentrated in an area about eight blocks wide and nine to
twelve blocks long lying to the north and east of the plant.  The area
encompasses all of tract 2303 and parts of 2302 and 2301.  If one takes all of
tracts 2303, 2301 and 2302 which lie within a 3/4 mile radius of the plant
and calculates the proportion of lung cancers to all cancers in this area
compared to the remaining census areas on the figure  4, the proportion is
44.4 percent near the plant and 16.8 percent in the outer areas.  For figure 5,
the proportions show similar differences for 1973-74 as in the previous three
years.  For the area within the defined census tract and 3/4 miles of the plant,
lung cancer represents 47.8 percent of the total cancers whereas in other areas
it is only 33.3 percent.  The northerly direction of this lung cancer excess is

-------
not compatible with the strong wind directions in that area.  These winds
arise in the northwest and west and should have carried contaminants to the
east and southeast of the plant.  The particles may have been moved by
gentler winds and deposited nearby.
Soil Sampling
     The highest arsenic levels are shown in figure 6.  In most cases these
levels occurred at 2 inches suggesting higher contamination in the past.
Occasionally, as in the area adjacent to the plant, the levels were highest
at one inch.  In general, arsenic levels were highest where lung cancer
mortality was also highest.  The mean arsenic level from 20 sample sites in
tract 2303 was 63 ppm of arsenic.  Even the omission of samples from the park
adjacent to the plant only reduced the mean arsenic level to 38 ppm.  Tracts
2301 and 2404 had means of 6 ppm and 2302 a mean of 4 ppm based on only 2 to
4 sample sites.  All sites in the park had high levels except for an area which
has  been turned over and resodded and in which low arsenic levels were present.
The  one inch  levels near the fence were as high as 695 and 226 ppm whereas at
the  opposite  side of the park the values were only 29 to 97 ppm at one inch
but  as high as 46 to 161 ppm at two inches deep.  From the soil levels of the
original samples, the data indicated high levels within a 3/8 mile radius of
the  plant.  It was also apparent that higher levels were found in a northerly
direction along the railroad lines.
Hospital Validation
     The deaths of the total 14 years of study were included in the sample
and  stratified by control and index census tracts.  The sample for hospital
record review included all deaths for residents in the index tracts.  Deaths
for  control tracts were stratified by age, race, and  sex and three time periods,

-------
1958-62, 1966-67 and 1968-74.  Four control deaths were selected randomly from
each stratum for each index tract death within the same stratum.  For the
following analyses no attempt was made to expand the sample to the original
population size.
     The hospital abstract form included information on the following variables:
          1.  Cancer diagnosis
          2.  Final diagnosis other than cancer
          3.  Source of information for cancer diagnosis
          4.  Arsenic-associated symptoms
          5.  Personal characteristics as smoking and occupation
          6.  Description of pathological specimens; operative or autopsy 7indings
Verification of the identification of the correct individual on the hospital
record was done by name, birthdate, residence, and date of death.
     Records were reviewed in eleven of the hospitals in Baltimore City.  The
remaining five non-cooperating hospitals were small and did not limit sub-
stantially the number of records reviewed.
     All possible medical conditions found on record review were listed and
coded by the same nosologist who coded all the death certificates.  The cell
types were classified, in general, according to the Manual of Tumor Nomenclature
and Coding.  Since this coding scheme does not appropriately classify the
cells of several tumors, especially those of non-solid origin, a revision of
the scheme was made to include these cancers if we felt that their frequency
was sufficient to warrant specific classification.

     The causes of death were grouped into two time periods which represented
the use of the 7th and 8th ICDA codes and  grouped into causes as listed on  the

-------
certificate by the first two digits of the code.   These causes were then
compared to the first four medical  conditions or  diagnoses as noted on the
hospital records.
     The problem of validation of death certificate information was reviewed
further by a physician who examined the data on the abstract forms.  As indicated
in table 6, there was complete agreement in diagnoses to four digits in the
ICDA code in only 75.0 percent of the cancer deaths.  If classification to
three digits only is used we will correctly verify 80.7 percent of the cases
listed on the certificate.  In 1.8 percent of cases metastatic lesions were
identified on the death certificate as underlying and in another 5.5 percent
multiple cancers were listed on the certificate and the primary site varied
from that listed on the hospital record.  There was no cancer diagnosis listed
anywhere on the hospital record for 2.7 percent of deaths.  A further examination
of the method of diagnosis of cases was attempted in order to demonstrate
whether differences in the methods might have changed the accuracy of death
certification.  Data from autopsy and histological examination of tissue
were used for the diagnosis of 82 percent of the cases with complete agreement
in records and 88.9 percent of cases where the agreement was less than perfect.
Therefore, the consistency of cancer diagnosis on hospital record and death
certificate is not related to the method by which the cancer was identified.
     The method of cancer diagnosis differed only slightly in the larger
hospitals with 69.7 to 91.4 percent of cases diagnosed by autopsy or histology.
An examination of the differences in diagnosis by census tract has not been
completed, but it is unlikely that there will be variations in results since
we have included all hospitals used by individuals from the index area in the
above evaluation.
                                        30

-------
     An examination of the hospital records for possible arsenic-associated
symptoms included gastrointestinal signs, skin lesions, Mee's lines on nails,
neurological or neuromuscular symptoms, cardiovascular disease, stroke and
asthma.  Only respiratory symptoms were slightly higher in the index tract
but since there were so many records in which there was no comment about
these symptoms, it is difficult to interpret the small variation.  We also
sought information on diseases for which arsenic might have been used as a
treatment, such as syphilis, trypanosomiasis and amebiasis and the results
indicated no higher frequency of these conditions among residents of index
tracts.
     Both smoking and drinking histories were abstracted from hospital records.
Drinking habits were rarely recorded and smoking histories were also frequently
missing.  Table 7 indicates the smoking characteristics of lung cancer deaths
in index and control tracts as determined from the hospital records.  For
46.6 percent of the patients, the smoking histories are unknown.  Despite that
fact, we attempted to compare the smoking levels in index versus the control
tracts.  The percent of smokers is slightly higher in the index tracts but
the difference is not impressive.  If one includes only those charts with a
recorded history, almost all cases are positive for smoking in both index and
control tracts.
     The original hypothesis was that if arsenic had caused the lung cancers,
the cell type of lesions from the index tract might differ compared to other
areas with an expected predominance of small cell or oat cell tumors in the
exposed tracts.  The data in table 8 would indicate that the cell types differ
very little from index to control tracts.
                                            3.1

-------
DISCUSSION
     An excess mortality from lung cancer has been demonstrated among men
living in a highly industrialized area of South Baltimore over a period from
1966 through 1974.  The death rate is significantly higher than in control
tracts in the later years.
     The area surrounding the pesticide plant has high levels of arsenic in
the soil which corresponds  generally to the same areas where a high proportion
of lung cancers to other cancers has occurred.   There was no attempt to correlate
directly arsenic levels to  residences of lung cancer deaths.
     The review of hospital  records did not indicate that the excess of lung
cancer deaths had occurred  because of variations in diagnostic practices,
cell types or other factors.  The information on other risk factors was poorly
ascertained from hospital records.
     There are some definite questions which arise in regard to the data.
Why did the excess risk appear primarily in the late 60's and early 70's when
the plant had existed and produced arsenical products since the early 1900's?
The discrepancy could indicate that the plant did not account for the excess
but some other local industry or occupational group accounts for the excess.
It is also possible that the men in the area had a higher frequency of smoking
and smoked a higher dose of cigarettes than did populations in the rest of the
city.  It is possible that  selective mobility of younger, healthier males has
left the area with a high risk among the remaining group.
     The sudden rise in lung cancer might be related to the destruction of the
old plant in 1952.  Such an undertaking could have spread dust diffusely through-
out the community.  Under these circumstances we must ask why the concentration
of lung cancer in the area  does not coincide with the assumed wind spread of
                                           3,:

-------
particles.  It is necessary to further examine the mortality in the 1950
period to determine whether an excess existed at any time before the destruction
of the old plant.  It would be interesting to see if the appearance of the
excess risk of lung cancer in the community coincided with that found in the
workers within the pesticide plant.  If we presume that arsenic may not be
causing the excess then it would be necessary to examine the mortality experience
of workers in other industries in the area, especially the natural gas plant,
to see if they have an excess lung cancer mortality.  In almost all cases,
workers within an industry should have higher exposure and a greater risk of
disease than the general public.  It is necessary to investigate whether the
increase in lung cancer can be related to a change in production or methods of
operation of any of the businesses.  For example, differences in handling
arsenic, changes in formulation of pesticides or the conversion of the gas
plant from carburetted water gas to oil gas production could have created
variations in level of type of pollution.
     The fact that the excess lung cancer mortality has occurred only in men
raises the question as to whether another environmental factor, differences
in smoking characteristics, or occupation has caused the increased death
rate in tract 2303.  It is possible that smoking plus an  environmental
pollutant are required to produce the excess of cancer.  The rates in older
women then could be lower because they did not smoke and the possible
synergistic effect of cigarettes and the environmental factor were not
observed.  Many of these questions might be determined by a community survey.
     Further sampling for arsenic should be done to determine at what distance
the levels actually returned to background.  It was first  thought that the
high levels along railroad lines might indicate a relationship to previous coal
use.   However, further investigation showed that the arsenic content of coal
                                         33

-------
in local  use did not reach levels  as  high  as  those measured along  the  tracks.
Rail  transport of materials from the  plant may  have  been  related to the high
levels.   Further investigation of  this  possibility is  needed.   It  appeared that
use of herbicides did not explain  the arsenic levels in the rail beds.
     In summary, men living in close  proximity  to a  chemical  plant which
produced arsenicals have a higher  risk  of  lung  cancer  than comparable  individuals
in other areas of the city.  The distribution of arsenic  in the soil near the
plant and along the railroad line  is  higher than in  control areas.
                                           34

-------
Cause
                                   Table 1
 Average Annual Crude Death Rate per 100,000 for  Index and Control  Census  Tracts
                                 Deaths 1958-62
                                     Hales
                                    Match I
 2303         2302         2404        Adjacent      South        Central       North
#    Rate    #    Rate    #    Rate    #    Rate    #    Rate    #     Rate    #    Rate
Oral
Pancreas
Lung
Prostate
All Cancer

00
c.t Cause
Oral
Pancreas
Lung
Prostate
All Cancer
0
1
4
2
11


0
0
3
1
10
-
18.5
74.1
37.1
203.9

2301
White
Rate
-
-
75.2
25.1
250.6
2 21.1
0
12 126.8
2 21.1
26 274.8

Nonwhite
# Rate
0
1 16.8
2 33.5
2 33.5
16 268.2
1 9.
2 19.
8 75.
3 28.
32 303.

5 5
0 3
9 23
4 3
5 56
Match I
Adjacent
White Nonwhite
# Rate # Rate
1 21.8
0
2 43.6
1 21.8
8 174.3 1
0
1 22.9
1 22.9
1 22.9
2 274.9
19
11
88
11
216
I
.3 10
.6 8
.8 79
.6 19
.1 244

8.3
6.6
65.4
15.7
202.1

20
31
193
43
615

Central
White Nonwhite
# Rate # Rate #
0
1
5
0
14
-
17.7
88.7
-
248.2
0
1 18.
6 113.
0
17 322.
3
9 2
6 12
5
0 43
7.6 11
11.8 18
73.6 60
16.4 23
234.6 293

8.9
14.6
48.8
18.7
238.2

North
White Nonwhite
Rate # Rate
18.4 0
12.2 0
73.4 8
30.6 3
263.2 39
-
-
39.3
14.7
191.6

-------
CO
                                                         Table 2
                       Average Annual Crude Death Rate per 100,000 for  Index and Control  Census  Tracts
                                                        Deaths  1968-74
                                                            Males
                                                           Match  I
                                                        Census  Tracts

Cause
Oral
Pancreas
Lung
Prostate
All Cancer



Cause
Oral
Pancreas
Lung
Prostate
All Cancer

*
2
2
18
1
33


White
2303
Rate
33.9
33.9
305.0
16.9
559.2

2301

i Rate #
0
0
7 113
1 16
17 275
1
3
.3 13
.2 3
.0 27
2302
#
0
0
12
1
26

2404 Adjacent
Rate #
-
-
103
8
223

0
1
.3 12
.6 2
.8 24

Rate #
4
8.0 3
96.1 24
16.0 4
192.2 86
Match II
Rate
12.7
9.6
76.4
12.7
273.9

Adjacent
Black
Rate
16.9
50.7
219.5
50.7
455.9

#
2
0
13
4
34
White
Rate
20.5
-
133.1
41.0
348.2
Black
# Rate
2 20.3
.3 30.5
16 162.6
4 40.6
42 426.7

#
1
2
4
1
17
South .Central
# Rate I
7 6.8 6
9 8.8 9
111 108.3 60
15 14.6 25
254 247.9 206

Central
White Black
Rate # Rate
17.5 0
35.1 0
70.1 5 102.8
17.5 2 41.1
298.0 13 267.2
Rate
9.0
13.4
89.6
37.3
307.7

North
#
5
3
48
7
no

Rate
11.7
7.0
112.4
16.4
257.7

North

#
0
0
15
3
41
White
Rate
-
-
124.5
24.9
340.3
Black
i Rate
1 6.9
1 6.9
6 41.3
3 20.7
28 192.9

-------
                                  Table 3
Average Annual Crude Death Rate per 100,000 for Index and Control Census Tracts
                                 Deaths 1958-62
                                     Females
                                     Match I

Cause
Oral
Pancreas
Lung
Breast
Cervic
All Cancer
2303
# Rate
0
0
0
3 57.5
0 T
6 115.1
2302
#
0
2
0
2
1
17
Rate
-
19.9
-
19.9
10.0
169.2
2404
# Rate
0
0
0
3 28.7
1 9.6
13 124.4
Adjacent
# Rate #






Match


Cause
Oral
Pancreas
Lung
Breast
Cervix
All Cancer
2301
White
# Rate
0
0
0
3 71.4
1 23.8
6 142.9


Nonwhite
#
0
0
1
2
1
10
Rate
-
-
15.8
31.5
15.8
157.6
Adjacent
White
# Rate
1 27.4
1 27.4
0
3 82.2
1 27.4
10 274.0

0
0
6 22
8 30
3 11
13 162
II

Nonwhite
#
0
1
0
1
1
8
Rate
-
27.3
-
27.3
27.3
218.6
2
8
.6 8
.2 28
.3 15
.3 186


White
South
Rate
1.6
6.4
6.4
22.5
12.0
149.3

Central
Central
#
2
23
18
81
34
460


Nonwhite
if Rate #
0
0
1 19
2 39
2 39
9 177
0
0
.7 0
.5 1
.5 1
.7 6
Rate
-
-
-
17.1
17.1
102.6
Rate
0.7
8.4
6.5
29.4
12.4
167.2


North
#
4
12
15
51
12
Rate
2.9
8.8
11.0
37.5
8.8
263 193.3

North
White
$
0
1
1
8
5
41
Rate
-
5.8
5.8
46.6
29.1
238.9


Nonwhite
# Rate
0
0
0
4 18.3
3 13.8
21 96.3

-------
Co
                                                        Table  4
                      Average Annual Crude Death Rate per 100,000 for Index and Control Census Tracts
                                                       Deaths  1968-74
                                                           Females
                                                           Match I
2303
Cause
Oral
Pancreas
Lung
Breast
Cervix
All Cancer


#
2
0
1
1
0
9


Rate
33.3
-
16.6
16.6
-
149.9

2301
White
Cause
Oral
Pancreas
Lung
Breast
Cervix
All Cancer
#
0
1
2
5
0
15
Rate
-
16.4
32.8
82.0
-
246.0
2302
2404
# Rate #
0
2 16
3 25
1 8
3 25
14 117

0
.8 2
.2 1
.4 2
.2 0
.8 17

Rate
-
15.0
7.5
15.0
/
127.3

Adjacent South Central
#
4
1
7
8
3
56
Match II
Adjacent
Nonwhite
# Rate
0
0
1 15.4
3 46.2
2 30.8
18 277.1
White
# Rate
0
1 10.9
2 21.9
6 65.7
3 32.8
26 284.6
Nonwhite
#
1
2
7
4
1
25
Rate
9.7
19.3
67.6
38.6
9.7
241.5
Rate #
12.1 3
3.0 11
21.1 16
24.1 25
9.1 6
169.0 165

Rate #
2.7 3
9.9 10
14.4 18
22.5 29
5.4 5
148.7 175

Central
White
# Rate
0
1 16.7
1 16.7
2 33.4
0
14 233.6
Nonwhite
# Rate
0
0
1 18.1
0
1 18.1
5 90.5
Rate
4.1
13.6
24.5
39.5
6.8
238.2

North
White
# Rate
0
3 23.0
3 23.0
8 61.4
2 15.4
34 261.1
North
#
2
1
9
9
0
78


Rate
4.3
2.2
19.5
19.5
-
169.2


Nonwhite
#
0
0
0
3
1
16
Rate
-
-
-
18.1
6.0
96.3

-------
CO
 I
1 U U 1 C J
Tine-Adjusted Si-R's Based on Average Annual Baltimore City Rates*
Deaths 1958 - 1962 and 1966 - 1974
Match I White Males

Cause
Oral
Storacn
Colon
Rectum
Pancreas
Lung
Prostate
Bladder
Lymphonas
All Cancer


Cause
CO
CO Oral
Stomacr.
Colon
Rectum
Pancreas
Lung
Breast
Cervix
Bladder
Lymphonias
All Cane or
* Averauc i

obs
3
2
3
1
5
25
3
0
2
54


obs

2
1
1
4
0
1
5
0
0
1
18
innua
2303
SMR
2.45
1.60
1.22
1.09
4.15
2.74
1.59
-
0.88
1.94

2303
SMR

6.31
1.32
0.45
6.12
-
0.71
1.29
-
-
0.65
0.96
1 Baltimo
2302
obs
2
3
9
3
0
30
3
3
4
67

SMR
0.88
1.18
1.74
1.62
-
1.72
0.68
1.38
0.89
1.21

2302
obs

0
3
7
0
4
4
5
5
0
4
41
re Cit
SMR

-
1.74
1.36
-
2.32
1.34
0.61
2.30
-
1.19
1.00
v rates
240-1
obs SMR
1 0.43
4 1.59
6 1.21
4 2.19
3 1.31
23 1.31
5 1.26
2 0.97
2 0.44
63 1.15

2404
obs SMR

1 1.40
0
5 0.94
2 1.32
2 1.13
1 0.31
6 0.68
3 1.27
0
3 0.86
39 0.91
; (based on
Adjacent
obs
9
10
9
6
7
57
7
5
10
168
Match I
Adjacent
obs

4
5
13
3
1
15
17
6
0
6
109
deaths ir
Controls
SMR
1.43
1.44
0.62
1.18
1.09
1.19
0.56
0.85
0.80
1.10
South
obs
18
23
43
14
21
206
36
20
26
543
Controls
SMR
0.88
0.98
0.94
0.85
1.01
1.35
0.93
1.03
0.62
1.10
Central
obs
29
60
67
34
43
282
74
37
65
891
Controls
SMR
0.91
1.40
0.91
1.25
1.27
1.21
1.15
1.11
1.02
1.13
North
obs
16
17
38
15
21
118
31
28
37
423
Controls
SMR
0.89
0.70
0.90
0.97
1.10
0.93
0.79
1.45
1.06
0.96
o
o>
3
O
(V
T
O
rt
Ol
^
<<
3
DO
Ol
€-»•
-5
1
3
o>
3
O
l/l


White Females
Controls
SMR

2.06
0.96
0.83
0.69
0.19
1.70
0.70
0.97
-
0.61
0.90
1 1958-196;
South
obs

5
14
41
14
20
26
58
25
10
36
390
') were
Controls
SMR

0.84
0.89
0.89
1.08
1.34
0.99
0.75
1.13
1.33
1.22
1.05
aoolied t
Central
obs

5
37
81
33
38
37
119
43
15
52'
682
o the 19
Control s
SMR

0.57
1.30
1.01
1.57
1.59
1.05
0.94
1.14
1.15
1.05
1.10
60 match c
North
obs

6
18
40
13
13
23
61
13
6
24
350
lomilat
Control s
SMR

1.14
1.01
0.82
1.03
0.90
1.14
0.82
0.61
0.74
0.83
0.96
ion and werf














l
                    wc'ighled for five years.  Average annual  Baltimore City rates (based  on  deaths  in  1968-1972) were applied  to  the 1970
                    motet, population and were weighted for nine vears.

-------
                             Table 6
      Level of Agreement between Death Certificate Cancer
                 Cause and Hospital Diagnosis
                               Number     %
Complete Agreement              555     75.0
(4 digits in ICDA code)                        J     80.7
Agreement to 3 digits            42      5.7
Agreement to 2 digits            31      4.2
Metastasis entered on D.C.       13      1.8
  as underlying
Multiple cancers on D.C.         41      5.5
  Primary site not stated
Other                            18      2.4
No cancer at autopsy or          20      2.7
  biopsy
No records available             20      2.7
           Total                740

-------
                         Table 7
     Sroking History in Lung Cancers from Index and
                 Control  Tracts by Sex
                  Smoking  . Non-Smoking   Unknown   Total
                  No.  %     No.   %      No.  %     No..
Index Tracts
Male
Ferna
Ccntro
Male
Ferna

le
1 Tracts

le
30
6

63
5
59
67

52
22
2
3

0
0
4
33

0
0
19
0

58
18
37
0

43
78
51
9

121
23

-------
                   Table 8
  Cell Types of Lung Cancer  by  Census Tract
Oat  Squcir.aijs  Adenocarcinoma   Epidermoid  Other  UK  Total
Ire


Cth


Con


ct 2393
Male
Female
er Index
Male
Female
trol
Male
Fernale

2
-

3
-

12
4

8
1

18
3

57
4

-
1

2
3

16
9

1 1
-

1 1
1

4 3
4

4
-

8
-

18
1

16
2

33
7

no
22

-------
Figure 1.  Map of Baltimore City showing location of 1970 index and control
           census tracts.

Figure 2.  Map of Baltimore City showing location of 1960 index and control
           census tracts.

Figure 3.  Age- and time-adjusted rates per 100,000 for tract 2303 and  North
           and Adjacent Controls.  White males.  All cancer and lung cancers.
           (semi-logarithmic scale)

Figure 4.  Spot map showing cancer deaths for 1970-72 by residence at death,
           excluding chemical plant employees.

Figure 5.  Spot map showing cancer deaths for 1973-74 by residence at death.

Figure 6.  Arsenic level in soil - ppm.  Highest value at each site.
           Summer 1976 and Spring 1977

-------
                              References
Baetjer, A., Lilienfeld, A., and Levin M. (1975).  Cancer and occupational
     exposure to inorganic arsenic.          In abstracts.  18th International
     Congress on Occupational Health, Brighton, England,  393.
Blot, W., and Fraumeni J. (1975).  Arsenical air pollution and lung cancer.
     Lancet      2 ,142-144.
Braun, W. (1958).  Carcinoma of the skin and the internal organs caused
     by arsenic:  Delayed occupational lesions due to arsenic.  German
     Med Monthly 3, 321-324.
Neubauer, 0. (1947).   Arsenical cancer:  A review.  Brit <1  Cancer 1,  192-251.
Ott, M.G., Holder B.B., and Gordon, H.L. (1974).  Respiratory cancer  and
     occupational exposure to arsenicals.  Arch Environ Health 29, 250-255.
Tseng, W.P., Chu, H.M., How, S.W., Fong, J.M., Lin, C.S., and Yeh S.  (1968).
     Prevalence of skin cancer in an endemic area of chronic arsenicism
     in Taiwan.  J_ Nat Cancer Instit 40, 453-463.
                                           44

-------
                             DISCUSSION

     DR. HUNT:  I think we do have about five minutes for questions.
We can proceed on that basis.  Are there any questions for Dr. Matanoski?
     DR. O'CONOR:  It is a very impressive study.  I have two questions.
One has to do with the characteristics or definition of the population
in the indexed census tracts and how that compares with the control
census tract.  The other has to do with similar industries to the one
that is now apparently defunct in your area.  What kind of controls or
regulations are in effect now for plants to benefit from the kind of
experience which you have described?
     DR. MATANOSKI:  The control tracts were actually matched for the
several characteristics to the index tract, namely age, race, sex,
socioeconomic status and percent at poverty level.  So they were
similar in these characteristics with very small deviation.  One thing
I did not emphasize is that the data that you have seen was subsequently
examined with employees excluded, and that did not change the observed
difference, the epidemic still persisted.  We had a list of all employees
provided through the industry which allowed us to accomplish this task.
     Your second question related to what my advice would be for control
of this situation.  The problem in this plant was apparently from dust.
At least when we observed the operation externally, it had a very
high dust level and the material was circulating very close to the ground.
The plant did not have high stacks.  Thus, the material was not moved
very far away.  We had not anticipated this distribution of arsenic.
We expected the material perhaps to be carried by wind currents away
from the plant and further out into the population.  Instead, the arsenic
remained in the very community close to the plant.  If one could manage
local dust problems, this type of spread could be avoided.

-------
       EPIDEMIOLOGIC STUDY OF A POPULATION PREVIOUSLY EXPOSED
                          TO HEXACHLOROBENZENE
          1222                 3
A. Gocmen , D. Cripps  , H. Peters , G. T. Bryan , and C. R. Morris

1                                                        2
 Hacettepe  University  Medical  School, Ankara,  Tud
-------
This is a picture of two sisters.  The younger sister who is age 17 years appears on the
left.  She has a normal appearance and is of normal stature.  The sister on the  right
was exposed to HCB during her childhood. As you can see, the HCB exposed individual
had scarring  on the face and  hands.  You can  also see the smaller  hands, the short
fingers, and the arthritis of the hands.  These characteristics are common among many
of the HCB survivors.

-------
This picture depicts the enlarged thyroid observed  in several of the HCB exposed
individuals.

-------
This picture depicts another example of an HCB-exposed individual with an enlarged
thyroid.   The  presence of enlarged thyroids in the  HCB exposed  population  are
presently being viewed  with suspicion in light of  recent experimental studies in which
thyroid tumors have been observed in hamsters exposed to HCB.  We are presently
trying  to determine whether the enlarged, goiter-like symptoms are endemic to this
Turkish population or are a manifestation of the HCB exposure. We plan to request
the pathologic findings  on those individuals who elect to have their thyroid condition
treated surgically and  have  requested thyroid scan  data on  all individuals  whether
surgically treated or not.  As  necessary, individuals with thyroid adenoma will  be
studied in more detail at the Hacettepe Hospital in Ankara.
                                          4;)

-------
This picture depicts the  extensive facial scarring,  the shortened fingers, and the
painless arthritis observed in several of these HCB exposed individuals.

-------
This study  represents  a unique oportunity to observe  the human effects of HCB
exposure. " The current health status of previous exposed individuals may provide us
with additional information relevant to the potential chronic effects of HCB. We have
a particular interest in the status of youngsters born of mothers who were 6-12 years
of age at the time of their initial exposure.

We have been most fortunate and very appreciative of  the coopration shown  by  the
Turkish authorities and clinicians.  Dr.  Ayhan  Gocmen, a  Turkist  clinician from
Hacettepe Medical School of Ankara, Turkey, has been indispensable in his efforts to
assist us in the clinical aspects of the study. Dr. Gocmen was initially involved in this
problem when he was a resident working in Eastern Turkey on this particular problem
shortly after the  initial  incidence  was  observed  and  has  been  associated with
evaluating the problem ever since. He has been able to assist us in visiting a total of
10 villages in which over  100 individuals have been identified as survivors of their
initial acute exposure  some  20 years ago.  Records indicate that nearly 5,000 villages
may have been involved to various extents in this epidemic area suggesting a sizable
population for study.    Dr.  Gocmen continues to expand our investigaiton and  has
acquired the services  of several paramedical personnel  to assist him in  identifying
HCB  exposed individuals.   Indoctrination of personnel  at various medical  centers
throughout the epidemic area has also made them more  aware of these disorders.

Preliminary information to  date indicate that fecal and urinary porphyrins are still
being  excreted  in significant amounts  by  several individuals.   Urine  and  stool
porphyrins have been obtained on 100 individuals with  clinical  evidence of  porphyria.
These  data are being  compared with Turkist and U.  S. control  individuals.   The
preliminary results indicate that five subjects are still  porphyric after 20 years. Four
of these subjects have moderately increased excretion of porphyrins whereas the fifth
subject had a urinary  uroporphyrin of  1,607 micrograms/liter compared with Turkist
and U. S. controls of  only 5.17  and  9.0  micrograms/liter, respectively.  This  latter
individual also had a stool  uroporphyrin of 189.2 microgram/gram dry weight compared
with controls of 2.09 and 2.80 micrograms/gram, respectively.

Further follow-up studies are under way including the  identification of  additional
exposed individuals and quantitative analysis of HCB and porphyrin levels.

In addition, animal studies are also in progress in an attempt to correlate animal and
human symptomology as a  result  of a range of  HCB exposures.   These correlations will
be based upon both acute and chronic animal exposures and those estimated in HCB
exposed human populations.

In summary, we have observed clinical symptoms of porphyria including the excretion
of urinary  and stool porphyrins as well as HCB in maternal milk from HCB exposed
individuals.  These data suggest that HCB was accumulated in body tissues and  fat
stores for at least 20 years  from initial exposure.  These findings would support  our
continuing concern regarding the potential chronic effects of chlorinated hydrocarbons
and, in particular, the chlorinated benzenes, on human health and the environment.
                                            51

-------
                                     Discussion


Dr. Kraybill, NCI:   Is there a  parellelism in the United  States, because there was  an
episode, I believe in Louisiana or Texas, where cattle got  quite an insult  from HCB?
Would this be a population that one would be looking at?

Dr. Morris,  EPA:  There is a strong possibiliity that some of the clinical data that we
develope in  the Turkish incident can be correlated to other situations here in the United
States.  Certainly, the Louisiana incident is a prime example.  The Agency  does have a
fairly sizeable data base on some of that and we are going to be  looking at that data base
as we proceed in this particular study.

Dr. Plotnick, NIOSH:  Is there any indication of the levels of HCB in the original grain
that was  eaten.   Also, are  there  samples still available  and  have  impurities  been
analyzed. Some of these things appear to  me to be  more related, or possibly related, to
the impurities than to the unchanged HCB itself.

Dr. Morris,  EPA:  Dr. Gocmen's brother-in-law is a member of the Turkist  Department
of Agriculture. In fact, he was very helpful in getting us the records of distribution. We
are attempting now to methodically go through the list and identify storage facilities at
distribution points to see if there might be any residual sacks of grain which might  be
made available to us for analysis.

In terms of your  question about contamination, we  would  agree with you.  A variety of
potential  contaminants with which many of us are aware may be of equal concern.  I am
hopeful that when our research  team returns from their next visit that we might  have
some samples to  analyze. Of course, after 20 years, I do  not know exactly  what this is
going to mean either, but it certainly needs to be evaluated.

Dr. Fraumeni, NCI:  Are there  some clinical or experimental observations suggesting a
relationship to cancer  with this agent?

Dr.  Morris,  EPA:   Yes.  There  are animal experimental  data  showing an oncogenic
response in hamsters and mice when they are exposed to hexachlorobenzene.

Dr. Fraumeni, NCI:  Do you know what type of cancer?

Dr. Morris,  EPA:  The thyroid gland  appears to be a major  target organ. This is why I
indicated  in my talk that we are interested in these particular patients with  enlarged
thyroids.  I believe there has also been liver involvement.

Dr. Fraumeni, NCI:  There is a condition called porphyria  tada, which is associated with
cirrhosis,  which in turn predisposes to liver carcinoma. Do these patients have cirrhosis?

Dr. Morris,  EPA:  I do not recall seeing it.  We have noted enlarged  livers in some of
these patients; that is true.   During my last visit in October 1979, we  had one of the
patients die of leukemia. I do not know what that means, because it is just one patient
and he was only 26 years old. I think there may be a lot  more pathology once we  have
identified the  population.  I appreciate that the study  presently funded  through the
NCI/EPA  activity is a  five year study, but we are just now  at the breakpoint  of 20 years.
I think  in the next 20 years this particular population may provide us with a  lot  more
data.  So it is possible that this may be  one of those activities, which Dr.  Kraybill
mentioned, that we might consider extending.

-------
Cancer in Southern Louisiana:  Progress Report of a Case-Control Study
                 of Lung, Stomach, and Pancreas Cancer

                        Linda W. Pickle, Ph.D.


     Maps of U.S. cancer mortality by county for the period 1950-69
pointed to southern Louisiana as a high risk area for cancers of the
lung, pancreas, and stomach.  This interview study of newly diagnosed
cases of cancer at these sites and their appropriate controls was
designed to investigate the causes of the high rates and to follow
up leads generated by a recent study of 10,000 death certificates in
the area.  Information is being gathered by face-to-face interviews on
diet, ethnic background and alcohol consumption, occupational, medical,
and residential histories, as well as other factors related to the
unique Acadian culture of the study a-rea.
     Interviewing began in several large hospitals in June, 1979, and
has been expanded to include 22 hospitals covering the entire southern
half of the state.  Since the great majority of lung, pancreas, and
stomach cancer cases are diagnosed by anatomic pathology and cytology
specimens, the primary source of cases has been through pathologists'
reports.  In addition, though, the abstractor/interviewers work closely
with one physician on the regular staff at each hospital to ensure
that no cases are missed.  Controls are selected from hospital admissions
lists or medical records matched to the cases by age, sex, race, and
hospital.
                                          53

-------
      Interviewing  teams are  based  in both New Orleans and Baton Rouge
 to  reduce  travel time.  As of April 1,  1980, 998 interviews have been
 completed,  including  450  cases  of  lung  cancer, 68 stomach cancers,
 and 48 pancreas  cancers which represent an average of 83% of available
 study subjects.  We anticipate  continuation of the study in order to
 complete interviews of  1200  lung cancer cases, 150 pancreatic cancer
 cases, 200 stomach cancer cases, and an equal number of controls.
      In addition to the information obtained through interviews,
 manufacturing industries  with over 50 employees in the area were located
 and mapped for future comparison to the residences of cases and controls.
 It  may be  possible to utilize EPA  maps  of various emission levels in the
 study area to further define the residential exposures of the study subjects
      Interviewing  will  continue for at  least one more year.  No results
 are available at this time.
 Acknowledgement:
      This research is  being  conducted  by  Louisiana State University,
 Dr.  Pelayo Correa, Principal  Investigator,  under NCI/EPA Contract
#N01  CP 91023.
                   NO DISCUSSION FOLLOWING THIS PAPER

                                        54

-------
              SUPPORT SERVICES FOR STUDY OF BLADDER CANCER IN
              NEW HAMPSHIRE, VERMONT, AND MAINE (NEW ENGLAND)

                              Dr.  Robert Hoover
                          National Cancer Institute
                                   Abstract


     Field studies are being carried out in seven New Hampshire,  nine
Vermont, and seven Maine counties to determine environmental  and
possibly host factors which may be responsible for the high rates of
bladder cancer in both men and women.  The study involves a case-
control next-of-kin survey of at least three hundred persons  in each
group, a case-control incidence survey targeted for one hundred cases,
characterization of industrial and other relevant features of the study
area, and collection and analysis of air and water samples.  The  inter-
viewing, recording, and other field work are carried out under the
support services contract with Westat, Incorporated of Rockville, Md.
A considerable amount of time has been devoted to professional con-
tracts and discussions for carrying out the study.  The questions asked
are those used for the national saccharin/bladder cancer survey,  plus
investigation of the importance of edible bracken fern for possible
carcinogenic factors and possible importance of French Canadian back-
ground.
                          NO MANUSCRIPT RECEIVED
                                          55

-------
        A Case-Control  Study of Lung  Cancer  Near  A  Zinc  Smelter



    Linda M.  Pottern,  William J.  Blot,  and Joseph F.  Fraumeni, Jr.








Background and rationale



     In 1969 a cohort  study of employees  of  a  large copper  smelter  in



the Western United States revealed  a  3-fold  increased  risk  of  lung



cancer, reaching 8-fold among workers most heavily  exposed  to  arsenic



trioxide (1).  This finding was substantiated  by  other studies of copper



smelter workers in the U.S. (2,3),  Japan  (4),  and Sweden (5).  Although



many suspect chemicals are present  in the smelter environment, inorganic



arsenic has been implicated as the  respiratory carcinogen.  This is also



the case in other industries, including the  manufacturing of arsenical



pesticides (6,7), where exposures to  inorganic arsenic are  relatively



heavy.  The consistent epidemiologic  evidence  linking  occupational



arsenic exposure to lung cancer has been  sufficient to label arsenic a



carcinogen, even though the agent has not induced tumors in laboratory



animals.



     In the 1970's environmental  measures of stack  emissions from copper



smelters indicated high airborne  levels of a number of pollutants including



arsenic (8).   Substantial  amounts of  inorganic arsenic were then detected



in the soil and air near a copper smelter, and in neighborhood families



the levels of urinary  arsenic were  as high as  in  smelter workers (9).



The possibility that smelter emissions  into  the general  community might



pose a cancer risk was raised by  the  significantly  increased mortality



from lung cancer in male and female residents  of  counties with copper,



lead, or zinc smelters and refiners (10). Inorganic  arsenic is often a
                                          56

-------
component of these ores, more so for copper ores,  and during processing
it is released as an airborne or solid by-product.  It was felt that
work exposures alone would not completely account  for the large excess
mortality in male residents or the increased risk  in females living in
the communities.
     To determine whether air pollutants such as arsenic from nonferrous
smelters may contribute to the risk of respiratory cancer in surrounding
communities, and to evaluate the confounding or modifying effects of
cigarette smoking and work exposures, we planned a case-control interview
study of lung cancer in areas of the U.S. where the smelters were located.
The original study design called for parallel studies near several  non-
ferrous smelters (copper, lead, zinc) around the country.  This was
scaled down when the State of Montana initiated with the EPA an interview
study centering about two copper smelters, and CDC-NIOSH developed
related projects with respect to lead smelters. Therefore we decided to
focus on a case-control study of lung cancer in a  tri-county area of
eastern Pennsylvania where a zinc smelter is located.

Methods and data collection
     Since the survival rate of lung cancer is relatively short and
there were no cancer registries or other means of  rapidly identifying
all newly diagnosed cancers, we selected cases and controls from a
computer listing of death certificates supplied by the state of Pennsylvania,
Death certificates were drawn for Northampton and  Lehigh county residents
who died of lung cancer during the years 1976-1977 and Carbon county

-------
residents who died of lung cancer during the years 1974-77.   A total  of



447 lung cancer cases and an equal  number of controls who died of other



causes (excluding lung diseases and suicide) were identified.  Medical



records were sought on each lung cancer patient for further details on



the disease, including the method of diagnosis and histologic type.



Field operations were conducted through a support service contract with



Lehigh University.



     Interviews were conducted with the next-of-kin of cancer cases and



controls, using a standardized questionnaire that solicited information



on smoking habits, occupational and residential histories, associated



medical conditions, and family history of cancer.  To date,  interviews



have been completed on 430 lung cancer cases and 426 controls.  This



represents a remarkably high response rate of 96%.  The medical  records



were available on 389 cases, with pathologic confirmation of primary



lung cancer  in 89%.  The interview data have been entered into computer



readable form and analysis will begin shortly.  The residences of the



cases and controls are being plotted on a grid map.  For each individual



it should be possible to determine residential proximity to the smelter's



two stacks,  and to estimate exposures to several pollutants including



arsenic based on data from prior environmental surveys conducted primarily



by EPA.







Summary



     A case-control study of lung cancer is underway in a tri-county



area of eastern Pennsylvania in the vicinity of a zinc smelter.  The



next-of-kin  of approximately 430 patients who died of lung cancer, and

-------
426 controls, have been interviewed.  From these data we hope to clarify
the role of occupational and neighborhood exposures to smelter pollutants
as risk factors in lung cancer.

-------
                              References







1.   Lee,  A.M.  and Fraumeni,  J.F.  Jr.:  Arsenic and respiratory cancer



     in man:   An occupational  study.   J.  Nat.  Cancer  Inst. 42:1045-1052,



     1969.








2.   Rencher,  A.C., Carter, M.W.,  and  McKee,  D.W.:  A retrospective



     epidemiological  study of mortality at  a  large western copper smelter.



     J. Occup.  Med. 19:754-758,  1977.







3.   Pinto, S.S., Henderson,  V., and Enterline, P.E.:  Mortality experience



     of arsenic-exposed  workers.   Arch. Environ. Health 33:325-338,



     1978.







4.   Kuratsune, M., Tokudome,  S.,  Shirakusa,  T., et al :  Occupational



     lung cancer among copper smelters.   Int.  J. Cancer 13:552-558,



     1974.







5.   Axelson,  0., Dahlgren, E.,  Jansson,  C.D., and Rehnlund, S.O.:



     Arsenic  exposure and  mortality:   A case  referrent study from a



     Swedish  copper smelter.   Br.  J. Ind. Med. 35:8-15, 1978.







6.   Mabuchi,  K., Lilienfeld,  A.M., and Snell, L.M.:  Cancer and occupational



     exposure  to arsenic:   A  study of  pesticide workers.  Prey. Medicine



     9:51-77,  1980.
                                            60

-------
7.   U.S. DHEW.  NIOSH.  Criteria for a  Recommended  Standard...Occupational



     exposure to Inorganic Arsenic.   New Criteria  -  1975.   HEW  Publ. 75-



     149.  Washington, D.C.:   U.S.  Government Printing  Office,  1975.








8.   Ott, M.G., Holder, B.B., and Gordon, H.L.:   Respiratory  cancer and



     occupational exposure to arsenicals.  Arch  Environ Health  29:250-



     255, 1974.








9.   Mil ham, S. Jr., and Strong, T.:  Human arsenic  exposure  in relation



     to a copper smelter.  Environ.  Research 7:176-182, 1974.







10.  Blot, W.J., and Fraumeni, J.F.  Jr.:  Arsenical  air pollution and



     lung cancer.  Lancet 142-146, 1975.
                                            61

-------
                                  Discussion
Dr. Kraybill, NCI:   Are you going to be measuring  the  ambient levels of these
cations or elements in air, water, diet, et cetera?

Dr. Blot,  NCI:   Yes.  One of the reasons for  picking the Pennsylvania area was
because CDC, together with EPA, conducted an environmental survey several years
ago  in the  environs of the Palmerton zinc smelter.  There  are soil, air  and
household dust measurements available  that we hope  to be able to correlate with
the information that we have on residences from the interviews.

-------
                              General Discussion
Dr. Cortesi, EPA:  If anybody wants to throw out a good subject for discussion, I
think now is the time.

I would like to throw a subject  out for people to consider.  It is a very parochial
point of view.  EPA is in the number picking business.  For epidemiology to be  of
much use  to us, we need a little help in picking numbers, and "go/no go" is not it. I
would like to hear any discussion on any subject, but on this I would like to say that
in regard  to epidemiology, where exposure is quesitonable or non-existent, it is  of
limited use to us.  Epidemiological  studies  where you cannot reduce the relative
risk, that  you feel quite certain  that you could have detected, is not apt to be very
useful  to  us if we  get  a  negative  result.   I think the recent studies on bladder
cancer and saccharin,  the three studies that have come out, are a good example.
As you know,  a couple of investigators talked about detecting a relative risk  of
1.15. Another group , Wynder and coworkers, indicated a risk of 2.5. Relative risk
value,  being the smallest that could be detected for a lot  of EPA type problems,
cannot be constituted as an epidemiological study which tells us we do not have  to
worry  about something when there is a  widespread pollutant with very large
numbers of people exposed.

Who else would like to make some comments?

Dr. Keefer, NCI:  I have a question on a totally different subject. I wanted to ask
the people involved with the HCB exposures how a thing like that could happen.  It
seems  that for  a large number of years a large amount of material was getting into
the human food supply.  More recently, Dr. Kraybill indicated the  same kind  of
thing happened in the United States in cattle rather than people. I  wonder if you
could identify the factors which contributed to this disaster  and comment on how
we can prevent that kind  of  thing  from  happening again somewhere else in the
world.

Dr. Morris, EPA: Actually, you  have two situations in terms of the exposure.  The
Turkish situation was  one in  which there  were starvation  conditions.   You are
talking about  an area of  the  country which is  very  poor.   The government does
subsidize them a great deal.  It actually provides many of them the seed grain.  In
that case, when we had that situation, the  people who received the grain had the
choice of  eating today and you do not  worry about what the seed grain is going  to
do  on  your land a year  from now.  Even more  importantly, the  extent  of
information of the health and environmental effects of HCB were very limited, and
for some effects, totally non-existent. So the people in this time of austerity did
in fact substitute the grain, because they did not have other food sources.  At the
time, they were unaware that there was going to be a problem associated with  it.
Unfortunately,  the  communication at that  time,  compared with now, was a great
deal different.  A lot of these people up in the hill country of south-eastern Turkey
can be lost and no one  is going to get too excited, because no one knows what it is
about.  They died of natural  causes  and  so  on.  But it  certainly came to the
attention  of the Turkish  government in the late  1950's and  that  is when they
obviously put a stop to it and tried to do something about the situation.

In terms  of  the exposure  in  the  Louisiana area,  that  was  a result  of  HCB
production.  They  were actually going to  the  dump with the HCB as  a manu-
                                           63

-------
factoring by-product.  A lot of our data comes from situations in which HCB was
spilled from open trucks along the roadside on the way to the dump.  The exposures
were different  in that sense because  it got in the ground and in the water supply.
These exposures were different but the end effects may not be different.  That  is
what we are looking at.

Dr. O'Conor, NCI: Could you identify some of the contaminants that you suspect?

Dr. Morris, EPA: Certianly one of them is  pentachlorophenol, which may well be a
contaminant along with some of the other chlorinated benzenes.  We have evidence
that these compounds themselves are problems.  I think those  particular contami-
nants have  a fairly extensive track record. That  will always continue to plague
EPA as we proceed through the rulemaking on various  existing chemicals. We do
have concerns about the effects of the contaminants of  the chemicals which we are
asking to  be tested.  This is encountered  in the bioassay program, too.  That  is
certainly the case with HCB.

We are doing some very interesting studies, which are  very  preliminary at  this
time, with the HCB that is now available to us.  That is, we are trying to look at a
series  of purifications and  then looking at  the  acute and subchronic toxicities to
see if there are differences from a strictly toxic point of view.  I do not have the
data on that activity yet, but I would  hope that I can tell you what we found in that
area in our next progress report.

Dr. Cortesi, EPA:  Before  you go away, let me ask you another question.  Don't
chlorinated  dioxins and chlorinated dibenzofurans always show up in these  sorts of
chemicals?

Dr. Morris,  EPA:  Well, I am reluctant to  say "always."  But oftentimes when you
have the chlorination process there is the  possibility of getting other  chlorinated
by-products.  It is true that we  have stopped  making HCB, but HCB and other
chlorinated  hydrocarbons have been reported in  drinking water following chlorina-
tion treatment.  So that continues to plague us as well.

Dr. Bull, EPA:  I have never heard of HCB being  a by-product of chlorination. Can
you expand on that?

Dr. Morris,  EPA:  Yes. I will provide you with my references on the presence of
HCB in drinking  water.  The presence of  HCB  in chlorinated  drinking water  was
reported  in  an  earlier EPA  publication.   The  extent  to  which  chlorination
contributes  to  the  chlorination of organic  compounds  is certainly  not clear.  In
fact,  you  might  check with Diane Courtney down at  RTP.  Apparently, she has
written a review paper on HCB which  you might look at.

Dr. Bull, EPA:  No, I meant as  a HCB chlorination product of disinfection. That  is
a little harder to see.

Dr. Morris,  EPA: I will be pleased to find  you  the reference and perhaps you can
clarify the report for me.  I would appreciate your thoughts on the matter.

Dr. Hoover, NCI:  I  would like to get back to your  original question which you
started the discussion with. I think in making a  decision about whether a chemical
is  a carcinogen or not or  whether it is a big or  a little carcinogen, we  cannot  rely
                                        6-1

-------
on any  one specific  methodology.  We cannot rely on chemical  structure.  We
cannot rely on the bioassay. We cannot rely on epidemiology solely.  I think we
need to make a synthesis of the observations. The saccharin issue is a good case in
point of the laboratory animal experimentation which did not identify saccharin as
a carcinogen at the levels at which it is usually consumed by the American public.
The problem of low level risk  is not unique to the epidemiologists. It is just as
much of a problem for the laboratory animal person.

The  laboratory animal experimenters  have  the  luxury of  being able to  give
enormous doses in order to  produce effects that then can make some assumptions
about extrapolation to a low dose and  some  other  people  can  make assumptions
about interspecies extrapolation.  I think the value of epidemiology is in doing the
same thing, which is finding abnormally  exposed people for whom elevations in risk
may be detectable and for whom you do not have  the  interspecies extrapolation
problem.  You still  do, however, have the dose response extrapolation problem,
with the saccharin issue again being a case in point.

If the overall risk is 1.04, 1.06,  1.1, or some four to eleven percent excess, I do not
think I  or anybody else who does these kinds of studies  would claim causality
associated with that level of risk. In fact, chance is the least of our worries down
in that zone.  All you have to do is have the groups be different with respect to one
cigarette  a  day and you get that kind of a  difference.  Most of  the concern  about
saccharin from the large bladder cancer study was in the 50 to 60 percent excess
risk for those who took upwards of 320  or more milligrams of saccharin per day as
an average, which was the observation and which was some cause for concern.

So, I think that probably when the epidemiologist  gets through looking  at his high
dose people, and  when the laboratory animal experimenter gets done, probably the
only way we can  determine what is a logical  risk is by trying to integrate the two
results.

Dr. Cortesi, EPA:   In reference to Dr. Hoover's remarks, I did not mean  to  be
critical of epidemiology but to  emphasize that a lot of what we want to do in EPA
is thinking ahead in the design of the experiment  and could you give us some help
on what the dose was?  It makes life an awful lot easier when it comes to telling
people that they have to spend  $500,000 a year to do something they do not want to
do.

Dr. Spirtas, NCI:  I  have  just one comment on your question.  I  think that the
setting of standards is probably going to be done in one of two ways, either by the
way that  it is being done now, by formula  or picking a magic number, or else by a
panel of experts.  I believe that the present tendency in regulations is to try to
have this picking done by  the government agencies and that is causing a  lot of
controversy. There is room to  think about  having independent panels of experts or
relying on some sort of consensus standard. This may evolve eventually. But there
are alternative ways to pick standards besides having  a regulatory agency pick
them.

Dr. Cortesi, EPA:  That is  a very good point.  Even if  you  know everything  about
the health effects of a substance, the picking of a standard is a political process
and I firmly believe that  it should be a political process.  I do  not think that you
ought to have a panel of experts do it. But I  do  think that this is something that is
well worth a lot of consideration because it is at the core of the problem. There is
                                            6->
                                           O

-------
misapprehension by a lot of the public that when you set a health based standard
that what you know about  the health should or should  not be, depending on your
point of view, the predominant effect on where you set the level.

-------
 FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:
     PROGRESS ON JOINT ENVIRONMENTAL AND
         OCCUPATIONAL CANCER STUDIES
              Tuesday Morning, May 6
EPIDEMIOLOGICAL/STATISTICAL SESSION (CONTINUED)

              SESSION CHAIRPERSON
                Dr. George Burton
              National Cancer Institute
                           67

-------
                INDUSTRIAL EMISSIONS AND CANCER INCIDENCE
                         IN CONTRA COSTA COUNTY:
                  PROGRESS ON THE EPIDEMIOLOGICAL STUDY
                  U.  S.  Environmental Protection Agency
                          Grant No.  R806396010
EPA Project Officer:
Principal Investigator:
Project Director:
Wilson B. Riggan, Ph.D. ,
Epidemiology Branch, Population Studies Division
Health Effects Research Laboratory
Research Triangle Park, North Carolina  27711

Donald E. Austin, M.D., M.P.H., Chief
Resources for Cancer Epidemiology
State of California
Department of Health Services
1450 Broadway
Oakland, California  94612

William Mandel,  M.D.
Resources for Cancer Epidemiology
State of California
Department of Health Services
1450 Broadway
Oakland, California  94612
Presented at the First NCI/EPA/NIOSH Collaborative Workshop:
on Joint Environmental and Occupational  Cancer Studies

                  Sheraton/Potomac,  Rockville, Maryland
                              May 6-8,  1980
                                     Progress

-------
                                ABSTRACT
                INDUSTRIAL EMISSIONS AND CANCER INCIDENCE
                         IN CONTRA COSTA COUNTY:
                  PROGRESS ON THE EPIDEMIOLOGICAL STUDY
     This research effort contains the following tasks:  cancer incidence
analyses, occupational monitoring, case control studies, and industrial
emission analyses.
     Cancer incidence analyses compare the age, sex, race, and site
specific cancer incidence rates in the industrialized areas with the
non-industrialized areas of Contra Costa County.   Analyses for years
1972-1977 are complete and analyses for years 1969-1971 are in progress.
     Occupational monitoring determines whether any labor union or
occupational group has higher incidence rates for any cancer sites than
do other union or occupational groups in Contra Costa County.   Names of
union members, employed persons, and professional groups are being
merged with the cancer incidence files of Resources for Cancer Epidemiology
Section (RCE).  This task has been completed for six (6) occupational
groups totaling over 6,000 members.
     Case control studies attempt to identify environmental factors
associated with cancer incidence in Contra Costa County.  The sample
design and the number and selection of cases and controls depended on
results of the incidence analyses and occupational monitoring.  Selection
of cases, controls, and design of questionnaire are about 75% complete.
Cases and controls or their families will be interviewed to obtain
length of residence, socio-economic status, smoking habits, occupation,
and exposure to other pollutants.

-------
     Industrial  emission analyses include the collection of air samples



and inorganic chemical  analyses to determine the levels of ambient air



pollution by census tract.   The Ames Salmonella test is being used to



test for mutagenicity of industrial emissions.   Collection of air samples



and chemical analyses of inorganic fraction are complete.



Acknowledgment:



     This research is supported by EPA Grant #R806396010 with the State



of California, Department of Health Services, Resources for Cancer



Epidemiology Section.
                                        f*H /'
                                        70

-------
                INDUSTRIAL EMISSIONS AND CANCER INCIDENCE
                         IN CONTRA COSTA COUNTY:
                  PROGRESS ON THE EPIDEMIOLOGICAL STUDY
INTRODUCTION
     Contra Costa County continues to be the major industrialized county
in the San Francisco Bay Area.   This industrial  complex,  in addition  to
five major petroleum refineries and many petrochemical  plants,  included
Kaiser's major shipbuilding center during World War II.   At present,  70
percent of the shipping going through the Golden Gate Bridge either
enters or leaves ports in Contra Costa County.   Sixty-eight percent of
the total stationary air pollution emissions in the San Francisco Bay
Area originates in Contra Costa County.
     Considerable speculation exists on how much air pollution  contributes
to cancer mortality in urban areas.  Previous studies have linked air
pollution to four anatomic sites—lung,  stomach, prostate, and  lymphoma.
Mortality study of U. S. counties with petroleum industries found other
sites with greater than expected frequencies.  These results raise the
question of how much of the excess cancer mortality is due to occupational
exposures, to ambient air pollution exposure, or to other relevant
variables.
     Investigators encounter several problems in trying to identify or
quantitate the contribution of each variable to the greater-than-expected
cancer mortality.  These problems are:
     o    Small size of study population;
     o    Latent period for cancer development;
                                        71

-------
     o    Population density;
     o    Smoking effect; and
     o    Other associated variables rwt -measured or included in the
          study.
     The monograph on cancer mortality by counties (1950-1969) by Mason
and McKay  revealed high site specific cancer mortality rates in industrialized
counties with petroleum refineries and petroleum chemical plants.
     As part of the NCI/EPA collaborative research program, EPA completed
the review process and funded a grant request from Contra Costa County
Department of Health.  This coincided with the publication of Blot's
                               2
report in Science.  Blot et al.   compared cancer mortality rates in 39
counties with petroleum refineries employing at least 100 people with
117 control counties.  Both men and women had a significantly higher
lung cancer mortality rate in the petroleum refinery counties than in
the control counties.  High lung cancer mortality rates among women
suggests ambient or personal (smoking or in-door) rather than occupational
exposure.  Men also had a significantly higher cancer mortality rate for
several other site specific cancers.  This publicity created many community
pressures to take action in what was locally being called "cancer county."
Industry countered by saying that San Francisco and Alameda counties had
cancer rates as high as Contra Costa County.  This issue became an
emotional and a political battle between various adversaries which
prevented the Contra Costa Department of Health from getting started on
the study.
     As a result of this impasse, the funds were transferred to California
State Department of Health Services, Resources for Cancer Epidemiology,
where the California Tumor Registry is located.

-------
     California State legislature and OSHA funded Dr. Austin's group to
extend the cancer study to four additional counties (Figure 1).  These
are:  San Francisco, Alameda, San Mateo, and Marin.  San Francisco and
Alameda counties have cancer mortality rates comparable to Contra Costa
County.
     Bear with me a moment please; we were not out of the woods yet.
During the summer of 1978, the voters of California passed "Proposition
13", which resulted in the Executive Branch of the State Government
freezing all personnel actions, including filling of positions.  Thanks
to the effort and support of the Department of Health Services, the
effort and support of Mr. Paul DeFalco, Jr., Administrator, Region IX
and his staff, and the picketing of Dr. Austin's office and the Health
Department by various environmental groups, Dr. Austin was given permission
in late June to complete the staffing of the project.  This was completed
by August 1, 1979.
     The northern part of Contra Costa County is heavily industrialized
with five major petroleum refineries, many petrochemical plants, and was
the home of Kaiser shipbuilding during World War II.  Many complaints
have been raised by individuals and groups about the air quality in
certain sections of Contra Costa County.
     Population census was collected and data published at the census
tract  level for both 1970 and 1975 in Contra Costa County.  Cancer
incidence data has been collected in Contra Costa County since 1969.
The Third National Cancer Survey collected incidence data between 1969
and 1971.   The California Tumor Registry collected incidence data from
1972 onwards.

-------
     This study contains four major tasks.   These are:
     o    Cancer incidence analyses;
     o    Occupational  monitoring;
     o    Case control  studies;
     o    Industrial  emissions analyses
Cancer Incidence Analyses:
     Preliminary analyses revealed a dramatic difference in lung cancer
rates for males between the industrialized and non-industrialized parts
of Contra Costa County for years 1972-1975.   Years 1967-1971 are being
added.  The investigators have gone back to the 1970 and 1975 census
data to validate the classification of each census tract as industrial
or non-industrial.   The analysis is being rerun for the extended period
on the reclassified (validated)  (industrial-non-industrial) census
tracts.
     The size of the difference  between lung cancer incidence rates in
the two parts of the county gives the highest priority for further
investigation to this cancer site.   Current funding for this grant
includes the case-control study for lung cancer incidence.
Occupational Monitoring:
     The purpose of occupational monitoring is to determine whether any
labor union or occupational group has a higher incidence rate for any
cancer site than other union or occupational groups in Contra Costa
County.  Names of union members, employed persons, and professional
groups are being merged with the cancer incidence files of Resources for
Cancer Epidemiology Section (RCE).   RCE has collected identifying infor-
mation on various cohort groups (Table 3).  Matching has been completed
for these.  RCE has contacted many additional groups (Table 4).

-------
     Monitoring of occupational groups is an attempt to identify working
groups with special cancer risks who, by their residential  patterns,  may
affect the observed cancer rates of the population in a specific geographic
area.
Case-Control Study:
     The past two months have been spent in planning this study.  A
questionnaire has been circulated for comment and is being prepared for
printing.  The study is scheduled to start June 2, 1980.
     Approximately 150 cases (lung cancer, men and women, white and
black, ages 20-74) and 300 controls will be stratified and matched by
age, sex, and race.  The primary intent of the case-control study is  to
identify the major factors associated with the difference in lung cancer
rates between the industrialized and non-industrial!zed parts of Contra
Costa County.  Some of the factors included are smoking history, occupation
history, and geographic location of residence history.
     This phase of the study examines the problems of latent period,
occupational and environmental exposure history, migration history,
smoking  history, and socioeconomic class.  The investigator may be able
to get some suggestion as to the effect of family income below the
poverty  line since 15% of the workforce in Contra Costa County falls
into this socio-economic class (Tables 1 and 2).
Industrial Emission Analyses:
     Air emission collection sites were established in November 1978.
The 15 station network contains 5 permanent (part of the Bay Area Air
Management District) stations and 10 temporary stations (Figure 2).  Air
samples  collected from November 1978 through October 1978 were analyzed
                                        7f>

-------
for total  suspended participates, inorganic substances (lead, nitrates,



sulfates),  benzene-soluble organics, polycyclic aromatic hydrocarbons,



and mutagenicity using the Ames test.   Gas and meteorological data were



collected duirng this one-year period.



     Data are now being used in a modeling technique to characterize



Contra Costa County census tracts by estimating a value for each of the



air pollutants measured.



ACCOMPLISHMENTS



     The first major accomplishment:  We have a research effort underway



after being caught between the various industrial, environmental, and



political adversaries for two years.  This project is staffed with a



highly motivated and qualified staff who are interested in obtaining



sound, valid scientific results.



     Established the following Technical Advisory Committee:



     o    Dr. James Sandberg, Meterologist;



     o    Dr. Alice Whitemore, Biostatistician; and



     o    Dr. Warren Winklestein, Jr., Epidemiologist.



     Many staff tasks and activities have been completed during the past



six months, for example, review of  literature (Attachment A); meetings



with Technical Advisory Committee, and the Citizen's Liaison Committee;



meeting with persons, agencies and organizations in the effort to develop



cooperative program and project relationships; and preparation of numerous



in-house reports.



     Drs. Austin and Mandel presented a paper at the 72nd Annual Meeting



of the American Institute of Chemical Engineers, November 28, 1979, San

-------
                      o
Francisco, California.   They discussed the role and importance of


population monitoring as a step to cancer prevention in professional


chemists.


     Analyses of air samples have been completed for five standard air


pollutants in samples collected during July 1979 to October 1979.   These


samples are currently being analyzed for chemical carcinogens and for


mutagenic activity, with a scheduled completion date of July 1980.


     The findings and progress in the past six months are described in a


report presented at the second symposium and the application of short-term


bioassays in the fractionation and analysis of complex environmental

                                                        4
mixtures at the Williamsburg, Virginia, March 4-7, 1980.


     Dr. Austin discovered, as a collateral development of the Contra


Costa County and Bay Area cancer incidence study, what appears to be a


sharp increase in incidence of malignant melanoma among employees of the


Lawrence Livermore Laboratory.  More than 18 months' work went into


completing the status report,  which is being reanalyzed by a panel of


experts  assembled by the Department of Energy.  Nineteen cases of malignant


melanoma occurred among laboratory employees between 1972 and 1977


including three deaths, among them the Laboratory Director.  Since then


six more cases have been reported, three in 1980.

-------
                               REFERENCES



1.    Mason, T.  J.  and F. W. McKay, U.S. Cancer Mortality by County:



     1950-69 (Government Printing Office, Washington, D.C., 1973).



2.    Blot, W. J.,  L. A. Brinton, J. F. Fraumeni, Jr., and B. J. Stone,



     Cancer Mortality in U.S. Counties with Petroleum Industries:



     Science, 198:51-53, 1977.



3.    Austin, D. F. and W. Mandel, Population Monitoring:  A Step to



     Cancer Prevention in Professional Chemists.  Presented at the 72nd



     Annual Meeting, American Institute of Chemical Engineers, November 8,



     1979, San Francisco, California.



4.    Flessel, C. P. et al., The Integration of the Ames Bioassay and



     Chemical Analyses in an Epidemiological Cancer Incidence Study.



     In:   Proceedings of the Second Symposium on the Application of



     Short-Term Bioassays in the Fractionation and Analysis of Complex



     Environmental Mixtures, Williamsburg, Virginia, March 4-7, 1980, in



     process.



5.    Austin, D. F., A Study of Cancer  Incidence in Lawrence Livermore



     Laboratory Employees, California  Department of Health Services,



     Report No. 7, April 17, 1980.

-------
Figure 1. Bay area environmental cancer study counties.

-------
                     SAN
                    PABLO
                     BAY
                                                 PITTSBURG'
                                                        ANTIOCH
                                                 •CONCORD
MARINCO
                           RODEO-PINOLE
                               MARTINEZ
                                                  CONCORD -TREAT
                        RICHMOND
                         1ELCERRITO
                                                                       BRENTWOODj
                                            LAFAYETTE
                   SAN
                FRANSISCO
                   CO.
                                                         LIVERMORE
                                                                       PERMANENT
                                                                       TEMPORARY
Figure 2. Location of sampling stations in Contra Costa county, CA.

-------
                                 Table  1
                           CONTRA COSTA COUNTY
                  CIVILIAN LABOR FORCE  BY  RACE  AND  SEX
                             Projected, 1980
                         Contra Costa County
                         (excluding Richmond)
Richmond
Total
Total
Male
Female
Total white
Male
Female
Total black
Male
Female
Total others
Male
Female
(Spanish-American)**
Male
Female
288,500
170,370
118,130
273,010
162,120
110,890
7,880
4,090
3,790
7,610
4,160
3,450
(23,810)
(14,490)
(9,330)
36,000
18,790
17,210
14,570
8,100
6,470
19,310
9,610
9,700
2,110
1,080
1,030
(3,860)
(2,100)
(1,760)
324,500
189,160
135,340
287,580
170,220
117,360
27,190
13,700
13,490
9,720
5,240
4,480
(27,680)
(16,590)
(11,090)
*6,810 are age 65+ (2.1%)

**A1ready counted


Source:   State of California,  Employment Development Department, May  1, 1979
                                      81

-------
                                 Table 2
                           CONTRA COSTA COUNTY
                    PERSONS BELOW POVERTY LEVEL, 1980

White
Black
Other
Total
Contra Costa County
(excluding Richmond)
31,970
4,890
1,050
37,910
Richmond
2,250
8,110
390
10,750
Total
34,220
13,000
1,440
48,660
Source:   State of California, Employment Development Department, May 1979,
         Projected 1980.

-------
                                 Table 3
                         STATUS OF PAST COHORTS

1.
2.
3.
4.
5.
6.
No.
Unions updated by EHA*
Unions to be updated**
Asbestos workers
Dry cleaners
Union members
Fabric care licensees
Cosmetologists
Livermore Radiation Lab
Total
groups
5
12
1
1
1
1
21
No. locals
14
60
1
1
-
-
76
Approximate
No. persons
6,000
24,000
250
4,000
20,500
6,000
60,750
*1.  Bakers
 2.  Painters
 3.  Plasterers
 4.  Plumbers & Steamfitters
 5.  Roofers
**1.  Cement Masons/Plasters
  2.  Cooks/bartenders
  3.  Firefighters
  4.  Hod Carriers
  5.  Industrial Iron Workers
  6.  Lathers & Plasterers
  7.  Laborers
  8.  Oil, Chemical and Atomic Workers
  9.  Painters
 10.  Paint Makers
 11.  Plumbers
 12.  Steamfitters
                                         8 a

-------
                                 Table 4
                    POSSIBLE NEW COHORTS IN SF-0 SMSA
Group                                           Approximate No.  persons


State of California                                        18,000

Operating Engineers                                         6,900

Teamsters                                                   6,500

Contra Costa County                                         6,000

State University                                            4,200

Apprentices                                                 3,900

American Chemical Society                                   2,000

Oil, Chemical, Atomic Workers (1-1978)                      1,600

Int'l Association Flight Attendants                         1,000

City of Hayward                                               800

California Veterinarian Association                           450

Oil, Chemical, Atomic Workers (1-326)                         350

   Total                                                   51,700
                                      84

-------
                              ATTACHMENT A

  SEVEN SELECTED MAJOR REFERENCE SOURCES ON OCCUPATIONAL CARCINOGENESIS
                        (Listed Chronologically)


I.    Compilation of some 28,500 citations, including approximatley 5,800
     reprints, published during 1960 and 1975.

II.   Approximately 1,450 abstract? on occupational and environmental
     carcinogenic hazards (1969 to 19743

III.  Proceedings of 1976 conference ofr occupational carcinogenesis.

IV.   Listing of 233 references published from 1963 to 1974.

V.    Listing of 584 abstracts of current cancer research on occupational
     and environmental carcinogenesis.

VI.   Abstracts on cancer research epidemiology with many citations on
     occupations and cancer.

VII.  Listing of 148 references irr a summary article on occupationally-related
     carcinogens.

-------
I.    U.  S.  Department of Labor.   Citations on Occupations and Cancer.
     1960 to 1975.

     Compilations of 28,498 citations (including approximately 5,784
     reprints) on occupational cancer published during 1960 to 1975.
     The citations were gathered from secondary sources (e.g., Index
     Medicus and Excerpta Medica).   The citations have been classified
     by subject headings.  Selected subject bibliographies have been
     prepared.  This reference source was prepared under contract with
     the Occupational Cancer Data Bank of the George Washington University.

II.   Abstracts and Indexes to Selected Literature on Occupational and
     Environmental Carcinogenic Hazards.   Prepared by the Franklin
     Institute Research Laboratories, July 10, 1975 (338 pp).

     This compilation of approximately 1,450 abstracts was prepared for
     a conference on cancer registries and occupational cancers held in
     1975.   The abstracts were obtained from Volumes 7 through 11 (1969
     to 1973) and the first two issues of Volume 12 (1974) of Carcinogenesis
     Abstracts.  Where abstracts were not present, they were taken from
     Excerpta Medica, Biological Abstracts, Chemical Abstracts or from
     the article.  There are three extensive indexes; namely,  agents,
     sites, and agents-site-tumors.

     The pages are not numbered.  The references are listed by the
     volume and reference number in Carcinogenesis Abstracts.   The
     publication brings together many abstracts dealing with occupational
     and environmental carcinogenic hazards.

III. Occupational Carcinogenesis, Vol. 271, Annals of the New York
     Academy of Sciences, 1976 (560 pp).

     This monograph contains the proceedings of a conference held on May
     28, 1976.  The many articles are grouped into nine sections with
     references following each article.  These references serve as a
     comprehensive and varied listing of citations in the field of
     occupational Carcinogenesis.

IV.   Decoufle, P:  A Retrospective Survey of Cancer in Relation to
     Occupation.  DHEW (NIOSH) Publication No. 77-178, 1977 (215 pp).

     This study, performed at the Roswell Park Memorial Institute contains
     233 references from the American and European literature during the
     years 1963 and 1974.  The references cover occupations, industries,
     hazards, and cancers.

-------
V.   Special Listing.  Current Cancer Research on Occupational  and
     Environmental Carcinogenesis, July 18, 1979.   U.  S.  Department of
     Health, Education and Welfare.   Public Health Service.   National
     Institutes of Health.  National Cancer Institute (85 pp).

     This listing contains 584 abstracts of current research projects  in
     seven major categories; namely, asbestos; metallic salts or oxides;
     organic chemicals, specific occupations; pollution and other environmental
     factors; detection and measurement of environmental  carcinogens;
     and other studies on environmental and occupational  carcinogenesis.

     The only index consists of an alphabetical listing of investigators.

VI.  Directory of On-Going Research in Cancer Epidemiology, IARC Publications
     No. 28, World Health Organization, International Agency Research  on
     Cancer, Lyon, 1979 (672 pp).

VII. Schottenfeld, DM; Haas, JF:  Carcinogens in the Work Place.   CA-  A
     Cancer Journal for Physicians'29:144-173, May/June 1979.

     This summary article on occupationally-related carcinogens contains
     148 references.  These references covery many of the important
     current topics in the field and can serve as an excellent source
     for persons interested in the subject matter.
 NO DISCUSSION FOLLOWING THIS PAPER
                                         87

-------
       An Etiologic  Study of Respiratory  Cancer  in  Coastal Texas
                      T.J.  Mason  and  L.W.  Pickle
      Environmental  Epidemiology  Branch,  National  Cancer  Institute

     Since the turn  of the century there  has  been  a rapidly  rising
incidence of lung cancer in the United  States and  other countries.
Recent mappings of U.S.  cancer mortality  statistics on a  county level
(Mason, et al, 1975, 1976)  revealed elevated  lung  cancer  morta-lity  for
white males in the northeast and  near large metropolitan  areas, with the
highest rates clustered in contiguous counties along the  Gulf  of  Mexico
and the southeast Atlantic coast.  Rates  among white women and among
blacks were also high in urban areas  but  did  not show the southern
coastal excess.
     Although smoking accounts for a  large fraction of lung  cancer,
other environmental  determinants  are  involved to some extent,  and may
act synergistically with tobacco  smoke  as in  the case of  asbestos workers
and uranium miners (Selikoff and  Hammond,  1975).  The precise  contribution
of occupational factors to the overall  risk of lung cancer in  the U.S.
is uncertain, but the hazards may be  more conspicuous and easier  to
identify in high risk areas than  elsewhere.   As  an  initial step toward
explaining the geographic variation of  lung cancer  in the United  States,
a series of correlation studies was conducted (Blot, et al,  1976, 1977,
1979; Hoover and Fraumeni, 1975)  linking  cancer  mortality rates with
demographic and environmental data available  at  the county level.   These
studies suggested that lung cancer mortality  rates  were elevated  in
counties where the petrochemical, paper manufacturing, smelting,  and
shipbuilding industries are concentrated. Mortality rates for cancer  of

-------
the larynx, particularly for white males, were also elevated in counties
where the shipbuilding industry had been concentrated during World
War II. These excesses at the county level suggest that the environmental
hazards associated with certain occupational exposures may have spread
beyond the workplace.
     More definitive case-control epidemiologic studies have recently
been conducted in several high-risk areas of the country, particularly
along the seacoast.  In these areas, lung cancer patients and controls,
or their next-of-kin, were interviewed to determine their lifetime
histories of occupation, residence, tobacco consumption, and other
environmental exposures.  Interview studies of male residents of several
counties in Georgia and coastal Virginia have shown an approximate 70%
elevation of respiratory cancer risk among men employed in the shipbuilding
industry during World War II (Blot, et al, 1978, 1980). Similar studies
are now being conducted in Jacksonville, Florida, and in southern Louisiana.
     Death certificate studies in Louisiana and Texas have suggested
excess mortality from various cancers among former employees of the oil
refining and shipbuilding industries (Gottlieb, et al, 1979; Thomas et
al, 1980). Elevated lung cancer mortality was also seen among residents
of Louisiana towns where the petroleum industry was a major employer,
again raising the possibility that the exposures had spread beyond the
workplace.
     In order to pursue leads generated by these and other studies we
have contracted with the University of Texas School of Public Health to
conduct a case-control interview study of respiratory cancer in the
Texas Gulf Coast area, a major shipbuilding center during World War II
and site of a major concentration of petrochemical plants today.

-------
Interviews will  be conducted with approximately 2000 respiratory cancer
patients and 2000 controls,  or their next-of-kin,  among white residents
of Jefferson, Orange, Chambers, Galveston,  Harris, and Brazoria counties
in Texas.  Cases are all  white residents of the study area  between the
ages of 30 and 79 who were diagnosed as having lung cancer  between
July 1, 1976, and June 30, 1980, or laryngeal  cancer (males only)
between July 1,  1975 and  June 30, 1980.  Male  lung cancer cases will  not
be ascertained from Harris county (Houston) so that the study will
include approximately equal  numbers of male and female subjects.  Principal
sources for case ascertainment will be the  admissions/discharge lists
and medical records of all adult medical/surgical  and cancer hospitals
in the study area.  In addition, pathology  reports and state death
certificate tapes will be utilized.
     Appropriate controls matched by sex, race, age, vital  status, and
county of residence (Harris vs. other) will be selected from the general
population.  Probable sources of control selection will be  through
drivers' license records  and Medicare records  for  those aged 65 and
over.  Potential comparison subjects with chronic  respiratory diseases
or other smoking-related  diseases will be excluded.  Information will be
gathered by face-to-face  interviews on lifetime histories of residence,
occupation, tobacco and alcohol consumption, medical history, and cancer
history among the subject's family.  A brief dietary supplement will  be
used to determine the subject's level of micronutrient intake, including
vitamin A.
     Most of the preliminary work has been  completed for this project.
Hospitals in the area have been contacted and the staff has been assembled.
Interviewer training will be conducted in July, with the actual field
work commencing by August.  Interviewing is expected to continue for one
year.  Preliminary results should be available late in 1981.
                                       90

-------
                            Acknowledgement







     This study is being conducted  by  the University of Texas School of



Public Health,  Patricia  Buffi er,  Ph.D.,  Principal  Investigator, under



NCI/EPA contract #N01CP91025.
                                    91

-------
                          REFERENCES

Blot, W. J. and J. F. Fraumeni, Jr. (1976).   Geographic patterns of lung
     cancer:  Industrial  correlations.   Am.  J.  Epidemiol..  103,  539-550.

Blot, W. J., Brinton, L.  A., Fraumeni,  J.  F-  Jr.,  and B.  J. Stone (1977).
     Cancer mortality in  U.S. counties  with  petroleum industries.
     Science, 198, 51-53.

Blot, W. J., Harrington,  J. M., Toledo, A.,  Hoover,  R., Health,  C.  W.,  Jr.
     and J. R. Fraumeni,  Jr. (1978).  Lung cancer  after employment  in
     shipyards during World War II.  N. Enql.  J. Med.,  299, 620-624.

Blot, W. J., Morris, L. E., Stroube, R., Tagnon,  I.,  and  Fraumeni,  J. F.,  Jr.
     (in press).  Lung and laryngeal cancer  following shipya-d employment
     in coastal Virginia.  J. Nat. Cancer Inst.

Blot, W. J. Stone, B. J., Fraumeni, J.  F.  Jr,  and  Morris,  L.  E.  (1979).
     Cancer mortality in  U.S. counties  with  shipyard  industries  during
     World War II.  Environ. Res.. 18_,  281-290.

Gottlieb, M. S., Pickle,  L. W., Blot, W. J.  and  Fraumeni,  J.  F., Jr. (1979).
     Lung cancer in Louisiana:  A death certificate analysis.
     J. Nat. Cancer Inst., 63_, 1131-1138.

Hoover, R. and J. R. Fraumeni, Jr. (1969).  Cancer mortality  in  U.S.
     counties with chemical industries.  Environ.  Res., ^,  196-207.

Mason, T. J., McKay, F. W., Hoover, R., Blot,  W. J. and J.  F.  Fraumeni,  Jr.
     (1975).  Atlas of Cancer Mortality for  U.S. Counties:  1950-1969.
     Department of Health, Education, and Welfare  Publ. No. (NIH) 75-780,
     U.S. Gov- Print Off., Washington,  D.C.

Mason, T. J., McKay, F. W., Hoover, R., Blot,  W. J.,  and  J. F. Fraumeni  Jr.
     (1976).  Atlas of Cancer Mortality Among  U.S. Nonwhites:   1950-1969.
     Department of Health, Education, and Welfare  Publ. No. (NIH) 76-1204
     U.S. Gov- Printing Off., Washington,  D.C,

Selikoff, I. J. and Hammond, E. C. (1975).  Multiple  risk  factors in
     environmental cancer.  In Fraumeni, J.  F-  Jr. (Ed.),  Persons at High
     Risk of Cancer:  An  Approach to Cancer  Etiology  and  Control,
     Academic Press, New York.

Thomas, T. L., Decoufle,  P., and Moure-Eraso,  R.  (1980).   Mortality among
     workers employed in  petroleum refining  and  petrochemical  plants.
     J. Occup. Med., 22^,  97-103.

-------
                              Discussion

Dr. Bellin, EPA:  In terms of history, are you going to question people as
to their occupational history and health and how far back do you go?

Dr. Mason, NCI:  To age 12.  We are taking lifetime occupational histories,
lifetime cigarette smoking, pipe, cigar or any other tobacco products.
Because it is both lung and larynx in Texas, we will also be taking detailed
alcohol consumption histories.  We also take lifetime residential  histories
which get at townships, because there is a growing interest and commitment
on our part to utilize information with regard to specific components of
municipal water.  So we are addressing the issue of when you lived, where
you lived, what kind of water did you drink.  This will tie in also with
the parallel study which we are doing in New Jersey where there is ample
opportunity for the same chemical exposure but with slightly different factors
related to the exposure, not the least of which is climate.

Dr. Saffiotti, NCI:  I was interested in your point about the study related
to vitamin A levels.  I would like to have some further information on how
you are going to approach that.

Dr. Mason, NCI:  Carefully.

Dr. Saffiotti, NCI:  Particularly in this respect, that there are  two possible
types of data that you can try to correlate.  One is" simply vitamin A intake
in the diet.  The other is to try to get some biological marker data, as more
and more information has developed in the last decade on the role  of vitamin A
in mechanisms of carcinogenesis.  That role becomes more and more  complex.
It may well be lihkedto the availability of receptor sites and to  host charac-
teristics more than the total amount of retinoids taken  with the  diet.   What
kind of parameter are you going to relate?

Dr. Mason, NCI:  What we are doing is we are concentrating on something like
28 particular foods, looking at usual portions, looking at methods of pre-
paration, getting at estimates of levels of consumption.  It will  be done
with both cases and controls.  These are general population controls in both
instances.  We feel that taking this abbreviated form on what is going to be
about 3,000 cases, which gives us a total of 6,000 persons, will improve our
understanding with regard to what is happening in this instance.  Then we
are taking the detailed, approximately 45 minute dietary questionnaire,  and
interviewing a sample of both cases and controls in New Jersey.  After that,
I believe, we would then be in the position to consider perhaps some of the
other things.

Right now, we are of the opinion that we can prioritize a series of foodstuffs.
We can get at the vitamin A, vitamin C, etc. issues.  We can do it in identi-
fied places with reasonable understanding of other factors as they relate.
We obviously have to have all the detail on the person with regard to smoking
and occupation and other such factors as they relate to the disease, and then
we have an opportunity to follow it up.  So it may well happen that next
year at this time, when some preliminary things come from this, we could con-
sider some of the subsequent more clinical laboratory types of investigations
of some samples of persons in these places.  That is how it is currently en-
visioned.


                                        93

-------
        ENVIRONMENTAL HEALTH DATA BASE FOR NEW JERSEY


                         Thomas Mason, Ph. D.
                       Field Studies and Statistics
                 Division of Cancer Cause and Prevention
                        National Cancer Institute
                          Bethesda, Maryland


We have an historic interest in the State of New Jersey.   In 1975, when
I presented this map of bladder cancer  among white males  for the first
time to the Society of Epidemiologic Research, the State  Epidemiologist
from New Jersey said, well, I assume we are going to have to follow this
up.  I said, yes, that seemed prudent.   For the past year, we have had a
very large commitment by the Environmental  Epidemiology Branch to the
State Department of Health and now the  State Department of Environmental
Protection in an attempt to try to characterize exposures and to quantify
exposures and to utilize historic information to look for surrogates.

Let me explain.  Last year when we were doing the National Survey of
Environment and Health, which addressed the saccharin issue, we identified
1,257 newly diagnosed bladder cancer cases  in the State of New Jersey,
1/3 of our total case load.  This is over 12 months.  It  is not an in-
significant problem.  In the lung cancer study which I  talked about earlier,
we will identify 1,100 a year in five counties.

Cancer is an interesting problem in the State of New Jersey because you
have an opportunity to look at some very specific exposures in real time.
As we were able to show in the National  Survey of Environment and Health,
we can get results in humans in a comparable time frame for that which is
possible in experimental animals.

Well, the New Jersey Department of Environmental Protection is really the
EPA equivalent in New Jersey.  The EPA has  given over to  them a number, if
you will, of its responsibilities.  They have certified a number of labs.
They have historically attempted to characterize ambient  exposures.  They
are in the business of monitoring discharge permits, and  things such as
that.

In an attempt to better understand and to follow up on some of the things
that we had done when we identified associations between  bladder cancer
and the non-chloroform tri-halomethane complex, we decided to enter into
a contract with the Department of Environmental  Protection in New Jersey
to do the following:  (1)  To define the service areas and average produc-
tion, population served and raw water sources for all major water pur-
veyors; (2) to determine historic and current chlorination practices of the
major purveyors; (3) to identify those dischargers which may be potential
pollution sources to unprotected surface water supplies;   (4) to compile
the results of all historic routine water quality monitoring for all major
purveyors; and  (5) to compile the results of all monitoring for toxic
                                    94

-------
and carcinogenic contaminants conducted by the U.S. EPA in the New Jersey
Program on Environmental Carcinogenesis and Toxic Substances on both raw
surface and ground water supplies and finished drinking water.

A large portion of these data were available in hard copy in filing cabinets
in New Jersey and had not been computerized.  We felt it important to fund
this particular project.  The timing of it got a little messed up 
-------
                                 Discussion

Dr. Bull, EPA:  I am a little curious about what other organic parameters
you are looking at in the drinking water other than THM.

Dr. Mason, NCI:  It is my understanding, and if Ken Cantor were here he could
tell you exactly which ones are being looked at, but I think it is a reasonably
broad spectrum, as far as characteristics.

Dr. Bull, EPA:  Do you mean GC?

Dr. Mason, NCI:  I believe, yes.  That was my understanding going into this,
that this would be the appropriate mechanism to use.  We wanted to get at the
THM's, but we also wanted to get a reasonable representation of what is there.

Dr. Bull, EPA:  The total organic carbon and total organic chlorine.

Dr. Mason, NCI:  Yes.

Unidentified Speaker:  I may have missed this in the presentation, but how
far back does the THM data go?

Dr. Mason, NCI:  The THM data are current.  It is for the past year or so.
What we are doing is we are going back historically.  A number of people
have argued that you could relate some of the metals, if you will.  If you
could look at metals and other such things as are currently in the water and
develop a reasonably predictive association, you could go back historically
and look at chlorination levels and metals, which is all you are going to
have historically.  See, if that were true, could you say something with
regard to potential levels?  If you know parallel to that all of the dis-
charges, which they have, could you then say something? They have an idea
and they have some estimate, although I  will not say exactly with what
precision, but at least they can have an estimate there with regard to
what could have been.  If you recall, what we are trying to do is to
complement our earlier study and afford yet another stratified analysis
of  bladder cancer in New Jersey.

Unidentified Speaker:  Is colon turbidity still factored in?

Dr. Mason, NCI:  I believe so, yes.  That was my understanding.

Dr. Kraybill, NCI:  What water sources are you going to be dealing with?
Is  this going to be ground water or surface water.

Dr. Mason, NCI:  All sources.  So, basically, what we are interested in is
all water sources. There are a number of measurements that have been done
with regard to private wells.  I do not want to get into that, but we are
interested in all of the different ways in which persons have gotten water.
You  can characterize the purveyors as their usual source and then you have

-------
information with regard to dischargers as well  as historic industrial
profiles which get at an estimate of potential  levels of contaminants  and
contaminants of a certain kind.  But it is an attempt to try to characterize
the water as it is being consumed by persons.

Dr. Kraybill, NCI:  How do you epidemiologists  control for the other
variants?

Dr. Mason, NCI:  We already have all the details on the persons.

Dr. Kraybill, NCI:  I am referring to air pollutants and diet.

Dr. Mason, NCI:  We have the history of smoking, occupation and residence
histories on large numbers of cases and controls.   So this is an attempt
now to add an additional component to that, looking at some of the specifics
in the water, because there are some certain hypotheses which have been
suggested and there are some consistent findings.

-------
 FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:
    PROGRESS ON JOINT ENVIRONMENTAL AND
        OCCUPATIONAL CANCER STUDIES
             Tuesday Afternoon, May 6
EPIDEMIOLOGICALSTATISTICAL SESSION (CONTINUED)
             SESSION CHAIRPERSON
               Dr. Kenneth Bridbord
  National Institute for Occupational Safety and Health
                   98

-------
                      IDENTIFICATION OF HIGH RISK
                  OCCUPATIONAL GROUPS USING NOHS/RTECS
                           David H. Pedersen
                Hazard Section, Surveillance Branch, DSHEFS
                          NIOSH - Cincinnati, Ohio


     The National Institute for Occupational Safety and Health (NIOSH) has
developed two data files, the Registry for the Toxic Effects of Chemical Sub-
stances (RTECS) and the National Occupational Hazard Survey (NOHS).   The data
from these two files can be used to assess the relative potential health risk
to occupational and industrial groups arising from workplace chemical exposure.

     In the course of this research effort, algorithms were developed which can
be used to:  (1) rank-order chemicals by their relative potential toxicological
effect, (2) rank-order chemicals by simultaneous consideration of their relative
toxicity and the number'of people exposed to each under industrial conditions,
(3) rank-order industries and occupations by relative potential health risk
due to aggregate chemical exposure.

     The algorithms producing these indices have been designed to accommodate a
wide variety of research interests.  Accordingly, indices of risk can be pro-
duced which emphasize such outcomes as dermatitits or cancer or which consider
only certain types of animal or exposure route data.  The purpose of this pre-
sentation is to describe how the indices produced by this modelling procedure
have potential application in such disparate activities as prioritization of
occupational health research needs, identification of occupations and industries
with previously unrecognized hazardous chemical exposure problems and assistance
in the allocation of health service resources.
                                            9,1

-------
              IDENTIFICATION OF HIGH RISK INDUSTRIAL AND
              OCCUPATIONAL GROUPS USING RTECS/NOHS DATA


The Surveillance Branch, Division of Surveillance, Hazard Evaluations,
and Field Studies (DSHEFS), National Institute for Occupational Safety
and Health (NIOSH), has a responsibility for the identification of high
risk portions of the work force and for providing comparative assessment
of their risk.

Two of the tools available to us in the accomplishment of this task are:

     The National Occupational Hazard Survey (NOHS) data base, which
     associates potential health hazards with occupational groups,
     industry types, and occupations within industries, and

     The NIOSH Registry for the Toxic Effects of Chemical Substances
     (RTECS), which is a compilation of published chemical toxicological
     data derived from review of the international literature.

Consideration of the data in RTECS and NOHS resulted in the initiation of
a NIOSH contract (*) effort, in which I served as project officer.  This
report is based on the results of that contract, which developed algorithms
designed to:

1.  Calculate an index number for each chemical found in both RTECS and
    NOHS which is representative of the relative health risk posed by
    an individual chemical.

2.  Calculate an index number for each of those chemicals which modifies
    the individual chemical's index number by considering observations of
    chemical use in the workplace.

3.  Calculate an index number which measures the relative health risk posed
    by the entire range of chemical exposures of workers in specific industries,
    occupations, and occupations within industries.
                 DEVELOPMENT OF A CHEMICAL RISK INDEX NUMBER

Development of a system to  calculate a value known as the Hazard Risk Index
(HRI) was the first task in this project.

The initial step in this task was to find out which chemicals were in both
NOHS and RTECS.  By matching on chemical abstract system (CAS) numbers, a
total of 1,904 chemicals common to both bases was found.

The toxicity data in RTECS are reported by species, route of administration,
and test end point, defined as "test classes," shown in Figure 1.  Matrices
were prepared which show the number of RTECS compounds in each of the test
classes, as shown in Table 1.

(1)
   NIOSH contract f;210-78-0076, "Identification of High Risk Occupational
   Groups and Industrial Processes Using RTECS/NOHS Data"
                                            100

-------
   TXDS:    jhl-hmn TCLo:5000 mg/m3/7M TFX:CNS
            b     d
AH B A AM 116.131,36
       i     J
       b    j
a.    An acronym which stands for "Toxic Dose".
b.    This is an abbreviation for the route of administration or entry
      of this substances.
c.    This is an abbreviation for the species.
d.    This is the type of dose reported.
e.    This is the dose which caused the toxic effect.
f.    The first part of this notation, "TFX," is  an  acronym which
      stands for "Toxic Effect."  The last part of this  notation
      referes to the organ system affected by the dose administered.
g.    This is a code denoting the reference from  which the  toxic data
      was derived.
h.    Volume number of the reference.
i     Page number of the reference.
j.    These two digits stand for the year of publication,  i.e.,  1936.
        Figure  1. A  Typical  Toxic Dose Entry From RTECS
                                    101

-------
                                                          LD50
SPECIES   iat  ice  lev  idu  ihl  imp  ipc  ipl  ipr   irn  lap  itr  ivg  ivn   mul  ocu  orl   par  rec  skn  sub
cat
ctl
dog
dorn
frg
mam
mky
mu a
pig
rat
rbt
tod
bir
rod
hum
^
o
re
12 83 59

2 1 22 3 173 155 3

3 235

2 15 21
9 283 9217 6118 2 4728 50
1 4
8 21 27 4 1 1611 3 855 5065 18
4 12 62 3 502 5 359 4

5 13 485
4 176 48 402 1
1 12

5 14

5 21

11

2 6
1 37 2377
1
7 232 632
930 106

15 7
45 91
2

                             Table  I.   An  Example RTECS Species/Route/Test End Point  Data Distribution Matrix

-------
It became obvious that certain test classes contained a very small
number of compounds.  A decision was therefore made to eliminate from
further consideration those test classes with fewer than 100 chemicals.
Almost all test classes with acute toxicity end points met this criteria,
but significant portions of the chronic test data were excluded.  To
compensate for this, chronic test classes excluded by the 100 chemical
threshold were combined by grouping species and routes into an "any
species, any route," (any-any) test class. Certain similar species
were also grouped.  The final result of this effort was the 66 test
classes shown in Table 2.

Because the RTECS dose data is reported in a number of different units,
It was necessary to standardize the dose units by converting inhalation
data to parts per million, and all other data to milligrams per kilogram
of unit body weight.

Three special treatments of the data were necessary.  First, all units
were expressed in molar form.  Second, the dosage was expressed in
log form.  Third, the Draize procedure (J. Pharmacol, Exp. Ther.,
82:277-419, 1944) was used as a standard for irritation data.

It was necessary to develop a method for comparison of data across test
classes as well as from chemical to chemical within a test class since no
single test class covered all the chemicals common to RTECS and NOHS.
This meant that dose data had to be normalized by expressing each dose
in a test class (d.) as a function of the range of doses in that class
using the formula:

                   dmax " di
             dn  * djnax - dm-in       WHERE:   dn    =  normalized dose

                                              di    -  observed dose

                                                    ™  maximum observed dose
                                                       in the test class

                                                    ™  minimum observed dose
                                                       in the test class
This normalization technique was selected because it results in positive
numbers that range from zero to 1.0 within class.  The maximum and minimum
values for each test class and the number of RTECS records in each class
are shown in Table 3.

The final step in manipulation of the dose data was to develop user options
for (1) the generation of neoplastic dose data from carcinogesesis dose
data, and (2) the use of neoplastic data as estimators of carcinogenic
potential when only one of these effects was reported for a chemical.
                                             103

-------
          Table 2.  Test Classes* Selected  for Use in HRIN Algorithm
              any-any
              any-any
              any-any
              any-any
              any-rat
              eye-rbt
              ihl-mus
              ihl-mus
              ihl-rat
              ihl-rat
              ihl-rod
              ipr-mus
              ipr-mus
              ipr-mus
              ipr-rat
              ipr-rat
              ipr-rat
              ipr-rod
              ipr-rod
              ivn-cat
              ivn-dog
              ivn-dog
              ivn-mus
              ivn-mus
              ivn-rat
              ivn-rat
              ivn-rbt
              ivn-rbt
              orl-bir
              orl-cat
              orl-dog
              or1-dog
              orl-hum
CAR
NEO
TER
TFX
TER
SSSS
LCLo
LC50
LCLo
LC50
LCLo
LDLo
LD50
NEO
LDLo
LD50
NEO
LDLo
LD50
LDLo
LDLo
LD50
LDLo
LD50
LDLo
LD50
LDLo
LD50
LD50
LDLo
LDLo
LD50
LDLo
orl-mus
orl-mus
orl-mus
orl-mus
orl-rat
orl-rat
orl-rat
orl-rat
orl-rbt
orl-rbt
orl-rod
orl-rod
par-mus
skn-mus
skn-mus
skn-rat
skn-rbt
skn-rbt
skn-rbt
sub-cat
sub-dog
sub-frg
sub-mus
sub-mus
sub-mus
sub-mus
sub-rat
sub-rat
sub-rat
sub-rat
sub-rbt
sub-rbt
sub-rod
CAR
LDLo
LD50
NEO
CAR
LDLo
LD50
NEO
LDLo
LD50
LDLo
LD50
LDLo
CAR
NEO
LD50
LDLo
LD50
SSSS
LDLo
LDLo
LDLo
CAR
LDLo
LD50
NEO
CAR
LDLo
LD50
NEO
LDLo
LD50
LDLo
*  Test classes  are  defined in terms of route, species,  and end  point,
   RTECS abbreviations are used.
                                        104

-------
                Table 3.  Test  Class Maximum and Minimum Values *
touc«/Specics/
End Poinc
«ny-«u? CAS
say-any HEO
«ny-«ny TSS
«ay-«ay T?X
«ay-r*i TEZ
cye-rbt SSS5
ibl*«u* LCLo
ial-mu« LC50
ihl-r«: LCLo
ihl-r«£ LC50
ihl-rod LCLo
ipr-snu LOLo
ipr-mii L050
ipr-nuj KEO
ipr-r«t LOLo
ipr-rat LD50
ipr-r*r KEO
ipr~rod LCLo
ipr-rod LC50
irft-cct LCLo
irn-dog LCLo
i*n-dog LD50
ivxt-aus LCLo
i*n-nu« LC50
iva-r«: LCLo
iva-rct L050
iTn-rbt LCLo
iTD-rbt L050
ocl-bir LC50
•rl-emt LCLo
erl-doj LCLo
•rl-doj LC50
0rl^buB I^Lo
Mlalaun

12893003
12195::?
10085552
065U512
10313734
116:S9S7
21609436
201C0418
19725555
20421799
18572S76
15290316
06641913
15345643
13241 £28
07215536
15860363
14538331
15305370
13423269
09734997
04144268
13164S15
03551912
10791159
05137520
12634115
03861710
14037132
15514176
14438037
14979315
11644989
MucLzDuo

30554932
30543671
31512924
37581161
33122360
24873E71
34366760
34296066
33704575
38927490
33328629
30532089
27435379
28895523
26961945
27464935
28742188
26767990
26781998
26754135
26747482
25808472
24922546
27475143
26112823
27532867
27748184
26865097
27337250
27173447
28327942
27969330
29163559
Number a
Records
000321
000426
000342
000690
000307
000977
000390
00016B
000436
000215
000103
002266
009486
000137
000820
001663
000102
000105
000183
000209
000260
000179
000290
007194
000240
000896
000489
000507
000323
000100
000147
000166
000151
ttouce/Spedes/
lad Poinc
orl-Bu* CAH
orl-BUi LCLo
orl-mii LC50
orl-Bui HEO
erl-rtt CAS
orl-r«t LCLo
erl-rct LC50
arl-r«t KEO
erl-rbt LOLo
OTl-rb: LD50
orl-rod LCLo
erl-rod L050
p«r-nut LCLo
•ka-nuc CAR
•kn— anit KEO
•ka-rcc LC50
•kn-rbt LCLo
•kn-rbt LC50
•tat-rbt SSSS
•ob-«*t LCLo
gab-dog LCLo
•nb-frg LOLo
»nb-nu« CAS
•nb-snj» LCLo
•ob-mu LC50
•ob-vat KEO
•*-r*t CAR
•ob-r*t LOLo
•at>-r»C L050
•ab-r«c KEO
•id»-rbt LOLo
•i*-Tbt L050
•rf>— rod LOLo
MlnXnun

17073349
15211675
14356787
17073349
09751596
14739038
12556482
09751596
11923172
15354968
16635117
09109745
14527393
12487536
11557886
17482300
16135483
15237360
11646735
14986702
13979462
13137839
12368568
10381526
09079959
12368568
15980739
13529510
06131088
14729737
12730731
12637354
13611102
Maximum

31223984
27426605
28504761
33312714
31990952
27848068
28425430
33078674
29729187
27222244
26897141
27822723
27226227
30521286
32004257
27268005
28377426
32222321
27764832
26004883
27173447
280S2977
2S777573
27069092
28026978
30981949
32076889
27692139
28936172
32076889
26787949
24392731
26759323
Number c
Rt cords
000182
000401
005022
000311
000454
000380
005134
000670
000312
000384
000128
000431
000157
000167
000412
000245
000152
OC0968
OC1310
000103
000105
000164
000195
C00720
OU2426
OJ0567
0)0222
GI 10329
000623
Or0442
01 '0291
OW099
000213
*  Units for the minimum and maximum values are In millimoles  per kilogram
body weight, with the  decimal point after the second digit.
                                        lor)

-------
A preliminary Hazard Risk Index (HRI) algorithm was developed using the
normalized dose for each test class reported for a chemical, modified
by the weighting factors shown in Table 4.  Weights were based on the
assumptions that:  (1) higher animals most closely duplicate human
response, (2) chronic outcomes are of greater importance than acute,
(3) a large test class data population gives better comparisons between
chemicals, (4) multiple-species testing of a chemical gives a better
estimate of overall toxic effect, and (5) those routes most closely
duplicating human industrial exposure (i.e. inhalation and skin/eye
contact) are more relevant than other experimental modes of adminis-
tration.

The algorithm multiplied the normalized dose number for each chemical
by the relevant weights, summed the resulting figures, and divided the
sum by the total of the weight factors to produce a weighted average
indicating the relative toxicity of the chemical across all reported
toxic effects as shown in Figure 2.

This preliminary model was submitted to a panel of toxicologists from
industry, government, and academia.  The panel concluded that:  (1) the
rank-ordering of hazards was not entirely correct, (2) misplacment of
chemicals in the ranking resulted in part from combining acute, chronic
and irritation data,  (3) weighting factors seemed to contribute to improper
chemical ranking, (4) the fact that no one test class or small group of
test classes provided total coverage of the RTECS data was a significant
limitation,  (5) the use of normalized dose data appeared reasonable and
useful,  (6) the expression of dose data in molar form was consistent with
the current views of pharmacologists and toxicologists, (7) chronic
effects were insufficiently emphasized, (8) the procedure for handling
primary  irritation data could be improved, (9) the RTECS data may be
skewed since only the lowest reported dose at which a particular effect
appears  is listed and the data is in all cases unevaluated.  In recognition
of these comments, the second version of the HRI was produced.  This
version produced a risk index composed of data on acute toxicity, primary
irritation, carcinogenicity, neoplastigenicity, teratogenicity, and other
toxic effects categories (sub-HRI's) for each chemical.  The HRI for the
chemical was calculated by averaging those sub-HRI's for which there was data
as the HRI for the chemical (See Figure 3).  This averaging procedure was
adopted  to avoid the assumption that missing data equates to zero effect.

The final output of the algorithm is shown as Figure 4.  The final form of
the algorithm allows the user considerable flexibility in choosing which
toxic effects to emphasize.  The impact of any sub-HRI upon the final HRI
can be increased or eliminated by using a multiplier (a, c, e, g, i, and k),
or the user can cause chemicals with data in an area such as carcinogenisis
to be emphasized by using sufficiently large constants (b, d, f, h, j, and
1).  In addition, the test class weights previously discussed can be used.
In short, the "dials" on the model can be set to adapt the algorithm to any
desired toxicological priority desired.  In the project model discussed
here, all multipliers and weights were set at 1.0 and all constants at zero.
                                            lOu

-------
                   Table  &.  Weighting Factors Used  in First
                    Calculation of Hazard Risk Index Numbers
                        Weight
               Characteristic
Test Species:
5
2
1
0.5
hum
cat, dog
rat, mus, rbt, rod
bir, frg
Test End Point:
5
4
2
1
CAR, wrr
KEO
TER
All others
Number of Chemicals in Test Class:
                             3
                              »
                             2
                             1
               1000
               500-1000
               less than 500
Number of Different Species:
                             1.2
                             1.1
                             1.0
               3 or more
               2
               1
Route of Administration:
                                            IHL, eye, skn
                                            All others

-------
HRIN CALCULATION BASED ON RTECS DATA AS OF SEPTEMBER 1978
H     AJ9625000
ipr-sus LCLo
orl-rat LD50
HI *   6.00

H     KL7525000
eye-rbt ssss
ipr-sus LDLo
orl-rat LD50
skn-rbt SSSS
UI =  26.00

H     KJ9100000
eye-rbt SSSS
ipr-irus LD50
orl-rat LD50
orl-rod LD50
skn-rbt LD50
HI =  23.00

H     VB8225000
ipr-mus LDLo
orl-rat LD50
HI =   6.00
              ACETIC ACID,  IRIFLUORO-
           2323        3.00       .£21
           5064        3.00       .350
          N SPEC =   2               HRIN
                              .479
              EIHANOL,  2,2'-CttEIHYLIMINO)DI-
            679
           2323
           5064
           1009
          N SPEC =  3
        8.00
        3.
        3.
00
00
       12.00
.406
.445
.153
.((38
                      HRIN
                    .475
              EIHANOL,  2-C2-3UIOXYETHOXY)-
            679       8.00      .598
           9217       3.00      .167
           £064       3.00      .152
            402       1.00      .160
            930
          N SPEC =  4
        8.00
        .393
                      HRIN =  .472
              QUINOLINE,  6-EIHOXY-1,2-DIHYDRO-2,2,4-TRIMEIHYL-
           2323       3.00      .545
           £064       3.00      .303
          N SPEC =  2               HRIN =  .466
H     TH4330000     -PHOSPHORUS SESQUISULFIDE
orl-rbt LDLo      309       1.00      .461
HI =   1.00     N SPEC =  1               HRIN =  .461
H     E02975000
ihl-rat LCLo
orl-rat LD50
skn-rbt LD50
skn-rbt SSSS
HI =  27.00
            443
           £064
              BUTYLAMINE
           1009
          N SPEC
        4.00
        3.00
        8.00
       12.00
        .371
        .264
        .439
        .452
                      HRIN =  .457
H
XU0175000
eye-rht ssss
orl-cat LDLo
orl-rat LDLo
orl-rbt LDLo
sub-cat LDLo
sub-rbt LDLo
UI =  16.00

R     AI7700000
ipr-nus LDLo
ipr-rat LD50
iwn-rbt LDLo
KI «   7.00
TOLUENE,  2,4,6-TRINITRO-
            679
            100
            891
            309
            103
            297
          N SPEC =
        8.00
        2.
        2.
        1.
        2.
        1.
00
00
00
00
00
.440
.237
.337
.372
.347
.262
                      HRIN =  .450
              ACETIC ACID,  HERCAFTO-.  HONOSODIUM SALT
           2323       3.00       .503
           1610       3.00       .242
            484       1.00       .370
          N SPEC =  3               HRIN =  .446
        Figure  2.  Sample Page  from First HRI  Draft
                             loa

-------
                    FIGURE 3  -  THE HRIN ALGORITHM



HRIN  =  (aAT 0 b) + (cPI & d) + (eCAR €> f)  + (gNEO 9 h)  + (1TER 0 j)  + (kTFX 0 1)

                                       N
Where:

       AT  *  the weighted average of all acute toxicity  normalized doses,
              termed the acute toxicity sub-HRIN

       PI  -  the primary irritation sub-HRIN

       CAR »  the carcinogenic sub-HRIN

       NEO «=  the neoplastic sub-HRIN

       TER -  the teratogenic sub-HRIN

       TFX «  the sub-HRIN derived from all  other chronic toxic effects

       Lower case letter a through 1 = weighting factors  for which values
                                       are selected by the user

       N   «  the number of sub-HRIN1s for which there are data

       ^   «  an addition that is performed  only if the associated sub-HRIN
              is not equal to zero
                                          101)

-------
IH IN CAICULXTIOH IA9CD ON RTCCS D»T» US OF JANUARY 1979 03/17/79
muc »• i.oo B- .00 c« i.oo o- .00 c» i.oo r- .00 o- t.oo M- .00
I* 1.00 J« .00 K- I.OO I-
SC« 1 20 II TJ2'l50000 PHYSOSTICtllNE, SALICYLATE CUD
ony-nny TFX 690
ipt-m\ia ID50 9'l86
ivn-»us LDOO 7I9M
or 1 mis I.n!i0 5022
E till -mus I.D50 2'l26
culi-ibt ll)l.o 291
Biih-rLt 1050 99
AT • .638 PI « .OOC
SEQ 1 21 II EV2700000
niiy-nny TED 3'I2
ipt-nus 1050 9'IOA
orl-doI'I2
AT * .161 PI • .OOC
SEQ 1 22 II DUO 175000
i|>t-irii3 l,DLo 2266
• lib-rot MEO M'I2
AT » .199 fl * .00
SCO 1 23 II UX6825000
onf-onv TFX 690
Jpr-nms IBHO 9M66
il't-rot LD50 1663
orl luc I.D50 323
orl-luim into 151
otl-xus 1.000 5022
oxl-tnt I.D50 5I3'I
GHii-t:il IDSO 245
sl'.n-tlit I.D50 968
sub-mils L050 2126
lub-int LD50 625
M =• .699 PI • .OOC
SEQ 1 2'l II UL2275000
ipt-miis IDT.O 9>l06
irt-tiit I.DI.O 820
iun-cnt IDLo 209
i vn- dor) LDIo 260
ivn-uuis LDSO 7I9'I
jun-rnt IDGO 896
inn-Mil ini,o '189
ocl-liii 11)00 323
oil-cat LDLo tOO
.00
.00
.00
.00
.00
.00
.00
CAR •
CADMIUM
.00
.00
.00
.00
.00
.00
.00
.00
CAR •
BEHZOIC
.00
.00
CAR •
823
558
506
799
6I'I
698
651
.000 NEO - .000 TER • .000 TFX •
SULMIE (tit) '
8'l5
mo
183
£21
'132
507
80S
831
.895 HEO • .709 TER • .815 TFX •
ACID. o-(6-(EIIIVLAHINO)-3-(ETHYtimNO)-l
899
637
.637 NEO • .637 IER • .000 TFX «







.813 IIRIN • .730









.000 HRIH • .716
,7-DINETIIYt-3M-XllNIMEN-9-n»-i


.000 IIRIN • .724
PVROPIIOSPIIORIC ACID. IETHAETIIYL ESTER
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
CAR -
7M8
527
5S6
807
675
701
786
997
823
601
513
.000 NEO • .000 TER • .000 IfX •










.748 HR1N - .713
STRYCHNINE
.00
.00
.00
.00
.00
.00
.00
.00
.00
527
733
856
68M
503
«<9«l
817
735
9'lO









                                                                 HEO-3U-TES
                                                              .00 CM-SU'TES
Figure 4.  HRI Sample Index Page

-------
      DEVELOPMENT OF RISK INDICES USING BOTH RTECS AND NOHS DATA

It was decided that since the HRI list was in fact a "relative risk"
list, we should express exposed workers as a function of observed NOHS
worker groups.  (A worker group is-defined as a population within NOHS
with the same occupation, industry, or chemical exposure characteristics,
See Table 5.)  This was accomplished by expressing people exposed as:

                               (PES)
                               (PEN)

     Where, by specified worker group:

     PES is defined as the number of people noted in the NOHS data
         base as potentially exposed to a specific chemical hazard.

     PEN is defined as the number of people noted in the NOHS data
         base as exposed to any chemical, physical, or biological hazard.

This "worker exposed ratio" allowed comparison of relative worker group
risk.  For example, we can compare the benzene exposure of one out of
the forty (1/40) plumbers observed in SIC "A" to twelve out of the
ninety (12/90) plumbers observed in SIC "B".
By assuming that part-time exposures (less than four hours per work day)
resulted in half the risk of full-time exposures,  exposure duration is
represented in the algorithm by the expression (0.5 + 0.5 PFT), where
PFT = the percent of workers exposed to a chemical hazard more than four
hours per work day.

To integrate information on the control practices  used in conjunction
with chemical exposures, "controlled exposures" were expressed as percent
of all exposures controlled (PC).  Since few control measures are 100%
effective, the 90% control achieved by an acceptable respirator is used
to express the general effectiveness of control measures.  The expression
for input of control data therefore became (1.0 -  0.9 PC).
                  WORKER GROUP RTECS/NOHS RISK INDICES

Worker group-specific risk indices are produced in the four general
categories described below.
                   THE ADJUSTED HAZARD RISK INDEX (AHRI)

This index is based on the rationale that assessing the risk posed by a
chemical hazard is dependent upon both the chemical's toxicity and the
extent of worker exposure.  That is, a highly toxic chemical with very
restricted use may be of less concern than a widely used chemical which
is less toxic.
                                                Ill

-------
                      Table .5.  NOHS Data Files


                Number
File Number   of Records               Worker Group

     1          1,904        All workers without regard to industry or

                             occupation.

     2         29,548        Workers within an industry at the 2-digit

                             SIC level, without regard to occupation.

     3         74,859        Workers within an industry at the 3-digit

                             SIC level, without regard to occupation.

     A        113,248        Workers within an industry at the 4-digit

                             SIC level, without regard to occupation.

     5         78,078        Workers within an occupation without regard
                •i\
                             to industry.

     6        305,466        Workers within an occupation within an

                             industry at the 2-digit SIC level.

     7        474,094        Workers within an occupation within an

                             industry at the 3-digit SIC level.

     8        550,020        Workers within an occupation within an

                             industry at the 4-digit SIC level.
                                       1.U

-------
The algorithm used for production of this index is expressed as :

     AHRIN - (HRIN) (fff) (1.0-0. 9 PC) (0. 5 + 0.5 PFT) (K)
     WHERE:

       AHRIN - Adjusted Hazard Risk Index Number

       HRIN  «= Hazard Risk Index Number

       PES   - Number of people observed exposed to a specific
               chemical hazard in NOHS

       PEN   * Number of people observed exposed to any NOHS
               hazard

       PC    - Percent of controlled exposures

       PFT   * Percent of full-time exposures

       K     »A constant used to remove leading zeroes

An example of the output from the AHRIN algorithm is shown as Figure 5.
All input factors are shown in the printout.


                     THE INDUSTRIAL RISK INDEX (IRI)

The rationale of the IRI is that industry worker group risk is a function of
the summed risk associated with the chemicals to which the group is exposed.

Examination of the algorithm results indicates two products of this algorithm.
This became apparent when considering the algorithm equation, which is
expressed as:

     IRIN «=   (HRIN) (Zli) (1.0-0. 9 PC) (0.5 + 0.5 PFT) (N)
                     PEN
     WHERE:

       IRIN  = Industrial Risk Index Number

             • The summation of:

       HRIN  = Hazard Risk Index Number

       PES   " Number of people (specified SIC) exposed to an NOHS chemical
               hazard

       PEN   m Number of people (specified SIC) exposed to any NOHS hazard

       PC    " Percent of controlled exposures

       PFT   • Percent of full-time exposures

       N     " Population (specified SIC) from census data

                                                   li.i

-------
a im IN CAiciiiATioH n»sen ON tin in BAIR RS or 01/27/79     RTECS DRIB  01/79
  •< t.OO b«   .00 c-  1.00 d"  .00 ••• 1.00 fi   .00 g> 1.00 h-  .00 1-  1.00 j<
HUN DRIB 11/02/79     NOII9 DRIB 01/79
 .00 N- 1.00  1"  .00 CKR-MU-VCn lirO-SH"VC9
8M C
HEQ i
sr.« •
SEQ 1
SCO 1
SEQ g
SEQ 1
SEQ 1
SE4 1
SEQ 1
SEQ 1
SEQ •
sr« l
SEQ 1
SCO Q
SEQ 1
SC« 1
1
IIRIII
i
MRIH
3
tin IM
M
liniH
9
IIRIM
6
nniii
^
'MRIH
•
IIRIH
9
IIRIM
10
IIRIH
II
IIR1H
It
IIRIH
13
IIRIM
I'l
IIRIH
15
IIRIH
16
IIRIH
17
IIRIH
XR2275000
SEQ • z'ii
tit 190000
SEQ • 163
HO 6 300000
SEQ "ID 15
XS5250000
SEQ • 728
VV73IOOOO
sr.Q • 905
HT6050000
3E«I -138!
KJ2975000
SEQ » 761
ROI200000
SEQ • 372
i:x'i55onoo
SEQ '109(1
Elfi'175000
SEQ •= 9A2
ZIII'IDOOOO
SEQ • '109
VZM725000
SEQ -1208
CU7700000
SEQ - 211
JJ7000000
StQ '1178
CC9'ISOOOO
SEQ • 6'I2
VZ227SOOO
SEQ -1033
CI6M75000
SEQ ° 217
TITRHIUIt nxiOf
IIRIH - .506 PES' 111372
XYLEME (pined)
IIRIH ' . 5'lf, PtS- 107316
ETHYL fILCOIIOL
IIRIH ° . 195 TES* I9fl!»'l0
lOlllEME
IIRIH - .361 rtS' Il'l30'l
SILICA, Minnnious runco
IIRIH " .3?.'! PEH" lOZnilS
inopnorvt ULCOIIOL
linill ' .231 PFS-- 13010'!
rniAiir, 1. 1, i-Tniciit.ono-
IIRIII ' .353 PE5» 7755 1
ALUIIIIIIIM OXIDE (2'3)
IIRIH - .'(55 PES' A6220
CTIIYLEHE, TRICIIlOnO-
linill • .207 PES« 107291
2-DUTAIIOMC
IIRIH - .311 PES- 66'I76
ZINC CHLORIDE
IIRIH - .'I'l 5 PTS- HM660
SODIUM CHLORIDE
IIRIH • . 2'I9 PES« 100011
PEH-
PEN-
PEH-
PEH-
PEH-
PEH-
PEH-
ft.H-
PEI<«
PEH-
PEH-
PEH-
51!i!>G9
S'IS569
5>I5569
5*15569
5>l5569
S'15569
5M5569
5'I55C9
5'I5S69
5 '15569
545569
5M5569
2-CYCLOIIEXEH-l-OHE, 3, 5, 5-IRIHEIIIYL-
IIRIII " .57.0 PEfJ- 39338 PEM> 5'I5569
nrpiirnjLMtiME
IIRIH - .271 PES« M7071
niirninriY TRisuirioe
IIRIH • .379 PES" II199M
SODIUM DORATE
IIRIM • .300 PES» 67793
A5IRESTOS
IIRIH • .517 PES' I|II'I2
PEH-
PCII-
PEH=
PEN"
5'I5569
5'I5569
545569
S'I5569
PC-
PC"
PC-
PC-
PC-
PC-
PC-
PC"
PC-
PC"
PC"
PC"
PC-
PC"
PC-
PC-
PC"
.MO
.MS
.2*1
.33
.46
.26
.23
.39
.36
.32
.15
.36
.22
.00
.00
.26
.MS
PfT"
prr-
pri-
prt-
prr-
pri-
pri-
prt-
prt-
prt-
prt-
prt-
pri-
pri"
prt-
prt-
prt-
.27
.04
.01
.03
.22
.03
.01
.10
.01
.04
.05
.02
.01
.29
.00
.01
.16
RIIRIH •
RIIRIH -
RIIRIH •
JUIRIH -
AIIRIH •
RIIRIII -
RIIRIH *
AIIRIH >
AliniH "
RIIRIH >
RIIRIH •
RIIRIH •
AIIRIH "
RIIRIH •
RIIRIH -
RIIRIH •
RIIRIH •
11. 988
31.764
28. 123
27.387
21.852
21.761
20. I!"
20.026
19.330
18.231
16.55)
15.892
15.213
15.086
11.617
14.459
13.473
                              Figure  5 .  A1IRI Sample Index  Page

-------
Removal of the "N" modifier converts this algorithm to an expression of
relative individual risk, rather than relative industry risk.

In reviewing NOHS worker group population data, a number of instances
were noted where the observed PEN was less than 10.  When this occurred,
the PES/PEN ratios frequently approached 1.0.  It was felt that this
was probably an artifact of particular observations rather than a
probability sample.  It was decided that NOHS worker group data should
be used only when the observed PES/PEN proportion was within a selected
range of the Gaussian approximation.  Using a confidence interval of .95
with an accuracy of + 0.25, a minimum PEN size of 16 is derived which
was used for this project.  Based on these calculations, a 95% confidence
interval of the IRIN based on the PEN value is provided in the output.

An example of the IRI output is shown as Figure 6.


                THE OCCUPATIONAL RISK INDEX (ORI)

The rationale of the ORI is the same as that of the IRI for occupational
groups.

The algorithm for production of the ORI is identical to that used for
the IRI except that the NOHS data elements are for occupations regardless
of industry.

The use of census data as a modifier and the expression of individual
versus group relative risk apply equally to the ORI, as do the confidence
interval and minimum sample size discussions.

An example of the ORI output is shown as Figure 7.
          THE OCCUPATION WITHIN INDUSTRY RISK INDEX (OWIRI)

The rationale of the OWIRI is identical to that of the IRI and ORI for
occupations within two-, three-, and four-digit SIC's.

The algorithm for the production of the OWIRI varies from that used for
the IRI and ORI in two particulars:  (1) the NOHS data elements are
occupation within industry specific, and (2) no census data is available
at this level of specificity, so a constant to remove leading zeroes
is ued.

The OWIRI output is shown in Figure 8.

Th sample size and confidence interval provisions are equivalent to
those in the IRI and ORI.
                                              11.1

-------
IMI c»irui.»noH Busen OH HRIH DMTR >!t or 03/17/79 RT
•« 1.00 b« .00 o- 1.00 d« .00 e« 1.00 {• .00 «•
c»R-3U'»r.!» Mr.o-su"YF.n cEHnun DUTA SH'tes
SC4 1
8C9 0
sc9 e
St« 1
SM 1
SCO 1
SC4 1
SCQ 1
SE9 1
fitq I
Stfl 1
SE4 1
SE4 1
SE« 1
SF* 1
sr.« a
SR« 1
1
1
3
'I
5
•
7
1
9
10
11
It
13
I'l
19
16
17
SIC 806
SIC 551
SIC 580
SIC 55'l
SIC 508
SIC 809
SIC 801
SIC 599
SIC S'll
SIC 739
SIC MSI
SIC 509
SIC 531
SIC 371
SIC 651
SIC 275
SIC 701
iinnrims
HimnF.n or CIIEIUCIUS » 856
MF.U MID USED CM! DEALCHS
Hunnr.R nr ciir.iucM,s - 293
F.AIIIIG dim nniMKina PLACES
MUIII1F.R OF CIIEIMCftl.5 = Z'I9
nnnoLiiiE rcnvicf: SIAIIOIIS
HIIMHFR OF Clir.lUCM.r, = I9'l
iinciiiiiFnY, F.qiiiriifHT. AND SUPPLIES
MiiMiiF.n OF riiriiicni.s > z 511
HOTOR VEHICLE!: AMD EqUIPHEllI
IIUMRFR Or CIIEIIICAI..1 ° 583
BF.AI, tr.IATE OPERATORS AND LESSORS
iniitnEn OF cnriiiCM.s = 298
CnilMERCIAL rnillTIUQ
iiiiiintn or CHEMICALS « 336
MOTELS. TOUniST COURTS, ADD HOTELS
                                   01/79    RUM BUTE  11/01/7*
                                   .00 1" 1.00 1*  .00 h« 1.00 1*
                                             16
HOIM nm vi/79
.00
                             tIJH  NOUS SAMPLE-
IRIN •
IRIH •
l/-
1RIN -
IRIN '
!/-
IRIH -
IRIH •
IRIN "
IRIH «
IRIH •
IRIH •
IRIH •
IRIH •
IRIN -
IRIH "
IRIH -
IRIH •
IRIH -
705. At*
7R.101
181.697
7l.l»0fi
M«|7.3»3
55lo«9
268.525
R. P.' 7.
228.77.6
21. 1'ir.
202.135
116.373
192.092
3.81ft
191.693
20.1VI
iei.8'17
21.73'!
I7'l.816
3'l.96.1
109.132
10.675
162. 152
N.858
153.057
1.215
112.503
12.401
135.528
10.073
132.191
5.899
Figure  6 •  IBI Sample Index  Page

-------
ORI CALCULATION BUSED on IIRIN DATA us or  01/17/79     mcs DATE 01/79      HUH DATE 11/01/79     NOUS  MTE 01/79
      • •  1.00  li«  .00 0- 1.00 i1«  .00 ••>  1.00 {•  .00 «• 1.00 h«   .00 1"  1.00 1'  .00 It* 1.00 1"   .00
     CAR-SU'VES NEO-SH*»ES     CEHSIIS DATA  BM'YE.1     IIIH HOIIS SAMPLE-   16
RE9 0     I    OCC 903
  JANITORS AHD SEXTONS
        or  CHEMICALS »   B>I'I
I   OCC OOn


3   OCC 075


M   occ 9'iii


S   OCC 925
SEQ  I


SEQ  I


SCQ  0


SE«  I


BTA  I     6   occ ool


SC4  I     7   OCC 2(10


srq  •     •   occ IBI


SC4  I     9   OCC 305


SE4  I    tO   OCC 902


SE4  I    11   OCC 510


SCO  0    11   OCC 623


SE«  I    13   OCC 602


StO  I    I'l   OCC 912


SE4  0    15   OCC 715


SK4  I    16   OCC 'I'll


SCQ  I    17   OCC 004
  AUTOMODUR llfCIIMIICS AMD APPRENTICES
nuiuirn  or  CHEMICALS =   537
                           REGISTERED  HUllSfS
                         HUltnF.R Or Clir.llICALS
                         512
  MAiRnnEssr.nn nnp ro^nr.TOLoaisra
liunnr.it  or  cnrnlcnin »   219

  MunSIMG  A1DF.5.  ORDERLIES, AND ATTENDANTS
Nunnr.R  or  CIIF:IIICAI,S *   322

  SEcnr.iftnir.s. IIEPICAL AMR N.E.C.
MIIIIRr.R  Or  CMF.IIICni.S <   'I'I2

  SALESMEN AMD SALES CLERKS. N.E.C.
Munnr.n  or  CIIF.IIICHLS "   330

  iiEnyv EQUIPIIEMT IIF.CIIAIIICS. met. DIESEL
HUlinER  Or  CIIF.IIICALS =   798
                           BOOKKEEPERS
                         NUIIRER Or Cllf;llICni.R
                         2'l6
  CLEAIIERf! AMD CIIARHOHEH
Miiiinrn or  CHEMICALS -   6W

  PAIIITERS,  COHSTRURTIOH AMD HAIHTEHAHCE
NUItnER or  CIIFIIJCALS -   376

  GARAGE UORKERS AHD CAS STATION ATTENDANTS
NUIIRER Or  Clir.llICALS «   2'I2
                           ASSEMBLERS
                                or CIIEIIICALS
                         7l'i
  COOKS,  EXCEI'T  FPIVATC HOUSEHOLD
Nunnr.R or cur.mcALS -   2'i9
  TRUCK DRIVERS
NHIinER OF CIIEIIICALS
                                                306
  rOREIIEH. N.E.C.
MWIfU.R or CHEMICALS •  1101

  ELECTRICIAUS  AHD APPRENTICES
liimnr.R or CHEMICALS =   6ns
OR IN  «
    •/-

OR1H  «
    •/-

ORIH  -
    «/-

ORUI  •
    «/-

ORIN  -
    •/-

ORIH  »
    •'-

ORIH  •
    •/-

ORIH  »
    */-

ORIH  »
    «/-

ORIH  •
    «/-

ORIN  •
    */-

ORIN  -
    */-

ORIN  »
    «/-

ORIN  •
    »/-

ORIN  «
    I/-

ORIN  •
    »/-

ORIH  •
    */-
                                                                                                                      N67.2SS
                                                                                                                       27.5H2

                                                                                                                      4G3.0riA
                                                                                                                       io.«//

                                                                                                                      '102.510
                                                                                                                       87.076

                                                                                                                      276.730
                                                                                                                       »?..'r/'i

                                                                                                                      233.001
                                                                                                                      230.515
                                                                                                                        2.IA1

                                                                                                                      226. 98'!
                                                                                                                        «.5f. i

                                                                                                                      no 879
                                                                                                                        7.209

                                                                                                                      IB0.1G7
                                                                                                                        6.677

                                                                                                                      149.376
                                                                                                                        i.3B'i

                                                                                                                      IM'1.508
                                                                                                                       21.0/7

                                                                                                                      1U2.525
                                                                                                                        5.n72

                                                                                                                      141.665
                                                                                                                       27.2ft.i

                                                                                                                      138.521
                                                                                                                        "i.fti'i

                                                                                                                      IJ'I 988
                                                                                                                      133.203
                                                                                                                        '1. 620

                                                                                                                      127.706
                                                                                                                        i.e?.;
                                 Figure  7.  ORI Sample  Index  Page

-------
OW1RI CALCULATION BASED OH UK III DATA A3 Or 03/27/79 RTEC5 DATE 01
M* 1.00 b* .00 c- 1.00 ri> .00 f 1.00 f« .00 o- 1.00 h* .00
CAR-SUITES NEO-SU'TES CENSUS DATA RU-HO HIM HOIIS SAMPLE- 16
8tf 1
SE« 0
SE, .
5E« C
sr<» 8
SE« 8
SC4 8
3E4 8
SC4 9
SE4 8
SE« 8
SE« 1
SED 8
SE« 8
•SE4 8
se« i
SE4 8
1 OCC 903 JANITORS AHD SEXTOHf]
NlllinER or CHEMICAL!] = 227
1 OCC 473 AUTOMOnil.E MECHANIC!!
IHII1MER OF CHEMICALS • 121
3 OCC I'll ADULT EDUCATION TEAI.'IIEP;
iiiiMiirn or CIII.IIICAI.S • 122
1 OCC 635 METAL PLATERS
itiinnrn or CHEMICALS • IOB
9 occ IIA'I orricE MACHINE
HHintr.it or CIIFIIICALS • lo'i
• OCC 473 AUTOnoniLF. IIECIIAHIC.1
IIIIIIIIF.il or CIIFIIICALS • 192
7 OCC '181 HEAVY EQUIPMENT MECHANICS, IHCl. DIESEL
IIUIIIIF.il or CHEMICALS • 166
• OCC 915 UAIIERS
NlllinER or CHEMICALS « 67
9 OCC 552 TELEPHONE INStALLERS AMD REPAIRMEN
NlllinER or CHEMICALS « 6«
10 OCC 950 HOUSEKEEPERS. EXC. PRIVATE HOUSEHOLD
iiminrfl or CHEMICALS • 60
11 OCC '170 AIR CnilDIllOMIHa, HEATIIIG. AHD RErRIOERR
IIUIinER or CHEMICALS • 69
12 OCC 510 PAINTERS, COHninUCI 1011 AND IIAIMIENANCE
HIIMnER OF CHEMICALS * 90
13 OCC 751 CONSTRUCTION LAnORERS, EXC. CARPEHTER' H
iiuiinEn or CHEMICALS • 121
14 OCC 933 ATTEHDAHTS, PERnOIIAL SERVICE, H.E.C.
IIDIinER OF CHEMICAL!; * 106
15 OCC 514 PATTERN AHD MODEL MAKERS, EXC. PAPER
HUIIRER Or CHEMICALS =• IM2
16 OCC 902 CLEANERS AHD CIIAni-IOMEN
HIIIIRER Or CHEMICALS " 109
17 occ 623 OARnnr WORKERS niin^nAS EIAIION ATTEHDKHT-
SIC '15
SIC 79
SIC 3Jt
SIC 32
SIC 59
SIC 45
SIC 45
SIC 34
SIC 17
SIC 32
SIC 53
SIC 28
SIC 49
SIC 79
SIC 79
SIC 59
SIC 37
                                             HUH Dm  It/01/79     HOWS BUTE
                                        t.M 1"  .00 *•  1.00 1"   .00
                                      TRANSPORTATION  HT MR


                                      AMUSEMENT  G  RECREATION SERVICES, NEC


                                      STONE,  CLAY,  AND GLASS PRODUCTS


                                      STONE.  CLAY.  Ann GLASS PRODUCTS


                                      MISCELLANEOUS RETAIL STORES


                                      TRANSPORTATION  BY KIR


                                      TRANSPORTATION  BY KIR


                                      MnniCATED METAL PRODUCTS


                                      SPECIAL TRADE CONTRACTORS


                                      STOIIE.  CLAY,  AMD GLASS PRODUCTS


                                      RETAIL  GENERAL  MERCHANDISE


                                      CHEMICALS  HMD ALLIED PRODUCTS


                                      ELECTRIC.  GAS.  MID SANITARY SERVICES


                                      AMUSEMENT  C RECREATION SERVICES. NEC


                                      AMUSEMENT  G RECREATION SERVICES, NEC


                                      MISCELLANEOUS RETAIL STORES


                                      TRANSPORTATION  EQUIPMENT
• 1/79
OUIRt'
OUIRl-
OUIRI"
OUIRI'
I /-
OUIRI'
OUIRI'
OUIRI'
OUIRI'
OUIRI'
OUIRI*
l/-
OUIRI*
OUIRI-
OU1BI'
OUIRI'
OUIRI*
OUIRI*
• /-
OUIRI'
IS 303
It 981
2.3/0
I.Vifl
11.615
2.235
11.270
I.Z'I'I
10.981
2.661
2 '. .If. ;
10.7'U
I.2IG
10.694
10.361
10.283
9.873
2.Jf.ri
9.805
.«on
2!'lfi1
9.R29
.'IA2
9.500
.576
9 Jt74
1.495
Figure   8.  OUIRI  Sample  Index  Page

-------
                     APPLICATION OF THE RISK INDICES

This system is designed to respond to user priorities.   For example,
user emphasis of potential carcinogens, and examination of the potential
exposure patterns of the various NOHS worker groups to  chemicals with
this potential effect accomplishes, several purposes:  (1)  the user
Is provided with a listing of all the potential carcinogens in RTECS
and a listing of their other toxicological effects, (2) the user is
presented a rank-ordered list of chemicals (AHRI) exhibiting carcino-
genic potential based on both their toxicologic potential  and the
number of workers affected, (3) the user is provided a  risk-ordered
listing of various worker groups to aid in establishing priorities for
field investigation or research, (A) using risk indices and chemical
lists relevant to specific worker groups, a user could  derive an
overall risk number for a specified geographical area.   Specifically,
this concept might be utilized to forecast comparative  health system
demands in an area where the industrial composition is  changing.  For
example, if a new facility of SIC X will double the employment in
SIC X in the area, a rough approximation of the additional health
system demand could be made.
                            CONCLUSION

This system, as presented, is a trial effort.  Our intent is to make
it useful to the occupational health community.  To this end, in-house
research is continuing in several areas.

Through manipulation of the algorithm, we are examining the effect of
each component on the resulting risk index numbers.

We are also producing risk indices based only on specific outcomes
(e.g. skin irritation) in order to identify worker groups at high risk
and to examine the contribution of specific outcomes to the overall risk
index number for various worker groups.

We intend to compare the results of these efforts with published data from
such sources as the Bureau of Labor Statistics, Social Security Administration,
and the National Center for Health Statistics using statistical techniques.
The results of these conclusions will be used to direct necessary changes
in the system.

Finally, we intend to continue evaluating alternate sources of toxicological
and chemical exposure data, and to supplement or replace the data used in
this project.
                                                 Hi)

-------
Unidentified Speaker:
slide?
                                Discussion
Can you explain  some  of  the  figures on this
Mr. Pedersen (NIOSH):   I  will  give it  a  try.   Let  us take number 4, if
you will, Occupation 635, metal  platers.  They are exposed to 108 of the
chemicals from the matched set of 1,904,  according to the NOHS data
base.  That occupation within  SIC 32 has  the  risk  index number expressed
at the right-hand side of the  page.  In  that  particular line we are
addressing only metal  platers  employed in SIC 32.
                                      120

-------
                        Industrial  Hygiene

                      Study of Workers Exposed

                          to Nitrosamines
                          WRITTEN BY:
                         John M.  Fajen
                  Industrial Hygiene Sectioa
                 Industry-wide Studies Branch
Division of Surveillance, Hazard  Evaluations,  and Field Studies
     National Institute for Occupation.il  Safety and Health
                       Cincinnati, Ohio
          First NCI/EPA/NIOSH Collaborative Workshop
              Progress on Joint Environmental and
                  Occupational Cancer Studies
                      Rockville, Maryland
                         May 6-8, 1980

-------
                              ABSTRACT
Under the NIOSH-sponsored contract,  a total of 45 plant surveys were
conducted at 37 separate manufacturing plants. The industries surveyed
were the azo dyes, fish processing,  fish meal, cutting fluid manufac-
turers and users, rubber and tanning. Airborne concentrations of Thermal
Energy Analyzer (TEA) responsive compounds were found in all the indus-
tries except fish processing.  The dye industry had airborne TEA respon-
sive material as high as 40 ug/cu m,  but they were not identified. Air
levels of N-nitrosodiethanolamine were detected at 0.08 ng/cu m in a
plant which uses cutting fluids. A fish meal factory was found to con-
tain N-nitrosodimethylaraine (NDMA) at 0.06 ug/cu m. In a chrome tannery
NDMA was identified at 47 ug/cu m. The rubber industry has airborn
levels of N-nitrosomorpholine  as high as 250 ug/cu m.

This study has resulted in an  increased understanding of mans1  exposure
to performed N-nitrosamines. It is conceivable, from the information
that has been generated in this study, that nitrosamine exposure as
large as that in the tire and  rubber  industry exist in other industry
not yet surveyed.

-------
                            INTRODUCTION




The National Institute for Occupational Safety and Health (NIOSH) has




conducted, under contract, environmental monitoring in a wide variety




of industrial facilities to determine workers' exposure to N-nitrosa-




raines. These compounds have been demonstrated to be highly toxic and




potent carcinogens in laboratory animals (Druckrey et al., 1967).




N-nitrosamines consist of a large family of compounds of which more than




100 of the 130 different N-nitroso compounds tested have been shown to




be carcinogenic in a wide variety of animal species (Druckrey et al.,




1969; Magee & Barnes, 1967; Magee & Schoental, 1964; Magee et al., 1976).




Some of these compounds have been shown to be carcinogenic in rats with




doses as  low as 1 to 5 ppm of N-nitrosodimethylamine (NDMA) and N-nitro-




sodiethylatuine (NDEA) in the diet.









N-nitrosamines are the N-nitroso derivatives of secondary amines with




the general formula R^/N-NO, R1 and R« being virtually any organic group.




One of the simplest members of this family of compounds.is N-nitroso-




dimethylamine £„ x-N-NO. This compound is also a regulated carcinogen




under part  1910..016 of the United States Occupational Safety and Health




Standards. H-nitiosamines may be formed by the reaction of secondary




amines (Mirvish, 1975; Scanlan, 1975) and nitrogen oxides, however, under




appropriate conditions primary and tertiary amines (Smith, 1967; Ohshima




& Kawabata, 1977) can also be nitrosated to produce these compounds.




The NO, or nitrosyl part of the compound, can be derived from nitrogen




oxides such as NO, N02» N2°4» or N2°3 (Challis et al., 1977) or from
                                     123

-------
nitrous acid or nitrite  salts  (Mirvish,  1975; Scanlan,  1975). N-nitrosa-




mines can also be formed by  transnitrosation whereby  other nitro  or




nitroso compounds serve  as the  nitrosating  agent  (Singer et al.,  1977;




Buglass et al., 1974). N-nitrosamines are commonly made by the reaction




of a secondary amine with sodium nitrite at acidic pH,  however, depend-




ing on the reactant and  catalysts that are  used,  N-nitrosation can also




occur at neutral or alkaline pH (Fine, 1979). Compounds known to  catalyze




N-nitrosation include formaldehyde (Keefer  & Roller,  1973), chloral




(Reefer & Roller, 1973), ozone  (Fine, I977a), and some  metal ions




(Keefer, 1976).









The amine fragment of the N-nitroso compounds can be  found in a large




variety of both man-made and natural products.  Secondary amines such




as dimethylamine, diethylamine  and morpholine are produced in large




quantities and are used  in both consumer and industrial products. These




products are, for example, used in agricultural chemicals, detergents,




rust inhibitors, rubber  additives, solvents, drugs, plastics, leather




tanning, textiles, cosmetics and synthetic  cutting and  grinding fluids.




Given the widespread use of  secondary ami.nes and  the  ever present nitro-




gen oxides of an industrial  society,  the likelihood of  N-nitrosamines




being found in these products  or in an ir.dustrial situation where these




compounds may occur together,  is high.









Recent advances in detection have made it possible to examine consumer




and industrial products  and  the environment for N-nitrosamines. It has
                                    124

-------
been found chat substantial  numbers  of  people  are  indeed exposed.  Levels




which have been determined in commercial  products  and  environmental  sam-




ples range from parts per billion to percent amounts.  Six human  popula-




tions have been identified as having a  potential exposure to  signifi-




cantly higher than background levels of carcinogenic N-nitrosamines.




They are, chemical workers at a  rocket  fuel  factory making unsymmetrical




dimethylhydrazine (UDMH) from NDMA (Fine et  al., I976a), agricultural




workers handling pesticides  contaminated with  nitrosamines (Fine,  et




al., 1977b), machinists using synthetic cutting and grinding  fluids  con-




taminated with N-nitrosodiethanolamine  (NDELA) (Fan et al., 1977b),  per-




sons using facial cosmetics contaminated with  N-nitrosodiethanolamine




(Fan et al., 1977a), rubber chemical workers exposed  to N-nitrosomor-




pholine (NMOR) (Fajen et al., 1979), and. leather tanners exposed to




N-nitrosodimethylamine in tannery air (Rounbehler  et al., 1979). The




probability that certain other occupations may involve exposure  to




N-nitrosamines is the basis of this NIOSH-sponsored study. While direct




evidence for the carcinogenicity of N-nitroso  compounds in man is




presently lacking, it is unlikely that  man alone will  be uniquely




resistant to their carcinogenic action.









                            STUDY DESIGN




During the study, a total of 51 on-site plant  visits were conducted  at




37 separate manufacturing plants. These plants represented 5  different




industries. The study was intended to determine if there was  worker  ex-




posure to N-nitrosamines. The basis for selecting  the  plants  that were




surveyed included:

-------
     -Known  or  suspected  use  of  N-nitroso  compounds




     -Known  use of  products  likely  to  contain N-nitroso compounds as




      impurities




     -Use of chemicals  that  could give rise to N-nitroso compounds




     -Epidemiological  data which, along with the possibility of worker




      exposure  to N-nitroso  compounds, suggested a higher than usual




      risk of worker exposure to an environmental carcinogen




     -Results of the study as it proceeded









In order to  illustrate the level of N-nitrosamines found in the fac-




tory environment, the  data from  the rubber and leather tanning industries




will be discussed.  Nineteen  factories  were visited, nine associated with




the manufacture of rubber and tire  products, and ten associated with




the leather trade.  The rubber and tire industry was chosen because a




variety of amines,  nitrosamines, nitrosd and nitro compounds are used




in various aspects of the manufacturing process, including N-nitroso-




diphenylamiae  (ND?hA), diethanolamine, and morpholine based accelerators




(Rubber Wor'.d,  N.Y., 1975).  In addition, rubber industry workers have




been identified in several epidemiological studies as suffering from




excess mortality from cancer of  a variety of organs including lung,




bladder, stomach, and prostate (McMichael et al., 1976). The leather




industry was chosen because  the  tanning process includes a dimethylamine




salt used in the dehairing processes (Walker et al., 1976). What little




data is available on leather workers show an increased

-------
tumor incidence of nasopharyngeal  and  bladder  cancers  among  shoe  and


bootmakers (Cole, et al.,  1972). Another  group stated  that  the  increased


buccal, larynx, nasopharyngeal  and bladder cancer  incidence  they  observed


among "leather industry operatives" (job  sites not specified)


might be related to the tanning process  (Viadana,  et al., 1976).  In both


cases the specific agents  responsible  for the  excess cancer  deaths have


not been clearly identified.






                          DATA COLLECTION


The laboratory apparatus,  including a  Shimadzu programmable  gas chromato-


graph, a high pressure liquid chromatograph,  fume  hood,  and  two Thermal

                     TM
Energy Analysers (TEA  ) are located inside a  fully equipped,  self-


contained mobile laboratory (Krull, et al.., 1978), which was parked


nearby each site.






The majority of the air samples were collected by  two  methods.  The first


method pulled air at a flow rate of between 1  and  2 fc/min through a


glass impin.'er containing 45 ml of IN  KOH. The second  method used a

            j^
Thermosorb/11  tube which consisted of  15  mm ID x 20 mm length  tubes con-


taining a mixture of magnesium silicate  and an amine trapping  (com-


plexing) agent and a nitrosating inhibitor. The air samples  were  col-


lected by drawing air through the  traps  at a constant  rate  of  from 1.5


to 6 A/rain for 5 to 200 minutes using  a  Bendix C115 pump or  a  10  l/min

                                        P
metal bellows air pump. The Thermosorb/N   tubes were used to absorb


N-nitroso compounds and to assess  the  presence of  nitrosating  agents


in the sampled air (Rounbehler, et al.,  1979). The nitrosating capacity
                                         127

-------
of the sampled air was estimated  by  measuring  the  amount  of N-nitrosomor-


pholine formed from the reaction  of  morpholine,  which was spiked  on  the

            ^
Thermosorb/N  adsorbant,  and  whatever  nitrosating  agent may have


been present in the air.  The  other trapping methods  used  in collecting


samples were dry cellulose fiber  traps,  alkaline cellulose fiber  and


Tenax GC cartridges.





Analytical techniques for the quantitative analysis  of NDMA in  air at


levels down to 0.001 yg/m  are available in the  literature (Fine, et


al., 1976b; Fine, et al., I977c). Analysis of  nitrosamines has  been


greatly simplified by the availability of the  TEA  (-Fine,  et al.,  I975b)


designed to be nitroso specific.  The TEA is used as  a detector  for both


gas chromatography (GC-TEA) (Fine, et  al., 1976a)  and high pressure


liquid chromatography (HPLC-TEA)  (Fine,  et al.,  l976c). The TEA sim-


plifies the analysis because  virtually no cleanup  of the  air  sample  is


required.





                              RESULTS


The industries surveyed by the NIOSH sponsored study were the azo dyes...


fish processing, fish meal, cutting  fluid manufacturers and users, rubber


and tanning. Airborne concentrations of  TEA responsive compounds  were


found in all the industries except fish  processing.  The dye industry


had airborne TEA responsive material as  high as  40 ug/m  , but they were


not identified. Air levels of NDELA  were detected  at 0.08 ng/m  in a


plant that used cutting fluids. A fish meal factory  was found to  contain
                                      128

-------
0.06 yg/m  of NDMA. The data generated on the leather and  rubber



industries will be discussed in further detail.
In an aircraft tire factory,  NMOR was found  to be present  at  levels



between 0.6 and 27 yg/m .  All 16 air samples which were collected  inside



the factory were positive for NMOR,  with the average  NMOR  level  being



4.85 yg/ra . The highest NMOR levels  were found where  rubber was  being



cured and extruded. Further evidence as to the identity of NMOR  was



obtained by combining the samples and identifying the NMOR by GC-high



resolution mass spectrometry (GC-MS).
In a chemical plant which was manufacturing rubber chemicals  three



N-nitrosamines were found: NMOR,  NDMA and NDPhA.  The NMOR levels varied


              3                             3
from 0.07 yg/ra  in the lunchroom to 4.6 yg/m  near the NDPhA  reactor.



The highest level of NDMA detected was 0.3 yg/m . NDPhA,  which  was  being



produced in the factory, was found to be present  at most  of the sites



sampled. The highest NDPhA level found was 47 yg/m . A dirt sample,



scraped frcm a staircase in the factory, contained 731 yg/g of  NMOR and



15,000 yg/g of NDPhA. The presence of NMOR in the dirt sample was con-



firmed by GC-MS.
Further investigations were made at five tire manufacturing plants to



determine if the levels in the aircraft tire plant were unique  or if



indeed nitrcsamines were ubiquitous in the rubber industry. Each plant



had NMOR levels similar to the aircraft tire plant (0.6 to 27 yg/m ),

                               3
however, one plant had 248 yg/m  of NMOR.  Table I summarizes the



N-nitroso compounds found in the rubber industry survey.




                                    121)

-------
Ac the leather tannery,  NDMA was  found  to  be  present  in  all  the  air




samples which were taken inside the  factory.  The  NDMA level  varied from




0.1 yg/m  in the lunchroom,  to 47 yg/m   inside  the  tannery,  adjacent




to the coloring and fat  liquoring process. The  average NDMA  level in




the 19 air samples which were collected was 17  yg/m .  Further  evidence




as to the identity of the NDMA was obtained by  combining the samples




and identifying the NDMA by  GC-MS. Air  samples  taken  at  the  doping area




were also found to contain NMOR at the  2 ug/m  level,  plus smaller




amounts of two unidentified  TEA responsive compounds.
Further investigation of the leather  industry was also made  to  determine




if N-nitrosodimethylamine was unique  to  this plant. Nine  other  tanneries




were surveyed for N-nitrosamines.  Four tanneries had  levels  ranging  from




0.03 - 10.8 yg/m ; N-nitrosamines  were not detected in the other  five




plants. Table II summarizes the  levels of nitrosamines found in the  first




tannery which was surveyed.
                            DISCUSSION




The question arises as to the  source  of  the  airborne N-nitrosamines  in




these two industries.









N-nitrosomorpholine was found  in both the  chemical  factory  and  the




factory where tires were being produced. In  the  chemical  factory,




N-nitrosoraorpholine was found  as an impurity in  the morpholine  (0.8  yg/g)




and in the product, bismorpholinecarbamylsulfonamide  (.4  to .7  yg/g),
                                  130

-------
which is used as an accelerator.  Also,  the  steam  condensate  contained




0.002 yg/g, possibly from the use of  morpholine as  a  corrosion  inhibitor




in the steam process equipment.








We believe this to be the first  report  of N-nitrosomorpholine as an air




pollutant. While the effects of  NMOR  by the inhalation  route have not




been tested, and its actions in  humans  are  unknown, studies  in  animals




by both oral and parenteral dosing have shown it  to be  carcinogenic to




a variety of species. Shank and  Newberne (1976) have reported  increased




incidence of liver angiosarcomas (15%), and lung  angiosarcomas  (9%) in




rats fed a diet containing 5 yg/g (5ppm) NMOR.








In the leather tannery, NDMA was found  at 0.5 yg/kg in  an aqueous  solu-




tion of dimethylamine sulfate which is  a depilatory agent in the unhair-




ing step. However, the NDMA impurity  in the dimethylamine sulfate  is




insufficient, in this plant, to account for the  level of NDMA  in the




plant environment. The causitive agent  or agents  responsible for the




total environmental  load of NDMA in the tannery has not been found;




however, it can be speculated that the  source of  the airborne  NDMA may




be due to gas phase nitrosation of airborne amines  by nitrogen oxides.




It can be further speculated that the source of  the airborne amines  in




these plants may be the dimethylamine sulfate and the dimethylamine




produced or released during the unhairing process.
                                      131

-------
The significance of the tannery findings  may be inferred from the results


of a recent study by Moiseev and Benemansky (1975).  They reported that


30 male Wistar rats breathing air containing 220 yg/ra  NDMA 24 hours/day


for 25 months showed an incidence of  malignant tumors of 83% in the ex-


posed as ccmpared to 13% in the control animals. These tumors were mainly


of the liver and kidney. However, it  is not possible to extrapolate the


animal data on NMOR or NDMA, as these compounds have not been identified


as human carcinogens.
                              SUMMARY


Under the NIOSH-sponsored contract,  a total  of  51  plant surveys were


conducted at 37 separate manufacturing plants.  The industries  investi-


gated were the fish processing,  fish meal, manufacturers and users of


cutting fluids, azo dye, leather and rubber.





NDELA, NM03, NDMA, NDPhA were found  in the  environmental air of several


factories. In a chrome tannery,  NDMA was identified as  high as 47 ug/ra

                    3
and NMOR at 248 ,ig/m  in a rubber tire plant.





This study has resulted in an increased understanding of mans" exposure


to preformed N-nitrosamines. It  is conceivable, from the information


that has been generated in this  study, that  nitrosamine exposure as large


as that in the tire and rubber industry exists  in  other industries not


yet surveyed.





NIOSH is continuing its research on  nitrosamines in the industrial envi-


ronment.

-------
                             REFERENCES
Buglass, A.J., Challis, B.C., & Osborne,  M.R,  (1974).  N-nitroso Com-
pounds in the Environment. In: Bogovski,  P.  &  Walker;  E.A.,  eds.,  Inter-
national Agency for Research on Cancer,  Lyon (IARC Scientific Publica-
tions No. 9), pp. 94-100

Challis, B.C., Edwards, A., Hunma,  R.R.,  Kyrtopoulos,  S.A.  &
Outran), J.R. (1977). Rapid Formation of  N-nitrosamines from Nitrogen
Oxides Under Neutral and Alkaline Conditions.  In:  Walker,  E.A.,
Castegnaro, M., Gricuite, L., & Lyle, R.E.,  eds.,  Environmental Aspects
of N-nitroso Compounds, Lyon, International  Agency for Research on
Cancer (IARC Scientific Publications No.  19),  pp.  127-142.

Cole, P-, Hoover, R. & Friedell, G.H. (1972).  Occupation and Cancer of
the Lower Urinary Tract. Cancer 29, 1250; Archeson, E.D.,  Cowdell, R.H.
& Jolles, B. (1970). Nasal Cancer in the Northamptonshire Boot  and Shoe
Industry. Br. Med. J.. 1, 385.

Oruckrey, H.R., Preussmann, R., Ivankovic, E.  & Schmahl, D.  (1967).
Organotrope Carcinogen Wirkungen Bei 65  Verschiedenen N-Nitroso-
verbindungen and BD-Ratten. Z. Krebsforsch., 69,  103-201.

Oruckrey, H., Preussmann, R. & Ivankovic, S.,  (1969).  N-nitroso Com-
pounds in Organotropic and Transplacental Carcinogenesis.  In: Biological
Effects of Alkylating Agents, Ann.  N.Y.  A:ad.  Sci., 163, 676-695.

Fine, D.H., Rounbehler, D.P., Pellizzari, E.D.. Bunch, J.E.,
Berkley, R.W., McCrae, J., Bursey,  J.T.,  Sawicki,  E.,  Krost, K. &
Detnarrais, G.A. (I976b). N-nitrosodimethylamine in Air. Bull. Env.
Contam. & Tox., 15, 639-746.

Fajen, J.M., Carson, G.A., Rounbehler, D.P., Fan,  T.Y., Vita, R.,
Goff, U.E., Wolf, M.H., Edwards, G.S., Fine, D.H., Reinhold, V. &
Biemann, K.  (1979). N-nitrosamines in the Rubber and Tire Industry.
Science. 20:>, 1262-1264.

Fan, T.Y., Goff, U., Song, L., Fine, D.H., Arsenault,  G.P.,  & Biemann,
K. (I977a). N-nitrosodiethanolamine in Consumer Cosmetics,  Lotions and
Shampoos. Food Cosmet. Toxicol. 15, 423.

Fan, T.Y., Morrison, J., Rounbehler, D.P., Ross,  R., Fine,  D.H., Miles,
W. & Sen, N.P. (I977b). N-nitrosodiethanolamine in Synthetic Cutting
Fluids, a Part Per Hundred Impurity. Science,  196, 70-71.
                                       133

-------
Fine, D.H. & Rounbehler,  D.P.  (I975a).  Trace  Analysis  of  Volatile
N-nitroso Compounds by Combined Gas  Chromatography  and Thermal  Energy
Analysis. J. Chromatogr., 109,  271-279.

Fine, D.H., Rufeh, F., Lieb,  D. & Rounbehler,  D.P.  (1975b).  Description
of the Thermal Energy Analyzer  (TEA)  for  Trace Determination of Volatile
and Non-volatile N-nitroso Compounds. Anal. Chem.,  47, 1188-1191.

Fine, D.H., Rounbehler, D.P., Belcher,  N.M. & Epstein, S.S.  (I976a).
N-nitroso Compounds: Detection  in Ambient Air. Science, 192, 1328-1330.

Fine, D.H., Rounbehler, D.P., Pellizzari, E.D.,  Bunch, J.E.,
Berkley, R.W., McCrae, J., Bursey, J.T.,  Sawicki, E.,  Krost, K. &
Demarrais, G.A. (I976b).  N-nitrosodimethylamine in  Air. Bull. Env.
Contam. & Tox. , 15, 639-746.

Fine, D.H., Rounbehler, D.P., & Silvergleid,  A.  (I976c).  Trace Analysis
of Polar and Nonpolar N-nitroso Compounds by  Combined  High Performance
Liquid Chromatography and Thermal Energy  Analysis,  Proc.  2nd Int.  Symp.
Nitrite Meat Prod., Tinbergen,  B.J.,  &  Krol,  B.,  eds., PUDOC, Zeist,
The Netherlands, p. 191.

Fine, D.H.  (1977a). Human Exposure to Preformed Nitrosamine Compounds.
In: Toxic Substances in the Air Environment.  Spengler, J.D., ed.,  Air
Pollution Control Assoc., Pittsburgh, 168-181.

Fine, D.H., Ross, R., Rounbehler, D.P-, Fan,  T.Y.,  & Morrison,  J.
(1977b). N-nitroso Compound Impurities  in Herbicide Formulation.
J. Agr. Fd. Chem., 25, 1416.

Fine, D.H., Rounbehler, D.P.,  Sawicki,  E. & Krost,  K.  (I977c).  Deter-
mination of Dimethylnitrosamine in Air  and Water by Thermal  Energy Anal-
ysis. Environ. Sci. Technol.,  11, 577-580.

Fine, D.H., (1979). N-nitroso Compounds in the Environment.       In:
Advances  in Environmental Science and Technology, Pitto,  J.N.,  ed., Wiley
Interscience,  New York, Vol.  9.

Keefer, L.K. & Roller, D.P. (1973).  N-nitrosation by Nitrite Ion in
Neutral and Basic Medium. Science, 181, 1245-1246.

Keefer, L.K. (1976). Promotion  of N-nitrosation by  Metal  Complexes.
In: Environmental N-nitroso Compounds Analysis and  Formation,
Walker, E.A.,  Bogovski, P. & Gricuite,  L., eds., Lyon, International
Agency for Research on Cancer (IARC  Scientific Publications No. 14),
pp. 153-159.

Krull, I.S., Fan, T.Y., Fine,  D.H.,  Young, R. & Egan,  E.  (1978).
A Mobile Laboratory for N-nitroso Analysis,  Research/Development Maga-
zine, p. 117.


                                         134

-------
Materials and Compounding Ingredients  for Rubber and Plastics,  Rubber
World. N.Y., 1975.

Miruish, S.S. (1975). Formation of  N-nitroso Compounds:  Chemistry,
Kinetics and In Vivo Occurrence.  Toxicol. Appl.  Pharmacol.,  31,  325-351.

Magee, P.N. & Schoental, R.  (1964). Carcinogenesis by Nitroso  Compounds.
Br. Med. Bull.. 20, 102-106.

Magee, P.N. & Barnes, J.M. (1967).  Carcinogenic  Nitroso  Compounds.
Adv. Cancer Res., 10, 163-246.

Magee, P.N., Montesano, R.,  Preussmann,  R.  (1976). N-nitroso Compounds
and Related Carcinogens. In:  Chemical  Carcinogens, Searle,  C.E.,  ed.,
ACS Monograph #173, American Chemical  Society, Washington,  D.C.

McMichael, A.J., Andjelkovic, D.A.  & Tyroler, H.A. (1976).  Cancer
Mortality Among Rubber Workers: An  Epidemiologic Study.  Annals  N.Y.
Academy Sci., 271, 125-137;  Peters, J.M., Monson, R.R.,  Burgess,  W.A.
& Fine, L.J. (1976). Occupational Disease in the Rubber  Industry, Env.
Health Perspectives, 17, 31-34; Monson,  R.R. & Nakano, K.K.  (1976),
Mortality Among Rubber Workers, Am. J. Epidemiology, 103,  284-296;
Mancuso, T.F. (1976). Problems  and  Perspective in Epidemiological Study
of Occupational Health Hazards  in the  Rubber Industry, Env.  Health
Perspectives, 17, 21.

Moiseev, G.E. & Benemanskii, V.V. (1975). Carcinogenic Activity of Low
Concentrations of Nitrosodimethylamine in Inhalation. Vopr.  Omkol.,  21,
107.

Ohshima, H. & Kawabata, T. (1977).  Mechanism of  N-nitrosodimethylamine
Formation from Trimethylamine and Trimethylaminoxide. In:  Walker, E.A. ,
Castegnaro, M., Gricuite, L. & Lyle, R.E.,  eds., Environmental  Aspects.
of N-nitroso Compounds, Lyon, International Agency for Research on Can-
cer (IARC Scientific Publications,  No. 19), pp.  143-154.

Rounbehler, D.P., Krull, I.S.,  Goff, U.E.,  Mills, K.M.,  Morrison, J.,
Edwards, G.S., Fine, D.H., Fajen, J.M.,  Carson,  G.A. and Reinhold, V.,
(1979). Exposure to N-nitrosodimethylaraine in a  Leather  Tannery. Fd.
Cosmet. Toxicol. (In Press).

Rounbehler, D.P., Reisch, J.W., Coombs,  J.R., &  Fine, D.H.  (1979).
Nitrosamine Air Sampling Sorbents Compared for Quantitative Collection
and Artifact Formation, presented at ASTM,  Boulder, Colorado,  August
7, 1979.

Scanlan, R.A. (1975). N-nitrosaraines in Foods. CRC Crit. Rev.  Food
Technology. 5, 357-402.
                                     135

-------
Shank, R.C. & Newberne,  P.M.  (1976).  Dose-response Study of the Carcino-
genicity of Dietary Sodium Nitrite and Morpholine in Rats and Hamsters,
Fd. Cosmet. Toxicol., 14, 1-8.

Singer, S.S., Lijinsky,  W. &  Singer,  G.M.  (1977). Transnitrosation: An
Important Aspect of the  Chemistry of  Aliphatic Nitrosamines.
In: Walker; E.A., Castegnaro, M., Gricuite,  L. ,  & Lyle,  R.E., eds.,
Environmental Aspects of N-nitroso Compounds,  Lyon, International Agency
for Research on Cancer (IARC  Scientific Publications No. 19),
Smith, P.A.S. &. Loeppky, R.N.  (1967).  Nitrosative Cleavage of Tertiary
Amines. J. Am. Chem. Soc.,  89,  1147-1157.

Viadana, E., Bross, I.D.J., Houten,  L.  (1976).  Cancer Experience of Men
Exposed to Inhalation of Chemicals or  to Combustion Products. J. Occup.
   nr, 48, 787,
Walker, P., Gordon, J., Thomas,  L.,  Ouellette,  R.  (1976).  Environmental
Assessment of Atmospheric Nitrosamines.  In:  MITRE  Technical Report
No-.-^-152, pp. 72-87.

-------
                                     Table I
                      Nitrosamines in Air Samples Collected
                         at Four Rubber Industry Plants
    Location
                                      NMOR,
NDMA,
vg/nf
NDPhA,
 Ug/m
NYPR*,
 Ug/nT
1. Tire Chemical Factory
DMA tank
Chemical storage
BMCS centrifuge
BMCS reactor
BMCS drier
BMCS discharge
NDPhA reactor
NDPhA decantor
Lunch room
Outdoors
2. Industrial Rubber Products
Solution area
Banbury machining
Batch off mill area
Office
3. Aircraft Tire Factory
Curing press
Extruder
Extruder
Warm-up and mixing
Cooling pool
Cutting area
Large tire curing
Small tire curing
Batchstock storage
Finishing and inspection
Office
Outdoors
4. Synthetic Rubber and Latex
Four areas
S. Other Tire Plants
Curing
Extruder
Warm-up mill
Calendar
Warehouse
Outdoors

0.9, 4
1.5
3.4
3.0
0.7
1.6
4.1, S
4.6, 3
0.07
0
Factory
0
0
0
0

2.2, 4
1.7, 2
27, 12
2.2, 1
3.3
2.2
7.1, 2
4.6
2.5
0.6
1.0
0
Factory
0

1.5, 1
22,0. 9
2.8, .
248
3.7
0

.6





.1
.9








.9
.4

.3


.6








.3, ,03
.2, 8,5
73





0.08
0.3
0.3
0
0.1
0.2, 0.1
0.05, 0.07
0
0

0.14
0.14
0.09
0.07

0
0
0
0
0
0
0
0
0
0
0
0

0

.24
2.0, .03

2.9
.38
0

0.8, 0.6
17
0.2
0.9
0.3
0
47, 0
12, 25
0.7
0

0
0
0
0

0
0
0
0
0
0
0
0
0
0
0
0

0

0
0
0
12.4 2.6

0
*N-nitrosopyrrolidine

-------
               Table 2;
N-Nitrosodimethylamine in Air Samples
    Collected in a Chrome Tannery

M
O
•H
•P
el
M
V
5-
4->
4)
^
in
o
•H
+>
at
ft
V
<§•
fr
o

Leather
Process NDMA (Area Air Samples)
Hide Restoring
Soaking
Fleshing
Unhairing
Bating
Pickling
Chrom Tanning
- Splitting
Re-Tanning,
Coloring, and
Fat- Liquoring
Color Set-Out
Drying
Conditioning
•Staking
Buffing
Doping Area
•Seasoning
Spray Finish —
Grading
Shipping
Lunch Room
Outside Plant












*











N.S.
izn

	 1

N.S.
3
1
N.D.

]



























••J

3

j



i













3



































13








•






|







]

"i









"1
— '


•

         10
15
20    25

  yg NDMA/ra
30
3
35   40   45   50.
55
            138

-------
                               Discussion
Dr. Kraybill (NCI):  I was intrigued when you mentioned fish processing.
What was your rationale there?

Mr. Fajen (NIOSH):  The natural amines that are in the fish.  That
is the one industry that did not contain any airborne nitrosamines.

Dr. Kraybill (NCI):  That is correct.

Mr. Fajen (NIOSH):  But your common fish sandwich that you can buy
commercially does contain measurable levels of dimethyl-nitrosamine.
The reason we went there was because of your natural amines that are in
fish.

Dr. Kraybill (NCI):  From where do you get your nitrite source?

Mr. Fajen (NIOSH):  In the environment that you are working in, in the
plant itself, one of the sources of nitrosation that we are interested
in is your propane-powered forklifts.  Especially in the tanning
industry, they use an awful lot of propane-powered forklifts.

Unidentified Speaker:  How specific was your method for detecting
other nitroso compounds, other than the ones you may have been looking
for initially?  Did you detect other peaks and attempt to identify the
substances that appeared?

Mr. Fajen (NIOSH):  The mobile laboratory contains a library of
standards.  We can only detect those that we have standards on board
for.  We carry about ten different standards on each survey.  When I say
we found nitrosomorpholine, we would do a screening for many different
nitrosamines and  if one comes out at a specific retention time, we would
try to match it with the standard.  We also have GC and HPLC capabilities
In both these industries, rubber and leather, the samples were confirmed
by GC mass spectrometry.  So we are not riding on the Thermal Energy
Analyzer alone.
                                         13;)

-------
          MORTALITY AND INDUSTRIAL HYGIENE STUDY OF


         WORKERS EXPOSED TO POLYCHLORINATED BIPHENYLS
                        David P-  Brown


                          Mark Jones
       U.S. Department of Health, Education and Welfare

                               c
                    Public Health Service


                  Center for Disease Control


    National  Institute  for Occupational Safety and Health


Division of Surveillance, Hazard  Evaluations and Field Studies


                    Cincinnati,  Ohio 45226
                                   11I)

-------
                            ACKNOWLEDGEMENTS
The authors.would like to express  their  appreciation for the work of
many individuals who helped  to successfully complete this study
including the — guidance of Joseph  Wagoner, Richard Lemen and Richard
Waxweiler; the assistance of the clerical and secretarial staff in the
Biometry Section; the data entry and data analysis provided by the
Southwest Ohio Regional Computer Center; the cooperation of the
companies and labor unions chosen  for  the study; and for the
information provided by the  New York State Department of Health.
                                        141

-------
                                ABSTRACT







Due to the demonstrated toxic  effects  from polychlorinated  biphenyls



(PCB's) on exposed animals,  the National  Institute  for  Occupational



Safety and Health (NTOSH)  conducted  a  retrospective cohort  mortality



study of two worker populations who  manufactured  electrical



capacitors.  The two study cohorts included  968 workers from  Plant  1



and 1599 from Plant 2 who  were employed for  at least three  months  in



areas of the plants where  PCB's were used.







The vital status of over 97 percent  of the two cohorts  was  determined



as of January 1, 1976 and  38,890 person-years were  accumulated.



All-cause mortality was lower  than expected  (163  obs. vs. 174 exp.) as



well as all cancer mortality (39 obs.  vs.  40.6 exp.).   Rectal  cancer



(4 obs. vs. 1.07 exp.) and liver cancer (3  obs. vs.  0.88 exp.)



excesses were noteworthy although not  statistically significant.   Tn



one of the plants, the observed mortality due to  cirrhosis  of the



liver was also elevated.  The  results  are discussed with regard to  the



detailed industrial hygiene surveys  conducted in  each plant.

-------
                              INTRODUCTION

Polychlorinated biphenyls (PCB's) are a class  of compounds  composed  of
biphenyl molecules with a varying number of substituted  chlorine
atoms.  In commercially prepared PCB's, the weight-percent  of chlorine
has varied between 21 and 68 percent.  In some preparations,  there has
also been some degree of contamination by chlorodibenzofurans. (1)

The primary use of PCB's has been as a liquid  insulating material  in
electrical capacitors and transformers, and the greatest potential for
occupational exposure has been in the manufacturing and  repair of
these components.  PCB's have also been used in heat exchange units,
hydraulic systems, vacuum pumps, gas transmission turbines,
plasticizers, adhesives, pesticide extenders,  paints, and carbonless
copying papers.

As of 1971, PCB's were sold only for use in closed systems.  According
to the Toxic Substances Control Act of 1976, specific rules and
regulations were promulgated to limit the manufacture and use of
PCB's.  This Act stipulated that all U.S. production of  PCB's end
January 1, 1979 and that all U.S. sale and distribution  of PCB's end
July 1, 1979.  However, continual exposure to PCB's will occur among
workers who maintain transformers and capacitors, and among the
general population mainly through contaminated food.

-------
During the past few years,  there  has  been  a  great  deal  of  interest  in



studying the health effects among individuals  exposed to PCB's.   This



interest has been stimulated by:   (a)  the  tendency for  PCB's  to



accumulate in tissues and in certain  organs,  (2, 3) (b) the stability



of PCB's and their persistence in the environment, (4,  5)  and (c) the



demonstrated long term effects in exposed  laboratory animals  (6-13)  -



including liver tumors and  other  liver diseases.   Much  of  this



interest was expressed at the National  Conference  on Polychlorinated



Biphenyls in November, 1975 (14)  and  the toxicity  of PCB's has been



extensively reviewed in the NTOSH Criteria Document on  PCB's.  (15)







Tn order to determine whether or  not  past  occupational  exposure  to



commercially produced PCB's has caused any long term health effects,



NTOSH initiated an epidemiologic  study among workers in two capacitor



manufacturing plants.  Tn conjunction with this study,  detailed



industrial hygiene surveys  were also  conducted by  NTOSH.







                       DESCRTPTTON OF FACTLTTTES







The two facilities chosen for the study were selected after



preliminary walk-through  surveys  were conducted at numerous types of



plants where PCB's were used.  Both of the plants  manufacture



electrical capacitors.  These plants  were  selected because of their



large workforce, the early dates  (1938 & 1946) at  which PCB's were
                                      144

-------
first used at these plants,  the potential  for  exposure  to  PCB's with
little potential for exposure to other  known toxic contaminants, and
the availability of records  necessary to identify individuals  to be
included in the study population.  At the time the study was initiated
both plants were still using PCB's.   Plant 1  is located in New York
State and is divided into two manufacturing facilities  within  close
proximity.  One facility produced small industrial capacitors  using
PCB's since 1946 and the other produced large  PCB filled power
capacitors since 1951.  The  type of PCB's used has varied  over the
years from "Aroclor" (Aroclor is a Monsanto tradename)  1254 (54
percent chlorine) to 1242 (42 percent chlorine) to 1016 (41 percent
chlorine).  Several other kinds of oils were  used, but  in  a small
number of capacitors.  These oils included castor oil,  dibutyl
sebaceate, diotylthai ate and mineral oil.

Plant 2 is located in Masachusetts where the  use of  PCB's  to
manufacture capacitors started in  1938.  This plant also  changed the
type of PCB's used from "Aroclor" 1254  to 1242 to 1016. Until 1972,
other types of capacitors which did not contain PCB's were made at
                                     c
this plant including mica, electrolytic and tubular.  Castor oil was
used in lieu of PCB's to produce the large power capacitors at this
plant.
                                        14.T)

-------
Both plants assemble the capacitors using the same general  techniques,
whether they are the small or large types.  The following is a brief
description of the assembly process:

A.  Winding and Pre-assembly - The inner components of the capacitor
    are made of paper, foil and sometimes plastic film, wound
    together, which are subsequently loaded into metal casings.  This
    job is done in an enclosed dust-free room where there is minimal
    exposure to PCB's.

B.  Impregnation - The pre~assembled capacitors are filled or
    impregnated with the PCB's.  Within this area there is  potential
    for exposure to PCB's.

C.  Final Assembly - The tops of the capacitors are closed  using
    various techniques - crimpping (rubber stoppers) or soldering,
    which involves some exposure to PCB's.  The capacitors  are washed
    to remove excess PCB's by running them through a detergent wash or
    a degreaser such as trichloroethylene.  Finally, they are sent
    through the final operations involving drying, testing,  and
    painting.

Other areas of importance where there is potential exposure to PCB's
in the plants, include the laboratory and the area where rejected

-------
capacitors are rebuilt.  Approximately 10  percent  of  the two
workforces have been employed in jobs  where  there  has been  potential
exposure to PCB's.

Historically, the workforce at Plant 1 has been  approximately 50
percent white males and 50 percent white females.   Plant 2  has had a
less homogeneous workforce with two-thirds being female and reflects
the general ethnic make-up of the area,  which  is largely Cape Verdean
and Portuguese.

                                METHODS

A retrospective cohort study of mortality  was  conducted to  determine
whether or not individuals occupationally  exposed  to  PCB's  have
experienced any increase in cause-specific mortality.  The  study
cohorts were defined as all workers who accumulated at least three
months of employment any time between  January  1, 1946 and January 1,
1976 for Plant 1 and January 1, 1940 and January 1, 1976 for Plant 2,
in areas of the plants where there was a potential  for exposure to
PCB's.  These exposed jobs were designated by  the  companies and
verified by the labor unions, and through  the  industrial hygiene
surveys.
                                          14?

-------
An effort was made to determine the vital  status  (alive  or deceased)



of each individual in the cohorts  as of  January 1,  1976.  Vital  status



was determined through records  maintained  by  Federal  and  State



agencies, including the Social  Security  Administration,  state motor



vehicle registration, and state vital  statistics  offices.  For  those



individuals who could not be located through  these  sources,  U.S.



Postal Mail Correction Services and other  follow-up searches were



used.  For all those who were known to be  deceased, death certificates



were requested and causes of death were  interpreted by a qualified



nosologist according to the International  Classification  of  Diseases



(TCDA) in effect at the time of death  and  then converted  to  the  7th



Revision of the TCDA.  Those who had an  unknown vital status were



assumed to be alive as of January  1, 1976  so  that the true risk  of



mortality was not overestimated.  Those  who died  after January  1, 1976



were considered alive for purposes of analysis.







Person-years were accumulated for  each worker starting at the point in



time when three months of employment in  exposed jobs was completed and



ending at the date of death or  the study end  date  (1/1/76), whichever



occurred first.  Using a modified  life table  computer program similar



to that described by Cutler, (16)  the person-years  for each  cohort



were combined into five-year calendar and  five-year age  time periods



and multiplied by the corresponding U.S. white male (for male cohort



members) and U.S. white female  (for female cohort members)
                                          148

-------
cause-specific mortality rates to yield the expected  number  of
deaths.  Person-years were additionally distributed by five-year
exposure and five-year latency (number  of years  from  date  first
employed in exposed jobs to date of death or study end date)
categories.  Observed and expected cause-specific deaths were compared
and differences were tested using the Poisson distribution.

The detailed industrial hygiene surveys included personal
time-weighted air samples of employees  from selected  job titles,  as
well as area air samples.  Tn both plants, samples were taken for
PCB's  (Aroclor 1016), trichloroethylene, lead, tin, and zinc.   Tn
addition, samples for toluene, methyl isobutyl ketone (MIBK), aluminum
and iron were taken at Plant 1.  These  surveys were designed to
characterize the exposures occurring at the time of the survey  and may
not represent exposures of previous years, especially those  of  Plant  1
where  exposures may have been reduced because of new  production
techniques which had recently been initiated.

                                 RESULTS

There were a total of 2,567 workers who met the  definition of the
study  cohort.  Table 1 gives a breakdown of vital status ascertainment
and the number of person-years within each sub-cohort. The  vital
status ascertainment was more than 97 percent complete.

-------
The possibility of missing records  from the  personnel  files  that  were



used to assemble the Plant 1 cohort was questioned  at  the  initiation



of the study.  Tn an effort to determine whether  or not  eligible



workers were missing from the plant 1  cohort,  a validity check was



conducted by the New York State Department of  Health (personal



communication from Phil  Taylor, NYSDH, April,  1980), similar to that



described by Marsh et aK (17).  Social Security  Administration (SSA)



quarterly earning statements (SSA form 941)  from  1945-1965 were



obtained and compared to the names  appearing on the microfilmed



personnel records which  were used to assemble  the cohort.  The results



of this comparison yielded 35 additional  workers  (3.5  percent of



cohort) who should have  been included  in the plant  1 study cohort.



The vital status of these missing workers is not  known at  this time,



however, the NYSDH is currently trying to ascertain this information.



Nevertheless, the validity check confirmed that only a small  portion



of the total population  at risk was missing  from  the study cohort and



the results should not be biased.  According to plant  officials,  there



was no reason to believe that the personnel  file  system  at Plant  2 was



missing records, and it  appeared from  our inspection that  the



personnel file system had been maintained intact.







Tables 2 and 3 summarize the number of deaths  observed from  the study



cohorts and the number of deaths expected.   The all-cause  mortality is



lower than expected in each cohort, with an  SMR  (SMR = observed
                                        150

-------
deaths/expected deaths x 100)  of 99 (73  obs.  vs.  73.2 exp.) for Plant
1 and an SMR of 89 (90 obs.  vs.  100.84 exp.)  for  Plant 2.  These SMR's
may be affected by the "healthy  worker effect".  (18)  There was no
increase in observed mortality for  any of  the major causes of death
listed in Table 2.

Table 3 lists the observed and expected  number of deaths by specific
cancer cause and for cirrhosis of the liver.  When both cohorts were
combined, the observed number  of deaths  was more  than that expected
for cancer of the rectum (4 obs. vs.  1.07  exp.)  and liver cancer -
ICD=155, 156A (3 obs. vs. 0.88 exp.). The only  statistically
significant difference (at p<0.05)  in observed versus expected deaths
occurred in females from Plant 2 for  cancer of the rectum (3 obs. vs.
0.46 exp.; p<0.05).  For both  cohorts combined,  there were 6 deaths
due to cirrhosis of the liver, while 5.47  were expected.  Five of
these cases were from the Plant  2 cohort while 3.1 were expected.
According to hospital reports, at least  three of the six persons who
died of cirrhosis of the liver were known  to  have consumed alcohol on
a regular basis.

The relationship between latency and the mortality from all cancer,
cancer of the rectum, liver cancer, and  cirrhosis of the liver is
shown in Table 4.  For "all cancer" there  is  no  apparent pattern in
either cohort.  However, for cancer of the rectum, there is a slight
                                      151

-------
increase with an increase in the latency period.   The risk of



mortality due to cirrhosis of the liver does not  show a consistent



increase with an increase in the latency periods, however there is a



greater risk after a 20 year period.







The relationship between these same causes of mortality and length of



employment in PCB exposed areas of the plants is  given in Table 5.  As



indicated in the table there is no increase in mortality with



increasing lengths of exposure, except for cirrhosis of the liver,



however; the numbers in this comparison are small.







The industrial hygiene survey results of area and personal sampling



for PCB's (Aroclor 1016) are summarized in Tables 6 and 7-  Due to



differences in the production processes, the results by specific jobs



or work areas are not comparable between the two  plants.   However,



relative comparisons can be made and  the range of concentrations



observed in Plant 1 were lower than those in Plant 2.   In Plant 1, the



time weighted average (TWA) personal  air samples  ranged from


       3            3
24 pg/m  to 393 pg/m  and the TWA area air samples ranged from



3 pg/m  to 476 pg/m .  In plant 2, the TWA personal air samples


                    3             3
ranged from 170 pg/m  to 1260 pg/nr and the TWA area air samples


                   3            3
ranged from 50 pg/nr to 810 pg/nr.  The current OSHA standard and



ACGIH TLV for chlorodiphenyl (42 percent chlorine)  is 1000 pg/m^.



There is no current OSHA standard of  ACGTH TLV for Aroclor 1016.

-------
Trichloroethylene was measured  near  the degreasers  in both plants.
Out of eleven area air samples  from  Plant 1, all were below 35 ppm
except for two which measured 195  ppm  and 321 ppm.  At Plant 2, three
area air samples were taken which  ranged from 53.4  ppm to 77.5 ppm.
The OSHA standard for trichloroethylene is  100 ppm  based on a 8 hour
time weighted average.

Even though most exposures to trichloroethylene were usually below the
TLV, an attempt was made to exclude  workers from the study who were
employed around the trichloroethylene  degreasers.

Area air samples were measured  for tin, lead and zinc near the
soldering operations.  There  were  no detectable levels for tin at
either plant.  Out of four samples collected for lead and zinc at
Plant 1, lead was detected in one  sample at a level of 12 yg/m ,
zinc was detected on two samples at  levels  of 8 and 24 pg/m .  At
Plant 2, fifteen samples were collected for lead and zinc.  All but
one of these samples showed no  detectable  levels for lead, the one
                               3
detectable sample was 41.2 yg/m .  Six of the fifteen samples found
                                    f              3
concentrations of zinc ranging  from  2.3 to  94.1 yg/m .  The current
OSHA standard for lead and zinc oxide  (reported as  zinc) are
       o           o
50 pg/nr and 5 mg/rrr respectively.
                                        153

-------
Both personal  and area samples  were  taken around the welding


operations for measuring aluminum  and  iron at  Plant 1.  The aluminum


samples ranged from non-detectable to  233 yg/m  and the iron samples


from 47 yg/m3  to 123 yg/m3.   The ACGIH TLV for aluminum


(Al?0,) is 10  mg/m3 and the  current  OSHA standard for  iron oxide


(measured as iron) is 10 mg/m .
Twelve personal samples were  collected for toluene and MTBK during


painting operations at Plant  1.   Toluene concentrations ranged from


0.48 to 22 ppm and MTBK ranged  from  2 to 5 ppm.  The current OSHA


standard for toluene is 200 ppm  and  100 ppm for MTBK.




Although the exposures to PCB's  at the time of the surveys (Plant 1 -


April, 1977; Plant 2 - March,  1977), were relatively higher in Plant


2, the, historic levels of exposure  may have been more equivalent.  Tt


is these exposures that occurred 20  to 30 years ago that are more


relevant when considering the occupational cancer risk among the study


cohorts.  The PCB mixtures used  during these time periods were Aroclor


1254 and 1242, whereas, Aroclor  1016 was first used in 1971.  Tn

                                    c
addition, several different stabilizers have been added to the PCB's


(1 percent or less by weight)  used at Plant 1 since the early 1960's.


These include potential carcinogens  such as diglyceride


ether-disphenol-a and more recently, vinyl cyclohexene dioxide.  Tt is


not known which stabilizers have been used at Plant 2.
                                       154

-------
                               DISCUSSION

There are few previous epidemiologic studies that have examined the
long term health effects of humans exposed to PCB's.   Individuals
poisoned by rice oil heavily contaminated with PCB's  (Yusho Disease)
have been studied extensively years after the incident took place in
Japan in 1968. (19, 20)  However, the rice oil contaminant also
contained polychlorinated dibenzofurans,  and quarter  phenyls in higher
concentrations than that found in commercially prepared PCB's.   A high
prevalence of skin and eye conditions were noted in the Yusho
patients.  In addition, there were clinical  and laboratory findings
that included changes in the microanatomy of liver cells and a
decreased concentration of bilirubin in the serum of  these
individuals. (21, 22)

Early reports regarding the health effects from occupational exposure
to PCB's include chloracne (23), digestive disturbances, eye
irritation, liver injury and impotence. (24, 25)  Most of these
findings have been reported as case histories.

Tn a recent study of volunteers conducted by the Mount Sinai School of
Medicine (26), 326 workers who were employed at Plant 1 were
examined.  The most prevalent symptoms noted were dermatological, and
those of the central nervous system.  There was a low prevalence of

-------
abnormal  liver findings  on  physical  examination.  However, a subgroup
exposed to PCB's were found to  have  liver enzyme changes different
from those of a normal,  non-exposed  group.   Tn addition, abnormal SGOT
levels were associated with plasma levels of PCB's.  There was a
relatively high prevalence  of decreased  lung capacity among a subgroup
of 243 workers tested. (27)

In a preliminary report, Bahn (28) reported  an increase in deaths due
to malignant melanoma (2 obs. vs. 0.04 exp.) and cancer of the
pancreas among 51 research  and  development employees and 41 refinery
plant employees at a New Jersey petrochemical facility.  These
individuals were considered to  have  had  some exposure to Aroclor 1254
during various periods between  1949  and  1957, along with exposure to
other toxic and potentially carcinogenic compounds.

Tn a summary of case histories  (G. Roush.  Written communication to
NTOSH, September, 1976)  among approximately  300 workers employed in
the manufacturing of PCB's, no  malignant melanomas or pancreatic
cancers were observed.  However,  among the death certificates of 50
                                     (
former workers at this manufacturing facility, seven cases of lung
cancer were observed whereas 2.7  cases were  expected.  The findings
were preliminary and were not adjusted for age or smoking.
                                      156

-------
The previously reported findings  of an  increased risk for mortality
due to malignant melanoma,  cancer of the  pancreas, and lung cancer
among workers exposed to PCB's  were not corroborated  in the present
study.  There were no observed  deaths due to malignant melanoma and
only 1 observed death from pancreatic cancer while 1.77 were
expected.  There were 7 observed  deaths from respiratory system
cancer, whereas 7.69 were expected. The only categories of cancer in
which the number of observed deaths were  greater than expected were
for cancer of the rectum and cancer of  the  liver and  only slight
increases for cancer of intestine except  rectum, and breast cancer.
When both cohorts and sex groups  were combined  none of the excesses
were statistically significant  at p<0,05.  However, the excess in
liver cancer is noteworthy because it is  consistent with the
toxicology data observed in laboratory  animals  exposed to PCB's, where
effects have been noted in  the liver (6-13).   The slight increase in
deaths due to cirrhosis of the  liver in the Plant 2 cohort is also
consistent with the notion that PCB's have  a toxic effect on the liver.

In most occupational health studies where cancer mortality is being
                                     t
assessed, latency is an important variable; the hypothesis being that
there is an increased risk of mortality once a  certain time period
after initial exposure has elapsed.  Tn this study, this hypothesis is
difficult to examine due to the small number of deaths.  None of the
causes of death analyzed according to latency clearly demonstrated
                                          15?

-------
this association.   Rectal  cancer  showed  a  slight  increase with  an
increase in latency and cirrhosis of  the liver showed an increase  in
risk with an increase in latency  after 20  years.

There was no relationship  between increasing  lengths of employment  in
PCB exposed jobs and the risk  of  mortality due to cancer or  cirrhosis
of the 1iver.

When the cancer mortality  is examined by plant,  it  is evident that
most of the excesses occur in  plant 2, especially among the  female
group.  This finding may be related to heavier exposures to  PCB's at
plant 2 as indicated by the industiral hygiene results.  Tn  addition,
there was an opportunity for earlier  exposures at plant 2, potentially
allowing for a longer latency  period.  However,  this difference  in
mortality may be a function of the size  of the cohorts (plant 1  only
has half the number of person-years as plant  2) and thus simply  be a
statistical quirk.

A potential confounding variable  or interaction variable in  this study
is the possible effect of  alcohol  ing^stion on the observed  increase
(at Plant 2) in mortality  from cirrhosis of the  liver.  However, this
cannot be properly assessed in the present study  since not enough is
known about the ingestion  of alcohol  among the entire study  cohort.
                                       158

-------
                              CONCLUSIONS

Due to a relatively small  number of deaths,  conclusions  drawn  from  the
results of this study are tentative.

Despite these study limitations, observed  excesses  for liver cancer
and cirrhosis of the liver are consistent  with  previously reported
findings on experimental animals exposed to  PCB's,  and suggest that
there may be an association between these  causes  of death and
occupational exposure to PCB's (Aroclor 1254 and  1242).   The observed
excess in cancer of the rectum related to  PCB workers was unexpected
and needs further investigation.

-------
                              REFERENCES

1. Hutzinger,  0.,  Safe,  S.,  Zitko,  V.  1974.   The  Chemistry of  PCB's.
   Cleveland,  The  Chemical  Rubber  Co.  Press.

2. Yobs, A.R.  1972.   Levels  of polychorinated biphenyls  in adipose
   tissue of the general population of the  nation.   Environ. Health
   Perspect.,  Experimental  issue No.  1:   79-81.

3. Price, H.A.,  Welch,  R.L.  1972.   Occurrence of  polychorinated
   biphenyls in  humans.   Environ.  Health  Prospect.,  Experimental
   issue No. 1:   73-78.

4. Jensen, S., Johnels,  A.G.,  Olsson,  M.  and  Otterlind,  G.  1969.   DDT
   and PCB in  marine animals  from  Swedish waters.   Nature,
   224:247-250.

5. Jensen, S.  1972.   The PCB  story, Ambio,  1:123-131.

6. Von Wedel,  H.,  Holla, W.A., Denton,  J. 1943.   Observations  on  the
   toxic effects resulting  from exposures to  chlorinated naphthalene
   and chlorinated phenyls  with suggestions for prevention.  Rubber
   Age 54:419-26.
                                      160

-------
 7.    Miller,  J.W.  1944.   Pathologic changes  in animals exposed to a
      commercial  chlorinated diphenyl.   Public Health Rep. 59:1085-93.

 8.    Bruckner,  J.V.,  Khanna,  K.L., Cornish,  H.H. 1974.
      Polychlorinated  biphenyl-induced alteration of biologic
      parameters  in the rat.   Toxicol. Appl.  Pharmacol. 28:189-99.

 9.    Kimbrough,  R.D., Linder,  R.E., Gaines,  T.B. 1972.  Morphological
      changes  in  livers of rats fed polychlorinated biphenyls.  Arch.
      Environ. Health  25:354-£4.

10.    Kimbrough,  R.D., Linder,  R.E., Burse, V.W., Jennings, R.W.
      1973.  Adenofibrosis in  the  rat  liver — With persistence of
      polychlorinated  biphenyls in adipose tissue.  Arch. Environ.
      Health 27:390-95.

11.    Kimbrough,  R.D., Linder,  R.E. 1974.  Induction of adenofibrosis
      and hepatomas of the liver in BALB/cJ mice by polychlorinated
      biphenyls (Aroclor 1254).  J. Natl. Cancer Tnst. 53:547-52.

12.    Allen, J.R.,  Abrahamson,  L.J., Norback, D.H. 1973.  Biological
      effects  of polychlorinated biphenyls and triphenyls on the
      subhuman primate.  Environ.  Res. 6:344-54.
                                       161

-------
13.   Vos,  J.G., Notenboom-Ram,  E.  1972.   Comparative toxicity study



      of 2,4,5,2',4',5'-hexachlorobiphenyl  and a polychlorinated



      biphenyl mixture in rabbits.   Toxicol.  Appl.  Pharmacol.



      23:563-78.







14.   Proceedings of the National  Conference on Polychlorinated



      Biphenyls.  EPA, Office of Toxic Substances,  Washington, D.C.,



      March, 1976.







15.   NTOSH, CDC, PHS, DHEW. 1977-   Criteria for a  recommended



      standard...Occupational Exposure to polychlorinated biphenyls



      (PCB's).







16.   Cutler, S.J., Ederer, F.  1958.   Maximum Utilization of  the life



      table methods in analyzing survival.   J. Chron. Dis.,  8:699-709.







17.   Marsh, G.M., Enterline, P.E., 1979.  A Method for Verifying the



      completeness of cohorts used in  occupational  mortality studies.



      JOM, In Press.







18.   McMichael, A.J., Haynes, S.G., Tyroler, H.A.   1975.



      Observations on the evaluation of occupational mortality data.



      J. Occup. Med.  17:128-131-

-------
19.   Kuratsune, M.t Masuda, Y., Nagayama,  J.  1976.   Some of the
      Recent Findings Concerning Yusho,  in  Proceedings  of the National
      Conference on Polychlorinated Biphenyls,  November 19-21,  1975*
      Chicago, EPA-560/6-75-004.  U.S.  Environmental  Protection
      Agency, Office of Toxic Substances,  1976,  pp.  14-29.

20.   Urabe, H. 1974.  [Foreward, The fourth  reports  of the study of
      "Yusho" and PCB.]  Fukuoka Acta Med.  65:1-4,  (Jap).

21.   Hirayama, C., Trisa, T.,  Yamamoto,  T.  1969.  [Fine structural
      changes of the liver in a patient  with  chlorobiphenyls
      intoxication.]  Fukuoka Acta Med.  60:455-56  (Oap).

22.   Hirayama, C., Okumura, M., Nagai,  J., Masuda,  Y.  1974.
      Hypobilirubin in patients with polychlorinated  biphenyls
      poisoning.  Clin. Chim. Acta. 55:97-100.

23.   Meigs, J.W., Albom, J.J., Kartin,  B.L.  1954.  Cloracne from an
      unusual exposure to Aroclor.  JAMA 154:1417-18  .

24.   Schwartz, L. 1936. Dermatitis from synthetic  resins and waxes.
      Am. J. Public Health 26:586-92.
                                        163

-------
25.   Drinker, C.K., Warren, M.F., Bennett, G.A. 1937- The problem of



      possible systemic effects from certain chlorinated



      hydrocarbons.  J. Tnd. Hyg. Toxicol. 19:283-99.







26.   Fischbein, A., Wolff, M.S., Lilis, R., Thornton, J., Selikoff,



      T.J. 1979.   Clinical findings among PCB-exposed capacitor



      manufacturing workers.  New York Academcy of Sciences.



      320:703-715.







27-   Warshaw, R., Fischbein, A., Thornton, J., Miller, A., Selikoff,



      I.J. 1979.   Decrease in Vital  Capacity in PCB-Exposed Workers in



      a Capacitor Manufacturing Facility.  New York Academy of



      Sciences.  320:277-284.







28.   Bahn, A.K., Rosenwaike, T., Herrmann, N., Grover, P., Stellman,



      J., O'Leary, K.:  Melanoma after exposure to PCB's.  N. Engl. J.



      Med. 295:450, 1976.

-------
            Known to be alive
            Known to be deceased
            Unknown vital status
            Total
            Person-Years
                                                    Table 1
                                          Vital Status of PCB Workers
                                           Plant 1                   Plant 2
                                   Males  Females  Total
                         Males  Females  Total
                           633     836    1,469
                            28      62       90
                            14      26       40
                           675     924    1,599
7,800   5,181    12,981    9,191  16,718   25,909
520
55
8
583
360
18
7
385
880
73
15
968
                                                      Grand
                                                      Total
 2,349
   163
    55(2%)
 2,567
38,890
cn
en

-------
                                        Table 2
                    Observed and Expected  Deaths  (0/E) According to
                            Major Causes Among PCB Workers
    Cause of Death
(7th Revision ICD No.)

All Malignant Neoplasms
  (140-205)

Nervous System
  (330-334, 345)

Circulatory System
  (400-468)

Accidents
  (800-962)

All Other Causes

All Causes
       Plant 1
     Plant 2
  Males   Females
Males   Females
Total    (SMR)
 9/ 8.97   4/ 6.62    3/ 6.13  23/18.90    39/ 40.62   (96)


 3/3.08   1/1.92    2/1.78   5/5-52    11/12.30   (89)


26/22.05   7/6.64    14/13.87  13/19.30    60/61.86   (97)


 7/5.86   1/1.06    3/7.26   2/3-46    13/17.64   (74)

10/12.06   5/ 4.94    6/ 9.53  19/15.09    40/ 41.62   (97)

55/52.02  18/21.18    28/38.57  62/62.27   163/174.04   (94)

-------
                                        Table 3
            Observed and Expected Deaths (0/E) According to Specific Cancer
                   Causes and  Cirrhosis  of  the  Liver Among  PCB Workers
     Cause of Death

 (7th Revision ICD No.)

All Malignant Neoplasms
  (140-205)
  Stomach
    (151)
  Intestine exp. Rectum
    (152, 153)
  Rectum
    (154)
  Biliary Pass   Livej^
    Liver not specified
    (155, 156A)
  Pancreas
    (157)
  Respiratory System
    (160-164)
  Breast
    (170)
  Lymphatic & Hematopoietic
    (200-205)
  Other
Cirrhosis of Liver
  (581)
    Plant 1
     Plant 2
Total
Males
9/8.97
0/0.48
1/0.79
1/0.28
1/0.19
0/0.50
5/3.15
0/1 .00
1/3.58
Females
4/6
0/0
0/0
0/0
0/0
1/0
VO
1/1
0/0
1/2
.62
.20
.67
.16
.14
.24
.66
.83
.50
.22
Males
3/6
1/0
0/0
0/0
0/0
0/0
0/2
0/0
2/1
.13
.28
.51
.17
.11
.32
.14
.83
.77
Females
23/18
o/
3/
3/
2/
O/
V
6/
2/
6/
0
1
0
0
0
1
4
1
6
.90
.58
.93
.46*
.44
.71
.74
.94
.56
.56
39/40
V
4/
4/
3/
I/
1
3
1
0
1
7/ 7
11 6
21 3
10/14
.62
.54
.90
.07*
.88
.77
.69
.77
.89
.13
(96)
(65)
(103)
(374)
(341)
(56)
(91)
(103)
(51)
(71)
1/1.66  0/0.71
2/1.22   3/ 1.88     6/ 5.47   (110)
* p<0.05

-------
c:
GO
                                                    Table 4

                               Observed and Expected  Deaths According to Latency!
                                               Among PCB Workers

                                                T.  All Cancers

0
6
3
4

0
0
1

1
0
n

i
0
0
Plant 1
E
4.83
6.07
4.65
TT.
0.13
0.17
" 0.13
ITT.
0.09
0.13
n 1 1
IV.
0.79
0.98
0.59

SMR
124
49
86

0
6
16
4
Plant 2
E
7.29
10.24
7.50
Cancer of Rectum (TCD =


	
769
Liver
1111
	

0
2
1
Ca nc er
1
1
n
Cirrhosis of
127
	
	
1
1
3
0.19
0.26
0.18
(TCD = 155,
0.15
0.23
0 1 fi
Liver (TCD =
0.95
1.32
0.85

SMR
82
156
53
154)


769
555
156A)
666
435

581)
105
76
353
              Latency
              (years)

             < 10 yrs.
            10-<20 yrs.
             > 20 yrs.
             < 10 yrs.
            10-<20 yrs.
             > 20 yrs.
             < 10 yrs.
            10-<20 yrs.
             > 20 yrs.
             < 10 yrs.
            10-<20 yrs.
             > 20 yrs.

            1  Latency = number of years  from date  first employed  in exposed job.

            *  p<0.05

0
12
19
8
0
2
2
2
1
0
2
1
3
Plants 1 &
E
12.12
16.31
12.15
0.32
0.43
0.31
0.24
0.36
0.27
1.74
2.30
1.44
2
SMR
99
116
66

465
645
833*
277
— —
115
43
208

-------
                                                     Table 5
c:
                                Observed  and  Expected  Deaths According to Length of
                                            Exposure Among PCB Workers

                                          I.   All  Cancers  (ICD =  140-205)
Length of
Employment
< 5
5-9
10-14
15-19
> 20
yrs.
yrs.
yrs.
yrs.
yrs.
0
11
1
0
1
0
Plant
E
11.49
2.59
0.84
0.59
0.08
1
SMR
96
39
	
169
	
Plant 2
0
20
2
3
1
0
E
17.97
3.65
2.01
0.89
0.51
SMR
111
55
149
112
	
0
31
3
3
2
0
Plants 1 & 2
E
29
6
2
1
0
.46
.24
.85
.48
.59
SIW
112
48
105
135
	
                                         II.   Cancer of Rectum  (ICD =  154)
< 5 yrs.
5-9 yrs.
10-14 yrs.
15-19 yrs.
> 20 yrs.
1
0
0
0
0
0.32
0.08
•0.03
0.02
0.001
313
	
	
	
	
0
0
2
0
0
0.45
0.10
0.05
0.02
0.01
222
	
4000
	
	
2
0
2
0
0
0.77
0.18
0.08
0.04
0.01
260
	
2500
	
____
< 5 yrs.
5-9 yrs.
10-14 yrs.
15-19 yrs.
>_ 20 yrs.
1
0
0
0
0
0.24
0.06
0.02
0.01
0.001
                                       III.  Liver Cancer (ICD - 155,  156A)
                                          417
2
0
0
0
0
0.40
0.08
0.05
0.01
0.01
500
3
0
0
0
0
                                      IV.  Cirrhosis of the Liver (ICD = 581)
0.64
0.14
0.07
0.02
0.01
469
< 5
5-9
10-14
15-19
> 20
yrs.
yrs.
yrs.
yrs.
yrs.
1
0
0
0
0
1.76
0.38
o.n
0.09
0.02
57
	
	
	
	
2
1
1
1
0
2.22
0.46
0.22
0.12
0.07
90
217
455
833
	
3
1
1
'1
0
3
0
0
0
0
.98
.84
.33
.21
.09
75
119
303
476
	

-------
                                         Table  6
          Concentrations of PCB's  (Aroclor 1016)  at  Plant  1  -  Taken  April,  1977
                       A.  Power Capacitor Manufacturing Facility
  Job Titles

Recovery
 Repair
Salvage
 Operator
EMF Operator
Treat Helper
Treat Operator
Repair

Moveman
 (Sealing Area)
Moveman
 (Testing  Soldi
   827
Testing
Packer
Treat Operator

Rework & Final
 Assembly
Maintenance
Rework Tester
Rework Packer
Rework
 Tester Solder
   Personal Air Samples
          Total
        Sampling
No. of    Time       TWA*
Samples (minutes)  (  g/m )
                                                                ~Area Air Samples
        Location
2
1
1
2
2
1
B.
2
ing Area)
115
3
3
2


2
1
1
1
840
426
431
867
731
422
Small
689


1290
1287
845


824
404
433
435
298
155
115
80
66
50
Capacitor
393
3

218
199
160


152
150
140
132
Test & Paint
Assembly
Shipping
Storage
Wi nding

Manufacturing Faci
Soldering
1306

Shipping
Winding
Can
Manufacturing
Cover
Manufacturing



2
2
1
1
1

lity
2
220

2
2
2


2



           Total
         Sampling
No. of     Time       TWA*
Samples  (minutes)  ( g/m )
                                                       840
                                                       834
                                            41
2
1
1
1
851
426
427
420
29
16
14
3
2
220
2
2
2
782
Assembly
838
828
836
476
2
56
54
51
                                            45
           271
24
  TWA is calculated over the total sampling time period.

-------
                                        Table 7

         Concentrations of PCB's (Aroclor 1016) at Plant 2 - Taken March,  1977
                  Personal Air Samples
                                     Area  Air  Samples



Oob Titles
Degreaser
Solder
Tanker
Moveman
(Soldering
Area)
Heat Soak
Operator
Tester
Pump Mechanic


No. of
Samples
1
3
9
«
3


3
3
1
Total
Sampl ing
Time
(minutes)
381
884
2120

752


872
917
377


TWA*
( 9/m )
1,260
1,060
850

720


630
290
280



Location
Impregnation
Pump Room
Testing

Pre-assembly


Shipping
Winding
Cover


No. of
Samples
2
3
5

4


2
4
3
Total
Sampling
Time
(minutes)
176
1079
1424

1213


741
637
1089


TWA*
( 9/m )
810
490
320

140


90
70
60
Floorman
(pre-assembly)
                                             Manufacturing
1683
170   Office
741
50
* The TWA is calculated over the total sampling time period.

-------
                              Discussion


Unidentified Speaker:  I know your numbers are small, but were
you able to do separate analyses by sex?  In particular, were there
liver cancers in women?

Mr. Brown (NIOSH):  I did separate analyses by sex.  I will have to go
back to check to see about the liver cancers in the women.  I am not
sure.

Unidentified Speaker:  In both plants, TCE was used as a degreaser and
it is known that TCE has induced liver cancer in laboratory animals.
How did you separate those exposed to TCE from those exposed to PCB's to
determine whether or not some of the excess is attributable to TCE?

Mr. Brown (NIOSH):  Based on the work histories that we got from the
plant, we could tell who worked around the TCE degreaser and we
eliminated them from our study.

-------
                     Painting Trades Study
                        Progress  Report
                         Dennis Zaebst
                         Marie Haring
                         Shiro Tanaka
             NCI/EPA/NIOSH  Collaborative Workshop
                           May,  1980
                 Industry-wide  Studies Branch
Division of Surveillance,  Hazard  Evaluations and Field Studies
     National  Institute for Occupational  Safety and Health
                        Cincinnati, Ohio
                                  173

-------
                                ABSTRACT
A NIOSH study of health hazards of workers applying paints and coatings
was begun in 1978.  The study was strated because of reports of adverse
health effects including bronchitis and chest x-ray changes in an earlier
NIOSH pilot study.

A literature review was conducted and nine industries that have painting
operations were selected for study:  construction/maintenance, auto
manufacturing, shipbuilding, and furniture manufacturing,  appliance
manufacturing, railroad car manufacturing, aircraft manufacturing and
maintenance, bus, truck, farm, and construction machinery  manufacturing,
metal furniture manufacturing.  Fifty-one walk through surveys to obtain
information on paint usage and exposure,  completeness of personnel records
and other basic information were conducted in the nine industries.

The information obtained in the walk through surveys is being analyzed
in order to decide which industries and specific sites are most suitable
for indepth industrial hygiene and/or retrospective cohort mortality studies.
Cross-sectional medical studies may be conducted if warranted by early
results of the mortality or industrial hygiene studies.
                                      I'M

-------
                               INTRODUCTION






Purpose of the Painting Contract






A NIOSH study of health hazards of workers applying paints and coatings was




begun  in  1978.   The  purpose  of  this  study  is  to  evaluate,  on  an




industrywide basis, possible health effects resulting from worker exposures




to a variety of coating processes.  It is a three  phase  study,  including




industrial hygiene, epidemiological, and cross-sectional medical surveys.






Background






The study was initiated for several reasons:






1.  Only a limited amount of significant research has been conducted




    to estimate acute and chronic effects caused by exposure to




    paints and coatings.






2.  During the past decade, there has been a growing concern that




    the materials in coatings may present potential health hazards




    to workers applying the coatings on a regular basis, for example,




    i.e. metallic pigments and additives, certain catalysts and




    activators, and solvents.






3.  Several recent but limited occupational studies have suggested that




    painters experience an increased risk of developing certain diseases,




    including among others, central nervous system and respiratory




    effects, and cancer (1-8).






The motivating force behind the project was a pilot study conducted under a




NIOSH contract in 1975 by Mt. Sinai  (1)  which  clinically  surveyed  1000

-------
members  of  the  International  Brotherhood  of Painters and Allied Trades




(I.B.P.A.T.).  The study results indicated that among  the  various  trades




Included in the union (e.g. general painter, dry wall construction workers,




wood finishers, paint factory workers and many others, Table 1}, anesthetic




effects  on  the central nervous system were more frequently reported among




general painters, particularly those who had reported  working  with  epoxy




paints, and among those classified as metal painters and sandblasters.






The  study  also  reported  symptoms  of  respiratory  irritation,  such as




increased frequency of bronchitis as  seen  among  sandblasters  and  paint




factory  workers.   Chest  x-ray  changes  such as plural thickening or the




occurrence of  small rounded opacities  were  more  frequent  among  general




painters, paint manufacturers, and sandblasters.






Although  this  study  had  many  inherent  limitations  - since it focused




largely on construction painting and related trades -  it  did  indicate  a




need   to  look more  carefully  at  the potfsntial hazards of paints, paint




application processes,  and  the  mortality  and  morbidity  experience  of



involved workers.






However,  it should be said again that at lea.st part of the impetus for the




present study vas the lack of well defined ani controlled studies  relating



to application of paint.






Scope  and Limitations






For  the  purpose of this study, we have defined the "painting" industry to




include those establishments that use coating products in the finishing  of




manufactured  products  and  in new construction, building maintenance, and




rehabilitation, specifically excluding paint manufacturing.  ^e  have  also







                                        176

-------
defined paint as "a mixture of pigment or -vehicle (as oil or water)  applied



to a surface to form a thin film adhering to a substrate".   The  group  of



finishes  which  are  considered  treatments  because  they penetrate  the




surface, and a variety  of  other  specialized  finishing  processes  (e.g.



metalizing) were excluded from the study.






While  attempting  to define the areas of study or scope of the project, we



became aware of the immense  variety  of  coatings  used  in  construction/



maintenance  or  for product finishing, the number of raw materials used in



paint manufacturer the variety of application methods, and the  variety  of



settings  in  which paint is applied.  Although we have tried to maintain a



broad perspective on these aspects during this study, we  have  necessarily




had  to  pre-select  at  several  stages,  the  coating types to study, the



environments and settings in which to study the associated hazards, and the



specific  sites  and  populations  to  be included.  Although this may have



caused us to  overlook  significant  aspect;;  of  paint  application  which



deserved   study,   given   the  complexity  of  the  subject  and  funding



limitations, this was unavoidable.








                              ACCOMPLISHMENTS






Selection of Study Sites






The initial task was to select groups of painters which would be reasonably




representative of all painters in the United States, or  failing  that,  to




select  groups to study which would be representative of the largest number




of painters, or those with the most hazardous exposures.
                                         i?;

-------
The criteria developed to facilitate group selection followed a three point




decision scheme based on resin type, industrial use and the availability of




a study population:





 (1) Resins - paints are frequently classified by resin»type.  We



    concentrated mainly on those (5-7) resin types which were in



    widespread and increasing use, or those which exhibited unique




    toacicity in other studies or case reports.






    Table 2 shows the estimated total 1974 consumption of resins in



    paints and coatings  (9) -  The three resin types in largest use-were alkyd



    vinyl, and acrylic.  Although of substantially lower volume usage,



    epoxy and urethane paints were also selected because of their



    reported toxicity and potential hazards in use, as well as their



    projected increasing use in industry.  Cellulosic based paints



    were also selected because it was known that these were used



    primarily, and in fact nearly exclusively, in the wood furniture



    industry.






 2.  Industrial Use - the selection of industries to study was then



    based on the typ<2 and volume of resin used, and the extent of exposure



    based on the method of application, number of workers exposed, and



    the  projected usage.






 3-  Population at Risk - additional criteria for the final selection



    of industries were those related to selecting populations for




    aortality studies, based on the size and stability of the popula-




    tion, the length of use of the selected paints  (or age of the




    industry), and information obtained prior to actual site visits

-------
    on the condition and availability of records.






After   examining  approximately  18-20  different  industrial  categories,



including architectural and maintenance painting, and  eliminating  several




due  to  such  factors  as  having  unusual or unique coating processes, or



difficulties in identifying suitable sites or  populations,  nine  possible




areas were chosen for study (Table 3)






In  summary,  each  of  the  illustrated industrial areas were selected for




study based on their high volume use of the selected paint resin types, the



number  of  painters  in  the  industry, application methods, and potential



exposure.  Also, it became obvious that  by  working  within  these  chosen



industrial  categories,  we  could assess effects of exposure to many other



paint resin types not originally selected for study.






Groups  9 and 10 - Construction and Maintenance, and the Mortality Study of



the IBPAT Membership, are included  as  separate  treatments  of  the  same



"category".   Because  of  the  apparent lack of large, stable workforce or



adequate records documenting worker exposure at work  sites  or  contractor



offices,  we  investigated whether the mortality s-nadies could be conducted




using   available  records  of  the  international  painter's   union.    In



conjunction  with  this  study,  industrial  hygiene  studies  were  to  be



conducted   to   explore   potential   exposures    at-   a    number    of



construction/maintenance worksites identified by the union.






Subsequently,  thirty-nine walk through surveys were conducted in the eight




manufacturing industries, five in each industry  (4 in ship  building),  and




surveys  were  conducted  at  all  eleven  construction/maintenance  sites




identified, for a total of 50 walk through surveys conducted.  These sites,







                                          17;)

-------
including  plant  and  construction   locations, were  selected in  general to



include (to the extent possible)  the  multiple  environments present in  each




industry,  and  to  include  where possible  large,  medium,  and  some small



establishments.  However,  this was not possible in all  categories  due  to



the  limited  amount of painting being done-and the small number  of workers




at some of the smallest sites.






Walk Through Survey Results






Table  4  presents  an  overall  summary  of the  findings in the  nine areas



examined in the walk through surveys. This is presented primarily'to  show



examples  of  some of the data and criteria we used in selecting  industries



for in-depth epidemiology studies.  These include: the number  of  painters



and "halo" workers potentially exposed (as of measure of total U.S. impact)



and the percent of total production workers who  are painters, the degree of



potential  exposure,  and  an  assessment  of  the  workforce  and  records



available for study.  Not shown but additional important criteria used were



the  types  of  coatings used, the kinds of potential hazards seen, and the



use of the newer kinds of coatings in increasing use.






A brief look at this summary indicates the following:






1.  The three categories with the largest number of painters are in



    order: construction and maintenance, autmobile manufacturing, and




    %»od furniture.  This, among other factors,  led to the selection



    of construction and wood furniture for in-depth studies.  Automo-




    bile OEM was a viable potential study, but exposures were sub-




    jectively low, and records would be  difficult to search.






2.  Metal furniture and appliances were  ruled out because tne






                                          180

-------
    nunsber of painters was low,  hazardous exposures were (subjectively)




    low, and records were marginal.






3.  Railroad equipment manufacturing was ruled out because of the



    small population available for study, and poor to marginal



    records.  Worker exposures,  however, were judged to be relatively




    high and potentially hazardous.






4.  This left large transportation equipment, aircraft, and ship



    building.  Although all exhibited variable and potentially



    hazardous exposures, aircraft was selected because of the availabi-




    lity of records at one site, and in particular because of the



    extensive use of the urethane paints, which are in increasing




    use in several other industries.






The characteristics  of  painting in each of the three selected industries



 (wood furniture, aircraft, construction/maintenance) and additional reasons



for the selection of each, are discussed below.








Wood Furniture .Finishing






Table   5 preser.~s some of the characteristics of finishing processes in the



Wood Furniture industry.  Previous  research  in  this  industry  does  net



address the problem of the finisher.  Most work concentrates on the hazards



and effects of wood dust exposure rather than  those  caused  by  finishing



products.   Our  planned studies concentrating in this area of the industry




will complement the research  already  investigating  wood  and  wood  dust



exposures.






Nitrocellulose   lacquers  are  traditionally  utilized  in  the  furniture





                                           181

-------
finishing  process.    Although  other  finish   types   may   be   adopted,




nitrocellulose  is the finish of choice.  In many operations, the finish is




applied by hand or using a hand-held  conventional  air  or  airless  spray




technique.






Environmental  controls  are  usually  limited  to  back draft booths, with




minimal use of personal protective equipment despite the fact that  workers




often  conduct  sanding and wiping operations outside the booths.  This may




cause worker exposure to evaporating solvent vapors.






The  primary  hazards, based on subjective observations and material safety




data sheets on the products in use, were exposure to  complex  mixtures  of




aliphatic  and  aromatic  hydrocarbons,  ketones,  plus  other solvents and




chemicals.   It  was  our  impression  that  although  local  exhaust   was




extensively  used,  overall control of solvent vapors in many instances was




marginal.   Confirmation  of  this,  of  course,  would  require  extensive




environmental surveys of several of these sites.






Finally,  a  i'ew  ot  the  sites  were found to contain data suitable for a




retrospective mortality study of wood furniture finishers.   This  part  of




the  furniture  industry  is  composed  of a large number of highly skilled




workers who remain classified as a finisher once the person  is  identified




as  such.   Tharefora,  it  is felt that long term finishers will have been




exposed to the nitrocellulose lacquers despite the rather high mobility  of




the  workforce.   The  cohort is anticipated to include approximately 4500-



5000 workers, working over an estimated 30 years.

-------
Aircraft Manufacturing





Comprehensive  surveys  in  the  aircraft industry (Table 6)  should provide



data on the chronic and acute health effects of urethane-type coatings,   as



well  as  zinc  chromate  and  epoxy  constituents.   Although  used as  top



coatings on aircraft for approximately 20 years, other industries are  just



beginning to use urethanes with any frequency.  Because of their durability



and fast-curing nature, urethanes are coming into increasing use and should




be  more extensively investigated to ensure proper worker protection in the



future.  Several health hazard evaluations conducted by NIOSH in  the past



few  years,  and  other  reports! have  indicated  that  exposure  to  free



isocyanate may occur in the use and application of  these  paints  (10-13).



Also,  a  recently published PMR study of zinc chromate exposed painters in



the aircraft overhaul industry has indicated  an  increased  risk  of  lung



cancer in these workers (16)..






Urethane   (usually  2-part  applied  by  handheld  air  apray) are the most




frequently used topcoats; primers vary depending on the  geographical  area



and  type  of  aircraft  service  intended  but often include zinc chromate



and/or epoxy primers.  Small parts painting is usually done  in  ventilated



booths but whole aircraft (particularly large aircraft) are done in hangars



with variable ventilation.  Personal protection is extensively used due  to




the  difficulties  in  using  engineering  controls.   Principal  potential




exposures include solvents,  lead and/or zinc chromate,  methylene  chloride




(frequent  in paint stripping formulations), and isocyanates (frequently in



pre-polymer form).






Currently,  a  study  to  assess  the feasibility of conducting a mortality




investigation in the aircraft manufacturing industry is underway.   Results






                                       183

-------
of the inquiry will be released within the.next few weeks.








Construction/Maintenance  Painting  Studies of construction and maintenance




painting would help to characterize exposures and disease processes in  the




single   largest  group of painters in the United States, and would include




situations  (along with wood finishing)  similar  to  those  in  which  non-




professional  painters may engage.  Previous research is limited to the Mt.




Sinai pilot study cited earlier, and several reports from Sweden which have




indicated psychiatic and neurological changes among house painters (14-15).






Table  7  summarizes  environmental  findings from nine of the walk through




surveys conducted at the eleven  construction/maintenance  sites.   The  11




sites included visits to two contractor's offices to examine feasibility of




using contractor records for a potential epidemiology study; actual working




sites were not visited.






The  remaining  nine  sites  can be classified as new construction painting




only/ or maintenance painting only.  Maintenance  painting  sites  includail




four  sites  under  five  contractors,  one  of which involved a complex of




commercial buildings, and three involved industrial  maintenance  paintincj.




Maintenance  painting  of  building complexes uses 70%-80% flat water based.




paints applied with roller or brush, and lesser quantities of solvent based.




enamels  are  used.   The painting is often done in unventilated areas with




inadequate or no respiratory protection and solvent  exposures  potentially




could  be  quite  high; it should be noted that some so-called "water base"




paints do contain volatile toxics such as glycol ethers.






Maintenance  painting  in  industry requires a large variety of paint types




depending on the application, geographical area, etc., but typical  were  2







                                         184

-------
system   epoxy,   latex,   alkyd   enamels,   epoxy.   polyamides,  and  (at



chemical/refinery locations) additional types such inorganic zinc  primers,




vinyls,  and  rarely,  urethanes.   They  are  usually applied by brush, or



handheld air or airless spray, often in open areas (outdoors) or in  closed



rooms  of  spaces  with  variable  ventilation  and  degree  of respiratory




protection.  Potential  exposure  include  solvents   (MEK,  Toluene,  MIBK,



Xylene,  Methylene  Chloride),  inorganic  zinc, chromates and (since these




painters are often involved in surface preparation)  silica  from  abrasive




blasting.   In  general,  respiratory  protection  and  work practices were




better controlled at the industrial sites.






Four  sites  of  construction  painting  were  also visited; two were power



plants,  (one  coal,  one  nuclear),  one  was  a  nuclear  waste  treatment



facility,   and   one   was   a    large  hone/apartment  complex.   In  the



home/apartment painting, latex and alkyd enamel were applied by  brush  and



roller in closed,  unventilated  spaces, and with little or no respiratory



protection.  At the power  plant,  2  part   epoxies  are  extensively  used



 (particularly at the nuclear facilities) and are applied in rooms or spaces




with variable ventilation by brush, roller,  or spray.  At the coal  powered



plant, polyuiethane coatings and zinc chronate primers are also used.  In



all of the industrial construction sites, respirators of various types  ara




available  on  demand,  but  are only occasionally required for specific jobs



 (e.g.  tank lining, sandblasting).  Potential exposures  in  home/apartment



painting are primarily  volatile organics, but in the  case of the industrial




construction painting,  also include various  solvents and zinc chromate  and




rarely NCO from polyurethanes, sensitizing agents  (amines, glycidyl ethers)



present in epoxies and  free silica.

-------
The  mortality  study of the membership of.the International Brotherhood of




Painters and Allied Trades (IBPAT)  includes a cohort size of  approximately



300,000.   Deaths  are  identified through the death benefit and disability



records available at the union.    The  anticipated  number  of  deaths  are



estimated  at  12,000  for  the  cohort.  This population size should yield




stable mortality rates even for rare causes of death.








                             IN-DEPTH STUDIES






In   summary,   three   industrial  categories  were  chosen  for  in-depth
                                                                  «


epidemiological   study;   namely    wood    furniture,    aircraft,    and



construction/maintenance.   The  mortality studies will utilize the records



of the  IBPAT  (the international painters union). the records of  one  large



site  in  the aircraft industry, and those of possibly two or more sites in



the wood furniture industry.  In each of these categories, three plants  or



construction   locations   will   be  selected  for  participation  in  the



comprehensive industrial hygiene surveys.  In addition, industrial  hygiene



surveys will be conducted at approximately one representative site in each



of the  remaining industries, for a total of  15  surveys.   Cross-sectional



medical studias will be conducted as warranted following completion of the



industrial hygiene surveys and the release  of  at  least  the  preliminajry



findings of tha mortality studies.






The  planned  studies  should increase the available information concerning



both the potential  and  existing  hazards  of  paint  application  in  the




workplace.   However,  the studies will necessarily fall short of achieving




an in-depth evaluation of all the  hazards  involved,  given  the  inherent




variability of conditions and processes between industries and even between
                                      186

-------
plants in the same Industry.





Also,  environmental  conditions and paint processes have changed, and will



continue to change, in many of the industries surveyed, making it difficult




to  correlate  specific causes of death or illness with specific exposures.




Evaluation of past exposures in many cases will  rely  on  historical  data



obtained  and  kept  by individual companies (sparse or nonexistent in many




cases), and available histories of process changes,  engineering  controls,




work practices, and use of protective equipment.






At  the  present  time,  the NIOSH contractor (Johns Hopkins University) is



preparing a detailed protocol for the total in-depth study.  This  protocol



will  undergo  extensive  internal  and  external technical and statistical



review.






We  anticipate beginning the in-depth industrial hygiene and epidemiological



studies  in  the  fall  of  1980.   Cross-sectional  medical  surveys,   if



warranted, will begin approximately one year later and will be based on the



findings of this industrial hygiene studies, and possibly on  the  basis  of




early results of the mortality studies.  Given thsse parameters we estimate



that the final report of the study will be released in mid-1982.
                                       187

-------
                                REFERENCES






1.  Selikoff, I.J.  1975.   Investigation of Health Hazards in the Painting




    Trades.  Unpublished Final Report.   USDHEW(NIOSH)  Contract CDC-99-74-91,






2.  Dunn, J.E., G. Linden, and L.  Breshaw.  1960.  Lung Cancer Mortality




    E3^>erience of Men in Certain Occupations in California. Am. J. Public




    Health.  50:1475-1487.






3.  Breshaw, L., L. Hoaglin, G. Rasmussen, and H.K. Abrams.  1954.




    Occupations and Cigarette Smoking as Factors in Lung Cancer.  Am. J.




    Public Health.  44:171-181.






4.  Brady, L.W.  1977.  Cancer of the Bladder.  Phila. Med. 73:181-187.






5.  Viadana, E., I.D.J. Bross, and L. Hauten.  1976.  Cancer Experience




    of Men Exposed to Inhalation of Chemicals or Combustion Products.




    J. Occ. Med.  18:787-792.






6.  Miller, A.B.  1S-77.  The Etiology of Bladder Cancer from the




    Epidemiolcgical Viewpoint.  Cancer Research.  37:2939-2942.






7.  Williams,  R.R., N.L. Stegens, and J.R. Goldsmith.  1977.  Associatioivs




    of Cancer  Site and Type with Occupation  and Industry from the Third




    National Cincer Survey Interview.  J. Nat, Cane. Inst.  59:1147-1185.






8.  Enterline, P.E., and M.F. McKiever.  1963.  Differential Mortality




    from Lung  Cancer by Occupation.  J. Occ. Med.  3:283-290.






9.  Unpublished Report.  1977.  Estimated Consumption of Resins in Paints




    and Coatings.  Institute of Applied Technology, Washington D.C.
                                      188

-------
10. Hervin, R.L., and T.W. Thobum.   1975.   Health Hazard Evaluation




    Report, HE-72-96-237.  USDHEW(NIOSH),  Cincinnati,  Ohio.






11. Okawa, M.T., and W. Kieth.  1977.  Health Hazard Evaluation Report,




    HE-75-195-396.  USDHEW(NIOSH),  Cincinnati, Ohio.






12. Seeman, J., and U. Walcki.  1976.  Formation of Toxic Isocyanate




    Vapors on Thermal Decomposition of Polyurethane Paints and Their




    Polyfunctional Hardeners.  Zentralblatt Fur Arbeitsmedizin,




    Arbeitsschultz Und Prophylaxe.   26:2-9.






13. Hardy, H.L., and J.M. Devine.   1979.   Use of Organic Isocyanates




    in Industry - Some Industrial Hygiene Aspects.  Ann. Occ. Hyg.




    22:421-427.






14. Hane, H., O. Axelson, J. Blume, C. Hogstedt, L. Sundell, and




    B. Ydreborg.  1977.  Psychological Function Changes Among House




    Painters.  Scand. J. Work, Envir., Health.  3:91-99.






15. Knave, B.  1976.  Health- Hazards in the Use of Solvents.  National




    Board of Occupational Safety and Health, Stockholm, Sweden, Report



    AMMF 112/76.






16.  Dalager, N.A., T.J. Mason, J.F- Fraumeni, R. Hoover, and W.W.




     Payne.  1980.  Cancer Mortality Among Workers Exposed to Zinc




     Chromate Paints.  J. Occ. Med.  22:25-29.
                                      189

-------
INVESTIGATION OF HEALTH HAZARDS
    IN THE PAINTING TRADES
   NIOSH  CONTRACT 210-77-0096
    THE JOHNS HOPKINS UNIVERSITY
  SCHOOL  OF HYGIENE AND PUBLIC HEALTH
           1978 - 1982
                  190

-------
       SELECTED INDUSTRIES
1.  Wood Furniture Finishing
2;  Aircraft
3.  Construction/Maintenance
                     191

-------
           IMPETUS FOR STUDY


1.  Limited Research
2.  Growing Concern About Hazards
3.  Recent Studies
4.  NIOSH Pilot Study (1975)

-------
                       SCOPE

PAINTING INDUSTRY:   "Those establishments which use coating
products in the finishing of manufactured products and in
new construction, building maintenance, and rehabilitation".

PAINT:  "A mixture  of pigment or vehicle (as oil or water)
which can be applied or spread over a surface to form a
thin -Mlm or coating which adheres to the substrate (base
material)".
                                 lea

-------
         SITE SELECTION

    DECISION SCHEME:

1.  Select Coatings for Study by Resin Type
2.  Survey Industrial Use
3.  Survey Populations Available
                     194

-------
               SITE SELECTION

    INDUSTRIAL USE:

1.  Resin/Paint Type Used
2.  Volume of Use
3.  Extent of Exposure
    a.  Application Method
    b.  No. Current Painters
    c.  Projected Usage

-------
         SITE SELECTION

    POPULATION AT RISK

1.  Population Size and Stability
2.  No. of Years of Painting (Plant Age)
3.  Records

-------
                                  TABLE  1

                              IBPAT TRADE MEMBERS
    Trade
  Reported
  Symptoms
   Trade
1. General Painter    CNS,  X-ray  changes     6.  Sandblasters
2. Drywall
   Construction Workers
3. Wood Finishers
4. Metal Painters

5. Paint Factory
   Workers
CNS, respiratory
 7. Glaziers
 8. Sign Painters
 9. Scenic Artists
10. Carpet Layers
  Reported
  Symptoms

CNS, respiratory,
X-ray changes
REF.  Selikoff, I.J.   1975.   Investigation of  Health  Hazards  in  the
      Painting Trades.   Unpublished  Final Report.   USDHEW(NIOSH)
      CDC-99-74-91.
                                           19

-------
                           TABLE 2

ESTIMATED TOTAL CONSUMPTION OF RESINS  IN PAINTS AND COATINGS - 1974
                     (Millions of Pounds)

    Alkyd                                       750.0
    Vinyl (Water-based)                          292.0
    Acrylic (Water-based)                        228.9
    Acrylic (Powder)                              90.0
    Epo*y (Reactive)                              76.6
    Amino                                        73.7
    Vinyl (Solvent-based)                         64.1
    Acrylic (T/P Solvent)                         62.0
    Unseed Oil                                  57.6
    Cellulosic (Solvent-based)                    55.1
    Urethane (Reactive)                           48.4
                                       198

-------
                                 TABLE 3

                            WALK THROUGH SURVEYS

               Industry                               No.  of Sites

 1.  Aircraft Manufacturing                                5
 2.  Wood Furniture                                        5
 3.  Major Appliance Manufacturing                         5
 4.  Metal Furniture and Fixtures                          5
 6.  Large Transportation Equipment                        5
 7.  Shipbuilding                                          4
 8.  Automobile O.E.M.                                     5
 9.  Construction/Maintenance                             11
10.  Mortality Study - IBPAT

                                                   Total  50

-------
Table 4
Industry
1. Auto O.E.M.
2. Large Transport
Equipment
3. Aircraft
4. Shipbuilding
5. Railroad
Equip.
6. Wood Furn.
7. netal Furn.
8. Appliances
9. Construction/
Maintenance
\
Total Painters
and Halo
# %
31,590
5,020
3,780
4,260
1,000
21,632
8,213
4,740
195,000
9
2
3.5
3
2.7
16.9
4.3
3.0
N.A.
Method of
Application
Manual A1r
Spray
Dipping
Manual Spray
Dipping
Flow Coating
Manual Air
Spray
Manual Spray
(air, airless)
Manual Air
Spr?y
Manual Spray
(air, airless)
Automatic
Manual Spray
Auto Electro-
static, manual
Electrostatic
All
Potential
Exposure
Probably
Low
Highly
Variable
Highly
Variable
Variable
High
Medium
Low
Low
Highly
Variable
Personnel &
Other Records
.Adequate
Adequate
Adequate
in two
Poor to
Adequate
Poor to
Adequate
Adequate
Marginal
Marginal
N.A,

-------
                      TABLE  5

             WOOD FURNITURE FINISHING

Previous Research - Limited
Resin Type - Nitrocellulose Lacquers (20 + years use)
Application Method - *Handheld  Conventional, air or airless
                        Spray Techniques;
                     *Back Draft Booths
Personal Protection - *Minimal, Heavy  Reliance on Engineering
                          Control
Hazards - *Aliphatic, Aromatic  Hydrocarbons
          *Ketones
          *0ther Solvents

Cohort Availability - *Relatively Sckilled Worker Stays
                         in Same Type  Job;
                      *Many Halo  Exposures
                      *Followup to 30  Years
                      *Estimated 4500-5000 Persons
                               201

-------
                     TABLE  6

              AIRCRAFT MANUFACTURING

Previous Research - *Health Hazard  Evaluations
                    *PMR Study  of Zinc  Chromate
                        Exposed Workers  (16).
Resin Type - Urethane Topcoats  (20  + years  use)
             Zinc Chromate, Epoxy Primers
Application Methods - *Handheld,  Conventional Air  Spray
                      *Booths or  Large  Hangars
Personal Protection - Used  Extensively
Hazards - *Urethanes (NCO,  frequently pre-polymer)
          *Lead
          *Zinc Chromate
          *Solvents
          *Methylene Chloride
Cohort Availability - Feasibility Assessment
                                 202

-------
                      TABLE  7

                CONSTRUCTION-MAINTENANCE

                   IBPAT MORTALITY

Previous Research - *Pilot Study  by Mt. Sinai  (1975)
                    *Scandinavian Studies of House Painters


            Maintenance                  Construction

Resin Type - *A1kyds                     *Alkyd
             *Acry1ics                    *Acrylic
             *V1nyls                     *Vinyl
             *Epoxy; Epoxy Polyamide      *Epoxy  (Reactive)
             *Urethane                    *Urethane
             *Inorganic Zinc  Primers      *Zinc Chromate  Primers

Application - Varies with  Job Requirements

Personal Protection - Variable, Available but  not Required

Hazards -    *Solvents                    *Solvents
             *Inorganic Zinc              *Zinc Chromate
             *Chromate                    *NCO (rare)
             *Silica                       Silica
             *NCO                          Sensitizing  Agents
             *Sensitizing  Agents
              (Amines, Gycidyl Ethers)

Cohort Availability - 300,000 Current and Previous Union
                     Members; Potentially  12,000 Deaths
                             203

-------
                                  Discussion
Dr.  Keefer (NCI):   On one of your slides,  you  singled out methylene
chloride which is  used so much in stripping  operations.   It was  listed
together with several other chemicals that I though  were  known  to  be
particularly hazardous.   Do you have any comments  on the  toxicity  and
hazards associated with  methylene chloride itself?  I am  back two  or
three years ago when some of the people from Dow and elsewhere,  I
think, cleared the material as a carcinogen.   I just wondered if you
have any special reasons for indicting it in this  way.

Mr.  Zaebst (NIOSH):  The list of chemicals shown in  the slide just gave
some examples of some of the hazards seen. Methylene chloride has  been
studied in several NIOSH health hazard evaluations involving several
different aircraft overhaul and construction facilities.   Overexposures
were found.  As far as the toxicology, I believe that it  should  be
looked at in more detail.
                                      204

-------
                              General Discussion
Dr. Bridbord (NIOSH):  We do have some time for discussion, if anyone
would like to comment on any of the four papers or ask additional
questions.  Please come to the microphone, if you do.

Dr. Kraybill (NCI):  I would like to ask a question of Larry Keefer.
Methylene chloride is now being tested and I was not familiar with the
Dow statement.  Did they clear methylene chloride?  They did a study?

Dr. Keefer (NCI):  All I recall is that Chemical Engineering News  a few
years ago had a small column which said that Dow's recent studies  show
that methylene chloride is not carcinogenic.  Other than that, I really
cannot say.  I have not seen any original data.

Dr. Cantor (EPA):  I do not think that they ever published that informa-
tion I think they made it available through a testing  program.  It was a
small local inhalation study.  It was more pharmacokinetics than anything
else.  I think it was a simple bioassay.  The NCI testing program  had
methylene chloride on test halfway through a two year  gavage study.  I
think they were right in the process of starting a new study.

Dr. Keefer (NCI):  I have a news clipping.  You can read it if you would
like. But that is all I have.
                                      205

-------
 FIRST NC1/EPA/NIOSH COLLABORATIVE WORKSHOP:
     PROGRESS ON 3OINT ENVIRONMENTAL AND
         OCCUPATIONAL CANCER STUDIES
             Tuesday Afternoon, May 6
EPIDEMIOLOGICAL/STATISTICAL SESSION (CONTINUED)
              SESSION CHAIRPERSON
               Dr. Joseph Fraumeni
              National Cancer Institute

-------
    Mortality Study of Workers Employed at Organochlorine
                Pesticide  Manufacturing  Plants
                       David  P.  Brown
                      David Ditraglia 2
                                       2
                      Tsukasa Namekata
                                     2
                       Norman Iverson
        U.S. Department  of  Health,  Education and Welfare
                    Public Health Service
                  Center for  Disease  Control
    National Institute for Occupational Safety and Health
Division of Surveillance, Hazard  Evaluations and Field Studies
                      Cincinnati,  Ohio
                   2
                     University of  Illinois
                   School of Public Health
                          May, 1980
                             20V

-------
                                ABSTRACT

    A retrospective cohort study was conducted to examine the
mortality of workers employed in the manufacture of the chlorinated
hydrocarbon pesticides, chlordane,  heptachlor, DDT and
aldrin/dieldrin/endrin.  Four manufacturing plants were selected for
study, and each cohort included all workers employed for at least 6
months prior to January, 1964.  The entire study group totalled
approximately 2100 individuals.  Vital status ascertainment for these
cohorts ranged from 90% to 97% complete, the cut-off date for
follow-up was December 31, 1976.
    In general, there are too few deaths in this study to make any
meaningful conclusions.  The standardized mortality ratio (SMR) for
all causes in each cohort is below expected (100), ranging from 66 to
82, probably reflecting the "healthy worker effect".  For "all
malignant neoplasms", the SMR's range from 68 to 91  and for
respiratory cancer from 55 to 132.   In the aldrin/dieldrin/endrin
cohort, pneumonia and "other respiratory diseases" were significantly
above that expected.  These causes  of death need to be examined in
more detail.
    It is recommended that these cohorts be followed for several more
years and the mortality patterns re-examined.

-------
Introduction

The organochlorine (OC) pesticides such as DDT,  Aldrin,  Dieldrin,
Lindane, Chlordane, Heptachlor,  Toxaphene and Mirex  have been  an
important class of compounds in  terms of production  volume  and use.
Due to the widespread use of these pesticides during the past  30
years, there has been an opportunity for widespread  exposure to
workers who manufacture, formulate and apply these compounds,  and
ultimately to those in the general population through ingestion of
contaminated food and general environmental pollution.

There has been a great deal of concern about the long term, latent
health effects from exposure to X pesticides.  These chemicals have a
tendency to penetrate cell membranes and to be stored in the body
fat.   Some of these chemicals have been shown to be toxic  to  the
                                  2-3
liver and kidney in exposed humans    and benign and malignant
tumors in the liver have been induced in experimental animals
                                                  4  9
chronically exposed to several of the OC compounds   .   There  are
also reports of effects on the hematopoietic system  among individuals
exposed to Chlordane10, DDT11, Dieldrin11 and Lindane12"14.

In order to.determine whether or not exposure to certain OC pesticides
is associated with an increased risk of mortality due to chronic
diseases an epidemiologic mortality study was initiated  by  the
National Institute for Occupational Safety and Health (NIOSH). The
                                    20

-------
study was carried out under  contract  by the  University  of  Illinois,
School of Public Health.   The  contract stipulated  that  the study
design be a retrospective cohort  mortality study.   The  study cohort
was defined as all workers at  selected pesticide plants who were
employed for at least six months, either continuously or
intermittently between January 1, 1940 and December 31, 1964.

The original intent of this  study was to examine the mortality  of
workers employed in OC pesticide  formulating plants.  However,  after
visits had been made to numerous  formulating plants around the  country
it was determined that this  part  of the pesticide  industry was  not
suitable for an epidemiologic  study.   The formulating plants are
usually small operations, where work  is seasonal,  the turnover  rate is
high, exposures are multiple and  records needed  to conduct an
epidemiologic study are not  generally kept.   Therefore, the emphasis
of the study was shifted to  OC pesticide manufacturing  plants.   These
plants offerred a more suitable population for the investigation.
However, the exposures at these plants are probably lower  than  at
formulating plants and also  include the chemical precursors of  the
final technical grade pesticide.

An attempt was made to include OC pesticide  manufacturing  plants in
the U.S. which began operations at least 25-30 years ago,  which had
relatively large workforces  and had records  available to identify a
study cohort.  A list of the major OC pesticide  manufacturing plants
                                    210

-------
was assembled from sources such as the Farm Chemical  Handbook,  trade
commission reports, and EPA.  Based on accummulated information,  a
number of potential facilities were contacted and walk through  surveys
were conducted to make the final selection for the study.

Four facilities were eventually chosen for the study.  Table 1  gives a
description of the four plants.  Plant No. 1 is located in Illinois
and has manufactured chlordane since 1946.  Plant No. 2 located in
Tennessee, has produced Heptachlor since 1951.  Also, in 1953 Endrin
was manufactured in a pilot operation at this plant and by 1955 full
scale commercial production of Endrin began.  Other products at plant
No. 2 have included hydrogen gas, chlorine, and chlorendic anhydride.
Plant No. 3 located in Colorado, has manufactured a variety of
pesticides.  In 1946 production of Aldrin and Dieldrin began and
continued until the 1970's.  Endrin production began in 1953 and
continued until 1965.  In 1955 this plant started manufacturing an
organobromine pesticide and in 1956 the production of organophosphates
was started.  Plant No. 4 is located in California and DDT has  been
its sole product since 1947.

Methods

The study population consisted of four separate cohorts from the  four
pesticide plants.  For purposes of future analysis some of these
cohorts may be combined.   However, in this presentation the results of
                                   211

-------
each cohort will be examined separately.  The cohorts were defined as



all workers at each plant who had achieved at least 6 months



employment prior to December 31, 1964.  This cutoff date was selected



to allow for accrual of sufficient time or latency for manifestation



of disease.







An effort was made to determine the vital status (alive or deceased)



of each member in the study cohorts as of December 31, 1976.  Vital



status was ascertained through records maintained by federal and state



agencies, including the Social Security Administration and state motor



vehicle offices.  For those individuals whose vital status could not



be determined through those sources U.S. Postal  Mail  Correction



Services and other follow-up searches were used.  For all those known



to be deceased, death certificates were obtained and causes of death



were coded by a nosologist according to the International



Classification of Diseases (ICDA) in effect at the time of death.



Those who had an unknown vital status were assumed to be alive as of



December 31, 1976 so that the true risk of mortality was not



overestimated.  Those who died after December 31, 1976 were considered



alive for purposes of this analysis.







In each cohort, person-years at risk of dying were accumulated for



each worker starting when six months of employment were completed and



ending either at the date of death or at the study end date of



12/31/76 whichever occurred first.  Using a modified life table

-------
                                                     15
analysis program, similar to that described by Cutler  ,  the
person-years for each cohort were combined into five-year calendar  and
five-year age time periods and multiplied by the corresponding  U.S.
white male cause-specific mortality rate to yield the expected  number
of deaths.  Person-years were additionally distributed by five-year
exposure and five-year latency (number of years from  date first
employed) categories.  The observed and corresponding expected  deaths
were compared and differences were tested using the Poisson
Distribution.

Results

The results of the vital status ascertainment and the total number  of
person-years for.each cohort are given in Table 2.  Even  with the
efforts previously described several  of the cohorts have  an unknown
vital status of 10 percent.

Table 3 summarizes the observed and expected deaths by specific
cause.  For the category of "all  causes", the SMR's  (observed
deaths/expected deaths x 100) range from 66 to 86.  Assuming the
record systems used to identify the cohorts were complete, these low
SMR's probably reflect the healthy worker effect which has been noted
in other studies of occupational  groups^,  and possibly the lack of
complete vital  status ascertainment.   Mortality due to all malignant
neoplasms was also lower than expected, with SMR's ranging from 68  to
                                  213

-------
91.  Other major causes of death in the cohorts including diseases of
the circulatory system are also generally lower than expected.   The
only major category where there was a significant increase in observed
deaths over expected was for pneumonia (11 observed vs.  4.3 expected,
SMR = 255, p<0.01) and for "other respiratory diseases"  (11 observed
vs. 5.2 expected deaths, SMR = 213, p<0.05), at plant 3.

Table 4 summarizes observed and expected deaths by specific type of
cancer.  As stated previously, there is a deficit in observed deaths
due to "all malignant neoplasms" in each plant studied.   Although
there are no statistically significant excesses or deficits in
mortality among any specific cancer site, several sites  are of  note.
In plant 1 there are 3 observed deaths due to stomach cancer when 0.99
were expected.  In plant 3 there are slight excesses in  cancer  of the
esophagus (2 observed vs. 0.85 expected), cancer of the  rectum  (3
observed vs. 1.24 expected), liver cancer (2 observed vs. 0.57
expected), and cancer of the lymphatic and hematopoietic system (6
obs. vs. 4.09 exp.).  In addition there is a deficit in  respiratory
cancer (7 observed vs. 12.64 expected).

An analysis by latency is presented in Table 5 for "all  malignant
neoplasms".  In this type of analysis, one looks for trends to examine
whether or not the risk of mortality as measured by the  SMR is
associated with any particular latency period.  Plants 2 and 4 show a
consistent increase in the risk of cancer mortality with an increase

-------
in the latency period,  however  the  numbers  involved  in  this analysis
are small.  Since respiratory disease  was  increased  in  plant 3,
latency was also examined for this  cause of death.   There was  a
statistically significant increase  in  mortality due  to  respiratory
disease during the 10-20 year latency  period (12 obs. vs. 4.51 exp.;
p<0.05), which decreased during the greater than 20  years latency
period (8 obs. vs. 3.96 exp.; 0.05
-------
In the plant 3 cohort there  was  a  significant  increase  in deaths due
to non-malignant respiratory disease  especially  among those with at
least 10 years of latency.   In contrast  to  this  finding there was  a
deficit in deaths due to respiratory  cancer.   These findings need  to
be examined further to determine whether there is  a true association
between respiratory disease  and  employment  at  this plant.

There was no excess in cause specific deaths in  the plant 4 cohort.
However, when the deaths from malignant  neoplasms  are examined  by
latency, there is an increase in risk with  an  increase  in the length
of the latent period.  The numbers in this  analysis are small and  this
trend could be due to chance alone.

Due to the small number of workers included in this study the
statistical power does not enable  one to conclude  that  there is no
association between cause-specific mortality and employment at  the
study plants.  The primary reason  for these small  numbers is due to
the rapid turnover at these  plants, and  thus most  workers who were
hired left before they achieved  6  months of employment.  This was
especially true at plant 4 where approximately seventy  percent  of  the
employees worked less than 6 months.

Although this study has not  identified a specific  cancer risk
associated with employment at certain types of OC  pesticide
manufacturers, it points to  several causes  (stomach cancer in plant 1,
                                     216

-------
esophagus, rectum, liver and lymphatic/hematopoietic  cancer  in  plant
3) that should be examined further.   An  attempt  should  be made  to
determine if there are any common  exposures  among  those who  died from
these causes of death.  Additional  analyses  are  also  necessary  to
determine, if possible, whether  or  not the excess  in  respiratory
disease is associated with specific occupational exposure at plant  3.
Finally, the mortality experience  in each  of these cohorts should be
followed for several more years  with a better ascertainment  of  vital
status to increase the statistical  power of  the  analysis so  that more
confident conclusions can be made.
                                       211

-------
REFERENCES








 1.  Smith MI,  Stohlman EF.   The  pharmacologic action of



     2,2 bis(p-chlorophenyl)  1,1,1-trichloroethane and its estimation



     in the tissues and body fluids.   Public Health Rep. 59:984-93,



     1944.








 2.  NIOSH, CDC, USPH,  DHEW.   Special  Hazard Review for



     Aldrin/Dieldrin.   September,  1978 (Publication No. 78-201).








 3.  NIOSH, CDC, USPH,  DHEW.   Special  Hazard Review for DDT.



     September, 1978 (Publication  No.  78-200).








 4.  Evaluation of Carcinogenic,  Teratogenic and  Mutagenic Activities



     of Selected Pesticides  and Industrial  Chemicals - Vol.  1,



     Carcinogenic Study.  Springfield, Va.   U.S.  Dept. of Commerce,



     National  Technical  Information Service, 1968, 92 pp.



     (NTIS PB 233 159).







 5.  Ulland BM, Page NP, Squire RA, Weisburger EK, Cypher RL.   A



     carcinogenicity assay of mirex in Charles River CD rats.   J.



     Natl. Cancer Inst. 58:133-40,  1977.








 6.  IARC Working Group.  Some Organochlorine Pesticides in,  IARC



     Monographs on the  Evaluation  of the  Carcinogenic Risk of

-------
     Chemicals to Man.   Lyon,  International  Agency for  Research on
     Cancer,  1974, Vol.  5,  pp.  25-38,  125-56.

 7.  Advisory Committee  on  Aldrin  and  Dieldrin.   Report of the
     Aldrin/Dieldrin Advisory  Committee to William D.  Ruckelshaus,
     Administrator, U.S. Environmetal  Protection Agency.  EPA, 1972,
     pp. 2, 3, 41-46.

 8.  Bioassay of Chlorodane for Possible Carcinogenicity,  DHEW (NIH)
     publication no. 77-808, NCI-CG-IR-8.  Bethesda,  Md.  U.S. Dept.
     of Health, Education and  Welfare, Public  Health  Service,  NIH, NCI
     Division of Cancer  Cause  and  Prevention,  1977, pp. v-vi.

 9.  Bioassay of Heptachlor for Possible Carcinogenicity,  DHEW (NIH)
     publication no. 77-809, NCI-CG-TR-9.  Bethesda,  Md.  U.S. Dept.
     of Health, Education and  Welfare, Public  Health  Service,  NIH, NCI
     Division of Cancer  and Prevention, 1977,  pp. v-vi.

10.  Furie B, Trubowitz  S.   Insecticides and blood dyscrasias  -
     Chlorodane exposure and self-limited refractory megaloblastic
     anemia.   J. Am. Med. Assoc. 235:1720-22,  1976.

11.  Sanchez-Medal L, Castanedo JP,  Garcia-Rojas F.  Insecticides and
     aplastic anemia. N. Engl. J. Med. 269:1365-67,  1963.

-------
12.  Erslev AJ.  Drug-induced blood dyscrasias - I.   Aplastic anemia.



     J. Am. Med. Assoc. 188:531-32, 1964.







13.  Samuels AJ, Mil by TH.   Human exposure to Lindane - Clinical



     hematological and biochemical effects.  J. Occup. Med.  13:147-51,



     1971.







14.  West I.  Lindane and hematologic reactions.  Arch.  Environ.



     Health 15:97-101, 1967.







15.  Cutler SJ, Ederer F.  Maximum Utilization of the life table



     methods in analyzing survival.  J.  Chron. Dis.  8:699-709,  1958.







16.  McMichael AJ, Haynes SG, Tyroler HA.   Observations  on the



     evaluation of occupational  mortality  data.  J.  Occup. Med.



     17:128-131, 1975.







17. 'Wang HA, MacMahon B.  Mortality of Workers employed in  the



     manufacture of chlordane and heptachlor.  J. Occup. Med.



     21:745-748, 1979.

-------
                                                             lable 1

                Description of Plants Included in the Study of Organochlorine Pesticide Manufacturers
PLANT 1
Date began OC
pesticide production 1946
OC pesticides produced - Chlordane
Other pesticides none
produced
PLANT 2
1951
- Heptachlor
- Endrin
none
PLANT 3 PLANT 4
1946 1947
- Aldrin - DDT
- Dieldrin
- Endrin
- organobromines none
- organophosphates
Other chemicals at plant
  - Chlorine        - Chlorine        - Numerous precursors
Dicyclopentadiene  - Chlorendic anhydride
                  Hexachlorocyclopentadiene
                   - vinyl chloride
                                     tri-chloroacetaldehyde
                                       - sulfuric acid
                                     -Monochlorobenzene
Location
    Illinois
Tennessee
Colorado
California
ro
ro

-------
                                     Table 2

                  Vital Status Follow-up of Workers in Study of
                      Organochlorine  Pesticide  Manufacturers

Known to be alive
Known to be deceased
Unknown vital status
Total
PLANT 1
259 (79%)
59 (18%)
9 (3%)
327
PLANT 2
265 (87%)
24 (8%)
16 (5%)
305
PLANT 3
870 (75%)
173 (15%)
112 (10%)
1,155
PLANT 4
278 (79%)
42 (11%)
34 (10%)
354
Person-years of observation  8,354        5,672         24,939         7,601

-------
                                        Table  3
          Observed/Expected Deaths According to Major Cause Among Workers in
                    Study of Organochlorine  Pesticide Manufacturers
CAUSE OF DEATH
(7th Revision I CD No.)
All Malignant Neoplasms
(140-205)
Nervous System Diseases
(330-334)
Circulatory System Disease
(400-468)
Non-malignant
Respiratory System Disease
(470-527)
Accidents
(800-962)
Suicide
(963, 970-979)
All Other Causes
All Causes
1 SMR = Observed Deaths y
PLANT 1
11/15.89
(69)1
7f 6.03
(116)
28/40.25
(70)
11 4.58
(22)
6/ 6.28
(96)
11 2.12
(47)
5/12.01
(42)
59/87.16
(68)
100.
PLANT 2
61 6.60
(91)
11 1.85
(54)
12/15.35
(78)

11 3.96
(25)
-
4/ 8.57
(47)
24/36.33
(66)

PLANT 3
31/ 37.79
(82)
9/ 13.32
(68)
69/ 90.17
(77)
22/ 10.40
(212)*
111 18.45
(60)
10/ 5.99
(167)
21/ 29.74
(71)
173/205.86
(84)

PLANT 4
61 8.86
(68)
-
17/20.61
(82)
11 2.28
(44)
4/ 5.46
(73)
-
14/11.79
(119)
42/49.00
(86)

         Expected Deaths
*  p<0.01
Blanks represent no observed deaths.

-------
                                        Table 4

              Observed/Expected Deaths According  to  Specific  Cancer Causes
                 Among Workers  in Organochlorine Pesticide Manufacturing



       CAUSE OF DEATH               PLANT 1       PLANT 2        PLANT 3       PLANT 4
 (Seventh Revision ICD No.)
All Malignant Neoplasms
(140-205)
Esophagus
(150)
Stomach
(152,153)
Intestine
(152,153)
Rectum
(154)
Liver
(155J56A)
Pancreas
(157)
Respiratory
(160-164)
Bladder and Urinary
(180-181)
Other and Unspecified
(1566,165,190-199)
Lymphatic and Hematopoietic
(200-205)
Others
11/15.89
(69)1
-
31 0.99
(303)
-
11 0.56
(178)
-
I/ 0.91
(110)
6/ 5.43
(110)
-
-
-
-
6/6.60
(91)
-
-
1/0.57
(175)
-
-
-
3/2.45
(122)
1/0.15
(666)
1/0.88
(114)
-
-
31/37.79
(82)
21 0.85
(235)
11 2.09
(48)
I/ 3.35
(30)
3/ 1.24
(242)
21 0.89
(225)
11 2.10
(48)
7/12.64
(55)
11 1.10
(91)
61 4.80
(125)
6/ 4.09
(147)
H-
6/8.86
(68)
-
1/0.44
(227)
-
-
-
1/0.49
(204)
4/3.19
(125)
-
-
-
-
       = Observed  Deaths x 100.
         Expected  Deaths


Blanks  represent no observed deaths.


                                              224

-------
                                Table 5

         Observed and  Expected Deaths Due to Malignant Neoplasms
      According to Latency! Among Workers  in  Study of Organochlorine
                         Pesticide Manufacturers
Years

PLANT


PLANT


PLANT


PLANT

Since First Employed
< 10 yrs.
1 10 - < 20 yrs.
> 20 yrs.
< 10 yrs.
2 10 - < 20 yrs.
> 20 yrs.
< 10 yrs.
3 10 - < 20 yrs.
> 20 yrs.
< 10 yrs.
4 10 - < 20 yrs.
> 20 yrs.
Observed
1
4
6
0
3
3
4
18
9
0
1
5
Expected
1.52
4.43
9.94
1.46
3.30
1.85
7.55
16.25
14.00
1.40
3.68
3.78
SMR
66
90
60
.
91
162
53
111
64

27
132
Latency = number of years from date first  employed.
                                  225

-------
                                  Table 6

 Observed and Expected Deaths from Respiratory Disease According to Latency
Anong Workers at Plant 3 in Study of Organochlorine Pesticide Manufacturers
Years Since First Employed
< 10 yrs.
10 - < 20 yrs.
> 20 yrs.
Observed
2
12
8
Expected
1.07
4.51
3.96
SMR
187
266*
202
p<0.05

-------
                                Discussion
Dr. Kraybill (NCI):  This paper interested me very much, because of the
selection of the chemicals, these pesticides, particularly DDT.  We have
been waiting for many years for the story about DDT.  Correct me if I
am wrong, but we have no human data yet to show that DDT or DDE has been
carcinogenic in man.  People opine that maybe if you did, then formulators,
since they are getting a good exposure, probably more so than in the
industrial plant, could show a causation.  I am wondering if you could
combine a population.  Is that plant still producing DDT in California?

Mr. Brown (NIOSH):   It is.

Dr. Kraybill (NCI):  Are there formulators for DDT?

Mr. Brown (NIOSH):   I am sure there are.  Most of the formulators
formulate thousands of different chemicals depending on the season and
on the day.

Dr. Kraybill (NCI):  Well,  I am not an epidemiologist and that is why I
am asking stupid questions.  If you could combine them, maybe the
numbers would be sufficient to draw a conclusion.  But if you could do
this on DDT and put that issue to bed once and for all, that would be a
great contribution.

Mr. Brown (NIOSH):   One thing that we may do is look at these plants and
have a shorter cut-off period for length of employment and we would get
more people in the study that way.  I do not know if the short-term
employment people are as important though.

Dr. Keefer (NCI):  I was wondering also about the agricultural workers
themselves who might be using these chemicals.  I do not suppose you
would have any better luck with them, but I wanted to ask.

I also wondered about other types of pesticides. -1 do not know about
the usage patterns at all,  but some of the dinitrobenzene types of
pesticides and herbicides have been shown to contain relatively large
amounts of nitrosamine contaminants.  These are several orders of
magnitude higher concentrations in the commodity than some of the
things that we worry about like beer and scotch.  I was wondering
whether it would be possible to follow the experience of people with
those kinds of pesticides as opposed to the chlorinated hydrocarbons or
in addition to the chorinated hydrocarbons.

Mr. Brown (NIOSH):   On your first question, agricultural workers are a
very difficult group to follow for epidemiological work.  They are
migratory.  There are usually no records of these people or who they
are.  Many of them are illegally in this country.  It is a very
difficult study to accomplish.  Plus, it is out of NIOSH's purview; I
think it is EPA's.

-------
As far as doing work on other compounds, I think NIOSH is looking at
other pesticides.  In fact, they are looking at some other formulators.
I am not sure if they are concentrating on looking at nitrosamines.   I
think John, who presented a paper on nitrosamines before, may be looking
at some of those in his survey.  As far as looking at the mortality
outcomes, I do not think there are any plans to do that right now.

Dr. Caldwell:  Is there any reason that you can think of as to why the
people leave in such a short period of time?  Did any of the health
hazard walk-throughs maybe indicate that people are leaving in six
months there because they are developing acute illness or acute hyper-
sensitivity, so that there is a reason why they are gone.

Mr. Brown (NIOSH):  I do not really know.   A lot of this is seasonal
work and a lot of people come in, especially in the DDT plant in Los
Angeles, for the summertime and they produce tremendous quantities of
DDT and then there is a slowdown and so they leave.  A lot of people are
there for summer jobs.

The other thing about studying other pesticides is we are trying to  look
at people who were employed by the government in some of these pest
eradication programs.  A lot of those people used DDT years back, such
as in the fringe beetle eradication program and some other things
through the Agriculture Department.

Dr. Galbraith (EPA):  Was there any difference in the inert ingredients
in the pesticide formulations in the four different plants?

Mr. Brown (NIOSH):  Well, since these were not formulators, these plants
only made the technical grade product.  So there was not any addition of
inert ingredients at these plants.  They sold these concentrated tech-
nical grades to a formulator and the formulators are the ones who add
the inert ingredients and mix them up.

Dr. Fraumeni (NCI):  Thank you very much.
                                     228

-------
PRELIMINARY FINDINGS OF AN EPIDEMIOLOGIC STUDY OF TALC WORKERS

                    (INDUSTRIAL HYGIENE PORTION)
                                  by

                              Alice Greif e
            National Institute for Occupational Safety and Health
         Appalachian Laboratory for Occupational Safety and Health
                    Morgantown, West Virginia  26505

-------
                              INTRODUCTION









     Talc, a magnesium silicate mineral, is mined in several geographic




areas in the United States.  The ore bodies examined in this study were




Montana, Texas, and North Carolina.  We examined seven (7) mines and eight




(8) mills.




     The purpose of the study was to characterize the talc, evaluate the




workers' exposure, and ascertain the chronic effects of exposure (Table A).




The environment of each facility was characterized as to total and respirable




dust concentrations, percent (%) free silica, trace element concentrations,




percent (%) fibrous minerals, calcite, and dolomite.  Individual exposure




was determined by personal respirable breathing zone samples on all




participating employees.  Estimates of exposure for each job were obtained




from the personal samples.




     The mines in Montana and Texas are typical open-pit operations, while




the mine examined in North Carolina is underground employing square set




timbers and stopes.









                                RESULTS









     Personal respirable breathing zone samples (PRBZ)  were collected from




each participating employee and Time Weighted Averages  (TWA) were obtained.




The TWA's were utilized to derive geometric mean values for each job examined.




These mean values were then used to develop cumulative  exposures.




     The time weighted averages for each employee were  grouped together




according to the actual job performed on the day of the study.   The geo-
                                          280

-------
metric mean  for  each  job  classification was  then  calculated  from  the




grouped TWA's.




     The geometric mean for the dust levels  in the mine and  mill  are




presented by region (Table B & C).




     Each ore body was analyzed for the following trace elements:  Iron,




Manganese, Calcium, Aluminum, Zinc, and Nickle (Table D).  These  trace




elements were selected to compare the talc examined in this  study with




the talc examined in New York and Vermont.




     Montana talc had the lowest concentrations of trace elements of the




three regions examined.  The trace element concentrations were slightly




higher in North  Carolina.  Texas talc differed most significantly from




the other regions by its extremely large concentration of calcium.




     The mineral composition of bulk samples also indicated higher calcium




value in the Texas.  This talc had a much larger percentage of dolomite




(CaMg(C)~) and a slightly larger percentage  of calcite (CaCO-) than the




other two regions (Table E).




     Examination of bulk samples of talc from each region for free silica




demonstrated the same trend as other contaminants (Table F).  Montana talc




had < .8% which  was the limit of detection.  North Carolina had a slightly




higher percentage, while Texas had the highest observed silica content.




     Respirable  dust samples revealed the silica content in Montana and




North Carolina to be generally below the limit of detection.  The Texas




talc had slightly higher levels of respirable silica.




     Analysis for the presence of fibrous minerals was two-fold.  The first




analysis was with light microscopy utilizing phase, contrast techniques.
                                       231

-------
Light microscopy was used as a screening tool to detect the presence of




fibers.  Further analysis of samples from each region was performed




utilizing analytical transmission electron microscopy (Table G).




Fibrous minerals were not detected in any samples of Montana talc.




     There were two fibrous minerals identified in the Texas talc:




tremolite and antigorite.




     Antigorite, a serpentine mineral was the major constituent.  The




fibers of both minerals ranged from 0.5 to 3.0 ym in diameter and 4 to




30 ym in length.




     The morphology of the North Carolina talc was identified as




acicular.  The acicular particles had aspect ratios ranging from 5 to




1 to 100 to 1 with some diameters  /..I ym.   These acicular particles may




have resulted from mechanical destruction of plates.

-------
                  TABLE A









INDUSTRIAL HYGIENE CHARACTERIZATION OF TALC









   I.  Personal Respirable Breathing Zone Samples




  II.  Trace Element Concentration




 III.  Mineral Composition




  IV.  Fibrous Minerals




   V.  Free Silica

-------
                  TABLE B

PERSONAL RESPIRABLE BREATHING ZONE SAMPLES
           (AVG •= GEOMETRIC MEAN)
Job
Montana
Bagger
Labman
Fork Lift Op.
Mill Operator
Laborer
Foreman
Boiler Operator
Front-End Op.
Maintenance
Welder
Wash Plant Op.
Sorter
Driller
Truck Driver
Miner
Shovel Operator
Calciner Operator

Texas
Bagger
Stacker
Forklift Op.
Mill Operator
Laborer
Foreman
Front-End Op.
Maintenance
Sorter
Driller
Avg 3
(mg/m )

2.8
.3
.5
1.0
1.4
.6
.1
.8
.4
6.3
1.4
1.6
.1
.3
.4
.2
.6
.86

3.1
1.6
2.3
38.4
1.3
1.3
1.3
1.0
.6
.7
Variance

1.9
2.5
1.8
2.5
2.0
2.8
4.9
6.1
4.3
	
8.7
2.3
	
1.8
1.9
3.6
™~"~"
.59

2.9
	
1.5
	
4.5
3.6
5.3
3.6
1.8
2.0
Number of
Samples

29
4
5
7
6
14
2
14
16
1
2
50
1
15
2
5
1
174

3
1
2
1
10
4
7
10
2
2

-------
                                TABLE B
                              (continued)

              PERSONAL RESPIRABLE BREATHING ZONE SAMPLES
                         (AVG = GEOMETRIC MEAN)
Job
Avg 3
(mg/m )
Variance
Number of
Samples
Texas

Truck Driver          .9
Miner                 .1
Shovel Operator       .3
Calciner Operator    1.1
Crusher Operator     1.7
Welder               8.5

                     TT08

North Carlinia

Bagger                .9
Mill Operator         .9
Laborer               .2
Foreman               .9
Maintenance           .03
Driller               .1
Hoist Operator        . 1
Miner                 .3
Grader                .4
Packer               1.2
Cutter               1.2
Rounder               .9
Officer Personnel     .1

                     T21
               1.5
               1.5
               1.2
                .52
               5.8

               5.8
               1.4
               2.2
               4.2
               4.8

               2.5

              16.5

                774
                          5
                          2
                          2
                          1
                          1
                          1
                         54
                          1
                          1
                          9
                          1
                          2
                          3
                          2
                          9
                          7
                          1
                          4
                          1
                          3

                         44~
ALL REGIONS
.72
.68
275
                                     235

-------
Region
                                TABLE C

                   SUMMARY OF RESPIRABLE DUST SAMPLES
                         (AVG = GEOMETRIC MEAN)
Avg (mg/m )
                                                       95% Confidence
                                                       Range of Mean
Montana
Mill
Mine

1.1
.66

.85
.47

- 1.41
- .92
Texas

     Mill

     Mine


North Carlina


     Mill

     Mine
                           1.56

                            .45
                            .26

                            .14
                              2.54 - .96

                               .18 - .71
                               .13 - .51

                               .07 - .31
CONCLUSIONS
     Mill - Baggers and Mill Operators  had highest exposures.

     Mine - Truck Drivers and Front-end Loader Operators had highest exposure.

-------
                                TABLE D

                              TRACE METALS
                               (mg/m )
Montana

Iron
 .05
Manganese
  £.01
Calcium
 .05
Aluminum
  .2
Zinc
Z..01
Nickel
 Z.-01
 .01
   .01
 Limit of Detection

 .03           .1
                .01
            .01
North Carolina
Iron
 .05
Manganese
  L> 02
Calcium
 .05
Aluminum
  .2
Zinc
Z..02
Nickel
 Z.-02
 .02
   .02
 Limits of Detection

 .02           .04
               .02
            .02
Texas

Iron
 .5
Manganese
  Z..08
Calcium
8.0
Aluminum
  .04
Zinc
 .08
Nickel
 £.08
             .08
                Limits of Detection

                .2            .2
                            .08
                          .08
                                             f-».
                                              (

-------
                                TABLE E




                  MINERAL COMPOSITION OF BULK SAMPLES




               AVERAGE PERCENTAGE (RANGE IN PARENTHESIS)









                                   Calcite             Dolomitji




Montana                               ^1                     1



                                   (0-.8)                (0-3)







Texas                                  1                     -^



                                   (0-3)                 (7-20)







North Carolina                         0                     3



                                       0                 (1-4)

-------
                                TABLE F




                        FREE SILICA BULK SAMPLES
Montana                  L*8% (Limit of Detection)






Texas                    2.23%






North Carolina           1.45%

-------
                 TABLE G




             FIBROUS MINERALS
Montana
     None Detected






Texas




     Tremollte




     Antigorite






North Carolina




     Acicular Particles
                      2'HI

-------
     PRELIMINARY FINDINGS OF AN EPIDEMIOLOGIC STUDY OF TALC WORKERS
                              John Gamble
                             Alice Greife
                             John Hancock
          National Institute for Occupational Safety and Health

       Appalachian Laboratory for Occupational Safety and Health

                        944 Chestnut Ridge Road

                     Morgantown, West Virginia 26505
Presented at the NCI/EPA/NIOSH Workshop on Environmental and Occupational
Carcinogenesis, May 6-8, 1980.

-------
INTRODUCTION




Talc is a mineral with a wide variety of uses in paint, paper, ceramics,




cosmetics, plastics, roofing products, textile material, rubber,




lubricants, corrosion proofing composition fire extinguishing powders,




cereal polishing, water filtration, insecticides to name a few.  Pure talc




is a hydrated magnesium silicate, but the talc found in nature has a quite




variable chemical composition.  The mineral contaminant in talc of most




concern is asbestos, which can produce a clinical condition resembling that




seen on exposure to asbestos per se.  The possible hazards from exposure




to talc free of asbestos contamination is less well documented.  The




purpose of this study was to ascertain the effects on the respiratory




system (symptoms, lung function, radiographic) of exposure to talc dust




thought prior to the study to contain no asbestos.  Two hundred and ninety-




nine talc workers mining and milling talc from Montana, Texas, and North




Carolina were studied in this cross-sectional prevalence study.  The




mineralogy of the talc and exposure of the workers were just discussed by




Ms. Greife.  In this paper we will report on the chronic or long-term




effects of exposure.






The specific questions addressed in this paper are:  What is the prevalence




of respiratory symptoms,  radiographic changes and reduced lung function




among these talc workers?  What  are the dose-response relations?  How




does "morbidity" of the study populations compare to that of other mining




populations?
                              2,4-2

-------
METHODS



The study population consisted of workers mining and milling talc from



three regions of the United States:  Montana, Texas, and North Carolina.



Although several different companies may be involved, the results for



each region are combined, as the characteristics of the talc in each



region are similar.  Over 90% of the workers participated in the study.





All workers were administered a British Medical Research Council respiratory



questionnaire by trained interviewers.  Most of the interviews in Texas were



conducted in Spanish.  Non-talc work history was obtained in the interview;



work experience at the talc facility was obtained from company records.



Standard posteroanterior chest radiograms were read by three "B" readers



using the ILO U/C 1971 scheme.  The films were read independently without



knowledge of age, occupation, or smoking history.  The median of the three



readings was used for analysis.  Flow volume curves from a minimum of 5



forced maneuvers were obtained and recorded on magnetic tape using an



Ohio 800 rolling seal spirometer.  Values from the maximum envelope were



used for analysis.  Before and after shift spirometry was administered to



workers on the day shift, and personal environmental samples were also



collected on these workers.  The results of the personal environmental



sampling were used to evaluate (1) dose-response relations of talc dust



with acute changes in pulmonary function over the shift (APFT = after



shift PFT minus before shift PFT); and (2) estimate talc dust exposure



for each job.  This estimate was then used to calculate cumulative'talc



dust exposure by multiplying job exposure by job years to nearest month,



and adding the results of each multiplication.  The units for cumulative


                 3              3
exposure are mg/m  x years (mg/m -years).  Sputums were collected on



workers 35 years or older.





                                        243

-------
The prevalence of symptoms and pleural thickening were compared to 3 mining




populations, after indirect adjustment for smoking,  and using the. age




distribution of all populations.






Internal comparisons of prevalence and dose-response relationships will




be examined first (Tables 5rl4).   Then comparisons with external control




populations will be made (Tables  16-18).   Dose-response relationships




and external comparisons for lung function are in Tables 19-21.








RESULTS




All of the Texas talc workers were male,  while about 20% of the Montana and




North Carolina talc workers were  female.   The North  Carolina population had




the highest proportion of smokers (62%)  and lowest proportion of ex-smokers




(17%).  The highest proportion of nonsmokers (33%) and lowest proportion




of smokers (46%) were in Montana.  Cigarettes smoked per day was similar




among smokers and ex-smokers in North Carolina and Montana (approximately




a pack a day), but was 1/3 a pack a day less in Texas.  The North Carolina




population had worked on average  about 3 1/2 years longer (10 years) than




the workers in Montana (7 years)  and Texas (6 years), but cumulative




exposure in North Carolina was one-half (3 mg/m -years) that of Montana



               3                                                    3
workers (6 mg/m -years) and about 1/4 that of Texas  workers (11 mg/m -years)




(Table 1).  Average exposure (cumulative exposure divided by years worked)



showed the same ranking.






Only 11% of the workers in Montana and Texas had worked 10 .or more years




compared to 38% in North Carolina.  Most of the study population in Montana




and Texas  had worked less than 5  years (66% and 73%  respectively).
                                         244

-------
About 20% in all regions had worked from 5-9 years.  North Carolina had


the lowest cumulative exposure in each years worked category, especially


in the less than 10 years category.  The men working 5-9 years in Texas

                                                              3
had the highest mean cumulative exposure of any group (25 mg/m -years);


twice that of the next highest group of Montana workers with 10 or more

                     3
years tenure (12 mg/m -years).  Age, years worked, and cumulative


exposure were correlated enough to potentially confound any dose-response


association.



Table 3 summarizes the characteristics of the low, medium, and high


cumulative exposure groups by region.  North Carolina had a higher


proportion of workers in the low exposure group and a lower proportion in


the medium and high exposure groups.  The North Carolina population was


older and had more years exposure in each cumulative exposure group.  Mean


cumulative exposure was also lowest in the low and high exposure group


from North Carolina.  The Montana and Texas populations were generally

                                                                  3
similar except for the very high mean cumulative exposure (40 mg/m -years)


in the Texas high exposure group.  Smokers comprised about 50% of all


exposure groups excapt for the medium exposure group from North Carolina


where 90% were smokers.



Table 4 summarizes by region the frequency of working in non-talc jobs


where exposure to respiratory irritants was possible.  The frequency was


generally low and similar in each region.



There were only 2 cases of pneumoconiosis, both grade 1 small rounded


opacities in Texas and Montana.  This number is too small to analyze further.

-------
Cytology on sputums collected from workers greater than or equal to 35




years of age revealed no cytology suggestive of malignancy.






Tables 5-14 summarize the prevalence of cough, phlegm, shortness of breath,




pleural thickening, and obstruction by region, smoking, and exposure.




The overall prevalence of cpugh was 18%, 17%, and 27% in Montana, Texas,




and North Carolina.  Cough increased with age in all smoking groups in




Montana and in the nonsmoking and ex-smoki.ng category in Texas.  No




increase with age was observed in North Carolina.  Smokers had a higher




prevalence of cough than nonsmokers and ex-smokers in Montana, and a




higher prevalence than nonsmokers in North Carolina (Table 5).  The only




statistically significant difference was between smokers and ex-smokers




in Montana.  Cough showed no apparent association with either years worked




or with cumulative exposure, although the medium exposure group had a




higher prevalence than the low exposure group (Table 6).






The overall prevalence of phlegm was 18%, 17%, and 25% in Montana, Texas,




and North Carolina.  Phlegm did not increase with age in Montana (all




smoking categories), Texas (nonsmokers and smokers), or nonsmokers in




North Carolina.  Among smokers in North Carolina-and ex-smokers in Texas,




phlegm increased with age.  Smokers had a higher prevalence of phlegm




than nonsmokers in Montana and North Carolina; ex-smokers were




intermediate.  Ex-smokers had the highest prevalence in Texas, and non-




smoker? were intermediate (Table 7).  None of these differences were




statistically significant.  There was no consistent tendency for the




prevalence of phlegm to increase with years worked.  In Texas and North
                                         2 41.

-------
Carolina (but not Montana), the prevalence of phlegm Increased with




Increasing cumulative exposure (but was not statistically significant)




(Table 8).






The prevalence of dyspnea was low compared to cough and phlegm with 4%,




9%, and 6% in Montana, Texas, and North Carolina complaining of shortness




of breath when walking on level ground with people their own age.  The




rates increased with age in all smoking groups and regions (except smokers




in North Carolina where prevalence was zero).  There was no apparent




association of dyspnea with smoking, however.  Smokers often had the lowest




prevalence of dyspnea (Table 9).   No differences were statistically




significant.  There was no consistent increase with increasing years




worked (there was no dyspnea in any region among the 5-9 year tenure group).




In North Carolina the prevalence was elevated in the high cumulative




exposure group compared to the low cumulative exposure, but the difference




was not statistically significant (Table 10).






The overall prevalence of pleural thickening was 4%, 13%, and 18% in




Montana, Texas, and North Carolina.   Pleural thickening increased with




increasing age in all regions but was significant only in Texas.  Nonsmokers




had the lowest prevalence (only one nonsmoker had pleural thickening), while




the prevalence among smokers and ex-smokers was similar (Table 11).




Prevalence of pleural thickening increased slightly with increasing years




worked, but did not increase with increased cumulative exposure (Table 12).
The overall prevalence of obstruction (FEV /FVC <.70) was 18%, 13%, and




23% in Montana, Texas, and North Carolina.  The prevalence of obstruction

-------
increased with age in all regions and smoking categories, but was significant




only among nonsmokers, smokers, and total in Montana and total in North



Carolina.  Smokers had a higher prevalence than nonsmokers, but only in



Montana, and the differences were not statistically significant (Table 13).



There was no apparent association with cumulative exposure or years worked




(Table 14).





Tables 16 through 18 compare adjusted prevalence of symptoms, pleural



thickening, and obstruction of the talc population with 3 mining populations:



878 potash miners, 503 coal workers who had never worked underground, and



7942 coal miners who had worked only underground.  The potash miners were


                           (1 2)
part of a MSHA/NIOSH study.  '    The coal workers are from the second



round of the National Coal Workers Study.   Demographic characteristics of



these populations are summarized in Table 15.






The adjusted rate of cough among underground coal miners was higher overall



and in the 40 year or older age group compared to potash, aboveground coal



and talc workers.  There was little difference in the prevalence rates



among these latter groups and no detectable difference among the talc



regions in both age categories.  Phlegm showed a somewhat similar pattern



except the underground coal miners had more phlegm in both age groups.



Again the talc groups showed no detectable difference from each other.




Montana had a lower prevalence than aboveground coal workers in the older



age catagory (Table 16) .






The overall prevalence of dyspnea was significantly higher in the under-




ground coal miners (24%)  than all other populations.  All the other

-------
populations had rates ranging from 5% (Montana) to 14%         (aboveground




coal).  Rates in the 40 or more year group  were 7-15% compared to 28%




and 41% in the coal populations.  In the less than 40 year old group




underground coal was again high with 10%, while all other population rates




were 5% or less (Table 17).






Pleural thickening was elevated in the talc populations compared to the




comparison populations.  Prevalence was elevated in the younger talc




workers, but only the increase in North Carolina was statistically




significant.  Montana had the lowest prevalence of pleural thickening among




the talc populations, but the differences were not significant (Table 17).






There were no detectable differences in the prevalence of obstruction among




any of the populations (Table 18).






Table 19 summarizes the results of multiple regression models of pulmonary




function (FEV , FVC, Peak Flow, FEF  , FEF  ) with the predictor variables




sex, mine, age, height, pack years and cumulative exposure.   Sex,  age,  and




height were significant variables for FEV.., FVC, and peak flow. • Of these




3 variables, only age significantly reduced the variability  in flow rates




(FEFcQ, FEF7,.).  Pack years was generally significant for FEV..  and flow




rates, but not for FVC.  Mine, region, and cumulative exposure were not




significant.






Essentially the same results occur when the class variables  smoking status




and cumulative exposure group replace the continuous variables pack years




and cumulative exposure.   Adjusted mean values by smoking status,  region,




and cumulative exposure group are summarized in Table 20.
                                         240

-------
In Table 21, pulmonary function of the combined male talc populations is




compared to the control populations (after adjustments for age, height,




and smoking).   FEV ,  FVC, and flow rates at low lung volumes (FEF -,




FEF?5) were reduced compared to the coal populations.  Peak flow was




about the same as underground coal, and elevated compared to aboveground




coal.  When compared to potash miners, flow rates were reduced, but there




was no detectable difference in FEV  and FVC.









DISCUSSION




Interpretation of the data from this study has the inherent problems of all




cross-sectional prevalence studies.  These include the lack of any past




environmental measures so cumulative exposure  is based on current exposure




levels only.  In addition the numbers of workers in North Carolina and




Texas was small, especially after stratification.   The small numbers result




in very wide confidence intervals.   Exposure time (years worked)  was short




compared to other mining populations.   The mean years worked for  the




potash and coal populations was 16, 18,  and 15 years compared to  6,  7, and




10 years for the talc populations.   This is a  short time for the




development of chronic symptoms,  pneumoconiosis and impaired pulmonary




function caused by work exposure.






None of the health variables were consistently or strongly related to the




exposure variables (years worked,  cumulative exposure).   Cumulative




exposure is only a crude estimate of past exposures, based on current




levels taken over a short time period.  Since  past environmental  measures
                                         200

-------
were not taken, past exposures for each job were assumed (for the purposes



of calculation) to be the same as current exposures.  Thus cumulative



exposure is only an estimate, and how much season of the year, changes



in the talc composition, humidity, and other factors affect this estimate



are not known.






Age was consistently associated with increased prevalence of cough, dyspnea,



pleural thickening, and decreased pulmonary function.  The calculated loss



of pulmonary function with age was comparable to values from other cross-


                  (3-5)
sectional studies,      and there was little difference in the age



coefficients among the regions.  Except for FVC and FEF,.,.,  mean adjusted



pulmonary function values were not statistically different among the



regions.  Mean FVC was largest in the Montana population and least in Texas.



The rank order for FEF5_ was reversed.  FVC was reduced in the high



exposure group, but the reduction was not large.  The high exposure group



did not have the lowest value for any of the other lung function



parameters.  The reductions in lung function compared to the coal



populations (particularly FEV. and FVC) were not large.  It is interesting



that lung function was reduced compared to the coal populations, despite



their higher prevalence of respiratory symptoms.  The relationship between



smoking and pulmonary function was as might be expected.  Smokers generally



had the poorest values, and nonsmokers the best.






Thus there were no apparent dose-response relationships with symptoms or



lung function, and no apparent excess symptoms or large reductions in lung



function compared to the control populations.  This does not mean that
                                         251

-------
talc may not have an effect on these health parameters.  A dose-response

relation can be obscured by an Inaccurate estimate of exposure.  And the

comparison populations used in this report were exposed to respiratory

irritants (e.g., coal dust, diesel fumes, sylvite or KC1 and NaCl),

thereby possibly increasing the prevalence of symptoms and reducing lung

function.  Comparison with a blue collar "nonexposed" population is now


underway and will be reported elsewhere.



The same criticism regarding the comparison populations may not be valid

for pleural thickening, which was quite low in all the non-talc populations.

The pleural thickening was associated with years worked (somewhat confounded

with age).  But the increased prevalence occurred in all 3 regions, despite

the difference in exposure, and difference in talc composition.



In this study age and years worked were associated with an increase of

pleural thickening, although one worker with pleural thickening was in

the 20-29 year age category, and 4 in the 30-39 year age group.  Three of

these had worked 5 years or less.   While pleural calcification is rare

in individuals under 40, uncalcified plaques are "quite often" seen in

individuals less than 40.     Ochs and Smith    report on several cases

where as little as a years time interval was necessary for the appearance

of pleural thickening.



Asbestos (particularly anthophyllite) from either occupational or community

exposure is believed to cause an increased prevalence of pleural
           fa\
thickening.     Talc contaminated with asbestos (tremolite and antho-


phyllite) as seen under the light microscope and EM has also been associated
                                         ,52

-------
                                     REFERENCES

 1.  Attfield, M:  The effect of exposure to silica and diesel exhaust in under-
     ground metal and nonmetal miners.  Industrial Hygiene for Mining and
     Tunneling - Proceedings of an ACGIH Topical Symposium.  November 6-7, 1978.

 2.  Sutton, GW, Weems, GW, Schutz, LA, Trabant, GD: Summary report of the
     environmental results of the MSHA/NIOSH Silica/Diesel exhaust study.
     Industrial Hygiene for Mining and Tunneling - Proceedings of an ACGIH
     Topical Symposium, November 6-7, 1978.

 3.  Morris, JF, Koski, A and Breese, JD:  Normal values and evaluation of forced
     end - expiratory flow. , Am Rev Resp Dis., 111:755-762, 1975.

 4.  Bass, H:  The flow-volume loop.  Chest, 63:171-176, 1973.

 5.  Ashford, JR, Brown, S, Morgan, DC, and Rae, S:  The pulmonary ventilatory
     function of coal miners in the United Kingdom.  Am Rev Resp Dis., 97:810-926,
     1967.

 6.  Fletcher, DE and Edge, JR:  The early radiological changes in pulmonary and
     pleural asbestosis.  Radio1., 21:355-365, 1970.

 7.  Ochs, CW and Smiths JP:  Chronic pleural thickening: Some observations on
     cause and pathogeneses.  Military Med., 141:77-81, 1976.

 8.  Sargent, EN, Jacobson, G, and Gordonson, JS:  Pleural Plaques: A signpost of
     asbestos dust inhalation.  Seminars in Roentgenology, 12:287-297, 1977.

 9.  Gamble, JF, Fellner, W, and DiMeo, MJ: An epidemiologic study of a group of
     talc workers.  Am Rev Resp Dis., 119:741, 1979.

10.  Dement, JM and Zumwalde, RD:  Occupational exposures to talcs containing
     asbestiform minerals in Dusts and Disease, Lemen, R, and Dement, JM (eds).
     Pathotox Publishers, Inc., Park Forest South, Illinois, 1979.

11.  Rubino, CF, Scansetti, G, Piolatto, G, Gay, G:  Mortality and morbidity among
     talc miners and millers in Italy.  Institute of Occupational Health of Turin
     University, 1977- 12 pages.

12.  Delaude, A:  Talc-related pathology.  Bull Acad Nat Med., 161:405-409, 1977.

13.  Messite, J, Reddin, G, Kleinfeld, M:  Pulmonary Talcosis, a clinical and
     environmental study.  Arch Ind Hlth., 20:408-413, 1959.

14.  Fine, LJ, Peters, MJ, Burgess, WA, DiBerardinis, LJ: Studies of respiratory
     morbidity in rubber workers.  Part IV. Respiratory Morbidity in Talc Workers.
     Arch Env Hlth.,  31:195-200, 1976.

15.  Wegmen, DH, Burgess, WA,  and Peters, MJ: Talc Workers Health Study Report.
     Harvard School of Public Health, unpublished paper.

16.  Meurman, LO: Pleural fibrocalcific plaques and asbestos exposure. Env Res.,
     2:30-46, 1968.

-------
                              TABLE  1




DEMOGRAPHIC CHARACTERISTICS  OF THE TALC WORKER POPULATIONS BY REGION

n
AGE (S.D.)
HEIGHT (S.D.)
WEIGHT (S.D.)
YEARS WORKED (S.D.)
CUMULATIVE EXPOSURE (S.D.)
AVERAGE EXPOSURE (S.D.)
(mg/m3)
NONSMOKERS (%)
EX-SMOKERS (%)
PACK YEARS (S.D.)
CIGARETTES /DAY (S.D.)
SMOKERS (%)
PACK YEARS (S.D.)
CIGARETTES /DAY (S.D.)
MONTANA
177
34.9 (11.5)
175.5 ( 8.8)
77.8 (13.5)
6.6 ( 6.3)
5.9 ( 7.6)
1.21 (.94)
33
21
15.7 (17.9)
23 (15)
45
17.9 (16.9)
20.4 (11.0)
TEXAS
71
38.0 (13.7)
173.0 ( 6.9)
78.3 (15.1)
5.5 ( 5.7)
11.3 (45.1)
2.64 (7.12)
20
27
13.3 (20.7)
12 (14)
54
14.3 (19.7)
14.5 (11.1)
NORTH CAROLINA
51
43.1 (12.6)
172.5 ( 8.3)
78.2 (16.3)
10.1 ( 8.6)
3.0 ( 4.8)
0.28 (0.33)
21
17
18.2 (16.5)
21.4 (15.7)
62
23.7 (21.8)
20.4 (10.0)
                                     2 5«I

-------
                                      TABLE ?




  A.  CHARACTERISTICS OF TALC WORKERS BY REGION AND YEARS WORKED  CATEGORIES

MONTANA
5
5-9
£10
TEXAS
5
5-9
>10
NORTH CAROLINA
5
5-9
>10

n
92
39
46

39
21
11
19
6
26
AGE
Mean (S.D.)
28.6 ( 8.8)
36.5 (10.4)
46.2 ( 7.8)

34.8 (14.1)
39.6 (13.3)
46.5 ( 9.2)
35.3 (13.8)
43.8 (12.0)
48.6 ( 8.6)
YEARS
WORKED
Mean (S.D.)
1.9 (1.3)
7.2 (1.5)
15.4 (4.9)

1.6 (1.3)
6.7 (1.6)
16.6 (4.6)
1.6 (1.8)
7.2 (1.2)
17.1 (6.0)
CUMULATIVE
EXPOSURE
Mean (S.D.)
2.3 ( 2.4)
7.3 ( 6.1)
12.0 (10.7)

4.4 (11.5)
24.9 (80.8)
9.8 (11.1)
0.52 (1.22)
0.47 (0.32)
5.3 (5.8)
AVERAGE
EXPOSURE
Mean (S.D.)
1.47 (1.01)
1.05 (0.87)
0.84 (0.67)

3.02 (7.45)
2.98 (8.32)
0.64 (0.71)
0.25 (0.32)
0.07 ( .05)
0.34 ( .36)
B.  CORRELATION OF AGE, YEARS WORKED, AND CUMULATIVE  EXPOSURE BY REGION




                                r (95% C.I.)

AGE BY YEARS EXPOSED
AGE BY CUMULATIVE
EXPOSURE
YEARS EXPOSURE BY
CUMULATIVE EXPOSURE
MONTANA
.63
(.48 to .78)
.41
(.26 to .56)
.48
(.33 to .63)
TEXAS
.36
(.12 to .48)
.12
(-.12 to .36)
.12
(-.12 to .36)
NORTH CAROLINA
.51
(.23 to .79)
.33
(.05 to .61)
.44
(.16 to .72)
                                             255

-------
                                       TABLE 3
              CHARACTERISTICS  OF TALC WORKERS EXPOSURE GROUPS BY REGION

LOW
MONTANA MEDIUM
HIGH
LOW
TEXAS MEDIUM
HIGH
LOW
NORTH CAROLINA MEDIUM
HIGH
n
n (%)
54 (31)
64 (36)
59 (33)
27 (38)
26 (37)
18 (25)
32 (63)
10 (20)
9 (18)
AGE
Mean (S.D.)
33.4 (10.5)
32.0 (11.4)
39.3 (11.5)
33.3 (13.1)
40.2 (13.6)
42.1 (13.4)
40.6 (13.7)
44.3 ( 8.5)
50.6 ( 9.4)
YEARS
WORKED
Mean (S.D.)
5.6 (7.4)
4.0 (4.3)
10.3 (5.2)
4.9 (7.0)
4.2 (3.8)
8.1 (5.4)
7.5 (8.0)
13.5 (9.7)
15.9 (4.9)
CUMULATIVE
EXPOSURE
Mean (S.D.)
0.48 (0.33)
2.72 (1.38)
14.4 (7.7)
0.37 (0.30)
2.93 (1.48)
39.9 (84.9)
0.30 (0.29)
3.53 (2.12)
11.8 (4.7)
AVERAGE
EXPOSURE
Mean (S.D.)
0.54 (0.81)
1.44 (1.01)
1.57 ( .58)
0.57 (0.70)
1.09 (0.71)
7.98 (12.91)
0.11 (0.19)
0.38 (0.30)
0.75 (0.24)
LOW CUMULATIVE EXPOSURE  = <2 mg/m  -years
MEDIUM CUMULATIVE EXPOSURE =2-5.9 mg/m3 -years
HIGH CUMULATIVE EXPOSURE =  >6 mg/m3 -years
AVERAGE EXPOSURE = Z(cumulative exposure/years worked)

-------
                               TABLE 4




             PREVALENCE (%) BY REGION OF OTHER OCCUPATIONS

HAVE YOU EVER WORKED ....
IN A QUARRY?
IN A FOUNDRY?
IN A POTTERY?
IN A COTTON, FLAX, OR HEMP MILL?
WITH ASBESTOS?
MONTANA
n (%)

3 (1.7)
2 (1.1)
2 (1.1)
1 (0.6)
4 (2.3)
TEXAS
n (%)

2 (2.8)
0
1 (1-4)
13 (18.3)
i
1 (1.4)
NORTH CAROLINA
n (|)

6 (11.8)
0
0
4 (7.8)*
0
*95% confidence intervals do not overlap.

-------
                                  TABLE 5

     PREVALENCE OF COUGH AMONG TALC WORKERS BY AGE, SMOKING, AND REGION
                                            AGE

MONTANA
NONSMOKER
EX-SMOKER
SMOKER.
TOTAL
TEXAS
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
NORTH CAROLINA
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
40
% (95% C.I.)
7 (6-19)
0 (0-16)
27 (16 - 40)
16 (10 - 24)
10 ( 1 - 40)
11 ( 1 - 44)
15 ( 6 - 34)
13 ( 5 - 28)
0 (0-50)
33 ( 2 - 87)
38 (17 - 67)
29 (13 - 51)
>40
% (95% C.I.)
19 ( 5 - 42)
10 ( 2 - 28)
38 (19 - 58)
23 (13 - 35)
25 ( 1 - 75)
40 (15 - 73)
8 (0-35)
23 (11 - 42)
0 (0-40)
17 ( 1 - 60)
37 (15 - 64)
26 (12 - 45)
TOTAL
-% (95% C.I.)
10 ( 4 - 21)
5 ( 1 - 17)*
29 (19 - 40)*
18
14 ( 3 - 39)
26 (11 - 50)
13 ( 5 - 28)
17 ( 9 - 28)
0 (0-25)
22 ( 4 - 56)
38 (21 - 58)
27
COUGH = Answering yes to the question:  "Do you usually cough on most days
        for as much as three months each year?"

SUMMARY:  Increase with age except in North Carolina and among smokers in
          Texas.

          Smokers had highest prevalence except among 40 or more year old
          smokers in Texas.  Association with ex-smokers variable.   Only
          significant differences was between smokers and ex-smokers in
          Montana.
* 95% confidence intervals do not overlap.

-------
                               TABLE 6

              PREVALENCE OF COUGH BY EXPOSURE AND REGION

YEARS WORKED
5
5-9
>10
CUMULATIVE EXPOSURE
LOW ( 2)
MEDIUM (2-6)
HIGH ( 6)
MONTANA
% (95% C.I.)
17 (11 - 25)
14 ( 6 - 30)
30 ( 9 - 36)
15 ( 6 - 28)
17 ( 9 - 28)
17 ( 9 - 29)
TEXAS
% (95% C.I.)
20 (10 - 35)
18 ( 3 - 50)
0 (0-32)
7 ( 1 - 22).
31 (15 - 51)
11 ( 2 - 33)
NORTH CAROLINA
% (95% C.I.)
30 (14 - 53)
33 (12 - 65)
20 ( 7 - 41)
19 ( 9 - 36)
50 (22 - 78)
22 ( 4 - 56)
COUGH = Answering yes to the question: "Do you usually cough on
        most days for as much as three months each year?"

SUMMARY:  No consistent tendency to increase with increasing years
          worked.
          No tendency for prevalence to increase with increasing
          cumulative exposure.

          No statistically significant association with either
          exposure variable.

-------
                               TABLE 7

PREVALENCE OF PHLEGM AMONG TALC WORKERS BY AGE,  SMOKING,  AND REGION

MONTANA
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
TEXAS
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
NORTH CAROLINA
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
40
% (95% C.I.)

12 ( 4 - 27)
17 ( 5 - 38)
27 (17 - 40)
20 (13 - 29)

20 ( 6 - 50)
22 ( 4 - 56)
8 ( 1 - 24)
14 ( 5 - 29)

0 ( 0 - 50)
0 ( 0 - 63)
23 ( 7 - 52)
14 ( 4 - 34)
>40
% (95% C.I.)

6 ( 0 - 25)
14 ( 4 - 34)
23 (13 - 50)
17 ( 8 - 29)

25 ( 1 - 75)
40 (15 - 73)
8 ( 0 - 35)
23 (11 - 39)

0 ( 0 - 40)
33 ( 6 - 73)
42 (22 - 66)
32 (16 - 51)
TOTAL
% (95% C.I.)

10 ( 4 - 21)
10 ( 6 - 30)
26 (16 - 36)
18

21 ( 6 - 50)
32 (15 - 57)
8 ( 2 - 21)
17

0 ( 0 - 25)
22 ( 4 - 56)
34 (18 - 54)
25
  PHLEGM = Answering  yes  to  the question: "Do you usually bring up
           phlegm from your  chest  for as much as three months each
           year?"

  SUMMARY:   Increased prevalence with age only among both smoking
            categories in North Carolina; ex-smokers and nonsmokers
            in  Texas.

            Association with smoking in Montana and North. Carolina.
            Ex-smokers highest and smokers lowest prevalence in Texas.

            No  statistically significant differences by age or smoking.
                                    2liU

-------
                                TABLE  8

              PREVALENCE OF PHLEGM BY  EXPOSUEE AND REGION

YEARS WORKED
5
5-9
_10
CUMULATIVE EXPOSURE
LOW ( 2)
MEDIUM (2-6)
HIGH ( 6)
MONTANA
% (95% C.I.)
19 (13 - 27)
12 ( 4 - 26)
30 (14 - 53)
17 ( 8 - 30)
18 ( 9 - 30)
17 ( 8 - 30)
TEXAS
% (95% C.I.)
20 (10 - 34)
9 (6-37)
13 ( 1 - 50)
7 (1-22)
27 (11 - 64)
17 ( 5 - 38)
NORTH CAROLINA
% (95% C.I.)
20 ( •? - 41)
33 (12 - 65)
25 (10 - 47)
13 ( 5 - 29)
50 (22 - 78)
33 (10 - 71)
PHLEGM = Answering yes to the  question: "Do you usually bring up
         phlegm from your chest  for as much as three months each
         year?"

SUMMARY:  No consistent tendency to increase with years worked.

          Higher prevalence in Medium and High Exposure groups in
          Texas and North Carolina, but not statistically significant.
                                  261

-------
                              TABLE 9

PREVALENCE OF DYSPNEA AMONG TALC WORKERS BY AGE, SMOKING, AND REGION
                                        AGE

MONTANA
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
TEXAS
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
NORTH CAROLINA
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
40
% (95% C.I.)
2 ( 1 - 19)
6 ( 0 - 24)
2(0- 9)
2(0- 6)
10 ( 1 - 40)
0 ( 0 - 29)
0 ( 0 - 12)
2 ( 0 - 12)
0 ( 0 - 50)
0 ( 0 - 63)
0 ( 0 - 23)
0 ( 0 - 14)
>_40
% (95% C.I.)
6 ( 0 - 25)
10. ( 2 - 28)
5 ( 0 - 21)
7 ( 2 - 16)
25 ( 1 - 75)
20 ( 4 - 60)
8 ( 3 - 45)
19 ( 8 - 37)
17 ( 0 - 40)
33 ( 6 - 73)
0 (0-15)
10 ( 3 - 24)
TOTAL
% (95% C.I.)
3 (0-11)
8 (2-21)
2 ( 0 - 8)
4
14 ( 3 - 39)
11 ( 2 - 32)
5 (0-16)
9
9 (0-37)
22 ( 4 - 56)
0 (0-10)
6
 DYSPNEA = Answering yes to the question:  "Do  you get  short  of
           breath walking with people  your own age on  level
           ground?"

 SUMMARY:  Prevalence increased with age.

           No association with smoking.

           None of the differences  were  statistically  significant.

-------
                                  TABLE 10

                PREVALENCE OF DYSPNEA BY EXPOSURE AND REGION
•
YEARS WORKED
5
5-9
^10
CUMULATIVE EXPOSURE
LOW ( 2)
MEDIUM (2-6)
HIGH ( 6)
MONTANA
% (95% C.I.)
5 (3 - 12)
0 (0 - 9)
5 (0 - 22)
6 (1 - 16)
2 (0 - 7)
3 (0 - 11)
TEXAS
% (95% C.I.)
10 (4 - 22)
0 (0 - 25)
13 (0 - 50)
7 (1 - 22)
8 (1 - 23)
11 (2 - 33)
NORTH CAROLINA
%-(95% C.I.)
5 (0 - 22)
0 (0 - 24)
10 (2 - 29)
3 (0 - 17)
0 (0 - 27)
22 (4 - 56)
DYSPNEA = Answering yes to the question:  "Do you get short of breath
          walking with people your own age on level ground?"

SUMMARY:  No consistent increase with increasing years worked.
          No consistent increase with increasing cumulative exposure.

-------
                                   TABLE 11

PREVALENCE OF PLEURAL THICKENING AMONG TALC WORKERS BY AGE, SMOKING AND REGION
                                           AGE
*
MONTANA
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
TEXAS
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
NORTH CAROLINA
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
40
% (95% C.I.)

0 (0 - 11)
14 (1 - 55)
2 (0 - 10)
2 (0 - 7)

0 (0 - 27)
13 (1 - 50)
0 (0 - 12)*
2 (0 - 12)*

0 (0 - 50)
33 (2 - 87)
8 (0 - 33)
10 (2 - 28)
£40
% (95% C.I.)

0 (0-17)
10 ( 2 - 29)
14 ( 4 - 34)
9 (3-19)

25 ( 1 - 75)
22 ( 4 - 56)
42 (18 - 71)*
32 (15 - 52)*

0 (0-50)
33 ( 6 - 73)
28 (12 - 56)
24 (10 - 41)
TOTAL
% (95% C.I.)

0 ( 0 - 8)
8 (3-27)
5 (2-13)
4

7 (0-31)
18 ( 5 - 42)
13 ( 4 - 27)
13

0 (0-27)
33 (10 - 71)
19 ( 9 - 36)
18
    SUMMARY:   Increased prevalence with increased age (significant only
              in  Texas among smokers and combined).

              No  relationship with smoking (lowest prevalence in nonsmokers,
              highest prevalence in ex-smokers') .
    *95% confidence intervals do not overlap.
                                      264

-------
                                  TABLE 12

PREVALENCE OF PLEURAL THICKENING AMONG TALC WORKERS BY EXPOSURE AND REGION

YEARS WORKED
5
5-9
>10
CUMULATIVE EXPOSURE
LOW ( 2)
MEDIUM (2-6)
HIGH (>6)
MONTANA
% (95% C.I.)
3 (1 - 8)
5 (0 - 17)
11 (2 - 33)
2 (0 - 11)
2 (0 - 11)
7 (2 - 16)
TEXAS
%(95% C.I.)
10 ( 7 - 33)
18 ( 3 - 50)
29 ( 5 - 66)
4 ( 0 - 19)
23 (11 - 42)
6 ( 0 - 25)
NORTH CAROLINA
% (95% C.I.)
15 (4 - 35)
18 (3 - 50)
21 (8 - 43)
13 (5 - 29)
10 (0 - 40)
    SUMMARY:  Tendency for prevalence to increase with increasing years
              worked.
              No consistent increase with increased cumulative exposure.
                                         > r

-------
                                         TABLE  13

PREVALENCE OF OBSTRUCTION (FEV /FVC  .70) AMONG TALC WORKERS BY AGE,  SMOKING AND  REGION
                                               AGE

MONTANA
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
TEXAS
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
NORTH CAROLINA
NONSMOKER
EX-SMOKER
SMOKER
TOTAL
40
% (95% C.I.)
2 (0 - 11)*
6 (0 - 25)
14 (6 - 25)*
9 (4 - 17)*
10 (0 - 40)
0 (0 - 50)
4 (0 - 19)
5 (1 - 17)
0 (0 - 50)
0 (0 - 63)
0 (0 - 29)
0 (0 - 16)*
40
% (95% C.I.)
29 (12 - 54)*
33 (14 - 55)
45 (26 - 67)*
37 (25 - 51)*
50 (10 - 90)
10 ( 0 - 40)
33 (12 - 65)
27 (11 - 47)
33 ( 6 - 73)
33 ( 6 - 73)
39 (16 - 63)
37 (21 - 56)*
TOTAL
% (95% C.I.)
10 ( 4 - 21)
21 (10 - 36)
22 (14 - 33)
18
21 ( 6 - 50)
5 (0-22)
14 ( 5 - 30)
13
18 ( 3 - 50)
22 ( 4 - 56)
26 (11 - 44)
23
        SUMMARY:  Obstruction increased with age in all  smoking  categories
                  (significant in Montana nonsmokers,  smokers  and  combined,
                  and combined in North Carolina).

                  Obstruction had a tendency to  be  higher  in smokers, but
                  only in Montana.
        *95% confidence intervals do not overlap.
                                             261.

-------
                                      TABLE  14

PREVALENCE OF OBSTRUCTION (FEV /FVC  .70)  AMONG  TALC WORKERS BY EXPOSURE AND REGION

YEARS WORKED
5
5-9
_10
CUMULATIVE EXPOSURE
LOW ( 2)
MEDIUM (2-6)
HIGH ( 6)
MONTANA
% (95% C.I.)
15 ( 9 - 23)
22 (10 - 38)
30 (14 - 53)
28 (16 - 43)
8 (3-18)
17 ( 9 - 29)
TEXAS
% (95% C.I.)
16 ( 7 - 30)
9 ( 0 - 37)
0 (0-32)
11 ( 3 - 27)
15 ( 5 - 33)
11 ( 2 - 33)
NORTH CAROLINA
% (95% C.I.)
17 ( 5 - 38)
20 ( 4 - 60)
32 (15 - 57)
23 (10 - 40)
29 ( 5 - 66)
22 ( 4 - 56)
      SUMMARY:  Tendency to increase with years worked in Montana and
                North Carolina.

                No association with cumulative exposure.
                                            26';

-------
                                 TABLE 15

         CHARACTERISTICS OF COMPARISON POPULATIONS FOR TALC STUDY

n
AGE (S.D.)
HEIGHT (cm) (S.D.)
YEARS WORKED (S.D.)
(RANGE)
NONSMOKERS (%)
EX-SMOKERS (%)
MEAN PACK YEARS (S.D.)
MEAN CIGARETTES/DAY (S.D.)
SMOKERS (%)
MEAN PACK YEARS (S.D.)
MEAN CIGARETTES /DAY (S.D.)
MEAN CURRENT NO 2
CONCENTRATION (ppm)
MEAN CURRENT TOTAL DUST
(mg/m3)
RESPIRABLE DUST
POTASH
875
41 (13)
176 ( 6)
16 (13)
(0-50)
20
28
23 (20)
25 (14)
52
28 (23)
25 (12)
*0.90
*3.45
N.A.
ABOVEGROUND
COAL
509
44 (12)
175 ( 6)
18 (13)
(0-55)
22
32
24 (19)
23 (12)
46
27 (18)
22 ( 9)
N.A.
N.A.
1.44 -H-
UNDERGROUND
COAL
5722
39 (13)
174 ( 6)
15 (13)
(0-56)
21
23
17 (18)
19 (12)
56
17 (14)
17 ( 8)
N.A.
N.A.
1.36 -H-
* Personal samples,  from (1,2)

N.A. = Not available.

-H- Collected between the first  and  second rounds of the National Coalworkers1
   Study.  The 25 coal mines were in both the first and second rounds of
   examinations of the coal study.
                                    268

-------
                                TABLE 16

  COMPARATIVE RATES (%) OF COUGH AND PHLEGM AMONG TALC WORKERS COMPARED
  TO POTASH MINERS, ABOVEGROUND AND UNDERGROUND COAL MINERS, STRATIFIED
  BY AGE AND INDIRECTLY ADJUSTED FOR SMOKING.
                                             AGE

COUGH
MONTANA
TEXAS
NORTH CAROLINA
POTASH
ABOVEGROUND COAL
UNDERGROUND COAL
PHLEGM
MONTANA
TEXAS
NORTH CAROLINA
POTASH
ABOVEGROUND COAL
UNDERGROUND COAL
<40
% (95% C.I.)
16 (10 - 24)
13 ( 5 - 27)
28 (11 - 52)
20 (16 - 24)
16 (10 - 23)
18 (16 - 20)
21 (14 - 29)
14 ( 6 - ,28)
13 ( 2 - 35)
25 (21 - 29)
18 (13 - 25)
32 (31 - 35)
>40
% (95% C.I.)
25 (14 - 37)
21 ( 7 - 42)
24 (10 - 43)
30 (26 - 34)
35 (30 - 41)
45 (43 - 46)
18 ( 9 - 30)
21 ( 7 - 41)
32 (16 - 51)
34 (30 - 38)
41 (35 - 47)
50 (46 - 53)
TOTAL
% (95% C.I.)
21 (15 - 28)
17 ( 9 - 28)
26 (15 - 40)
25 (21 - 29)
25 (21 - 29)
30 (29 - 32)
19 (13 - 26)
16 ( 8 - 26)
15 ( 6 - 29)
29 (25 - 33)
29 (25 - 33)
41 (39 - 41)
SUMMARY

COUGH:  All talc populations Underground coal in >^40 age group and overall.
        No difference among talc and other comparison populations.

PHLEGM:  <40 - Underground coal had greater prevalence than all populations
               except North Carolina.   No differences among the other
               populations.
         >40 - Talc populations no different from each other.
               Montana and Texas 
-------
                                TABLE 17

 COMPARATIVE RATES (%) OF DYSPNEA AND PLEUBAL THICKENING AMONG TALC
 WORKERS COMPARED TO POTASH MINERS, ABOVEGROUND AND UNDERGROUND COAL
 MINERS, STRATIFIED BY AGE AND INDIRECTLY ADJUSTED FOR SMOKING
                                          AGE

DYSPNEA
MONTANA
TEXAS
NORTH CAROLINA
POTASH
ABOVEGROUND COAL
UNDERGROUND COAL
PLEURAL . THICKENING
MONTANA
TEXAS
NORTH CAROLINA
POTASH
ABOVEGROUND COAL
UNDERGROUND COAL
<40
% (95% C.I.)
3 (1 - 8)
4 (0 - 15)
0 (0 - 16)
5 (3 - 7)
2 (1 - 7)
10 (9 - 11)
2 (0 - 6)
3 (0 - 14)
12 (2 - 34)
0 (0 - 1)
0 (0 - 3)
0.2 (0-0.3)
>40
% (95% C.I.)
7 ( 2 - 16)
15 ( 4 - 35)
14 ( 4 - 31)
12 ( 9 - 16)
28 (23 - 34)
41 (40 - 42)
11 ( 4 - 22)
32 (15 - 53)
25 (11 - 44)
3(2- 4)
1(0- 3)
1 (.5 -1.5)
TOTAL
% (95% C.I.)
5 ( 2 - 10)
9 ( 4 - 18)
6 ( 1 - 17)
8 ( 6 - 10)
14 (11 - 17)
24 (23 - 25)
6 ( 3 - 11)
17 ( 9 - 28)
18 ( 9 - 31)
2(1- 3)
0.3 (0 - 1)
1 (.5 - 1.5)
SUMMARY

DYSPNEA:  <40


          >40
No difference among talc, potash, and aboveground coal
populations.  Montana less than underground coal.
No difference among talc and potash populations, and all
had less dyspnea than underground coal.  Montana had
less prevalence than aboveground coal.
        Total - No difference among talc and potash populations.
                Underground coal had greater prevalence than all
                populations.   Montana had less dyspnea than aboveground
                coal.

PLEURAL THICKENING:  <40 - No difference among talc populations.   North
                           Carolina elevated compared to potash and
                           underground coal.

           >_40 and Total - No differences among talc populations.  All
                           populations had greater prevalence than
                           nontalc populations.
                                       2 VI)

-------
                              TABLE 18

COMPARATIVE RATES OF OBSTRUCTION (FEV../FVC <.70) AMONG TALC WORKERS
                COMPARED TO OTHER MINING POPULATIONS.
       STRATIFIED BY AGE AND INDIRECTLY ADJUSTED FOR SMOKING
                                       AGE

MONTANA
TEXAS
NORTH CAROLINA
POTASH MINERS
UNDERGROUND COAL
ABOVEGROUND COAL
<40
% (95% C.I.)
8 (4-15)
5 (5-17)
0 (0-17)
9 (6-12)
11 (10 - 12)
8 (4-14)
>40
% (95% C.I.)
37 (25 - 50)
29 (15 - 51)
36 (21 - 56)
33 (29 - 37)
32 (31 - 33)
31 (26 - 36)
TOTAL
% (95% C.I.)
22 (16 - 29)
16 ( 9 - 27)
17 ( 8 - 30)
20 (17 - 23)
21 (20 - 22)
19 (16 - 23)
 SUMMARY;   <40 - No differences
            >40 - No differences
          Total - No differences
                                    i i **' 1
                                    d ( 1

-------
                                               TABLE 19
     SUMMARY OF PULMONARY FUNCTION REGRESSION MODELS AMONG TALC WORKERS (COMBINED AND BY REGION)
  MODEL:  PFT = a+gsex) +B(mine) +e3 (age) +3^(height) +65(pack years)  +36 (cumulative exposure)



FEVj^ (mL)
Montana
Texas
North Carolina
FVC (mL)
Montana
Texas
North Carolina
PEAK FLOW (mL/sec)
Montana
Texas
North Carolina
FEF5Q (mL/sec)
Montana
Texas
North Carolina
FEF7_ (mL/sec)
Montana
Texas
North Carolina

SEX

-572
-608
— _
(-212)
-865
-1038
— —
(-341)
-1592
-1674
	
(-1298)
N.S.
N.S.
_ —
N.S.
N.S.
N.S.
___
N.S.

MINE

N.S.
N.S.
N.S.
	
N.S.
N.S.
*
	
N.S.
N.S.
N.S.
	
N.S.
N.S.
*
	
N.S.
N.S.
N.S.
	

AGE

-30
-30
-26
-25
-19
-13
-21
(- 9)
-39
-35
(-25)
-54
-51
-57
-33
-60
-37
-42
-26
-40

HEIGHT

+ 46
+ 40
+ 33
+ 77
+ 64
+ 53
+ 57
+108
+ 80
+ 74
+ 63
+115
+ 27
+ 30
(- 5)
(+ 36)
+ 12
(+ 14)
(- 2)
+ 23)

PACK YEARS

- 7
- 8
(- 8)
(- 9)
(- 3)
(- 3)
(- 8)
(- 6)
-15
(-13)
(-17)
-25
-20
-19
(-22)
(-17)
- 9
-10
(- 8)
(- 6)
CUMULATIVE EXPOSLRE
(95% CONFIDENCE INTERVAL)
UPPER LOWER
(-3 +3)
(- 27 + 1)
(-4 +3)
(- 31 + 61)
(-5 +5)
- 35 - 6 *
(-5 +3)
(- 84 + 26)
(- 12 + 5)
(- 57 + 56)
(- 13 + 7)
(-179 +137)
(-4 + 11)
(- 36 + 32)
(-6 + 12)
(- 55 + 94)
(-4 +3)
(- 18 + 14)
(-5 +3)
(-7 + 52)
2
r

.64
.71
.48
.60
.65
.74
.48
.62
.47
.50
.21
.54
.36
.40
.22
.44
.51
.55
.40
.63
N.S. or ( ) = not statistically  significant.

If * or no  ( ), then p  <.05.

-------
                                      TABLE 20




        MEAN ADJUSTED VALUES OF PULMONARY FUNCTION OF TALC WORKERS (n = 292)
MODEL:  PFT = a+g^sex)  +3-(age)  +3,(height)  +3, (smoking status) +e_(region)
        +$, (cumulative exposure group)
                                               MEAN (S.E.)

SMOKING STATUS
NONSMOKERS
EX-SMOKERS
SMOKERS
REGION
MONTANA
TEXAS
NORTH CAROLINA
CUMULATIVE EXPOSURE
GROUP
LOW
MEDIUM
HIGH
FEV.!^
(mL)
3.72* (.09)
3.59 (.09)
3.50 (.08)
3.58 (.06)
3.52 (.10)
3.71 (.11)
3.10 (.08)
3.68 (.08)
3.53 (.09)
FVC
(mL)
4.59 (.10)
4.50 (.09)
4.48 (.08)
4.65* (.06)
4.35 (.11)
4.57 (.12)
4.64* (.08)
4.59* (.09)
4.35 (.10)
PEAK FLOW
(mL/sec)
8.32 (.23)
8.62* (.25)
7.94 (.21)
8.51 (.16)
8.45 (.27)
7.92 (.29)
8.27 (.20)
8.28 (.23)
8.33 (.25)
FEF
*^50
(mL/sec)
4.63 (.19)
4.57 (.21)
4.20 (.18)
4.07* (.13)
4.72 (.23)
4.61 (.24)
4.23 (.17)
4.57 (.19)
4.59 (.21)
FEF
75
(mL/sec)
1.66 (.09)
1.47 (.10)
1.40 (.08)
1.41 (.06)
1.51 (.10)
1.61 (.11)
1.42 (.08)
1.59 (.09)
1.53 (.10)
  Significant difference at .05 level.

-------
                                              TABLE 21

             MEAN PERCENT PREDICTED PULMONARY FUNCTION OF MONTANA, TEXAS, NORTH CAROLINA
          TALC WORKERS COMPARED TO COMPARISON GROUPS, ADJUSTED FOR AGE, HEIGHT, AND SMOKING

COMPARISON POPULATIONS
MALES ONLY (n = 251)
POTASH
UNDERGROUND COAL
ABOVEGROUND COAL
% PREDICTED PULMONARY FUNCTION = (observed/predicted) x 100
FEV
7, (S.E.)
98.85 (1.01)
97.55 (1.01)*
96.60 (1.01)*
FVC
% (S.E.)
99.60 (.84)
95.09 (.80)*
96.62 (.83)*
PEAK FLOW
% (S.E.)
93.19 (1.03)*
100.19 (1.13)
112.43 (1.29)4
FEF
% (S.E.)
95.62 (2.10)*
95.62 (2.17)*
92.93 (2.00)*
FEF
25
% (S.ET)
88.23 (3.12)*
82.58 (2.75)*
80.76 (3.92)*
* »  >2 S.E. less than 100

+ -  >2 S.E. greater than 100

-------
                                 Discussion
Dr. Fraumeni (NCI):  As this study progresses, will you be able to shed
any light on the relationship of talc exposure and cancer.

Dr. Gamble (NIOSH):  The population here is probably not too good for a
mortality study at this point.  The exposure histories, as we showed,
are relatively short, even in North Carolina where it is only ten years.
I think this population should be followed-up because of the pleural
thickening and the concern for possible mesothelioma.  But T* do not know
that we are going to have any answer for that for awhile.

I think the pleural thickening is of possible concern because of the
relationship of pleural thickening in asbestos exposure.  It is
interesting that in Montana, where no asbestos fibers have been found,
there was still an increase in pleural thickening.  These populations
should be followed.
                                            r«

-------
                   LUNG CANCER IN

      THE NATIONAL COAL WORKERS'  AUTOPSY STUDY
V. Vallyathan, M.  Attfield,  R.  Althouse,  N.  Rodman*

            C. Boyd* and F.  H.  Y.  Green


 Pathology Section,  Lab. Investigations Branch,
 Division of Respiratory Disease Studies, Na-tional
 Institute of Occupational Safety and Health,  944
 Chestnut Ridge Road,  Morgantown,  West Virginia
                                            26505

                        and
*
 Department of Pathology, School of Medicine,  West
 Virginia University Medical Center, Morgantown,
 West Virginia  26506
                               276

-------
                             Summary



    The 2410 cases in the National Coal Workers' Autopsy Study



were analyzed to determine whether factors in the underground



mining environment influenced the incidence or histogenesis of



lung cancer.  The major factor in the development of lung cancer



in coal workers appeared to be cigarette smoking.  We could find



no effect due to duration of underground exposure.   An interesting



finding was an apparent increase in the number of cases with adeno-



carcinoma.  This latter finding supports the concept that the



histogenesis of bronchial neoplasm is influenced by environmental



factors.
                                    277

-------
                        Introduction



   It is being increasingly recognized that most respiratory



cancers are associated to some extent with environmental factors.



Among the environmental factors, cigarette smoking, asbestos



exposure and exposure in the metal and mineral mining industries




have been considered significantly influential in the development



of lung cancer.  (Stocks, 1966; 1967; Selikoff, et al.  1974;



Axelson and Sundell; 1976, Archer, et al.  1973; Newman,  et al.



1974; Auerbach, et al. 1975; Wagoner, et al.  1967; 1973).  There



is also evidence suggesting a direct relationship between air



pollution and pulmonary cancer (Carnow and Meier, 1973;  Hagstrom,



et al. 1967; Schneiderman and Levin, 1972; Levin, et al.  1960).



   There are many reasons for suspecting that the environment of



the coal mine may influence the incidence  of  lung cancer.  Several



studies have been reported in the literature  mainly based on death



certificate and necropsy data (Kennaway and Kennaway, 1947;  1953;



James, 1955; Carroll, 1963; Enterline, 1964;  1972; Liddell,  1973;



Costello, et al.  1974; Ortmeyer, et al.  1974; Rooke, et  al.  1979;



Scarano, et al. 1972; Mooney, 1975; Abraham,  1978; Cochrane, et al.



1979).  The majority of these studies have shown a slightly decreased



incidence of lung cancer in coal workers.   It is difficult to draw



firm conclusions from these studies as most of them have suffered



from important epidemic!ogical limitations, for example,  the




majority of studies have not been adequately  controlled  for smoking.



   In the United States squamous cell carcinoma is considered the



most common type of lung neoplasm (Clifton and Luomnnen,  1968).




Also, squnmous cell carcinoma is known to  be  the most prevalent

-------
type of cancer in smokers  (Kreyberg, .1962; Doll, et al. 1957;



Vincent, et al. 1965; Weiss, et al. 1972).  On the other hand,



small cell carcinoma has been reported to be the most prevalent



type of lung neoplasm in uranium miners  (Archer, et al. 1973;



Auerbach, et al.  1975) and adenocarcinoma is more prevalent in



asbestos workers (Spencer, 1977; Whitwell, et al. 1974; Heuper,



1966; Hourihane and McCaughey, 1966).  Thus, knowledge of the



frequency distribution by histological type in a given population



may provide valuable clues to the etiology of lung cancer.



   The purpose of this study was to determine whether there is an



association between histological type of coal workers' lung cancer



and factors in the underground mining environment.  The study was



based on the material in the National Coal Workers' Autopsy Study



(NCWAS).  Comparisons were made between histological type, years



of underground exposure, specific occupation within the mine and



the type and severity of coal workers' pneumoconiosis (CWP).








                    Materials and Methods



   In 1969 the United States Congress passed the Coal Mine Health



and Safety Act, which provides free autopsies for all underground



coal workers.   Since its inception from 1972 through 1977 we have



collected 2410 cases from 22 states.  Each case submitted to this



progra.m included  full demographic data, occupational and smoking



histories together with a detailed autopsy report and pulmonary



tissue.  The NCWAS population is similar to the general working



miner population with regard to geographic distribution, occupation



within the mines  and smoking history.  However, the mean age, total



number of years worked and the smoking history are higher than the

-------
general working miner population (Abraham, 1978).



   The pathological material consisted of 3 or more histological



sections (1 x 1.5 cm) with 3 corresponding or different blocks of



tissue which had been prepared from post-mortem tissues.  Histo-



logic evaluation of the type of neoplasm was determined by 4



pathologists according to the WHO classification (Kreyberg, et al.




1967) with minor changes as outlined below.



1.  Epidermoid or squamous carcinoma;



    a.  Squamous cells with keratin, keratin pearls and intracel



        lular bridges (well differentiated).



    b.  Squamous cells with intracellular bridges  or pre-keratin



        (moderately differentiated).



    c.  Squamous cells with characteristic growth  pattern, sheet-



        like arrangements and occasional cells with pre-keratin



        (poorly differentiated).



2.  Small cell anaplastic carcinoma.



    a.  Small cell lymphccytic (oat cell) type.



    b.  Small cell polygonal type.



    c.  Small cell fusiform type.



3.  Adenocarcinoma



    a.  Acinar type with mucin.




    b.  Papillary type bronchoalveolar adenoca.rcinoma with mucin.



    c.  Poorly differentiated acinar type with occasional mucin



        containing cells.




4.  Mixed squamous cell and adenocarcinoma with keratin and mucnn.



5.  a.  Large cell carcinoma with mucin.



    b.  Large cell carcinoma without mucin.
                                 28C

-------
    c.   Large cell carcinoma with giant cells.



    d.   Clear cell carcinoma.



6.  Mesothelioma.



    a.   Diffuse.



    b.   Localized.



7.  Others (Leiomyosarcoma,  fibroscarcoma,  etc.



   In 202 cases there was sufficient material available to make a



microscopic assessment.   Evaluation on a minimum of 3 histological



slides  were made independently by all the members of the panel;



that is, without prior knowledge of autopsy findings or smoking



histories or years of mining history; and results recorded.   The



same group of slides were re-evaluated simultaneously by the panel



one week later using a multi-headed microscope  and the results of



earlier independent interpretations were compared.  When disagree-



ments were found among the members of the panel  or when mixed tumor



patterns and poor differentiation were observed, additional slides



stained with a battery of special stains were obtained.  The histo-



logical sections,  stained with H § E, keratin stain, Fontana stain



and mucicarmine stains were re-examined by the  panel.  When these



preparations were studied by the panel a unanimous opinion on the



type of carcinoma was usually reached.  The total number of cases



disputed or demanding further characterization  by special stains



for a unanimous opinion was 66.



   Histopathological evaluation of the pneumoconiosis was perform-



ed on all the 202  cases by two of the panel members  (V.V. and F.H.



Y.G.).   In this evaluation it was assumed that  at post-mortem the



lung tissues selected for microscopic examination by the partici



pating  NCWAS pathologists were representative of the whole lung
                                   281

-------
for the purposes of determining the extent, severity and type of
pncumoconiosis.  Classification and grading of coal workers pneumo-
coniosis was made according to standards established by a panel of
the College of American Pathologists (Kleinerman, et al. 1979).
   The 221 cases identified with lung cancer were matched with
cases without lung cancer.  The following groupings were chosen:
   1.  Lung cancer cases matched with all non-lung cancer cases.
       The matching variables were:  race,  exact age, mining
       regions represented by eastern Pennsylvania (anthracite)
       the rest of appalachia and the mid west (bituminous)  and west,
       smoking status (non-smoker,  cigarette smoker (current or
       ex), pipe or cigar smoker).   A total of 175 matches were
       found.
   2.  The same as (1) above with the exception that smoking
       status was replaced by pack years (5 pack-year intervals).
       A total of 135 matches were found.
   Statistical tests were performed to  determine the relationship
between occupational exposure, smoking,  job differences and the
prevalence of lung cancer.   The analyses  were made using the matched
pair t-tests.

                           Results
   Table I presents relevant descriptive information for the entire
sample.   The 2410 miners in the NCWAS had an average age at death
of 64.0 + 11.0 years with 27.0 ± 13.0 years of underground mining.
Seventy-two percent of the total sample had a history of cigarette
smoking with a mean pack year smoking history of 25.0 + 19.0.  Two
hundred and twenty one (221) cases were  identified in which carcin-

-------
oma of the lung was mentioned on the autopsy .report.   This repre-



sented 8.8% of the total sample.  The mean age at death of miners



with lung cancer was '65.0 + 10.0 years.  In the 221 lung cancer



cases 200 (90%) smoked cigarettes, 3 were pipe or cigar smokers



only and 18 were non-smokers.  In 183 cases lung cancer was deter-



mined by the pathologist to be the underlying cause of death.



Table II shows the distribution of lung cancer cases  and cases with



cancer at other sites in 22 coal mining states.  The  proportion of



lung cancer cases in Illinois and Ohio appears to be  increased.



   Analysis of matched group 1 showed a significant difference



(P = 0,03) in mean pack years between the smokers in  the two



groups.  The mean pack years for lung cancer cases was 31.0 ± 20.0



compared to 24.0 + 22.0 for non-lung cancer cases.



   Among the 175 matched lung cancer cases 13 were non-smokers



with an average underground work history of 26.0 + 17.0 years.



The average underground work history of the matched control popu-



lation was 35.0 + 12.0 years.  This difference was not statistically



significant (P = 0.12) .



   As there was a significant difference in the amount smoked



between the lung cancer and non-lung cancer cases we  felt a more



precise, matching of smoking history \\ras required to determine an



independent effect due to occupation.  Group 2 illustrates this



matching.  In both smoking and non-smoking miners no  significant



differences were detected between years underground,  specific occu-



pation and the presence of lung cancer.



   Table III shows the relationship between the histological type



of lung cancer and age, mining and smoking histories  in 202 cases



on which histologic.nl material was nvailnble.  The predominant
                                2 8 a

-------
cell type was adenocarcinoma (301),  followed by small cell (28%)
and epidermoid (24%)  carcinomas.   For each of these three major hist-
ological types of lung cancer no  significant differences were  detected in
mean age, pack years  smoking history or years in underground mining.
   The type of CWP lesion and its frequency of occurrence with
the different cellular types of lung cancer are shown in Table IV.
There appears to be a lower incidence of macular CWP in cases with
small cell carcinoma.
   Due to the unexpectedly high incidence of adenocarcinoma in
this series an attempt was made to  determine the site or origin of
these tumors within the lung.  These details were extracted from
the clinical, surgical and autopsy  reports.  Out of a total of 60
cases of adenocarcinoma 23% of the  tumors originated in the upper
lobes, and 16% originated in the  lower lobes.  The remaining tumors
either originated from the trachea,  right middle lobe,  main stem
bronchi or the site of origi7i could  not be determined.

                         Discussion
   Whether exposure to coal mine  dust increases the chances for
developing carcinoma  of the lung-  is  a matter of considerable impor-
tance which has not been resolved in spite of several investigations
(Green and Laquer, "1980).   There  are many reasons for suspecting
the environment of the coal mine  may influence the incidence of
lung cancer.   Coal mine dust is known to contain polycvclic aromatic
hydrocarbons and several metal carcinogens such as beryllium, cad-
mium, chromium, cobalt, lead, maganosc and nickel (Berg and Burbank,
1972).  Moreover,  cool dust due to  its high adsorbing ability may
facilitate the transportation of  polycyclic hydrocarbons from

-------
cigarette smoke.  It has been shown that cigarette smoking miners
have an 8 fold increase of lung cancer deaths as compared with non-
smoking coal miners (Jacobson, 1976).  Several studies suggest that
carcinoma of the lung is less frequent in coal miners than in non-
miners of a similar age group (James, 1955; Doll, 1959; Goldman,
1965; Costello, et al. 1974; Ortmeyer, et al.  1974; Liddell, 1973;
Kennaway, and Kennaway, 1947; Rooke, et al. 1979; Cochrane, et al.
1979).  In contrast to these reports other mortality studies con-
ducted in the United States have shown an increased SMR for lung
cancer in coal workers (Enterline, 1964; 19.7-2-; Mooney, 1975).  The
cause of this disparity in findings is a reflection on the complex-
ity of the problem.  The difficulties in determining the true
incidence of lung cancer in £oal miners are mainly due to regional
variations in coal mines a:nd dust concentrations, difficulties in
obtaining control populations from the same areas, and separation
of the influence of mining independent of smoking.
   In this study we could find no evidence that the development of
lung cancer is influenced by duration of underground exposure.
However, we have confirmed the well recognized ettological relation
ship between cigarette smoking and lung cancer.
   We found a greater percentage of adenocarcinoma and small cell
carcinoma in the NCWAS cases than reported in comparable studies.
There is some evidence that lung cancers due to smoking are mainly
squamous or small cell in type and adenocarcinomas are endogenous
growths (Kreyburg,  1962;  ]967;  Weiss, et al.  1972).  In this study
90% of the NCWAS cases with lung cancer had a history of cigarette
smoking.   Thus, the descrepnncy noted in this study between predom-
inant histologicnl  type and smoking history may be related to

-------
occupation.   In a study of histological cell -types in asbestos



workers with lung cancer, adenocarcinoma was also found to be the



most common type (Whrtwell, et al. 1974).  Ionizing radiation, on



the other hand, is known to induce predominantly small cell, undif



ferentiated type of carcinomas and the relative frequency of this



type of tumor rises with cumulative radiation exposure (Saccomanno



1971; Archer, et al. 1973; Horacek, et al. 1977;Auerbach, et al.



1975).  The relative predominance of adenocarcinoma in the NCWAS



cases raises the interesting possibility that factors in the coal



mining environment influence the histogenesis of bronchial carcin-



omas.  However, this conclusion cannot be drawn with certainty as



a suitable control group of non-miners with lung  cancer were not



available for us to study.  There is also some evidence that the



incidence of adenocarcinoma of the lung in the general population



 is increasing (Vincent, et al. 1977).



   In conclusion, we would like to stress that this report is



provisional.  We are currently updating the study to include cases



submitted during 1978 and 1979, and are tracking  down additional



material on cases lacking histological sections of the lung tumor.



We hope to better define the relationship between smoking and lung



cancer in coal workers and to compare the severity of CWP in lung



cancer cases with matched controls.

-------
                                                    TABLE  I


                                             NCWAS CASE DESCRIPTION




                                       SAMPLE SIZE                  2410


                                       MEAN AGE + S.D.              64 ± 11


                                       SMOKERS                      1709 (72%)


                                       NON - SMOKERS                619  (26%)


                                       PIPE SMDKERS                 47   (2%)


                                       MEAN PACK - YEARS + S.D.     25 ± 19


                                       MEAN MINING YEARS + S.D.     27 ± 13
ro
oc

-------
                          TABLE  II



PREVALENCE  OF LUNG CANCER AND CANCER OF OTHER ORGANS BY STATE
STATE
ALABAMA
ARKANSAS
CALIFORNIA
COLORADO
DIST. OF COLUMBIA
ILLINOIS
INDIANA
KANSAS
KENTUCKY
MARYLAND
MISSOURI
NEW MEXICO
OHIO
OKLAHOMA
PENNSYLVANIA
TENNESSEE
TEXAS
UTAH
VIRGINIA
WASHINGTON
WEST VIRGINIA
WYOMING
TOTALS
TOTAL DEATHS
3
3
1
33
1
88
8
8
120
6
1
5
89
4
1176
3
1
9
71
1
666
38
2336
ALL CANCERS ^
EXCEPT LUNG
0
1
1
5 15
0
13 15
1
4
9 8
1
0
1
10 11
1
163 14
0
0
0
6 8
9
91 14
9 24
316 14
LUNG CANCER
1
1
0
2
0
16
0
1
8
1
0
0
16
1
103
1
0
0
5
0
56
1
213
%
-
-
-
6
-
18
-
-.
7
-
-
-
18
-
9
-
-
-
7
-
8
3
9
                              288

-------
,	I
                          TABLE  III

CARCINOMA OF THE LUNG:  CELL TYPES BY AGE,  SMOKING AND MINING


                                          Pack     Mining
                                          Yrs.      Yrs.

                                           28*       30+
                                           20        14

                                           33*       31+
                                           19        12"

                                           30*       32+
                                           23        11"

                                           35*       33+
                                           16        14"


                                           30*       41 +
                                           18        14"


                                           37*       19+
                                           27        13"

                                           32*
                                            3


  *MEAN + 2 STANDARD DEVIATION
Adeno - Carcinoma
Epidermoid
Small Cell
Large Cell
Mixed
Adeno -Carcinoma
Epidermoid
Broncho -Alveolar
All Cases
%
30
24
28
9
8
1
100
Age
63*
10*
68*
10
64*
9
67*
10
65*
9
62*
10
65*
2

-------
                         TABLE  IV



THE TYPE OF LESION AND ITS FREQUENCY OF OCCURRENCE IN THE



                 CASES WITH LUNG CANCER
Cell Type
Adeno carcinoma (60)
Epidermoid (48)
Small Cell (56)
Large Cell (19)
Mixed
Adenocarcinoma (16)
Epidermoid
Broncho-Alveolar (3)
No-CWP
Lesions
23
17
34
16
12
33
Macule Nodule PMF
77 23 5
81 35 8
64 23 5
74 32 16
88 44 19
67
Silicosis
8
8
2
16
6
--

-------
                         References

Abraham, J. L.  Recent advances in pneumoconiosis.  The'
pathologists role in etiologic diagnosis.  The Lung:  TAP
Monograph No. 19, Williams and Wilkins Co., Baltimore, p. 96
1978.

Archer, V. E., Wagoner, J. K. and Lund.in, F. E. Uranium
mining and cigarette smoking effects on man. J. Occup. Med.
15:204, 1973.

Auerback, 0., Garfinkel, L. Histological type of lung cancer
in relation to smoking habits, year of diagnosis and sites of
metastasis.  Chest, 67:382, 1975.

Axelson, 0. and Sundill, L.  Mining, lung cancer and smoking.
Scand. J. Work Environ. Health, 4:46, 1976.

Berg, J. and Burbank, F.  Correlations between carcinogenic
trace metals in water supplies and cancer mortality.  Ann. N. Y.
Acad. Sci., 199:249, 1972.

Carnow, B. and Meier, P.  Air pollution and pulmonary cancer.
Arch. Environ. Health, 27:207, 1973.

Carroll, R. The influence of lung scars on primary lung cancer.
J. Path. Bact., 83:293, 1962.

Clifton, E.E. and Luomanen, K. J.  Relationship of pathology
to diagnosis and treatment.  In:  Lung Cancer:  A study of 5000
memorial cases (W.L. Watson, Ed.), p. 376, 1968.

Cochrane, A. L.,  Haley, T.J.L., Moore, R., and Hole, D.   The
mortality of men in the Rhondo Fach, 1950-1970. Brit. J.  Ind.
Med., 36:15, 1979.

Costello, J., Ortmeyer, C.E. and Morgan, W.K.C.  Mortality from
lung cancer in U. S. coal miners.  Am. J. Pub. Health, 64:222,
1974.

Doll, R., Hill, A.B. and Kreyberg, L.  The significance of cell
type in relation to the etiology of lung cancer. Brit. J. Cancer,
11:43, 1957.

Doll, R. 'Occupational lung cancer.  A review.  Brit. J.  Ind.
Med.  16:181, 1959

Enterline, P.E.   A review of mortality data for American coal
miners.   Ann.  N.  Y.  Acad.  Sci. 200:260, 1972.

Enterline, P.  E., Mortality rates among coal miners. Am.  J. Pub.
Health,  54:758,  1964.

Goldman, K, P.   Mortality of coal miners from carcinoma of the
lung.  Brit, J.  Ind. Med.  22:72, 1965.

Green, F. H. Y.  and Laquer, W. Coal workers pneumoconiosis.
Pathology Annual, 1980 (In press).


                                    291

-------
Hagstrom, R.  M.,  Sprague, H. A., Landau, E.  The Nashville air
pollution study.   VII Mortality from cancer in relation.to air
pollution.  Arch. Environ. Health 15:237, 1967.

Horacek, J.,  Placek, V. and Seve, J.  Histologic types of
bfonchogenic  cancer in relation to different conditions of
radiation exposure.  Cancer, 40:832, 1977.

Hourihane, D. B.  and McCaughey, W.T.E.  Pathological aspects of
asbestosis.   Postgraduate Med.  J. 42:613, 1966.

Hueper, W.C.   Occupational and environmental cancer of the
respiratory tract.   Recent Results in Cancer Research, Vol. 3,
p. 43, Springer,  Berlin, 1966.

Jacobson, M.   Dust exposure,lung disease, and coal miners
mortality.  Ph.D. Thesis, Edinburgh University, 1976.

James, W. R.  L.   Primary lung cancer in south Wales coal
workers with  pneumoconiosis. Brit. J. Ind. Med. 12:87, 1955.

Kennaway, E.  L.  and Kennaway, N. M.   A further study of the
incidence of cancer of the lung and larynx.  Brit. J. Cancer,
1:260, 1947.

Kennaway, F.  L.  and Kennaswy, N. M.   The incidence of cancer of
the lung in coal  miners in England and Wales.  Brit. J. Cancer
7:10, 1953.

Kleinerman,  J. ,  Green, F., Harley, R. A., Lapp, L., Laquer, W.,
Naeye, R. L., Pratt, P., Taylor, G., Wiot, J.  and Wyatt, J.
Pathology standards for coal workers' pneumoconiosis. Arch.
Path, and Lab.  Med. 103:375, 1979,

Kreyberg, L.   Histological lung cancer types.   A morphological
and biological  correlation, Acta Patholo. Microbiol. Scand.
Suppl. 157:1, 1962.

Kreyberg, L., Liebow, A.A., Uehlinger, E.A.  Histological typing
of lung.tumors.   International  Histological Classification of
Tumors, No.  1,  Geneva, WHO, 1967.

Levin, M.L.   Cancer incidence in urban and rural areas of New
York State.  J.  Natl. Cancer. Inst. 24:1234, 1960.

Liddell, P.  D.  K.  Mortality of British coal miners in 1961.
Brit. J. Ind. Med.  30:15, 1973.

Mooney, F. S.  Coal workers' pneumoconiosis and carcinoma of
the lung.  Lancet,  1:45, 1975.

Newman, J. A.,  Archer, V. E., Saccomanno, C.,  Kuchner, M. ,
Auerbach, 0., Grondahl, R. D. and Wilson, J. C.  Histologic
types of bronchogcnic carcinoma among members of copper mining
and smelting  communities.  Ann. N. Y. Acad. Sci. 271:260, 1976.

Ortmcyer, C.  F,.,  Costcllo, J.,  Morgan, W.K.C., Swcckcr, S. and
Peterson, M.   The mortality of Appalachian coal miners, 1963 to
197].  Arch.  F.nviron. Health, 29:67, 1974.

-------
Rooke, G. B.,  Ward, R. G.,  Dempsey, A. N., Dowler, J. B. and
Whitaker, C. J.  Carcinoma of the lung in Lancashire coal
miners.  Thorax, 34:229, 1979.

Saccomanno, G., Archer, V.E., and Auerbach, 0.  Histologic
types of lung  cancer among uranium miners.  Cancer, 27:515, 1971.

Scarano, D., Fadali, A. M.  and Lemole, G. M.  Carcinoma of the
lung and anthracosilicosis.  Chest, 62:251, 1972.

Schneiderman,  M.A.  and Levin, D.L.  Trends in lung cancer.
Mortality, incidence, diagnosis and treatment, smoking and
urbanization.   Cancer, 30:1320, 1972.

Selikoff, I. J. , Churg, J.  and Hammond, B.C.  Asbes.tos exposure
and neoplasia. JAMA,188:22, 1974.

Spencer, H.  Pathology of the Lung.  Third Ed., Vol. 2. Pergamon
Press, New York, p. 773, 1977.

Stocks, P.  Recent epidemiological studies of lung cancer
mortality, cigarette smoking and air pollution, with discussion
of a new hypothesis of causation.  Brit.  J. Can. 20:595, 1966.

Stock, P.  Lung cancer and bronchitis in relation to cigarette
smoking and fuel consumption in twenty countries.  Brit. J.
Prev. Soc. Med. 21:181, 1967.

Vincent, T.N., Satterfield, J.V. and Ackerman, L. V.  Carcinoma
of the lung in women.  Cancer, 18:559, 1965.

Vincent, R. G., Pickren, J., Lane, W.W.,  Bross, I., Takita, H.
Houten, L., Gutierrez, A.C. and Rzepka, T.  The changing histo-
pathology of lung cancer.  A review of 1682 cases.  Cancer,
39:1647, 1977.

Wagoner, J. K., Archer, V.E., Carroll, B. E., Holaday, D.A.
and Lawrence,  P.A.   Cancer mortality patterns among U.S. uranium
miners and millers, 1950 through 1962.  J. Natl, Cancer Inst.
32:738, 1964.

Wagoner, J.K., Miller, R.W. and Lundin, F.E.  Unusual cancer
mortality among a group of underground metal miners. New. Eng.
J. Med. 296:284, 1973.

Weiss, W., Boucot,  K. R., Seidman. H., and Carnahan, W.J.  Risk
of lung cancer according to histoiogic type and cigarette dosage.
JAMA, 13:799,  1972,

Weiss, W., Moser, R. L. and Auerbach, 0.   Lung cancer in chloro-
methyl ether workers.  Am.  Rev. Resp. Dis. 120:1031, 1979.

Whitwell, F.,  Newhouse, M.L., and Bennett, D.R.  A study of the
histological cell types of lung cancer in workers suffering from
asbestosis in the United Kingdom.  Brit.  J. Ind. Med. 31:298. 1974

-------
                               Discussion
Dr.  Lingeman:   I would like to know if there is any correlation between
the type of cancer and the amount of silicon present in the coal.  For
example, do you know how much silicon is present in the coal mined in
Pennsylvania,  versus that mined in Ohio or in Illinois?  Is there a high
level of silicon in the coal  from these states?

Dr.  Vallyathan (NIOSH):  Yes, there is a difference in the amount of
silicon present in the different types of coal.  We have not determined
whether there is any correlation between the frequency or type of cancer
and the silicon content of the coal mine dust.

Dr.  Spirtas (NCI):  From the abstract I take it for granted that this is
a case control study.   Is that correct?

Dr.  Vallyathan (NIOSH):  No,  it is not a case control  study.  We have
not been able to obtain a suitable control autopsy population.  There-
fore all our comparisons are internal.

Dr.  Spirtas (NCI):  What is your study design?   Is there some way that
you will try to determine whether this series of cases are represen-
tative cases or is this a report on a series of cases?

Dr.  Vallyathan (NIOSH):  The National Coal Workers' Autopsy (NCWAS)
cases represent only about ten percent of the deaths in the Nations'
coal workers.   It is a selected population due  to its  voluntary nature.
The NCWAS population is demographically similar in many respects to the
living miners in our National coal study.

Dr.  Spirtas (NCI):  I  take it that the voluntary part  of the program is
on the part of the physician.  Is there some reason-, to suspect that'
certain types of physicians or certain types of cases  are volunteered
for autopsy?

Dr.  Vallyathan (NIOSH):  No,  it is not the physician who determines
whether to submit a case or not.  An autopsy is requested and submitted
by the next-of-kin.

Dr.  Spirtas (NCI):  Is this for a claim for benefits under the Coal
Mining Act?

Dr.  Vallyathan (NIOSH):  Yes.

Dr.  Marcy (EPA):  I presume these cases represent a stable population
in terms of their work history in the industry  and residency in a
state.  Have you verified their residency and other occupational
exposures?
                                     23-1

-------
Dr. Vallyathan (NIOSH):  Yes, these cases do represent a stable population
with a coal mining history of 10 or more years in the individual states.
We have ascertained the residency of all cases included in this study.
However, details of occupational histories other than in the coal mining
industry are not available.

Dr. Kraybill (NCI):  Did you make any radioactivity measurements?   I
understand when coal is burned an effluent of radioactive material  is
released.  Is there any radium or uranium type ores connected with  coal?
Is there a difference in the radioactivity levels of coal mine to coal
mine from state to state?

Dr. Vallyathan (NIOSH):  No, we did not make any radioactivity measure-
ments in the different mines.  There is some information available  to us
on the radon daughters in different coal mines.  This level of radio-
activity seems to be insignificant.  I am not aware of the presence of
any uranium or radium ores in the coal mining areas.

Dr. Blot (NCI):  Other than the proportional histology analyses that you
have presented today, could you say again how you might be able to  use
this data to get at the question of whether or not coal miners have an
increased lung cancer risk?

Dr. Vallyathan (NIOSH):  We hope to answer that question by a case
control epidemiological study.  However, I have mentioned some of the
difficulties that we have encountered.  If all the variables are
adequately controlled, the question of whether the coal miners have an
increased risk of lung cancer can be ascertained.  Some of the earlier
studies have attempted to differentiate the effect of cigarette smoking
and a limited number of cases have shown a low incidence of lung cancer
in coal miners.  In this respect it is important to note that the
latency period for the inorganic type of minerals to induce lung cancer
is probably in the range of 20 to 30 years.  Induction of cancer by coal
dust may not be evident if other causes such as smoking or shortened
lifespan from pneumoconiosis occurs before the expression of the cancer.

Dr. Blot (NCI):  How are you going to use the data that you have avail-
able?  I think I am getting at the question that Bob Spirtas was bringing
up.  How are you going to use that data to answer this question?

Dr. Vallyathan (NIOSH):  It can be answered only by a case control
study.

Dr. Blot (NCI):  A case control study among all coal miners and their
autopsies and then looking for differences in lengths of employment?

Dr. Vallyathan (NIOSH):  Yes, that is what we plan to do, and we are
also in the process of getting a case control series of lung cancer
cases from non-mining populations.

-------
Dr.  Fraumeni (NCI):   I would just like to ask one question for the
future.   I would be  very cautious in saying, as you have a number of
times, that you have evaluated the incidence of lung cancer.   The data
that you have,  is some sort of a proportionate frequency series; but
does NIOSH have any  plans to conduct a cohort analysis of coal miners to
determine once  and for all  whether or not the risk of lung cancer is
increased or decreased or the same as the general population?  Because I
think without a cohort study and lifetable analysis, you will never be
able to evaluate the issue.  Can anybody answer that?

Dr.  Vallyathan  (NIOSH):  As far as I know, there are no cohort studies
which have been initiated.   Regarding your first comment, I definitely
agree that incidence is the wrong term to be used.  I probably should
have said "prevalence".

Dr.  Bridbord (NIOSH):   First of all, there was a recent cohort mortality
study sponsored by NIOSH, which did not clarify the lung cancer issue
but did suggest an increase in stomach cancer.  The second point to note
is that the recent Mine Safety and Health Amendments Act does give NIOSH
considerable new responsibilities in the area of mining in general,
which is not exclusively coal mining but really reaches out to the total
spectrum of mining.   I think we would have to weigh the answer to that
question in terms of the opportunities that would be available to do
that additional study and look at our total mining responsibilities.
But I do not think that the Institute is completely satisfied that we
have resolved the question.  So, at some point in the future, I think
there might be  a chance to  embark upon a larger study, but that would be
a competing issue in terms  of the total mining research needs.

Dr.  Fraumeni (NCI):   Is there any evidence with regard to the relation-
ship between the cell  types and the amount of pulmonary fibrosis?  For
example, with the cases of  adenocarcinoma of the lung, did they have
more pulmonary  fibrosis or  any evidence of what has been called Caplan's
syndrome, which is a hypersensitivity reaction with pulmonary fibrosis
and rheumatoid  arthritis?

Dr.  Vallyathan  (NIOSH):  This was illustrated by the last slide.  May I
have the last slide  again,  please?  The extent and severity of pneumon-
coniosis, which has  been graded based on the type and presence of the
different types of lesions, that is; simple macule, nodule, PMF, which
is progressive  massive fibrosis and silicosis, are scored for each type
of cancer.  We  have  not found any definite association with adeno-
carcinoma and PMF.  However, there is a lower prevalence of pneumo-
coniosis with small  cell carcinoma.   In small cell carcinoma  only 64% of
the  cases had pneumoconiosis.  Caplan's syndrome was not observed in
these coal workers.
                                   29f.

-------
FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:
   PROGRESS ON 3OINT ENVIRONMENTAL AND
       OCCUPATIONAL CANCER STUDIES
            Wednesday Morning, May 7
 METHODOLOGY/EXPERIMENTAL MODELS SESSION
            SESSION CHAIRPERSON
                Dr. Carl Morris
         Environmental Protection Agency
                         297

-------
            EVALUATION OF THE TRANSFORMATION ASSAY USING C3H IQTh CELLS
             FOR USE IN SCREENING CHEMICALS FOR CARCINOGENIC POTENTIAL
                             Thomas P.  Cameron, D.V.M.
                             National  Cancer Institute

In the past few years, considerable effort has been directed to defining batteries
or combinations of in vitro test methods which could provide rapid, sensitive, and
reliable means for assessing the carcinogenic potential  of chemical compounds.
These various in vitro test systems can be broadly divided into three major
categories, namely (1) those which detect mutagenic or  chromosomal changes in
micro-organisms or mammalian cells; (2) those in which  there is induction of
morphological transformation in mammalian cells in culture; and (3) those in which
interactions between the chemical and  target macromolecules such as DNA can be
assessed.  Experimental evidence has been accumulated which shows that there is a
positive correlation between the in vivo carcinogenicity of many chemicals and
their capacity to elicit a response in  in vitro systems.

Approximately one hundred different model systems have  been identified as having
been used, in varying degrees, for assessing the carcinogenic and/or mutagenic
potential of chemicals.  In some instances, limited numbers of chemicals have
been tested and in many cases they have been restricted to well characterized
direct-acting agents.  Although such a  review and such  compilations of test data
are extremely valuable, certain problems are encountered with data derived solely
from the literature.  For any single test method, there can be small but critical
differences in the methods used which  can result in apparent non-reproducibility
of results.  Another problem relates to the fact that the tests, in general, have
been performed with the full knowledge  of the in vivo carcinogenic activity of
the chemicals.  Such a situation can introduce a certain amount of bias.

-------
In the utilization of any in vitro  method  as  a  valid indicator  of  the  carcinogenic




potential of chemicals, there  is a basic requirement to know that (I) the metho-




dology is well-defined and the critical elements  of the procedure are recognized; (2)




reproducible results can be obtained not  only within a specific laboratory but can be




obtained equally well  among different  laboratories; (3)  there  is experience with a




broad spectrum  of chemicals of diverse  structure and biological activity in order to




recognize that  certain types  of  chemicals may give unique  negative or positive




results  in some  tests; and (4) there are methods and approaches  for  analysis and




interpretation of the experimental results.  The  existence  of  well-defined and




evaluated assay  methods, which  can be  exploited  for  the  assay  of  carcinogenic




potential, would then provide the means to  examine a large number  of chemicals and




aid in setting priorities for long-term animal carcinogenicity bioassays.






The purpose  of this  project  is   to  evaluate and  determine  the usefulness and




reliability of an in vitro  transformation  assay  using C3H  IOT Ife cells  as a candidate




for one of  a battery of short-term assays  for  the initial  determination of the




carcinogenic potential of chemicals.  In addition the reproducibility of the system




will be  assessed since the studies are being conducted in  two  laboratories simultane-




ously.






The workscopes of the two  contracts for this effort were sharply defined so  as to




initially emphasize  the  methodological aspect.   The  specific objectives  to  be




approached in parallel  were as follows:








      I)    Propagate and store a  large quantity of mycoplasma-free, low passage




      IOT '/2 cells.

-------
2)   Characterize the behavior of the cell population with respect to plating




efficiency, generation time, saturation density, karyotype stability, cell mor-




phology,  cell size distribution, absence of growth in soft agar, and absence of




tumor formation in C3H mice.








3)   Identify and  obtain  large lots of fetal bovine serum that  yield optimum




growth characteristics and transformation response with direct acting carcino-




gens and polycyclic aromatic hydrocarbons.








4)   Establish the IOT fe cell  transformation assay and its concomitant plating




efficiency  in the absence of an exogenous metabolic activation system, using




model chemical compounds.








5)   Evaluate  the transformation  assay  using selected known  chemical car-




cinogens and non-carcinogens by scoring  for Type III  foci and by determining




the ability of cells from Type III foci to grow in soft agar and produce tumors




when injected into irradiated C3H mice.








6)   Develop and  characterize a mamalian  S-9 activating system for incor-




poration  into the C3H  IOT fz transformation assay.








7)   Determine whether  the  sensitivity  and reproducibility of the assay can




be further  improved by examination of certain baseline factors such as plating




density and assay  interval,  passage number of target cells, and step-down of



serum concentration.
                                       800

-------
      8)    Test  a  series  of chemicals  (supplied under  code) consisting  of both




      carcinogens and non-carcinogens  for  their  transforming  capacity in  IOT  '/z




      cells.  The chemicals will be tested with and without an exogenous metabolic




      activation system.








The progress  made  in  these areas by the two contractors has been satisfactory.




Early passage (P-5)  cells received from  Dr.  Heidelberger  were subcultured by the




prescribed protcol (once every  10 days  at 5  X 10 cells/dish) and large aliquots of




cells  from P6, 7 and 8 frozen in liquid nitrogen to form a  uniform stock for future




studies.  The cells were determined to have the culture characteristics for C3H IOT




'/z cells previously reported  in the literature.  In addition, lots of fetal bovine  serum




have  been identified that provide the expected saturation  densitiy and  chemically-




induced morphological transformation.








These cells  have been characterized and it has been found that they have a plating




efficiency (100 cells/60  mm  dish)  of  23%,  a saturation density  of  5.9  X  I05




cells/60 mm dish and a generation time of approximately  16 hours.  The cells were




also examined by transmission electron microscopy and it was observed that there is




overlapping  and underlapping of cytoplasm between  cells.  This suggests  that cell




contact may not be a controlling variable in  the density dependent inhibition of cell




division.








Other significant findings have been reported by these two contractors.







Transformation assays with eight known chemical compounds showed that benzo(a)-




pyrene (B(a)P),  3-methylcholanthrene (3-MC), dibenz(a,h)anthracene and  7,12-di-




methylbenz(a)anthracene gave a positive response by producing  Type III  foci  in the
                                              301

-------
absence of an exogenous metabolic activation system.  Chemicals not active  in the




assay were the non-carcinogens,  anthracene and  phenanthrene,  and two  known




carcinogens,   2-acetylaminofluorene  (2-AAF) and  N-methyl-N-nitro-N-nitroso-




guanidine (MNNG).  In the assays positive for transformation, the absolute  transfor-




mation frequency and the dose at which a positive response was induced varied from




experiment to  experiment.   In general, 3-MC induced  a  more  reproducible  trans-




formation  response and dose-dependent effect than B(a)P, although variation from




experiment to experiment was still evident.








An acceptable lot of  rat liver  S-9 has been  prepared  from Aroclor-1254 induced




Fischer 344 male rats and  is being used to determine the critical parameters  for




chemically transforming IOT '/2 cells in  the presence of such an exogenously supplied




source of  mammalian metabolizing activity.  Cytotoxicity and transformation assays




have  been conducted  using B(a)P, 2-AAF, diethylnitrosamine  (DEN), and 3-MC.




When the  suspension  assay  was  used,  only  those  cells  exposed to  B(a)P  in  the




presence of a metabolically active S-9 preparation  exhibited morphological trans-




formation. The negative results obtained thus far with the other compounds tested




may be attributable  to  various  factors such  as  (I)  the  relatively short  exposure




period (2-4 hours) used; (2) the limitations associated with  a suspension assay; (3)




failure of the S-9 to activate the  chemicals to forms capable of  transforming  IOT h




cells; (4)  metabolic  inactivation  and  detoxification of  the chemicals by  the  S-9




preparation; or (5) the failure of  transformed  cells  or cells  in the process  of  being




transformed to exhibit the transformed  phenotype.








An amplificaton  assay (Level II  transformation  assay) is  also  being assessed to




determine  whether non-expressed transformed cells are present  in  the  standard
                                              302

-------
transformation assay.  This  assay  involves replating of the treated cell population


when they just reach confluency and scoring the replated cells for Type III  foci in


2-4 weeks.




MNNG  has been tested using  this  amplification modification.  At  a dose of 0.5


ug/MNNG/ml, a single  Type II  focus was detected in the standard transformation


assay.  The Level  II  transformation  assay gave rise to numerous Type III foci. The


positive control, 3-MC  (2.5  ug/ml) induced the formation of 17 Type III foci  in the


standard transformation assay and no augmentation  in transformation was obtained


in the Level II assay.




Studies have also  been  conducted to test various factors which may affect and/or


optimize the transformation assay.  It was observed  with 3-MC treatment that the


total number  of foci (II and III) increased  with increase in exposure time  to this


chemical,  but there was no enhancement  of  transformation  frequency when  the


treated cultures were maintained in medium containing 2%  and 5% serum rather


than the standard 10% serum concentration.




The  constitutive activity of arylhydrocarbon hydroxylase of the C3H IOT '/2 cells was

                                               o
also  determined by measuring the conversion of  H-B(a)P and it was compared  to


the activity in BALB/C-3T3 clone 1-13 cells.  The time course of B(a)P metabolism


(0.3  ug/ml) with 2X10   cells is linear with  incubation time for both 3T3 and IOT fc


cells. However, the  IOT fe cells possess a 15-fold higher activity than the 3T3 cells.




For  the next  contract  period the overall objective  will be to further  develop and


validate a  reliable, reproducible assay system  for neoplastic transformation using
                                           303

-------
C3H IOT !fe cells, specifically identifying those factors which influence the assay and




then determining those procedures  which  would  lead  to their standardization.




Emphasis  will be placed  on developing the  exogenous enzyme  activation system




because of absent or insufficient enzyme levels in the target cells.








Simultaneously,  because  of the  progress  already made  by  two laboratories  in




standardizing, we are moving ahead on  phase II,  the  initiation of assays of coded




samples supplied by the NCI.  Cytotoxicity effects have already  been run on more




than 10 of the  coded  samples and we expect that this  effort will  move forward




rapidly in  the final contract year FY 81.
                                           304

-------
                                 Discussion
Dr. Page, EPA:  On these assays, does there appear to be any good correlation
between  the  cytotoxicity and transformation capabiiity?  Is there any positive
correlation you could see?

Dr. Cameron, NCI: Tranformation and the cytotoxicity?

Dr. Page, EPA: Cytotoxicity for the cell cultures?

Dr. Cameron, NCI:  I  don't have an answer for that.  I would mention that the
samples came out of the bioassay program.

Dr. Kelsey, NCI:  Was I correct in hearing  that they did not pick up AAF and
MNNG in the standard assay?

Dr. Cameron, NCI: Right.

Dr. Kelsey, NCI:   Would that not bother  you in terms  of  giving them coded
samples?

Dr. Cameron, NCI: It bothers the contractors.

Dr. Kelsey, NCI:  I mean until they can get some knowns to work, or at least some
basic knowns, wouldn't it be advisable?

Dr. Cameron, NCI: I think it would bother anybody using the system.

Dr. Kelsey, NCI:  Are there plans to use maybe hepatocytes or something like that
as an activating system?

Dr. Cameron, NCI: Not in the protocol.

Dr. Hegyeli, NCI:  Dr. Alcian made a study to compare in vitro and in vivo toxicity,
and he inferred that the distribution of a certain substance between optimal and
the aqueous  phase has a very  important role in determining the  ratio that was
detected from in vitro to in vivo. My quesiton is whether there was any study done
as far as the solubility  and what kind of solvent was used for the different kinds of
chemicals?

Dr.  Cameron,  NCI:   The solubility  factors  were  determined  by  a different
laboratory, a chemical  analysis  laboratory, and  they  were  supplied  by that
laboratory.  So that  is worked  out for the  contractor.  He receives the chemical
and the solubility data simultaneously.  I hope that answered the question.
                                    305

-------
                  NCI/EPA COLLABORATIVE PROGRAM PRO3ECT
                                 FISCAL YEAR 1980
I.     TITLE
      Human Epithelial Cell Metabolic Activation Systems  for  Use with  Human Cell
      Mutagenesis                 (R80556310-02)*

                     J. Justin McCormick and Veronica M. Maher
                              Carcinogenesis Laboratory
                              Michigan State University

II.    ABSTRACT

             The need  to  develop  ever  more adequate test systems which  can reliably
      determine the possible toxic, mutagenic, teratogenic and/or carcinogenic effects of
      exposure  of  the  human  population to chemicals in the environment  is becoming
      more and more evident.   It is, of  course, not possible to carry out such testing in
      humans, and  therefore, information on risk estimates has to be extrapolated to man
      from results obtained in test systems which utilize microorganisms, mammalian
      cells in  culture, or  animals.   Mammalial  cells in culture have  increased in
      importance in biological  testing because they retain many of the characteristics of
      the  target cells  at  risk and, yet, can be grown into very large populations  and
      analyzed  and characterized by a wide  range of biochemical and genetic techniques
      that are not  possible with whole animals.  We have developed a quantitative system
      for measuring the cytotoxic and  mutagenic effect of radiation and direct-acting
      chemicals in diploid human skin fibroblasts  in culture and  have shown that these
      effects are directly related to the capacity of  human  cells  to repair damage to
      DNA.   Since the human population  is far  more likely  to be exposed to parent
      compounds than to direct-acting model compounds, we are extending the capability
      of  the diploid   human  fibroblast cell   mutagenesis system  to  include  parent
      compounds,  or  mixtures of  chemicals,  which  require  metabolic activation  by
      coupling it with  cell-mediated activation.  Cell  lines derived from human tumors
      are being utilized as feeder layers to provide metabolic  activation of carcinogenic
      agents.   To find cells  capable of  metabolizing the various test chemicals, we
      prescreen our series of tumor  cell lines for evidence  that  the  chemical causes
      interference with DNA  synthesis. (The  compound under investigation is admin-
      istered to the candidate metabolizing cells  over a  wide range of concentrations.
      After  48  hrs. of incubation, the amount of semiconservative DNA  synthesis is
      measured by incorporation of radioactively-labeled thymidine and compared with
      that of untreated control cells.)  When a cell line appears to activate a particular
      chemical  to  a form which interferes  with DNA  synthesis, it is then examined for
      ability to act as  a metabolizing layer  for target  cells.  To enhance the sensitivity,
      the target cells in this co-cultivation  system are usually excision repair defective
      XP cells  which are  examined for  evidence of cell killing (loss of colony formation).
      We have found that normal fibroblasts may be used as target cells once conditions
      are optimized.  Using this prescreen, we have identified cells capable of activating
      a  wide range of  agents, including  polycyclic aromatic hydrocarbons, aflatoxins,
      aromatic  amides  and  amines, nitrosamines,  nitrogen-containing  polycyclics, etc.
      We are now  investigating the usefulness of these cells lines:  (a) to quantitate the
      cytotoxic effect  of environmental  agents by measuring cells' loss  of ability to
      clone, (b) to quantitate  the  induction of mutations using several  markers, and (c)
      using radioactive parent compounds to determine the number and kinds of adducts
      formed on DNA by covalent attachment of chemical  residues.   The activation
      systems developed are not limited in  application to the  human fibroblast mutation
      assay and can be  expected to have more general applicability.

*Project officer, Dr. Michael D. Waters,  Environmental  Protection Agency.

-------
HI.    INTRODUCTION

             It  is  now  realized  that most chemical carcinogens  require  metabolic
      activation before they can exert their carcinogenic effect.  James and Elizabeth
      Miller were the first workers  who clearly realized the importance of  metabolic
      activation and its  near universal  application to chemical carcinogens in their
      studies of the 1960's (1).  This led to a realization that assays of the mutagenic or
      carcinogenic potential of compounds using bacteria, fungi, or mammalian or human
      cells  in culture required a source of metabolic activation of the carcinogens if the
      cells  being used were not  able to  metabolize the compounds themselves  as was
      most  frequently the case. This led  Mailing to develop the use of liver homogenates
      to provide the metabolic activation  for  such in vitro systems (2).  These liver
      homogenates  have  been  widely used  and  adapted as  in the Ames  test  with
      Salmonella and may consist of microsomal preparations or of  S-9  preparations of
      liver  homogenates  (3).  Once the need for  metabolic  activation was realized,  a
      parallel development  also took place  in mammalian cell culture where researchers
      made use of feeder layers of metabolizing cells such as primary liver hepatocytes
      which could metabolize many of the carcinogens  that one wished to examine (4-6).
      In practice, the target cells which one wishes to mutagenize or  transform into
      cancer cells are placed in contact with  the feeder layer cells and carcinogen is
      added for 24  or 48 hours.    The  cells  are then  trypsinized and replated  for
      determination  of the mutagenic or  transformation  response.   The studies  with
      feeder layers  have  generally been qualitative. That is, the authors have shown  a
      plus or minus response but have not  usually shown a dose response or other more
      elaborate kinetic analysis of the interaction of the carcinogen with the cells.

             The value of  using activation systems derived  from liver or  other cell
      homogenates has been questioned since high concentrations of benzo(a)pyrene (BP),
      e.g.,  1,320 nmol per mg of microsomal protein, produced extensive DNA adducts,
      but only a small percentage of these represented the principal cellular DNA adduct
      formed from  BP,  viz., the N^-guanine adduct formed by  the anti 7,8-diol 9,10-
      epoxide of BP (7).  In contrast, low concentrations of BP, e.g., 15 nmol per mg of
      protein, produced the diol epoxide DNA  adduct of BP exclusively (8).  A recent
      report by Santella,  et. al. (9) indicates  that  with appropriate  induction  of
      microsomal enzymes, an S-9 fraction can also be shown to catalyze the binding of
      BP to DNA exclusively through  the  diol epoxide pathway  when a low substrate
      concentration (12.5  nmol per mg of  protein) is used.  Jhe maximum extent of DNA
      modification in these latter studies  was  8.1 per 10  nucleosides.  Using  diploid
      human fibroblasts which are totally lacking in excision repair of BP adducts from
      DNA   we  have  determined that   250-fold  greater  levels of DNA binding  are
      necessary to  see significant  induction of  mutations to thioguanine resistance.
      Even  higher initial  levels of bound adducts  are  required to observe mutations in
      normally-excising human fibroblasts.  The only practical  way to  obtain greater
      binding with S-9 fractions is to use  higher  concentrations of BP since  the  S-9
      fraction itself is toxic.  Unfortunately, as noted above, this higher concentration
      can be expected to  generate the wrong adducts.  Similar problems have been noted
      with dimethylbenz(a)anthracene (DMBA) metabolism by Bigger, et. al. (10) when the
      rat liver  microsomes were  used for  carcinogen  metabolism.  At  a high  concen-
      tration of DMBA, K region DNA adducts were formed while at low concentrations,
      bay region adducts  were formed.  Most signifcantly, they found that when intact
      mouse cells were used for metabolism, no such qualitative  differences were found
      in the DNA adducts formed over a 40 fold  dose range. Since the K region as well as
      the bay region adducts of BP and DMBA are mutagenic (11), it seems  likely that
                                      807

-------
      many of the reports of mutagenicity of these compounds in short term tests which
      make use of S-9  fractions or microsomal protein for metabolic activation are the
      result of formation of K region adducts.  Thus, it appears that the assays used have
      given the "correct"  answers but for the  wrong reasons.   It should, therefore, be
      clear that at least for metabolism of polycyclic aromatic hydrocarbons, intact cells
      are  to  be preferred to microsomal or S-9 preparations.   Detailed comparative
      studies  of metabolism of other compounds have  not yet been carried out, so the
      problem may well extend to other classes  of compounds.

             We have extensively studied the effect of active derivatives of carcinogenic
      agents  on human fibroblasts  derived from normal  individuals as  well as repair
      deficient fibroblasts derived from xeroderma pigmentosum patients (11-15). Because
      such fibroblasts  have either  no or  extremely low levels of metabolic activating
      activity for carcinogenic agents, it is necessary to utilize direct acting carcinogens
      with such cells to see their effect.  However, to make  broader use of this assay, as
      well as  to explore the effect of agents that require  metabolic activation or of
      unknown chemical agents (such as in complex mixtures), we have been studying the
      ability  of feeder layers derived from  various  cell types to  provide metabolic
      activation for this cell system.  Because we were working with human fibroblasts as
      target  cells,   it seemed  ideal  to  select human cells for metabolic  activation.
      Primary epithelial cells derived from various  tissue of man would seem  likely to
      provide the ideal system for metabolic activation. However, there are a number of
      problems connected with  obtaining and standardizing any  such  epithelial cell
      system.  First, it is difficult to obtain pure populations  of epithelial cells in culture
      since fibroblasts tend to overgrow the  epithelial cell cultures.   Second,  it is
      extremely likely  that epithelial cells derived from various human donors (and even
      from the same donor at different times) will show different levels of metabolizing
      ability  for  a particular  carcinogen.  This will  make it  extremely  difficult  to
      standardize any assay based on the use of primary epithelial cells. Third, epithelial
      cells from organs such as  liver which might be extremely  useful for metabolizing
      carcinogens  are  not  readily available for use, show extreme variability between
      various donors, and are not easily adapted to cell culture. In addition, working with
      human liver material poses the health threat of hepatitis infection.  To get  around
      all the  limitations  of metabolic activation systems  for cells in culture discussed
      above, we have chosen to use tumor cell  lines derived  from various human tissues.
      These  human cell lines  maintain the ability to activate various carcinogens and,
      have an infinite  lifespan  in  culture.  These properties suggested they would  be
      extremely useful  as feeder layers.


IV.    OVERALL OBJECTIVE

             The objectives of the research are to develop procedures that will allow for
      more adequate in vitro testing of environmental chemicals  and to shed light on the
      mechanisms involved in mutagenic processes and related events. The use of diploid
      human  cells in culture for  environmental  research  is very useful first  of all,
      because of the relevance to man of  results obtained with such cells and secondly,
      because the existence of DNA repair deficient human cells derived from xeroderma
      pigmentosum (XP) patients which have been characterized and shown to be unable
      to remove many  different types of carcinogen residues covalently bound to  DNA
      (11,12) allows  one to  determine the potency of various  chemicals without the
      interference of excision processes operating during the  period when the target cells
      are  being incubated  with  the chemical.   Furthermore, comparing the effects of
      various agents in excision repair-proficient and deficient human target cells allows
      one  to  study the biological  effects of  DNA excision  repair  on  the mutagenic
      process.

-------
V.    MA3OR FINDINGS AND PROGRESS

      A.     Development  of  a Short Term Assay for Identifying Human Epithelial Cell
             Lines Capable of Metabolizing Parent Carcinogens into Reactive Forms:

                   In our  initial studies  of human cell-mediated metabolism, we made
             use of benzo(a)pyrene BP, as our model  compound.  One of the reasons for
             choosing BP  is that there is a  simple  rapid  technique for  measuring its
             metabolism, viz., conversion of tritiated BP into the water soluble product.
             We used this  assay  to screen  17 cell lines and found  several which could
             metabolize BP.  When these were identified we assayed them for ability to
             serve as a metabolizing layer capable of activating not only BP but a whole
             series of polycyclic hydrocarbons into forms which are cytotoxic to repair
             deficient diploid  human skin fibroblasts.

                   However,  not all carcinogens which require metabolic  activation for
             activity can  easily  be obtained in  tritiated form nor is metabolism  of a
             compound to a water-soluble form necessarily an indication of activation to
             a  form which can bind to DNA.  Therefore, we modified a DNA synthesis
             inhibition  assay  developed by  Painter  (17) which  uses the inhibition of
             incorporation  of  H-thymidine into DNA  as an  indication of metabolic
             activation of carcinogen and resulting DNA damage.
                                       h
                   In this  method, 5 x 10 actively growing cells to be assayed for  ability
             to metabolize a  carcinogen  are plated per 16mm well  (Costar 3524  Tissue
             Culture Cluster)  in 1 ml medium. Approximately 24 hr later, these cells are
             treated with carcinogen by adding 1 ml of carcinogen solution to the 1 ml of
             medium already  present in the well.  After 48 hr incubation,  the treatment
             is removed and replaced with 1 ml  of medium  containing  H-Tdr (5uCi/ml).
             After a two hour labeling period, the medium  is removed and the cells are
             rinsed with PBS.  The cells are removed in 1 ml of trypsin-EDTA, diluted to 4
             ml with phosphate buffered saline (PBS),  rinsed with PBS two times on a
             glass membrane filter, and then rinsed two times with 10% chilled TCA. The
             filters are then removed,  treated in 1 ml 0.5N HC1 for 60 minutes at 90  and
             counted in a liquid scintillation counter.   The  data  are calculated and
             presented as  percentage  of the amount  incorporated  by  the untreated
             control cells.

      B.     Metabolizing Strains Identified:

             1.      Interference with DNA Synthesis:

                          Using  the DNA  synthesis inhibition assay described  above,
                   many human carcinoma cell lines were examined for their ability to
                   metabolically activate several classes of carcinogens to forms which
                   would  interact with cellular DNA  and thus inhibit DNA synthesis.
                   Two cell  lines  which  we  had  already  found  to be  capable of
                   metabolizing  benzo(a)pyrene were examined in this assay as positive
                   controls.  Figure 1  shows  the  results of this experiment.  Both cell
                   lines show inhibition of DNA  synthesis,  703  to a greater extent than
                   835. This inhibition data agrees with the level of metabolism of BP
                   to water-soluble compounds found in these  cells,  i.e.  703 is greater
                   than 835.  Figure 2 shows the results of exposing PC-3, 703, XP and
                   normal  human  fibroblasts (NF)  to dimethylnitrosamine (DMN)  and
                   diethylnitrosamine (DEN) and  measuring the inhibition of thymidine
                   incorporation.

-------
               BP  CONCENTRATION ( H.M)

                    10       20       30
IUU<

_J
zo
o rr
— t-
£z 40
§8
{J ^
&^
Ko
O i nni
OH i^^1
ZJZ
— UJ
u.0
-r-i CY" \ f~\
PS I0
" Q.
«S i
01
. 1
c ' ' ' A-
t 835 " -
v^ • BP I
i i i

1 ' ' B
1 703
T BP
i i i
0 10 20 30
Figure 1. Inhibition of tritiated thymidine incorporation in tumor lines 835 and 703 after a
W hour pre-treatment with BP at various concentrations.


                             an

-------
     o
     a:
     o
     o
     o
     UJ
     o
     cr
     UJ
     a.
     ro
     O


     O
     (T
     O
     a.
     a:
     o
     o
        100
100
          10
                       50
                          100
 150  225
DMN
            0    10   20  30   40   50   60

               CARCINOGEN CONCENTRATION  (mM)
Figure 2.  Inhibition of tritiated thymidine incorporation in tumor lines o 562,* PC-3, and

o 703, and in • XP and • normal human fibroblasts after a 48 hour treatment with various

doses of DEN or DMN.
                              311

-------
2.      Direct Cell Killing:

             To  further  examine  the  biological  consequences  of  the
       metabolic  activation  of  the  carcinogens  discussed  above,   we
       examined  the  cytotoxicity  of  the  compounds  directly  on  the
       metabolizing cells themselves by treating the cells with carcinogens
       at the indicated concentrations for 48 hr at 37  in Eagles Minimum
       Essential Medium containing  10% fetal bovine serum  and antibiotics.
       At the  end of  the exposure  period,  the  cells were trypsinized,
       replated at cloning density, and  cultured for approximately two weeks
       until  macroscopic clones  developed.  The results of exposing 703, 835
       and 549  cells to BP  are shown in Figure 3.  It is obvious that 703 cells
       shows the greatest  toxicity per dose,  these results were the results
       obtained above with 703 and 835 cells.  The results  of exposing 703
       cells  (polycyclic hydrocarbon metabolizing cells) to aflatoxin  B. are
       shown in Figure 4.  The aflatoxin B. appears to be very cytotoxlc to
       these cells.

3.      Cell-mediated  Cytotoxicity in Xeroderma Pigmentosum Cells:

             Since it is always possible that direct cytotoxicity assays are
       not as sensitive as they could be due to DNA repair in the carcinoma
       cell lines, we  used  the  cell-mediated assay previously developed in
       this laboratory in which  lethally irradiated metabolizing cells are co-
       cultivated with  target  xeroderma pigmentosum  cells  (XP)  in the
       presence of the carcinogen requiring activation for a period of 48 hr
       at 37°.  At the end of the exposure period the cells  are trypsinized,
       replated, and the XP cells allowed to clone.  The percent survival of
       the cloning ability of XP cells was determiend by dividing the cloning
       efficiency of co-cultivated XP cells in the presence of carcinogen by
       the cloning efficiency of co-cultivated  XP  cells in  the absence of
       carcinogen, multiplied by 100.

             Optimal   conditions   for  co-cultivation   were   previously
       established using BP  as  the  model carcinogen.   The  effect  of
       increasing  the number  of  metabolizing  cells  and  changing  the
       concentration of BP was investigated. The cytotoxicity of BP in the
       XP target cells was  found to increase with increasing number of
       metabolizing cells and BP concentration.  In addition, the number of
       induced  mutations to thioguanine resistance in the  XP target cells
       also  increased with increasing  BP concentration  in the presence of
       metabolizing  cells.    Using  the  cell-mediated cytotoxicity  assay
       system developed for BP  other carcinogens and metabolizing cells
       were examined. (Figure 5-8)

-------
                      100
                              CONCENTRATION B(a)P (JJ.M)
                              0.2    0.4    0.6    0.8
                        I -
Figure 3. Direct cytotoxic action of BP on 549, 835 and 703 tumor cells.  The cells were
incubated for 48 hours with BP at various concentrations and then assayed for cloning
ability.
                      0      2      4      6      8     IP
                       AFB, CONCENTRATION (jxM)
Figure 4. Direct cytotoxic action of AFB. on 703 tumor cells.  The cells were incubated
for 48 hours with AFB. at various concentrations and then assayed for cloning ability.
                                      313

-------
                                    CONCENTRATION
                                02   0.4    0.6   0.8
                         100
                          10 -
                                                  B(o)P
                                                  7,8-diolofB(a)P
                                                A BA
                                                • DMBA
                                                  7-MeBA


                                                       C—
                             CELL-MEDIATED CYTOTOXICITY IN XP
Figure 5.  Cell-mediated cytotoxicity of XP cells coincubated for 48 hours with lethally-
irradiated  835 tumor cells  in the presence of  various  concentrations  of polycyclic
aromatic hydrocarbons or  their derivatives.  After the  48 hour coincubation,  the cells
were trypsinized and the XP cells assayed for cloning ability.
                                              314

-------
                AFB, CONCENTRATION
            100
              80
              60

              40
         CD
       I  _
         CD   20
835
                                                 40
Figure 6. Cell-mediated cytotoxicity of XP cells coincubated for 48 hours with lethally-
irradiated 835 or 703 tumor cells in the presence of various doses of aflatoxin B..  After
the 48 hour coincubation, the cells were trypsinized and XP cells assayed for cloning
ability.
                               SIS

-------
        I00«
         50
    _J
    CD
 CL
 X
          10
T—q
    z      0      4       8      12      16
    g         DEN CONCENTRATION (mM)
    o
> IUU1
1
w 50
1—
PERCEN'
» 0
> 	 1 	 i I i i _
F 700 :
t
[V,
1 1 1 1 1
           4,4' METHYLENE BIS(2-CHLOROANILINE)CONC.(^xM)

Figure 7.  Cell-mediated cytotoxicity  of XP cells coincubated for 48 hours with
lethally-irradiated  PC-3 or  700  tumor  cells  in  the  presence  of  various
concentrations of DEN or 4.4'  methylene bis (2-chloroaniline).  After 48 hours of
coincubation, the cells were trypsinized and the XP cells assayed for cloning
ability.
                                 SIS

-------
              CARCINOGEN  CONCENTRATION
         100-
           50
       >-
       h-
       QQ
       <   10
       e>
    0.^100
                                              549  :
                    0.2    0.4    0.6    0.8    1.0
             '0     0.2   0.4   0.6    0.8     1.0

              CARCINOGEN CONCENTRATION
Figure 8.  Cell-mediated cytotoxicity of XP cells coincubated for 48 hours with lethally-
irradiated 549 or 703 tumor cells in the presence of various concentrations of polycyclic
aromatic hydrocarbons or their derivatives. After 48 hours coincubation, the cells were
trypsinized and the XP cells assayed for survival.

                                 sr;

-------
4.     Cell-mediated Mutagenicity in Xeroderma Pigmentosum Target Cells:

             We  have tested the system for use as source  of  metabolic
      activation  of  two  parent polycyclic aromatic  hydrocarbons  viz.,
      benzo(a)pyrene  and dimethylbenz(a)anthracene.   To assay  the co-
      cultivated  target cells for the frequency of BP-induced mutations to
      6-thioguanine resistance, cells were trypsinized and counted as above
      and plated into 250 ml flasks to allow for expression of mutations at
      cell densities which  allowed  surviving  target  cells  to replicate
      without  reaching  confluence.  (Sufficient  numbers of  flasks  werg
      employed  for each determination  to insure a minimum  of 10
      surviving target cells at the beginning fo the expression period.) The
      cells  were trypsinized, pooled, and replated into flasks one or more
      times  to  insure  logarithmic  growth.    The  total  length  of  the
      expression period  was adjusted for each experimental  determination
      to  allow the cells to undergo  at least 4.5 population doublings before
      selection was begun, i.e., 7 to 10 days (15).

             At the end of,the expression period, the cells were trypsinized,
      pooled, and 1-2 x 10  cells plated into selective medium at a density
      of  500 cells/sq cm ( 180 x 60mm dishes per  point).  To  determine
      their cloning efficiency  at the time of selection, these cells  were
      further diluted and plated at  cloning densities into medium identical
      to  selective medium, but lacking 6-thioguanine.  A reconstruction
      experiment with HPRT~ Lesch Nyhan cells (18) accompanying  each
      determination indicated  that  the  efficiency  of recovery of  6-
      thioguanine resistant colonies in the  experiments reported here was
      85%.  Selection was continued for 14 to 18 day one refceding.  The
      frequency  of mutations  to 6-thioguanine resistance was calculated
      from the probability of  a mutant per dish using the P(0) method (13).
      The results are shown in Figure 9.

             It will be seen that DMBA is more mutagenic than BP when
      these are compared as a function of the concentration administered.
      This  is  also  true   as  a function  of  the cytotoxic effect  of the
      hydrocarbons.   (Compare these data with  those  of Fig.  8  for the
      percent survival of cells exposed to DMBA and in Fig. 11 for that of
      cells exposed to BP for 48 hrs.)
                                    318

-------
                    o
                    2
                    UJ
                    Z>
                    O
                    UJ
                     c
                    Q
                    UJ
                    O

                    Q
                       
-------
                  High Pressure Liquid Chromatography Analysis of B(a)P-DNA Adducts
                  Produced in Co-cultivated Human Cells:

                        Since previous data had shown that BP induces an increase in
                  the frequency of mutations to thioguanine resistance in target human
                  diploid fibroblasts co-cultivated with human tumor cells capable of
                  metabolizing BP, this suggested an interaction of BP metabolites with
                  cellular  DNA.    To investigate  the  extent  and  nature  of  this
                  interaction, we determined the  number of BP-DNA adducts formed
                  during a 48 hr exposure of co-cultivated target cells and metabolizing
                  cells to tritiated  BP per 10  moles of DNA nucleotides were found to
                  be, respectively,  1.4,   4.1,  and  12.   These results  confirm  that
                  metabolites of  BP  were binding  to  the cellular DNA  and  showed a
                  direct relationship  between the concentration of BP in the medium
                  and the number of BP-DNA adducts formed.

                        To identify these  BP-DNA adducts, we analyzed those formed
                  at   the    highest   concentration   using    high   pressure  liquid
                  Chromatography (HPLC).  Figure 10  shows  the HPLC elution profile
                  obtained.  The  major peak,  82% of  the radioactivity associated with
                  the adducts, co-chromatographed with the optical  standard-adduct
                  formed between  deoxyguanosine (dG) and  anti BPDE.   This  cell-
                  mediated BP-DNA  adduct also co-chromatographed  with the adduct
                  formed in normal diploid human  skin fibroblasts exposed for 2 hr to
                  anti  BPDE (12).    The  minor  radioactive  peak  in  Figure 12  co-
                  chromatographed with  the optical standard formed between dG and
                  syn BPDE.   These  results  are consistent  with reports on BP-DNA
                  adducts formed in human explant tissue from lung (19), colon (20), and
                  bronchus (21) as  well ^s cells derived from  human  lung carcinoma,
                  A549, in which the N  -dG-anti  BPDE  adduct was the major adduct
                  observed.
                         o_
                         Q
                               0   10   20  30  40  50  60  70  80  90  100
                                                TIME (MIN)
Figure 10.  HPLC profile of the DNA nucleoside adducts formed in 835 cells coincubated
with XP cells. The major peak is the N  -dG-anti BPDE adduct and the minor peak is the
N2dG-syn BPDE adduct.

-------
      C.    Time-Dependent Dose Response of XP Cells to BP Using Cell-Mediated
            Metabolic Activation:

                   To further characterize our assay system to changing doses of BP for
            varying times of exposure, logarithmically growing XP cells were plated at
            200,000  per 30mm diameter  culture  dish.   After  2k hours, X-irradiated
            transformed epithelial cells were plated at  a final attached cell density
            equal  to  that  of the XP cells.  The  cultures were exposed to increasing
            concentration of BP for  24 or 48 hours and then replicate sets of treated
            cells  were trypsinized and  replated  at  cloning densities.   In order  to
            determine  the  number  of XP  cells  in  co-cultivated  treatment dishes
            accurately, a dish  containing XP cells alone without epithelial cells was
            counted.  As shown in Figure  11, the percent survival of XP cells decreased
            with increasing concentration of BP and in addition, the 48 hr incubation
            resulted in greater cytotoxicity than the 24 hr.
0
                                  CONCENTRATION B(a)P
                                  0.2    0.4    0.6    0.8
                           10 -
Figure 11.  Cell-mediated cytotoxicity  of  XP cells incubated  for  24 or 48 hours with
lethally-irradiated 835 tumor cells in the presence of various concentrations of BP. After
the coincubation, the cells were trypsinized and the XP cells assayed for cloning ability.
                                          3 21

-------
VII.   SIGNIFICANCE

             In  vitro assays of  mutagenicity and transformation are used as short term
      tests to determine the potential danger of unknown agents as well as to carry on
      mechanistic studies regarding the mechanisms of mutagenesis and transformation.
      In both types of studies, one is frequently faced with  the need to utilize agents
      which require  metabolic  activation.  It is clear from  published studies that  the
      incubation of target cells with  preparations from liver homogenates such as  S-9
      fractions or microsomal  preparations, results  in the  production of  potentially
      mutagenic and transforming adducts in the DNA  of the cells, but that these may be
      different from those actually produced in the DNA of the cells used to make  the
      homogenate.   Thus, the  conclusion  that  a particular compound  is  potentially
      mutagenic may be the correct one, but for the wrong reasons, i.e., the wrong DNA
      adduct was produced.  It is obvious that any reliance on short term assays which
      makes  use of  the  metabolizing ability of cell  homogenates may - be  held up to
      question  precisely  because of  this problem.  This  poses  potential problems  for
      regulating agencies as well as for those interested in mechanistic studies.  The  use
      of cells able to metabolize carcinogens as feeder layers allows one to produce  the
      expected DNA adducts in target cells which are the same as those produced in vivo
      and will therefore be useful for many types of studies.

-------
                             REFERENCES

Miller, E. C. and Miller, 3. A. Pharm Rev., 18, 805 (1966).

Mailing, H. V. Mutation Res., 13, 425 (1971).

McCann, 3. and Ames, B. N. Proc. Natl. Acad. Sci., 73, 950 (1976).

Huberman, E.  and Sachs,  L. Int. 3^ Cancer, 13, 326 (1974).

Langenbach, R., et. al. Nature, 276, 277 (1978).

Newbold, R. F., et. al. Mutation Res., 43, 101  (1977).

King, H.  W. S., Thompson, M. H.,  Brookes, P.  Cancer Res., 34,1263-1269 (1975).

Meehan,  T., Straub, K., and Calvin, M. Nature. 269, 725-727 (1977).

Santella, R.  M.,  Grunberger, D., and  Winstein, I.  B.   Mutation Res., 61, 181-189
(1979).

Bigger, C. A. H., Tomaszewski, 3. E., and Dipple, A.  Carcinogenesis, j, 15-20 (1980).

Maher, V. M., McCormick, 3. 3.,  Grover, P. L., and Sims, P.  Mutation Res., 43,117
(1977).

Yang,  L. L., Maher, V. M., and McCormick, 3. 3.  Proc.  Natl. Acad. Sci.  (In press,
1980).

Maher, V. M., and Wessel, 3. E. Mutation Res., 28, 277 (1975).

Heflich,  R. H., Hazard, R.  M.,  Lommel,  L., Scribner, 3. D.,  Maher,  V. M., and
McCormick, 3. 3. Chem. Biol. Interact., 29, 43-56 (1980).

Maher, V. M., Dorney, D. 3., Mendrala, A. L., Konze-Thomas, B., and McCormick,
3. 3.  Mutation Res., 62, 311 (1979).

Krahn, D. F., and Heidelberger, C. Mutation  Res., 46, 27-44 (1977).

Painter,  R. B. 3n. Environ. Pathol. Toxicol., 2, 65-72 (1978).

Maher, V.  M., and  McCormick, 3. 3. Chemical Mutagens, Principles and Methods
for their Detection, 6, 309-329 (1979).

Shinohara, K., and Cerutti, P. A.  Cancer Lett., 31, 303 (1977).

Autrup, H., Harris, C. D., Trump, B. F., and  3effrey, A.  M. Cancer Res., 384, 3689
(1978).

3effrey,  A. M., Weinstein, I. B.,  3ennette, K. W., Grzeskowiak, K., Nakanishi,  K.,
Harvey, R. G., Autrup, H., and Harris, C.  Nature,  2691, 348 (1977).

Feldman, G.,  Remsen, 3., Shinohara,  K., and Cerutti, P.  Nature, 2741, 796 (1978).
                                       32;*

-------
                              Discussion

Dr. Morris, EPA:  Certainly within the regulatory sphere we are interested
in batteries of tests and so on for predictors and do you envision then
that as we proceed, certainly in the rule-making exercise for Section A of
TSCA we might be involved in not only selecting the cell type for the
mutation or transformation, whatever assay but we are, also, maybe wanting
to hone in on the specificity of the cell line for metabolizing?  Is that
the direction we are going in?

Dr. Waters, EPA:  Yes, I think that is definitely true, especially as we
progress to higher levels of testing.  I think what I have said in terms
of the specificity of metabolism should not be taken as an indictment of
an Ames test and its S-9 preparation because I still believe that this is
a very good detection system.  However, as we attempt to extend results
from detection systems to confirmatory in vitro bioassays, and to the
level of the in vivo bioassay we must be much more concerned about the
specificity of the metabolism.  We must be concerned about it if we are
to have really relevant confirmatory bioassays.  So, I would see the kind
of work that is being performed under this grant at Michigan State University
as being indicative of the type of testing that ought to be carried out in
the confirmatory phase of assessment of compounds.  Does that answer your
question?

Dr. Morris, EPA:  Yes.

Dr. Kraybill, NCI:  I am Dr. Kraybill, NCI.  I am still not quite clear then
because I feel that using the S-9 fraction here, you are getting the result,
but the relevancy of that result to an in vivo system; after all, that is
what we are interested in, what happens in man, how do you look then upon
the Ames system with the S-9 fraction here, as an indicator or—

Dr. Waters, EPA:  Yes, as an indicator system.

Dr. Kraybill, NCI:  Just simply that?

Dr. Waters, EPA:  Right, and in point of fact, it is the correlation that
has been developed between Ames test results for many compounds and results
of whole animal carcinogenesis bioassays that is the real strength of that
assay.  I think that these correlations argue very strongly for the use of
this kind of test as a detection system.  However, when the question of
specificity of metabolism arises, as we progress from the detection level
to the confirmatory level of testing we should be concerned about the
formation of specific metabolic products in the intact animal and possibly
in man.  I think that is the reason that we need to be doing this kind of
research—to solidify our confidence in the kind of metabolism carried out
in confirmatory bioassays.

-------
Dr. Chandler, NIOSH:  Jerry Chandler, NIOSH.  Mike, there are two facts
that we have to keep in mind, I suppose.  One is that many of the reactive
intermediates that are generated by microsomal oxidase systems are extremely
short-lived and may not be as long as the BAP half life or reactive inter-
mediate.  Secondly, when you use two cell systems, one is an activation
system.  You are requiring that the compound must diffuse out of the one
cell across the other membrane into another cell.  Do you think this is
going to be generally applicable?

Dr. Waters, EPA:  The data that has been obtained up to now by a number of
investigators, Huberman and Sacks in Israel, Weibel in Germany, indicate
that despite our concerns, the ones that you mentioned, that a compound
has to be metabolized in one cell, get out of that cell and into another
to show its effect, despite those concerns, in many cases, with a number
of compounds that require metabolic activation, the systems are working.
I believe it must not be terribly difficult for this cell-cell interaction
to occur.  The key element may be the proximity of the two cell types.
In fact, it appears that they must be in close proximity to one another.
I don't know how far one can separate the activation cell on the one hand
and the indicator cell on the other.  We have performed some experiments
in our laboratory where we have interposed dialysis membranes, and it is
still possible under those circumstances, in the case of compounds, for
the metabolites to get out and to enter the indicator cells.  So I don't
yet know what the limits are, but I do think, that based on the results
obtained thus far, that it does seem feasible to use these kinds of
systems.  Also, it is indicative, I think, of an important concern that
we probably should have, and that is that metabolism that occurs in one
cell type in an intact organism may be highly influential on another cell
type.  That is something we need to keep in mind.

Dr. Hegyeli, NCI:  Hegyeli, NCI.  Mike, I see from your description of the
procedure that these cell lines, the feeder cell lines are derived from
human tumors.  Did you ever try to use primary cells?

Dr. Waters, EPA:  This is, of course, not our own work but that of
Drs. McCormick and Maher.  We have used primary cells in our laboratory,
yes, and they have as well but not as a part of this particular program
to date.  We have used primary hepatocytes, and we are using primary
bladder cells and primary lung cells.  As a specific example, Dr. Robert
Langenbach in our laboratory has worked with a series of nitrosamine
analogs.  He finds that using whole cell hepatocyte activation, the
correlation that is obtained for mutagenic activity in V-79 cells and
carcinogenic activity in whole animal systems is much better than that
obtained if mutagenic activity in an Ames test using S-9 activation is
compared with carcinogenic activity in the intact animal.  Please, let
me say again, I am not indicting the Ames test.  I am simply indicating
that for certain kinds of chemicals whole cell metabolism may be more
relevant to the intact animal, and I agree with you that primary cell
                                       325

-------
metabolism may even be better than that observed in many cell lines.
However, primary cells in culture do have some significant disadvantages.
They are much harder to reproducibly prepare and control.  If we can
select cell lines that can carry out a broad range of metabolic activities,
then I think they should be, in theory, preferable to the primary cells
because of the difficulty of having to recover cells from the intact
animal for each experiment.

Dr. Herberman, NCI:  Herberman, NCI.  Is there some concern that different
cell lines might metabolize the same agent somewhat differently and that
if you use as the primary screen the direct cytotoxicity that that might
not reflect the metabolites that you would be interested in for mutagenic
or carcinogenic effects?

Dr. Waters, EPA:  Yes, the Painter screen was used primarily because of
its rapidity, and they were able to examine a series of 19 different cell
lines very quickly.  However, before they would recommend the use of these
cell lines for screening purposes it would be essential to confirm that
the metabolic activation capability possessed by these lines is indicative
of the type of metabolism that we are looking for in vivo.  So, this is,
in fact, what they have done in one part of the study.  Did I answer the
first part of your question?  Would you repeat that?  Maybe I did not quite
catch all of it.

Dr. Herberman, NCI:  The question was whether the primary screen that was
being used was the best one or whether it would be better to go directly
to a mutagenic or carcinogenic one?

Dr. Waters, EPA:  I think it would, but of course, it takes a lot longer to
do that, and they are doing it secondarily, and as long as it is done before
you propose using those cells, I think it is probably okay.

Dr. Morris, EPA:  Thank you, Mike.
                                       321

-------
STUDIES OF ORGAN-ASSOCIATED ANTIGENS AND OTHER MARKERS IN HUMAN TUMORS
        WHICH MAY BE USEFUL FOR THE DIAGNOSIS OF MALIGNANT DISEASES


                                Ronald Herberman, M. D.

                              Laboratory of Immunodiagnosis
                         Division of Cancer Biology and Diagnosis
                                National Cancer Institute
                                   Bethesda, Maryland

  The main rationale for these projects is to help identify test procedures that could pick
  out individuals in a population  who might have been exposed  to carcinogenic agents and
  would be the ones who would actually be coming  down with malignant disease. This is a
  generally  important problem,  to be  able  to screen general  populations  or  high-risk
  populations and to identify the relatively small numbers of individuals who harbor occult
  malignancies.

  The particular objectives in  this area have been  to focus on immunologic markers that
  might be useful to diagnose certain  types of human malignancies, particularly in this type
  of screening approach, and the  focus for the contracts that are being supported under this
  program has been primarily  leukemia and  breast cancer, but these  projects also  have
  possibilities for other types of cancer.

  There are some particular goals  to keep in mind that are needed to  be successful with
  tests  of this  type.    I might  note  at  the outset that accurate detection of occult
  malignancies is  a  very  difficult  task  to accomplish,  although  it  is  obviously quite
  important.

  One of the first criteria that one would be concerned with would be to have a test with a
  high degree of specificity.  Particularly, it would be desirable to have a test that would be
  able to  not only detect the presence  of cancer but would be able to get some indications
  for the type of cancer or where it would be arising. This is one of the reasons to put some
  emphasis on markers that would have some tissue or organ association.

  The other side of the coin is to have tests  or markers which would have a high level of
  sensitivity.  There are many tests  in  which the markers are positive  with cancers when
  they are present  at a fairly advanced stage, when there is a  lot of tumor present.  It is
  considerably more difficult for a test to be positive in patients with small amounts of
  tumor, when it is early, localized, and at a point that is treatable.  That is clearly the
  stage  that one would like to have  sensitive tests, in order to be  able to begin or alter
  therapy and thereby affect the course of  disease.

  Finally, if one  is  going  to  develop  assays  that might be valuable for screening of
  populations, even quite high-risk populations,  this would involve  testing of quite large
  numbers of individuals.  Therefore, it would be  necessary to have an asssay procedure
  which would be practical for  large  numbers  of specimens  and  would,  also, be quite
  reproducible.'

  There has been  quite  a large  effort in this direction.  The National Cancer Institute
  specifically has many grants  and  contracts  that have been designed to develop immuno-
  diagnostic or  other  diagnostic markers for cancer.  The  two projects  that are being
  supported under this program  essentially  are  just a small  subsegment  of  this type of
  approach, but they are ones that were thought at the time of  initiation of the program to
  be ones  that were promising for these  goals.
                                            32'

-------
The first particular project that we are concerned with is at the University of Minnesota
and has focused on leukemias and lymphomas.  The investigators  involved  with this are
Drs. Kersey and Tucker Lebien.

The particular  approach that they started out with was to prepare heterologous antisera,
primarily  in rabbits,  that  would react against antigens that would  be associated with
various types of human leukemias, either in a quite specific way for the leukemias, or that
would  see  differentiation  antigens,  that would not be  entirely leukemia specific.   The
latter  type might still be  quite  useful  as a  detection  system  since  in the  normal
development of the  hematopoietic cells these markers  might only be present in a very
small proportion of cells or in early stage of development.  Thus, in this situation, one
would  be  focusing more  on  quantitative differences  rather  than on  real  qualitative
differences.

The approach  which has been adopted  during the past year  or so has been to shift away
from the more classical approach of just injecting  cellular  or subceilular materials into
rabbits or other species, and to go to an approach which in general seems to be much more
promising for  specificity and, also,  for developing  large amounts  of  the same  reagents.
This involves the production of monoclonal antibodies in mice that might be able to see
the same or perhaps an additional series of antigens.

It  may be worthwhile  to first briefly describe the methodology of monoclonal antibodies.
Cells or subceilular materials are injected into mice and, at  a time thought to be optimal
for getting  an  immune response, spleen cells are taken from the mice.  These lymphoid
cells are co-cultured  with an established  myeloma cell line, preferably a myeloma cell
line which itself does not make antibodies or immunoglobulins.  The transformed cells are
fused with the B cells that are in the spleens of the immunized animals; this  fusion occurs
quite readily in the presence of polyethylene glycol.

After fusion, the cells are cultured for a period of time in so-called "HAT" medium which
is  a combination of cytotoxic agents to which the parental myeloma line is quite sensitive.
Thus,  the  unfused cells  would be  eliminated  and only the fused  products that  have
resistance to these materials would  be  able to survive.  Supernatants from surviving cells
in  various wells are screened for the reactivity that one is trying to measure.  This is the
most critical part of this  methodology and one  has to  be clever enough to have a good
screen. The type of screens that the group in Minnesota has used has involved the use of
paired  cell lines; on  the one  hand,  the leukemia line  to look  for positive antibody
reactivity against leukemia associated antigens  and on  the  other  hand, an  autolygous  B
cell line transformed  with Epstein-Barr virus as the  negative part of the screen.  Thus,
they are looking for antibodies that would react with the leukemic line and not with the 3
cell line from  the same individual.   They have been  able to derive several clones from
these fusions that have looked quite promising for antibodies  that  could  be useful to
detect leukemia-associated or differentiation antigens.  One of these monoclonal anti-
bodies  was raised against a T cell acute lymphocytic leukemia line, HSB.  This particular
monoclonal  antibody has not been nearly as selective as one would like, at least in the
peripheral blood since  it reacted with a large majority of T  cells among peripheral blood
mononuclear cells.  However, it has reacted with a series of acute leukemias and leukemic
cell lines but not with other types of acute leukemias.

-------
There have also been some monoclonal antibodies that have been raised to another acute
lymphocytic leukemia cell line, NALM-6.  One monoclonal antibody to NALM-6 reacted
to both the ALL line and the autolygous B cell line, and it appears to be detecting some B
cell associated antigens.  In the peripheral blood it seems to be reacting with more or all
normal B cells, but in the bone marrow it seems to be much more selective, reacting with
very few of the cells.  For the screen against leukemias it has been quite effective in
picking  up all of the  B  type chronic  lymphocytic leukemias  and a  series  of  acute
lymphocytic leukemias.

The third monoclonal antibody that they have raised seems to be the most interesting one.
This reacts  against  an  ALL-associated antigen which seems to be very similar, if not
identical to one which they had  previusly  identified by heterologous  antiserum.   The
specificity for this  antigen seems to be a differentiation antigen associated with pre-B
cells.  Very small numbers of normal  B cells in the bone marrow react with this, but
almost no cells in the normal peripheral blood react.  Yet, the antiserum reacts with a
considerable portion of acute lymphocytic leukemias, mainly those which are classified as
the most common type  of ALL, the so-called "non-T, non-B"  ALL and also  reacts with
so-called  "pre-B ALL"  and also  with the cells from chronic  myelogenous leukemias in
blast crisis.  This last group is particularly of interest because those types of leukemias
have been asociated with  some of the carcinogenic effects seen with cytotoxic therapies
in cancer patients.

Although the  first two  antibodies that I described have not been nearly as selective as
would have been hoped,  one of the promising aspects to the overall study  is that among 40
different individual  leukemias  that have been screened, there was essentially no over-
lapping between leukemias positive with one of these antibodies compared to the others.
Thus, by using a panel of these three reagents it was possible to pick up  almost all of the
leukemias whereas  each  antibody  detected only  a  subset.   This may be  the  type of
approach which will  be taken in this area, to combine several markers since one particular
reagent may not be sufficiently sensitive or specific.

The other project which is being supported in this area is at Emory University with Dr. R.
Chawla as the principal  investigator.  The focus has been  on a marker called EDC1.

EDC1 is a protein which has been found in the urine  of cancer patients.  It is a relatively
low molecular weight glycoprotein.  It has been possible to identify EDC1 in the urine of
the majority of patients with metastatic breast cancer,  and a variety of other types of
solid malignancies.

This is  not  a  tumor-specific  protein but is actually a degradation product of a normal
serum protein, inter-alpha-trypsin inhibitor, that is present in everyone's  serum that has a
much higher molecular  weight (170,000).  Under normal  circumstances the serum protein
is not degraded to produce detectable levels of EDC1. Therefore, the focus in most of the
studies  of  Dr. Chawla has been to screen the  urine  for the lower  molecular  weight
compound.  The problem  with this approach has been  that the antibodies that he has
available have cross-reacted immunologically between the serum protein  and EDC1.

Therefore, he  was faced  with the  task of finding a method to more specifically detect
EDC1.  To a certain extent, screening in the urine provided a biologic distinction because
the large molecular  weight protein under normal circumstances would not be filtered in
the glomeruli, whereas the lower  molecular weight  one would.  The problem with this
approach is that with any impairment of glomerular function, there could be some leakage
into the urine  of the higher molecular weight material.
                                           321

-------
Dr. Chawla has corrected for that to a large extent by making a ratio of the amount of
immunologically active material detected to the creatinine that was excreted into  the
urine.   However, this was  not entirely selective.   He  has been  able  to develop  a
radioimmunoassay which is much more sensitive than the early ones.  However,  it detects
both proteins,  and there have been some false positives in tests  on  urine which  are
probably related  to the subtle impairments in  renal function which can occur in some
tumor-bearing individuals.

Dr. Chawla has been trying  other  ways to discriminate between the two proteins.  One
possibility  was two-dimensional crossed immunoclectropheresis, since the charge, as well
as the size of these molecules, was different. However, although the initial studies looked
very promising, with clear separation of the two proteins, there have been some normal
individuals who have  had an  unexpected degree of heterogeneity, with a  peak of an
intermediate molecule between the one and the  other.  This has caused some confusion in
the assay.

The more  recent  thing that Dr. Chawla has come  up with has been to precipitate  the
parent molecule with  sulfa-salicylic acid.  This seems to be quite effective in precipi-
tating the  normal serum protein without affecting the lower molecular weight compound.
This seems to be a very good  basis for an assay for distinction.

The objective of  Dr.  Chawla's project, once he works out this  technical problem, is to
screen a large number  of sera and urine which he is currently collecting from patients who
are being evaluated for the initial diagnosis for  breast cancer.  There are specimens from
patients with breast cancer and also sera from  women  who turn out to have  only  benign
breast disease.

At this time, with metastatic breast cancer  patients,  the  large majority have levels of
EDC1 that are above the range that is seen in either  normal individuals or  in patients with
a  variety  of benign diseases, including benign breast lesions.  Elevations also have been
seen in a variety of other types  of malignancies, including  acute myelogenous  leukemia.
The hope is that with the  radioimmunoassay, which is much more sensitive than the assay
used to generate  the current data, and some kind of selection procedure,  there  may be  a
specific and sensitive procedure suitable for the  ultimate objective of this  project.

-------
                                     Discussion
Dr. Kraybill, NCI: We received a call the other day from an authority in the South Dakota
State Department of Health.  He has come across some records of 1400 tests that were run
with a high indication of association  of another biochemical marker, hyaluronidase, that
might indicate tumorigenicity in people.  Have you heard anything about  this particular
marker?

Dr. Herberman,  NCI:  I am not aware of detailed documentation that hyaluronidase  is
useful.  There have been a series of enzymes that have been  put  forward as possible
discriminants. In fact, my experience has been that investigators do a study by comparing
advanced cancer with normals and see some very interesting differences.

Unfortunately,  you  can get  the same  kind of  differences very frequently  with  the
erythrocyte sedimentation  rate.  The  bigger  difficulty  is to be able to get adequate
discrimination between  cancer, particularly at a  localized stage, and  benign diseases,
especially of the same organ system as the cancer.

As part of  the NCI's program we  have a serum  bank in which specimens are sent to
investigators that have assays that they think are promising for making  these kind of
discriminations.   Over about the five-year period that this has been available, there have
been about  200 or so assays that have been screened, including a number of enzymatic
assays of the type you have mentioned.  Unfortunately, only a handful of  them has had
significant levels of discrimination when tested with the  various groups of patients and
controls that I mentioned.

Dr. O'Conor, NCI:  I would  like to  raise a general and somewhat provocative question.  I
wonder whether you would  comment on the potential of the three-dimensional system
that Norman Anderson has  at  Argonne.   I think  you are familiar  with it.   Would  you
comment about where that might fit into this whole problem?

Dr. Herberman,  NCI:  I think this  is a very impressive type of methodology.  It has the
ability, as I understand it, to look at a very wide cross section of metabolic products and
not only immunologic but,  also, biochemical types of markers.  It  might well pull out
something which would be quite interesting.

With the serum bank that  has been available there have been some things in the sera that
have looked to be quite discriminating.  For example, Dr. Phyllis Brown in Rhode Island
has been looking at liquid  chromatographic separation of nucleosides that are in sera.
There  are  several nucleosides that she has been able  to identify  that  seem  to have
considerably different patterns  in cancer  patients, including some  early stage cancer
patients, compared to some  benign diseases and normals.  Certainly the Anderson type of
 approach should  be able to  pick up this as well as a variety of other things.  I guess one of
the concerns that I have with that  aproach  is  that you may be  really flooded with
information.  There will  be so much information gathered that it  may  be  difficult to
actually sort out the relatively small number of interesting or important differences from
the background of so many other  nondiscrirninating materials.  The other thing  that  I
would have to add is that, although this technology is quite intriguing as an  approach, I am
not aware that it has been validated for this kind of application. We  will really just have
to wait and  see  how well it will do.  Unfortunately, among the many potential tests
screened thus far, there has been a low yield of tests that have held up under scrutiny.  A
very small proportion of tests can discriminate between  known early cancers and  known
benign diseases and  even  fewer have the levels of sensitivity and specificity to  screen
even  a high-risk population.   For  screening you need  extraordinarily  high  levels of
                                        38.1

-------
sensitivity  and specificity and  I am not  convinced that there are any  tests that are
adequate for screening for the types of  cancers that we are concerned with in the United
States.

Dr. Morris, EPA:  As I understand, in the past with breast cancer  we had HCG and CEA
and one other test which I cannot remember now that was used in a battery approach.  Do
you feel that we are far enough along to consider some type  of battery even at this point
or do you feel that'is premature?

Dr. Herberman,  NCI:  I think the battery approach probably will be the way to go.  The
problem is to decide what to put into the battery. My view is that each of the tests in the
battery has to, on its own right, be reasonably powerful.

Some of the evaluations that have been done  have included  some things in the batteries
that are marginal, and if you put together a few marginal tests, you seem to end up with
marginal data in the aggregate with their problems adding up  as well as their strengths.

In fact, the combination of tests for breast cancer that you are  mentioning is an example
of the problems in this area.  HCG is a marginal marker for breast cancer.

Dr.  Morris, EPA:  You are  saying  that for  high-risk  groups it is  far  enough  along,
determining,  you know, malignant  disease  versus benign disease.   Do  we  have  enough
information or should we start considering this. I was thinking about my ACB population
for example that  are potentially at risk.  Do we have  enough information  even in the
present battery,  admitting  we  may wind  up  with  some benign  in  the process?  How
significant  would it be, even at this point to approach that  with some  of  these selected
populations, not broadly but in select cases?

Dr.  Herberman,  NCI:  I think  an example that one can think about specifically in this
regard  is  CEA (carcinoembryonic  antigen).  If one  looks  at  cancers,  carcinomas for
example, versus  a normal healthy population, it actually performs rather well, in that with
colon cancer,  breast  cancer, lung cancer, at  least 50 percent or more  of patients have
elevated levels of CEA as compared to less than 5 percent of the normal population.  That
sounds pretty good, but there is, also, a range of benign diseases with elevated levels.  The
experience has been that when  CEA is  applied for screening, most of the elevations are
due  to  benign diseases since they are much more  prevalent in the population than the
cancers. For every cancer that  may be  picked up, there may  be 50 or 100 benign diseases.
In addition, the problem with markers of this type is that they  tend  to increase in their
levels  or even in  their detection  rate  with more advanced  disease.  Thus, most of the
cancers that were picked up  were advanced and untreatable.

The  one test which has really shown promise  and has been applied at a large-scale level
for  screening  has been  alpha  fetoprotein.   The  Chinese  have  utilized  this test for
screening in a high-risk area and have shown  reasonably  well that they could use this  in
field conditions.  They could identify a significant number, at least, of  people with lever
cancer, including some surgically resectable liver cancers.   However, the yield of really
treatable cases was very low for the number of people that they had to screen, which has
been in the neighborhood  of one-half million. At  most 10 or 20  treatable liver cancers
were picked up.

-------
                   EFFECTS OF CARCINOGENS, MUTAGENS
                  AND TERATOGENS ON NON-HUMAN SPECIES-
                            AQUATIC ANIMALS
                                 by
                    John A. Couch and Peter W. Schoor
                    U.S. Environmental Protection Agency
                    Environmental Research Laboratory
                    Sabine Island
                    Gulf Breeze, Florida 32561

     A research program using aquatic systems and organisms to study the
fate and effects of carcinogens in the aquatic environment has been underway
during the FY's 78 and 79 at the Gulf Breeze EPA Laboratory.   The two major
investigative, disciplinary areas have been in pathobiology and biochemistry.
A fish-carcinogen assay system has been developed that involves laboratory
controlled long-term exposure of fish to suspect carcinogens  followed by
histopathology and physiology of exposed fish, induced growths, and related
disorders.  This system appears adequate for carcinogen tests for fish.   A
two year field study is underway of tumor, disease prevalence, and carcin-
ogen residue or metabolites in fish and shellfish in variously polluted
estuaries along the northern Gulf of Mexico.  It is too early to predict
the significance of the findings in this study, but several new tumor types
have been discovered in fish.  Biochemical studies have revealed that fish
may respond in enzymatic reactions in ways similar to mammals exposed to the
same carcinogens, and that fish may prove to be adequate supplemental, bio-
logical monitors of carcinogens in the environment. It was shown that aside
from induction of oxygenase activity, transferases responsible for detoxi-
fication reactions are also induced, the latter by metabolites rather than
parent compound.  The significance of this is that while the  oxygenases  are
responsible for producing ultimate carcinogens from procarcinogens, the  trans-
ferases are involved in the excretion and hence detoxification of the oxidized
metabolites.  An extra-mural (grant) program consisting of eight grants  to
Academic investigators has supplemented specific areas of investigation  in
the overall project.  Findings from individual grants and cooperative agree-
ments are reported for work completed to date.
Acknowledgement:
     This research is supported by an Interagency Agreement between the
     National Cancer Institute and the Environmental Protection Agency.
                                        333

-------
                           Ill Introduction
    A major problem faced by the National Cancer Institute and the
Environmental Protection Agency is that of aiding in determining the
fate and effects of carcinogenic pollutants in the larger environment.
It is one thing to track the routes,  fates, and effects of a pollutant
in an industrial system or even in a  metropolitan area, but quite another
to understand the behavior of single  pollutants in the total environment.
Exposure of human populations to most carcinogens occurs insidiously in
the general environments of air, land,  and water.  This does not minimize,
in any way, the necessity to study the  behavior of carcinogens in special,
limited environments,  such as industrial, urban, or suburban complexes,
but we know relatively little about the risks of human exposure to pollutants,
such as mutagens and carcinogens, in  the general environment.

    One way to approach the problem of  understanding risks of general ex-
posure is to study wildlife populations that are widespread, but which  live
in environments where exposure to ambient pollutants is certain.  The use
of wildlife populations as  surrogates  for human populations may be con-
sidered to be a novel expansion or logical extension of the use of labor-
atory animals and animal models of human diseases as alternatives to use
of human subjects.  The sharing of biologic characteristics by phylogeneti-
cally diverse species makes certain comparative approaches possible.

    Problems arise, however, when such factors as selection of sensitive,
indicative species, geographically adequate populations, and indicative
segments of the air, land, water Biosphere are considered.  In this regard,
Office of Research and Development laboratories, such as the Gulf Breeze
Environmental Research Laboratory, have exemplary, pilot research programs
that are investigating the use of aquatic animal species as indicators of
the presence and potential effects of toxics, particularly of certain  car-
                                        331

-------
cinogens, in the larger environment.  The Gulf Breeze pilot research program




has been under way since August, 1978 and is supported jointly by the Office




of Research and Development and the National Cancer Institute through an




interagency agreement.  The Gulf Breeze studies are based on the premise




that the aquatic portion of the biosphere (water, biota, and sediment) is




the ultimate "sink" for the runoff, fallout, and discharge of most toxic




pollutants.  In addition, animals living in the relatively efficient solvent




water are more intimately exposed (total exposure through body surfaces,




gills, alimentary tracts) than probably are most species living in terres-




trial or air environments and are less likely to easily escape a dissolved




or carried pollutant.




    At Gulf Breeze, researchers are studying species of fish and shellfish




along the Northern Gulf of Mexico (Florida, Alabama, Mississippi) in order




to determine which are good indicators of the role of carcinogenic agents




in the environment.  Fish, oyster, and clam populations are sampled monthly




for determination of prevalence of tumors, cellular diseases indicative of




pollution, and chemical analyses of residues of potential carcinogenic




chemicals.  Sampling stations are located in both polluted and clean estuaries,




as well as offshore in'relatively pristine waters.  Select species of fish




are exposed in long-term assays in the laboratory to determine their specific




tissue, cellular, and biochemical responses to known chemical carcinogens




that may occur in the environment.




                            IV Objectives




Objective 1:




Carcinogenic or suspect carcinogenic substances often enter the aquatic




environment as pollutants and pose a multi-threat to aquatic ecosystems.




Because most carcinogens are mutagens, the major risks  to aquatic ecosystems
                                        33T

-------
and component species are long-term mutagenlc and teratogenic effects ex-




pressed both at the organismic and population levels.   Commercially and




ecologically valuable species such as fishes, oysters,  and shrimps may be




adversely affected by carcinogens in the form of cellular proliferative




diseases and mutagenic effects.  Populations of valuable aquatic species




impacted by mutagenic substances will be studied in order to predict effects




on population stability and survival.




Objective 2:




The intake of carcinogens, mutagens, and teratogens by man comes in part




through his food.  Our chief concern in this area is the need to determine




if aquatic species accumulate and convert procarcinogens to proximal car-




cinogens and thus pose a direct carcinogenic threat to man in his consumption




of seafood.  It is therefore essential to know the routes, rates, and




reservoirs and metabolism involved in the accumulation of these compounds




in aquatic food webs.




Objective 3:




Aquatic organisms may be exposed to carcinogens from run-off, fallout and




discharge of pollutants into the aquatic portion of the biosphere which




behaves as the ultimate-pollutant "sink".  Select species in the aquatic




environment have potential to be used as sentinel or indicator systems to




reflect the presence, behavior and effects of carcinogens, mutagens, and




teratogens.  Comparative laboratory and field studies of select species




for uptake, accumulation, and effects of known carcinogens will reveal the




best modes for utilization of aquatic species as indicators.  Results .of




studies with sentinel species will be used to determine when further testing




is required and may offer pertinent information on mechanisms of effects.
                                       336

-------
                     V Methodological Approaches




    To contribute to the realization of the preceding goals, we have.underway




both in-house and extra-mural complementary projects.  The overall project




is divided into two major disciplinary approaches:   1) Pathobiology and 2)




Biochemistry.  Therefore, methods outlined below and results in the next




section will be reported under these two complementary disciplinary headings.




Results of grants in progress during FY 79 will be  included at the end of




this report.  The disciplinary area of each grant will be identified by




project office (Couch - pathobiology; Schoor - biochemistry).




1.  Pathobiological Methods




    a)  Fish carcinogen assays, toxicity, and histopathology of induced lesions:




        Aquatic species such as oysters and fishes  that are exposed to




        carcinogens in special flow-through laboratory systems will be studied




        to determine relatively toxicity (lethality and dysfunction as criteria)




        and structural effects (histopathological and morphological changes




        as criteria).




        Long -term exposures to low concentrations  of select carcinogens will




        be carried out to determine if tumors or cellular proliferative disorders




        similar to those reported in feral specimens of oysters and fishes can




        be induced.  Exposure of oysters and fish for periods of one year or




        more are underway.  Information from these  tests should aid in deter-




        mining if tumors found in feral invertebrates and fishes in field




        monitoring studies are indicative of chemical carcinogenesis related




        to specific  pollutants as determined experimentally.  Studies similar




        to this for other systems are being supported through- cooperative




        agreements (Couch, Martin, Hendricks, Sinnhueber - see grant progress




        reports at end of annual report.)




    b)  Field epizootiological survey of cellular diseases and carcinogen




        residues in feral fish and shellfish:  A survey of tumors and cellular








                                          337

-------
       diseases  in oysters,  clams,  and  fish concomitant  with a point source




       pollutant survey is underway for a period  of  at least two years  along




       the Northeastern Gulf Coast  between Pensacola, Florida and Pascagoula,




       Mississippi.   Public  requests for tumor  bearing specimens have also




       been made.   The purpose  shall be to see  if there  is  a correlation




       between tumor  prevalence in  aquatic species and pollutant prevalence




       in specific coastal areas.   Four stations  are sampled monthly for




       oysters and fish which will  be examined  grossly and  histologically




       for neoplasms  or related disorders.  A concomitant survey of  indust-




       rial, agricultural and domestic  pollution  at  or near the sample




       stations  has been made.   Attempts to isolate  specific carcinogenic




       pollutants or  complexes  will be  made by  collecting tissue, and sediment




       samples for chemical  analyses as dictated  by  knowledge of pollutant




       sources.   A 1979-1980 cooperative agreement between  Dr.  John  Laseter




       and Gulf  Breeze will  permit  the  detailed qualitative and quantitative




       analyses  of tissue and sediment  samples  for the presence of suspect




       or overt  carcinogens. A similar study along  the  Oregon Coast of




       shellfish tumors and  carcinogens has been  supported  through a grant




       from the  project to Dr.  Michael  Mix, (Couch,  Mix).




2.  Biochemical Methods




   a)   Induction Studies: Aquatic  species such  as mullet, killifish, flounder,




       sea catfish and others have  been and will  continue to be exposed to




       inducers  of microsomal mixed-function oxygenase  (MFC1) activity either




       by intraperitoneal injection or  direct exposure in seawater.   Since




       times of  possibly one year might be necessary to  induce MFO activity




       by water  exposure, the direct injection  of inducer is used at the




       start of  the investigations  in order to  optimize  other parameters




       such as metabolite and conjugation reactions. The long-term,  low-

-------
       exposure route in seawater will  follow.   (Schoor, Melius,  Strength).




   b)   Metabolite Identification;  Metabolites  from the MFO reactions will




       be identified using  high pressure  liquid  chromatography coupled  with




       fluorescence detection  and confirmed  by stopped-flow fluorescence




       scanning.   Metabolite standards  for benzo(a)pyrene  (BaP) have been




       obtained from the Illinois Institute  of Technology  through the




       courtesy of NCI.   All the phenolic compounds have been chromatographed




       and their fluorescence   and emission  spectra have been obtained  in




       appropriate solvents.   All spectra will be  stored on discs for later




       data manipulation.  (Melius,  Schoor).




   c)   Conjunction and Excretion Studies




       Rats are being used  to  make a series  of conjugation products in  vivo




       by injection of   C-labelled BaP.   They will include glucuronides,




       glutathiones, and sulfates.  Their occurence in fish will  then be




       ascertained by comparison to the standards  produced in the rat.  This




       will be helpful in determining the final  disposition of a  carcinogen  like




       BaP within the animal and in what  forms the parent  compound is finally




       passed back into the seawater.  (Strength,  Schoor).




                    VI  Major  Findings  and Progress




     As noted earlier under Methodological Approaches results and findings




of studies to date are reported under the two  disciplinary areas  of Pathobiology




and Biochemistry with grant cooperative agreement  complementary studies




reported at the end of this paper.




1.  Pathobiology




    a)   Fish carcinogen assay: Major advances  have been made during FY  79




        in design, utilization and evaluation  of a laboratory, flowing-water,




        carcinogen assay system.  The system is  designed  to control water

-------
     temperature,  photoperiod,  flow rates,  and nutritional status of




     fish subjects for long periods and is  being evaluated in a long-




     term (11 months to date)  exposure of sheepshead minnows to the suspect




     carcinogenic  herbicide, Trifluralin.  To date,  70 to 90% of fish




     exposed continuously from fertilized eggs to 1-3 yg/£ Trifluralin




     have developed abnormal vertebral growths that  superficially re-




     semble osteomas,  or benigh boney tumors during  their early juvenile,




     and young adult periods.   Methods for  study of  these lesions have




     been developed and include histological,  histochemical and radio-




     graphic approaches.  The vertebral growths can  be located and diagnosed




     in samples of experimental fish taken  directly  from the assay system,




     anethetised and then radiographed in a laboratory x-ray unit.  Con-




     firmation of  any specific vertebral lesions is  then made by histo-




     logical examination of individual specimens prepared in special ways.




     Large numbers of control fish and feral fish (sheepshead minnows)




     have been examined similarly with no findings of growths.  We conclude,




     therefore, that the vertebral growths  are experimentally induced in




     our assay system.  This study continues in FY 80.




b)    Field epizootiological survey: This study has been underway since




     August, 1978.  To date, samples of fish,  oysters, and clams have been




     collected and examined monthly for the prevalence of tumors, or




     cellular diseases.  Thousands of (>40,000) fish have been examined




     grossly and internally for lesions; and over 7,000 oysters and several




     hundred clams have been examined for lesions histologically.  Fish




     and oyster tissues, and sediment samples have been collected for




     chemical analyses, underway at present.  Another approach has also




     yielded good  results - that of circulating requests for tumor bearing




     specimens to  the public (fishermen - sport and  commercial, clubs
                                      3 'i r

-------
and packing houses).   So far several specimens have been brought in




with cancer-like or truely neoplastic lesions.




   Summarized results are as follow:  Fish - numerous tumor-like




lesions with several  neoplasms diagnosed from field collected specimens;




oysters - two cases of leukemic-like (blood cell proliferative)




disease - one case from a polluted harbor, another from clean waters'




clams - several cases of external growths (polyps) to be diagnosed.




   Perhaps the most exciting and interesting finds were the specimens




brought to us by an aquaculture group in response to our public




circulars requesting  specimens.  Mr. Bill Tremble of the Mariculture




Center at Gulf Shores, Alabama presented us with three specimens of




Fundulus grandis, the Gulf Killifish, that had large white to yellow




growths on their heads and bodies.  Histologic examination of these




cultured fish revealed that the growths were probably very invasive




pigment cell tumors that grew rapidly.  These large, older fish had




been reared in ponds  treated with certain chemicals for control of




parasites and receiving ambient water from the intercoastal waterway.




The availability of cultured Fundulus with a history of invasive




neoplasia may provide us with an animal model with which to study




certain forms of neoplasia.  To date, the tumor has been identified




as an erythrophoroma, the tumor cells identified and biochemical




characterization of pigments has been completed.




   After only one and one-half years of study, from which only a




portion of the specimens have been analyzed, we cannot draw conclusions




or final correlations concerning disease prevalence and pollutant




occurence.
                                 341

-------
2.   Biochemistry




    a)   A Schoeffel RRS-1000  Spectrofluorometer was installed in the spring




        of 1979 and has been  used in obtaining the excitation and emission




        spectra of metabolites of BaP.   Twelve different phenols for BaP and




        all spectra were obtained.   The speculation that there would be




        enough characterizable differences in each compound's spectra was




        found sound for the phenols.  So far, the quinones have been shown




        to be extremely unstable in light.  It was shown that in the case




        of 3-OH and 9-OH BaP  the stopped-flow HPLC scan clearly identified




        the two isomers.  In  a mixture of the twelve phenols, ten can




        now be separated and  quantitated.




    b)   Enzymes studies to date are (1) UDP-glucuronosyl tranferase (E.C,2.4.1.17),




        (2) 3-phospoadenosine-5-phosphosulfate sulfotransferase (E.G.2.8.2.1),




        (3) UDP-glucose dehydrogenase (E.C.I.1.1.22),  (4)  glucose-1-phosphate




        uridyltransferase (E.G.2.7.7 .9), (5)  b-glucuronidase (E.C.3.2.1.31),




        and (6) aryl sulfatase (E.C.3.1.6.1).  We have observed the induction




        of the above transferases in rat tissue, but no induction of the




        dehydrogenase activity was found.  For UDP-glucuronosyl transferase




        induction by phenobarbital was 3-5 times, by 3-methylcholanthrene 2-3




        times, by phenanthrene 1.5-1.8 times, and by BaP 1.5-2 times the value




        of the control.  For  the 3-phosphoadenosine-5-phosphosulfate sulfo-




        transf erase induction by PB was 2 times, 3-MC 2 times, and weak for




        both Ph and BaP. All  inducers were fed in the diet at 0.1%.  It was




        observed that in the  case of 3-MC there was a large increase in liver




        size, which was not observed in the other cases.  It is presently




        suspected that the transferases are induced by the metabolites of




        the inducers. (Strength).
                                         34L

-------
c)  Our studies have shown that the mullet (Mugil cephalus),the sea




    catfish (Arius fells) and the gulf killifish (Fundulus grandis) possess




    MFO systems which are inducible by  Aroclor^ 1254 (mullet)  and by 3-MC




    (sea catfish and gulf killifish).   We have employed cytochrome b^ and




    cytochrome £450  reductase assays, carbon monoxide difference spectro-




    scopy, high-pressure liquid chromatography, and the Salmonella/microsome




    mutagenicity assay to monitor induction and evaluate PAH metabolism.




    Our studies indicate that these organisms possess MPO systems inducible




    by AroclorR 1254 and 3-methylcholanthrene and capable of benzo(a)pyrene




    metabolism.  B(a)P treated gulf killifish did not appear to metabolize




    B(a)P as efficiently and seemed to produce lower levels of B(a)P




    metabolites than did 3-MC treated gulf killifish.  This may have




    resulted because (1) B(a)P is a less effective inducer than 3-MC or




    (2) in vivo B(a)P metabolite may have induced the conjugation systems




    which would result in ethyl acetate insoluble and nonmutagenic in vitro




    B(a)P metabolites.  These studies show that certain similarities exist




    in the mechanics of fish and mammalian MFO systems even though differences




    exist in the activities of these systems. (Melius).




d)  It has been demonstrated that the mullet is able to hydroxylate B(a)P,




    that this activity is inducible by 3-MC, and that the metabolite profile




    and enzymatic activity are similar to those found in rats.  Proliferation




    and enlargement of ER strongly suggests that the mullet hepatocytes




    behave similarly to other vertebrate hepatocytes when exposed to enzyme-




    inducing chemicals.  Nuclear changes in the hepatocytes of the injected




    fish reflect cellular necrosis due to cellular intoxication.  In spite of




    feeding, early loss of liver glycogen may be indicative of nutritional




    as well as other stress such as confinement. (Schoor, Couch).
                                  34,':

-------
               VII Significance to Biomedical Research




                   and Program Needs of NCI and EPA




Progress achieved in FY 79 in this project indicate the following contributions




to biomedical interests and needs of NCI and EPA:




1.  The fish carcinogen assay system functions well for long-term testing of




    chemicals against the sheepshead minnow, a common Gulf and Atlantic coastal




    species.  This system can be considered now a  complementary test system




    to those used routinely for rodent assays at NCI and elsewhere.   New




    chemicals and some new species will be tested  in the future.   The system




    has been used in FY 79 to demonstrate the induction of a specific tumor-like




    lesion in a fish with a suspect carcinogen (Trifluralin).




2.  To date, the field study has shown that fish and shellfish from  the




    Northern Gulf of Mexico suffer from a variety  of lesions similar to lesions




    found in higher animals such as mammals (i.e.  pigment cell tumor in




    Fundulus).   This study is too incomplete, at present, to draw conclusions




    concerning early warning or sentinel capabilities of the aquatic populations




    under investigation.




3.  Biochemical and correlated structural responses of fish liver systems seem




    to be relatively similar to responses of mammals to certain carcinogens.




    This permits future comparative studies to determine if biochemical methods




    may be incorporated in early warning or sentinel monitoring project with




    fish.  The biochemical studies suggest that fish may serve as animal models




    in carcinogen (preneoplasia) studies to complement mammalian studies.
                                       344

-------
                VII Proposed Course - Future Plans









We plan to pursue the following efforts in FY 80-81 in the project:



1.  Continue and expand our use of the fish carcinogen assay system by



    testing new compounds and,  perhaps, by using Fundulus as a test organism.



    Carcinogens found in the field monitoring study may also be tested against



    select fish species.  Improve diagnostic techniques for preneoplastic



    and neoplastic lesions in fish and shellfish.



2.  Continue in the second year of the field (epizootiology) study of tumors



    in feral fish and shellfish populations.  Beginning in FY 80 will be the



    chemical analytical portion of the study.



3.  Continue and expand the biochemistry effort in the study of metabolism,



    conjugation, and excretion of carcinogens in fish and invertebrates.



    Examine the ways in which man may be exposed to carcinogens or their



    metabolites via the aquatic environment.



4.  Several new cooperative agreements for extramural studies will begin



    in FY 80.
                                   34
c
ci

-------
        IX Date  Contract  Initiated and Period of Contract Planned









Initiated                                      Expiration Date




October 1978                                   September 30, 1984









         X Contractors  Project Director




Dr. Herman Kraybill/NCI




           Project  Officers  for NCI or EPA




NCI:                Dr. Herman Kraybill




EPA/ORD:            Dr. Wayne Galbraith




EPA/Gulf Breeze:    Dr. John A. Couch, Dr. Hank Enos
                                346

-------
XI  Grants and Cooperative Agreements Funded Progress Reports
                              347

-------
Coop.  Agreement:             Development of a Carcinogen Assay System
                            Utilizing Estuarine Fishes


Identification Number:      R806212


Principle Investigator:     Dr. B.J. Martin


Project Officer:            John Couch


1) Long-term experiments in progress:


   a. Ictalurus injected i.p. with 50 yl 10% DENA (diethylnitrosamine) in


      distilled water.


   b. Cyprinodon injected i.p. with 50 yl 5% DENA


   c. Cyprinodon injected i.p. with 50 yl 1% DENA


   d. Ictalurus and Cyprinodon injected i.p. with 125 yl BEN (benzidine).


      Saturated solution in distilled water.


   e. Cyprinodon exposed to weekly contaminations of DENA at 10 ppm.


   f. Cyprinodon exposed to weekly contaminations of BEN at 1 ppm and at 10 ppm.


   g. Ictalurus and Cyprinodon fed 1 ml DENA/100 gm dry food.


   h. Ictalurus and Cyprinodon fed 0.5 gm BEN/100 gm dry food.

                                                                 3
2) Bioaccumulation studies have been conducted in which 50 yl of  H-BaP


   (benzo(a)pyrene) was adsorbed to food pellets fed to Ictalurus. The results


   indicated the highest level of label in the liver ^at 6 hours after ingestion.


   These results, when compared to previously accomplished bioaccumulation


   studies, suggest that fish are more likely to accumulate polycyclic aromatic


   hydrocarbons as a  result of their being adsorbed to ingested materials than


   through direct uptake from the water column.


3) A  dechlcrination   technique has been developed that allows increased  resolution


   of cellular morphology of the early embryonic development of Cyprinodon


   (Moreno, M.S. Thesis, Univ. Southern Mississippi, 1979).  Experiments are


   being conducted utilizing this technique to determine the effects  of  BEN and


   DENA on early embryonic development.
                                           348

-------
4)  Studies with a sheepshead (Archosargus probatocephalus)  cell line have




   established the following levels of acute toxicity:   BaP 2.0 ygm per ml




   media,  BEN 0.2 mgm per ml media, and DENA 2.0 mgm per ml media.   Subacute




   toxicity is evidenced by altered cell morphology and reduced growth rate.




   Vacuolization is a characteristic stress response for SHF^l and  BaP




   elicits this response at lowest concentrations.   After a number  of sub-




   cultivations, SHF-1 cells exposed BaP in the range of 100 ngm to 20 ngm per




   ml media developed multilavered foci.  These foci of cells that  apparently




   lack contact inhibition are about 1 mm in diameter.   Multilayered foci




   have also been observed in one instance of exposure  to BEN.   The foci




   do not  appear in controls.




5)  Classic acute toxicity studies have established  that the LC-50 of BEN for




   Ictalurus is 50 ppm.  Studies are underway to make this same determination




   for BEN with respect to Cyprinodon.




6)  A thorough histologic study of the G.I. tract of Cyprinodon has  been




   completed and is being prepared for publication.  A similar study of the




   blood cell morphology of Cyprinodon is nearing completion.




7)  Preliminary studies indicate a striking elevation in leukocyte (mostly




   eosinophil-like cells) counts in Cyprinodon exposed  to DENA.  When




   Cyprinodon are injected with foreign RBC's and later their splenic blood




   is exposed to the antigenic RBC's, immune rosettes can be observed.  Efforts




   are now underway to use this technique to evaluate the immunocompetence- of




   exposed and unexposed Cyprinodon.
                                  34S

-------
 Coop.  Agreement:             Utilization of Indigenous Populations of Bivalve
                             Mollusks for Monitoring, Surveillance, and
                             Assessing the Public Health Significance of
                             Polycyclic Aromatic Hydrocarbons and other Chemical
                             Carcinogens in Bays and Estuarines

Identification Number:       R806224

Principle Investigator:      Dr. Michael Mix

Project Officer:             Dr. John Couch


1.   Methodology

       We have nearly completed development of a method that has a high degree

of precision for measuring polynuclear aromatic hydrocarbons (PNAH) in environ-

mental samples.  This method is considered by analytical chemists to constitute

 the state-of-the-art for analyzing environmental levels of PNAH in shellfish

tissues and sediments.  We intend to publish this method in Analytical Chemistry.

      Eighteen PNAH, including 12 EPA priority pollutants (*) can be quantified

using reverse phase high pressure liquid chromatography (HPLC)  with UV and

fluorescent detectors; the PNAH include fluorene*, phenanthrene*, fluoranthene*,

pyrene*, benzo(c)phenanthrene, triphenylene, benzo(a)anthracene, chrysene*,

(benzo(j)fluoranthene-benzo(e)pyrene), benzo(b)fluoranthene*, benzo(k)fluoranthene*

dibenz(a,c)anthracene, benzo(a)pyrene*, dibenz(a,h)anthracene*, benzo(g,h,i)-

perylene*, ideno(l,2,3-c,d)pyrene* (Figure 1).

2.   Sampling and Determination of PNAH Body Burdens

        Beginning in October, 1978, bivalve mollusks have been sampled bimonthly

from three Oregon bays.  These include Coos Bay (Mya arenaria and Tresus capax),

Yaquina Bay (Mytilus edulis, M. arenaria and Crassostrea gigas) and Tillamook

Bay (M. edulis, M_. arenaria and £. gigas) .  Table 1 contains data on PNAH

concentrations in samples that have been analyzed to date.  All samples will

be analyzed by September, 1980.

        Tissues from clams (Coos Bay) and mussels (Yaquina Bay) have also
                                    35C

-------
been prepared for histological examination to detect the presence of cellular




proliferative disorders.  Those slides will be examined during the summer (1980)




3.  Trace Metal Analysis




       During 1979, trace metals were measured in mussels using atomic




absorption spectrophotometry (AA) and neutron activation analysis (NAA).




The purposes of the study were to determine:  if there were correlations




between quantities of trace metals and concentrations of PNAH; and if NAA and




AA could be used to supplement HPLC data for identifying or "fingerprinting"




point sources of PNAH.  The data obtained was not sufficient to identify  any




correlations.  However, it was determined that there were correlations between




metal and PNAH body burdens.  Recently, these studies have been extended  to




measure levels of trace metals as a function of season and degree of metabolic




activity (Table 2).  The purpose of this study is to identify those factors




that may influence uptake and retention of various contaminants.  Finally,  we




became interested in measuring two inorganic carcinogens, nickle and arsenic




and determined that AA is suitable for Ni while modified NAA methods can  be




used for As.  We will conduct preliminary studies on As levels in shellfish




this summer.




4.  Other studies




       The following studies are in progress or will be initiated this spring:




determining the existence and prevalence of cellular proliferative disorders




in bivalve mollusks; Pollicipes polymerus preliminary bioassay studies; arid




measurement of PNAH levels in other marine organisms.
                                      351

-------
.METHOC-  c




  INJECT
                  Ifi  41:29     1
-O

C

i

r>



H>
                  1120

                                 LV'X


                                                           •2 r f K

                                                                                           /^\ v\4-V\.r(,M f A <



                                                                                                               C/)
                                                                                                               n>
                                                                                                               o
                                                                                                               D


                                                                                                               "0
                                                                                                               o
                                                                                                               in

-------
ns of PNAH in Mussels (Mytilus edulis) and Oysters (Crassostrea gigas)  from Yaquina Bay, Oregon.
YJ
8 4/20/79
19.5
17.8
6.7
1.7
4.8
4.1
14.0
4.2

1.5

1.7
1.0
*5 0.2
^ 0.4
0.6
Yl
7/2/79
19.0
16.5
6.0
1.7
4.6
4.2
14.2
3.2

1.4

1.6
1.0
0.2
0.4
0.6
Yl
7/30/79
10.7
4.6
1.1
2.6
2.5
2.3
7.7
2.3

0.8

1.0
0.5
0.2
0.3
0.5
Yl
10/8/79
17.5
14.1
6.8
1.5
3-3
3-1
10.4
3-2

1.2

1.3
0.8
0-3
0.4
0.6
Yl
11/14/79
__._
76.4
4.2
2.1
1.3
1.8
3.6
2.5

0.3

0.3
0.3
0.0
0. 1
0.1
Y2
10/2/78
ns
254.9
288.5
61.6
112.1
152.0
161.3
51.8

19.5

39.2
13.6
14.3
ns
3.0
Y2
4/20/79
235.0
211. 2
227.2
48.2
88.9
121. 1
129.2
40.7

15.0

31.3
10.6
10.0
5.4
2.2
Y4
1/24/79
.*-__•-
317-3
247.4
129-9
74.0
95-8
112.0
74.4

13-2

7-8
2.0
1.7
2.7
0.7
Y4
3/21/79
_ — . —
178.0
140.5
125. 1
88.6
179-4
157.8
54.0

21.3

14.7
4.2
3-2
5-2
1.3
Y4
4/20/79
464.9
431.2
234.2
168.2
124.4
189.6
191.9
86.2

24.3

15.3
3.8
3.7
2.4
1.7
Y4
5/31/79
_ _ _ •»
335.8
254.8
138.4
77-0
102.5
119.1
79-1

14.0

8.3
nd
nd
nd
nd

-------
Table 2.  Concentrations of trace metals in soft  tissues  of M.  edulis  in  Yaquina  Bay  at  Station
          Y-l.  Values are in yg/g.
Sample Date




CO
cn

Manganese
Nickel
Copper
Zi nc
Cadmi um
9/10/79
4.5
2.3
15-2
219.1
5.5
10/23/79
5.1
1.6
11.3
125.4
12.0
11/14/79
4.9
0.9
11.0
137.0
12.5
1/14/80
6.6
4.1
12.6
156.3
10.8
2/11/80
7.2
4.5
8.7
177.3
8.1
2/26/80
8.1
6.8
10.0
162. 1
10.8
3/10/80
7.6
5.1
7.7
131.8
8.0
3/24/80
5-9
5-7
9.2
126.4
8.1

-------
Coop. Agreement:             Rainbow Trout:  Model for Carcinogenesis




Identification Number:       R807016




Principle Investigator:      Dr. Jerry Hendricks




Project Officer:             Dr. John Couch








     In compliance with the special conditions required in the Cooperative




Agreement Act of 1977, the following is a quarterly progress report for




the  above project, CR-807016-01-1.




Methods and Materials




     Three compounds  (aroclor 1260, toxaphene, benzo(a)pyrene)? each at 2 dose




levels, were fed to  groups of 80 rainbow trout for 7-9 weeks to determine




maximum tolerated doses and effects on selected liver mixed function oxidase




(MFO) parameters.  Protein levels, cytochrome P-450 content, benzo(a)pyrene




monooxygenase activity, ethoxyresorufin-0-deethylase (EROD) and ethoxycoumarin-




0-deethylase (ECOD)  activities were determined at 1,3,5,6,7 and 9 week




intervals and compared to levels in control fish fed our standard basal diet.




The  groups of fish were started on the following staggered schedule to facil-




itate sampling:  benzo(a)pyrene 2-13-80; aroclor and^toxaphene 2-20-80.  Dietary




levels of 500 and 1,000 ppm were used for each compound.  However, partway




through the short-term feeding trials both dose levels of toxaphene were found




to be toxic for rainbow trout.  Affected fish became anorexic and hyperactive




with a few mortalities occurring.  On 3-7-80 two new groups of fish were started




on diets containing  toxaphene at lower levels of 250 and 125 ppm, respectively.




    Maximum tolerated doses of each compound determined from the short-term




feeding trials were used for initiating a long-term diet study.  On 4-7-80,




6 duplicate lots (12 total) of 100 rainbow trout fingerlings were placed in




perforated buckets (2 per tank) and started on separate diets each containing




one of the three compounds at the following dose levels: benzo(a)pyrene -







                                       355

-------
1,000 ppm;  aroclor 1260 - 500 ppm;  toxaphene - 100 ppm.   Control fish re-




ceived our  standard basal diet.   These diets will be fed for 18 months to




determine if there is a carcinogenic response to any of  the three compounds.




Samples of  20 fish will be removed  from each tank at 6 (10-6-80) and 12




(4-6-81)  months for histopathology  and bioaccumulation assays.   The remaining




60 fish will be killed at 18 months (10-5-81).   Multiple tissue samples will




be taken from all fish.  Other data will include diet consumption,  fish




weight gain and liver weight to  body weight ratios.




    Two additional routes of exposure to benzo(a)pyrene  will be used to deter-




mine its carcinogenicity in rainbow trout.   Fifty gram rainbow trout will be




injected subcutaneously with benzo(a)pyrene, dissolved in propylene glycol




and observed for 12 months; fertile rainbow trout embryos will  be exposed to




a solution  of benzo(a)pyrene and held for 12-18 months.




Results




    Rainbow trout tolerated the  higher dose levels of aroclor and benzo(a)-




pyrene during 7 weeks of feeding.   However, the lower dose level of aroclor




1260 was used for the long-term  feeding trial due to the following criteria:




    1) Experiments in rats have  shown that  lower doses  of 200-300 ppm are




       sufficient carcinogenic levels




    2) Slower weight gain in our fish on the higher dose suggested that a




       lower dose may be more tolerable during an 18 month feeding trial




    3) Other research at our laboratory will involve the use of additional




       aroclor compounds in diets at levels of 500 ppm.   The use of a 500




       ppm  level of aroclor 1260 would facilitate comparison with these




       other results.




    Both the adjusted lower dose levels of  toxaphene were tolerated for at




least 31 days although fish did  not feed as vigorously as control fish and
                                   35G

-------
body weights remained stationary.  Therefore,  a still lower dose level of




100 ppm was chosen for the long-term feeding trial of toxaphene.




    Results of the liver MFO determinations from trout on the short term




feeding trials are presented in Table 1,   Due  to the breakdown of a fluori-




meter used in both deethylase assays, no  3 week samples were taken of trout




fed the benzo(a)pyrene and control diets.   Early in the experiment a basic




modification in the preparation of liver  microsomes used in the 4 assays




invalidated comparison of the 1st week samples of benzo(a)pyrene and




control fish with subsequent samples.  Despite these difficulties sufficient




data was obtained for a valid analysis.  Data  in Table 1 indicate the following:




    1) All 4 liver MFO parameters were higher  in fish fed diets containing




       the 3 compounds than in fish on the control diet




    2) The levels of all MFO parameters appeared to be dose responsive




       except in fish fed the toxaphene diet where levels remained similar




       regardless of dose.




Conclusions




    All three compounds appeared to be inducers of the 4 liver MFO para-




meters that were examined.  Increased induction appeared early (1-3 weeks)




in the experiment with benzo(a)pyrene being more potent as an inducer than




aroclor 1260.  Because toxaphene dose levels were considerably lower, its




induction potential could not be compared to the other two compounds.




However, at the dose levels used, MFO induction in fish fed toxaphene was




considerably lower than in fish receiving aroclor 1260 or benzo(a)pyrene.




Induction of drug metabolizing enzymes by some compounds has been recognized




as an early step towards carcinogenesis.   The  outcome of our long-term




feeding trials will determine if such a correlation can be made with aroclor




1260,  benzo(a)pyrene or toxaphene.
                                    357

-------
                Table 1.   Mean mixed function oxidase values in livers of rainbow
                          trout fed diets containing benzo[ajpyrene,  arochlor 1260
                          and toxaphene.*
  DATE
WEEK
   EROD
nM/mg/min
                                   **
500 ppm BENZ[A]PYRENE
  2-20-80
  3-05-80
  3-19-80
  4-02-80
  4-09-80
  1
  3
  5
  7
  9
1000 ppm BENZ[A]PYRENE
  2-20-80
  3-05-80
  3-19-80
  4-02-80
  4-09-80
  3
  5
  7
  9
500 ppm AROCHLOR 1260
  2-27-80
  3-12-80
  3-26-80
  4-09-80
  1
  3
  5
  7
  7.752 ±5.179
  4.6033±0.7859
  3.2936±0.6058
                       /. 0.3336±0.1665
                       '•  0.3293±0.3258
                       :  1.8468±1.416
                       '  1.6732±0.6527
1000 ppm AROCHLOR 1260

  2-27-80        1
  3-12-80        3
  3-26-80        5
  4-09-80        7

125 ppm TOXAPHENE

  3-12-80        1
  3-26-80        3
  4-09-80        5
  4-16-80        6
                         0.5567+0.070
                         1.4143+0.4521
                         4.2619+1.0117
                         1 .4356+0.8846
                       /• 0.0395+0.015
                         0.1058 ±0.0197
                      ECOD
                    nM/mg/min
                     AHH
                  nM/mg/min
 i  2.007 ±0.361
 •  1.5310±0.5766
 ,;0.7010±0.293
                                         0.2575±0.0219
                                         0.2431+0.0744
                                       V 0.0767±0.0104
                0.3744+0.0110
                0.2902+0.1496
                0.2872+0. 0034
                                       .  0.6232±0.1039  /i. 0.5751 ±0.1 018
                                       ;  0.5760±0.2613  '0.7169+0.1080
                                       '!<  0.2038±0.0255  ! 0.8670±0.0492
             /'• 0.0675 ±0.027
               0.0853±0.0081
             :  0.1724+0.1025
             '  0.2427±0.0529
0.1157±0.028
0.2080±0.0374
0.3610±0.1637
0.2671+0.1745
                          0.0498±0.0132
                          0.0572±0.0028
250 ppm TQXAPHENE
  3-12-80
  3-26-80
  4-09-80
  4-16-80

CONTROL
  1
  3
  5
  6
                       •  0.1850+0.0111   0.0354+0.0054
/• 0.0461 +0.0200   0.0470±0.0112
 ; 0.1193+0.0935   0.0603+0.0135
 I
 • 0.1654+0.0168   0.0293±0.0101
2-20-80
3-05-80
3-19-80
4-02-80
4-16-80
1
3
5
7
9
                        0.0457iO.0008
                        0.0199±0.0028
                        0.0456±0.0031
                         0.0412+0.008
                         0.0249±0.003
                         0.0384±0.0145
 * ± = standard deviation
** nano moles per mg of protein per minute
                                  0.0222+0.0129
                                  0.0852+0.0030
                                  0.2689±0.1536
                                  0.1271±0.0544
                                          0.0496+0.0279
                                          0.2678±0.0627
                                          0.5242±0.0252
                                          0.3723±0.0454
                                  0.0276+0.0390
                                  0.0255±0.0093
                                  0.0281±0.0032
                                  0.0945±0.0100
                     0
                0.0272±0.0002
                0.0680±0.0159
                0.0642+0.0003
                                  0.0106±0.0149
                                  0.0096±0.0029
                                  0.0141+0.0057
                                                                             P-450
                                                                             nM/mg
                                               0.5307+0.023
                                               0.5595+0.0632
                                               0.4037+0.0349
                                0.5501+0.008
                                0.7829±0.0229
                                0.5349+0.0345
                                0.5406+0.083
                                0.3753±0.0483
                                0.3769±0.0127
                                                  0.6678±0.078
                                                  0.5035±0.009
                                                  0.3326+0.0291
                                0.6276i0.042
                                0.4008iO.0473

                                0.4168+0.0265
                                                                        0.6510+0.044
                                                                        0.4058+0.0122

                                                                        0.3901+0.0097
                                0.4367+0.015
                                0.3640+0.0286
                                0.4693+0.0082
                                        358

-------
GOOD. Agreement:            Oxidation and Conjugation of Carcinogen
                            Hydrocarbons in Marine Animals

Identification Number:      R806368

Principle Investigator:     Dr. D.R. Strength

Project Officer:            Dr. W.P. Schoor


     The oxidative metabolism of polycyclic aromatic hydrocarbons (PAH)

results in products of high cellular toxicity or mutagenicity.  The types

of metabolites identified in several other laboratories include  phenols,

dihydrodiols and quinones.  In addition, several secondary metabolites

of PAH compounds have been identified: these include glucuronide, sulfate

sulfhydryl and DNA conjugates.  The primary purpose of this project is to

identify the types of oxidation products produced from selected PAH

compounds in certain fish and shellfish.  The extent and types of

conjugation products are under study and the enzyme systems involved in the

conjugation process are under study to ascertain the effects of exposure

of the organisms to PAH upon their metabolism.  The results of this investi-

gation indicate that the glucuronosyltransferase and sulfotransferase

enzymes are induced by exposure of rats to PAH compounds by feeding or

injecting phenobarbital, 3-methylcholanthrene, phenanthrene and benzo(a)pyrene

Exposure time required for induction was between 8 and 12 days; requirements

for the induction of the enzymes involved in the conjugation process suggested

that metabolites (possibly oxidation products of the PAH compounds) may be

the actual inducers of the enzymes catalyzing the conjugation reactions.

Phenobarbital, 3-methylcholanthrene, phenanthrene and benzo(a)pyrene were

inducers of the transferases and are listed in the order of most effective

to least effective as inducer.

     The enzyme systems, UDP-glucuronosyl transferase, sulfate transferase,

UDPG-dehydrogenase, g-glucuronidase and arylsulfatase were each studied
                                     35

-------
in tissues of rat,  mullet, oyster,  clam and mussel.   Attempts to demonstrate




induction of the transferases in the marine animals involved mandatory limita-




tions to exposure time and dosage.   Generally UDPG-dehydrogenase, arylsulfatase




and 3-glucuronidase were not observed to be induced by PAH compounds.  In




marine animals, unequivocal induction of the transferases has not been demon-




strated.  Modifications in exposure time and dosage in channel catfish and




Fundulus are currently in progress  to enhance the conditions favorable to




induction.  Additional studies are  in progress to develop conditions for the




maintenance of cells from several species in order to expose the cells directly




to the PAH compounds and their metabolites.  Liver cells from rat livers and




channel catfish livers have been successfully maintained for periods sufficient




to effect exposure and observe effects.




     3-Methylcholanthrene administered to mullet as a single dose or fed to




rats has a profound effect upon the mass and gross appearance of the liver.




The mass expressed as g of liver/lOOg of body of animal, both rat and mullet




is significantly greater in treated animals than in their respective controls.




     Oxidation products of phenanthrene and benzo(a)pyrene produced by




incubation of the compounds with microsomes of rat liver and microsomes



                                                            35
of channel catfish liver, respectively form conjugates with   SO^ that can




be demonstrated by separation on thin layer chromatograms.  Arylsulfatase




liberates the radioactive sulfate from the bound, mobile radioactive fraction




recovered from the thin layer chromatograms. Indications are that hydroxyaryl




metabolites of phenanthrene and benzo(a)pyrene are conjugated with sulfate




by the enzyme of the cytosol or rat liver.




     Work planned for the future includes the identification of the compounds




that conjugate with sulfate and glucuronic acid.  Improvements in regulation




of treatment time and dose is essential to future studies with marine animals.
                                       3GC

-------
Feeding  of the compounds or repeated injections will be required to simulate




the conditions found  to be the most effective for induction  of  the transferase




enzymes in rats.
                                      361

-------
Coop. Agreement:             Metabolism of Polyaromatic Hydrocarbons by
                             Mixed Function Oxidases of Marine Organisms

Identification Number:       CR806213020

Principle Investigator:      Dr. Paul Melius

Project Officer:             Dr. W.P. Schoor

    In the first year of this project we investigated the induction of the

mixed function oxidase enzyme system in gulf mullet (Mugil cephalus) and

the Killifish (Fundulus grandis).   Aroclor 1254 and 3-methylcholanthrene were

used as the inducing agents and NADPH-cytochrome P^Q reductase and cyto

chrome P^Q were measured in the fish livers.  No significant changes were

observed.  In in vitro studies, using microsome preparations from the mullet

and killifish it was found that benzo(a)pyrene (B(a)P) was oxidized to a

variety of metabolites comparable to those found when rat microsomes are

used.  The microsomes from the livers of induced fish gave positive results

in the Ames Salmonella test for mutagenicity.  This work was reported at the

"4th International Symposium on Polynuclear Aromatic Hydrocarbons" held at

Ohio State University in 1979.

    We reported on the metabolite patterns of B(a)P in mullet liver microsome

preparations compared to the rat at the ASTM 4th Symposium on Aquatic Toxicol-

ogy held in Chicago, October 16, 17, 1979.  The total metabolites of B(a)P

in the mullet were formed at about one fifth the amount as the rat under the

in vitro conditions we used.  We were able to detect increases in the

9,10-diol B(a)P, 4,5-diol-B(a)P, 7,8-diol-B(a)P, quinones, 9-"OH^B(a)P arid

3-OH-B(a)P.

   More recently we have initiated studies with the Tilapia aurea. The

inducing agents were trans-stilbene oxide (TSO), Aroclor 1254, g-naptho-

flavone (BNF) and 3-methylcholanthrene (3-MC)-  Cytochrome P45Q and b^,

NADPH-cytochrome PA.-Q and NADH cytochrome be reductases, glutamate-pyruvate

-------
and glutamate-oxaloacetate transaminases, N-demethylase and cholinesterase




were assayed in control-and induced fish.  Enzyme induction decreases in the




following order:  Aroclor 1254 (200 mg/kg) > ISO (200 mg/kg), BNF (40 mg/kg) >




3 MC (20 mg/kg).  Male fish had greater induction capacity than the female




fish.  The results of these experiments have been submitted to Biochemical




Pharmacology and Toxicology.




    We have developed a new analytical procedure to assay for epoxide hydrase




by determining 1,2-diols using lead periodate and atomic absorption spectro-




photometry.  Epoxide hydrase is one of the enzymes of the mixed function




oxidase system.  This work has been reported in Analytical Chemistry 52, 602,




1980.




    An extensive study of the kinetic product patterns and mechanistic path-




ways of B(a)P metabolism in Aroclor-treated mullet has been made and a man-




uscript has been submitted to Biochemical Journal.  The aryl hydrocarbon




hydorxylase activity has been found to be constant for all metabolites studied




except 3-OH-B(a)P and B(a)P-9, 10-diol which increased non-linearly to 22-fold




and 17 fold at higher concentration, respectively.  Incubation temperature




greatly altered metabolite patterns; phenols and diols required longer reac-




tion times at lower temperature to obtain optimum activity, quinones changed




from a steady state at 25°C to optimated kinetics at 37°C, triols and tetrols




were relatively unchanged with temperature.  A steady state intermediate




radical, 6-oxobenzo(a)pyrene was postulated for mediating the enzymatic for-




mation of the quinones as has been previously suggested by Lesko et al (1975).

-------
Coop.  Agreement:



Identification Number:

Principle Investigator:

Project Officer:
GC-MS Analysis of Potential Carcino-
genic Organic Pollutants in Aquatic
Organisms and Sediments

CR807160011

Dr. John Laseter

Dr. John Couch
     This analytical study was undertaken as part of the Pathobiology

Section of the EPA/NCI Collaborative Carcinogens in the Aquatic Environment

Program. The primary objective of the study is to obtain quantitative

measurements of specific organic pollutants with carcinogenic potential in

sediment and biota samples collected in three Gulf Coast bays and in an off-

shore comparison site. As originally conceived,  the program consisted of two

phases. In Phase I,  replicate analyses of eight  specific organic carcinogens

would be carried out in sediment, oyster tissue, and fish tissue. These data

would then be used to estimate the number of samples required to achieve the

desired analytical precision in Phase II of the  program. A qualitative screen-

ing of the three matrices collected at each bay  and the offshore site for 112

of the EPA organic unambiguous priority pollutants, and any other potentially

carcinogenic organic pollutant detected, would also be conducted in Phase I.

Phase II of the program would be devoted to monitoring the carcinogenic com-

pounds detected during Phase I in samples collected periodically throughout a

one year period.

     The results of  our initial analyses indicated that the samples were too

complex to analyze for the eight specific carcinogens by gas chromatography

without the use of specific detectors. This restriction dictated that the

emphasis of Phase I  be shifted away from replicate analyses of the eight

selected carcinogens, towards the screening of the samples for potential organic

carcinogens. The screening, which was to be done by combined gas chromatography
                                      364

-------
and mass spectrometry ( GC-MS ),  was upgraded from a qualitative analysis to




a qualitative/semi-quantitative analysis.  The screening exercise was constructed




around but not limited to analysis of the  organic priority pollutants.  That is,




volatile (purgable), base-neutral extractable, acid extractable, and pesticide




fractions were collected from representative samples of each of the three




matrices from the three bays and  the offshore comparison site.  Quantitative




analytical procedures were available for pesticides in sediments and in tissues,




but not for the other (semi-volatile) fractions.  This necessitated a period of




development prior to the beginning of analyses of the volatile base-neutral




extractable, and acid extractable organics.




      The pesticide analyses,which consisted of qualitative and quantitative




GC analyses, are in their final confirmation stage. Thirty-three pesticide




fractions from eleven representative samples have been analyzed by GC-EC




(electron capture detection). Polychloridated biphenyls (PCB) were not detect-




ed in any of the samples analyzed. Of the 28 priority pollutant pesticides and




PCB's searched for, B-BHC, heptachlor, DDE, and trifluralin (not a priority




pollutant) were identified by GC-EC in the samples treated. Several unidenti-




fied halogenated compounds were also detected and will be analyzed by GC-MS




during the confirmation phase of  the pesticide analyses.




      Two computerized mass spectral search procedures were sequenced on the




GC-MS data system in such a manner that specific priority pollutant organics




were identified and quantitated and then all peaks above threshold were searched




against the NIH-EPA mass spectral library and quantitated, if successfully




identified. These two search routines were used in the analysis of the volatile,




the base-neutral extractable, and the acid extractable organic fractions. The




total processing time for a complete detailed mass spectral search using both




search procedures ranged from 6 to 12 hours and yielded data on from 50 to 200






                                       36E

-------
or more peaks per run.




      Because of the problems with emulsions,  it  was necessary to use two




separate aliquots of each sample for extraction of the base-neutral and acid




organics. To improve recovery, acid-neutral rather than just acid extractable




organics were extracted from the sediment samples. Neutral compounds of bio-




genie origin such as fatty acid methyl esters,.normal alkanes, and fatty acid




ethyl esters dominated  the analyses of the base-neutral and the acid-neutral




extracts. The acid extractable fractions contained free fatty acids as well as




fatty acid methyl esters (probably due to overloading the system) of biogenic




origin. The sediment extracts contained low molecular weight hydrocarbons,  pro-




bably of petroleum origin, as well as several  polynuclear aromatic hydrocarbons




apparently derived from pyrolytic sources. Without counting extraction, fraction-




ation, or actual instrument analytical time, approximately 150 hours of data




treatment time have been consumed on these analyses.




      Volatile organics from sediments, oyster tissue, and fish tissue were




analyzed using a dynamic headspace technique which involved a total purging,




trapping and drying time of six hours per sample. These data will be processed




using the sequenced mass spectral search procedures described above.




      Upon completion of the screening portion of this program, specific organic




carcinogenic pollutants will be selected for analysis during Phase II. The




carcinogens selected for quantification will be analyzed in replicate samples




collected quarterly from the four bays and the offshore comparison site.
                                        36C

-------
             XII.PUBLICATIONS  RESULTING FROM PROGRAM




Bunting, D.B. 1979.  An Evaluation of Benzo(a)pyrene Metabolism in an




    Oyster (Ostrea edulis)-Bacteria System.  M.S.




Coker, S. 1980. Benzo(a)pyrene Metabolism in the Sea Catfish (Arius felius).




    Ala. Acad. Sci. Meeting, March 21, 1980.




Couch, John A. 1979.  Carcinogens in the Aquatic Environment.   Interagency




    Collaborative Group on Environment Carcinogenesis, NIH, Bethesda,




    Maryland, January 10, 1979.




Couch, J.A. 1979.   The American Oyster (Crassostrea virginica)  as an




    indicator of carcinogens in the Aquatic Environment.  In,  Animal




    Models and Wildlife as Monitors, 65-84.   National Academy of Sciences,




    Washington, D.C.




Couch, J.A. 1979.   Vertebral dysplasia in young fish exposed to the herbicide




    Trifluralin.  Journal of Fish Diseases,  2:35-42.




Couch, J.A. 1979.   Pollution Ecology of Penaeid Shrimps. In, Pollution




    Ecology of Estuarine Invertebrates, 236-258.  Hart/Fuller (eds.) Academic




    Press, N.Y.




Elam, D., M.V. Kilgore, B. Tan, and W.P. Schoor. 1979.  Mixed Function Oxidase




    Inducibility in the Mullet and Killy Fish.  4th Intl. Symp. on PAH,




    Columbus, Ohio, Oct. 4, 1979.




Elam, D., M.V. Kilgore and W.P- Schoor. 1979. Induction of Mixed Function




    Oxidase and Metabolism of Polyaromatic Hydrocarbons in Marine Organisms.




    llth Intl. Congress of Biochemistry, Toronto, Canada, July 11, 1979.




Gregory, P.E., P.N. howard-Peebles, R.D. Ellender andB.J. Martin. 1980.




    Banding of chromosomes from three established marine fish cell lines.




    Copeia (accepted).




Gregory, P.E., P.N. Howard-Peebles, R.D. Ellender and B.J. Martin. 1980.




    Analysis of a marine fish cell line from sheepshead:  chromosomal altera-




    tions in Archosargus probatocephalus. J. of Heredity (accepted).







                                        367

-------
Harshbarger, John C.  and John A. Couch. 1978.  Symposium on Neoplasms in




    Invertebrates:  Highlights and Reflections.  XI Ann. Meet. Soc. Invert.




    Path. Sept., Prague, Czech.




Hemingway, S.J. 1979. (1) Storage Sites of Benzo(a)pyrene in contaminated




    Bay Mussels (Mytilus edulis) inhabiting different sites in Yaquina




    Bay, Oregon.  (2) Uptake and depuration of Benzo(a)pyrene in contaminated




    Bay Mussels (Mytilus edulis) inhabiting different sites on Yaquina Bay. M.S




Hendricks, J.D., R.O. Sinnhuber, J.E.  Nixon, J.H. Wales, M.S. Masri, and




    D.P.H. Hsieh (1980).  Carcinogenic response of rainbow trout (Salmo




    gairdneri) to aflatoxin Q-^ and synergistic effect of cyclopropenoid fatty




    acids.  J. Natl.  Cancer Inst. 64:523-528.




Hendricks, J.D., R.O. Sinnhuber, J.H.  Wales, M.E. Stack, and D.P.H. Hsieh




    1980.  The hepatocarcinogenicity of sterigmatocystin and versicolorin




    A to rainbow trout embryos.  J.  Natl.  Cancer Inst. 64:000-000 (June issue).




Hendricks, J.D., R.A. Scanlan, J.L.  Williams, R.O. Sinnhuber, and M.P. Grieco




    1980.  The carcinogenicity of N-methyl-N'-nitro-N-nitrosoguanidine to




    the livers and kidneys of rainbow trout (Salmo gairdneri) exposed as




    embryos.  J. Natl. Cancer Inst.  64:000-000 (June issue).




Hendricks, J.D., T.P. Putnam, and R.O. Sinnhuber. 1980.   Null effect of




    dietry Aroclor 1254 on hepatocellular  carcinoma incidence in rainbow




    trout (Salmo gairdneri) exposed  to aflatoxin B^ as embryos. J. Environ.




    Pathol.  Toxicol.  (In press).




Handricks, J.D., R.O. Sinnhuber, M.  Henderson, and D.R.  Buhler. 1980.  Liver




    and kidney pathology in rainbow trout  (Salmo gairdneri) exposed to dietary




    pyrrolizidine (Senecio) alkaloids.  (to be submitted to Am. J. Pathol.)

-------
Hendricks, J.D., W.T. Stott, T.P. Putnam,  and R.O.  Sinnhuber„  1980.  En-




    hancement of aflatoxin B^ hepatocarcinogenesis  in rainbow trout (Salmo




    gairdneri) embryos by prior exposure of gravid  females to dietary Aroclor




    1254.  Proc. 4th Ann. Symp. on Aquatic Toxicol., ASTM (In press).




Hendricks,  J.D., J.H. Wales, R.O. Sinnhuber, J.E.  Nixon, P.M. Loveland,




    and R.A. Scanlan. 1980.  Rainbow trout (Salmo gairdneri) embryos:   A




    sensitive animal model for experimental carcinogenesis.   Fed. Proc.




    (In press).




Hendricks, J.D., R.O. Sinnhuber, P.M. Loveland, N.E. Pawlowski, and J.E.




    Nixon. 1980.  Hepatocarcinogenicity of glandless cottonseeds and refined




    cottonseed oil to rainbow trout (Salmo gairdneri).   Science (In press).




Kilgore and D. Elam. 1979.  Mixed-Function Oxidase  Inducibility in Marine




    Organisms. Ala. Acad. of Sciences Meeting, April 14, Florence, Ala.




Kilgore, M.V., D. Elam. 1979.  A comparative study  of induction of mixed




    function oxidase activity in the rat,  mullet and killifish.  Ala.




    Acad. Sci. Meeting, March 30, 1979.




Long, R.L. and B.J. Martin.  1980.  Morphology of peripheral blood cells of




    Cyprinodon variegatus. ABSTR. J. Miss. Acad. Sci.,  Vol.  XXV, Suppl:122.




Martin, B.J. 1980.  Effects of petroleum compounds  on estuarine fishes.




    Ecol. Res. Series. EPA-600/3-80-019, Jan. 1980.




Martin, B.J. and W.W. Greenwich.  1980.  Exposure of two telost species to




    polycyclic aromatic hydrocarbons. ABSTR. J. Miss. Acad.  Sci., Vol. XXV,




    Suppl:87.




Melius, P., B. Tan, M. Kilgore, D. Elam, W.P. Schoor. 1979.   Product pattern




    and kinetic  analysis of benzo(a)pyrene metabolism in Marine Organism by




    HPLC.  ACS 31st Southeastern Regional Meeting,  Florence, Ala. April 14, 1979,
                                    36:

-------
Melius, P., B.  Tan,  M.  Kilgore,  D.  Elam.  1979.  Metabolites of B(a)P in




    Aroclor induced  fish.   Fourth ASTM Symposium on Aquatic Toxicology,




    Chicago, Illinois.  Oct. 16,  17.




Melius, P., D.  Elam, M. Kilgore, B. Tan and W.P. Schoor.   1979.   Mixed




    Function oxidase inducibility and polyaromatic hydrocarbon metabolism




    in mullet,  sea catfish, and  gulf killifish.  Fourth International




    Symposium on Polynuclear Aromatic Hydrocarbons, Columbus, Ohio.




Mix, M.C., R.T. Riley,  K.I. King, S.R.  Trenholm, and R.L.  Schaffer. 1977.




    Chemical carcinogens in the  marine environment. Benzo(a)pyrene in




    economically important bivalve mollusks from Oregon estuaries. In:




    Fate and Effects of Petroleum Hydrocarbons  in Marine  Organisms and




    Ecosystems, D.A. Wolfe (ed), 421-431.  Pergamon Press,  N.Y.




Mix, M.C. and R.L. Schaffer. 1979.   Benzo(a)pyrene concentrations in mussels




    (Mytilus edulis) from Yaquina Bay,  Oregon during June, 1976-June, 1978.




    Bulletin of Environmental Contamination and Toxicology, 23:67—684.




Mix, M.C., J.W. Hawkes  and A.K.  Sparks.  1979.   Observations on  the ultra-




    structure of large  cells associated with putative neoplastic disorders




    of mussels, Mytilus edulis,  from Yaquina Bay, Oregon.  Journal of Invert-




    ebrate Pathology, 34:41-56.




Mix, M.C. 1979.  Chemical Carcinogens in Bivalve Mollusks from Oregon Estuaries.




    EPA Ecological Research Series, EPA-600/3-79-034. 33.




Mix, M.C., D.L. Bunting and D.T. Abbott.  1979.   Preliminary "studies to evaluate




    the potential of using embryo and larval stages of the goose barnacle,




    Pollicipes  polymerus,  for marine bioassays.  Proceedings of  the Second




    Biennial Crustacean Health Workshop.  Texas  A&M Univ., TAMU-SG-79-114:361-381




    College Station, Texas.
                                     37C

-------
Mix, M.C., S.R. Trenholm, and K.I. King.  1979,   Benzo(a)pyrene body burdens




    and the prevalence of cellular proliferatiye disorders in mussels,




    Mytilus edulis, from Yaquina Bay. Oregon. In, Animals as Monitors of




    Environmental Pollutants, 52-62.  National Academy of Sciences,




    Washington, D.C.




Mix, M.C. and W.P. Breese.  A cellular proliferative disorder in Ostrea,




    chilerisis from Chilor, Chile, South America.  Submitted to Journal




    of Invertebrate Pathology, In Press.




Mix, M.C. Cellular proliferative disorders in bivalve mollusks from




    contaminated marine environments.  To be submitted to Journal of




    Environmental Pathology and Toxicology.




Moreno, M. and B.J. Martin. 1979.  Embryologic development of sheepshead




    minnow,  (Cyprinodon variegatu_s). ABSTR. J. Miss. Acad. Sci. Vol. XXIV,




    Suppl.:116.




Porter, R.D. and B.J. Martin. 1980.  The histology of the post pharyngeal




    digestive tract of the sheepshead minnow, Cyprinodon variegatus. ABSTR.




    J. Miss. Acad.  Sci., Vol. XXV,  Suppl:123.




Riley. R.T.  and M.C.-Mix. 1980.  Thin layer separation of citric and cycle




    intermediates,  lactic acid, and  the amino acid taurine.  Journal of




    Chromatography, 189:286-288.




Riley, R.T.  and M.C. Mix.  An ion-exchange technique for the determination




    of ammonia in small volumes of  seawater.  Submitted to Marine Chemistry.




Riley, R.T., M.C. Mix, and D.L. Bunting.  The uptake, accumulation, and




    metabolism of naphthalene by  the oyster, Ostrea edulis;!. Design of the




    dosing system,  and water quality.  Submitted  to Marine Biology.




Riley, R.T., R.L. Schaffer and M.C. Mix.  The uptake, accumulation  and




    metabolism of napththalene by  the oyster, Ostrea edulis:II.  Submitted




    to Marine Biology.






                                      37J

-------
Riley,  R.T.   Stimulatory effect of naphthalene on blucose transport in the


    oyster.   Submitted to Nature.


Riley,  R.T.  and M.C.  Mis.  The pathways of glucose metabolism in the oyster,


    Ostrea edulis.  To be submitted to Comparative Biochemistry and Physiology.


Riley,  R.T.  197S.  The effects of  chemical perturbation by naphthalene in


    glucose metabolism in the European Flay oyster, Ostrea edulis: An


    in  vivo kinetic analysis. Ph.D.


Schaffer,  R.L.  and  M.C. Mix.   A method for determining environmental levels


    of  polycyclic aromatic hydrocarbons in bivalve mollusks by reverse


    phase liquid chromatography.   To be submitted to Analytical Chemistry.


Schneider, S.R., J.D. Hendricks,  G.H. Constantine, and R,E. Larson. 1980.


    Tobramycin  nephrotoxidity in coho salmon at subtherapeutic doses.


    Toxicol. Appl.  Pharmacol. (In  press).


Schoor, W.P- and J.A. Couch.   1979.  Correlation of mixed-function oxidase


    activity with ultrastructural  changes  in the liver of a marine fish.


    Cancer Biochemistry and BioPhysics, 4:95-103.


Schoor, W.P- 1979.   Fluorimetric confirmation of metabolites of benzo(a)


    pyrene from a marine fish.  Fourth International Symposium on Polynuclear


    Aromatic Hydrocarbons, Columbus, Ohio, Oct. 2-4, Battelle Institute.


Smith,  A.C.  and M.C.  Mix.  1978.   The effects of sodium chloride concentration


    on  electrophoretic patterns of adductor muscle proteins from bivalve


    molluscs.  Comparative Biochemistry and Physiology (part B), 61:169-171.


Strength,  H.H.  Daron, J.L. Aull,  J.F. Wilson and W.P- Schoor. 1980.  The


    induction of glucuronide and sulfate transferases by phenobarbital and


    polycyclic  aromatic hydrocarbons.  Fed. Proc.  (In press).
                                    Si-, r
                                     ( £

-------
Tan, B. and M.V- Kilgore. 1980.  Determination of l,2r-Diols by direct




    atomic absorption with digested lead periodate.  Anal, Chem. 52:602.




Tan, B., D. Elam, M.V. Kilgore and WfP. Schoor.  Mixed function oxidase




    inducibility and polyaromatic hydrocarbon metabolism in mullet, sea




    catfish and gulf killifish.  4th Int. Symp. on Polynuclear Aromatic




    Hydrocarbons Chemistry and Biological Effects.  Battelle Press (In press).




Tan, B.  Responses of hepatic enzymes of tilapia to biphenyls and poly-




    aromatic hydrocarbons.  Submitted to Biochem. Pharmacology.




Tan, B.  Kinetic product patterns and mechanistic pathways of benzo(a)pyrene




    metabolism in Aroclor-treated mullet. Submitted to Biochemistry




    Journal.




Tan, B.  Comparative metabolism of polyaromatic hydrocarbons in marine




    organisms.  Invited review article being prepared for Journal of




    Comparative Biochemistry and Physiology.




Tan, B. 1979.  Product Pattern and kinetic analysis of benzo(a)pyrene




    metabolism in marine organisms by HPLC. S.E. Regional A.C.S. meeting




    Roanoke, Va. Oct. 25, 1979.




Tan, B., D. Elam, M.V. Kilgore and W.P. Schoor. 1979.  Metabolites of Benzo(a)




    pyrene in Aroclor 1254 treated mullet.  4th ASTM Symp. on Aquatic




    Toxicology.  Oct. 16-17, 1979f Chicago, 111.




Tan, B., 1980.'-:Indirect atomic absorption assay of epoxide hydrase. Ala.




    Acad. Sci. March 21.




Tan, B., P. Melius, M. Kilgore, D. Elam and W.P. Schoor.  1979.  Metabolites




    of benzo(a)pyrene in Aroclor-Induced fish.  XI International Congress




    of Biochemistry, Toronto, Canada, July 8-13.




Tan, B., M.V. Kilgore, D. Elam and W.P. Schoor.  Metabolites of benzo(a)pyrene




    in Aroclor 1254 treated mullet.  4th Symp. on Aquatic Toxicology, ASTM




    (In press).

-------
Tan, B. 1980.  Responses of hepatic enzymes of Tilapia to PCB and PAH




    Modifiers.  To be presented June 1, 1980,  American Assoc. Biol.




    Chemists Meeting, New Orelans.




Tan, B. 1980.  Product pattern, kinetic analysis and pathways for benzo(a)




    pyrene metabolism in marine organisms.   To be presented July 20-26,




    English Society of Analytical Chemist,  5th Intl. Symp., Lancaster,  England.




Tan, B.. 1980.  Indirect atomic absorption as'say of expoxide hydrase




    activity and determination of 1,2-Diols via digested lead periodate.




    To be presented July 20-26, English Society of Analytical Chemists,




    5th Intl. Symp., Lancaster, England.




Tan, B. and J. Grizzle. 1980.  The use of benzo(a)pyrene metabolism as  a




    biochemical indicator in control and tumored bullhead catfish.   To  be




    presented Oct. 28-30, 5th Intl. Symp. on Polynuclear Atomatic Hydrocarbons,




    Columbus, Ohio.




Tan, B. 1980.  Benzo(a)pyrene Metabolism in PAH and Biphenyl treated Tilapia.




    To be presented Oct. 28-30, 5th Intl. Symp on PAHs, Columbus, Ohio.




Trenholm, S. and M.C. Mix.  The lethal and  sublethal effects of ionizing




    radiation on juvenila Pacific oysters (Crassostrea gigas).  Radiation




    Research. (In press).




Trenholm, S.R. and M.C. Mix. 1978.   Regeneration of radiation damaged diges-




    tive tissue in juvenile Pacific oyster  (Crassostrea gigas).  Journal




    of Invertebrate Pathology 32:249-257,




Trenholm, S.R. 1978.  Effects of X- and Gamma irradation on the juvenile




    Pacific oyster, Crassostrea gigas. M.S.




Voss,  Sherri, and Jerry Hendricks.  1980.  Induction of hepatic microsomal




    enzymes in rainbow trout by dietary aroclor 1254 and the effect of




    cyclopropene fatty acis.  To be submitted to Archives Environ. Contamin.




    Toxicol.
                                 374

-------
                                  Discussion
Dr. Cooper, NCI:  In one of your early slides you showed us a very nice schematic
of the equipment you use to expose marine organisms to environmental hazards,
and at the  bottom of that slide was a little arrow that said, "Effluent line."  What
is on the other end?

Dr. Couch, EPA:  That question invariably comes up. Anyway, we have an effluent
pond next to the laboratory.  It is  an evaporation pond. It receives the effluent
from our test system.  So, effluent does not get out into the natural system bay, we
monitor this periodically, and  we have wells around the pond to test for seepage.
We use low concentrations (micrograms/liter) and in our test system that I showed
you, there is a very low flow rate. The fish we selected are good because they can
live  in a  very low flow-through of water; the turnover of  the water and the
concentration is very small in  that regard.  If you used larger  fish, you would  have
to use  a greater flow rate.

Dr. Morris, EPA:  Yes.

Dr. Baumel, NIOSH:    I am  not that familiar  with the  dynamics  and all the
pharmacokinetics of the metabolism of aquatic organisms.   It seems  to me you
have  an interesting problem  of re-uptake of  excreted metabolites.   You  have
somewhat a potential for some steady state or diffusion dynamics that go on that
don't  normally  happen,  say,  in a  normal  environmental  system  with  airborne
concentrations.

Have you looked at any of this? Have you looked at the formation of metabolites
or re-uptake possibilities as factors affecting the outcome of your experiment?

Dr. Couch, EPA:  In a flow-through system, when we expose the animals, much  of
the excreted metabolites would be  removed  by  the flowing  water turning in the
tanks. There may be some re-uptake of excreted metabolites by the test animals.
We have done no work on this specifically yet.

We have looked at the disappearance and persistence of the parent compound in the
water  column as I showed you.  Some of the disappearance is  due to adsorption  to
the glass of the aquarium or the system  itself.   For example, if we put oysters in
the system and expose  them  to benzopyrene,  a certain amount of the finite
concentration that we put in will be taken up by the oyster, and an equal amount  or
an even larger amount may be taken up, adsorbed by the glass of the system. So,
we do have adsorption  and  desorption as a problem.  Dr. Schoor has done some
solubility studies and  fate  studies in these systems to try to  keep track of the
compound,  but we have not  addressed what you mentioned specifically in terms  of
metabolites being taken up again. Dr. Schoor can talk with you about that in more
detail and tell you what his opinion is on that.

Dr. Hegyeli, NCI:  Bivalves  are excellent bioconcentrators, but at the same  time
they flush  out the toxic materials  or corpuscular elements very easily.  So, my
question is  whether this was considered when you were collecting the samples and
keeping them in flowing fresh water, and the other idea is about the resistance  of
bivalves to  carcinogenic substances.

-------
Dr. Couch,  EPA:   That is  a very complex set  of questions.  We do know that
oysters, for example, which we are very familiar with in our work, will accumulate
two different toxicants or carcinogens at different rates, depending on the kind of
chemical  to  which you are exposing  them.  For example,  with some of  the
pesticides such as DDT and some of the PCBs in earlier work at our laboratory, we
found  that  oysters  would concentrate  up to several  hundred thousands  times
ambient, but when we  exposed oysters to benzopyrene at environmentally realistic
levels of four parts per billion (benzopyrene in a flow-through  system), we  found
that oysters only accumulated 200 to 240 times ambient.  So, it depends on the kind
of chemical to which you are exposing them.  The natural cellular and functional
excretory mechanisms, secretory mechanisms of  the animals involved, in bivalves,
oysters for example, involve a phenomenon that vertebrates do not have.  If they
take up  macromolecules or  molecular  substances or even  larger than  macro-
molecular substances into their system, their leukocytes may pinocytose  or phago-
cytize  these substances and  actually crawl  across  the  epithelial layers, throw
themselves out into the mantle  cavity, and rid themselves of large quantities  of
these  substances.   Therefore, uptake  and retention  of  different substances by
bivalves depends entirely upon the substance and the bivalve involved.

In answer to the second part of your question, the bivalves have not proven to be as
good an assay organism for  car ci no genesis  in the laboratory as  have fish.  So, we
are concentrating  more on  fish  for  many different reasons.   There  are  several
interesting neoplastic diseases of bivalve mollusks that occur in nature, including
both carcinoma-like disorders and sarcoma-like disorders.  We have read a report
from Russia recently that Kudolay has been able to induce a leukemia  condition in
freshwater mussels with chemical carcinogens. So far this has not been substan-
tiated  in  this country; in our work  with  benzopyrene,  we were able to induce
various cellular  inflammatory reactions in oysters, but  we have not  successfully
induced a tumor in oysters, but we have only really started in that area.

Dr. Morris,  EPA:   I was intrigued by your model earlier where you  showed the
relationships  of man and the rat  and  fish and showing these overlaps.  This  is an
interesting concept because  I think  as we collect, and certainly  in our agency,
human  health effects data, ecotox data, at  some point in time we may  want to put
this together and ask questions about  these inter-relationships.  I think this kind of
approach is very useful to that way of  thinking.

Dr. Couch,  EPA:   One word  of  caution.  I think  that with any phylogenetic or
comparative approach you have to be  familiar enough with the two organisms that
you are comparing to know where the differences begin and the similarities end.

-------
     FIRST NCl/EPA/NIOSH COLLABORATIVE WORKSHOP:
        PROGRESS ON 3OINT ENVIRONMENTAL AND
            OCCUPATIONAL CANCER STUDIES
                 Wednesday Morning, May 7
METHODOLOGY/EXPERIMENTAL MODELS SESSION (CONTINUED)
                 SESSION CHAIRPERSON
                    Dr. 3ohn Cooper*
                 National Cancer Institute
     * Dr. Cooper substituted for Dr. Umberto Saffiotti, NCI
                             3'?*~
                              ( I

-------
    METABOLISM OF AZO DYES TO CARCINOGENIC AMINES
                Larry K. Lowry, Ph.D.




                William P. Tolos, B.S.




    Division of Biomedical and Behavioral Science




National Institute for Occupational Safety and Health









                 Mark Boeniger, M.S.




              Division of Surveillance,




        Hazard Evaluations, and Field Studies




National Institute for Occupational Safety and Health
                         and
                 Charles Nony, B.S.




                Malcolm Bowman, B.S.




     National Center for Toxicological Research




             Jefferson, Arkansas  72079









                   FDA 224-78-0004

-------
                                  ABSTRACT








The metabolism of the purified benzidine-based azo dye Direct Black 38 (DB-38)




and the 3-3'dichlorobenzidine-based Pigment Yellow 12 (PY-12) was studied in




the hamster.  A single oral dose of DB-38 containing 3.0 ppm benzidine (Bzd).




6.0 ppm 4-aminobiphenyl (4-ABP) and 670 ppm of 2,4-diaminoazobenzene (DiAmAzBz)




was administered at 100 mg/kg to 18 male Syrian golden hamsters.  Urine speci-




mens collected over a period of 8 days and analyzed by electron capture-gas




chromatography and high pressure liquid chromatography showed significant total




amounts of Bzd (10 yg), monoacetylbenzidine (MoAcBzd, 535 yg), diacetylbenzidine




(DiAcBzd, 28 yg) and 4-ABP (11 yg).  Levels of metabolites peaked at 8-16 hours




with MoAcBzd, the major metabolite, still quantitated after 7 days.  In addition,




alkaline hydrolyzable conjugates of Bzd (328 yg) and 4-ABP (613 yg) were found.




The level of excreted metabolites far exceeded the levels of Bzd and 4-ABP




present as impurities in the dye and represent metabolic breakdown of the dye.




hutagenic potential of DB-38 and the major metabolites evaluated with the Ames




Salmonella test indicated:  MoAcBzd and DiAcBzd - strong (TA 1538); 4-ABP -




moderate (TA 98 and 100); DB-38, Bzd, DiAmAzBz - weak tTA 100); hamster urine,




8-16 hours containing 26 yg MoAcBzd - moderate (TA 1538).  No compounds were




nutagenic in the absence of S-9 fraction.








In contrast to DB-38, studies vith PY-12 (100 mg/kg) did not produce any




detectable levels of the hypothetical netabolites indicating either no




•etabolisak or little absorption of the pigaent.

-------
                                ABBREVIATIONS









DB-38, Direct Black 38; PY-12, Pigment Yellow 12; Bzd, benzidine;




4-ABP, 4-aminobiphenyl; DiAmAzBz, 2,4-diamifioazobenzene; MoAcBzd,




monoacetylbenzidine; DiAcBzd, diacetylbenzidine; EC-GC, electron-




capture gas chromatography; HFB, heptafluorobutyryl; HPLC, high




pressure liquid chromatography.
ACKNOWLEDGMENT:









This effort was supported by an interagency agreement with the National Cancer




Institute.

-------
                                INTRODUCTION









Epidemiological investigations of workers in the dye industry have shown an




increase in bladder tumors over that  expected in age-adjusted cohort popula-




tions (1,2).  The benzidine-based dyes  represent one class of dyes and




include many of those most commonly used.









Benzidine (Bzd),  a structural  component of these dyes,  is also an impurity




and is a known human bladder carcinogen (3).   The hypothesis that the dyes




may be metabolized back to Bzd prompted studies  on the  carcinogenicity and




metabolism of these dyes.









Studies sponsored by the National Cancer Institute have shown that Direct




Black 38, Direct  Brown 95, and Direct Blue 6 produced liver tumors in rats




as early as five  weeks after continuous dosing .(4) .   Urine specimens contained




Bzd in the parts  per billion (ppb)  range.  Okajima,  et  al., found that Direct




Black 38 (DB-38)  produced bladder,  liver, and colon tumors in 46% of male rats




given 500 parts per million (ppm) of  the dye in  drinking water for 60 weeks (5).




Bzd was not detected in urine  specimens.









The metabolism of DB-38, Direct Brown 95, Direct Blue 6, and Direct Red 28 was




studied by Rinde  and Troll (6).  When single oral doses of the dyes were given




to Rhesus monkeys, Bzd and monoacetylbenzidine (MoAcBzd) were found in the urine.




It is known that  these commercial dyes  contain many impurities,  including Bzd.
                                             381

-------
The discovery that these dyes were carcinogenic in rats,  together with the




knowledge that they may be metabolized to Bzd in animals  and humans,  has




lead to the recommendation by the National institute for  Occupational Safety




and Health that workers no longer be suBject to the adverse health effects




of these dyes and that these dyes no longer be used (7,8).









Other concerns have been raised on the potential metabolic  fate of the




dichlorobenzidine-based pigment,  Pigment  Yellow 12 (PY-12)  (9).   Previous




research on the metabolism and carcinogenesis of benzidine-based azo  dyes




has been done with commercial dyes containing unspecified impurities.   The




question of the origin of Bzd in the urine of animals  dosed with these dyes




has not been answered.  This report describes a definitive  metabolism study




in the hamster of DB-38 containing defined levels of amine  impurities.




Specific methodologies were developed and used to identify  and  quantitate  dye




metabolites in urine.   Selected metabolites and the dye were also evaluated




for mutagenic. potential.   Preliminary metabolism studies  were also done with




PY-12.

-------
METHODS AND MATERIALS








DB-38 was purchased from GAF Corporation, New York,  N.Y.   PY-12 was purchased




from the Dry Color Manufacturers' Association, Nutley,  N.'J.   Other chemicals




were obtained from commercial sources or synthesized (9).   Purification of




the dye and pigment was conducted manually by exaustive liquid-liquid extraction.




Amine impurities were converted to their heptafluorobutytyl (HFB)  derivatives




and analyzed by electron-capture gas chromatography (EC-GC).   Stability and




recovery studies of the dye and potential metabolites in urine were conducted




prior to the dosing of animals.  Details of these procedures  are reported by




Nony and Bowman (10).








Potential metabolites of the dye, Bzd, and the pigment were synthesized and




HFB derivatives prepared.  Structures were confirmed by gas chromatography-




mass spectrometry.  Urine metabolites were extracted, derivatized and analyzed



                                                                63
by EC-GC using a 5% Dexil 300 on Anakrom Q glass column and a Ni   detector.




High pressure liquid chromatography (HPLC) was also used to analyze underiva-




tized urine extracts using a reverse phase column (v Bondapak C^g, Waters




Associates, Milford, MA.) and an ultraviolet detector at 295  nm.  Conjugated




metabolites were first hydrolyzed with sodium hydroxide,  then extracted and




analyzed by EC-GC and HPLC.  Details of the metabolite analysis are as reported




by Nony and Bowman (11).








Male Syrian golden hamsters obtained from ARS Sprague-Dawley, Madison, WI.,




weighing between 104-128 g were housed three to a cage with a total of 18
                                             383

-------
animals (6 cages) used with the dye.   Control urines were  collected  for  24  hours




prior to dosing.  Urine samples were  collected at  intervals of 0-8,  8-16, 16-24,




24-32, 32-48, 48-72, and 144-168 hours after dosing.   All  collections were  done




in the presence of dry ice.  Hamsters received a single  oral dose  of DB-38  in




water at 100 mg/kg body weight.









In a limited experiment, three hamsters were dosed orally  with 100 mg/kg of




PY-12 in trioctanoin.  Urines were collected as above.   Details  of the




experimental protocol are reported by Hony,  et al. (12).









Mutagenicity of urinary metabolites was evaluated  using  the Ames Salmonella




test with and without mouse liver microsomal metabolic activation  (9).









RESULTS









The Dyes









Figure 1 shows the structures of DB-38 and related substances with their




abbreviations.  DB-38 was analyzed upon receipt and was  found to contain




traces of Bzd and significant amounts of 4-aminobiphenyl (4-ABP) and diamino-




azobenzene (DiAmAzBz), both known carcinogens.  The latter two compounds have




not been previously reported as contaminants of azo dyes.  Table I shows the




levels of impurities before purification and immediately before  use. Note




that the levels of non-benzidine impurities were significantly reduced while




the level of Bzd actually increased.
                                             384

-------
Figure 2 shows the structures of PY-12 and related substances  along with




their abbreviations.  The pigment was analyzed and contained 89 ppm of




3,3'-dichlorobenzidine.  After purification,  the level  of dichlorobenzidine




was 0.3 ppm.  No other amine impurities were  detected.
    As Received




    After Purification
                                   Table I




                         Analysis of Direct Black 38




                         	Impurity, ppm
Benzidine    4-Aminobiphenyl    Diaminoazobenzene




  <0.1             150                9,200




   3.0             6.0                  670
Urinary Metabolites - Direct Black 38









Bzd (see Figure 1), raonoacetylbenzidine (MoAcBzd),  diacetylbenzidine (DiAcBzd),




and 4-ABP were positively identified in the urine of hamsters fed DB-38.   These




compounds were analyzed by both HPLC and EC-GC of HFB derivatives.   Table  II




shows the results of HPLC analysis of urine from hamsters given DB-38.  Bzd




excretion peaked at 0-8 hours but fell to control levels by 48 hours.   MoAcBzd




also peaked at 0-8 hours, but was 100 times more concentrated than Bzd.  Its




excretion returned to control levels after 168 hours.  DiAcBzd and 4-ABP  peaked




at 16 and 8 hours, respectively, before returning to control levels by  24  hours,




No DiAmAzBz was detected.
                                            385

-------
                                                Table II

  HPLC Analysis of Major Metabolites in Urine of Hamsters Fed One Dose (100 mg/kg)  of  Direct Black 38
a/
Total Amount Excreted (yg) of Indicated Metabolites (x ± SD)—
Sampling
Interval
(hr)
Pretreatment
(24 hr)
0- 8
8- 16
16- 24
24- 32
£ 32- 48
48- 72
144-168
Volume (ml) of
Urine (x ± SD)-
9.1 ±
2.3 ±
3.7 ±
2.7 ±
1.9 ±
6.2 ±
12. ±
8.5 ±
5.4
0.1
3.2
0.7
0.6
2.6
7.8
1.3
Monoacetyl- Diacetyl-
Benzidine Benzidine Benzidine
0.093 ± 0.014 0.142 ±
2.33 ±2.16 196. ±
1.89 ± 2.04 174. ±
1.89 ± 2.19 98.7 ±
0.258 ± 0.000 6.07 ±
0.524 ± 0.000 6.17 ±
ND 1.36 ±
ND 0.710 ±
0.044 0.175 ± 0.030
107. 2.78 ± 1.68
170. 4.69 ± 1.34
146. 2.39 ± 0.60
4.80 ND
6.73 ND
1.08 ND
0.188 ND
4-Aminobiphenyl
0.473 ± 0.028
4.28 ± 2.41
3.75 ± 2.86
2.02 ± 1.01
ND
ND
ND
ND
a/  Mean and standard deviation from five cages of three hamsters each.
    pretreatment sample background and recovery.

b/  Mean and standard deviation from five cages of three hamsters each.

ND  None detected above background.
Results were corrected for

-------
Figure 3 shows an EC-GC chromatogram from urine collected 8-16 hours after




dosing with DB-38.  The chromatogram clearly shows 4-ABP (1.4 ppm),  Bzd




(1.7 ppm), and MoAcBzd (78.7 ppm).  DiAcBzd and the alkaline hydrolyzable




conjugates of Bzd and 4-ABP were analyzed following hydrolysis to free amine




metabolites as previously described (9).









Table III shows the composite results of an EC-GC analysis of hamster urine




from dosed animals.  The results are similar to those from HPLC analysis but,




because of higher sensitivities, levels of metabolites have been detected at




longer intervals after dosing.  Table IV shows the results for alkaline




hydrolyzable conjugates analyzed by EC-GC after pretreatment.  It can be seen




that Bzd conjugates are present at about 30 times the level of free  Bzd but




that 4-ABP conjugates are present at about the same levels as the free




metabolites.  No DiAmAzBz was found in hamster urine by either HPLC  or EC-GC,









Urinary Metabolites - Pigment Yellow 12









The expected metabolites of PY-12 shown in Figure 2 were not detected in the




urine of hamsters dosed with PY-12.  The expected metabolites were chemically




synthesized, spiked into control urines and analyzed at the ppb level by EC-GC.




However, no traces of these expected metabolites were found in dosed hamster




urine.
                                           38

-------
                                               Table III

             EC-GC Analysis of Major Metabolites in Urine of Hamsters Fed Direct  Black 38
a/
Total Amount Excreted (ug) of Indicated Metabolites (x ± SD)—
Sampling
Interval Volume (ml) of
(hr) Urine (x ± SD)-'



GO
CO
m


Pi ^.treatment
(24 hr)
0- 8
8- 16
16- 24
24- 32
32- 48
48- 72
144-168
9.1 ±
2.3 ±
3.7 ±
2.7 ±
1.9 ±
6.2 ±
12. ±
8.5 ±
5.4
0.1
3.2
0.7
0.6
2.6
7.8
1.3
Monoacetyl- Diacetyl-
Benzidine Benzidine Benzidine
0.151 ± 0.217 0.361 ±
4.33 ± 2.39 216. ±
3.08 ±2.44 208. ±
2.16 ± 2.33 95.8 ±
0.259 ± 0.205 5.59 ±
0.470 ± 0.506 7.39 ±
0.226 ± 0.184 1.56 ±
ND 0.474 ±
0.103 0.087 ± 0.061
133. 10.7 ± 7.1
177. 10.3 ± 2.5
133. 5.69 ±5.11
4.88 0.064 ± 0.063
7.82 0.223 ± 0.212
0.994 0.626 ± 0.000
0.393 ND
4-Arainobiphenyl
0.152 ± 0.155
6.62 ± 1.72
3.26 ± 3.02
1.17 ± 1.52
0.162 ± 0.084
0.236 ± 0.078
ND
ND
a/  Mean and standard deviation from five cages of three hamsters each.   Results were corrected for
    pretreatmerit sample background and recovery.

W  Mean and standard deviation from five cages of three hamsters each.

ND  None detected above background.

-------
                                  Table IV

             EC-GC Analysis of Alkaline Hydrolyzable Conjugates
                in the Urine of Hamsters Fed Direct Black 38

                                        Total Amount Excreted
                                     	(yg)	
            Sampling Interval (hr.)   Benzidine  4-Aminobiphenyl

            Pretreatment (24 hrs.)      0.034        1.16

                     0-  8            103.            2.57

                     8- 16            154.            2.56

                    16- 24             45.5           ND

                    24- 32              5.59          ND

                    32- 48             13.9           ND

                    48- 72              6.41          ND

                   144-168              0.019         ND



            ND - None detected above background.



Mutagenicity Testing



The mutagenic potential of the identified metabolites of DB-38 is  shown  in

Table V.  The purified synthetic metabolites were assayed  with and without

S-9 fraction.  None of the compounds tested were mutagenic without activation.

All compounds showed some degree of mutagenic activity in  the presence of S-9

fraction with at least one of the tester strains.  The degree of mutagenicity

based on the number of revertants per plate were:  MoAcBzd and DiAcBzd - strong;

4-ABP - moderate; Bzd and DiAmAzBz - weak.   Mutagenic evaluation of  DB-38 hamster
                                       38£

-------
urine collected at pretreatment, 0-8 and 8-16 hours after dosing,  showed no

mutagenic activity without activation but mutagenic activity up  to 10 times

background in the 0-8 hour urine with activation.   These results are consistent

with the level of MoAcBzd in the urine,  the major  metabolite.



                                   Table V

                       Mutagenic Potential of Direct
                          Black 38 and Metabolites

                                               Mutagenic Response
           Compound	Tester  Strain    -S9	+S9

           Direct Black 38          TA 98
                                    TA 100         -         +

           Benzidine                TA 98          -         +
                                    TA 100         -         +

           Monoacetylbenzidine      TA 1538        -

           Diacetylbenzidine        TA 1538

           4-Aminobiphenyl          TA 98
                                    TA 100

           Diaminoazobenzene        TA 98          -
                                    TA 100
           -    Negative
           +    Weak (2 x background)
           -H-   Moderate (5 - 10 x background)
           +++  Strong (40 x background)
CONCLUSIONS



Purified DB-38, containing low defined levels of free Bzd as an impurity,  was

extensively metabolized to Bzd, the N-acetylbenzidine metabolites and to
                                          30C

-------
unspecified Bzd conjugates in hamsters.   About  8-10%  of  the Bzd  contained  in




the dye in azo linkage could be accounted for in the  urine as  Bzd  or  its




metabolites.  In addition, 4-ABP was found in amounts much greater than the




level of impurity, along with its conjugates.  These  findings  indicate that




considerable risk may exist for humans exposed  to DB-38  since  both Bzd and




4-ABP are well established bladder carcinogens.   In addition,  MoAcBzd, the




major metabolite, was found to be mutagenic with activation using  the Ames




Salmonella test.









In contrast, PY-12 did not produce detectable levels  of  expected metabolites,




indicating either no metabolism or little absorption  of  the pigment.
                                      39J

-------
                                 REFERENCES









1.    Yoshida,  0.  and M.  Miyaka.   1973.   Etiolo'gy of Bladder Cancer -




     "Metabolic"  Aspects in Nakohara,  W.,  T.  Hirayama, K. Nishioka and




     H.  Sugano (eds.).   Analytic and Experimental Epidemiology of Cancer,




     Proceedings  of the  3rd International  Symposium of the Princess




     Tamamatsu Cancer Research Fund, 1972.   Baltimore, University Park




     Press (pub.  1973)  pp.  31-39.




2.    Genin,  V.A.   1977-   [Formation of Blastomogenic Diphenylamino Derivatives




     as  a Result  of the  Metabolism of  Direct  Azo Dyes.]  Vopr. Onkol.




     2.3 (9): 50-52  (Rus) .




3.    Federal Register 39(20):3756-97,  January 29, 1974.




4.    13  Week Subchronic  Toxicity Studies of Direct Blue 6, Direct Black 38,




     and Direct Brown 95 Dyes.  National Cancer Institute.  Carcinogenesis




     Technical Report.   DREW Publication Mo.  (NIK) 78-1358, 1978.




5.    Okajima,  E. ,  K. lliramatsu,  I. Ishu, S. Matsujima, K. Yamada, and




     M.  Arai.   1975.  [Multiple  Tumors in  Rats After Oral Administration




     of  the Benzidine-type Dye Direct  Deep Black EX.]   Igaku no Ayumi




     92:291 (Jap).




6.    Rinde,  E. and  W. Troll.   1975.  Metabolic Reduction of Benzidine Azo




     Dyes to Benzidine in the Rhesus Monkey.   J. Nat.  Cancer. Soc. 55:181-182.




7.    Direct Black 38, Direct  Blue 6, and Direct Brown 95 Benzidine Derived




     Dyes.  Current Intelligence Bulletin  No. 24.  National Institute for




     Occupational Safety and  Health, Rockville, MD.  DUEW (NIOS1I) Publication




     78-148,  April  1978.

-------
8.    Special Hazard Review of Benzidine-Based '"lyes.  National Institute




     for Occupational Safety and Health.  DIIEW (NIOSH) Publication,




     October 1979.




9.    Metabolism of  Azo Dyes to Potentially Carcinogenic Amines.  Final




     Report of Interagency Agreement FDA 224-78-0004 Between the National




     Institute for  Occupational Safety and Health and the Food and Drug




     Administration,  March 1979.




10.   Nony, C.R. and M.C. Bowman.  1980.  Analysis, Purification and




     Stability:  Requirements for a Metabolism Study of an Azo Dye and




     Pigment.   J. of Analytical Toxicology (in press).




11.   Nony, C.R. and M.C. Bowman.  1980.  Trace Analysis of Potentially




     Carcinogenic Metabolites of an Azo Dye and Pigment in Hamster and




     Human Urine as Determined by Two Chromatographic Procedures.




     J. Chromatogr. Sci. 18(2):64-74.




12.   Nony, C.R., M.C. Bowman, T. Cairns, L.K. Lowry; and W.P. Tolos.  1980




     Metabolism of  an Azo Dye and Pigment by Hamsters to Potentially




     Carcinogenic Aromatic Amines as Determined by Analysis ot the Urine.




     J. of Analytical Toxicology (in press).

-------
                               FIGURE LEGENDS









Figure 1 - Direct Black 38 and Related Compounds.




Figure 2 - Pigment Yellow 12 and Related  Compounds.




Figure 3 - Electron-Capture Gas Chromatograms of Derivatized Extracts of




           Urine Collected From Hamsters  8-16 Hours After Treatment With




           Direct Black 38.
                                          394

-------
          NHa
                                     H2N   OH        ,.  .x
                                   N. ^J^JN=N-t'  ^

                                        T
                              Na03S

                     CX Direct Black 38
       S03Na
Diaminoazobenzene
   (DiAmAzBz)
4-Aminobiphenyl
    (4-ABP)
                                                       N-C-CH3
      Benzidine
       (Bzd)
Monoacetylbenzidine
    (MoAcBzd)
                   0  H
                        Diacetylbenzidine
                           (OiAcBzd)
                                 Figure  1

-------
            H3C
           HOC
      NH-OC-C-N=N
     CH3
     COH
                   C.I. Pigment Yellow 12
                                          Cl
                                                       H 0
                                                        I   It
                                                       N-C-CH3
3,3'- Dichlorobenzidine
     (DiCIBzd)
Monoacetyldichlorobenzidine
      (MoAcDiCIBzd)
                  Diacetyldichlorobenzidine
                       (DiAcDiCIBzd)
                                 Figure 2
                                       39C

-------
   1.6
LU
<
o
CO
   19
   Q.
   i
   b 0.8
   x
   CM
CD *-
-•» Itl
   CO

   I0-4
   C£
URINE COLLECTED FRCf^l HAMSTERS 8-16 HRS AFTER TREATMENT WITH
• DIRECT BLACK 38
 220°C
 5.0ug-equiv/inj.
          4-ABP
          (l.4ppm)
                    Bzd
                   (l.7ppm)
                                T
                                   260°C
                                   6.25ug-equiv/inj.
                                 Bzd
                                  MoAcBzd
                                 (78.7 ppm)
280°C
2.5ug-equiv/lnj.
                                                                 MoAcBzd
                                                                 (78.7ppm)
                     8
                          4     8      12
                              MINUTES
                                                         16
           8

-------
                                  Discussion
Dr. Jenkins, EPA: I would like to ask one question about the problems you would
have in the HPLC analysis of benzidine and I believe you said that you converted
that to the appropriate derivative.  The reason for my asking would be what other
types of compounds would interfere, come out at about the same peak,  and what
you are converting that to? I am interested in the congeners.

Dr.  Lowry,  N1OSH:    The derivative  you  mentioned  was a  heptafluorobutyryl
anhydride reaction with an aromatic amine to produce a fluoroelectron capture
sensitive group that would be picked up with the electron capture detector of the
gas chromatograph.

The  HPLC analysis of benzidine was done without derivatization using a reverse
phase C,o column. The congeners of benzidine are separated from benzidine under
these conditions.  More details of  the  methodology can  be  found in the paper
written by C.  R. Nony and M. C. Bowman published in the February issue of the
Journal of Chromatographic Science (Vol. 18, pages 64-75, 1980).

Dr. Cooper, NCI:  It was curious that on the benzidine results in the minus 24 to
zero  time period you  showed excretion of  benzidine  and  yet it  fell later to
undetectable levels.  Where was that benzidine coming from?

Dr. Lowry, NIOSH: It is possible you could call that noise.

Dr. Hegyeli, NCI: Was inhalation considered in this case as most of the workers
working with these types of chemicals were inhaling instead of ingesting them.

Dr. Lowry, NIOSH: Inhalation was not considered primarily for reasons of getting
the work done with the amount of money that was available to support the work.
Further work  is being done on two congeners of  the benzidine-based dyes,  one
toludene- based dye, Direct Red 2 and one  dianicidine-based dye, Direct Blue 15.
NCTR is about to start some work using some radio labeled material to look more
thoroughly at metabolism, absorption of the material, and tissue distribution.

-------
VOL.26, NO.3                                        Research Communications in
DECEMBER 1979                            Chemical Pathology and Pharmacology
         THE EFFECT OF  DIETARY DISULFIRAM UPON THE TISSUE DISTRIBUTION
               AND EXCRETION  OF  14C-1,2-DIBROMOETHANE IN THE RAT

                     HARRY  B. PLOTNICK, WALTER W. WEIGEL,
                  DONALD  E. RICHARDS, AND KENNETH L. CHEEVER

               U.S.  DEPARTMENT OF  HEALTH, EDUCATION, AND WELFARE
             U.S.  Public  Health  Service, Center for Disease Control
              National  Institute for Occupational Safety and Health
                  4676  Columbia  Parkway, Cincinnati, Ohio 45226

                                   ABSTRACT

    Dietary disulfiram  enhances  the toxicity of inhaled 1,2-dibromoethane in

    rats.   This study was undertaken to determine whether the differential

    toxicity noted was  associated  with alterations in the levels of the com-

    pound and/or its metabolites in the target organs.  A comparison of the

    levels of 14C in selected tissues of male rats, with and without dietary

    disulfiram, following the oral administration of 11(C-l,2-dibromoethane

    was made.  The results  indicated that levels of radioactivity in the

    target organs of animals  in  the disulfiram group were significantly

    elevated both at 24 and 48 hours following compound administration.  The

    data indicate a direct  correlation between tissue levels and the en-

    hancement of toxicity noted  in the disulfiram-treated rats in the in-

    halation study.   A  significant elevation in the levels of radioactivity

    in washed liver nuclei  obtained from animals receiving dietary disul-

    firam was also noted, suggesting a relationship between nuclear uptake

    and the increased incidence  of liver tumors appearing in the disulfiram

    group in the inhalation study.

                                  INTRODUCTION

    A study of the toxicity of 1,2-dibromoethane in rats, exposed at the

    current U.S.  occupational standard of 20 ppm, with and without disul-

    firam in the diet,  has  recently been completed by the Midwest Research
                                        39

-------
VOL.26, NO.3                                        Research Communications in
DECEMBER 1979                            Chemical Pathology and Pharmacology
     Institute, Kansas City, Missouri  (NIOSH, 1979).  Preliminary results

     noted at  the end of the eleventh month of the 18-month study were re-

     ported earlier  (Plotnick, 1978).  The final results of this study

     indicate  that 1,2-dibromoethane, at an exposure level of 20 ppm under

     simulated occupational exposure conditions (7 hours/day, 5 days/week),

     is  a carcinogen in both male and female Sprague-Dawley rats.  A most

     interesting finding in this study was an enhancement of carcinogenic and

     other toxic effects in the disulfiram-treated rats.  The combined ex-

     posure resulted in significant increases in tumors of the liver, spleen,

     kidney, and thyroid as well as atrophy of the testes when compared with

     those animals exposed to 1,2-dibromoethane alone.  The most profound

     disulfiram-related effect was an increase in the incidence of hepato-

     cellular  carcinoma.  The present study was undertaken to determine

     whether dietary disulfiram modifies the tissue distribution and excre-

     tion of orally-administered 1,2-dibromoethane, utilizing 1<4C-labeled

     material, in an attempt to explain the interaction noted.  Special

     attention was given to the organs affected in the chronic inhalation

     study.  In addition, liver nuclei were isolated for quantitation of

     radioactivity to determine whether the disulfiram diet was associated

     with a preferential distribution to this organelle.

                            MATERIALS AND METHODS

     Animals.  Male  rats of the Sprague-Dawley strain, weighing 100-125 g,

     were obtained from Charles River Breeding Laboratories, Inc., Wilmington,

     MA.  For  a one-week acclimation period, the animals were housed two per

     cage, under controlled environmental conditions, with a 12-hr light-dark

     schedule with the light cycle beginning at 7:00 a.m.  Pelletized Rodent
                                          40C

-------
VOL.26, NO. J                                        Research Communications in
DECEMBER 1979                            Chemical Pathology and Pharmacology
     Laboratory  Chow (Ralston Purina Co., St. Louis, MO) and tap water were

     available ad  libitum during this acclimation period.

     Chemicals.  Tetraethylthiuram disulfide  (disulfiram) was obtained from

     the Sigma Chemical  Co., St. Louis, MO.   [U-ll*C]l,2-dibromoethane was

     purchased from New  England Nuclear, Boston, MA.  Analysis of the radio-

     isotope by  the supplier prior to shipment showed this compound to have

     a radiochemical purity of 98% and a specific activity of 9.35 mCi/mmole.

     1,2-Dibromoethane  (98%) was obtained from MC/B Manufacturing Chemists,

     Norwood,  OH.

     Treatment Solution.  The treatment solution was prepared by mixing the

     [U-ll*C]l,2-dibromoethane with sufficient "unlabeled" 1,2-dibromoethane

     in corn oil (MC/B Manufacturing Chemists, Norwood, OH) to yield a con-

     centration  of 3.75  mg 1,2-dibromoethane/g of treatment solution with a

     specific activity of 6800 dpm/yg of 1,2-dibromoethane.

     Animal Treatment.   Upon initiation of  the experiment, 24 rats were

     weighed, randomly assigned  to two experimental groups of 12 animals

     each, and placed in individual stainless-steel metabolism cages design-

     ed for the  separate collection of urine  and feces.  The rats in one

     group received ground Rodent Laboratory  Chow containing 0.05% disul-

     firam, while  those  in the second group received plain ground Rodent

     Laboratory  Chow, as a control diet.  All animals had unlimited access

     to tap water  and their respective diets.  After 12 days on these diets,

     the rats were fasted overnight and subsequently given single 4-g/kg

     (15 mg 1,2-dibromoethane/kg) doses of  the treatment solution by oral

     intubation.  The rats were  returned to their metabolism cages, and

     urine and feces were collected from each rat at 24-hour intervals from
                                           40]

-------
VOL.26, NO.3                                        Research Communications in
DECEMBER 1979                            Chemical Pathology and Pharmacology
    the  time  of  treatment with  1,2-dibromoethane until  the  animal was kill-

    ed.   Six  rats  from  each  experimental  group were killed  at 24 and 48

    hours after  administration  of  the  treatment solution by exsanguination

    by cardiac puncture with a  heparinized  syringe following anesthetiza-

    tion by 100-mg/kg intraperitoneal  injections of sodium  pentobarbital

    (Nembutal Sodium, Abbott Laboratories,  North Chicago, IL).  In  addition

    to blood, the  liver, kidneys,  spleen,  testes, brain and suprarenal fat

    were removed at  autopsy  for analysis  of ll*C activity.   Plasma was

    obtained  by  centrifugation  of  an aliquot  of the heparinized blood.

    Isolation of Nuclei from Rat Liver.   Livers taken at autopsy were im-

    mediately placed in ice-cold 0.25M sucrose in TKM buffer solution

    [0.05M tris-(hydroxymethyl)-aminomethane, 0.025M KC1, and 0.005M

    Mgd2, adjusted  to  pH  7.5 at 20° C] .   The livers were individually

    homogenized  in two  volumes  of  0.25M^ sucrose in TKM  and  the nuclei from

    9.0 ml of the  homogenate (equivalent  to 3 g of whole liver) were iso-

    lated according  to  the method  of Blobel and Potter  (1966).  The

    nuclear pellets  were resuspended in 36  ml of TKM buffer and subsequent-

    ly resedimented  by  centrifugation  at  5000 x g for 10 minutes.  These

    washed nuclear pellets were then analyzed quantitatively for 1'*C.  A

    2-ml aliquot of  each liver  homogenate was taken as  the  liver sample to

    be analyzed  for  11(C as part of the tissue distribution  studies.

    Radioactivity  Assays.  All  biological samples, including urine  and

    feces, were  prepared and analyzed  for 14C activity  by the method of

    Weigel, Plotnick, and  Conner (1978) using a Beckman LS  8100 Liquid

    Scintillation  System.

    Statistical  Analysis of  Data.   The significance of  differences between

-------
VOL. 26, NO. 3                                        Research Communications in
DECEMBER 1979                            Chemical Pathology and Pharmacology
    groups was  determined by  Student's ^-test at a probability level of

    0.05.

                                  RESULTS

    At  both  time  intervals studied, tissue concentrations of  1I+C in the

    liver, kidneys,  spleen, testes, and brain were significantly higher in

    the rats receiving  disulfiram  in  the diet.  Urinary excretion of

    radioactivity during the  first 24 hours after compound administration

    was significantly depressed  in the disulfiram diet group.  A compar-

    ison of  tissue levels at  24  and 48 hours indicates that the rate of

    clearance of  *"*C from liver, kidneys, spleen, testes, and brain was

    appreciably lower in the  disulfiram group.  This was particularly true

    for clearance from  the testes.  While the levels (vg/g) in the testes

    of  the animals in the control  group at 24 and 48 hours were signifi-

    cantly different (p < 0.001),  a comparison of the corresponding levels

    in  the testes of the disulfiram group at the two time intervals indi-

    cated that  they did not differ significantly (p > 0.2).  There were no

    significant differences between the groups with respect to levels of

    JI*C in the  fat and  whole  blood at either time interval studied.  The

    tissue distribution data,  expressed both as tissue concentrations and

    as  a percentage of  the administered dose, appear in Tables 1 and 2.

    The levels  of radioactivity  in the washed liver nuclei obtained from

    the disulfiram-treated animals were significantly higher than those of

    the controls  at both 24 and  48 hours.  These data appear in Table 3.

-------
Table 1 - Effect of Dietary Disulfiram Upon the Distribution of  14C in Selected Tissues and  Body
          Fluids of Male Rats 24 Hours After a Single Oral Dose  of 15 mg/kg  [U-^C]!,2-Dibromoethane
                  TISSUE CONCENTRATIONS1
PERCENT OF DOSE2
TISSUE
LIVER
KIDNEYS
SPLEEN
TESTES
BRAIN
FAT3
BLOOD1*
PLASMA
URINE5

CONTROL DIET
4.
3.
1.
0.
0.
0.
0.
0.



78
32
00
49
41
35
90
46



+
+
+
+
+
+
+
+



0.
0.
0.
0.
0.
0.
0.
0.



24
42
03
05
04
04
05
04



DISULFIRAM DIET
6.
6.
1.
0.
0.
0.
0.
0.



52
82
56
88
64
48
95
56



+
+
+
+
+
+
+
+



0
1
0
0
0
0
0
0



.396
.376
.146
.106
.086
.07
.09
.05



1. Values represent mean concentrations in yg/g or pg/ml  (expressed as parent  compound)  ±  S.E.M.  of
   duplicate determinations on six animals.
2. Values represent the mean percentage of the administered radioactive dose ±  S.E.M.  of duplicate
   determinations on six animals.
3. 6% of body weight (Donaldson, 1924).
4. 9% of body weight (Donaldson, 1924).
5. n = 12 (includes 24 hour samples obtained from rats killed 48 hours after compound  administration).
6. Significantly different from control values (p < 0.05).
                                          o  <
                                          R  °
                                          rn  •
                                          f  **>
                                          i.  ffv
                                          CD  -
                                          5  Z
                                          2  P
                                          vo  U>
                                          vj
                                          >£
CONTROL
1.
0.
0.
0.
0.
0.
0.

72.
1.
79
21
02
04
02
15
59

38
65
+
+
+
+
+
+
+

+
+
DIET
0.
0.
<0
<0
<0
0.
0.

0.
0.
07
02
.01
.01
.01
02
03

98
28
DISULFIRAM DIET
2
0
0
0
0
0
0

64
1
.46
.45
.02
.07
.03
.20
.59

.86
.60
+
+
+
+
+
+
+

+
+
0.
0.
<0
0.
<0
0.
0.

1.
0.
166
096
.01
Ol6
.Ol6
02
05

945
37







n
3-
n
3,
n'
BJ
                                         ?1
                                         £J  2.
                                         3  «
                                         1  ft
                                         8  |
                                         5" vi
                                         90  —
                                         •<  3

-------
Table 2 - Effect of Dietary Disulfiram Upon the Distribution of 14C in Selected Tissues and Body
          Fluids of Male Rats 48 Hours After a Single Oral Dose of 15 mg/kg [U-1'*C]l,2-Dibromoethane
                  TISSUE CONCENTRATIONS1
PERCENT OF DOSE2
TISSUE
LIVER
KIDNEYS
SPLEEN
TESTES
BRAIN
FAT3
BLOOD14
PLASMA
Tnrnr<;5
CONTROL DIET
2.87 ± 0.33
1.06 ± 0.16
0.66 ± 0.03
0.19 ± 0.02
0.17 ± 0.02
0.44 ± 0.06
0.64 ± 0.07
0.22 ± 0.02

DISULFIRAM DIET
5.23 ± 0.386
4.31 ± 0.406
1.29 ± 0.126
0.72 ± 0.086
0.50 ± 0.036
0.53 ± 0.05
0.81 ± 0.05
0.39 ± 0.036

                                                                CONTROL DIET
                                                                 1.10 ± 0.12
                                                                 0.08 ± 0.01
                                                                 0.01 ± <0.01
                                                                 0.01 ± <0.01
                                                                 0.01 ± <0.01
                                                                 0.20 ± 0.03
                                                                 0.43 ± 0.04

                                                                73.54 ± 2.80
                                                                 2.42 ± 0.54
          DISULFIRAM DIET
           1.74 ± 0.106
           0.27 ± 0.036
           0.02 ± O.OO6
           0.06 ± 0.016
           0.03 ± <0.016
           0.23 + 0.02
           0.53 ± 0.03
          66.95 ± 2.48
           1.56 + 0.45
1.  Values represent mean concentrations in yg/g or ug/ml (expressed as parent compound)  ± S.E.M.  of
   duplicate determinations on five animals.
2.  Values represent the mean percentage of the administered radioactive dose ± S.E.M,  of duplicate
   determinations on five animals.
3.  6% of body weight (Donaldson, 1924).
4.  9% of body weight (Donaldson, 1924).
5.  Cumulative 48 hour excretion.
6.  Significantly different from control values (p < 0.05).
O <
S °
pfl •
7 KJ
I-*
E z
29
vo t*>
vj
<£
n
n'
                                       ? «
                                       « i/>
                                       ? ro
                                       O ti
                                       w n
                                       3 §
                                       a. 3

                                                                                                               II
                                                                                                               00  -•
                                                                                                               -<  3

-------
VOL.26, NO.3                                        Research Communications in
DECEMBER 1979                            Chemical Pathology and Pharmacology
     Table  3  -  Effect  of Dietary Disulfiram Upon the ll|C Content of Liver
               Nuclei  Isolated 24 or 48 Hours After Administration of a
               Single  Oral Dose of 15 mg/kg [U-^Cjl^-Dibromoethane

    TIME INTERVAL          CONTROL               DISULFIRAM

      24 Hours             687 ± 821             1773 ± 3142

      48 Hours             460 ± 42              1534 ± 1973
    1.  Results are  expressed as disintegrations per minute per pellet
       (Mean  ± S.E.M.) of duplicate determinations on six animals per
       group  at  24  hours and five animals per group at 48 hours.
    2.  Significantly different from control values (p < 0.01).
    3.  Significantly different from control values (p f 0.001).
                                 DISCUSSION

    The  study performed by Midwest Research Institute established that the

    addition of disulfiram to the diet of rats enhanced the toxicity of

    inhaled 1,2-dibromoethane.  The study reported here demonstrates that

    addition of disulfiram to the diet of male rats results in significant

    increases in  the  tissue levels of subsequently administered *"*C-l,2-di-

    bromoethane in  the organs studied.  This increase in tissue levels was

    evident at both 24 and 48 hours following administration of the halo-

    genated hydrocarbon and probably accounts for the interaction noted.

    Particularly  noteworthy is the increase in the amount of radioactivity

    in washed nuclei  isolated from the livers of rats in the disulfiram

    group.  At 24 and 48 hours there are 1.4-fold and 1.8-fold increases,

    respectively, in  the levels of radioactivity in the livers of animals

    in the disulfiram group as compared with the control animals.  The

    corresponding ratios obtained for the nuclear pellets are 2.6 and 3.3,

    respectively, indicating a non-uniform, preferential distribution to

    this organelle.   This increased nuclear uptake of 14C in the liver may
                                     40£

-------
VOL.26, NO.3                                        Research Communications in
DECEMBER 1979                            Chemical Pathology and Pharmacology
    account for the high percentage of hepatocellular carcinomas found in

    the inhalation study in the group exposed to 1,2-dibromoethane while

    receiving disulfiram in the diet when compared to those exposed to 1,2-

    dibromoethane and receiving a control diet (70% vs.  5%).   Slower clear-

    ance of   C from the testes in the disulfiram group  may explain the

    significantly greater incidence of testicular atrophy observed in the

    rats receiving the combined exposure in the inhalation study when com-

    pared with those exposed to the halogenated hydrocarbon alone (90% vs.

    4%).  The mechanism of the enhancement of carcinogenicity by disulfiram

    is presently unclear.  Researchers at the Southern Research Institute

    (Hill et al., 1978) have identified bromoacetaldehyde as  an intermedi-

    ate of 1,2-dibromoethane metabolism in the rat.   Based upon this

    observation, one could speculate that disulfiram, a  known inhibitor of

    aldehyde dehydrogenase, blocks the further oxidation of the bromoacet-

    aldehyde formed, resulting in increased tissue levels of  this inter-

    mediate.  While little is apparently known about the toxicity and bio-

    logical reactivity of bromoacetaldehyde, such information is available

    on another a-haloaldehyde, chloroacetaldehyde.  In the Ames mutagen

    assay, employing tester strain TA100 without activation,  chloroacetal-

    dehyde was found to be 746 times more active as a mutagen, per ymole,

    than vinyl chloride (McCann e^t £l., 1975).  In addition,  it is known

    that chloroacetaldehyde reacts non-enzymatically with the nucleic acid

    bases adenine and cytosine to form so-called "etheno" derivatives (Se-

    crist ££ al^., 1972).  Such a reaction, if it occurs  in vivo, could pro-

    duce significant alterations in nucleic acids.  It is also possible

    that disulfiram merely inhibits the excretion of 1,2-dibromoethane
                                         40'

-------
VOL.26, NO.3                                        Research Communications in
DECEMBER 1979                             Chemical Pathology and Pharmacology
    metabolites  and  that  the  increased  tissue  levels reflect  this  inter-

    ference with excretion.   This alternate hypothesis  is  supported by a

    recent study which  suggests  that disulfiram interferes with  the excre-

    tion  of barbital, a hypnotic agent  which does not undergo any  signifi-

    cant  biotransformation, resulting in an increased barbital sleeping

    time  in rats (Sharkawi and Cianflone, 1978).  Studies of  the biochem-

    ical  mechanism of this interaction  are currently in progress in our

    laboratory.

                                  REFERENCES
     Blobel,  G.  and  Potter, V.R.  (1966). Nuclei From Rat Liver: Isolation
     Method  that  Combines Purity with High Yield. Science 154: 1662-1665.

     Donaldson,  H.H.  (1924). The Rat Data and Reference Tables. The Wistar
     Institute of Anatomy and Biology, Philadelphia, pp. 181-298.

     Hill, D.L.,  Shih,  T., Johnston, T.P. and Struck, R.F.  (1978). Macro-
     molecular Binding  and Metabolism of the Carcinogen 1,2-Dibromoethane.
     Cancer  Res.  38:  2438-2442.

     McCann,  J.,  Simmon, V., Streitweiser, D. and Ames, B.N.  (1975). Muta-
     genicity of  chloroacetaldehyde, a possible metabolic product of 1,2-
     dichloroethane  (ethylene dichloride), chloroethanol (ethylene chloro-
     hydrin), vinyl  chloride, and cyclophosphamide. PFOG. Nat. Aaad. Soi.
     USA  72:  3190-3194.

     NIOSH  (1979). Carcinogenicity and Toxicity of  Inhaled  1,2-Dibromoethane
     in Rats With and Without Disulfiram in  the Diet. National Institute for
     Occupational Safety and Health, Cincinnati, Ohio, Contract No. 210-76-
     0131.  In preparation.

     Plotnick, H.B.  (1978). Carcinogenesis in Rats  of Combined Ethylene Di-
     bromide and  Disulfiram  (Letter to the Editor). J. Amer. Med. Aasn.
     239: 1609.

     Secrist.III, J.A., Barrio, J.R., Leonard, N.J. and Weber, G.  (1972).
     Fluorescent  Modification of Adenoslne-Containlng Coenzymes. Biological
     Activities  and  Spectroscopic Properties. Biochemistry  11: 3499-3506.

     Sharkawi, M. and Cianflone, D. (1978).  Disulfiram Enhances Pharmaco-
     logical Activity of Barbital and Impairs Its Urinary Elimination.
     Science 201: 543-544.

-------
VOL.26, NO.3
DECEMBER 1979
         Research Communications in
Chemical Pathology and Pharmacology
     Weigel,  W.W., Plotnick, H.B. and Conner, W.L.  (1978).  Tissue Distri-
     bution and Excretion of 1'*C-Epichlorohydrin in Male and Female Rats.
     Res.  C&m.  Chem. Pathol. Phcamacol. 20: 275-287.
     Mention of company  or product names does not constitute endorsement
     by the National  Institute for Occupational Safety and Health.
     From the Mechanism of Industrial Disease Section,  Experimental
     Toxicology Branch,  Division of Biomedical and Behavioral Science
                            Copyright© 1979 By
               PJD Publications Ltd., Box 966, Westbury, N.Y. 11590
                                         40£

-------
                                  Discussion
Unidentified Speaker: I was interested in the finding with dibromoethane and I am
wondering whether or not the finding might be extrapolated to bromomethanes or
bromopropanes.  Do you think that that might be the case?

Dr. Plotnick, NIOSH:  You don't get the same type of metabolites.   So, I would
doubt it. With bromomethane, is this CH3BR you are talking about?

Unidentified Speaker: For example,  some the compounds that Dr. Bull was talking
about - dibromochloromethane is one of them.

Dr. Plotnick, NIOSH:  They don't undergo  the  same type  of  biotransformation -
through that alcohol aldehyde type of  thing.   They  are  two carbon compounds
rather than one carbon.  Propanes, yes, possibly because I  really  think that things
like DBCP act after loss of one of their carbons.

Dr. Bull, EPA:  I thought those were very interesting results you had.  Trichloro-
ethylene, as you noted, does go through a  similar kind of  metabolic  path, and do
you have an epidemiologic cohort, or  whatever, identifying disulfiram?

Dr. Plotnick, NIOSH:  I am not sure whether we are going to do anything with it at
this point. It is very intermittent exposure in very few people.

Dr. Bull, EPA:  I mean  there is evidence  of interaction I think  in epidemiologic
populations bet wen disulfiram.

Secondly, would there be a possibility of interaction  with ethanol, because ethanol
obviously goes through the same sort of — the same enzymes are involved  and at
least  you could see a  competitive  kind of interaction.  Has anyone done  any
interaction studies between ethanol and —

Dr. Plotnick, NIOSH:  I  did precisely the same study with 5% ethanol in  drinking
water.  The only thing I can tell you is that there may have been one defect in
experimental design.  We stopped  the ethanol  the night before,  and  then treated
with the ethylene dibromide  orally  on the following morning.  Because of the rapid
detoxification of ethanol there is a possibility there  was not sufficient ethanol
there to result in this modification.  Therefore, it really should be done and they
should be drinking the ethanol up to the time of  treatment.

The other thing is that because of the fact that they  are nocturnal maybe what we
ought to do,  really with that  particular  study, is treat them after midnight when
they would be actively eating and drinking.

There is a study in the  literature by Raddicke and Stemmer. It was an EPA-funded
study at Cincinnati.  In this study six  hundred ppm vinyl chloride with 5% ethanol in
the drinking  water resulted in a shortened latent period and an increased number of
liver hemangiosarcomas.  We  used their method  except that, again,  you  have to
have a point at which you can cut off the  ethanol and that probably  is a problem,
but there was  no difference  and  we  did  the  same thing, 24 animals two time
intervals with no differences in tissue levels.
                                      4iC

-------
Dr. Bull, EPA:  How does a regulatory agency deal with an effect of a compound
that is produced only in conjunction with a second compound which is rarely used.

Dr. Plotnick, NIOSH:  With respect to the ethylene dibromide itself, since our study
as well as an NCI gavage study  reported in  1975, and a  more  recent inhalation
study at two other levels by NCI, have established that EDB is a  carcinogen in the
rat, for regulatory purposes  there  is no problem with that  one -  at least from
OSHA carcinogen policy guidelines.  There also  is a definition there that  anything
that enhances carcinogenicity or  acts in  a promoter type of a way is, also,  to be
considered a carcinogen.  However, while disulfiram is used industrially in rubber
manufacturing as an accelerator,  a related compound is used  as an accelerator or
as an antioxidant and  also is used in spraying fruit trees,  primarily in California.
There is alcohol intolerance in workers who spray their fruit trees in  that  area.
You have across-the-board regulatory problems there because some of it is EPA,
some  of it is OSHA, and a good deal of it because the major use of the tetraethyl is
antabuse or  related products, is FDA.
                                  411

-------
        IN DEPTH BIOCHEMICAL,  PHARMACOLOGICAL AND
      METABOLIC STUDIES ON TRIHALOMETHANES IN WATER
R.J.  Bull,  M.  Robinson,  T.J.  Brown,  F.L.  Mink,  R.D.  Lingg
           Health Effects Research Laboratory
           U.S. Environmental Protection Agency
                 Cincinnati,  Ohio  45268

                           and

                     Carrie Whitmire
                    Toxicology  Branch
             NCI/National Toxicology Program
                     Landow Building
                Bethesda, Maryland 00014

-------
                             INTRODUCTION







     The occurrence of trihalomethanes in drinking water represents




one of the most fundamental issues in environmental health.   These




chemicals are produced through reaction of chlorine used in  disin-




fection with organic material present in the source water (Bellar et




al. 1974; Rook 1974).  This discovery is currently forcing careful




consideration of trade-offs between 3 fundamentally different types  of




risk:









     1.   Carcinogenic risk associated with disinfection byproducts.









     2.   Risks associated with microbiologically-induced disease.









     3.   Non-cancer risks that may be associated with alternate dis-




          infectants to chlorine.









It is clear that a logical framework for making such trade-offs  does




not currently exist.  This is partially due to inadequate information




upon which to base estimates of risk. More fundamental is the lack of




a means of assigning value to acute vs. chronic effects of disease




causing agents.  In the final analysis, however, it is the quality of




relevant data that describes the dose response relationship  involved
                                    41J

-------
with the alternative risks which must be relied upon for decision




making.  The present project represents one attempt to provide an




adequate data base upon which carcinogenic risks associated with




disinfection byproducts may be estimated.









     The major evidence of carcinogenic risk associated with chlorin-




ation reaction products are several reports of chloroform-induced




tumors in experimental animals (Eschenbrenner and Miller, 1945; NCI,




1976; Roe et al.  1979)   These data do fairly clearly establish that




chloroform is carcinogenic in the mouse and the rat.  However, the




studies provide very little insight into the question of how these




data might be relevant to man.  It is obvious from the work of Roe et




al. (1979) that there are substantial strain as well as species differences




in response to chloroform.  Additionally, the dose-response informa-




tion which resulted from these studies has been extremely limited.









     The use of bioassay data in which maximally tolerated doses have




been utilized to ,ale rosl estimates at low doses has been criticized




on several grounds.  Perhaps the  oldest argument is that tumors might




arise from epigenetic mechanisms triggered by necrotic tissue damage.




Such a relationship has been observed specifically with chloroform,




although with limited numbers of animals/group (Eschenbrenner and




Miller 1945)   A second issue has been the appropriateness of extrapo-




lating data from the rat and mouse to the human when the extent of




chloroform metabolism in the three species differs considerably (Reitz
                                        414

-------
et al. 1978).  Third is the possibility that the metabolism of




chemical carcinogens may be dose-dependent;  raising a second question




about whether results obtained at maximally tolerated doses can be




extrapolated with confidence to environmental levels.









     The above problems are not trivial.   At the same time they are




not completely answerable because of the ethical limitations rightfully




placed on human experimentation.  Within that limitation the present




project addresses these issues.  The results presented here include




metabolic work being performed at the Health Effects Research Laboratory




of the U.S. EPA in Cincinnati and subchronic and chronic studies being




conducted under contract with Stanford Research Institute in Menlo




Park under the direction of D.C.L. Jones and T. Jorgensen.  These




efforts are expected to contribute to better estimates of the risks




posed by chloroform and the other trihalomethanes and to form the




foundation of better regulatory decisions.











                                METHODS







Animal Groups and Dose Levels









     Three hundred female B6C3F1 mice were obtained from Charles River




Breeding Laboratory and 250 male Osborne-Mendel rats from Camm Research




Institute.  Following two weeks of quarantine the animals were placed




on test at 6 weeks of age.  The animals were allocated to experimental
                                   41f

-------
groups by using a table of random numbers.   Cages were rotated on the

racks and racks were rotated within the room once each week throughout

the study



     The test groups,  levels of chloroform,  and number of animals

assigned to each group were as follows:
     Male Osborne-Mendel Rats
     Dose Level
       (ppm)
No.  of
 Rats
        0               40*
        0 (match cont.)  +30
      200                30
      400                30
      600                30
      900                30
     1800                30
Female B6C3F1 Mice
Dose Level          No. of
   (ppm)            Mice
    0               40*
    0 (match cont) +30
  200               30
  400               30
  600               30
  900               30
 1800               30
 2700               30
     •'-' Includes 10 for day 0 sacrifice.
     + The amount of water consumed by the high dose groups was
       determined daily and an equivalent amount given to the matched
       control group the ensuing day.
Clinical Chemistry



     Ten animals from each group were sacrificed at 30,  60 and 90 days

of the study.   Serum glutamate oxalacetate transaminase  (SCOT) and

lactic dehydrogenase (LDH) activity were measured in the mice using

the methods of Karmen (1955) and Wacker, et al.  (1956).   The clinical

chemistry measurements in the rats were measured according to

Technicon Manual Technical Publication # UA-3-0306B3,  March 1976 and

included blood urea nitrogen, lactic dehydrogenase, serum glutamate
                                    41C

-------
oxalacetate transaminase, serum glutamate pyruvate transaminase,  and




serum triglycerides.









     At the time of sacrifice complete necropsies were performed




according to the Guidelines for Carcinogen Bioassay in Small Rodents




(NCI-CG-TR-1, Technical Report Series #1, February 1976).   Major




organs and all suspected tumors and gross lesions were examined




microscopically.









Chloroform analysis









     Drinking water.   Pesticide-quality chloroform from Matheson Coleman




Bell all of the same lot number was used throughout this study.  Chemical




analysis indicated the presence of 30 ppb diethyl carbonate (DEC) in




the commerically obtained chloroform.  Therefore, the chloroform was




distilled just prior to the twice-weekly preparation of the dosing




solutions.  Fresh solutions of chloroform were prepared on Mondays and




Thursdays and the old solutions were discarded.  When the solutions




were thoroughly mixed, they were transferred to animal water bottles




via a syringe-activated teflon siphon.  Two samples for each chloroform




level were removed for analysis by the method of Bellar and Lichtenberg




(1974).  Sampling was done both on freshly prepared and discarded




water at the twice-weekly change in the animals' drinking water.   Second




Analysis of solutions being discarded revealed an average of 92.0 .1




1.4% of the initial chloroform concentrations.
                                      417

-------
     Serum.  Chloroform levels in serum were assayed using the purge




trap method of Bellar and Lichtenberg (1974) as modified by Peoples,




et al. (1979).









Organ Fat Analysis









     The procedure used to perform the assay of lipids in the rat




kidney and mouse liver in this study was as follows.  The organ tissue




was tamped dry on the exterior and weighed on an analytical balance.




The tissue then was homogenized with 4 ml of high purity water using a




Tektron Inc., polytron homogenizer.   The homogenate was added to a




separatory funnel containing 49 ml chloroform:  methanol, 2:1, the




mixture shaken for 30 seconds, and 8 ml of 0.018 N H2S04 added.   The




resulting mixture was shaken for 15  seconds, and the total contents




were added to a 15 ml centrifuge bottle, which  was spun at 2000 rpm




for 20 minutes.  The resulting suspension contained two layers separated




by a thin white protein disk.   The upper aqueous layer was drawn off




by suction, and a 20 ml fraction of  the bottom  layer (chloroform) was




evaporated to dryness in a tared 3-g test tube  in a water bath at




57°C.   Nitrogen gas was used to remove final traces of the solvent.




The dry test tube was placed in a dessicator overnight, and the net




weight, representing lipids, was determined.
                                    41£

-------
Metabolic work.










Chemicals and Solutions.  He-chloroform, 2.1 mCi/mmole and 99% minimum




radiochemical purity (mrcp); Hc_dibromochlorome thane, 2mCi/mmole and




98% mrcp; and Hc_t>rOmform, 2.1 mCi/mmole and 99% mrcp were purchased




from radiochemical suppliers.  Dosing solutions were prepared by mixing




the appropriate unlabeled trihalomethane (THM) (all 97% pure) with the




HC-THM  in corn oil  (Fisher Laboratory grade): for the rat, 100 mg-




THM/ml with a radioactive concentration of 16 uCi/ml, for the mouse




150 mg-THM/ml with a radioactive concentration of 32 uCi/ml.









Animals  and Dosing.  Male Sprague-Dawley rats (Charles River) were




fasted for 16 hours  prior to dosing.  The THM was administered, 100




mg/kg, in a single dose by  intragastric intubation.  The rats were




immediately placed,  individually,  in glass Roth-type metabolism cages




which were equipped  with urine-feces separator-collectors cooled to 4°C.




Fasting  from  food was  continued  throughout the 8 hour sampling period,




but water was allowed  ad libitum.









     The mice, B6C3F1  were  given single  150 mg/kg per oral  doses of




HC-THM  by intragastric intubation following a 16h fast.  Five mice




were placed in a  single metabolism cage.  Other procedures  were identical




with those used for  the rat.









Sampling of expired  air, urine and feces.  Room air was drawn  through




the chambers  by mechanical  pumps at a rate of 500 ml/min.   A series of
                                      41

-------
two glass impinger traps was placed between the cages and the pump to




collect expired ^C-THM and ^CC>2 from the air stream.  The first




trap, for collecting unchanged -^C-THM  contained as a solvent 100 ml




of a 9:1 mixture of ACS Grade Xylenes (Fisher Scientific) and scin-




tillation grade 2-methoxyethanol (Eastman).  The second trap, for




collecting -^CC^, contained 100 ml of a 5N ethanolamine (Eastman




scintillation grade) in scintillation grade 2-methoxyethanol.










     The efficiency of the expired air traps for separating and collec-




ting l^C-THM and 1^CC>2 was tested by first evaporating a simulated




dose of  l^C-THM  (dissolved in methanol) from an open dish in the metab-




olism cage.  Collection of l^C-THM by the  first trap was quantitative




at the completion of evaporation, but a small loss, 1.3-5.2%, of l^C-




THM was  experienced by continued pumping up to 8 hours.  In the second




step of  the test, -^CC^ from a calibrated  compressed air tank was




sparged  through  fresh traps for 10 minutes.  The 1^C02 was quantitatively




trapped  by the second trap:  there was no  -'-^C02 in  the first trap.




The  air  pumps were  then turned on and room air was  drawn through the




traps for 8 hours.  No loss of ^C02 occurred.
      Samples were  collected  from  the  traps at 0.5,  1.0,  1.5  and  2.0




hours  and  then  at  hourly  intervals up  to  8 hours.   During  the  runs




with  the rats,  0.5 ml  aliquots were taken from  the  expired air traps




at  these intervals and  assayed for radioactivity  by liquid scintillation




counting (LSC).  The sparging solutions remained  in the  traps.   In  the




experiments with the mice,  fresh  solution was placed in  the  traps at




each  sampling interval.

-------
     Urine and feces were collected at 2, 4 and 8 hours if available.




Urines were assayed for ^C by LSC: 0.5 ml aliquots were added to 12




ml Scintiverse (Fisher Scientific Co.).  Negligible amounts of ^C




were found in the feces, so after initial screening feces were dis-




carded.
Determination of   C in selected tissue and organs.    At 8 hours,




animals were sacrificed by anesthetizing with Nembutal.  After with-




drawing a blood sample from the inferior vena cava, selected tissue




and organ samples were excised and dissolved in solublizer.  The preparation




of the solublizer and method for determination of l^C in tissue samples




was that described by Dent and Johnson (1974).  Internal standardization




with 14c_ toluene was used to determine counting efficiency.











                                RESULTS









     As a result of experiments which are reported in the present




paper, dose levels were selected for the chronic study now underway.




For purposes of comparing the dose levels in the present work and the




chronic study with those used in the NCI bioassay (NCI, 1976) calcu-




lated intakes of chloroform are given in Table 1.  As can be seen the




doses obtained by rats and mice at 900 and 1800 mg/liter approximate




the doses administered in the NCI bioassay of 90 and 180 rag/Kg in the




male rat.  The 1800 mg/1 dose to the female B6C3F1 mouse approximates




the lowest time weighted dose of 238 mg/Kg.  (The NCI doses were
                                     421

-------
administered for 5 days a week by gavage whereas the present study




employs continuous exposure increasing the dose levels in the present




study relative to the NCI doses).









     Included in Table 1 are the number of animals included per exposure




group in the chronic study.  This design has been chosen to better




estimate the tumor incidence at the lower doses.  The biological and




clinical data presented in this paper are derived from the subchronic




study used to design the chronic study or from the interim sacrifices




conducted over the first 6 months of the chronic study.   In general




these measurements have been performed on groups of 10 animals per




time interval.









     Table 2 gives the average serum chloroform concentrations in rats




treated with different concentrations of chloroform in their drinking




water   Two trends are noted in the data.  First of all  there tends to




be a consistent increase in the levels of chloroform measured in the




serum of the control animals over time.  The increase of concentrations




over a 180 day period would argue for a serum half-time  of several years




rather than the   2 hr usually reported (Brown et al. 1974).  These




data suggest extraneous sources of chemical in serum which are measured




as chloroform.  Alternate sources of chloroform have not been found to




account for this increase (i.e. food and room air)   Despite this




pattern there appeared to be significant dose-related increases of




serum chloroform levels at 30 and 90 days of exposure.  The scatter of




this data prevented significant relationships with dose to be




consistently observed despite the use of early morning samples.








                                   42.

-------
     During periodic sacrifices clinical chemistry measurements have




been made in rats on 22 parameters and in mice for two.  For purposes




of this interim report we have chosen to include only those parameters




which are referrable to kidney or liver damage.









     In Table 3 the effects of chloroform treatment on blood urea




nitrogen (BUN) are displayed.  As can be seen there appears to be a




dose-related increase (evident primarily at 1800 mg/1) of BUN with




chloroform treatment.  However, this does not appear to be attributable




to chloroform since the control group matched for water intake shows




the same or even greater increases in BUN at all time points.









     Tables 4, 5 and 6 display the results of serum enzyme activities




in the rat that would be predicted to increase with liver damage (and




to some extent other organs), SCOT, SGPT and LDH respectively.  There




are no significant changes in SCOT, a transient increase in SGPT at 30




days of exposure at 1800 mg chloroform/1, and a decrease in LDH at 60




days.  This decrease in LDH activity has also been observed at the 180




day sacrifice of the chronic study (data not shown).









     Serum triglyceride concentrations were decreased by chloroform




treatment (Table 7)   This was a consistent observation at all 3




sacrifice times at a concentration of 1800 mg chloroform/1 and the




changes were not reflected in animals matched for drinking water




consumption.  A similar decrease in serum triglyceride levels was

-------
observed in animals exposed to 1800 mg chloroform/1 in the chronic




study (data not shown)










     Because of the limited availability of serum from the mouse it




was not possible to perform extensive clinical measurements.  Since




the B6C3F1 mouse has been shown to develop hepatocellular carcinomas,




SCOT and LDH were the parameters measured in mouse blood (Table 8).




As with the rat, the only statistically significant and consistent




changes in these enzyme activities were decreases in the normal




activity   In both cases the decreases were observed primarily at 1800




and 2700 mg chloroform/1.  In the case of LDH the decrease appeared to




be complicated by an increase in activity induced by restricted water




intake in the matched control.  It is not possible to determine the




extent to which the increase in activity by restriction of fluid in-




take may have modified the effect of chloroform at lower doses.










     As a result of appearance of increased liver fat in female mice




exposed to chloroform in the subchronic test (data not shown), the




chronic study was modified to allow liver fat content to be measured




in both rats and mice over the entire dose range.  Data from the 90




and 180 day sacrifices are shown in Table 9.  At both times definite




increases in liver fat were observed in female B6C3F1 mice.  The lowest




effective dose was 400 mg/1 at 90 days and 200 mg/1 at 180 days.  In




contrast, no increase in liver fat was observed in male Osborne-Mendel




rats at 90 days of exposure; the only significant increase being observed




after 180 days of exposure to 1800 mg/1.
                                      424

-------
     Histopathological evidence of fatty infiltration of the liver was




observed in the mice at all sacrifice periods of the subchronic study




(Table 10).  Despite the limited number of animals at each time and




dose, a rather clear relationship of this effect with chloroform




exposure could be seen.  No histopathological evidence of fatty infiltration




of the liver has been observed in the male Osborne-Mendel rat.









     No other histopathology could be clearly related to chloroform




exposure.  Particular attention was paid to the development of kidney




pathology in the male rat because this site was involved in malignant




changes in the NCI bioassay.









     The comparative metabolism of trihalomethanes in the mouse and




rat are currently under extensive study at the Health Effects Research




Laboratory of USEPA in Cincinnati.  Table 11 summarizes the overall




metabolism of each of the four trihalomethanes commonly found in




drinking water, trichloromethane  (TCM or chloroform), bromodichloro-




methane (BDCM), dibromochloromethane (DBCM) and tribromomethane (TBM,




bromoform).  In general, it can be seen that all of  the trihalomethanes




are more extensively metabolized  in the mouse than in the rat with 4




to 9 times higher proportions of  the dose being excreted as expired




CC-2.  This is paralleled by the appearance of approximately twice the




level of metabolites excreted in  the urine of the mouse.  There is




also a relatively higher retention of label in tissues  (with  the




exception of BDCM) in the mouse.

-------
     It is curious to note that the metabolism of chloroform parallels




that of bromoform more closely than the mixed trihalomethanes.   The




mixed trihalomethanes are metabolized to a much greater extent  by both




species (14 and 18% conversion to C02 in the rat and 81 and 72% in the




mouse for BDCM and DBCM, respectively).  The dose-response characteristics




of trihalomethane metabolism and identity of the metabolites are now




being studied.









     To obtain an estimate of the relative amounts of reactive  inter-




mediate available for binding macromolecules our laboratory has been




exploring the use of covalent binding carcinogens to hemoglobin that




has been demonstrated by Ehrenberg and coworkers (Osterman-Golkar et




al., 1976).  Experiments with chloroform have demonstrated a linear




binding of [^-^C]-chloroform to hemoglobin with single doses ranging




from 0.012 to 120 mg/kg body weight (Pereira and Chang, 1980).   The




binding of t^C]-chloroform to hemoglobin following a dose of 1.2




mg/kg has been compared in 3 strains of mouse and rat.  The data so




obtained is shown in Table 12.  In general it can be seen that  the




covalent binding of chloroform to hemoglobin is generally greater in




the rat than in the mouse.  Specifically, the degree of hemoglobin




binding in the Osborne-Mendel rat exceeds that observed in the  B6C3F1




mouse by a slight but insignificant amount.
                                        426

-------
                              DISCUSSION









     This preliminary assessment of factors which might influence




across-species extrapolation of carcinogenicity data for chloroform




can only lead to tentative conclusions.   Despite the incomplete nature




of the results, some patterns seem to be emerging.









     First, clinical parameters which have been developed for the




purpose of detecting acute liver damage appear to be of rather limited




usefulness in judging liver damage induced by chloroform on a chronic




basis.  Rather than the increased levels of these enzymes classically




found in serum following acute liver injury, the pattern observed with




some consistency in the present study was a decrease in these activities




This might be a result of the specific type of injury observed, fatty




infiltration of the liver, which occurred with little evidence of




gross necrosis.









     Despite the failure of the serum enzyme measurements  to detect




the injury, it is quite clear that the female B6C3F1 is much more




sensitive to the development of a fatty liver than the Osborne-Mendel




rat.  The doses at which this effect is noted are within the range and




extend much below those used in the NCI bioassay for chloroform (NCI,




1976)   Consequently, it is reasonable to conclude that this effect




was present in the study and perhaps even exacerbated by the use of




bolus doses resulting from gavage treatment.  Similarly, it is
                                    42'

-------
apparent that the Osborne-Mendel rat was much less sensitive to liver




injury induced by chloroform than the mouse.   It is difficult at this




time to determine how the reduction in serum triglyceride levels might




be related to the marginal but highly significant increase in liver




fat noted in the rat at the highest dose.  The coupling of these two




effects would suggest a relatively specific interference with lipid




metabolism.










     In general, these data support the contention that the ability of




chloroform to produce liver tumors is not separable from its ability




to damage the liver.  It is known that other treatments that produce




fatty livers, such as choline-deficient diets (Shinozuka, et al. 1978)




promote the development of liver tumors   No such relationship could




be drawn between chloroform-induced damage to the kidney and tumor




development in the Osborne-Mendel rat.










     It is premature to judge the contention that a difference in




metabolism of chloroform between species can account for different




susceptibilities to chloroform carcinogenesis (Reitz et al.  1978)




Previously reported differences in the fraction of chloroform metabo-




lized by mouse and rat (Brown et al. 1974) were confirmed in the




present work.  However, it appears that the levels of activated




metabolite which reach hemoglobin in the circulating red cell are




essentially the same in the male Osborne-Mendel rat and the B6C3F1




mouse.  This latter data may or may not accurately reflect the degree
                                  426

-------
to which chloroform covalently binds  to tissue  macromolecules,




particularly DNA, and the extent to which it  acts  as  an  initiator.  On




the other hand, the levels of activated intermediate  available  for




such interactions are even more tenuously related  to  gross measures of




metabolism upon which the argument is developed.   The extent  to which




chloroform interaction with DNA might differ  between  the two  species




is currently under investigation in our laboratory (Lin  and Pereira,




personal communication).
                                     429

-------
References







Bellar, T.A.,  Lichtenberg,  J.J.  and Kroner,  A.D.  (1974)  The occurrence




of organohalides in chlorinated drinking water.   JAWWA 66:703-706.









Brown, D.M., Langley, P.F., Smith, D. and Taylor, D.C. (1974)  Metabolism




of chloroform I.  The metabolism of [^C] chloroform by different




species.  Xenobiotica  4:151-163.









Dent, J.G.  and Johnson, P.  (1974)  "Solublization of Mammalian Tissue




for Scintillation Counting", Chapter II.  in: Liquid Scintillation




Counting,  (M.A. Crook and P  Johnson, eds.).  Heyden and Son, London.









Karmen, A.  (1955)  A note on the spectrophotometric assay of glutamic-




oxalacetic  transaminase in human blood serum.  J. Clin. Invest.




34:131-133.









Eschenbrenner, A.B. and Miller, E. (1945)  Induction of hepatomas in




mice by repeated oral administration of chloroform with observations




on sex differences.  JNCI 5:251-255









NCI, 1976.  Report on the carcinogenesis bioassay of chloroform.  Car-




cinogenesis Bioassay and Program Resources Branch, Carcinogenesis




Program, National Cancer Institute Bethesda, Maryland, March 1, 1976.









Osterman-Golkar, S., Ehrenberg, L., Segerback, D. and Hallstrom,  I.




(1976)  Evaluation of genetic risks of alkylating agents II.  Haemoglobin




as a dose monitor   Mutation Res.  34:1-10.






                                     43C

-------
Peoples, A.J., C.D. Pfaffenberger,  T.M. Shafik, and H.F. Enos.




Determination of volatile purgeable halogenated hydrocarbons in human




adipose tissue and blood serum:   Bull. Environ. Contam. Toxicol.  23:




244-249 (1979).









Pereira, M.A. and Chang, L.W. (1980)  Binding of chloroform to mouse




and rat hemoglobin as a dose monitor.  J. Environ. Path & Toxicol. 3.




In press.









Reitz, R.H., Gehring, P.J. and Park, C.N. (1978)  Carcinogenic risk




estimation for chloroform:  An alternative to EPA's procedures.  Fd.




Cosmet. Toxicol. 16:511-514.









Roe, F.J.C., Palmer, A.K., Worden,  A.N. and VanAbbe, N.J. (1979)  Safety




evaluation of toothpaste containing chloroform. I.  Long-term studies




in mice.  J. Environ. Path. & Toxicol. 2:799-819.









Rook, J.J. (1974)  Formation of haloforms during chlorination of natural




waters.  Jour. Water Treat. Exam. 22:234-243.









Shinozuka, H., Lombardi, B., Sell,  S. and Immarino, R.M. (1978)  Early




histopathological and functional alterations of ethionine liver carcinogenesis




in rats  fed  a choline deficient diet.  Cancer Res. 38:1092-1098.









Wacker, W.E.C., Ulmer, D.D., Vallee, B.L.  (1956)  Metalloenzymes and




myocardial infarction.  New England J. Med.  255:449-456.







                                   43J

-------
     Table 1.
Calculated Doses of Chloroform Corresponding to Levels
   Administered in Drinking Water to Rats and Mice
                  Osborne  Mendel
                    Male Rats
                                        B6C3F1 mice
Chloroform
mg/1
0
200
400
900
1800
M0b
n
330
330
150
50
50
50
Calculated dosea
mg/kg
0
19
38
75
151
-
n
430
430
150
50
50
50
Calculated dose
mg/kg
0
34
65
130
263
-
aCalculated from average water consumption/kg body weight in subchronic study

°MO = Group matched with 1800 mg/1 group for water consumption.

-------
     Table  2.   Serum Chloroform Levels in Rats Treated
                 with Chloroform in Drinking Water
Chloroform
mg/1
Control
Matched Control
200
400
600
900
1800
30 Day
0.60
0.89a
0.86a
1.26
1.12b
7.18a
4.17b
ug/1
60 Day
1.45
0.96
1.07
0.92
1.15
1.34
5.89
Blood
90 Day
6.69
-
13.0
13.5
-
10.9
40. 8a
180 Day
9.75
10.15
8.76
7.05
-
8.79
9.95
aSignificantly different  from corresponding control by one  way
 ANOVA and t-test  at  P <  0.05.
bSame as a except  P<0.01.
                                     433

-------
   Table 3.   Effect  ot  Chloroform on Blood Urea Nitrogen of the Rat

                                  BUNa
Chloroform
mg/1	30  days	60 days	90 days
0
200
400
600
900
1800
M0d
20 ±
22 +_
23 ±
22 ±
25 ±
26 ±
28 +
2
3
3C
4
10
2C
10C
22
23
22
23
23
25
26
± 2
± 2
± 3
± 1
± 1
±3b
± 2°
22
22
22
23
22
25
25
± 1
± 2
± 2
± 2
± 3
± 2C
+ 5
     I S.D.  ten animals

^Significantly different  from control by ANOVA and t-test P < 0.05

cSame as b with P < 0.01
                                     434

-------
     Table  4.   Effect of Chloroform on Serum Glutamate
                 Oxalacetic Transaminase in the Rat
Chloroform
mg/1
0
200
400
600
900
1800
M0b
30 day
174 ±
202 ±
144 ±
161 ±
151 ±
251 ±
166 ±
90
74
37
51
67
125
64
SCOT Activity3
60 day
138
129
117
122
124
104
142
± 48
± 28
± 24
± 37
± 46
1 30
± 35
90 day
155 ±
107 ±
110 ±
119 ±
107 ±
153 ±
183 ±
89
17
35
39
18
51
81
   serum jf S.D.  ten animals


= Group matched  with 1800 mg/1 group for water consumption.
                                 43?;

-------
     Table  5.  Effect of Chloroform on Serum Glutamate
                 Pyruvate Transaminase in the Rat

                                        SGPT Activity3
Chloroform
mg/1
0
200
400
600
900
1800
MOC
30 day
63 ±
83 ±
63 ±
60 ±
83 ±
112 ±
62 ±
21
33
11
17
64
53b
16
60 day
65 ±
66 ±
64 ±
66 ±
65 ±
65 ±
65 ±
18
13
16
17
13
20
11
90 day
62 ±
59 ±
72 ±
64 ±
61 ±
84 ±
74 ±
7
7
33
14
16
37
21
       serum JL S.D.  ten animals

"Significantly different  from control by ANOVA and t-test P ^0.05.

CMO = Group matched  with  1800 mg/1 group for water consumption.
                                   436

-------
     Table 6.   Effect  of  Chloroform Treatment on Serum Lactic
                    Dehydrogenase Activity in the Rat

                                       Serum LDH Activity
Chloroform
mg/1 drinking water
0
200
400
600
900
1800
M0d
30
1593
1651
1270
1577
1275
1616
1484
day
±504*
± 361
± 225
± 303
± 360
± 482
± 287
60
1370
1390
1190
1079
1014
676
1467
day
± 442
± 368
± 349
± 328
± 461
± 269C
± 319
90
2016
1859
1547
1091
818
1016
1383
day
+
+
+
+
+
+
+
1582
700
455
300
405
391
592
amu/ml 1. S.D. ten animals

^Significantly different from control  P < 0.05 by ANOVA and t-test

cSame as b but P< 0.01.

dMO = Group matched with 1800 mg/1  group  for water  consumption.
                                  437

-------
     Table  7.  Effect of Chloroform on Serum Triglyceride
                      Concentrations in the Rat
Chloroform
mg/1
0
200
400
600
900
1800
M0d
Serum Triglycerides3
30 day 60 day
87 ±
80 ±
96 ±
60 ±
313 ±
43 ±
123 ±
58
23
98
10
832
9b
210
77
68
86
81
62
36
53
± 26
± 16
± 21
± 30
± 25
± 15C
± 13b
90
78
93
91
80
93
38
84
day
± 33
± 43
± 53
± 24
± 18
± 21C
± 46
amg% i S.D.  ten animals

bSignificantly different  from control by ANOVA and t-test P   0.05

cSame as b except  P    0.01.

dMO = Group matched with  1800 mg/1 group for water consumption.
                                   438

-------
Table 8.  Serum Glutamate Oxalacetate  Transaminase  (SCOT) and Lactic
             Dehydrogenase (LDH)  Activity  in  Female Mice Receiving
                      Chloroform  in Their  Drinking Water
                   SCOT Activity
LDH Activity
Chloroform
mg/1
0.
200
400
600
900
1800
2700
M0d
30
328
225
233
196
246
298
156b
522
60
623
400
414
257
383
136b
619
349
90
353
609
201
392
234
176b
190
620b
30
909
738
767
601
913
898
653
1535b
60
1010
996
965
1114
1018
548C
1318
1413b
90
1222
1520
767b
1197
1160
731C
765C
1506
aAverage values for 10 animals expressed as mu/ml
bSignificantly different from control by ANOVA and t-test  P^CO.05
cSame as b except P< 0.01.
"MO = Group matched with 1800 mg/1 groups for water consumption.
                                   430

-------
Table 9.   Liver Fat  Content  in Mice and Rats Treated
                with Chloroform  in Drinking Water
                               Percent Liver Fata
Chloroform
mg/1
0
200
400
900
1800
90
3
3.
3
4
6,
B6C3F1
Mouse
Days 180 Days
.33
.45
.89C
.51°
.36C
5.82
7.93b
6.77
7.11b
10.40C
Osborne -Mendel
Rat
90 Days 180 Days
3.32
3.31
3.20
3.58
3.46
4.49
4.50
4.59
4.77
5.13C
aAverage value for 10  animals at each dose and time.

bSignificant at P<0.05  by ANOVA and t-test.

Significant at P<0.01  by ANOVA and t-test.
                               44C

-------
Table 10.    Histopathological Evidence of Fatty Infiltration
                       of Livers of B6C3F1 Mice

                        Number of animals with centrilobular  fatty change3
Chloroform
mg/1
0
200
400
600
900
1800
2700
30 Day
0
0
3
0
2
5
6
60 Day
0
0
0
0
0
0
5
90 Day
0
0
0
0
0
4
2
Total
0
0
3
0
2
9
13
aOut of 10 animals at each  dose group and time point.
                                    44J

-------
     Table  11.   Comparative Distribution of Labelled Carbon Eight Hours
                 Following Oral Administration of -"-^C-Trihalomethanes
                      to  the Sprague-Dawley Rata and B6C3F1 Mouseb
TCM
Rat
% C02 Expired
% Unmetabolized
Expired
% 14C Urine
% 14C Total Organ
Total 14C Recovery
6

64
2
4
78
.5

.8
.6
.3
.2
Mouse
49.

26,
4,
13.
94.
.55

.05
.91
.45
.47
BDCM
Rat
14

41
1
5
62
.3

.7
.4
.3
.7
Mouse
81.

7.
2.
3.
92.
.20

,18
.17
,18
71
DBCM
Rat
18.

48.
1.
2.
70.
Mouse
2

1
1
9
3
71.

12.
1.
5.
91.
58

31
90
02
63
TBM
Rat
4.

66.
2.
5.
78.
3

9
2
5
9
Mouse
39.68

5.70
4.62
12.18
62.23
aAverage not less  than  5 animals per group - dose administered 100 mg/kg
"Average of 4 groups  of 5  animals each - dose administered 150 mg/kg
                                     442

-------
     Table 12.   Chloroform Bound  to  Hemoglobin
                  in Mice  and  Rat Strains3
     Animal                        Amount  Bound
     Strain                        pmoles/Hb'5
1.    Rat

     a.    Osborne-Mendel                 109  +  19
     b.    Sprague-Dawley                 136  +_  10
     c.    Fisher-344                    152  +  14

2.    Mice

     a.    Swiss (CFN)                     85+7
     b.    B6C3F1                         96  +_  8
     c.    Swiss (SENCAR)                 104  +  10
a - Data of Pereira and Chang,  1980
b - The values in the table represent  the mean  +_
    standard error for 6 animals  in each group.
                               443

-------
                                  Discussion
Dr. Keefer, NCI:  I had always imagined that chloroform was a lot less soluble in
water than you indicated, but I have never seen any data. I am wondering whether
you had to use any special methods to get it into solution at almost 2 milliliters per
liter and,  more  importantly, whether you have  some data  showing  that  the
volatilization from the water solution on feeding was not contributing to loss from
the water and inhalation exposure as well.

Dr. Bull, EPA:  The room air has been  monitored, and there is no real indication of
that, but more basic to the problem is that the water bottles are changed twice
weekly, and the amounts of chloroform monitored on a weekly basis.  The levels of
chloroform in those water  bottles  compared to the initial  levels  were 92 plus or
minus  about 2%.  This was partially  attributable  to the use of a double balled
watering bottle which seems to cut down on the loss to the ambient environment
although there is  obviously still a small loss to the head space. In addition, there is
a very high turnover in the room air which probably accounts for the inability to
document any appearance of chloroform in the room air.

Regarding  solubility,  things which are called  non-soluble in  the  Handbook in
Chemistry and Physics are  often soluble to several percent.  The limit of solubility
of chloroform in water is  about  0.5 % at  room temperature.  Consequently, no
extraneous agents were needed to  get chloroform into solution, even at the highest
concentrations employed.

Dr. Keefer, NCI:  How did you do the measurement of the concentration in water?

Dr. Bull, EPA:  By the purge and trap method; standard methodology for looking at
trihalomethanes in water.

Dr. Keefer, NCI:  I am not sure what that is.

Dr. Bull, EPA:  It is a GC-mass spectrometric method.

Dr. Gregory, CPSC:  I really object to  the last sentence  in your  abstract that says
that such a condition is well known as a promoter of liver tumors in other species
as well.  I would object to the way it  is put. I certainly don't object to the fact
that many substances which produce fatty livers are  promoters.   I object  to the
indication that it is the condition itself that is the promoter, that is not true.

Dr. Bull, EPA:  That would be true, I  think. However, maybe it is a little  too all
inclusive.   Conditions  other than chemical  exposure, for example  a choline
deficient diet,  also produce a fatty liver and promote tumor developoment.  It is
difficult to see this procedure  acting as an initiator.

Unidentified Speaker:  I think what he is trying to say is the fatty liver does not
imply cancer.

Dr. Bull, EPA:  I did not mean to imply that either.

Dr. Chandler,  NIOSH:   The observation  of  a linear  increase  in hemoglobin
alkylation  or derivative  formation over five orders  of  magnitude is a very rare



                                       444

-------
observation, at least in my experiences.  Any biochemical phenomenon which would
be linear over five orders of magnitude leads one to wonder exactly what is being
measured with regard to metabolism.  I have three quesitons:  Are you certain that
its  reactive  intermediates are  being generated  through metabolism  that  are
alkylating hemoglobin  or is it a spontaneous reaction between hemoglobin and the
trihalomethane?  Also, have  you characterized  the nature  of the derivative of
hemoglobin?  It is established that macrophages can activate  many compounds. Is
it  possible that circulating  macrophages  could  be  the source of  the reactive
metabolism and not the liver or kidney?

Dr. Bullj EPA: That last question in particular is the reason why I think we have to
compare the formation of DNA adducts, or  at least macromolecular binding of the
chloroform in target tissues, with alkylation of hemoglobin.  The  source of the
alkylating group may be  all tissues or selectively from some.  We are hoping that
alkylated hemoglobin will be an integrator of all kinds of tissue, and it is certainly
possible  that  the  macrbphage might be responsible  for at  least  some of  that
activity.  The critical  issue at this point in  time is to see if that really accurately
reflects the degree to which the material is  covalently bound in target tissues.

I think it is  pretty clear at this point in  time that  a covalent  binding  with
chloroform is involved -  basically because of the way the  product is isolated.
However, we have not actually identified the product as of this  moment. That is
being worked on. Another argument is that probably the half  life of the adduct to
hemoglobin  approximates  the  half life of  the red  cell.  Therefore, if you were
looking at, say, carbon monoxide as  a product binding to heme  for example, you
would not expect that half life.
                                      445

-------
     FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:
        PROGRESS ON JOINT ENVIRONMENTAL AND
            OCCUPATIONAL CANCER STUDIES
               Wednesday Afternoon, May 7
METHODOLOGY/EXPERIMENTAL MODELS SESSION (CONTINUED)
                 SESSION CHAIRPERSON
                     Dr. C. C. Lee
              Environmental Protection Agency
                            446

-------
      Chronic Animal  Inhalation Study of Short
               (<5 yro) Asbestos Fibers

                   Progress Report
                         by
                  Stanley F. Platek

                   David  H, Groth

    Division of Biomedical and Behavioral Science

National Institute tor  Occupational Safety and Health

                  Taft  Laboratories

                4676  Columbia Parkway

               Cincinnati, Ohio  45226
            Contract Number:   210-77-0151

 International Research  and Development Corporation

              Mattawan,  Michigan  49071
                          44V

-------
         CHRONIC INHALATION STUDY OF SHORT (<5 ym) ASBESTOS FIBERS




                               Progress Report




An animal inhalation study was initiated in 1977 to study the chronic




biological effects of inhalation of short chrysotile asbestos fibers.  Rats




and monkeys were exposed for 7 hours/day, 5 days/week for 18 months to




specially prepared chrysotile.  Based upon daily chamber measurements, the




mean concentration of fibers in the chamber air was less than 1 rag/m  .  By




phase contrast and electron microscopy, the ratio of the number of chrysotile




fibers/cc <5 ym in length to the fibers >5 ym was established at approximately




265:1.  Rats were autopsied for examination 1, 3, 6, 12, and 18 months after




initiating exposures.  Histopathological examinations of the lung tissue have




so far revealed little or no pathological reaction to the inhaled asbestos.




Although asbestos fibers could not be seen in lung tissues by light microscopy,




they were seen in alveolar macrophages when examined by electron microscopy.




Six months after the last exposure date, i.e., 24 months after initiating




exposures, the remaining rats will be sacrificed for examination.  The monkeys




will be maintained and observed for signs of latent pulmonary disease for




approximately an additional seven years.

-------
INTRODUCTION









Asbestos has been implicated by numerous investigators  as playing a major




role in the debilitating human diseases of bronchogenic carcinoma,  mesothelioma,




and pulmonary fibrosis (1,2,3).  Previous studies  have  focused  on the inhalation




of asbestos fibers greater than 5 ym in length and largely disregarded the




effect of asbestos fibers less than 5 ym in length.









The purpose of this project is to study the relationship of exposure to




chrysotile asbestos fibers less than 5 ym long and the  development of chronic,




asbestos-associated diseases.  Chrysotile asbestos was  employed in this project




because more than 90% of the asbestos used industrially and commercially is




chrysotile (4,5).









MATERIALS AND METHODS









Short Fiber Preparation









The chrysotile used in this study was type 7TF1 chrysotile obtained from the




Johns-Manville Sales Corporation in Denver, Colorado.   Short asbestos fibers




were prepared by drying 500 grams of the fibers in an oven at 191°C.  After




cooling, the chrysotile was placed in a cylindrical ceramic (Burundum) ball




mill (7" x 8'j") with 120 cylindrical Burundum pellets,  each pellet measuring




13/16" x 10/16".  The mill was rotated at 73 rpm for 24 hours after which time




the chrysotile was removed and again dried in an oven at 191°C for 24 hours.  The




asbestos was then cooled and stored in tightly secured, double plastic bags.  By
                                        440

-------
this method of preparation,  99.98% of the resulting chrysotile fibers, as




viewed and sized by electron microscopy,  were less than 0.6 ym in diameter




and about 20% were longer than 5 urn.









In addition to the large and short asbestos fibers, the ball-milling process




produced amalgamated "balls" of chrysotile fibers which measured up to 10 ym




in diameter.  Figures 1, 2,  and 3 are scanning electron micrographs (SEM) of




the ball-milled asbestos shotting the  various sizes of fibers and "balls."









It has been reported that ball-milling is not a completely satisfactory method




for preparing short asbestos fibers for biological studies (6,7), however ball-




milling was the best method  available for this project in which the production




of more than 100 pounds of short asbestos fibers was required.









Animals









The experimental design incorporated  300  male Sprague-Pawley rats and 20 male




cynomolgus monkeys.  Each animal was  individually identified and randomly




assigned, half the rats and  monkeys to an asbestos exposure chamber and the




other half of the rats and monkeys to a control inhalation chamber.  Each rat




and monkey was individually  housed and provided food (Purina Basal Rat Diet




and Purina Monkey Chow, respectively) and water ad libitum, except during the




hours of inhalation exposure.
                                      45C

-------
                                n
All animals were exposed in 12 m  inhalation chambers  for 7  hours/day,  5  days/week


for 18 months.  At the end of the 18 month exposure period,  the  surviving rats,


both those exposed to chrysotile and the controls,  were scheduled  for a 6 month


post-exposure observation period of which at the time  of this  report, 5 months


have elapsed.




Table 1 shows the serial sacrifice schedule for the exposed  and  control rats


at 1, 3, 6, 12,  and 18 months after the initiation  of  the asbestos exposure.


A terminal sacrifice of all remaining rats will be  made at month 24.




                                   TABLE 1


                           EAT SACRIFICE SCHEDULE
Sacrifice Interval and No. of Rats Selected

Exposed rats
Control rats
1 mo.
5
5
3 mo.
15
15
6 mo.
15
15
12 mo.
15
15
18 mo.
15
15
24 mo.
85
85
The experimental design designates that all monkeys,  10 control  and 10 asbestos


exposed, will be maintained and observed for at  least seven years  after  the last


exposure day.




Animal Tissue Diagnostic Tests




Five rats from the exposed group and five rats from the control  group were


sacrificed at the end of one month's exposure.  Lung tissue was  taken for
                                        45J

-------
scanning electron microscopy examination for asbestos  fibers as well as liver,




kidney, spleen,  and tracheal and mesenteric lymph nodes  for histopathological




examination.   At the 3,  6,  12,  and 18 month exposure intervals, 15 rats/group




were sacrificed.  Of the 15 rats/group,  5 were used for  evaluation by scanning




electron microscopy and  cytochemical determination.  The relative amounts of




cellular acid phosphatase,  beta-glucuronidase,  and lactic dehydrogenase were




measured to determine the release of non-membrane bound  enzymes and the extent




of lysosomal exocytosis.









Samples of lung tissue,  liver,  kidney,  spleen,  and tracheal and mesenteric




lymph nodes were preserved in 3.0% phosphate-buffered  glutaraldehyde for




scanning electron microscopy examination.   Of the remaining 10  rats/group,




blood and half of the left lung were analyzed for silicon (Si)  by plasma




emission spectroscopy.   The other hair of each left lung was evaluated lor




relative amounts of hydroxyproline.   The right lung of each animal was fixed




in 10% formalin and processed with other body tissues  for gross and




histopathological examination.









Rats to be sacrificed at 24 months will receive the same evaluations previously




described for the interim sacrificed animals with 10 rats/group used for silicon




analysis and the remainder for histopathological examination.









Exposed and control monkeys will have the complete cytocheraical-silicon




analysis-histopathological evaluations conducted at the  time of their scheduled




sacrifice.
                                        452

-------
Chamber Asbestos Analysis Methods and Results





Conditioned air (humidity and temperature modified)  was used to disperse the

                                          3
asbestos into the dilution air of the 12 m  chamber  and to prevent fibers from


adhering to one another.   The air flow through the chamber was regulated for


six air changes per hour.  The chrysotile asbestos in the exposure chamber was


measured by three methods:  gravimetric analysis,  fiber length distribution


analysis, and by scanning electron microscopy.





The gravimetric analysis  consisted of drawing a sample of air from the exposure


chamber at 5.3 liters/minute for 60 minutes through  a 37 mm diameter fibrous


glass filter.  The pre-collection filter weight a'nd  post-collection filter


weight were used to determine total asbestos collection for a determined volume


of sampled air.  Three samples vcre collected at evenly cpaced intervals during


each 7 hour daily exposure and the concentration expressed in mg/m .  The mean


concentration of asbestos in the chamber air for the entire study was


0.95 mg/m3 + 0.26 (S.D.).





The UIOSH P and CAM 239 method was used in a modified form to determine the


chrysotile fiber length distribution in the exposure chamber (8).   The asbestos


concentration was not determined by this method.  A  sample of chamber air was


drawn at a flow rate of 5.3 1pm for 90 minutes through a 37 mm diameter, 0.8 urn


pore size cellulose ester filter.  The 5.3 1pm differed from the 1.7 1pm to


2.5 1pm prescribed in the NIOSH method.  The procedures as described in the KIOSH


method were only used to  clear the cellulose ester filter and to count asbestos


fibers longer than 5 pm in length.
                                        453

-------
A small wedge of the filter was placed on a microscope slide and the body of the




filter was cleared with a reagent containing dimethyl phthalate and diethyl




oxalate.  The slide was then viewed at a magnification of 400X in a phase con-




trast microscope and the sizing of fibers accomplished with the use of a Porton




graticule and hand counter.  Three chamber atmosphere samples were taken at




regular intervals during each 7 hour daily exposure.   The mean number of




chrysotile fibers/cc greater than 5 ym in length for  the entire study was




0.79 fibers/cc + 0.13 (S.D.).









Electron microscopy was used to monitor the number of fibers/cc less than 5 ym




in length.  On occasion, the >5 ym fibers were also measured by this method.




Chamber air was collected in the same manner used for samples drawn for fiber




length distribution except that a polycarbonate filter was substituted for the




cellulose ester filter.   The filLei conlaiuing Luc collected sample was mounted




on a carbon planchet and viewed through a scanning electron microscope.   Photo-




graphs were taken at a magnification of 2,OOOX and the negatives used to print




X5,000 enlargements of the field.  The asbestos fibers were then counted and




measured by hand.  Figure 4 shows a typical photo taken of a sample prepared




by this method.  Short (<5 ym)  and long (>5 ym) asbestos fibers can be seen as




well as a few of the previously described asbestos "balls."  Figure 5 shows a




higher magnification (X20,000)  of one of these generated asbestos "balls"




collected from the asbestos exposure air.  By electron microscopy, the number




of fibers/cc <5 ym in length was estimated at 210 (a  ratio of 265 fibers/cc




<5 ym to each fiber >5 ym).
                                        454

-------
DISCUSSION









At the time of this report,  there are less than two months of maintenance and




observation to complete before the June,  1980,  terminal sacrifice of the rats.




The final steps are being taken to arrange the  long-term holding of the control




and exposed monkeys.









The monthly body weights of  both exposed  and control rats and monkeys have




indicated normal weight gains.  Forty-six rats  (23 exposed and 23 control) have




died or were sacrificed moribund.   ITo pharmacotoxic signs were seen that could




be associated with exposure  to chrysotile.  Lung tissue from serial sacrificed




rats through the IS month sacrifice has revealed little or no pathological




reaction to the inhaled asbestos (Figure  6).









Some short asbestos fibers (0.5-1.0 ym) have been seen in the alveolar macrophages




by scanning transmission electron microscopy.   Due to the relatively small




number of macrophages seen in the alveoli, pieces of rat lung 1.0 cc in size




were ashed in a low-temperature plasma oven. The ash residue was then suspended




in distilled water and filtered on a polycarbonate filter.  Scanning electron




microscopy examination of this filter revealed  a number of short chrysotile




fibers and "balls" as seen in Figure 7.









The positive identification  of chrysotile asbestos in both alveolar macrophages




and in the low-temperature,  plasma-ashed  rat lung tissue was accomplished by




energy dispersive x-ray analysis.
                                        455

-------
The contractor is  compiling  the data on the interim rat sacrifice silicon




assays and the other  cytochemical tests for the final report due in September,




1980.   !k> tests of tissues have been taken on the two groups of monkeys since




none have been sacrificed or died thus far during the study.









It is hoped that the  results of this study will be of significant value in




the continuing assessment and updating of data on a highly hazardous industrial




and commercial substance.
                                     45b

-------
                                 REFERENCES









1.  Newhouse,  M.L., and Thompson, H.  Mesothelioma of Pleura and Peritoneum




    Following Exposure to Asbestos in the London Area.  Br. J. Ind. Med.




    22:261-269,  1965.




2.  Kawnerstein, M., Churg, J., McCaughey, W.T.Ei, and Selikoff, I.J.  Pathogenic




    Effects of Asbestos.  Arch. Path. Lab. Med., Vol. 101:623-627, December 1977.




3.  Shabad, L.M., Pylev, L.R., Krivosheeva, L.V., Kulagina, T.F., and llemenko, B.A.




    Experimental Studies on Asbestos Carcinogenicity.  J, Natl. Cancer Institute,




    Vol. 52:1175-1187, April 1974.




4.  Clifton, R.A.  Asbestos.  Minerals Yearbook.  Vol. 1:213-224, 1975.




5.  Selikoff,  I.J., Nicholson, and Langer.  Asbestos Air Polution.  Arch. Env.




    Health, Vol. 25:1-13, July 1972.




6.  Spurny» K.P.. , Stober» Opielp; ar>d Weiss-  On t-Tip Problem of Milling and




    Ultrasonic Treatment of Asbestos and Glass Fibers in Biological and




    Analytical Applications.  Am. Ind. Hyg. Assoc. J., Vol. 41:198-203,




    March 1980.




7.  Langer, A.M., Wolf, M.S., Rohl, A.N., and Selikoff, I.J.  Variations of




    Properties of Chrysotile Asbestos Subjected to Milling.  J. Tox. Env.




    Hlth. 4:173-188, 1978.




8.  Asbestos Fibers in Air, P and CAM 239, MIOSH Manual of Analytical Methods,




    Vol. 1:239-1-21 (2nd rj.), 1977-
                                      457

-------
                           PHOTOMICROGRAPH LEGENDS









Figure 1.  Ball-milled chrysotile (original X400,  SEM).









Figure 2.  Ball-milled chrysotile showing numerous short asbestos fibers among




           a few large fibers and "balls" (original X5,000,  SEM).









Figure 3.  Chrysotile asbestos "balls1' composed of numerous  short asbestos




           fibers (original X20.000,  SEM).









Figure 4.  Inhalation chamber sample  of chrysotile fibers (original X5,000,




           SEM).









Figure 5,  Inhalation chamber Cample  of chrysotile "ball" (original X2C,000,




           SEM).









Figure 6.  Terminal bronchus of 12-month chrysotile exposed  rat (original X200)









Figure 7.  Chrysotile fibers and "balls" from low-temperature ashed asbestos-




           exposed rat lung (original X5.000, SEM).
                                      458

-------
459

-------
460

-------
 '
."   -.
  3
46J

-------
46,;

-------
463

-------

-------
465

-------
                               Discussion

Dr. 0'Conor, NCI:  Yes, that is an elegant study.  I have two questions.
One, you say this is to test the chronic effects, I guess in these animals,
but you are sacrificing them all six months after the first exposure, no,
I mean after the termination of the exposure.

Mr. Platek, NIOSH:  You must realize that the average lifespan of the rat is
two years.

Dr. 0'Conor, NCI:  I was not thinking of the rats.  I was thinking of monkeys.

Mr. Platek, NIOSH:  Let me explain.  I was appointed project director about
one year ago.  The design was set up by another gentleman at NIOSH.  There is
talk at this moment of possibly doing pulmonary function tests on these
monkeys so the data will not be wasted.  Granted the monkeys will live much
longer than rats.

Dr. 0'Conor, NCI:  Have any of the monkeys been sacrificed to date?

Mr. Platek, NIOSH:  No, none whatsoever.  They will all be sacrificed or are
scheduled for sacrifice the 25th of June.

Dr. 0'Conor, NCI:  Is there any possible consideration of just holding the
monkeys rather than sacrificing?

Mr. Platek, NIOSH:  I would have to speak with Dr. Groth on that question,
but right now there is no consideration.

Dr. 0'Conor, NCI:  The other point is what do you think will be the implication,
let us say that these experiments are negative; what do you think will be the
implication in terms of the industry and regulatory action?

Mr. Platek, NIOSH:  We were concerned in talking this over, that is,  where
would a worker be occupationally exposed to purely small asbestos fibers,
and we don't know.  Brake shoe repair operations might be one of the few
places, but as I mentioned before, in any preparation or any exposure, long
fibers will be present.  However, this project was designed to test a standard
of two fibers per cc, greater than 5 micrometers in length.  In that standard
you don't take into consideration any of these smaller asbestos fibers, and
as you can see, our chamber samples here meet that standard and are below it.
In fact, they are less than half of the federal standard.  We can show that,
but you can also see there are multitudes of these small fibers.  This is why
the project was conducted to determine when only these small fibers are
present, do they have any adverse biological effect.
                                    46£

-------
Dr. Burton, NCI:  Was there any physical effect on the fibers of superheating
them several times compared to fibers that might have been obtained in the
original physical condition?

Mr. Platek, NIOSH:  I know what you are saying.  There have been numerous
reports that heating, sonicating, even ball milling of the asbestos can
create problems.  We have done no tests to my knowledge to test whether the
crystal structure of the asbestos was altered.  This can be determined by
x-ray and electron diffraction analysis.  We have within the last year
received the proper equipment on our transmission scope that we can do this
type of analysis.  That has not been done to date, but it will be performed
and the results will accompany the final report.

Dr. Cameron, NCI:  Just a further comment on the primate aspect.  As a
rhesus monkey they should have a lifespan—

Mr. Platek, NIOSH:  Cynamolgus.

Dr. Cameron, NCI:  Cynamolgus.  They are still rhesus, the rhesus families
have a lifespan in excess of 20 years.  So that is minimal.  I have a question
though, are you aware of the NIEHS asbestos study at a local lab?  I don't
know the particulars.  I wonder if you do?

Mr. Platek, NIOSH:  I have been told, in fact, we just learned last week,
there is evidently a lab that is doing an asbestos study; an intra-tracheal
study of short asbestos fibers.  I have no idea how they are doing it, but
I have been told that it is a one-dose intra-tracheal injection study.  I
have no idea of the dose, how they prepared the asbestos, who their source
was or the size range of the asbestos.

Dr. Lee, EPA:  I would like to make a comment.  If I am not mistaken, within
the EPA at our TP laboratory we have also undertaken a chronic study of the
asbestos in rats.  I am not too familiar with this project.  I wonder if
Dr. Waters is at liberty to give us some information?  Dr. Waters is not
here.  As I say, we understand that project is, also, near the end, and that
is a two year study for the rats.  You may want to get in touch with them.
The project is headed by Dr. Coffin.

Dr. Brown, OSHA:  Could you elaborate a little bit on the splitting of these
particular asbestos fibers?

Mr. Platek, NIOSH:  The splitting as far as the process we used?

Dr. Brown, OSHA:  Right.

Mr. Platek, NIOSH:  We did ball mill these asbestos fibers.  I explained the
method in which they were dried, ball milled for 24 hours and we dried them
again, and they they were shipped to the contractor.
                                        467

-------
Dr. Brown, OSHA:  Specifically what I mean was, was it vertically or
horizontally splitting?

Mr. Platek, NIOSH:  The ball milling split the asbestos bundles as well as
breaking the fibers into shorter lengths.  As I mentioned, asbestos "balls"
were also created.

Dr. Brown, OSHA:  The second question is where are the areas of the greatest
deposition?

Mr. Platek, NIOSH:  I honestly don't know yet.  As I said, or pointed out in
the slide that I showed a while back with the terminal bronchus, if you were
seeing a lot of asbestos in that lung you would probably expect to see the
asbestos fibers near the lymphatics and near the major blood vessels, but we
have not seen them.  Once again, we are talking about fibers that when viewed
by the light microscope are going to be far beyond the range of light
microscopy to resolve, and to develop a technique to do it by scanning
electron microscopy is another interest of mine in this project.

Dr. Lee, EPA:  One more question?

Dr. Hegyeli, NCI:  There are studies indicating that the physical size, the
diameter and the lengths of the fiber has much more importance in the
physiological response than the nature, the chemical nature of the substance,
including glass, metal and other fibers.  So, my question is what does this
study mean physiologically?

Mr. Platek, NIOSH:  From what we have seen so far, as in the previous slide
that you saw of the electron micrograph of the macrophage containing the
asbestos fibers, it would appear the macrophage is doing its job, and it is
engulfing the asbestos.  There was no adverse effect in that macrophage that
we could see, and it looks like they could be clearing themselves as they
are supposed to do.  That would be of significance in hopefully determining
that the short fibers really don't produce the problems by remaining in the
lung, and as you stated, it has been pointed out by numerous investigators
that the macrophages have difficulty engulfing the long fibers, and there-
fore you have the influx of fibroblasts, the laying down of fibrin and then
your fibrosis sets up.  I hope that answers your question.

Dr. Hegyeli, NCI:  My question is that in a practical sense you never
encounter these type of fibers.  You have mixed fibers, and most of them
are in the range, and as you indicated with the macrophage, at least there
are some scanning electron micrograph studies indicating that if it occurs
up inside the cell, it might serve really as a factor.

Mr. Platek, NIOSH:  I don't really know how to answer you any further than
what I have on that one.  I said that all environments with the possible
exception of some of the brake shoe removal operations where the asbestos
                                      468

-------
is under extreme pressures  and  can be broken down into much smaller fibers,
you are going to bave large fiber lengths and the smaller ones in all
exposures.  This was, once  again, mainly a project of testing a federal
standard and not where will the worker be exposed to these short fibers
because as of right now I do not know of a work environment that is strictly
small fiber exposure.
                                     469

-------
MUTAGENICITY TESTING OF SELECTED  INDUSTRIAL  CHEMICALS
    Tong-man Ong,  Geoffrey Taylor3,  John  Elliott,



         Carole A.  Golden  and  Randy  G. Moon
   Division of Respiratory  Disease  Studies,  NIOSH



              Morgantown, West  Virginia



              (T.  0.,  G.  T.,  and  J.  E.)



                         and



           Utah Biomedical  Test Laboratory



                Salt Lake City, Utah



               (C.  A.  G.  and  R. G.  M.)
    aPresent address:   Wausau  Medical  Center



                       2727  Plaza  Drive



                       Wausau,  Wisconsin    54401
                            470

-------
                                SUMMARY
     The mutagenicity of 147 industrial chemicals  and structurally
related compounds have been studied by the Utah Biomedical  Test
Laboratory at Salt Lake City,  Utah, under the contract with the
National Institute for Occupational Safety and Health.  The Salmonella
typhimuriurn-microsome plate incorporation test developed by Ames  and
Co-workers was  used as the assay system.   The assays  were conducted
with the tester strains TA 1535, TA 1537, TA 98 and TA 100  in  the
presence and absence of S-9 prepared from the liver of Aroclor 1254
pretreated Sprague Dawley rats.
     The results of these studies indicate that 120 of 147  compounds
were not mutagenic to any of the testers  tested with  or without
metabolic activation.  Twenty-three compounds were directly mutagenic
to one or more  tester strains, and the remaining four compounds
required metabolic activation  for their mutagenic  activities.
                                     47 i

-------
                              INTRODUCTION
     It is well  documented that many synthetic and naturally occurring
compounds can interact with the genetic material  and cause mutations in
somatic and/or germ cells.  Induction of mutations in these cells may
lead to one or more of the following deleterious  effects:   Genetic
disease, malformation, spontaneous  abortion,  and  cancer.   Recent studies
have estimated that more than  1000  known genetic  disorders and diseases
can be related to gene mutations.   It has been estimated  that
approximately 50% of all spontaneous abortions involve  chromosomal
defects and more than 0.5% of  total  live births in the  United States
carry clinically serious chromosomal aberrations^.  The large number of
chemicals created by modern industrial  technology may account for this
prevalence of genetic diseases  and  disorders  and  for the  high incidence
of cancer noted in humans in recent  years.
     During the past decade many short-term tests for mutagenesis have
been developed.   Among these tests,  the histidine reverse  mutation  system
of Salmonella typhimurium developed  by  Ames and co-workers^  is probably
the most sensitive and useful  system for mutagenesis testing.   In this
test system, the use of j_n yitro metabolic  activation has  helped detect
the mutagenic activity of promutagens2.   By using different tester  strains,
mutagenic specificity (base-pair substitution vs. frameshift mutation)
of chemicals can be determined.   With this  test system,   McCann et  al.^,
have shown that there is an excellent correlation between  mutagenicity
and carcinogenicity among the  300 compounds studied.
     Many industrial workers are routinely  exposed to occupation-related
chemicals.  To protect these workers from any potential mutagenic and
                                       47.

-------
carcinogenic hazards, it is  necessary to  detect and identify the
mutagenic activity of industrial  and related  chemicals.   A contract was,
therefore, initiated by NIOSH in  1977 for Utah  Biomedical  Testing
Laboratory to study the mutagenic activity of 147 industrial  chemicals
and related compounds in the Salmonella/microsomal  assay  system.
                          MATERIALS  AND METHODS
Bacterial Strains
     The bacteria used in this study were Salmonella typhimurium strains
TA 98, TA 100, TA-1535 and TA 1537.   They were  provided by Dr.  Bruce N.
Ames, University of California at Berkeley.   Upon receipt, each strain
was subjected to the appropriate  procedures  to  confirm its genotype.
Strains were reisolated, genotypes tested, and  new frozen  stocks prepared
at bimonthly intervals.  Stocks were stored  at  -80° C, and inocula  were
prepared fresh for each experiment by subculture in nutrient broth.
Mutagenesis Assay
     Most chemicals were tested by the plate  incorporation method.   Some
chemicals were tested by the pre-incubation  technique. Details of  the
methodology of Salmonella/microsome  mutagenesis assay have been described
by Ames and co-workers-^.  A  brief description of both assay systems  is
as follows:
1.  PI ate Incorporation Assay
         Agar plates containing histidine-deficient Vogel  Bonner
    Medium £^ and fortified  with^Z*  glucose  were prepared and incubated
    overnight at 37° C prior to use  (VBME plates).   Aliquots of top
    agar (0.6S Bacto-agar in 0.5% NaCl) were  melted on the day they
                                        473

-------
    were  to  be  used  and  maintained  at  45°  C.   Each  100  ml  aliquot



    of top agar was  supplemented  with  a  final  concentration  of



    0.05  mM  histidine  and  0.05  mM biotin immediately  before  use.



    Five  ml  of  a 16  hour nutrient broth  culture  of  the  appropriate



    Ames1  tester strain  of Salmonella  typhimurium was added  to the



    100 ml aliquot of  top  agar, and the  mixture  transferred  to



    sterile,  disposable  screw capped tubes  in  2.0 ml  aliquots  and



    held  at  45° C.   Test materials  were  added  to the  tubes,  mixed,



    and the  contents of  the tube  were  poured onto the surface  of



    VBME  plates. All  plates were incubated at 37°  C  for three



    days,  and the number of colonies was determined using  a  manual



    Quebec colony counter.



         S-9 was prepared  from  livers  of male  Sprague-Dawley rats



    (150-200 g) injected with Aroclor  1254  at  a  dose  of 1000/mg/kg



    five  days prior  to sacrifice.   For tests in  the presence of S-9,



    each  tube containing top agar,  inoculum and  test  material



    received 0.5 ml  of S-9 mix  containing  0.45 ml of  S-9 base^ and



    0.05  ml  of  S-9 fraction.  Tube  contents were poured onto the



    surface  of  VBME  plates  immediately after addition of the S-9.



2.  Pre.-incubation Assay (as above  with  the following exceptions)



         Test materials  were added  to  sterile, disposable  screw



    capped tubes and inoculated with 0.1 ml of a 16 hour nutrient



    broth  culture of the appropriate Ames1  tester strain of



    Salmonella  typhimurium (TA  1535, TA  1537,  TA 98,  or TA 100).



    The tubes were incubated at 37° C  for  30 minutes  with  shaking.



    Two ml aliquots  of top agar at  45° C were  then  added to  the

-------
    tubes,  the  tubes  were mixed, and the contents were poured onto
    the surface of VBME  plates.  For tests  in  the presence of S-9,
    each tube containing sample plus inoculum  received 0.5 ml of
    S-9 mix prior to  the 30 minute  incubation  at 37° C.
         Five different  doses of each chemical were tested with
    and without microsomal activation.  The highest concentration
    used was limited  by  solubility  and  toxicity.  Five different
    concentrations of S-9 (10, 20,  30,  40 and  50 yl per plate) were
    used for each dose in the plate incorporation test while only
    one concentration of S-9  (50 yl/plate) was used for each.dose
    in the  preincubation test.  All tests were performed in
    duplicate and each compound was tested  along with positive and
    negative controls.  The positive control compounds used are:
    Propylene oxide for  TA 1535 and TA  100, 9-aminoacridine for
    TA 1537, and 2-nitrofluorene for TA 98.  Ethidium bromide was
    used as a positive control compound for metabolic activation.
    All gases and volatile compounds were tested in a sealed jar.
Chemicals
     Chemicals  used in this study,  with the exception of platinum related
compounds,  were obtained from commercial chemical companies.  Platinum
related compounds were provided by,  Dr.  Dave Groth, NIOSH, Cincinnati,
Ohio.  Chemicals were dissolved in  sterile  distilled water (whenever
feasible) or in dimethyl sulfoxide  immediately before used.
                                    475

-------
                                 RESULTS
     The results are shown in  Tables  1  and  2.   Among the 147 compounds
tested, 120 were not mutagenic to any of the  testers either with or
without metabolic activation.   Twenty-three compounds were directly
mutagenic to one or more than  one tester.   The  remaining four compounds
(1,2,3-trichloropropane, dimethoxyethylphthalate,  2,4-dimethylam'line
and 2,6-dinitrotoluene)  required metabolic  activation for their mutagenic
activities.  A compound  is classified as mutagenic if it causes, a  dose
related increase in the  number of revertants  and  the increase is more
than two times the background  level.   The ranges  of background
revertants were 10-50 for TA 1535,  5-25 for TA  1537,  16-90 for TA  98
and 100-350 for TA 100.
     Two (nitromethane and monochloroethane)  of the 27 mutagenic
compounds were mutagenic only  for TA  1535  and two  other compounds
(dimethoxyethylphthalate and n-nitrosoaniline)  were mutagenic only for
TA 1537.  Pt (bipyridyl) Cl2 and 2,4,-dimethylaniline were mutagenic
only for TA 98 and TA 100, respectively. Several  compounds  were
mutagenic only for one or two  testers if tested v/ithout in vitro
microsomal activation.  However, they were  mutagenic for more than two
testers if tested with S-9 from liver of Aroclor  1254 pretreated rats.
Some compounds showed higher mutagenic  activity when a low concentration
(20 pi/plate) of S-9 was used, whereas  other  compounds required a  higher
concentration (50 ul/plate) of S-9  for  mutagenic  activity.
                                  476

-------
                               DISCUSSION
     In this study,  we found  that Industrial  chemicals  such  as
1,2,3-trichloropropane,  nitromethane,  2-nitropropane, methyl  bromide,
methylene chloride,  monochloroethane,  2,3-dinitrotoluene,
2,5-dinitrotoluene,  2,6-dinitrotoluene,  2,4,5-trinitrotoluene,
p-dinitrobenzene,  and m-dinitrobenzene are  mutagenic  for Salmonella
typhimurium.  More studies  need to be  conducted  in  mammalian  and  other
submammalian mutagenesis test systems, and  the potential mutagenic and
carcinogenic hazards of these compounds  for the  exposed population need
to be determined.
     Several interesting phenomena were  noted in this study.  Ptl^C^
was mutagenic for TA 98 and TA 100 if  it was  dissolved  in water.  If
dimethyl sulfoxide (DMSO) was used as  the  solvent,  however,  negative
results were obtained.  It  seems that  DMSO  could interact with  PtKgCl^
and inhibit or diminish the mutagenic  activity of this  compound.  This
result emphasizes  the importance of making  an effort  to dissolve
chemicals in aqueous solution for mutagenesis testing.
     S-9 from the liver of  mammals can activate  promutagens  to  mutagenic
metabolites and, in some instances,  enhance the  mutagenic activity of
directly acting mutagens.  A  concentration  of 50 yl S-9 per  plate is
used by most laboratories for mutagenesis  testing.  In  this  study,
however, the best mutagenic response of  several  compounds was found when
a lower concentration of S-9  (20 pi/plate)  was used.  Recently, we have
also found that m-aminophenol and the  dye,  direct blue-15 are mutagenic
only when these compounds were tested  with  S-9 from hamsters.
                                        477

-------
     There appears  to  be a structure and function relationship among
several  structurally related chemicals studied.  For instance,
riitrophenol  is  mutagenic if hydroxy and nitro groups are in a meta
arrangement,  but not mutagenic if both groups are in an otho arrangement
Similar results were found with dinitrobenzene,  m-dinitrobenzene was
mutagenic but o-dinitrobenzene was not.
                                   478

-------
                               REFERENCES
1.  Ames, B. N.   The Detection  of Chemical  Mutagens  with  Enteric
         Bacteria.   In Chemical  Mutagens:   Principles  and Methods  for
         their Detection (Hollaender,  A.,  ed.),  Vol.  1, New  York:
         Plenum Press, pp.  267-282,  1971.
2.  Ames, B. N.,  Durston, W.  E.,  Yamasaki,  E.  and  Lee, F^  D.   Carcinogens
         and Mutagens.  A simple  test  system  combining liver homogenates
         for activation and bacteria for detection.   Proc. Natl... Acad.
         Sci., USA,  70:2281-2285,  1973.
3.  Ames, B. N.,  McCann, J. and Yamasaki,  E.   Methods  for Detecting
         Carcinogens and Mutagens  with the  Salmonella/Mammalian Microsome
         Mutagenicity Test.  Mutation  Res., 31:347-364, 1975.
4.  Carr, D. J.  and  Gedeon, M.   Population  Cytogenetics of Human
         Abortuses.   In Population Cytogenetics  (Hook, E.  B.  and Porter,
         I.  H.,  ed.), New York:   Academic  Press, p.  1, 1977.
5.  McCann,  J.,  Choi, E., Yamasaki,  E. and  Ames, B.  N.  Detection  of
         Carcinogens as Mutagens  in  the Salmonel1 a/Microsome Test:
         Assay of 300 Chemicals.   Proc. Natl.  Acad.  Sci.,  USA, 72:5135-
         5139, 1975.
6.  Vogel, H. J.  and Bonner,  D.  M.   Acetylornithinase  of  E.  Coli partial
         purification and some  properties.  J. Biol.  Cherri.,  218:97-106,
         1956.
                                      479

-------
                         TABLE 1



    Compounds That Did Not Exhibit Mutagenic Activity



               for Salmonella Typhimurium








Alicyclic and Heterocyclic Compounds:



     Adenine (1500 yg)a



     Dipyridyl  (5000 yg)



     Furfuryl alcohol (15 yl)



     Hexachlorocyclopentadiene (0.0002 yl)



     2-mercaptobenzothiazole (500 yg)



     N-methyldicyclohexylamine (10 yl)



Aliphatic Amines:



     sec-butylamine (10 yl)



     2-dibutylamino ethanol (50 yl)



     2-diethyl ami no ethanol (30 yl)



     Ethanolamine (30 yl)



     Hexamethylenediamine (3 yg)



     Hexamethylenetetramine (5 yg)



     n-hexylamine (4.5 yl)



     n-pentylamine (8 yl)



     n-propylamine (10 yl)



Aliphatic Carboxylic Acids and Other Aliphatic Compounds:



     Acetone cyanohydrin (0.05 yl)



     Acetonitrile (300 yl)



     Acrolein (0.2 yl)



     Aery 1 amide  (50 yg)

-------
               TABLE 1  (CONT'D)
Acrylonitrile (25 pi)
Adiponitrile (100 yl)
Allylchloride (4 yl)
Butyl isocyanate (0.01 yl)
N-butyronitrile (100 yl)
DiethyVenetriamine pentaacetic acid (1 yl)
N,N-dimethylacetamide (1000 yl)
Ethylene (100%)
Ethyleneglycol-bis-(g-aminoethylether)-N,N'-tetra
     acetic acid (500 yg)
Glyconitrile (3500 yg)
Hexachlorobutadiene (500 yl)
Hexachloroethane (4000 yg)
Iso-butyronitrile (5 yl)
Malononitrile (0.5 yl)
Methyl ethyl ketone peroxide (0.4 yl)
Bis-(2-methoxyethyl) ether (500 yl)
Oxalic acid (2000 yg)
Pentachloroethane (5 yl)
Perchloroethylene (100 yl)
n-propyl isocyanate (0.01 yl)
Proprionitrile (200 yl)
Succinonitrile (4000 yg)
1,1,2,2-tetrachloroethane (2 yl)

                         48J

-------
                    TABLE 1  (CONT'D)
     Tetramethyl succinonitrile (4000 yg)
     1,1,1-trichloroethane (500 yl)
     Tn'chloroethylene (100 yl)
Aromatic Amines:
     m-aminophenol (1000   g)
     2,3-dimethylaniline  (2 yl)
     2,5-di methyl am'line  (2 pi)
     2,6-dimethylaniline  (4 yl)
     3,4-dimethylaniline  (1250 yg)
     3,5-dimethylaniline  (100 yl)
     0-methoxyaniline (30 yl)
     N-methylaniline  (60  yl)
     p-nitrobenzyl-N,N-propylamine  (5000 yg)
     N-phenyl-N'-2-octylparaphenylenediamine  (250 yl)
     Triphenylamine (2000 yg)
Aromatic Hydrocarbons:
     Benzoyl  peroxide (100 yg)
     Bisphenol  A (200 yg)
     m-dichlorobenzene (10 yl)
     Ethyl  benzene (0.2 yl)
     Hexachlorobenzene (150 yg)
     Naphthalene (500 ug)
     2,4-toluene diisocyanate (3 yl)
     1,3,5-trichlorobenzene (100 yg)


                              48^

-------
                    TABLE 1  (CONT'D)
     1,2,4-trimethyl benzene (0.3 yl)
     Triortho cresyl  phosphate (500 y)
     m-vinyl  toluene  (0.1  yl)
     0-vinyl  toluene  (0.1  yl)
     p-vinyl  toluene  (0.1  yl)
Metals, Metal Salts,  and Organometallics:
     Bed2 (5000 yg)
     Bipyridyl BeCl2  (300  yg)
     Bis (dimethylglyoxime) Pt(II)  (300  yg)
     Bis (2-pyridinaldoximinato) Pt(II)  (300  yg)
     Dichloro (2-formimidoyl  pyridine) Pt(II)  (100  yg)
     H(Pt adenine C13)  (300 yg)
Nitro Aromatics:
     o-dinitrobenzene (200 yg)
     4,6-dinitro-o-cresol  (150 yg)
     2,4-dinitrophenol  (250 yg)
     2,6-dinitrophenol  (250 yg)
     2,3-dinitrotoluene (50 yg)
     2,4-dinitrotoluene (400 yg)
     3,4-dinitrotoluene (500 yg)
     Metaoxon (2000 yg)
     Mononitrobenzene (4 yl)
     o-nitrophenol  (500 yg)
     p-nitrophenol  (500 yg)
     2,4,6-trirritrophenol  (1500  yg)

                             483

-------
                    TABLE 1  (CONT'D)
Solvents:



     Acetic acid (500 yl)



     Acetone (500 yl)



     Benzene (500 yl)



     Cyclohexanone (50 yl)



     N, N-dimethylacetamide (500 yl)



     Dimethylformamide (500 yl)



     Dimethylsulfoxide (500 yl)



     Ethanol (500 yl)



     2-ethoxyethanol  (20 yl)



     Glycerol (500 yl)



     2-methoxyethanol (500  pi)



     Phosphoric acid, in H20,  pH4 (500 yl)



     Potassium acid phthalate  (2%)  (500 ul)



     Sodium hydroxide, in  H20, pH 9.5 (500



     Sodium phosphate, dibasic (2%)  (500 yl



Miscellaneous:



     Antioxidant 2246 (200  yg)



     Butylene oxide (100 yl)



     Carbon disulphide (100 yl)



     Cyanogen (0.01%)



     Geltrol (20 yl)



     a-(histamine AlBr3)+ (2000  yg)



     Hydrogen sulphide (0.1%)
                              484

-------
                    TABLE 1  (CONT'D)
     Isopropyl  isocyanate  (0.01  pi)



     Methyl  parathion (300 pg)



     Nitrous oxide (90.6%)



     Phenyl  isocyanate (0.01  pi)



     Phosgene (0.0001%)



     0-terphenyl  (5000 pg)



     p-terphenyl  (1500' pg)



     Tert-butyl  isocyanate (0.1  pi)



     (Tetrazene  AlBr2)+  (500  pg)



     p-toluene  sulfonyl  isocyanate  (2  pi)
aThe number in parentheses  represents  the  highest



 concentration of chemical  tested  in terms  of  quantity



 per plate.
                             485

-------
                                                         TABLE 2


                                Compounds with Mutagenic Activity in Salmonella Typhimurium
Oo
cr

Compound
S-9a Dose/Plateb
No. of His+ Revertants/Platec
TA 1535 TA 1537 TA 98

TA 100
Alicyclic and Heterocyclic Compounds
Vinyl cylohexene dioxide


Di ethyl carbamoyl chloride

Ethylene oxide

2-nitropropane

Methyl bromide

Methylene chloride

+ 15 yl
15 yl
Aliphatic Carboxylic Acids
+e 200 M!
200 pi
+ 0.1 %f
0.1 %f
+ 50 yl
50 M!
+ 0.53 %f
0.53 %f
+ 500 yl
500 M!
669 (9) 28 (12) 283 (78)
647 (11)
and Other Aliphatic Compounds
82 (16) 21 (9) 157 (59)
107 (43)
>2000 (12)
592 (12)
+ - +
NT - NT
124 (8)
28 (13)
70 (22) - 632 (58)
88 (33) - 190 (26)
<2000 (226)
<2000 (285)

896 (183)
690 (261)
>2000 (195)
1640 (215)
>4000 (NT)
>4000 (141)
710 (159)
700 (151)
>2000 (142)
1836 (170)

-------
                                                     TABLE 2 (CONT'D)
                 Compound
S-9a    Dose/Plateb      TA 1535
                                                                           No. of His+ Revertants/Platec
              TA 1537
                                           TA 98
 TA  100
oo
~ T
      Monochloroethane
      Nitromethane
      1,2,3-Trichloropropane
      Dimethoxyethylphthalate
      2,4-dimethyl aniline"
      N-nitrosoaniline
      [Pt(bipyridine)  (adenine)]Cl2
5000 yl
5000 yl
  50 yl
  50 ul
 0.5 yl
 0.5 yl
 100 yl
 100 yl
   2 yl
   2 yl
 100 yg
 100 yg
594 (14)
159 (18)
146 (19)
105 (36)
385 (21)
                                        100 (9)
                                         34 (8)
                                         75 (14)
                                        143
  Metals, their Salts and Organometallics
 +                          NT           NT
            40 yg
                            NT
                          511 (40)
>2000  (219)
                                                                   446 (194)
                                                          NT
                                                         543 (221)

-------
TABLE 2 (CONT'D)

Compound S-9a Dose/Plateb TA 1535
Pt(bipyridyl)Cl2 + NT
10 yg
Pt K2 C14 + NT
100 yg
Nitroaromatics
m-dini trobenzene +e 200 yg
»U - 200 yg
Oo
CC1 p-di nitrobenzene -e 50 yg
50 yg
2,5-dinitrophenol +e 20 yg
20 yg
2,5-dinitrotoluene +e 200 yg 61 (21)
100 yg
2,6-dinitrotoluene + 2000 yg
2000 yg
No. of His+ Revertants/Platec
TA 1537 TA 98
NT NT
709 (50)
NT NT
267 (40)
219 (41)
160 (21)
90 (48)
505 (44)
142 (64)
351 (57)
112 (49)
233 (35)
57 (19) 337 (88)
-

TA 100
NT
-
NT
689 (221)
362 (180)
569 (255)
407 (172)
565 (209)
324 (193)
-
-
-
485 (243)
_

-------
TABLE 2 (CONT'D)
Compound S-9a Dose/Pi ateb
1-nitronaphthalene +e 100 yg
100 yg
m-nitrophenol +e 500 yg
500 yg
Paraoxon + 4 yl
4 yl
2,4,5-trinitrotoluene +e 10 yg
QD - '10 yg
O
Miscel
Propylene oxide +e 100 yl
100 yl
Styrene oxide + 5 yl
5 yl

TA 1535
-
-
36 (18)
-
54 (16)
71 (35)
-
laneous
168 (16)
120 (20)
NT
NT
No. of His"1" Revertants/Platec
TA 1537 TA 98
82 (38)
76 (22)
127 (33)
205 (20)
-
-
96 (52)
279 (44)
60 (22)
NT NT
NT NT

TA 100
1000 (332)
1080 (300)
505 (196)
-
536 (198)
565 (231)
486 (178)
384 (209)
2500 (332)
2300 (300)
>4000 (199)
>4000 (153)

-------
                                                    TABLE 2 (CONT'D)
Compound S-9a
Tetramethyl thiuram di sulphide +
-
Dose/Plateb
25 Mg
TOO yg
No. of His+ Revertants/Platec
TA 1535 TA 1537 TA 98 TA 100
43 (12) - - 510 (171)
453 (191)
     a + = Tested with metabolic activation; - = Tested without .metabolic activation.
     k Dose which gave the highest mutagenic response.
     c Number of revertants is an average of 2 plates; Number of spontaneous revertants is shown in parentheses;
       NT = not tested; - = Number of revertants is less than 2 times of the background.
ȣ*   H
^     Tested by the pre-incubation assay system.
     e The concentration of S-9 is 20 ^I/plate rather than 50 yl/plate.  Number of revertants decreased when 50
       S-9/plate was used.
       Percent in air.

-------
                                           TABLE 3



                                  Summary of the Test Results
Types of Compounds Tested
Ali cyclic and Heterocyclic
Aliphatic Amines
Aliphatic Carboxylic Acids
and other Aliphatics
Aromatic Amines and
Aromatic Hydrocarbons
Metals, their Salts
and Organometallics
Nitroaromatics
Solvents and Reagents
Miscellaneous
No. of Compounds
Tested
7
9
38

27

9

21
15
20
No. of Compounds
Without
Activation
1
0
7

1

3

8
0
3
Found to be
Activation
Required
0
0
1

2

0

1
0
0
Mutagenic
Total
1
0
8

3

3

9
0
3
Total
147
23
27

-------
                              Discussion

Dr. 0'Conor, NCI:  I could not read the slide of the list, and you say you
will have the list, but could you just tell us were there some metals that
were positive?

Dr. Elliott, NIOSH:  As you know, if you test the metals like cobalt, nickel
and cadmium, they are not positive in the standard Ames1 assay.  The com-
pounds that were positive in this study were organic platinum compounds.
Dr. Groth is interested in beryllium compounds and in the way that they
cause cancer.  He has synthesized the platinum compounds and then he is
going to repeat the same thing with beryllium.  We do not have the data on
those compounds, but beryllium metal and its salts, beryllium chloride,
beryllium nitrate, and beryllium oxide, were negative.

Dr. Lee, EPA:  On the same line, I have a question.  I noticed that on your
list there you did some of the toluene group, and you have about three or
four different isomers.  Did you check on most of those?

Dr. Elliott, NIOSH:  Yes.

Dr. Lee, EPA:  The reason is when I was at a Middle West research institute
we undertook an extensive mammalian toxicity study for the munitions com-
pounds, the compounds important to the Army, and in one group is trinitro-
toluene and dinitrotoluene.  Most of the isomers, unfortunately I don't
remember which is positive and which is negative and this data has not been
in the literature, but it is in the extensive report to the Army.

Dr. Elliott, NIOSH:  We have looked at several dinitrotoluenes.  The 2,3,
2,5 and 2,6 were negative, and I cannot remember which ones I had up on the
slide.  It does not make any difference, but—

Dr. Lee, EPA:  You had three of them.

Dr. Elliott, NIOSH:  The 2,5 and 2,6 dinitrotoluenes were positive along
with trinitrotoluene.

Dr. Lee, EPA:  One of the derivatives is the nitroaminotoluene, and if I
remember that is a positive.  I understand that, also, is one of the
metabolites, both by bacteria reaction and, also, biological changes in
the high species.  That is, also, positive.
                                      492

-------
                             FIGURE CAPTIONS









Figure 1.  Mean daily percent of intervals in which the  chop




sprayer and gelcoat sprayers 1, 2, and 3 placed molds  properly




within the spray booths, during baseline and training.









Figure 2.  Mean daily work-duration, breathing zone exposures




(ppm) for the resin-chop sprayer, rollout person,  gelcoat




sprayers 1, 2, and 3 and mold repair person, during baseline




and training.  Means across baseline and training  conditions




are denoted by (	) .
              NO HTS^USSIO-N FOLLCyviN'G THIS P
                               493

-------
        A Strategy to Validate Work Practices:

  An Application to the Reinforced Plastics Industry
       R. J. Conard, Ph.D.,  Research Associate,
           Department of Hunan Development,
                 University of Kansas

           B. L.  Hopkins, Ph.D.,  Professor^
           Department of Human Development,
                 University of Kansas

          H. Gordon Fitch, Ph.D., Professor,
                  School of  Business,
                 University  of Kansas

   Michael J. Smith, Ph.D.,  Research Psychologist,
National Institute for Occupational. Safety and Health

    W. Kent Anger, Ph.D., Research Psychologist,
National Institute for Occupational Safety and Health

     Richard F. Dangel, Ph.D., Assistant  Professor,
            Graduate School  of Social Work
           University of Texas at Arlington
                       494

-------
                                FOOTNOTE









From the Department of Human Development (Dr. Conard and Dr. Hopkins) and




the School of Business (Dr. Fitch), the University of Kansas, Lawrence,




the.Robert A. Taft Laboratories, the National Institute for Occupational




Safety and Health, Cincinnati (Dr. Smith and Dr. Anger), and the Graduate




School of Social Work, the University of Texas, Arlington (Dr. Dangel).








This research was supported in part by a contract (NIOSH 210-77-0040)




between the National Institute for Occupational Safety and Health and




the University of Kansas.








The help of Albert Stewart, M.A., in identifying work practices and




establishing air sampling procedures, and Robert Haynes, B.A.,




Robert Spencer, Ph.D., and Randall Pine, M.B.A., in conducting the




literature search and establishing the experimental procedures, is




gratefully acknowledged.  Especial gratitude is due Mr. Walt McGinnis,




President, Mr. Bill Holiday, Plant Manager, Mr. Ken Hans, Foreman, and




all of the employees of the Fiberglass Department, Labconco, for their




indispensible cooperation and interest.  We also wish to acknowledge




the assistance of Rick Hornung for providing parametric analyses of




certain data and Dr. Robert Mason for his critical review of the




manuscript.









Reprint requests to Department of Human Development, University of Kansas,




Lawrence, Kansas, 66045 (Dr. Hopkins).







                                 495

-------
                           SYNOPSIS/ABSTRACT









Most recommendations for work practices appear to  be  based  on  common




sense rather than an empirical analysis of the value  of  those  practices




to reduce exposures.  A strategy for validating recommended work




practices is presented and applied to a reinforced plastics manufacturing




plant.  Selected employees were trained to use several work practice




behaviors judged likely to reduce their exposures  to  styrene.   Observa-




tional data indicated that all of the work practices,  with  the exception




of respirator usage, changed as desired.   Indices  of  personal  exposures




decreased by up to 74 percent following training for  the workers with




the greatest exposures and, potentially,  the most  control over their




exposures.  The research is presented as  a model which could be generally




applied to validate work practices and to develop  methods by which




workers can be trained to participate in  their own occupational health




protection.
                                   496

-------
Recommendations of Industrial hygiene and occupational medicine commonly



Include work practices designed to protect people from hazards.  Examples




may be found in sections on work practices contained in criteria documents



published by the National Institute for Occupational Safety and Health


       1 2
(NIOSH) ' .  The execution of all work practices is some form of human




behavior.  Although work practices are widely recommended,  rarely have



researchers attempted to validate them by demonstrated usefulness to




reduce hazards.  Moreover, when there have been attempts to validate




work practices, the research methodology has not provided for conclusive




inferences.  In a literature search, the authors found 254 publications



that included various recommendations for work'practices, but. only 8 of




them included measures of hazards prior to, and after the practices were



recommended.  Moreover, the recommended work behaviors were not measured



in any of the eight papers which reported reduced hazards.   Therefore,



it can not be inferred that the use of the recommended work practices



was responsible for the reduced hazards.






                          o

Fitch, Hermann and Hopkins  outlined a strategy for applying technologies



of behavioral science to safety problems in 1976.  Central characteristics



of the strategy are that health-endangering and health-promoting




behaviors of workers are involved in individual exposure to hazards,



and that technology is available to measure the relevant behaviors,




to change them in prescribed ways, to maintain those that are acceptable,




and to evaluate the effectiveness of the behaviors to reduce the hazards.




The present paper extends that strategy to the validation of work practices,




particularly to work-practices designed or selected to control  exposures
                                  49?

-------
to toxic substances.   By including measurement of use of relevant work



practices (behaviors) and measurement of associated hazards,  the



strategy overcomes the inadequacies of methods employed in previous



reports, and provides for the relatively complete validation of recommended



work practices.  To demonstrate the strategy outlined above and the



effectiveness of certain work practices to. reduce exposures,  workers



exposed to styrene in the reinforced plastics industry were chosen for



study.







Styrene and reinforced plastics manufacturing were convenient for an



initial test of the strategy because many examples of human behavior



involvement with exposures result from open, person-performed processes



and there are a wide range of measurable exposures which could possibly



be reduced.







Human subjects exposed to styrene develop irritation of the mucous



membranes, particularly those of the eyes and nose.  Exposures at


                                                          4

200-300 ppm produce problems with coordination and balance .   Styrene



can be absorbed readily through the skin as well as from inhaled


   4 5
air ' .  The 8-hour time-weighted average (TWA) Federal standard for



styrene is 100 ppm^-







Manufacturing Processes







Manufacturing of reinforced, laminated plastic products, such as those



manufactured from styrane-containing resins, typically consists of a



series of operations'.  A mold that has the converse shape of the






                                    491

-------
desired product is cleaned and waxed and then moved to the gelcoat




sprayer who sprays a mixture such as pigmented polyester resin and styrene




monomer onto the moid with the compressed air sprayer.  The compressed air




sprayer is similar to a paint sprayer and is constructed to mix a catalyst




such as methyl ethyl ketone peroxide (MEK-p) with a resin such as a




resin diluted with styrene as it leaves the gun.








When styrene is used as a diluent-reactant it polymerizes with, the resin




after application to the mold; during this curing process, the mold is




set aside to allow the gelcoat layer to harden.  After hardening, the




cured gelcoat is given a reinforcing lamination of fibrous-glass.  The




lamination is applied with a second spray gun that shoots a mixture of




chopped fibrous-glass, resin-styrene mixture and catalyst.  The operator




of this machine is called the chop sprayer.








Immediately after application of the reinforcing lamination, additional




reinforcement may be built into the part by the integration of wooden




or metal members or woven fibrous-glass mats soaked in a resin-styrene-




catalyst mixture.  The reinforcement is typically bonded to the part




with a light spray of the chopped fibrous-glass mixture.








In the next operation, workers using rollers, much like those employed




for painting, roll the newly applied lamination to remove gas bubbles




from the mixture and to insure that the resin and fibrous-glass are




thoroughly compressed and mixed.  These workers are called rollout persons.
                                  498

-------
The molds and parts are again set aside to cure,  the parts  are  removed




from the mold, and the mold is inspected and repaired if  necessary.




The person who performs this latter operation is  called the mold  repair




person.









Although other finishing operations may be performed, and plants  differ




with respect to floor plans, engineering controls,  storage, and equipment,




the above described steps are typical and characterize a major  portion




of the reinforced-plastics industry.









Methods









Cooperation was secured from Labconco, a producer of reinforced plastic




laboratory equipment such as fume hoods and bacteriological glove boxes.




In obtaining cooperation of the company, there was  an understanding




that individual workers would have the option to  participate  as subjects




in the study after being told the general purposes  of the research.









Determination of High Exposure Areas and Jobs: Momentary air samples




(commonly called grab tube samples) were used to  determine  styrene




concentrations.  These samples were taken with a  Bendix,  Model  400,




Gastec pump and Zink styrene detector tubes to identify the plant




areas and jobs that involved relatively high exposures.  This method




sampled only small volumes of air (100 ml) over brief periods of time




(less than 30 sees) yielding an estimated accuracy of +25-35%.   The




high exposure areas, with momentary concentrations ranging  from 110-280 ppm,




were the two spray booths in which the gelcoat mixture and  the resin-chopped




                                    HOC

-------
fibrous-glass mixture were sprayed onto the molds.  Styrene was vaporized




and aerosolized as a result of being sprayed and due to the heat produced.




during polymerization.  These two processes appeared to be the source




of most of the styrene in the plant.  The rollout and curing areas of




the plant also yielded relatively high momentary concentrations of




styrene, ranging from 70-170 ppm.  The two spraying jobs and the job




in which the resin-fibrous-glass mixture was rolled out involved not




only the greatest momentary exposures for workers, but also the greatest




total time of relatively high exposure.









Several jobs, such as repairing molds and touching up blemishes in gelcoat




surfaces, occasionally introduced relatively small quantities of styrene




into the air.  Many jobs, such as those involved in finishing operations,




and moving parts from one area to another, introduced no or negligible




styrene into the plant.  Workers in these jobs were exposed to styrene




as a result of the ambient concentrations produced by the styrene-




introducing processes described above.  These momentary ambient concen-




trations typically ranged from 2-20 ppm.









From the high exposure jobs, three workers, the gelcoat sprayer, the




resin-chop sprayer, and one rollout person, were selected for further




study.  It was assumed that changes in work practices would have the




greatest likelihood of affecting the exposures of these workers.  To




determine if similar work practices would reduce the exposure of a worker




who was primarily contacting styrene only in ambient air, the mold




repair person was also included in the sample.
                                  50 J

-------
Development of Potentially Useful Work Practices;  Detailed observations




were made of these four selected workers to identify ways in which their




on-the-job behaviors might be resulting in styrene exposure.  Three




general classes of health-promoting work practices emerged:  1) using




appropriate personal protection, 2) avoiding high-exposure areas when




not necessary for production, and 3) taking advantage of existing




engineering controls.  Based on these three classes, the following




specific work practice behaviors were defined:




     1)  Gelcoat and Chop Sprayers:




         a)  Using appropriate personal protection




             1)  Wearing a respirator when working inside spray booths,




             2)  Keeping all skin below the neck, including hands, covered.




         b)  Avoiding high exposure areas




             1)  Staying out of spray booths except when spraying,




                 transferring or arranging parts.




         c)  Taking advantage of engineering controls




             1)  Activating booth exhaust ventilation before spraying.




             2)  Keeping doors to spray booths closed while spraying.




             3)  Placing molds to be sprayed directly in front of, and




                 close to, exhaust ventilation.




             4)  Spraying toward the exhaust ventilation.




             5)  Turning molds as necessary to maintain a downwind




                 spray direction.




             6)  Minimizing overspray on booth floors and walls.




             7)  Not directing spray toward self.




             8)  Not directing spray toward others.

-------
2)  Rollout Personnel:




    a)  Using appropriate personal protection




        1)  Wearing a respirator while in the rollout  area  or




            spray booth.




        2)  Wearing a respirator when working inside molds.




        3)  Keeping all skin below the neck,  including hands,




            covered.




    b)  Avoiding high-exposure areas




        1)  Performing rollout in the rollout area, not in  the




            spray booth or curing areas.




        2)  Avoiding the rollout area when not working.




        3)  Not entering the spray booth while the sprayer  is




            in operation.




    c)  Taking advantage of engineering controls




        1)  Activating floor fans and directing  them toward exhaust




            ventilation before rolling out.




        2)  Staying upwind of the part while rolling  out.




        3)  Turning molds as necessary to maintain an upwind




            position while rolling out.




3)  Mold Repair Personnel:




    a)  Using appropriate personal protection




        1)  Wearing a respirator when working with uncured  resin.




        2)  Keeping all skin below the neck,  including hands,




            covered.




    b)  Avoiding high exposure areas




        1)  Staying out of the mold repair area, if not working,




            while curing resin is present.






                              S03

-------
             2)  Keeping head at least 12 inches away from  uncured  resin




                 applications on molds.




             3)  Keeping resin containers covered at  all  times.




         c)  Taking advantage of engineering controls




             1)  Performing repairs within the mold repair  area.




             2)  Activating floor fans and directing  them toward exhaust




                 ventilation, before applying resin to molds.




             3)  Staying upwind of uncured resin applications.




             4)  Turning molds as necessary to maintain an  upwind




                 position to curing resin.









Measurement of Work Practices and Exposures:  Data were collected on




Tuesdays, Wednesdays, and Thursdays, for seventeen days,  on each of




the four selected workers.  There was greater absenteeism and more  time




was devoted to scheduling work and getting machinery  operating  on




Mondays, and Fridays were given largely to plant cleanup.   Therefore,




data were not collected on these days.  The gelcoat spraying job




underwent two personnel changes during the six-week period. The first




gelcoat employee worked on days one through three, the second on days




four through eight and the third on days nine through seventeen.









Five classes of data were taken:  behavioral data, styrene  exposure




data taken from work-duration breathing zone air samples,  styrene




exposure data from eight-hour breathing-zone and area air samples,




and mandelic acid levels taken from urine samples collected at  the




end of the work shifts.
                                504

-------
To establish whether training actually produced any changes in employees'




work practices, behavioral observations were conducted throughout each




day of data collection.  For each worker, the specific work practice




behaviors were defined in observable terms.  Each worker was watched




during the entire day by an observer who had been trained to recognize




and record instances of these defined behaviors.  However, observational




data were not continuously collected throughout the work day because




there were times during which work practices could not have affected




styrene exposures.  Observational data were only taken during the times




the four employees worked with curing resin.  For example, the sprayers




were observed whenever they were spraying or working inside the spray




booths, but not when they were taking breaks, assisting with.jobs such




as mold waxing, or participating in cleanup operations.  An observer




was positioned close enough to a worker to be able to see well, but far




enough away to not intertere with the employee's work.  Each observer




carried a clipboard and stopwatch.  The stopwatch was turned on, and




recording began, whenever the employee began working, and was turned off,




terminating recording, whenever the employee stopped working for longer




than one minute.  During each 15-second period of the observation, the




observer scored any instances of the targeted behaviors on a behavior




recording sheet.








The accuracy of observer recording was tested through frequent cross-




observer reliability checks.  At unannounced times, a second observer,




using the same observation procedures and  definitions, would make an




independent recording  of an employee's behavior,  simultaneously with the
                                 505

-------
assigned observer.  The two recordings were later compared to determine




the accuracy of the observers and the recording procedures.   Such




observer "reliability checks" were made on an average of 18 percent of




all observations, and were conducted during at least 75 percent of the




days of the study.  Total daily interobserver agreement scores across




all observations were 94 percent for  observations of the resin-chop




sprayer, 98 percent for the rollout person, 97 percent for the gelcoat




sprayer and 100 percent for the mold repair person".  Additional detail




regarding occurrence, nonoccurrence and chance agreement comparisons of




the observational data are available from the authors upon request.









To establish -whether the recommended work practices had any effect on




workers' exposures to styrene vapor, breathing-zone air samples x^/ere




collected by operating pumps at precisely the same times as the behavioral




observations were conducted, that is, only while the employees were




actually working with curing resin.  This procedure afforded a direct




means of assessing the effects of changed work practices on exposure




concentrations during the work periods in which work practice behaviors




could affect styrene exposures by eliminating thos'e times of the workday




when task-specific work practices would not be feasible to help reduce




exposures.  Operating the pumps only while work with available styrene




occurred provided a means to control for effects from variations of




production.  The amount of time a pump was operated should vary directly




with production.  Therefore, this measure, unlike an eight-hour sample,




should provide an estimate of average exposure relatively free from




changes in exposure resulting from production changes.
                                    5Gb

-------
Several eight-hour breathing-zone samples were collected, beginning


with day six, for the resin-cho'p sprayer and the mold repair person,


to provide overall daily TWA's in addition to the work-time exposures.


These pumps were worn by the employees throughout the workday, and


were turned off only during lunch periods or times when the employees


departed the plant.





Eight-hour samples were also taken daily in two areas of the plant which


were not directly involved with any of the specific work areas of the


participating employees.  One pump was located approximately 15 feet


outside of the resin-chop spray booth, and the other was located outside


the supervisor's office, near the middle of the production area.





Air sampling procedures were supervised by an AIHA-certified industrial


hygienist.  Air sampling pumps were calibrated daily at a flow rate


of 50 cc/min which was monitored regularly, and adjusted as necessary.


Air samples were drawn into charcoal-filled glass tubes for collection


of styrene and analyzed by the Utah Biomedical Laboratory using NIOSH

                 9
P&CAM method #127  with ethylbenzene as an internal standard.  Sample


results were adjusted for temperature, humidity and atmospheric pressure.





In order to further evaluate the effectiveness of the program, a 100 ml


urine sample was collected from each of the participating employees


once daily, during the last 30 minutes of the work shift.  Urine samples


were treated with 1 ml 6 N HC1 and frozen within two hours after
                                 507

-------
collection and were analyzed for mandelic acid by high pressure  liquid




chromatography.









Work Practice Training Procedures:  The above data collection procedures




were carried out during several baseline or pre-training days.   At the




end of baseline, one member of the research staff,.functioning as a trainer,




met once with each worker for ten to fifteen minutes prior to the beginning




of a day's work shift.  The trainer explained each of the recommended




work practices and how each could help to reduce exposures to styrene.




If a worker indicated a lack of understanding of a work practice, the




trainer demonstrated it for him or her.  The trainer remained with the




worker for another ten to fifteen minutes, as work was begun, to give




feedback on the use of the work practices and to correct any that were  not




being properly executed.  After this initial training, the trainer visited




each worker at unannounced times once or twice each day, for only a minute




or two, to provide brief encouragement and feedback on the employee *.s




continued use of the new work procedures.









Experimental Design:  To provide information on the extent to which changes




in data could be attributed to training, as opposed to uncontrolled




variables, not all workers were trained at the same time.  After baseline




data had been collected on all four workers for eight days, training was




introduced for two workers, the resin-chop sprayer and the rollout person.




Pre-training data collection was continued for the gelcoat sprayer and  nold




repair person for the next three days, before they, in turn were trained.




This experimental design allowed for a number of important comparisons.
                                   508

-------
Effects of training could be inferred from changes from baseline to post-




training data.  In addition, the fact that the gelcoat sprayer and mold




repair person were not trained until after the other two workers,  allows




their data to serve as controls for the data of the first-trained workers




for three days.  For example, if any changes in the data of the first-




trained workers should be attributable to uncontrolled variables such as




changes in plant policies or weather, these factors might also be reflected




in changes in data of the workers who were not yet trained.  Finally, the




fact that the workers were trained at different times provides a test of




the extent to which the training procedures were effective to produce




repeatable results at different points in time.  This design, called a




multiple baseline design, is frequently used in behavioral experiments   '









Results









Table 1 displays the percent of 15-second observation intervals in which




each of the targeted behaviors occurred during the baseline period




and following training.  With few exceptions the behaviors changed




as desired.  The  exceptions included several practices which were







already occurring at acceptable levels, e.g., the chop-sprayer's spraying




toward himself and turning on the booth exhaust.  Wearing a respirator




while working with resin was not adopted by the chop sprayer even though




it was a highly desirable work practice.  Training apparently
                                   509

-------
induced slight increases in the percent of time the chop sprayer,  the



rollout person, and the gelcoat sprayer were in their work areas but not



working.  This may 'have been due to the tine required to arrange and turn



molds.  The extent to which the number of data points, during baseline and



after training, deviated from the mean of all data points in the desired



direction was compared to the number expected by chance according  to a


                     T ?
binomial distribution  .  The probabilities that such results, for each



work behavior, would be obtained by chance are presented in the right column



of Table 1.







Figure 1 includes graphed data of the percent of observation intervals in



which the resin-chop sprayer and the gelcoat sprayers placed molds in the



spray booths properly, during baseline and after training.  During baseline,



the rate of correct mold placement was low for both the chop sprayer and



the gelcoat sprayers.  When training was given to the chop sprayer on the



ninth day of data collection, his rate of correct mold placement  increased



immediately while that of the gelcoat sprayer remained low until  he was



trained one week later.  Figure 1 provides a representative example of the



way in which the data of the gelcoat sprayer and mold repair person served



as a control for those of the chop sprayer and rollout person during days



nine, ten, and eleven of data collection.  In all cases in which the work



practices changed as desired, the percent of intervals in which they were



occurring remained relatively stable throughout the baseline period before



changing with the introduction of training.  The fact that training, and



the changes of the behaviors, did not occur for the gelcoat sprayer and



the mold repair person until a week after they occurred for the chop sprayer

-------
and rollout person is a good indication that the behavior changes xjere caused




•by training rather than by some unmeasured confounding.  The fact that the




behaviors of the gelcoat sprayer and the mold repair person generally




changed, as had the behaviors of the chop sprayer and rollout person one




week before, once training began for them, is a good indication that the




effects of training are replicable.








The behavioral data, when displayed as in Figure 1, also provide an index




of the rapidity with which the various work practices can be induced.  It




can be seen that the changes in the percent of intervals in which the




resin-chop sprayer and gelcoat sprayer placed molds properly in the spray




booths occurred almost immediately with the beginning of training.  This




was true of most of the work practices that changed as desired.  Exceptions




were the chop sprayer's working with uncovered skin, overspraying and




having the booth doors open while working; the rollout person's working




outside the rollout area; and the gelcoat sprayer's working with skin




uncovered and spraying toward himself.  In the cases in which the workers'




behaviors changed gradually, anywhere from one week to almost three weeks was




required for the full extent of change to occur.









The daily personal styrene exposures for all four  subjects,  taken during




the times they were working as defined above, during baseline and after




training, are presented in Figure 2.  The exposure of  the chop sprayer
                                    51J

-------
decreased from a mean of 150 ppm during baseline to 96 ppm after training,




a decrease of 36 percent; the rollout person from 121 ppm to 70 ppm,  a




42 percent decrease; and the gelcoat sprayers from 210 ppm to 91 ppm, a




57 percent decrease.  A statistically unreliable eight percent decrease




in mean personal exposure occurred for the mold repair person.  The




significance levels of the reductions in exposure, calculated according




to binomial probabilities, were 0.02, 0.01, and 0.'02 respectively for the




chop sprayer, the rollout person and the gelcoat sprayer.









Decreasing trends occurred in the personal samples of the chop sprayer




and the rollout person during baseline.  However, the daily exposures




during this time very closely correlated with production.  The correlation




suggests that production was simply introducing less styrene into the plant




rather than that these two workers were becoming less exposed to available




styrene during baseline.  This also suggests that the method of only




operating the breathing zone pumps during time worked only partially  controls




for variations of exposure due to changes in production.









There were immediate changes in exposures, following training, for the




two sprayers and the rollout person even though the production schedule




was increased.  The new levels of personal exposures remained relatively




stable throughout the post-training data collection.
                                   512.

-------
The eight-hour personal samples and eight-hour area samples indicate that




exposures were well below the Federal standard.  The personal samples




ranged from a low of 28 to a high of 54 ppm styrene, with a mean of 41  ppra,




for the chop sprayer, and from 3.2 to 13, with a mean of 7.5 ppm, for the




mold repair person.  General area samples ranged from 0.10 to 12 ppm.







Urine mandelic acid levels did not decrease for all workers following




training for the recommended work practices.  Micrograms of mandelic acid




per milligram of creatinine were divided by minutes worked because the




amount of available styrene, and consequently the concentration of mandelic




acid in the urine is in part dependent on duration of exposure.  This




index decreased from a mean of .66 during baseline to a mean of .17 after




training (binomial p < .002) for the gelcoat sprayer and from .46 to .30




(binomial p < .03) for the rollout person.  There was no change for the




mold repair person just as there was no change in his styrene exposure. A




slight increase in this index for the chop sprayer may be explained by  the




decreases in exposure concentration being counteracted by increased




exposure times.





Discussion





The training technology was sufficient to induce the desired changes in




most of the work practices and these were correlated with 36 to 57 percent




reductions in exposures to styrene vapor during the times measured for




the three workers who were receiving the greatest breathing zone exposures.




That there was little reduction of exposure for the mold repair person  was




not surprising.  A great portion of his exposure appeared to result from




styrene introduced by processes in other parts of the plant.  Little of




his exposure appeared to result from his occasional and brief work with
                                 513

-------
styrene.  Therefore, only a small percentage of his total exposure could




be avoided by his own work behaviors with the exception of wearing a




respirator.









Mandelic acid levels have been reported to be highly correlated with exposures




to styrene,-*-3>14,15,16,17 an(j they have the potential to reflect ingested




or percutaneously absorbed styrene as well as that inhaled.  Only the data




for the gelcoat sprayer and rollout person reflect a consistent change in




mandelic acid levels.  It is possible that these effects are apparent




because of the magnitude of the change in their exposure, and their use of




respirators.









The striking reduction in mandelic acid of the gelcoat sprayer likely




resulted from his training-induced use of a respirator, the only work




practice he, but none of the other workers, adopted during a substantial




percentage.of time worked with curing resins.  This suggests that wearing




a respirator may be an important work practice whenever high exposures




can not be reduced by engineering controls or other work practices.  In




turn, this observation highlights the importance of the difficulty in




getting some workers to wear respirators.  The chop sprayer declined to




wear a respirator at all times and the rollout person generally used one




only when working inside a mold.









The training carried out by the senior author was sufficiently simple and




straightforward that it could be done by plant personnel.  It should be




observed that none of the work practices were particularly complex and
                                  514

-------
all workers probably already had the necessary behaviors in their repertoires.




Therefore, training was a matter of prompting the workers to engage in the




behaviors.  Once the behaviors were occurring, there was, similarly,  little




difficulty in maintaining them for at least three weeks.  There has been




general skepticism that workers will reliably engage in protective behaviors




over long periods of time.  It remains to be seen to what extent this is




correct in the present case.








The greatest difficulty in validating work practices by measuring the*




extent to which they reduce exposures will probably result from the fact




that there are many other variables that will contribute to the amount of




toxic substances introduced into plant environments.  In the present




research these variables included such things as the ratio of styrene to




resin in the materials supplied to the plant, the amount of catalyst




introduced in the spraying  operations,   the sizes and shapes of the parts




being produced, and the weather.  These variables will change from day-to-day




and will, thereby, produce variations in exposure levels.  Such variations in




exposures will tend to hide effects that result from changed work behaviors.




In addition, whenever reductions of exposures are correlated with changed




work practices, there can be questions about whether the reduced exposures




result from the work practices or from unknown changes in the many other




variables.








In the reported research, the major variation in styrene levels probably




resulted from changes in rates of production.  Sales demands, breakdowns




and numbers of workers present sometimes interacted so that production,




and the amount of styrene introduced into the environment, might vary by
                                   sit
o

-------
a factor of two or three.  For the purposes of this experiment,  controlling




the rate of production was not reasonable because it would have  interfered




with the company's business.  Therefore, this factor had to be controlled




by some rational adjustment of data.  The adjustment took the form of




keeping sampling times proportional to production and weighting  mandelic




acid data by the reciprocal of time worked with curing resin. These




adjustments appear adequate for the exposure data, but partially




successful for the mandelic acid data.









If such measurement problems in estimating exposures can be solved -




and they must be solved to provide empirical bases for all approaches




to reducing exposures to toxic substances - the technology to validate




work practices would appear to be broadly applicable.  Once potentially




useful practices are identified, observational definitions and recording




methods can be developed and the reliability of the measurement  determined




at least for a large class of practices.  This allows for the examination




of the extent to which workers' behaviors change as prescribed.   If




appropriate measures of exposure change, as desired, with changes in the




behaviors, and particularly if this correlation is' replicated over workers,




the usefulness of the work procedures can be determined.  The technology




would appear to be sufficiently flexible to examine the effectiveness of




collections of simultaneously introduced work procedures, as was done in




the present case, or it could be applied to validate a single work practice




such as vacuuming rather than blowint* dust which contained asbestos




fibers18'19.
                                516

-------
If the technology Is applied to collections of work practices, it is



Impossible to infer, from positive results, that a single work practice



contributes to the overall reduction in exposures.  However, this may not



be a serious loss of information unless some of the work practices are so



difficult to implement that including them in the package endangers the



acceptance of the other practices.








Analyses of the .behavioral and exposure data will allow for determinations



of the ease with which work practices are adopted and are useful for



different workers within a single plant, the extent to which different



practices are adopted and are useful within different plants In a single



industry, and the generality of usefulness of work practices over different



industries.  As work practices are validated, they can be incorporated into



training programs.  If necessary, motivation programs can be developed to



encourage workers to use effective work practices.  The research strategy



used to validate work practices- will also be directly applicable to building



an empirical base for worker training and motivation programs.  Measurement



of worker behaviors and exposures will provide benchmarks against which the



effectiveness of training or motivation technologies can be compared and



the only means to validate the programs that are developed.








There has been recent emphasis on the importance of human behavior in



protecting workers from exposures to toxic substances^.  The lack, of a



technology to successfully Influence behavior, has been noted to be crucial.


                           71
For example, Dr. Anita Bahn   has stated,



     "...in general, modification of Individual behavior so as



     to reduce personal hazards is the principal impediment...
                                   51V

-------
     (in the industrial setting) today" (p.12).









It is our opinion that much of the pessimism about the prospects of




influencing behavior has resulted from cases in which there have .been




failures to produce desired behaviors because:  1) training has amounted




to little more than simple communication of  information;  2)  training




methods have not routinely included the extended follow up  necessary to




alter existing habits; or 3) little attention hae been paid to the




importance of motivating the person being trained.  Training and




motivation technology can be made arbitrarily powerful.   In some cases,




that power can be achieved without undue expense and complexity.   In




all cases, an appropriate strategy to validate proffered  work practices




is at hand.
                                 516

-------
                               REFERENCES









1.  Criteria for a recommended standard — occupational  exposure  to




    asbestos.  HEW Publication No.  (HSM) 72-10267,  1972.









2.  Criteria for a recommended standard — occupational  exposure  to




    inorganic arsenic, new criteria.   HEW Publication No.  (NIOSH)




    75-149, 1975.









3.  Fitch HG, Hermann J, Hopkins BL:   Safe and unsafe behavior  and




    its modification.  J Occup Ned  18: 618-622,  1976.









4.  Stewart RD, Dodd HC, Baretta ED,  Schaffer AW:   Human exposure to




    styrene vapor Arch Environ Health, 16: 656-662, 1968.









5.  Dutkiewiez T and Tyras G:   Studies on the skin  absorption properties




    of styrene in human beings.  Gigiena Truda i Professional'nye




    Zabolevaniia, 12(4): 35-39, 1968.









6.  Occupational Safety and Health  Administration,  Department of




    Labor:  Code of Federal Regulations, 29CFR 1910.1000,  Table Z-2.









7-  Modern Plastics Encyclopedia, Volume 51,  Number 10A, New York:




    McGraw-Hill, 1974.
                                 519

-------
 8.   Hopkins BL, Hermann JA:  Evaluating interobserver reliability of




     interval data.  J App Behav Anal, 10: 121-126, 1977.









 9.   NIOSH manual of analytical methods (2nd Edition), Fart 1,  Vol. 1,




     127-1-127-7, Cincinnati:  NIOSH, .1977.









10.   Baer DM, Wolf MM, Risley TR:  Some current dimensions of applied




     behavior analysis, J App Behav Anal,  1:  91-97, 1968.









11.   Hersen Mil, Barlow DH:  Single case experimental designs.  New York:




     Pergamon, 1976.









12.   Feller W:  An introduction to probability theory and  its applications.




     New York:  John Wiley and Sons, 1957.









13-   Bardodej Z, Bardodejova E:  Biotransformation of ethylbenzene,




     styrene and alpha-methylstyrene in man.  Amer Industr Hyg Assoc J,




     31:  206-209, 1970.









14.   Ohtsuji H, Ikeda II:  A rapid colorimetric nethod for  the determination




     of phenylglyoxylic and mandelic acids.  Brit J Industr Med,




     27:  150-154, 1970.









15.   Gotell F, Axelson 0, Lindelof B:  Field studies on human styrene




     exposure.  Work F.nv, 9:  76-83, 1972.

-------
16.  Ikeda M, Immura T, Hayashi M, Tabuchi T, Kara I:  Evaluation of

     hippuric, phenylglyoxylic and mandelic acid in urine as indices of

     styrene exposure.  Int Arch Arbeitsmed, 32:  93-101, 1974.




17.  'Engstrom K, Harkonen H, Kalliokoski P, Rantanen J:  Urinary mandelic

     acid concentration after occupational exposure to styrene and its
                                                »
     use as a biologic exposure test.  Scand J Work Env-Health, 2:

     21-26, 1976,




18.  Lee SL:  Removing dust from brake assemblies during vehicle

     servicing - alternative cleaning methods.  Aon Occup Hyg, 13:  33—36

     1970.




19.  Okawa MT, Polakoff PL:  Occupational Health Case Report - Number 7.

     J Occup Med, 16:  350-354, 1974.




20.  Lassiter DV:  Prevention of occupational cancer - toward an

     integrated program of governmental action, in V Saffioti and

     JK Waggoner (Eds.) Occupational Carcinogenesis, Ann New York Acad

     Sci, 271:  208-213, 1976.




21.  Bahn'AK:  Epidemiology, in Occupational Health and Safety Symposia,

     HEW Publication Number (NIOSH) 76-136, 1976.

-------
                              TABLE 1

THE MEAN PERCENT OF INTERVALS OF OCCURRENCE OF EACH OF THE WORK

BEHAVIORS DURING BASELINE (BL) AND AFTER TRAINING (TR), FOR EACH

OF THE FOUR WORKERS AND PROBABILITIES (p)  OF OBTAINING SUCH

CHANGES BY CHANCE.  N.S. INDICATES -THAT THE CHANGES WERE NOT

STATISTICALLY SIGNIFICANT.



Chop Sprayer                                BL      TR     p_

Wearing respirator while inside spray
                             booth           0       0    N.S.

Keeping skin covered                        24      76    <.03

Staying in spray booth when not working      5      12    N.S.

Activating booth exhaust ventilation       100     100    N.S.

Keeping booth doors closed while spraying   42      93    <.10

Placing molds properly                       8      93    <.001

Spraying toward exhaust ventilation         70      99    <.001

Turning molds                                2       4    <.04

Overspraying unnecessarily                  15       4    <.03

Spraying toward self                         .3      0    N.S.

Spraying toward others                      23       2    <.001
                               522.

-------
                        (TABLE 1 CONTINUED)

Rollout Person                                 BL     TR     p

Wearing respirator while working with
    uncured resin                               0      4     <.01

Working inside molds without respirator         A      .2    <.01

Keeping skin covered                            3     98     <.001

Performing rollout in rollout area              4     93     <.001

Remaining in rollout area when not working      2      3     N.S.

Exposing self to resin spray                   13      0     <.001

Using floor fans properly                      11     98     <.001

Staying upwind of part while rolling out       50     94     <.001

Turning molds                                   0      1     N.S.



Gelcoat Sprayer                                BL     TR     p

Wearing respirator while inside spray booth     2     99     <.03

Keeping skin covered                            0     89     <.03
                                                   1
Staying in spray booth when not working         9     10     N.S.

Activating booth exhaust ventilation           99    100     N.S.

Keeping booth doors closed while spraying       2     97     <.03

Placing molds properly                         10     94     <.03

Spraying toward exhaust ventilation            58     97     <.03

Turning molds                                   1      7     <.03

Overspraying unnecessarily                     10      3     <. 10

Spraying toward self                            1      .2    N.S.

Spraying toward others                          0      0     N.S.

-------
                       (TABLE 1 CONTINUED)
Mold Repair
BL     TR     p
Wearing respirator while working with
    uncured resin
Keeping skin covered

Remaining in repair areas when not working

Holding head too close to resin applications  6

Leaving resin containers uncovered

Performing repairs within the repair area

Using floor fans properly

Staying upwind of uncured resin
    applications

Turning molds
0
92
9
6
4
76
1
55
0
0
99
6
3
3
95
100
97
.3
N.S.
N.S.
N.S.
N.S.
N.S.
<.05
<.05
<.05
N.S.

-------
   lOO-i
                 BASELINE
          CHOP SPRAYER
                                            POST-TRAINING
   50-
g
    0-
  : lOO-i
UJ

O

Ctf
    0-
          jprayer-1
sprayer-2
spraysr-3
                                        10
                                                         IS
                                   DAYS
Figure  I
                                      Ox ""x
                                     & O

-------
  280-1
  140-
                              POST-TRAINING
        i  i  i  i   i  i  i  i I  i  i  i  i   i  i  i   i  i
                         DAYS
Figure  2
                          526

-------
     FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

        PROGRESS ON 3OINT ENVIRONMENTAL AND

            OCCUPATIONAL CANCER STUDIES
                Wednesday Afternoon, May 7
                       SESSION A


WORKING GROUP ON PROBLEMS, NEEDS AND NEW DIRECTIONS

               FOR EPIDEMIOLOGY STUDIES
                 SESSION CHAIRPERSONS
                   Dr. Kenneth Bridbord
      National Institute for Occupational Safety and Health

                   Dr. Joseph Fraumeni
                 National Cancer Institute
                             527

-------
SESSION A  -  WORKING GROUP  ON  PROBLEMS, NEEDS AND NEW DIRECTIONS
FOR EPIDEMIOLOGY STUDIES
Dr. Mason, NCI:  I understand that one of the things we  would like to talk about is
future studies that  we would like to have funded through one or another or  both of
the collaborative arrangements.  One would be the funding of follow-up studies in
places that  we have  identified through  some  of our  mapping  projects with inter-
mediate studies,  whether they be death certificate, hospital record room searches
and some  analytical case control interview or  cohort interview studies in the field.
Also,  if you will, the potential for using some of these monies to get a better handle
on levels  of  exposure to both qualify and quantify exposures in communities,  to
pursue some of the more clinical laboratory investigations, and these types of studies.

Dr. Riggan,  EPA:  I share Dr. Mason's interest.   Another area for  collaborative
studies  would  be  the  environmental  and occupational  health  impact  from  the
synthetic  fuel program.  We need to identify  both  occupational and environmental
cohorts who  may be followed over  time.   A collaborative study involving  these
agencies may be  the only way to establish cohorts with any hope of following them
over the time period required for meaningful results.

Dr. Bridbord, NIOSH:  I might add that we already have plans for  similar studies
actually under another  collaborative  program  with  EPA in the area  of energy.  So
there  are  and have been some  discussions about developing registries of exposed
workers and being in a position to  follow those over time.

Dr. Riggan, EPA: I was  not aware of the collaborative program of NIOSH and EPA in
the area of energy.

Dr. Brown,  NIOSH:   We  have some  information  on that, a cohort  study for coal
liquification and coal gasification. A couple of years ago, we checked into trying to
find a cohort study and were fairly unsuccessful in finding a good population.

Dr. Riggan, EPA:  This, I think is true today, but it may be very important in a few
years. I am interested in estimates of exposure both qualitatively and quantitatively.
This is a problem  that interferes with environmental  studies.  While it may be less of
a  problem with  occupational  studies,  it is  a  real  problem  with  environmental
epidemiology studies.

Dr. Fraumeni, NCI:  You are talking now about all studies across-the-board?

Dr. Riggan, EPA: I am talking about epidemiology studies, all studies,  where we have
a problem, especially with  air pollution and environmental exposure, what are  the
qualitative and quantitative  aspects.

Dr. Fraumeni, NCI:  How long have the coal gasification plants been in operation?

Dr. Brown, NIOSH:  There are some in Europe that have been in operation for a long
time,  and  that is one  possible place to try to get a cohort. I think there were some
pilot plants started  a long time ago in the United States but never any that were  put
into full-scale capacity with a large enough population.
                                     523

-------
Dr. Riggan, EPA:  Yes, but several are in the planning stage for this country where
baseline data could be collected on cohorts.  These could be monitored for changes
which may occur.

Dr. Brown, NIOSH:  I am just saying for a retrospective cohort  kind of study it would
be very difficult.

Dr. Riggan, EPA: There are some in Poland - Czechoslovakia.

Dr. Brown, NIOSH:  I think Germany ran her war machine on coal gasification.

Unidentified  Speaker:   The biggest ones now are in  South Africa; most of their
gasoline is produced through coal liquification.

Dr. Bridbord, NIOSH: We sent a team  of  people to South Africa about two years ago
looking at that possibility, and the records there and other complicating  factors made
it uncertain as to whether you could effectively do that.

Under PL480 there  is a cooperative  project between  EPA and  the Department  of
Energy and NIOSH.  I think it is in Yugoslavia, but it may have been Czechoslovakia,
an ongoing plant which offers promise for looking at the long-term implications.

In terms of this country the concept would be to basically build  a cohort looking  at
the pilot and demonstration plants and over time, as  Dr. Riggan said, to have some
baseline data, and thus be able to make some evaluations.

Dr. Sloan, NCI:  It  will take you a long time though to get results; if you can capture
something from those that have been ongoing for a long time, it will be helpful, too.

Dr. Brown, NIOSH:  But you could certainly do the industrial hygiene work.

Dr. Riggan, EPA:  If we don't start planning now, in 10 or 25 years we will be where
we are today.

Dr. Mason, NCI:  If  I might, since we have sort of gone away from introductions and
gotten into topics, one of the things that  I believe we would  be well advised to take
advantage of is the newly formed National Death Index.

We are in a position to set  up prospective studies, to collect those  14 data elements
that are going to be maintained and taken from every death certificate in the United
States from 1979 forward.   We are going to be lobbying very strongly to push them
back  prior to 1979.  However,  we are currently in the  position of ascertaining  vital
status and obtaining death certificate numbers  if we  have name, social security
number and place of birth and date of  birth.  I think it behooves us to no longer argue
that we do not have the wherewithall  to  follow large groups of persons.  I, for one,
would like to propose the registration of all persons resident in  the City of Duluth,
Minnesota, in 1970, since the question is  still unanswered as to  whether or not the
ingestion  of asbestos fibers  does, indeed, have a long-term carcinogenic risk.  We
know what happens when you inhale it.  I  believe that  there  does exist  a mechanism
for collecting the required information from residents  of Duluth that  would be cost
effective.   It cannot  cost that  much to move  parallel with  the census to get
information with regard to individuals who were resident. We need to get informa-
tion with regard to their duration of residence, and to  have the basic data such that
perhaps in 1985 and again in 1990 we can  ascertain the vital status of this population.

                                     529

-------
I think that this  is the type of project which we need to fund because you are not
talking about more than several hundreds of thousands of dollars relative to the total
budget which is small, and you are talking about the potential societal impact which
is quite large.

Dr. Bridbord,  NIOSH:   One can  make  those  same  arguments when you are talking
about  any new technology,  for  that matter, what is coming on  board,  such  as
recombinant DNA or various changes in the electronics industry, etc.

Dr. Mason, NCI:   Baseline  data needs to  be collected  from  persons  residing  in
selected  communities  which have realized exposures of specific interest.  You are
going to be looking at levels of exposure. Thus, you are going to have reasonable  dose
information,  and  I think that if  we  put in place a way to  efficiently follow these
persons,  we could make some  very strong points with regard  to  funding for other
studies.

Dr. Sloan, NCI: There is merit in following perhaps, through the death index or other
ways, the changes that may occur  from a major shift from oil  to coal in terms  of
human consequences of acid rain and whatnot; we are  apparently going to see a big
change in this country in what health effects we may expect.

Dr. Mason, NCI:   If we could prioritize the selection of places as a function of some
baseline  data  and baseline risk, yes.  I think almost anything that lends itself  to a
testable  hypothesis which  is consistent with either a known human response  or a
suspected human response  in relationship to laboratory findings is something which
we  must pursue.  These are the ones  that have the  potential to affect large  numbers
of persons in this  society.

Dr. Riggan, EPA:  I listed coal gasification  and liquification as one source of exposure
to toxic  chemicals which may expand very rapidly  in the near future; however, Dr.
Mason's proposal is more general.

Dr. Brown,  NIOSH:  The same sort  of thing could be done in the nuclear  industry
where people want to follow workers in nuclear power plants or areas around them.

Dr. Mason, NCI:  If you wanted to do an across-the-board study of every person  who
has ever  worked in a uranium mill in the United States  you could do it, and there are
some  recent changes with  regard to the Social Security Administration, independent
of any sort of right to access to their data which argues that if a  place is no longer
doing business  then the equivalent  to the  corporation  that  the Privacy Act is to an
individual no longer applies such that they  can release  to someone who is interested
in doing a study the names of every person who ever worked for that place.

So, you have a way in which to build large  rosters of persons, and I think if  you  look
at the massive class action suits that  are being filed i,n the names of miners and
millers _and if you look  at the fact  that a number of us have worked in the  area  of
radiation  carcinogenesis,  it does, indeed,  fit in  very nicely.   I think  it is an
appropriate  utilization of funding from the NCI/NIOSH collaborative agreement.   I
would like to approach the Social  Security  Administration  and obtain employment
rosters for every place that is no longer  doing business.  This information is sufficient
to follow  every one of these persons  with  the National Death Index, and I think  that
that is a reasonable project for  us to consider.

-------
Dr. Blair, NCI:  NCI is attempting to do that with the fur and leather industry.  There
are, however, other possible study approaches  using Social Security.  One method
would be a case-control  approach where work  histories for each study subject are
determined from SSA files.  I think NIOSH gets such information from them now, but
it is expensive.  This might be an appropriate time to launch feasibility  studies on
particular cancers to evaluate the usefulness of this aproach.

Dr. Mason, NCI:  I really  do think that this  is what this collaborative arrangement is
all about - to provide a reasonable level of funding to support studies which we are
interested in collectively.

Dr.  Blair,  NCI:   The Social Security Administration has data  valuable to  many
governmental organizations. It  is important to press  this agency  and to keep on
opening doors to obtain acccess.  There is congressional pressure now, to aid us, and
we want to maintain the momentum.

Dr. Mason, NCI:  I do believe that if it came from  us collectively with a  number of
specific types of studies that we are interested in we can lobby for it.

Dr. Bridbord, NIOSH:  I personally believe that  the more studies we could do  which
are truly collaborative, with people in both institutes working on them, is laudable.

Dr. Mason, NCI:  I agree.  Let us say  there is  something else that you would agree to
pursue; then it comes down to a policy type of  decision.  Is it more appropriate or less
appropriate for staff  members of  the  Cancer  Institute  to be involved in that
particular type of study?

I argue that if it has the potential for any carcinogenic effect, we should be involved
whether  or not that is the first priority.  There is a study that I would like to pursue
now and that is the epidemiology of occupational  exposures as they relate to the
general health of workers as well as their reproductive outcomes.  The effect upon
the fetus could be spontaneous abortion, teratogenic or carcinogenic.

Dr. Bridbord, NIOSH:  I would second that  idea and probably expand it to include
reproductive outcomes of  working populations, whether those were the result of male
or female exposure.

Dr. Mason, NCI: I agree.  We are currently negotiating with  some of the big unions to
get access to sufficient data in  order to permit this type  of study,  and  I think we
should pursue it.

Dr. Bridbord, NIOSH:   Here is an example on  the  NIOSH  side.  We are already
committed to a  major expansion of our own efforts, and it does not make a lot of
sense to have independent efforts. We should be sitting down and coordinating and
collaborating, and to the extent that it is reasonable, using some of the collaborative
money funds. We would  probably supplement that  activity from our base funds as
well.

Dr. Mason, NCI: I have members of my own staff who are ready, willing and capable
of working in this particular area, and I would love to have you exchange with me the
names of people that I should be talking to on this effort.

Dr. Chu, OSHA:  I have a concern here that needs to be addressed  in terms of the
overview of  these  studies  because  certainly now  that I am at OSHA  there is a

-------
question as to having the resources  available.   I am in the business of regulating
compounds and creating standards.   Somewhere along the line,  at  some juncture,
there should be a resource or an input available so that if OSHA was interested in
regulatng four or  five compounds that it have a resource available so  that it can say,
can we institute a study in these areas?  If the activities that you are talking about
are more fundamental or if this is a resource where  regulatory agencies have some
concern like the NTP and get these things done, then the issue should not be, let us
collaborate and get  something done.   It  should  be a question of going  to  the
regulatory  agencies and asking them what kind of priorities they have in terms of the
substances  that they  need studies on. Is this a forum for regulatory agencies that
come in? Is it a research forum for a better understanding of collaborative projects?
I need to get a clarification here.

Dr. Mason, NCI: The program is an attempt at all things for all people.

Dr. Bridbord, NIOSH:  First of all, in terms of the regulatory agency issue, we already
did program as part  of  this conference one of the three workshops to specifically
focus on that issue, among others, and that is Workshop C.

The other points to note are that there are already  a number of other infrastructures
through  which these  communications are exchanged,  and  a  number of  us  sit  on
different groups or at least have sufficient knowledge about those groups that  we are
aware of that within, actually not just OSHA.  It is really the Department of Labor
because there are three assistant secetaries involved.  There is a whole infrastructure
called the  NIOSH Planning Group where  we do sit down and  talk  to  representatives
from  all three assistant secretaries.

Dr. Chu, OSHA:  I am asking what is the role of this forum towards getting this done?
Does  it have that  kind of responsibility or doesn't it.

Dr. Bridbord, NIOSH:  This forum is meant  to  consider those issues among other
things, but the main thrust  of this forum is really  to build a  structure, if  possible,
that is as much a grassroots collaboration as it is something that is totally driven by
the particular needs of the more formal organizational structures.

Certainly the issue of the needs of the regulatory agencies is important.  In the case
of EPA there is, as you know, the issue of how EPA scientists relate to the needs of
the various  program offices, and  in the reproductive area  we talked about  the
interest between  NIOSH  and NCI potentially, but there is a whole  number of reasons
why  I would imagine people in  the EPA  Office of Toxic Substances would be very
interested in any collaboration.

Dr. Chu, OSHA:  That is a prime example of  where a regulatory agency, OSHA, has
already gone to court with regard to discrimination in the workplace  on this issue in
terms of the reproductive effect.

Dr. Mason, NCI:    Really what you  want  to know  is  whether  or  not, if  in your
particular set of  circumstances there was a sufficient body of knowledge to argue
that  there  should be a study of this compound, that you should be able to come to a
forum such  as this  and ask whether or not there are  sufficient numbers.  You have
identified it somehow, whether it  is through a  particular  plant  or  something,  but
maybe you  don't have sufficient numbers to really get your hand on it.  Maybe  it is an
alert  clinician response or  whatever, and now  you  want a larger  study  which is
industry  wide with regard to that  particular  exposure and with  regard  to,  in this
instance, reproductive outcomes.

                                      532

-------
Dr. Chu, OSHA: My question is, is this a resource that is available for us to come to
with these kinds of questions or is it not a resource. I am  trying to simply identify
that capability.

Dr. Bridbord, NIOSH: I think the answer to your question is in principle, yes, and that
this would be a matter of a case-by-case discussion among the various parties.  There
are many reasons, both technical, logistical, resource, et cetera, where an answer to
a given question may be yes or may be no or it may be let us  talk about it next year.

Dr. Mason, NCI: I  think we would be most interested in having that type of exchange.

Dr. Bridbord, NIOSH: To the extent that from the OSHA perspective and the NIOSH
Planning Group  within  the Department of Labor  and the other formal mechanisms
that are already established, those needs do at least  surface in these discussions, and
a good example would be in the area of reproductive effects.

Part of our  desire  to  move  into that area in addition to our  own  interest  and
recognition is the driving force out of the Department of Labor that encourages us to
recognize the importance of this.

A lot of discussions will come up as to which populations we look at, which exposures
are more important, and  all the issues that surround why you study  one group or
another group, but in principle that already is the driving force.

I am more than glad that Dr. Mason raised the point independently because that, from
my perspective, was one  that  I wanted to suggest  as at least  one of  the potential
areas where we  might somewhat modify our existing situation to at least allow some
case-by-case possibilities to move in that direction.

Dr. Sloan, NCI:  We are in  the position of taking the results of  what a good many of
you do and trying to apply them in prevention, diagnosis and treatment of patients
with cancer,  not doing the  research in  diagnosis and  treatment,  but applying what is
coming out of the research field that is ready for application.

We have a major responsibility for education of practicing physicians and organizing
resources in communities and through cancer centers to deliver better care.

One of the areas we have been really concerned about recently is trying to look ahead
and estimate  the burden that asbestos-related  disease  is going to represent to this
country over the next 20 years.

I don't believe anyone has  a very good idea  how serious that burden is going to be,
how many cases  of mesothelioma  we  are  going  to  have,  how many cases  of
asbestos-related lung cancer, and that  there are many populations around the country
which have really not been studied so that we don't know what the long-term effects
of their exposure would be; brake band workers would be one. I  think EPA is going to
try to do something about that, but I don't  really know.

We have heard a lot in the last couple of years about the hazards of  deteriorating
asbestos in schools, but we don't have any real evidence that anyone has ever had his
health endangered  by exposure through  a school situation.
                                     533

-------
Perhaps  custodial workers  and  others  who  have had  more  exposure to asbestos
through the maintenance work they have to do might be  groups that could be studied
more specifically.

So, we would like to stimulate or encourage the organization of more epidemiological
studies on  any kind  of cohort that can be identified that would  help  us  assess the
burden we have to look forward to and prepare for in the  future.

Dr.  Mason,  NCI:   Dr.  Sloan,  do  you  think  that  these talks of  collaborative
arrangements would  be a reasonable forum to consider research in behavior modifica-
tion? If  you think in terms of  one of the greatest risks which is  cigarette smoking,
and if you  think in terms of your all but impossible task of communicating to people
that they should quit, as  well as let us assume  that in the  next several years we are
going to  be, you know, really concentrating on nutrition and things like that, we are
going to  have to go out and start, if you will, preaching  to people that they need to
change their lifestyles.  Do you see that as the type of thing that would come under
this collaboration.

Dr. Sloan, NCI:  I am not sure because I have not been a part of the discussions of this
collaborative  group  before and I don't know whether behavioral research is part of
your area  of  responsibility or not.  The  smoking problem or stopping  smoking is
certainly the  one single  thing that people could do  that  would  help  them more in
preventing and slowing the development of asbestos-related disease  than anything
else.  It  has no effect on  mesothelioma, but  there  is a major program  in smoking
abatement organized under Dr. John Pinny in the Surgeon General's Office, and there
is a coordinating focus for  all of  NIH in the Cancer Institute which Dr. Diane Fink is
running at the moment, and I don't know how much, in addition  to that, is appropriate
to carry  out under this forum.

Dr. Bridbord, NIOSH: I think the  collaborative nature of these programs would by and
large have  to  respect and understand  the differing mandates,  with  the different
organizations  and not to  question  the  importance of  the  smoking issue which is
certainly a major public health hazard, but I would think that the EPA people  would
see their mission primarily in terms of the toxic  chemicals and the NIOSH mission is
primarily focused on the chemical and physical agents in the workplace.

There are many, many examples of  where NIOSH recommendations have  considered
the smoking issue, not only in terms of  the interaction but, also, some of  the safety
and other problems that  surround that in the workplace, but I  would be reluctant to
look at this particular pool of resources as the main focus for that work as opposed to
some of the others.

In the cancer  control area  the argument I would  make is that the control technology
issue, the control technology assessments and engineering studies in terms of actually
translating the information on toxic exposures to a prevention outcome is something
that I would suggest  we talk about as a possibility, but I think if you are talking about
studies that look at work practices and  really try to assess how effective  they  are in
reducing  personal exposure  in the workplace, there is one example.

Any of  our epidemiology  studies or cross-sectional studies  that  look  at  a toxic
exposure must consider the other  confounding factors.
                                    534

-------
Dr. Spirtas, NCI:  Dr. Sloan are you trying to identify additional occupations which
may be beginning to show an asbestos-related health problem.  Is that what you are
saying?

Dr. Sloan, NCI: If there are some that we have not heard about yet.

Dr. Spirtas, NCI:  That is an argument to  go ahead with the mesothelioma registry, to
find cases of mesothelioma and trace back over time to see where they worked, and
if something crops up like custodial workers.

Dr. Sloan, NCI:  What  mesothelioma registry are you talking about particularly, one
that doesn't exit yet or one that does?

Dr. Spirtas, NCI:  I am  talking about one that is new - in the formative stages.

Dr. Sloan, NCI: Under  whose aegis?

Dr. Brown, NIOSH:  It is proposed for NIOSH.  Did you start it when you were at
NIOSH or have something to do with getting that started?

Dr. Spirtas, NCI:  I was involved in getting it started.  We did not see it from the
point  of view of research as being a strong research effort because of the ubiguitous
nature of asbestos.

We know that asbestos causes lung cancer and mesothelioma, but if  there is a reason
such as the one that you have just described, there may be  workers in  occupations
who are in need of education or in need of further protection, and we should identify
those unknown occupations such as custodial workers, people in school buildings.  To
my thinking, Dr. Sloan's discussion of intervention strategies strengthens the argment
to push ahead.

Dr. Sloan,  NCI:  There is  also the fact  that there is some pretty  good work going
forward now on developing better methods of treating mesothelioma, and there are a
few long-term survivors  emerging from these trials that may make it very important
to be able to identify mesothelioma cases as early as possible.  I think we, also, are
anxious to  know, I think from your death  record list, Dr. Mason, that we need to know
where in the  country  they  are occurring and what the  economic burden  is on the
community, what the treatment resources  are going to be that are  required in these
areas.  The SEER program is showing  that there is definitely an  increase in the
shipyard areas which you might expect.

Dr. Brown, NIOSH:  Don't  you think that the National Occupational Hazards Survey
would be of some use  to try to get an estimate of the number of people that have
been exposed to asbestos and where they have been exposed. NIOSH is just planning
another round of the NOHS project.

Dr. Sloan, NCI: I assume you are going to do that.

Dr. Mason, NCI: What is the timing on the third one?

Dr. Bridbord, NIOSH:  It was supposed to start as a substantial  activity this summer.
We are facing the same issues, I think everybody else is, with perhaps the  exception
of some parts of EPA,  and that is the crunch on the ability to hire people, particulary
permanent people.  I think that has caused our progress to have been not quite as

                                    535

-------
active as we would have liked, but there is no question but that we are moving ahead
with the second survey, with some improvement in it, and it will be an  important
piece of information, I think, for all of the agencies.

Dr. Sloan, NCI: Asbestos will be a part of that?

Dr. Bridbord, NIOSH:  No question about that.

Dr. Spirtas, NCI:  I cannot tell you off the  top of my head  how many places asbestos
was seen in the first survey, but it  was a substantial number. In fact, it was not just
in shipyards, but asbestos is also  a component in insulation and other products. In my
opinion, one of the valuable things that comes out of an exercise like NOHS is not so
much picking numbers but it is telling you that asbestos was seen in a factory where
you would never have suspected  it.  Thus, there are reasons to explore the  NOHS file
for qualitative information that  may be as  important as the quantitative information
that comes out of it.

Dr. Sloan, NCI: Could I just ask one more question about asbestos?  The problem of
removal of deteriorating asbestos from schools is only a small part of a huge iceberg,
because an infinite number of buildings in this country  were treated in the same way,
including some of our most elegant apartment houses and whatnot, and we really have
very little knowledge about  the reality of this hazard.  Is there any way that through
your collaborative efforts  you could address this in any more positive way?

Dr. Bridbord,  NIOSH:  My guess is that from the occupation  perspective it would  be
fairly difficult. I  think that is a question  of the indoor  air pollution issue and how
EPA feels about that matter.

Dr. Riggan, EPA:  We are concerned and interested in indoor  air pollution.  Increased
insulation and reduced air  exchange  in the living areas may create air pollutant levels
consideraably  above the environment ambient levels.

Dr. Bridbord, NIOSH:  Actually, EPA has some studies of indoor air pollution, I think,
and carbon monoxide?  No?

Dr.  Riggan,  EPA:   EPA  has had  several  studies  including one at the  Lawrence
Berkeley Laboratory in California.  At  present  EPA  has  a  contract  with TRC  on
estimating individual's total  exposure.

Dr. Brown, NIOSH: We do a number of different  types of industrywide studies. Most
of our studies  historically have been retrospective cohort kinds of studies on cancer.

Dr. Fraumeni, NCI: Could you explain what you mean by industrywide?

Dr. Brown, NIOSH: Industrywide means,  if  we want to  study a certain type of agent,
such as trichloroethylene  or asbestos, we go out and we look for a population in  an
industry, not a specific industry, but an industry where there is exposure to  this agent
and try to generalize the outcome that  we see from  this exposure industrywide.
Maybe that is  the wrong interpretation of industrywide.

Dr. Fraumeni, NCI: Do you always start with a particular agent?

Dr. Brown, NIOSH: No.


                                    536

-------
Dr. Fraumeni, NCI:   For example, would the rubber industry investigations by the
University of North Carolina and by Harvard be called industrywide studies?

Dr. Brown, NIOSH:  Involving the rubber industry, yes.  Our studies start  from an
agent or from a certain occupational group. One thing that we are getting into lately
is  the  reproductive studies that Dr. Bridbord mentioned, and that is where people
should have some collaboration because the analytical methods are not well defined
yet. Nobody  really knows good ways to analyze all this data that has come in.  There
are thousands of variables that one has to look at and control for.  I think that would
be a good  topic to have  some collaboration on,  and I think in general the federal
agencies should  collaborate more  on analytical techniques  and  be willing to share
their  computing programs and expertise so that when one agency comes up with a
finding, the analytical techniques  that were used can  be accepted throughout the
Federal government.

Dr. Mason, NCI:  Yes, that is why I am, personally, very pleased that  your life table
analysis system is now sitting at the Parklawn computer because I am getting it. You
have treated  it as a black box for a long time until such time as you got it to a point
where you felt more comfortable with the whole thing, and the modifications of that
program relative to some other life table programs are something which we need. I
mean you allow a person to come in  and go out as a function of breaks in exposure.
You don't  have to make the  oversimplifying assumption  that  once exposed  you
continue exposure throughout that  particular employment which we know is not true
but which we have had to make the  assumption because the black box that we have
does not permit  that type of modification. So, I think we are making good strides.
Dr. Kreitel uses all of my mapping programs, and he  rides free  on my interagency
agreement  with  NOAA in order to make maps.  There is a growing interest in the
sharing of information and I am encouraged.

Dr. Bridbord, NIOSH:  I think particularly in the area of reproductive effects, this is
going to be  one where there will be a number  of areas of progress in terms of
methodology  in the next number of years, and that is an important one to be talking
about.

Dr. Mason, NCI:  Not  only in methodology.  It is  something which needs to have a
clinical laboratory  component.   It  is  something  where you not only need  good
industrial monitoring, but you need some very specific measures on the person, male
or female, and there are some fundamental questions that need to be addressed with
regard to the availability  of amniocentesis to working women and how many  of them
do take advantage of that and  whether or not the  amniotic fluid would not be a
reasonable  specimen to look at from the standpoint  of certain levels.  I, personally,
think  yes.    However, historically amniocentesis  had  a very strong  social class
gradient.  If you were not better off, you did not avail yourself of it.  So whether or
not we have made strides along those lines, whether or not 90 percent  of women who
are working  in,  let us say the  chemical  industry or whatever, would  indeed avail
themselves of this routinely  as  part of their health program at the company is a
question thjat I think needs to be addressed.

Dr. Bridbord, NIOSH:  Or just even  doing, in effect, case control studies, just even
looking at  exposure, getting that  technique for  various  reasons that are  already
indicated.

Dr. Mason, NCI:  As long as you can identify the characteristics of the  persons and
you can characterize that  population or that subpopulation  that they come from, then


                                      537

-------
it is reasonable to look at these particular comparisons, because I do think that it is a
good  thing from the  standpoint that the fluid can be  looked  at in a  reasonably
complete manner, and it has not been done, not at all uniformly.

Dr. Bridbord,  NIOSH:  Similarly, just understanding how to measure  people's sperm
counts and motility and what all that really means is important.

Dr. Blair, NCI:   I have a  couple of  general categories  of  projects.   One  we have
already talked about is maintaining some momentum to gain access to Social Security
files to enhance  our  research capabilities.  It seems that this group might  play an
important role because there are several government agencies involved.

Another is making an attempt to build a  comparison population of working persons
that can  be  used in occupational mortality studies.   I am sure  every agency has
thought about this at various times, but it is difficult to allocate the necessary time
or the resources. There are. a number of  cohort mortality studies completed where
there were not striking findings.  If  pooled, these  might provide a suitable worker
comparison population which would modify the  "healthy worker" problem  that arises
from use of the general population for comparison.

Dr. Fraumeni, NCI:  There are studies already done on cohorts whose risks were not
exceptional?

Dr. Blair, NCI:  The idea is to pool many completed cohorts.  You probably would not
want to include insulators because of their high  lung cancer rates, but others where
the findings were not so striking could be used.

Dr. Bridbord,  NIOSH:  We have already  started in the area of respiratory disease
where we are developing  out of our  laboratory  in  Morgantown a blue collar worker
control population to look at respiratory disease, but I think  your point is well taken,
and actually the  few data that are available on  groups that are, as far as we know,
unexposed  might  well  be confounded  by other things, such as alcohol, cigarettes, et
cetera. It would be hard to tease these apart, but they suggest mortality ratios, even
as low as 50 to  60 percent.   However, we must always  use the United States as a
whole as our control  group which would be analogous to trying to at least assess the
quantitative  impact  of  cigarette smoking,  looking  at a  group  of  smokers  and
comparing them  with a group that included 30 to 40 percent of smokers. I mean that
is, in effect,  what  we are doing in all of our occupational  studies at this  point; in
terms of identifying a high-risk situation, I think you can still do it.  Where you are in
the margin you lose something and then in terms of the  quantitative impact we  may
be understating that  by 25 to 50 percent.

Dr. Blair, NCI: Because various agencies already have suitable cohorts available, you
might be able  to put something together.

Dr. Brown, NIOSH:  I was  approached by Stanford Research Institute and they wanted
to do that very thing as a grant.

Dr. Blair, NCI:  My guess is  it ought  to be funded through a contract.  It would be a
time-consuming job and that  is why it does not get done.  Government agencies seem
to be the logical  group to initiate the effort.

Dr. Spirtas, NCI:  It might not be such a terrible job to utilize the Social Security's 1%
CWHS file.


                                     536

-------
Dr. Blair, NCI: Yes, and it could be that academic institutions would also contribute
cohorts.

Dr. Fraumeni, NCI:  Are you suggesting that the Social Security file be used for this
purpose?

Dr. Spirtas, NCI:  Yes.  Its population has been followed over time for vital status. I
think your two questions may be related.

Dr. Brown,  NIOSH:  How would  that  be different than using the  US population as
opposed to a 1% sample of the US  population?

Dr.  Mason, NCI:   They are people who  were  identified  while  working and  then
followed.

Dr. Bridbord,  NIOSH:  But you would still like to look at a group that included, for
example, perhaps only blue collar workers but who were not in certain types of  jobs,
or white collar workers.

Dr. Mason,  NCI:  The ideal set of circumstances would be to have a representative
sample of every standard occupational category so that you could lump or split in any
way what you wanted in order to get a reasonable group of persons to study.  I am
hypothesizing this particular outcome, to take this particular group here with interest
in a similar  age structure, and then I can look to see if there are differences. That is
what we  want, but until  such  time as we open up the floodgates, we are going to  have
to just keep chipping away, and it might well be reasonable to make that a tandem
request that this is  something which we need, that there is an identifiable need, not
just by us, but by you and by anyone who wants to work in this area.  We know it can
be done,  and you are the only resource that can permit this particular thing  and just
use it that way.

Dr. Fraumeni, NCI:  Dr. Blair, would you give us again your first point?

Dr. Blair, NCI: The first one was maintaining the momentum  of trying to get access
to Social Security files for establishing cohorts  from companies out  of business or
developing a fallback case control capability.

Then there are specific study areas  where there  is an urgent  need for epidemiologic
research. One is herbicide exposure and cancer.  The Swedish  studies need to  be
replicated in this country, either  by a case control approach or by identifying exposed
cohorts.  Secondly, another group mentioned recently in the New York Times and by
Walter Cronkite  is farmers.  There are  several  different mortality studies  that
suggest farmers have high rates  for several cancers, leukemia, multiple  myeloma,
lymphoma, pancreatic cancer,  despite a low overall total  mortality. Very  little is
known about what  the  particular hazards  might be, although you can list several
hundred  chemicals  farmers come in contact with.  This may be time» to develop
studies to pinpoint farm-related hazards.

Dr. Mason,  NCI:  It is like the grain inspectors  hypothesis.   If you could  really get
sufficient numbers of them and access to their records, you could find the response.

Dr. Bridbord, NIOSH:  You would probably never be able to  identify the  chemicals
that are responsible.

-------
Dr. Burton,  NCI: One study that has never been worked out is the epidemiology of
Burkitt's lymphoma in  the United States.  There have been many case reports and
immunologic studies on Burkitt's lymphoma patients and biologic  materials.  There
have been about 350 verified cases of Burkitt's lymphoma in the United States.  In the
U.S.  you don't have the suspect triggering factor  which  occurs in Africa and New
Guinea, that is, malaria, which is said to interact with the Epstein-Barr  virus to
induce Burkitt's lymphoma in certain individuals.

There have been a number of Burkitt's lymphoma clusters reported, presented as case
reports,  but I  am not  aware  of  any  study  which  attempts  to  determine  any
environmental factors  which are common  to the  Burkitt's lymphoma cases in the
United States.

Dr. Bridbord, NIOSH:  How many of those have occurred?

Dr. Burton,  NCI: There are about 350 verified from the American Burkitt's Tumor
Registry, based on information provided by Dr. Paul H. Levine of NCI.

Dr. Bridbord, NIOSH:  Over what period of time?

Dr. Burton,  NCI: Over 10 years.

Dr. Bridbord, NIOSH:  I would like to play devil's advocate with you.

Dr. Burton,  NCI: The incidence is probably less than 1 per million.

Dr. Bridbord, NIOSH:   I would find  that  hard to  argue  as a candidate under the
collaborative program, particularly from  NIOSH's perspective facing so many pro-
blems that are just huge in magnitude with many, many people exposed.

Dr. Burton,  NCI: Another aspect that has not  been studied in detail by the United
States  Government investigators  has been the followup of  nitrosamines in the
causation of nasopharyngeal  carcinoma.  Funds for this purpose have been provided to
some foreign investigators.   Dr. 3ohn Ho  in Hong Kong  has hypothesized  that the
triggering factors in nasopharyngeal carcinoma among those of Chinese origin are
nitrosamines in salted, dried fish.

Another factor is aflatoxin,  indeed, the mycotoxins  as a group.  Mycotoxins  have not
been carefully  studies  in relation to  human  cancer in the  United  States.  Aflatoxins
are produced in peanuts  and corn  in considerable amounts.  The  Department of
Agriculture  has long  been  interested in this.  Farmers,  of course,  are  those  who
handle peanuts and  corn and would probably be most exposed. You never see much in
the literature about this, not only the aflatoxins but the whole gamut of mycotoxins
in the causation of human cancer.

Dr.  Bridbord,  NIOSH:     I  suspect   of   all  the   things  you  described,  the
nitrosamine  area   is   one   where I  suspect NIOSH  will continue  to  do  some
follow-up work; and to the extent that we can identify cohorts we will have a decent
idea what the exposure might have been.  That would be something valuable.

Dr. Burton,  NCI: Nobody has ever been interested enough to really pursue that as a
subject in itself.  It has always been  a small part of some large study  for some other
purpose.
                                      540

-------
Dr. Brown,  NIOSH:   We  have a cohort that we are studying now  in the  leather
industry where there are some  documented  exposure to nitrosamines.   Dr. Fagen
talked about this yesterday. We have a leather plant where they have never used the
nitroso compound  and a leather plant which  has almost exactly the same  process
except they use a different agent for dehairing the hides which uses nitrosamines.

Dr. Mason, NCI:  Do you people have the capability as far as laboratory support if  we
went  out and actually in one of  our colon studies or whatever were able to take and
do it  the  same as they did it  in  Africa and take samples  of foodstuffs prior to
preparation and subsequent to preparation and get them to a laboratory?

Dr. Riggan, EPA:  Dr. Gardner's division in the Health Effects Research Laboratory
in Research Triangle Park, NCI, has the capability of quantifyng nitrosamines in food.
According to my understanding, the Food and Drug Administration has this capability
also.  The expertise  exists in the Division.  Administrative details would have to be
worked out.

Dr. Burton, NCI:  It could be while  you are pursuing this you may get on to something
as to  why nasopharyngeal  carcinoma occurs in the other groups, in the other cultural
groups.

Dr. Fraumeni, NCI: Dr. Keefer do you have any comments?

Dr. Keefer,  NCI:  One of the problems with  the NPC  study, I  believe, is that the
consensus of analytical chemists does not agree that nitrosamines have been reliably
implicated in causation of this tumor.  There have been some attempts to confirm
their  presence in local diets, and some failures to confirm them.

Dr. Burton, NCI:  That is what I  am saying.  Nobody has realy put enough money into
it as  a study in itself with the intention of determining the etiologic relationship of
nitrosamines to naso-pharyngeal carcinoma.

Dr. Keefer,  NCI:   That is not quite true.  Terry Gough, one of the best nitrosamines
analysts, went to Hong Kong to pick up some samples from  3ohn Ho, and I  am not
even sure what happened to them.  It has been a long, sad story.

Dr. Burton, NCI:   I know  about  that  study.  It was just a small study.  They found
levels in  certain  croakers, white  croakers and  yellow croakers,  but  nobody ever
determined the extent to which croaker fish are eaten in preference  to any other kind
of fish during one's lifetime. It is all called salted, dried fish. Recently Drs. Huang,
Ho. Gough and colleagues  have studied volatile nitrosamines in salted, preserved fish,
so such studies are continuing, but probably not with NCI support.

Dr. Keefer,  NCI:  As a matter  of  fact, in the meantime we had a contract in the
Carcinogenesis Program with  Mt. Sinai, a small feasibility study to find populations
anywhere in the world that could be studied for possible human effects of nitrosamine
exposure.  That contract was, also,  a failure for a lot  of procedural reasons; as a
consequence, most people in the field of epidemiology seemed  to  agree  that there
really was not any population worthy of study because of the small numbers and/or
the exposures  to a multiplicity of  carcinogens, and/or all the other things that you
epidemiologists complain about. I should probably admit at this juncture that I don't
really belong here in the  epidemiology session because  I am only  a  chemist, but  I
came anyway  because I think epidemiology is really where any approach  to cancer
prevention has to  start and end.  I  have devoted  over a  decade of  my career to  the


                                    54J

-------
laboratory study of carcinogenic nitrosamines, and I still have to say that there is no
evidence at all that anybody ever got cancer from nitrosamines.

Dr.  Bridbord, NIOSH:   I think if there  is any  chance  to get some  additional
information in that regard it will probably be through studies described where one has
an opportunity to  compare two groups  where at least there is  a  difference in that
exposure.

Dr. Sloan, NCI:  Isn't there some work  going on in  the nutrition program where they
are trying to relate vitamin C levels to nitrosamine in causation of  nasopharyngeal
cancer in China?

Dr. Keefer,  NCI:   Vitamin C is of significance primarily as a blocking agent for
nitrosamine formation. It turns out to be quite a good scavenger for nitrite, which is
one of the  precursors of nitrosamines.  You can demonstrate in animal systems that
you can block nitrosamine formation in a cancer  preventive sense using vitamin C.

I don't  know, however,  that  there is  any  evidence  that vitamin C helps  prevent
carcinogenesis by  preformed nitrosamines, which are present in many of our foods,
and  in the workplace as  many people  have  mentioned.  I don't think there is any
evidence that vitamin C can help you in that case.

Dr. Burton, NCI:   I thought you might be interested, talking about vitamin C, that in
the one intensive study done in Hong Kong on those in their 20's with naso-pharyngeal
carcinoma there was a very intensive lifetime food intake history done by Drs. E.  N.
and M.  L. Anderson, anthropologists from the University  of California at Riverside.
They determined that what was common to every one of the 25 or 30 cases that they
studied  intensively was that none of them had an intake of vitamin C in their  younger
days. They did not believe in  eating oranges or  limes or lemons  or anything that had
vitamin C in it. They also consumed salted, dried fish from early childhood.

Dr. Fraumeni, NCI:  From that perspective, how would you evaluate the role  of
nitrosamines  in human cancer?  What would be  the best  groups to look at - people
occupationally exposed  to nitrosamines, people  exposed to nitrates which are later
converted to  nitrosamines?

Dr.  Keefer,  NCI:   Let me  give  a  real amateur's  answer.    People  found  air
concentrations of  nitrosomorpholine in  one  rubber  plant  in Maryland  that approach
those which  are detectable in small  animal studies as being carcinogenic, in other
words, 250 micrograms per cubic meter  of nitrosomorpholine in the air.  These  levels
may well be sufficiently carcinogenic that you can detect their effects in a small
population. But the analytical data  are not adequate to launch a major study  because
after finding 250  ug/m   that one  day on one  test,  a new ventilation system was
installed and  the levels were much lower when the analytical equipment  was brought
back for a  confirmation  retest.  Would  such variations have happened in the  absence
of ventilation changes?   How long had those levels been there  before they were
discovered?  We don't have data on these points, so we can't say whether it would be
fruitful  to study these workers or not.

Dr. Bridbord, NIOSH:  The situation where you had  a reason to believe that one plant
had it and one plant didn't, even if you cannot  quantify exactly what that would  be
over a period of time, such as  was described, is about as good a natural opportunity to
look at that as any.


                                    54Z

-------
Dr. Mason, NCI:  As long as you can control for age and have information on other
such things. It at least has the potential of eliciting a finding.

Dr. Keefer,  NCI:   There are several situations that might  be worthy  of study.
Certainly the leather industry is one that I think should be pursued vigorously, both by
people with statistical  interest and by chemists.  I think there is  a lot more that
needs to be done in the chemistry area, much more systematic study  of how exposure
levels vary with certain parameters. So far I don't see that there is the massive input
of funds that I think  is required in that area.

Dr. Bridbord, NIOSH:   We  want  to raise the question that  regulatory agencies,  I
believe, OSHA has been petitioned to  set  a  standard for nitrosamines just for that
very reason.  It may be extremely difficult to do that, but in terms  of the  relevance
of pursuing this issue to the regulatory needs, that certainly would seem justified.

Dr. Keefer, NCI:  Another thing that comes to mind is this diphenylnitrosamine issue,
which I don't think we have seen the last of.  The conventional  wisdom for  years has
been that it is  a non-carcinogen.  Everybody "knows" it is not carcinogenic,  and
everybody is using it in rubber factories in massive amounts.

About  one year ago  Lijinsky's group came out with a study showing about 50 percent
of their rats got bladder tumors from this chemical and after looking  back in the
literature,  Dr. Lijinsky has  told me there was only one marginal animal study that
the conclusion of non-carcinogenicity was  based on.   So, people have been  exposed
tonwise to this particular material, which may well be a powerful carcinogen.

Dr. Burton, NCI:  One thing that Dr. Ho emphasizes  is that exposure to nitrosamines
could begin in early childhood, by chewing and swallowing salted,  dried fish.  This
occurs among southern  Chinese,  and often among Chinese-American and Chinese-
Hawaiians. Such investigations should continue to be pursued.

Dr. Keefer, NCI:  Let me throw out one more thing that you may not know about yet.
There has been a lot of nitrosamine news in  the  press but there is one story that is
just not in the  public  press yet  which  is that  90  percent  of the powdered  milk
prepared in this country contains a small amount of nitrosodimethylamine, the same
nitrosamine found in the leather industry.  When you mix the milk according to the
instructions on the box, the levels come out  to less  than  1 part per billion, which is
five-fold less than what people have  been making so  much fuss  about in beer. But in
my house we used to  drink  a  lot of powdered milk and oftentimes people use the
powder directly in cooking.  We used it in place of whole milk,  and I don't know how
many American  families or families  overseas might use milk powder in enough
quantity to get significant exposures while their children are in their formative years
physically.   That is  something that  we  might want  to think about preempting the
press on and get to work on  before it becomes senstional. I don't know that  anybody
should be alarmed by this, but I personally am.

Dr. Fraumeni, NCI:   How does it get in  the milk?

Dr. Keefer, NCI:  Dr. Scanlan at Oregon State looked for it as an  extension of his
studies  on beer, where nitrosamine  contamination had been traced to  the  malting
process in which  direct gas flames are used to dry  the sprouting  grain to  make the
malt.  The gas flames  apparently have  enough oxides of  nitrogen  or something in
them that they can nitrosate some dimethylnitrosamine precursor. This produces the
dimethylnitrosamine  in beer.


                                  543

-------
Similarly, when they spray the milk into drying drums heated with the same kind of
gas firing, there  are  precursors  there,  too,  which lead  to  dimethylnitrosamine.
Certainly that was the hypothesis  that Dr. Scanlan used in studying milk in the first
place.  So, any process in which you have direct gas firing on a foodstuff  appears to
have that potential, though detectable nitrosamines are not  formed  in every such
situation; some kinds of dried  coffee,  for example, have not had that contamination.
A product is a candidate for nitrosamine contamination if it is gas fired, but it does
not necessarily contain a nitrosamine just because it is directly dried.

Dr. Brown, NIOSH: They have taken it out of the beer though, right?

Dr. Keefer, NCI:  They can.  All they have to do is change the  process to indirect
heating.  Most American companies don't do that, only Coors who has always done it
that way can say that  they have  no detectable nitrosamines in their product.   It
seems that nobody has changed their process in this country. In Germany they  have.
In Germany, also, the rubber industry  does not now  have nitrosomorpholine contami-
nation to the extent we do.  They changed their process there too, whether by fiat or
what. They have made the changes. In America we  don't seem to  do that.

Dr. Mason, NCI: And it is  actually  dimethylnitrosamine in powdered milk?

Dr. Keefer, NCI:  Yes.   There are a number of other  foodstuffs  and some  other
industries that  quite probably merit consideration though I don't  know whether they
are worth studying. I cannot answer that question. I would sure be glad to  tell you or
anyone else anything I know about  the background and analytical backup so that we
can get on with preventing cancer.

As  a matter of fact, that is  really why I am  here.  I would like  to know why  the
interface between epidemiology and  chemistry  over  the years has  been such an
abysmal  failure. I need some suggestions as to what we can do differently.  There  are
situations out there that  statistically  just  cry  out for someone  to identify  the
causative factor in the food or somewhere in the environment to discover that this
thing  or  that  causes  cancer  in people.  We  have not  been able to identify  the
carcinogens in the "cancer gardens" you have described for us.  I don't know what to
do, but I  came here with that question in mind.  I would like to know how we can help
solve these very issues  you are talking about and particularly how we  chemists can
change what we are doing.

Dr. Fraumeni, NCI: You talked about analytical techniques to identify nitrosamine in
the workplace  or in  foods or in   the environment.   Do you have any analytical
techniques to measure body burden or other laboratory indices of exposure?

Dr. Keefer, NCI:  The analytical  methods are there for things  like dimethylnitro-
samine and nitrosomorpholine, which  are  the ones  that  were  discovered in massive
amounts, relatively speaking. The reason why it might be difficult to do body burden
studies is that the compounds are  metabolized rapidly, plus there is another reason.
New evidence suggests  that dimethylnitrosamine may  be a normal constituent  of
human  blood, so that we  would then be trying to  find  small levels of exogenously
produced  material on  top of a somewhat  smaller background.   People  have  not
quantified the blood levels  carefully  enough yet in their minds  to be able to  bring
themselves to publish it, but it may be that  these  kinds of levels  of dimethylnitro-
samine are there in the blood to start  off with.  So,  it would be  complex, but  the
analytical methodology  is available, well developed, and well confirmed  for  those
particular compounds.

                                    544

-------
Dr. Mason,  NCI:   And  not  so expensive that it wouid  break the bank  to do a
reasonable number?

Dr. Keefer,  NCI:  We have in  the contract program at  least 10 of the instruments
most commonly accepted for doing this kind of thing.  We don't  have one in our lab,
but we have good communications and working relationships with labs who do, and I
think we can arrange any analysis.

Dr. Sloan, NCI:  I wonder if someone here knows about the program Jeffrey Pearlman
is trying to work out in the National Center for Health Statistics.  As I understand it
the National  Center for  Health Statistics was ordered to develop a plan which they
are due to report to Congress in June that would provide some registration for  people
who had been exposed to some toxic substance and give us a basis for following them
over time.  That is  not  built into your  death  registry.  This is something entirely
separate.  Do you know about that?

Dr. Mason, NCI:  Not really, other than it has been fraught with all sorts of problems.
I don't know if they have  ever really staffed up to do half  of what they wanted. They
were in trouble personnel wise  over there. It is yet another interface that should be
pursued if it can come together.  Any reasonably defined identification of persons
because of an exposure or an outcome would be desirable. The National Center needs
to be  doing more of that, but it is my understanding, at least months back, that they
were not very happy with its productivity as yet.

Dr. Sloan, NCI:  I am sure Mr. Mazzocchi and  others have approached  some  of you
regarding the need for some kind of medical surveillance of people known to have
been exposed  to toxic substances, not all of  them carcinogens.  For example, the
Velsicol Company had workers  who were exposed to a great variety of solvents and
brominated compounds.  It has  now gone out of business and Mr. Mazzocchi is very
concerned that  we have no  mechanism, at  least within our division, to set up a
surveillance program or  fund it. These  are  workers who may turn up with serious
disease  conditions in the  future, particularly  because they have  been exposed to
multiple toxic compounds.  Is there any way of doing more to establish a surveillance
system for such exposed individuals?

Dr. Bridbord, NIOSH: I think in terms of providing routine medical care, I doubt that
either NCI or NIOSH or  certainly EPA would be able  to provide that to the  extent
that you are  talking about. Demonstration or research projects are another matter,
but each of those would have to be looked at  on an individual basis to understand the
circumstances that surrounded that.  I doubt that either organization would routinely
be able to take up ongoing medical surveillance just for the sake of that, you know,
just in terms of the enormity of what would probably be involved.

On the positive side, NIOSH is making some efforts through the primary care  health
delivery side of HEW  to eventually build up an expertise in occupational medicine
that could potentially deal  with these situations.  We are working,  for example, with
the National Health Service Corps to  establish an initiative in occupational medicine,
but these are rather long-term type things.  Public Health Service hospitals are
another example of trying  to get their interest, but the efforts so far have basically
been pilot types.  We are  probably many years away in terms of getting the full-term
benefits.

Dr. Sloan, NCI: Wouldn't your ERC's be able to tackle some of those situations, too?



                                       545

-------
Dr. Bridbord, NIOSH:  Not as a routine.

Dr. Sloan, NCI: Not as a routine, but this would be a special study.

Dr. Bridbord, NIOSH:  But the ERC's are not provided with money for research per se.
They  have money for teaching and education.  To  the  extent that  a given ERG
investigator would want to  come in,  for  example, to  NCI or to NIOSH for an
investigator-initiated  research project, there are  mechanisms in place.  Also,  to the
extent that the inhouse people in either organization identify a group at risk and want
to do  a cross-sectional medical or  mortality study that could be done,  but the place
where the system falls  down is in the routine surveillance of populations that don't
fall in those other categories.

Dr. Spirtas, NCI:  You might want to call Dr. Joyce Sals in NIOSH.  She did a  study.
Her number is 684-3284.

Dr. Mason, NCI:   Nobody can afford to  do  it,  and it comes down to a fundamental
question  as to what you mean  by surveillance.  Registration is one thing.  A routine
physical  is another thing.  You have  no  guaranty that you even know  what test to
perform.  That is the  problem, and the last thing that you want is everybody and his
neighbor  collecting sputums to look at cytology when some fundamental questions on
it are not even answered yet.

Dr. Keefer, NCI:  I guess I said my piece, but I will summarize once more by saying
that I will do anything I can to help you people identify carcinogens.

Dr. Fraumeni, NCI:   I  think that  would be  very helpful.  You are one of  the  few
laboratory people at NCI with a solid interest in epidemiology.  We would very much
like to meet again with you.

Dr. Keefer, NCI: It is just a suggestion that we try to work on the interface  between
our disciplines.  Actually, Mervish  first brought up this question over lunch one time
many years  ago, i.e.,  can we make a better two-way flow out of this epidemiology-
chemistry collaboration.

Dr. Mason, NCI:   Operationally, I would like to suggest,  since when push comes to
shove, if there is not  money  there, is  it really productive, why not in October and
then six  months later  in the fiscal year because of the way these things are funded,
we get together on some sort of formal  basis?  The  thing that you threw out with
regard to powdered milk was an interesting thing. I cannot for the life of me think of
how we can study it right now.  I  mean  it is something that should be taken under
advisement, but how  do you  really identify a group of individuals  where you could
quantify  that use, for what period of time.

I  think that these types  of things are productive from  the standpoint of saying that it
is an  interesting question.  We know what we want to look  for, at least have some
pretty good ideas about what we want to look for. How do we identify that  group of
individuals?

Dr. Bridbord, NIOSH:   From  the perspective of  linking  the chemistry  and the
epidemiology,  I suppose  NIOSH  is the place where we  have at least been endeavoring
to do that for many, many years. I would  not profess  we always get out medical
people and industrial  hygiene  people working at  100  percent  effectiveness, but we
certainly recognize that that  is extremely important.  On the EPA side, also, even


                                      546

-------
five, six, seven  years ago there  have been a number of workshops that were just
looking at the whole issue of the atmospheric transformation products with fairly
sophisticated atmospheric chemistry involved  and  the need to  get  the health
scientists and chemists together.  So, I certainly think it is extremely important.

Dr. Keefer, NCI:  I  will come anytime you want.  As I say, without your help, we
cannot answer the question of whether what we are doing in the chemistry laboratory
is worth anything in cancer prevention terms.

Dr. Mason, NCI:  Obviously, if what we are going to do  is go into a real life set of
circumstances and take things from people  and try to quantify their exposures, and
this is independent of the powdered milk  thing at the moment, we need laboratory
backup, and we, also, will need laboratory tests  that are not so expensive that we
cannot afford more than 10 of them.  That  has been the problem  with a number of
things, that the assays themselves are very, very expensive.

Dr. Keefer, NCI:  Chemical assays or some kind of biological work?

Dr. Mason, NCI: Either of them. What I am saying is that we recognize the need for
additional laboratories, if we just look at what we are  doing in  families and familial
aggregations of cancer.   We are spending a lot of money just for some fairly routine
stuff, and to do some additional things, I think we are responsive and would like very
much to continue on discussing.

Mr. Steelnack,  OSHA:  Our basic interest is the OSHA cancer  policy.  I am  here to
see exactly where epidemiology is going in the future since we are going to have to
be explaining that policy to our own people. One of the big problems I have is that
industry keeps coming back to us with what they call  negative epidemiolgy studies.
There is  no real definition of what it is, but industry  can always  come back with a
study saying that they surveyed their people, and they found that there is no increase
in cancer in their industry, and we should not be regulating it at  all.

Dr. Bridbord, NIOSH:  One has to ask about the quality of the data.  Negative studies
don't  prove safety.  They just define  potentially limits of risk, even good epidemio-
logy studies.

Mr. Steelnack, OSHA:  According to industry there is a difference between exposing
an animal to a large  amount and getting a small result  and to exposing a  worker to a
small amount and getting no response.

Dr. Bridbord, NIOSH: That is the argument  on the first cut, but if you really look at
all the dynamics of  the absorption, metabolism, retention  and excretion you  may
actually find that the large dose in the experimental animal actually is giving you an
equivalent to a much lower exposure in the human being.  I think a good example of
that is what you find with lead;  whereas  people were measuring doses for so many
years in experimental models based on body weight and food intake, when they finally
got around to looking at blood  leads,  situations where you would think the exposures
were orders of  magnitude above the  human situation were really resulting in blood
leads  that  were not that far different.    You  have  to look  at  the  case-by-case
situation.

Dr. Gass, OSHA: I have a little different problem.  I am interested in the design of
internal and external training  programs to educate not only our own inspectors but
also workers to risk  in industry.  I am not an epidemiologist, but I have  a couple of
questions that I think maybe the group should ponder.


                                       547

-------
OSHA probably is the United States' largest employer of industrial hygienists, by job
title. We probably have 680 or so inspectors out there, covering about ^ percent of
the work sites each year.  It is not enough, but that is about what our resource level
is.

The  question to you is since we only have  about 23  or  maybe  25 percent formally
trained  industrial hygienists on the  work force  and  those  are  the  ones that have
formal epidemiology training, just what should the routine, everyday, OSHA inspector
know about epidemiology  as he goes through the field,  and how could his training
program be so arranged as to give the most benefit from data collection and observa-
tional points of view? It is a very practical problem.

Dr. Spirtas, NCI: 3ust one.  Dr. Waxweiler  from NIOSH and I are going to put on a
session  at  AIHC to try to address this.  We talked to Dr.  Kelley about this,  and I
thought originally it  was meant to be a seminar, but it is  going to include definitional
terms and  study design, regarding what an industrial  hygienist needs to know about
epidemiology in order to be setting up programs  or evaluating contracts. If you  are
an industrial hygienist for a company or an inspector for  OSHA, our seminar will
address what you need to  look for and what  data  is required and then secondly, what
should you be doing on  your own,  and when  do you  pull in  a   person  trained
professionally in epidemiology.  We will give some examples of studies to bridge some
of the gap between theory and practice.  We think it is important to have practicing
epidemiologists talk  to industrial  hygienists and tell them some of the things that we
think they should be considering.  In addition we hope to describe what sorts of things
we do and what words we use.

Dr. Mason, NCI: When is this scheduled?

Dr. Spirtas, NCI:  May 21,  at the AIHC in Houston.

Dr. Gass, OSHA:  You know the problem with that is that we have travel restrictions
that are very tight, and I doubt  that one-third of  our industrial hygienists will  get
there.  Ordinarily we would not send but half our force anyway  because we have to
cover the shop. You know, industry  moves on and so does risk, and our meetings  are
not that important.

Dr. Blair, NCI: Don't you have regional training programs specifically for government
industrial hygienists?

Dr. Gass, OSHA:  Yes. That is part of the problem, but that is, also, our problem.
We are  the ones who formulate  that in the national office, and the question still
remains, you  know, what do  you teach internally?   You  know, you  are  teaching
industry as well in that particular problem.

Dr. Blair,  NCI:  But couldn't  you use  the  same sort of approach  to the regional
program and eliminate the travel problem?

Dr. Gass, OSHA:  If I had  known that ahead of time,  I probably would have sent
somebody.  Are you going  to be involved in that?

Dr. Bridbord, NIOSH: Why don't you think about sending a videotape crew even out of
your regional office, and then at least you have that on record?
                                      548

-------
Dr. Gass, OSHA:  I don't know. Would they allow that since it is a pay program? It is
a tuition program, right?

Dr. Spirtas, NCI:   We are involving an industrial hygienist who did his dissertation on
a  retrosective industrial hygiene study, and  another person whose  speciality is
medical surveillance.  So there will be a total of four speakers.

Dr. Gass, OSHA:  But you do recognize that as a problem within the association  and
at large.  Very few industrial hygienists that we  have on board  are really formally
trained  industrial  hygienists.  They  are  really chemists that have cross-trained or
they are physicists or engineers.

Dr. Spirtas, NCI:   We are willing to make the effort to try to put  together a useful
program, and there must be some way to work out logistical details.

Dr. Gass, OSHA:  We have issues that raise their ugly heads all the time, like I had an
industry group come at  me because of some training that we were  contemplating on
carcinogens in the painting industry.  Well,  the company sent a contingent to have a
little talk with me, and some of the pointed questions they asked were where is  the
epidemiology data that  shows that zinc in paints,  that lead in paints, that chromium
in paints actually  cause  cancer in any humans?  What effect does the vehicle have on
the actual  industrial hygiene collection?    3ust what does the method of collection
mean?  Are there any data points and where  are they?

It is hard for me to find that data, and that is one example.  Another example is when
we prepared  training  for inspectors  in the  oil well drilling industry we had a little
problem along with the  Census Bureau.  It  seems  as  though they drill a 10,000 foot
hole, tap a natural gas  reservoir, cap it, put a Christmas tree  on it and move on in
about three to four weeks, and by the time we come back with a follow-up citation
they are gone and it is the same problem in  construction.  I was happy to hear some
of your  points about a national registry, but that still does not tell us what we can do
with that epidemiology.  Are there  any studies of  oil field workers and the types of
compounds they are exposed to?

I made a list. In  our training program we found about 250 toxic substances that they
routinely handle.  Drillers are at  great risk because they are there most of their lives,
but mud engineers really do handle a lot of things including asbestos.   So, those  are
the kinds of things that I need to know - where I  can find the epidemiology studies
listed.  I would love to have job titles, agents, hazards, some kind of cross index from
the industry so that when I  have training problems or for that matter  when we  are
teaching inspectors to go out and do a job, it would be really nice to know the data in
advance.

Dr. Bridbord, NIOSH:  Have you ever gone to chat with people who do the National
Occupational Hazards Survey?

Dr. Gass, OSHA:  Yes, I called Dr. Sundin, to discuss these matters. I  called the  one
expert on oil  well driving. Are you aware that there are only 14 participating states,
and Texas, Louisiana and Oklahoma, who have the most oil wells, don't participate?
Where do we  get  data?   We would love to have  it for our training  programs and for
our own professional knowledge.

Dr. Chu, OHSA:  OSHA has subpoena powers  and so does NIOSH.



                                        549

-------
Dr. Gass, OSHA:  They won't collect it. Louisiana will not even take the data.  There
is  no record in the Louisiana system.   They  don't collect  it, no  mechanism for
collecting it, no state law.  This is someplace where we need to improve, I think.

Dr. Bridbord, NIOSH:  A couple of  comments.  One is  we are at least beginning to
look at the petroleum industry as part of  our work within the  institute.  I don't know
about a full range of operations that includes all the way from  the mining, in effect,  I
mean the drilling in this  case, but certainly we are  taking a  closer  look at that
generic area.

I suspect the issue of trying to get  more efficient recovery of oil which is already in
place, particularly as the price of oil has increased, will provide greater incentives to
go to secondary means to increase  the yield which in previous years would not have
been  much of an incentive, and one is potentially faced with all sorts of different
schemes and  mechanisms to increase that  yield. I would suspect that as that becomes
more generic that is a very valid question to ask.

The third point would be, I am not sure, maybe Dr. Sundin might be able to carry the
message  back to Cincinnati in  NOHS  2, is this  at  least a  group  or an industrial
situation where we might make some effort to clarify exposure?

Dr. Chu, OSHA:  One of the first things that happens in a regulatory program  that  I
am involved in  is that in spite of lot of occupational studies on workers, i.e., painters,
the problem, from a regulatory sense, is  difficult to regulate.  It has to  be flipped
over so that it is substance oriented.  We need identification of substances so that we
can get at those particular substances.  The regulatory group that I am with is caught
by the dilemma that when someone comes up with a PMR on dry cleaners, and we
cannot identify whether  or not it is TCE or  carbon  tetrachloride, unless we  go to
some generic classification, we need to have substance orientation.

Consistent with that, I am trying to take some leads from some of  the animal data
results in the  carcinogen  bioassays and  possible  look  at  whether or not these,  by
looking  backwards  and looking  at  the animal  studies  that are  positive  in several
species, you can say that here are some candidates, let  us look to see whether  or not
they have some kind of subset that would make a good epidemiological study.

We have  identified  compounds that are occupational, that are potential occupational
carcinogens and in the reviewing of  the data we have been able to capture target site
information.  We are now trying to capture and create a data bank of the chemicals
and target sites for those chemicals, full well realizing  that in not every single case
is there a one-to-one correspondence  between animals and humans, but cases  where
you  have  a rare tumor  in the human situation  that  will be referenced back  to
associate chemicals would be a part of  this picture.

We would like to get involved in tying together use information with this data bank of
chemicals and target sites.

We are now approaching  a  situation  to gather  epidemiological tools to be  used,
computer  programs, and I spoke about getting PMR  studies programs and maybe
getting the  life table programs.  I would hope these would be available  so that at
least  the tools  can be in place,  if in fact OSHA decides to get involved in  those
activities.   That  is  basically the  concern  that  we have; the  orientation toward
chemical specific entities so that we can regulate.  I think of all the concerns that
people have, this is probably the greatest frustration.


                                      550

-------
Dr, Bridbord,  NIOSH:  I would just comment that there are a number of linkages
already with NIOSH.  We  have had a number of policy meetings  with  the OSHA
directors  and the  directors of  the NIOSH offices and  divisions,  and  there are
mechanisms being established to facilitate, at least, in  terms of  the OSHA-NIOSH
relationship, our understanding  of what your needs are and we are doing the best we
can to meet those needs.

Dr. Brown, NIOSH:  As far as the epidemiological study in the single agents, most of
the situations  that we  encounter now are industries where there are multiple agents
involved,  and  the approach that we have  been taking lately is to  look  at  a  large
industry of, say,  10,000 workers, to look at the overall  mortality of that plant and
then to concentrate on a single cause  that is shown to be in excess such as stomach
cancer and do a case control to try and trace that back to a specific  agent.  That is
one way around these multiple agent problems or multiple exposure problems.

The single exposure populations are  all used up.

Dr. Chu,  OSHA:   I don't believe that.  When you have 200 bioassay  results, as a
chemist, I feel there  is a driving  need  for leads  that  do exist  and that it  is an
interdisciplinary concern and that, as new viewpoints come in, that perhaps it may be
a small cohort that  has a single  compound exposure.

Dr. Brown, NIOSH:   I was being facetious, there are some left, but most of the time
we do get involved  with populations that  have multiple exposures,  and they are hard
to separate out and see where the association lies.

Dr. Sloan, NCI:   Is there  any  group specifically working on interactions between
different carcinogens?

Dr. Bridbord,  NIOSH:  Our own group in  toxicology in Cincinnati, I guess as much as
anyone else, has at  least begun to approach  the question, an example  being the  study
we heard  this  morning. They have  looked at the interaction, for example, with high
temperatures and certain carcinogens, and I think initially you would really be looking
at this primarily through laboratory studies.

Dr. Spirtas, NCI:  I think we  do have a lot to gain by getting together and talking.  Is
there  anything that  we can do as a step in  that  direction that  is  not overly
bureaucratic and  complicated but would create a list of people  who call themselves
epidemiologists, who want to work with epidemiologists or who have some interest in
epidemiologic  studies in the  various agencies?  Is there any way we can  easily find
out who is in the field?

Dr. Riggan, EPA:  Would it  be helpful if I  compiled a list of individuals with their
expertise in the Epidemiology Branch, HERL/RTP? To whom should this list be sent.

Dr. Mason, NCI:  Send them to the branch chief.  He does not get enough mail.

Dr. Riggan, EPA:  I  would be glad to share.

Dr. Chu, OSHA:  In  statistics they have a federal directory of statisticians.

Dr. Mason, NCI:  The last thing in the world we need is  something like that.  If you
want  to do something along these  particular lines and  have  it be productive, then
earmark some of  these monies and have a workshop and use the  monies to pay travel
for individuals to  come to discuss, but we don't need another laundry list.


                                  55J

-------
Dr. Bridbord, NIOSH:  I think one of the questions that is going to come up is where
do you go from here.  One of the things that might make some sense is  to  at least
look at  the  issues  of  new  starts among  NCI, EPA  and  NIOSH  in the  area of
occupational  and  environmental  epidemiology  and  with  particular reference  to
cancer, but also, allowing at least some flexibility to look at reproductive effects as
a minimum start.  Also, to ask each of us, as we make our own tentative proposals on
program planning, to have the commitment that before each of the organizations
makes their signed and sealed approval on that, that each have a representative group
of people who come  and  sit around  the table and say  that here is what we as
collective agencies are planning to do in FY 1981,  and then to identify people with
mutual interests  where we  might find that two or  three of those studies  might
duplicate.

We might, after hearing each other's  presentations, decide  that we  left out three or
four things that individually we had forgotten about and  could begin to establish those
mechanisms and be thinking from  that perspective.

Dr. Mason, NCI: I agree. We have a lot  of internal program money that is committed
to these types of studies.  It does not have anything to do with this collaboration. I
think this should merely be viewed as a mechanism to stimulate  additional  studies,
not as the prime way to fund these types of studies  because there is  not enough
money in the  pot.

Dr. Bridbord, NIOSH: I think you both identify other studies that maybe, individually,
the agencies  had not perceived.   The other  thing  you may do is strengthen those
collaborations for example, just identifying another individual who has an interest in
that who might be willing to review a protocol or informally work on  something.  It is
just built that way.

Dr. Mason, NCI: Dr. Chu was proposing a directory and I like to argue with him.

Dr. Spritas, NCI:  If one did not exist.  There is a directory of statisticians, a little
yellow book that comes out, but I don't know of a directory of epidemiologists unless
SEER is going to put something out.

Dr. Chu, OSHA:  I disagree with  you  because you are working from  a position where
you are in  the catbird seat.  Of course, when you are  in the catbird seat you don't
care what else is out there, but the people who are out there and not in  the catbird
seat need to know of these  activities.

Dr. Mason, NCI:  There exists an easy way for you and for anybody else who  is not
that familiar  with what is going on. You make three phone calls.  You can get three
pieces of stuff in the mail, and you can sit down for yourself and just decide; these
are the types of things that they are doing, these are the types of problems that they
are addressing, and I think that is  true.

Dr. Bridbord, NIOSH:  There  are a number of assumptions that certain mechanisms
don't already  exist, and there really are two ways that those things get started.

One is just a scientist with an  interest, with some ability to effect the resource
allocation decisions and talking to those people and just with the power of the ideas
convincing them that there is worthiness in pursuing study A as opposed to study B as
opposed to study C, et cetera. Those processes exist within each of our respective
organizations, and they exist to some extent across them.


                                      552

-------
The other mechanism is the more formal organizational policy decision  mechanism
which is well  established, at least in terms of OSHA and NIOSH, and  one  of  the
reasons  why we specifically wanted OSHA to come here.  Dr. Leidel is on two-year
detail from  NIOSH and is a person who can understand some of  those  linkages in
health that occur, but I just perceive that your assumptions are basically that there
really isn't anything when there is something already and we are trying to —

Dr. Chu, OSHA:  I guess what I  am saying is that I have discovered  in the inhouse
group at OSHA that the tools, the basic tools like statistical programs, access to
computers, these kinds of things are not available which are just fundamental.

Dr. Bridbord, NIOSH:  That is a  decision for the  assistant secretary to make,  you
know, in terms of whether that person is going to commit their resources to building
the internal OSHA research capability versus giving Dr. Gass  20 more people to do
work on education, versus adding 30 more compliance officers.

Dr. Gass, OSHA:  But you just mixed apples and pears, because the  Act does not give
us  the  right  to  do research.    That  is  your domain,  but except  in standards
development; that is the only place where there  is some kind of research done  in that
sense of the word.

Most of us are compliance people who are  pointed and focused toward enforcing the
Act and the regulations.

Dr. Bridbord,  NIOSH:  Right, but to the  extent  that OSHA  does certain kinds of
analyses and research associated  with the  standard setting process,  that has been
going on for a long time.

Dr. Gass, OSHA:  That is an exception to the Act.

Dr. Bridbord, NIOSH: I mean the basic mechanism, at least, in terms of the research,
it would be difficult. The mechanism  is basically through NIOSH and through the
other organizations. I know there are a number of cases where NCI was asked to do
studies.

Dr. Spirtas,  NCI:  I have a couple  of small points. I think I have heard of some  studies
that may be of interest to some  of the  people  here today.  On your question about
substance versus  occupation, I think Sheila Hoar is doing a dissertation on that with
Alan  Morrison, a  computer  program  that  would pick up  studies  about  certain
occupations  or certain substances, and  provide a  cross-link between substances.  I
think in terms of the oil field worker studies, wasn't there a woman from  Tulane who
come up and gave a seminar?

Dr. Brown, NIOSH:  It was people on oil platforms and offshore.   That is not in our
domain.

Dr. Blot, NCI:  I missed the first hour,  and maybe I am repeating things that have
been said, but in terms of general cooperation between NCI, EPA and NIOSH,  I would
see three major areas.  The first concerns case control studies of certain cancers.
These could be conducted in high-risk areas in the  country, and focus on tumors that
have not been studied yet. For example, there is some indication that there  may be
an  occupational  and/or environmental  link between herbicide exposure and lym-
phomas  - an hypothesis that could be tested by the case-control study approach.  A
second case-control study is one that we are planning to do  within our  own group.
                                    553

-------
 This is a study of colon  cancer  in retirement areas of the South, which may be of
 general interest to EPA in that  a variety of enviornmental measures are suspected,
 particularly to  dietary factors.  The reason for this selection of Florida is that the
 rates in certain counties  there are quite low even though  many people  have  moved
 down  from the North where rates are high.   The  colon cancer  rates  in  these
 retirement areas  are the same as they are in  other southern areas where there are
 not retirees  even at older ages, whereas you might expect them  to be somewhere
 midway  between the low rates in the South and high rates in  the North.  We have
 been phasing up  now  for  an interview study contrasting dietary differences  and
 changes on diet, between cases  and controls,  and also are  attempting to build in a
 laboratory component, and  here  is  where  we might  possibly  call on EPA  for
-assistance.  There are certain  laboratory measurements  of interest,  e.g., micro-
 nutrients in the blood, vitamin A in particular, with perhaps fecal material analysis
 for some other constituents.

 The second general area concerns cohort studies, particularly of occupational groups.
 One such cohort study that might be worthwhile would be a follow-up of a cohort of
 shipyard  workers  who were employed during  World  War II. There are people now
 involved in studying cohorts of shipyard  employees.  The largest one, by the group at
 Hopkins, involves  employees who began  working in the shipyards in the  fifties.  This
 study keys on radiation exposures in nuclear yards, but there is  some interest among
 the investigators, if  additional money could be  provided, to go back in some  of the
 yards  and identify workers who began work in the 1930's and 1940's.  Job titles  are
 available, and it may be  possible to more fully evaluate the role of asbestos and get
 clues  to other shipyard  exposures  in  addition  to  evaluating  the possibility  of a
 radiation hazard.
                                                                     I
 A  third general  area  of NCI/EPA/NIOSH cooperation  would  be  in utilizing data
 resources  that  already exist and linking them  up.  In  particular there are  some
 systems here  in the US which would provide useful occupational  and/or environmental
 data if accessible - particularly the Social Security System. If  we  are not limited to
 the United States, there  are excellent resources available elsewhere, particularly in
 the Scandinavian  countries  where  perhaps a  concerted effort  by  NCI,  NIOSH,  and
 EPA might get some work going. For example, in Sweden there are cancer registry
 data listing all cancer cases in the country. Census data are taken every five years,
 and there  are rosters of industrial groups that have been specially put together by the
 Swedish version of OSHA,  and there happen  to be other rosters.  The  Swedes  are
 great at making lists of people,  and they all have social numbers  that can be linked
 rapidly.   There happens to be a roster of all prescription drugs issued in the country.
 So, there  is  potential for looking at these together.  Some people in the  US have
 talked  with  people from  Sweden about such  epidemiologic studies  but perhaps a
 greater stimulation might be beneficial.

 Dr. Bridbord, NIOSH: There recently was a US-Swedish conference on occupational
 safety and health where  about  40 representatives  from  Sweden, including  a  full
 spectrum  of  government,  labor  and  industry  representatives,   and   there  is a
 mechanism being  established to  facilitate the cooperative studies between the two
 countries which certainly  would include a number of the things you mentioned.

 The main collaboration would be Department of Labor to Department of Labor, but
 NIOSH has already tied into that from  a research perspective, and I would imagine
 that that umbrella could be extended.
                                     554

-------
Also, the Finns are coming to the United States in October, and they have also had a
number of activities and as of now there is the Third  Annual NIOSH Scientific
Symposium which  will be held in  Detroit  preceding  the  APHA  Meeting which will
feature not only the results of some NIOSH work but,  also, summaries of the ongoing
Finnish work which includes cancer, and a number of other things as well.

Mr. Boeniger,  NIOSH:  In our primary responsibility  to characterize  and determine
worker exposures we have to deal with biochemists and physicians during morbidity
studies, with control technology engineers for controlling and correcting exposure
conditions, and most frequently with epidemiologists, among others. I  don't like to be
redundant, but I would like the idea of trying to exchange directories of  some kind  of
listing  of  professionals in various  organizations.  Every once in a while I have been
able to extract from someone a telephone  directory from a particular group,  and  it
has been indispensable to me, but I know NIOSH has directories of their personnel and
how the organizations are broken down, and I imagine many of the other government
organizations have similar things. I wonder  about the possibility of exchanging them.

In terms of professionals  in various  areas  of  research in environmental health and
public health, I was wondering if anyone else had heard about an environmental  health
directory that a New York publisher  is coming out with, similar to one that they had
perceived in energy technology.  It is essentially going to be able to allow one to look
under a title area, such as, say, epidemiology or  more specifically, printing  trades
epidemiology, or something, and be able to identify individuals that have done work  in
that area.

Dr. Bridbord, NIOSH:  This just reminded  me that NIOSH does contain a system of
ongoing research in the  occupational health area on  computer drta  base  which
includes not only the United States but other countries as well. I would not say that
we have 100 percent on that system, but it is a reasonable place to start to get lists
of names of people who are currently involved.

Dr. Spirtas, NCI:   Obviously everyone in HEW enters projects  into the Smithsonian
Scientific Information Review System.  Is that the way it is for OSHA and EPA, also?

Dr. Bridbord, NIOSH:  Yes.

Dr. Riggan, EPA:  Yes, and on EPA's computers.

Working Group adjourned.
                                    555

-------
      FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

          PROGRESS ON 3OINT ENVIRONMENTAL AND

              OCCUPATIONAL CANCER STUDIES
                 Wednesday Afternoon, May 7




                        SESSION B


WORKING GROUP ON THE TOXICOLOGY. METHODS DEVELOPMENT,

              PROBLEMS AND NEW DIRECTIONS



                  SESSION CHAIRPERSONS
                     Dr. Gregory O'Conor
                   National Cancer Institute
                      Dr. Norbert Page
               Environmental Protection Agency
                              55b

-------
SESSION B  - WORKING GROUP ON TOXICOLOGY,  METHODS  DEVELOPMENT,
PROBLEMS  AND NEW DIRECTIONS
Dr. O'Conor, NCI:   One of the major objectives of this session is to examine  the
program as it exists and to make it more effective for the future.  I think we  can
begin to  taik about the areas  that  seem most  important for joint cooperation  and
collaboration.

The  program is still fairly new.  While there  were problems  at the  onset, it is
beginning  to  reach maturity now, and this workshop  is a fantastic opportunity to
 develop the  communication and interaction that is difficult when we are all back at
our own institutions.  I think it is time  now to try to identify  those specific areas
where this type of  collaboration  can be mutually beneficial  and  most productive
because the  funding has  been generous, and  I  hope it  will  continue that way.
However,  the funds  are limited and I think with the increased  interest in the total
program we are going to have to prioritize more carefully and identify those areas of
research and  application research where the benefit is going to be maximum.

Dr. Page,  EPA:  As I have listened to the papers presented  today,  it is very obvious
that we are only seeing one small part of research in many areas.  We have seen or
heard  results from some of the in vitro testing, some in vivo,  and  some  of  the
metabolism work which is just a drop in the  bucket of all  the  research going on in
these areas.  We have to take a perspective look at what is being  done in this program
in context with that underway  in the rest of the research community, identify those
areas that are particularly important for  the three agencies and, as Dr. O'Conor said,
attempt to develop a meaningful program within the constraints of this collaborative
effort.

We will start with some of the questions that have been submitted.

Question  1;   How  would you suggest that a group of  workers exposed to suspected
carcinogens could be effectively  monitored for the  presence  or  absence of early
cancer.  This question, I guess, is directed toward the immunodiagnostic  work, or
other markers, in early diagnosis of cancer.

We talked about the possibility of a battery of diagnostic tests, and that  seems,
certainly, to be a reasonable  approach to me, but I am not a  clinician.  However,
knowing what I do of the pitfalls of some of the early diagnostic tests, I think it is an
area of great importance; one certainly that NIOSH and OSHA have a great interest
in.

Dr. O'Conor, NCI:  The question is how would you  go about  handling  a  group of
workers exposed to suspected carcinogens, and monitoring them for the presence or
absence of early cancer.  Well, that  really is a question for an occupational physician
in the workplace, and  unfortunately, to my  knowledge, there  is no ideal way of
screening  individuals for cancer.  It is one of the major problems of cancer prevention
or early diagnosis  that  exists.  My response would  be  that it depends on first of all
identifying the  suspected carcinogen, and from our knowledge of what organ systems
different  types of compounds affect, this would determine the type of monitoring
activity, be it physical examination and laboratory testing, depending on what you are
looking for.
                                     557

-------
Broad screens in general have not been particularly effective for  the  detection of
cancer in the  early phases.  So the best approach is that the key  to screening and
early diagnosis is identification of high-risk groups and then developing selected tests
which will try  to focus on the particular type of cancer that one might expect in that
population group.

Dr. Lowry,  NIOSH:  I think I asked the question knowing that there really isn't an
answer, as you have said, but more or less to place the bottom line of why we are all
here in  perspective.   In other words, all of the basic studies that are done to show a
particular compound  to be carcinogenic,   mutagenic,  all of  the  different  testing
techniques that have been developed, these are all fine and good, and they need to be
done, but at some point in  time, you have to take all of this  data and try to apply it
to the human being.

At this  point in  time,  I would agree with you that the screening techniques, at least
those that I  am aware of, are not that effective, and there are a lot of false positives
and  false negatives.   In most situations you are dealing with a reasonably healthy
population of working people, rather than an acutely ill  hospitalized  patient who may
have lumps and bumps that are rather easily recognized.

Dr. Plotnick,  NIOSH:  At the  present time, and Dr. Lowry knows this  better  than  I
because he does a lot of them, these sample analyses, clinical chemistries, et cetera,
are  required  by OSHA  guidelines  for  pre-placement  or follow-up  medical
examinations.

We had a recent  discussion at a symposium about  testing methods, and it was the
consensus of  people  I  consider to  be  relatively expert in  the  area that clinical
diagnostic testing has not  saved one life because of the fact that  by the time that
these relatively  gross changes are picked up in clinical chemistry, it is generally too
late to do anything.

Dr. Hygyeli, NCI: I  would suggest approaching this from an entirely different  angle
- from the practical standpoint.  My experience is based on two large studies, one in
Tyler, Texas,  on asbestos-exposed workers  and a  second one in Louisville, Kentucky
on vinyl chloride workers. We also support  two new prevention programs in Berkeley,
California.  As you know, the National Academy  of Sciences has recommended that
every worker who  has been exposed to any kind of carcinogen or suspected carcinogen
should be notified. I feel that there has to  be a step beyond this since it creates only
anguish and  does not solve the problem. It is my recommendation that the follow-up
of these high  risk workers  should include  more than just notification.  For example,
our experience with the projects to date shows  that unless you have some kind of
medical  screening for early  detection  of  cancer or very simple medical test(s) to
determine damage due to exposure, the educational programs are not accepted by the
workers since  they can see no reason for or benefit resulting  from  them.  Also, the
majority of the  workers will not  participate unless  the  educational  program  is
provided on company  time.   In view of the current limitations in the therapy of
occupational cancers  it is important that the educational  program  include emphasis
on decreasing  the overall risk to cancer.   For instance, these workers need to know
not only about their exposure  to a given occupational  carcinogen and the resulting
risks, they also need to be educated about  risk multipliers such as exposure to other
toxic substances  and  behavioral patterns such as smoking,  drinking and nutritional
imbalance.
                                        558

-------
Dr. Turner, EPA:  I came from  industry and  was concerned  with  safety, health,
labeling and regulatory compliance. In my opinion the development of a standard for
vinyl  chloride by OSHA was accomplished  by regulatory action despite  the strong
resistance  by industry  because ample evidence for cancer  of the liver  in  plant
workers exposed to vinyl  chloride existed.  Industry  is now, conforming  to  these
standards and apparently has not been affected from an economic standpoint. Until a
good diagnostic test for carcinogenicity is developed which bears  a good correlation
with the formation of cancers in  humans, we should resort to  preventive measures
similar to those taken for  preventive medicine. Cigarettes and saccharin  are good
examples.  By following this  approach at  least one can avoid  problems  of defending
one's stand and political implications. It is much  easier to backoff on a cancer suspect
than a firm  statement of a chemical being a carcinogen when  the evidence disproves
it.

Dr. O'Conor, NCI:   We recognize that research  on markers and  better diagnostic
methods  certainly are a high priority, particularly in  relation  to identification  of
patients or individuals who  might have been exposed to chemical carcinogens. In the
meantime we must use the means at hand and put into place those practices which we
know can be effective, that is improved practices within the workplace and the
design of  monitoring systems which  are  selected on a  biological  rational  basis  to
survey or monitor the particular type of exposure that we are concerned with.

Dr. Morris, EPA:  I like the idea that Dr. Herberman and others are  working on in the
marker area, but I  think we  have to appreciate there are other chronic diseases in
this process. I realize we have had a lot  of emphasis and certainly NCI has focused
on the cancer problem.  On the other hand, I think at least from  our point of view,
there should be  a more equal distribution  of concern, perhaps from all  of us,  in
NIOSH, EPA and certainly the NCI, in  looking at other chronic diseases and the
potential markers for those.

Question 2;  Was inhalation  applicaton considered in the study?  This  is the major
route of exposure for  workers, lung macrophages, pyrogen products  and safe disposal.
(Refers to Dr. Lowry's presentaton.)

Dr. Lowry, NIOSH:   I made an attempt to answer that  during the session.  Basically
inhalation was not considered for two  very practical reasons - time and money.

I am  not  convinced that inhalation  is the major  or  only  route  of  exposure  in  an
environmental situation. I  believe Mr. Boeniger may be able to  answer the question.
Mr.  Boeniger do you have any comments on that as to what is considered a major
route of exposure?

Mr. Boeniger, NIOSH:  We took some environmental samples and tried to ascertain
what the respirable  and  non-respirable fraction of the particulates were and we found
that the majority of them were non-respirable.

Therefore, we feel, I guess,  that  any non-respirable particulates  that were inhaled
would  subsequently be  swallowed.   So,  oral routing  of dosing the animals would
probably be  most efficient and practical in terms of time and money.

Dr. Gregory, CPSC:  When the artists and craftspeople first petitioned OSHA and
CPSC about the benzidine  based dyes, the  problem of  possible skin  absorption was
brought up,  and  DETO felt this was very  impossible.  When Black 38 was applied to
                                      559

-------
the skin of rabbits the portion that was labeled with Carbon 14, that is the diphenyl
portion, was recovered in the feces and urine up to 90 percent. It was found out later
that the experiment was not done correctly, and they needed to put these Elizabethan
collars on the rabbits to keep them from licking off  the dye, but even when they did
this, they found out  that there was skin absorption of the amount that  10 percent of
the carbon-labeled substance was found in the urine and 5 percent in feces, so that
we do know  that  some portion, at least of  the benzidine  based dye  Black  38, is
absorbed through the skin.

Dr. Page, EPA:  I don't know that much about these dyes, but it is my understanding
they are not that  firmly fixed on particles, that they  can  elute, and  regardless of
whether the  exposure  is through the  intestinal tract or in the lung, you have  a fair
amount that is absorbed. Therefore, the issue is  really  one of systemic exposure, so
choosing a route so that you  get systemic exposure is a  reasonable and practical way
to go.

Dr. Lee, EPA:  Before I left  Kansas City, we did  a study on several dyes. Although I
am not at liberty to  say, the study was performed for a private  company.  We  had a
serious  discussion  concerning the  right  route of  administration.   Naturally, the
inhalation route is very expensive.  Finally, we chose the oral route for two reasons.
First, during  inhalation a good part of the material will be getting into the GI  tract.
Second, I understand  that this materal is mostly used for the copiers, typewriters and
so forth.  The chance for the operator to get the chemicals on their hand and into
their mouth is quite  great.  For these two reasons,  we  also chose the  oral route to
study these dyes.

Dr.  Page, EPA: Since we  have been talking  about  inhalation,  let us continue with
that subject.   We have a question pertaining to the chronic  inhalation  of  short
asbestos fibers, less than 5 microns.

Question 3;  What do  you think are the implications for the health and engineering
controls for the current OSHA asbestos standard? As a result of your findings, what
types of engineering controls would you recommend for the new asbestos standards?"

Mr.  Platek, NIOSH:  Some of that was answered in the talk, but I might comment on
another section of  it.  You said, "What precautions might we take from our findings?"
I honestly don't have  enough  findings to comment on any type  of  engineering controls.
At this point  we have not finished the study.

Question  fr;   How  do you  assess  the  potency of a carcinogen?    Can we  be
quantitative?

Dr.  Page, EPA: This  one, we could spend a  lot  of time talking about - how to go
about addressing potency and the attempts and the failures  and the successes, but let
us hear a few thoughts on it.

Dr. Saffiotti, NCI: I have my standard reaction to the word "potency" which is, let us
not use it.  We don't know what it really means to say potency of a carcinogen.  There
are more and more examples which would show that what might be a measure  of the
level of effect of a carcinogen in a certain set  of  conditions becomes a very different
type of effect in a  different  set of conditions.
                                       560

-------
There is a recent study, for example, of Lisa Prane from Bar Harbor in which she has
tested rnethylcholanthrenes by injection in 8 or  10 strains  of mice, inbred strains of
mice, at a certain  dose and got a level of response different in different strains so
that she could rank the responsiveness of these  different strains, and then she tried
the same experiment with a lower dose of methylcholanthrene and, lo and behold, the
rank of response was all different.  In fact, it was practically reversed, and this was
done  several times.   It was  published  in  Science last  year.   So, here you have
essentially a reverse  order of  response, and therefore an implication of different
potency, if you wish to use that term, in different biological systems, but it  is at a
different level depending on the level of carcinogen used.   There are many more of
these examples that make  it very difficult to utilize the concept of potency as if it
were  an inherent property of  the chemical  until someday when we will be  able
perhaps to narrow down  that definition to some particular biochemical interaction,
such as DNA binding.   We know already in DNA  binding, for example, that there are
all  the  host factors  that control  the  level of  binding and  enzymatic  activity,
penetration, distribution, passage  through membranes so that the same  carcinogen
would bind very differently under these different conditions.  Potency is really the
measure  of an  interaction  and therefore should not be used as a  measure  of  the
property of one of the components of the interaction, and I am  just sort of advising
everybody to stay away from it. In the IRLG report that we put together last year,
we  let it sort of sneak through  in  one  case. There was one phrase with the word
"potency" in it, but in most other cases we systematically  replaced it with the word
"sublevel of effect" which then required specifying where and how and under what
conditions.

Dr. Gregory, CPSC:   Dr.  Saffiotti, the  level of effect must certainly be detailed.
However, I would  like to  refer to  your  own  work  with benzidine,  or toluidine,
dianisidine, and 2-naphthylamine in hamsters.  I am certain that you would not put
toluidine or dianisidine in the same category of  potency as benzidine or 2-naphthyl-
amine, and yet we do  need to make some sort of an evaluation because we do need to
know in a very real sense how dangerous toluidine is compared to 2-naphthylamine.

Dr. Saffiotti, NCI:  I  agree with  the  obvious general concept behind all this.  The
difficulty is to be sure that we can make a predictive projection that would cover  all
the individuals.   Are  there some individuals that are particularly  susceptible  to
toiuidine carcinogens?  We  simply do not know that.

With epidemiology, in  some cases, we will eventually build up a sort of reasonable
basis for some  general assumptions.  Let me give you an example of some  of  the
difficulties in that area.  We  have always  thought  of 2-naphthylamine, the alpha
naphthylamine, as pretty much being in that category like toluidine, dianisidine and
so on in terms of being a much  less reactive  compound with lower toxicity and lower
effect in various biological systems.  We have recently done a fair  amount of work
with aromatic amines in the Ames system, and sure enough alpha-naphthylamine and
toluidine and dianisidine  were much less effective as mutagens  in the system than
benzidine, 4-amino-biphenyi, beta-naphthylamine, dichloro- benzidine and so on.

We  have  then  started doing  work on  cell  transformation with  the  3T3  mouse
fibroblast system and doing toxicity work in  that system, and there  is a result that I
am  still  not convinced about. I am still trying to see if there has been some mistake
there, but  alpha-naphthylamine is much more toxic to these mouse fibroblasts than
the beta compound. The first reaction I had when I saw the result was that somebody
had  mixed  the  samples,  and  this  has been checked  and  done over  again, and


                                       56J

-------
apparently there  is no such simple explanation.  We don't know what it is.  It may
eventually be sorted out.

Dr. Gregory, CPSC:  I believe  we need to be  very careful about  whether we are
talking about carcinogenicity  or cytotoxic effects, because they do not  necessarily
correlate  one with the other, and with respect to - we all know the differences in the
extremes, and that is that we  certainly  don't want to call saccharin the same sort of
carcinogen that could be identified  with bis(chloromethyl)ether.  We  know that we
have both animal and human  data for  bis(chloromethyl)ether,  that it is, indeed  a
carcinogen and a very potent one.

My question is whether  or not we are  satisfied  with saying that bis(chloromethyl)-
ether is a potent carcinogen and saccharin is a mild carcinogen or do they need to be
somewhat more quantitatively ranked.

The  National  Advisory  Board has come out with  some  of  the criteria that  are
necessary to be taken into account,  such as shortened latency period, dose response
effect, and whether it is metabolized in the bioassay system the same way that it is
metabolized in the human.  There were seven  criteria, some of them much more
important than others.  Is it not possible that you could come up with an algorithm
whereby you  could evaluate the substance  with respect to all the data, and  this,
again, Dr. Saffiotti, I agree with you, we don't have all  the answers yet, but  on the
basis of what data we now have, could we come up with an assessment of potency?

Dr. O'Conor,  NCI:  I think Bruce Ames  is trying  to do this, and maybe other  people
are, too.  The question is when you come up with it is it going to be really biologically
meaningful, beyond what we already know from  one extreme to the other.  It would
be ideal to be able to prioritize them in a  very  specific way, but I think what Dr.
Saffiotti and others are  saying is  that  we really don't have the type of information
today to do this  where it  would be  meaningful  in terms of human experience.  It
might be meaningful in terms of the specific animal experience.

Dr. Saffiotti, NCI: Perhaps in the continuation  of our workshop this may be  one of
the points that we might want to pick  up for consideration and recommendation of
further  work that could be done  usefully in collaboration  with the three  agencies
because we may  combine really useful  information from human studies, monitoring
and environmental measurement of  exposures,  biological  responses, and biological
models.  It may be one of the areas of quantitative studies of carcinogenesis that may
be an important one to keep in mind for  collaborative research.

Dr. Page, EPA:  My own  thinking,  being in a regulatory  agency and, also,  having
worked in the research environment and now trying to bridge that gap, knowing that
in the regulatory agency you  have to make estimates, is that you have to come up
with your best guess  as  to the likely impact of  exposure to a  particular chemical.
You cannot use only the innate biological  activity,  but  you have got to  look at the
exposure from  the viewpoint  of absorption  and target tissue deposition - how the
chemical passes into the body and has potential for interacting.

I  think that is what Dr. Gregory is alluding to, not simply potency from the viewpoint
of the initial biological activity, but  the total spectrum of what happens.  I think that
is what the regulatory agencies have to  come up with - a type of weight of evidence
assessment.
                                   562

-------
Dr. Kraybill, NCI:  We had discussions on this some years ago, two or three years ago,
and I think Dr. Saffiotti was involved.  Potency, as he indicates is hard to define.  If
you want  to call it  effects, that  might fit  the bill, but it  is conditioned on what
species you use and what strain you use. You know, you could  use the C3H mouse and
get a big response; yet in another strain you may not.

I wonder if it is not more  important to develop a classification system such as IARC
is using. We, and  Dr. Hueper, used to call them strong or weak carcinogens.  Well,
now, IARC  does  not do that.  They  limit it, and  in  their monographs  they say,
"carcinogens with sufficient evidence and carcinogens with limited evidence."  That
sort of  classification helps me a lot, and then you might classify  chemicals  into a
group where you have a physical-chemical phenomenon like an irritant effect. That
is important to know. Some derive their effect, like say a calcium compound from
becoming an irritant to the bladder wall where you get nephrocalcinosis or problems
of that sort; and then, last but not least, are  those that  would fall in the category of
promoters.  I noticed in a  recent article that  Boyland published, he labeled saccharin
a promoter. That makes all the difference in the world in my opinion. Now that has
gone from the classification of a weak carcinogen  to  a promoter.  So,  I think a
classification scheme might be a step, a first step in that direction.

Dr. O'Conor, NCI:  This is  as close to a consensus on this subject as I have ever heard.
Obviously,  it is a terribly important subject, and we have started,  and  we certainly
will come back to it probably today in another forum and certainly tomorrow  at the
Plenary Session.

 Question  5; Was DMSO used in the in vitro carcinogenic system?  Were radiolabeled
carcinogens used to see the uptake solubility tested?  Was liquid aqueous solubility
determined?

Dr. Gregory, CPSC:  The answers to those questions are yes, no and  no. I asked those
same questions of Dr. Dunkel and I think she is the best one to answer  that.  They  did
use DMSO.

Dr. O'Conor, NCI:  Are there  advantages and disadvantages in this  over the hamster
system? So much work has gone into the hamster system, I am wondering how far  we
should  continue to  go in  terms of development and refinement and  validation of a
series of these mammalian transformation systems or whether we should really be
putting  more concentrated  attention  into a couple of specific ones and  get  them
standardized, and  whether we should put more  attention into the  transformation
assays of human epithelial cells? I would like  to hear some comment on this.

Dr. Waters, EPA:  We have worked with the three leading transformation systems in
our laboratory and contract programs.  They are the  Balb 3T3 system, the C3H 10T Yi
system and the Syrian hamster embryo system.

I think it is fair  to say that the advantages of the mouse fibroblast systems are that
there  are  continuous  cell  cultures.  They are  very  reproducible.   They have some
spontaneous transformation to a  greater  or lesser  extent  in  certain  laboratories.
They have  a low metabolic activation capability  for  some  compounds.   This is a
disadvantage, and  for this reason the efforts have gone  into providing metabolic
activation either via co-cultivation with other cell types or with S-9  type metabolic
activation systems.
                                    563

-------
 The real advantage of the Syrian hamster embryo system as distinguished from the
other two is the fact that you retain in early passage Syrian hamster embryo cells a
fair amount of metabolic activation capability.

Theoretically, this should make it more useful as a detection system.  However, the
other systems may have advantages for mechanistic studies.  Their cell cycle kinetics
are very well defined;  the basic  enzymology and  so  forth is easier to establish
because of the fact that they are continuous cells in culture.

The variability of  the Syrian hamster embryo cells from preparation to preparation
can be overcome to some extent by freezing down large lots of characterized cells.
So, these are some of the trade-offs. I think you have advantages and disadvantages
depending on what you want to use the system for a screening system or a system for
mechanistic studies.

As for your other comment about the use of human cell systems, I think there is a
great deal of effort going into this area, both in the use of human fibroblast systems,
as well as human epithelial  cells  in culture.  Obviously, if we are going "to^study
car ci no genesis we want to study human  epithelial cells, but human fibroblasts may
offer a great deal of advantage in terms of their utility.

I think that the human cell work should be very much encouraged, but not to the
exclusion of systems such as 10T&, 3T3 and the Syrian  hamster embryo system that
may be more useful for routine screening and  perhaps confirmatory bioassays.  I
would put them in the latter category, confirmatory bioassays.

Dr. Page, EPA:   The Europeans seem very nervous about  the  possibility  of cell
transformation systems being introduced  in a battery of tests by the OECD.  You are
familiar with this type of regulatory scheme  for testing new chemicals.  How would
you assess the current  state  of the art as far  as regulatory requirements to have
transformation tests performed?

Dr. Waters,  EPA:  I think that transformation systems have come a  long way. I think
they are really not readily available in testing laboratories at the present time, and
this  would  be one of  the major  concerns  that anyone would have in terms  of
regulation requiring the testing in transformation systems as a key component.

There are laboratories that can reproducibly perform these assays.  I think the assays
themselves can be carried out in a reproducible fashion. There  are major difficulties
that we are  encountering, especially now, and they have to do with, for example, the
shortage of  serum.  This has  been  a severe  problem.   One  has to  characterize the
serum to be  used  in these systems to make  sure that it will provide a reproducible
system.

This kind  of  problem  has caused difficlty.  There have been many  laboratories that
have tried, for example the Syrian hamster embryo system, and had difficulty  with it.
There was a  period of time when hamsters were apparently  infected which produced
many difficulties  with  that  particular system.   The hamsters seem  to  have been
cleaned up at the present time.

I think it is those kinds of concerns that have warranted, perhaps, some concern over
the use of transformation  systems  in a regulatory context.  However, at the same
time  you have  to look at  what  alternatives exist.   If you  are moving towards


                                    564

-------
carcinogenesis as an endpoint, I think perhaps you have the sencar system that you
could use as an alternative.  It takes more time.  It may be a little more costly.  The
question is what else to do, and it is a difficult question right now.

Dr. Page, EPA:  In your sencar system there are only a few places that really have
that going, aren't there?

Dr. Water, EPA:  That is right, the same difficulty in some ways, but it is a whole
animal system and it does  get  around  some  of  the problems  that in vitro systems
inherently have. Maybe Dr. Saffiotti would  like to comment. He has been  heavily
involved with this question.

Dr. Saffiotti, NCI:  First of all, I would like to agree very wholeheartedly with Dr.
Water's comments,  I think that the  issue is  an open and complex one, and  one we
need to continue  to develop.   The  methods that have been  developed in the last
decade or less are beginning to be used in a number of laboratories.   Certainly they
are not ready to be written into some regulation for any industrial or toxicology lab
to apply and use them  because even the sophisticated laboratories sometimes have
difficulty with these delicate systems.

There are, however, several obvious needs in research development in this area which
once again I think  are  particularly important in the  cooperation  of three agencies
because they complement each other.   One  is the point  that Dr. Waters has
mentioned on the  increasing difficulties  with sera,  and  the  possibility  that  some
systems may be developed to have  cell transformation even  in chemically  defined
media or  at  least  with very small amounts  of  sera present which would minimize
some variations  due to that.

The other, as Dr.  Waters has mentioned, is the problem of  epithelial cells and  of
human cells, human cells both epithelial and connective tissue cells.

I think that  in  the next  decade or so we will probably see, from general research
interest in this  area which is rapidly expanding, the development of a  battery  of
methods that will come out of  different laboratories  that will offer reasonably good
models for studies of transformation in major different epithelial tissues.

In  the  last  few years we have had  progress  with  the  respiratory epithelium  in
Nettesheim's group for example.  Of course, the liver has been available for some
time, the skin system, that is through our laboratory, and Fusinick in Germany has
been particularly working on it.

There  is  work  now that  begins  to extend to kidney,  begins  to  extend  to the
endometrium, for example, a whole variety of target tissues,  colon.  Some of those
are just beginning and will take years to develop, but I think that one could conceiw
of a situation, maybe a decade from now, where we  will be able to use reasonably
standardized cell  culture systems  to ask questions as to whether certain types  of
chemicals are particularly interactive with or capable of transformation at selective
sites.   Obviously,  this will have to be paralleled with  the  experience in  in vivo
observations and the counterpart of the human tissue in  vitro, and we  will have a
pattern that will evolve of  knowing what carcinogens are  particularly active  in what
target tissues and what modifying factors are key factors.
                                      565

-------
So, I think that this is an area that could be very fertile for collaboration in which we
will,  in  fact,  benefit  from  the interaction that  comes  from  EPA and NIOSH
laboratories that see some of the human exposure problems combined with those that
we see in NCI and you know, stimulate interest in the whole field.

Dr. Page, EPA:  As you probably know, there  are a number of schemes that are being
proposed now for regulatory use which would  have the transformation assays sort  of
like  a second level in a tier scheme.  Once  you have results in bacterial  systems,
Ames  test,  or  some other system,  this  would trigger  the possibility of doing a
transformation assay  more or less from the true positives  I would guess. Then  based
upon results of that you could go on if you have a  positive to a longer term test,  or
else at that point conclude from a negative result, that there is very little potential
risk for carcinogenicity and stop at that level.

So, there is, certainly, in the  regulatory  scene, a big movement to try to use the
transformation assays. My concern is like Dr.  Waters that we don't push this  too fast,
but that  where there  is an available technique, we try to implement it, but do  it in a
meaningful way.

Unidentified  speaker:  Is  it possible or is  there any idea that transformation assays
might be useful for detecting possible teratogenic effects?

Dr. Page, EPA:  It has been my understanding that the correlation has  not been that
good  between  teratogenic effects  and positive  transformation results.   We are
considering this  kind of a  possible screen as to the rationale for whether to require a
teratogenic test.

Dr. Gregory, CPSC:   I remember that was discussed at the meeting over in Monaco,
and I believe that the conclusion was that many of the teratogens were indeed cell
transforming.   They  would transform  the  cells,  but  not everything that  would
transform the cells  was indeed a teratogen.  We had many transforming  types  of
substances that did not turn out to be teratogenic at all.

Dr. Page, EPA:   There are other mechanisms for  teratogenic effects  that are not
genetic in nature or would not  respond to a transformation assay.  That has  been my
understanding.

Dr. Waters, EPA: I would like to say that  I think most teratologists don't feel that  in
vitro  methods  are  really applicable, except for limited  chemical  classes   under
well-defined  and known conditions, and  so  I think  I would accurately reflect the
feeling of teratologists by saying that the answer to your original question is no.

Dr. O'Conor,  NCI:  Again, in a provocative vein, I am glad to hear you say  that you
are beginning to really talk seriously about the tiered system, using the in vitro test
for a screen.

I certainly appreciate what has been said in not pushing them too fast, but I wonder if
we can really afford  to not push as rapidly as possible, because the in vivo  tests,
really from simply an economic point of view, are incapable of doing the job that we
think probably needs to be done in  terms of  the number of compounds  and the
screening that would be ideal.
                                    566

-------
We hear at the National Toxicology Program that implementation of TSCA will take
care of a lot of this, but I wonder do we really think that industry is going to be able
to do a better job of the in vitro screening or invo testing with a lot more compounds
than the government has been able to do.  We have had enough trouble, plenty of
trouble, trying to grind out in vivo tests at great expense and I understand now that
we are only going to get about 40 a  year because  of budget restrictions from the
NTP.

Dr. Bull, EPA: I am  not going to do much more than repeat what other people have
already said. Basically this problem has been recognized  within the EPA at least that
people are familiar with the issue within the Office of Toxic Substances and other
program officers. Cell  transformation assays or any other assays to function at the
tier two level have to reject false positives or they really do not serve any function
not already met by tier  one tests such as the Ames test.  With that limitation, I think
our inability to say that these systems would rule out false positives and at the same
time not produce any false negatives makes their application at this point in  time
promising but still kind  of hypothetical.  I think there needs to be a  lot of work done
before any of these systems, in vitro or in vivo, can be considered fully validated for
regulatory purposes.

Dr. O'Conor, NCI:  To be argumentative, in terms of the in vivo tests, it  is certainly
clear whether it  is  positive or negative that in terms of human application we don't
really know whether they are false positives or negatives  in many instances.

Dr. Page, EPA: I think  it is quite certain that there will be a form of either a battery
or a tier scheme  come into existence, either on the international scene or within the
EPA or other regulatory agencies.  We are having difficulty in how  to structure this
tier scheme.  Certainly, a lower scheme is  going to be inexpensive  tests and I don't
think there is much doubt that a bacterial test belongs at that level.

I know Dr. Waters has proposed schemes and Dr. Lee now is working on one which will
possibly be put into place in the EPA.  The trouble, as I see it, is where to place this
transformation assay?   What  emphasis to put  on  that particular  scheme or that
system? It is not so much the bacterial system.  I think  it is pretty well agreed that
it will go into a lower level scheme.

Dr. Waters, EPA:  It seems to  me that if one is taking a tier testing approach and
directing  one's  attention to  carcinogenesis  then  the  appropriate place  for  the
oncogenic transformation assay  is  in  the  second tier, among the confirmatory
bioassays. I think most  people would agree with that.

A lot of people, though, are favoring the combined use of mutagenicity bioassays for
gene or point mutation together with oncogenic  transformation  bioassays, simul-
taneously testing for mutagenic and potential oncogenic effects.

If one is trying to determine whether a compound is negative, it seems to me  this
latter approach   makes  some  sense.   If you can test  for point  or gene mutation,
primary DNA damage,  chromosomal  effects and  oncogenic  transformation, and  a
compound does not come up positive in one  of those  assays,  I think you  have  a
reasonable assurance, at least in a preliminary fashion, that that chemical is not very
biologically active relative to mutagenesis and  presumptive carcinogenesis. So, I just
wanted to mention that  I think you can take the straightforward carcinogenesis tract
or the mutagenesis tract, or you  can also take a combined approach.  A battery of
                                    567

-------
tests  to  delineate  a  presumptive  negative  response  for  both mutagenesis  and
carcinogenesis might be the one that I mentioned.

Dr.  Morris,  EPA:   I  think  we  need to  consider  these short-term tests in  a
retrospective sense.  We are all talking about their use in a prospective, predictive
sense. I think there may be, and I think it needs to be looked at more thoroughly, the
use of these short-term tests retrospectively, that is in  your bioassay results in  the
areas of the gray zone, as they are sometimes called.

I think the time that we are detecting strong, positive carcinogens is rapidly passing.
We have a pretty good  handle on those now.  There may be a few sleepers out there
we have not pegged down, and we  are  looking for those obviously, but a lot  of  the
data that we are seeing falls into an equivocal area and, by itself, I cannot  decide
what  the bioassay means.  I think that in a retrospective sense, perhaps  these
additional clues  or additional  weights  of  evidence from  short test  studies may be
useful in a retrospective analysis.

Question  6;    Was  ultrasound treatment of  activating  cell  systems,  plus  ultra
filtration, plus exposure of xeroderma cells to the filtrate considered?

Unidentified Speaker:  The idea behind this is if you have a certain chemical which
does not get inside the cell you cannot expect any kind of activity. So, it goes to  the
same solubility and compartmentalization of a certain chemical.

My question  was if  this  particular type of  chemical  got into the  cell,  it was
metabolized, then if the cells leaches it out into the surrounding area, you mentioned
already that you did a preliminary study that through a millipore filter it  gets through
at least in certain cases.   So  the question comes up, going back to my  early studies
that radio labeled material was taken up by the cells.  We sonicated, broke up  the
cells and then ultra filtered it and used the filtrate for further studies.

Dr. Waters, EPA: Those kinds of studies are very useful in the sense that one can
track the pathway of the chemical,  but  I think  the  question that was  asked
immediately following my  presentation about the  half life of certain metabolites in
culture  is very important  there.  It  would only work if  one had a chemical with a
metabolite that had a relatively long half life. Some of them are in milliseconds.  So,
it  depends upon whether or not the chemical is sufficiently stable, and I think that it
is  important  to know something about  the stability of the chemical  that you  are
trying to test, too.  I think that is something that we tend to neglect in a lot  of  our
testing.

Question  7;   As  useful at the  Ames  system  is with  bacteral  cells  it has  some
handicaps as to relevance in metabolism extrapolation to in vivo systems. Do we feel
we should place more utilization of cell  culture, primary or tumor cells?

Dr. Waters, EPA:  Is the question either/or?

 Dr. Page, EPA:  It is  the priority we give to  the resources. Should we place more
resources on methods  toward utilization of cell  culture rather than  Ames type
systems of cell culture of the primary cells or tumor cells?

Dr. Waters, EPA: It  is always difficult when  you are in an audience and you have
some  people interested in some systems and some in others. I guess the answer that I
                                   568

-------
would give is that I think all of these systems are important.  We are supporting and
continuing to support all of them. I think they have different applications, and I think
that is  the key.  At the same time, as I  pointed out, in conjunction with the
metabolism studies at Michigan State University, it is clear that the metabolism, the
S-9 metabolism which  we  also used  right away with cell  culture systems is not
entirely accurate, if that is the term  we want to use in reflecting in vivo metabolism.
We know that is the case and yet we still say, let us go ahead and use it because we
understand more about it.  We have got a larger data base  at the present time.  I
think it would be foolish to make a shift to any other metabolic activation approach
for general screening at the present time, simply because we don't have the data
base.

With regard  to relative funding,  I  think Dr.  Saffiotti  made the case  that  there
certainly needs to be a great deal of emphasis on cell culture systems and especially
human cell culture systems.

There are some studies being done analogous  to  the one that  was mentioned, Norman
Anderson's study at Argonne. Similar work is  being done by Dr. Kakunaga  at NCI and
by Sachs in Israel, that may provide  the means to detect changes in specific proteins,
specific gene products that occur with cell transformation.

I think that these  kinds of studies in simple systems may provide important clues  as
to what is really involved in transformation,  what the gene regulatory mechanism is
that controls the transformation process.

This could be extremely important in the basic  understanding of carcinogenesis.  So,
these kinds of things definitely  should be supported.  They  are  going to be costly
because they are uncertain. I think it is a question of our objectives, if we want to  do
testing and want to test a large number  of  compounds,  then there are probably  no
immediate substitutes for the microbial systems, but if we want to become closer  to
the human situation and we want to learn more about the mechanisms of carcino-
genesis per se, then the  microbial systems don't help us.  So maybe that  answers  to
some extent, diplomatically, your question.

Dr. Bull,  EPA:  I  would like to add to what  Dr. Waters said and take it  a little bit
further.  I think of the things that we are struggling with, and  I think it was evident in
some of  the comments that Dr. Saffiotti  made  earlier, that  we have no standard  of
comparison in the whole area of chemical carcinogenesis.  One of the things that has
been neglected is research  to  establish the basic reasons  as  to why one  species
responds in  a certain way to a chemical carcinogen and another species does not
respond or responds in a somewhat different manner.  These questions are fundamen-
tal to the problem.   This is the key reason why we cannot answer questions very
clearly  when  it comes  to  considerations of potency.  I think  that  same  kind  of
approach suggested by Dr. Waters  needs to be taken into the whole animal arena  as
well.

Dr. O'Conor, NCI: We have another consensus that we don't  have the knowledge yet
to introduce  an effective in vitro  all encompassing screen, and  we need  a lot more
research in the mechanism of transformation. We certainly would buy that from the
NCI side which brings me to  a  general question in terms of the  philosophy of this
particular program.
                                 569

-------
The  NTP is  not represented in an  official  capacity at this  particular meeting.
Obviously one of the big interests is  in in vitro  tests and in the consideration  of
prioritization of compounds in terms of "potency."

This is clearly an area of research that the NTP must be interested in. Can we have
some discussion or does anyody have any viewpoints as to how or what emphasis this
particular program should  give  to  the  subject  which  is clearly of  interest  to
everybody here, and how this can best be coordinated with the NTP.

At the present most of the funding for that program  is coming  from the NCI with
some from NIEHS, but very little or none so far from EPA, and I don't think any from
NIOSH.

Dr. Galbraith, EPA:  The problem you mentioned was addressed by Dr. Vilma Hunt in
her presentation to the NTP Executive Committee in January 1980.

Dr. O'Conor,  NCI: That must have been the meeting I missed.

Dr.  Galbraith, EPA:   I  think  the  question  of  which research  areas should  be
concentrated  on  by  the  various government research programs  is one  that weighs
heavy on all  our  minds, particularly in the austere research funding period that  we
seem to be faced with.

Dr. Hunt addressed three categories of research that the Office  of Health Research
at EPA is concerned with. However, before reviewing these, I would like  to point out
that the  mandate of the EPA Office of Health Research is to generate, evaluate and
continually update the scientific and technological data  base necessary to  support
EPA regulations.

This  is  what  EPA  research is  all  about,  i.e.,  to  support regulations that are
promulgated  as a result  of EPA's effort  to enforce the  Clean  Air Act, the Clean
Water Act, the Safe Drinking Water Act, FIFRA, TCSA and the Resource Conserva-
tion and Recovery Act.

The  first category of  research (at least in the  view of the Office of Health Reearch)
includes  tasks that  must  be  addressed  by  EPA as  a  result of  (1) regulatory
requirements  and standard  setting responsibilities.  The  second category would be
research which EPA  must  address  but  collaboration  with  other programs would
prevent a duplication of  effort and conserve federal resources, that is collaboration
with other programs would result in an extension of our  own work, a verification  of
our own work and an increased understanding of  the  basic mechanisms that are  in
operation. The third category would be research which is more appropriate for other
agencies to perform.

Category I -  Research which must be addressed by EPA/ORD;

      A.   Methods Development

          Develop accurate and inexpensive screening methods for:

                carcinogenesis
                reproductive effects/teratology
                neurotoxicology
                                     570

-------
                cardiopulmonary effects
                effects on other target organs

     B.   Testing of Environmental Agents

          Toxicological  evaluation of  specified agents and  complex mixtures  to
          meet regulatory deadlines.

          Chemical analysis of environmental media and human tissues to document
          human exposures.

     C.   Criteria for Assessment of Test Data

          Improve data base for the interpretation of test results derived from
          animal studies.

          Develop data base to extrapolate animal toxicity data to human toxic
          dose.

Category  II  -  Research  which  EPA  must  address  but collaboration with  other
programs would be helpful;

     A.   Validation of Tests

          Validation of proposed methods
          Confirmation of test results

     B.   Fundamental Research in Support of Applied Research

          Development of animal models that mimic susceptible human populations.

          Characterize mechanisms of chemically induced disease processes.

          Development of new sensitive analytical detectiqn devices which identify
          environmental pollutants.

     C.   Collaborative Support

          Build a health effects data base for a wide range of chemicals.

          Interspecies toxicology.

Category III - Vital research which is more  appropriate for NCTR or other agencies
to perform;

     A.   Basic Research

          Mechanisms of action

          Etiology of disease processes

     B.   Special Testing Conditions
                                    57J

-------
           Performance  of the rodent carcinogenesis bioassays on potential environ-
           mental carcinogens identified by EPA.

           Examination  of  dose-response as  applied  to the  bioeffect  levels and
           presence or absence of a threshold dose phenomenon.

           Performance  of  long-term carcinogenesis studies  to  serve  as  reference
           for validation of short-term in vitro and in vivo methodologies applied  to
           environmental samples by EPA.

           Performance  of  long-term carcinogenesis and toxicity studies involving
           environmental samples  designed to determine risk to people of ambient
           levels of contaminants.  These would  be very large  scale experiments
           involving samples  selected  on the  basis  of EPA's  short term  testing
           program.

The categories are not absolute, and  the attempt to categorize research on the part
of EPA is provided only to give us a framework in which to discuss our  research and
to ensure that what we are doing is pertinent  to the EPA mandate.

There have only been two interaction bioassays  (additive risk)  studies, one  sponsored
by the NCI Bioassay Program, the  results of which  have not been published because
the statistical work is not complete. This study was  conducted  at SRI International.

The second study is currently in its initial phases at the NCTR.  We plan to look  at
four carcinogens found in drinking water. It will  be at least five years before we
have the results of that study and we will be fortunate to have them at that  time.

The other topic which falls into Dr. Hunt's third category is an understanding of the
strengths and limitations of structure/activity  relationships for toxicological  evalua-
tion.  Structure/activity  relationships have proven very valuable in the design of new
therapeutic agents.  There has been an increased  desire on the part of regulatory
agencies, including EPA, to  determine the status  of the usefulness of structure/-
activity relationships in predicting the toxic effects of chemicals.

Dr. O'Conor,  NCI: We see evidence that this is  an action program between the three
agencies, and I guess one of our jobs is to point  it out now in more specific directions,
but it still leaves the point, as to how do we divide the responsibility for certain types
of research between the monies available to this program and the NTP.

Dr. Page, EPA:  It seems to me that the uniqueness of this type  of a  collaborative
effort  would be the combined energies and expertise in directing the results of the
basic  research  program or basic  research  findings into application  in  addressing
regulatory needs.

Now the regulatory agencies often do rely  on results  coming out  of  the research
agencies and  often they are a little uncertain as to  how good some of these systems
are and how to use the systems.  We know that some of the systems require  additional
validation.    But a  program  like  this, joining forces  of  research and  regulatory
agencies can really bring to bear the type of expertise that is needed to push what I
call basic research results on into a mode that a regulatory agency can then utilize.
                                    572

-------
Dr. Kraybill, NCI:  I believe Dr. O'Conor was getting into a deeper subject. I think it
is  implied in his remark, why we didn't invite  a  broader  delegation here.  We had
considered it, but we  thought since this was  our first effort we should confine it to
EPA,  NIOSH and NCI, but maybe next year we may broaden our horizons and scope
and have other agencies.

The other thing is, I think Dr. Galbraith also defined a problem  we must be aware of,
that each of us has what we call  turf.  We have missions and for some of these
missions, I am  sure  between two of the agencies that were  mentioned, you cannot tell
where one leaves off and another one begins. They are very close, not  so  much with
NCI but I think in the environmental area some of the agencies get pretty close.

Dr. Gregory, CPSC: I think what Dr. Kraybill said is very appropriate.  We  all tend to
have our own turf and Dr. Galbraith talked about how EPA has defined  theirs.  There
needs to be some sort of  an  interagency coordination.  Quite often there is going to
be some overlap, especially in  areas  involving  NIOSH  and EPA.  They are going to
both be interested in some of the same things sometimes, and you cannot  avoid some
of that, but when things  are really  flagrantly out of control there needs  to be some
sort of an interagency collaboration so that one group will take the lead and provide
the results in  one area while the other agency will take the  lead and provide the
results in another.  We will  be  able to save  a lot  of money this way and make our
overall research much more dollar efficient.

Dr. Rausa,  EPA:  NIEHS publishes  an Annual Report to Congress entitled "Federal
Agency Support for Environmental Health Research.  The report contains a descrip-
tion,  by each  agency, of its area  of responsibility and program in  environmental
health research.  A copy of the  report may be  obtained  from Dr. Phil  Schambra,
Office of Interagency  Programs, NIEHS.

Question 8;  How can regulatory agencies keep from interfering with the development
of improved methods for toxicity testing or defining the types of data they  need to do
their  job, that  is required  protocols  for toxicology  testing, for example, the Ames
testing may get in the way of improved, more accurate, less expensive protocols.

Dr. Page, EPA:  This is  an  interesting question.  I know we are trying to deal with
development of guidelines and standards at this  time, and in the back of our mind we
want  to be sure we  don't  put something down which is going to stymie further
development of test method development.  So,  is there anyone who wants to take a
shot at that?

Dr. Waters, EPA:   I think maybe you can answer it better than I, but  is it not true
that with respect to testing  protocols under TSCA, for  example, you have  the option
to revise on a yearly basis the protocol? Isn't that true?

Dr. Page, EPA: We have more  than one option.  The law requires that  each year we
at least examine the existing standards and determine  whether they need  to be
revised.

Dr. Water,  EPA: At least under that  legislative mandate, I don't think  it necessarily
is a problem. Certainly, I think the scientific community is in favor of flexibility in
protocol application.
                                      573

-------
Dr. Page, EPA:  One of the greatest  difficulties we have, in fact, is the degree of
rigidity versus flexibility in setting our standards.

I  think we certainly want to lean toward flexibility, but then on the other hand, in the
regulatory agency, you have to have something that is enforceable.

So, you have the other  influence  of putting something down that  in  the  event  you
don't get the kind of data you need you have got something to enforce to go back  and
require additional data.  Under TSCA with a testing rule, we are required to put down
standards for the testing.  The question is to  what depth do we go in putting down
these standards.

Dr. Waters,  EPA: I would just like to use this  opportunity to say something about the
GENE-TOX  program which, if some of you have not heard about it, I think is probably
important to mention, and it relates  to this  question.  Understanding the need to
define  as precisely as possible protocols for testing, one of the  efforts that has been
supported by the Office of Pesticides and Toxic  Substances in EPA  has  been  the
so-called GENE-TOX  program,  the current status of tests in  the  area of genetic
toxicology.  One of the charges to the committees (and there are about 200 scientists
working on this program with 22 committees  on different short-term  tests) was to
attempt to come to an agreement as  to what the current acceptable  protocol is for
any given test system. The charge did not say that you have to define that protocol
such that the system could be used for testing now.  We simply wanted an evaluation
of where that test system stands. It was left as a judgment for the scientists that  are
intimately familiar with  the test system under evaluation to make. I think that was a
very enlightened  approach, and I hope that the information that comes out will  be
used in that  fashion.  I think it probably will be.

Dr. Kraybill, NCI: I have a practical question to ask Dr. Page since  he was with  the
testing program  at  one time.   You  remember that  guidelines for  testing were
developed by NCI, and lo and behold we discover one day that  EPA  comes in with a
guideline.  What  do you  do about situations like  this, when you have NIOSH, OSHA,
EPA, FDA,  and NCI each  with their  own  protocols.  I know it is bad to  hamstring
people  and lock them in,  but you  face the danger of one agency  being played  off
against the other, play A off against B.

Dr. Page, EPA:  I will give one quick comment and then  I see Dr. Morris wants to
address it.  You  have a very valid concern.   This  is one of  the  reasons for  the
Interagency  Regulatory Liaison Group -  to attempt to coordinate or harmonize  the
regulatory directions and activities. Under the Interagency Regulatory Liaison Group
there is a test standards  and guidelines group. This involves the regulatory agencies,
but NCI is participating on  the group, along with a couple of other research agencies.
Now there are going to be guidelines developed for chronic  toxicity.  They have
already developed a  number  of  them for acute  toxic effecs, but  the  guidelines
developed for chronic effects, including carcinogenicity, are only now being drafted,
and I can mention that it seems that  the  direction they are probably going in is to
stay as close as  possible to the  recommendations of the International Agency  for
Research on Cancer.  A recent conference was held, about one year ago in Hanover,
and certain  recommendations have come  out  of that work group.   The Interagency
Regulatory Liaison Group is trying to adhere as close as possible to their recommen-
dations.
                                       574

-------
Dr. Bull,  EPA:   I just wanted to make one comment, going back to the original
question about whether specifying guidelines, specifically guidelines for testing under
TSCA negates the ability to develop methodology. I don't think that is the intent of
it.  I think that is a misunderstanding and to clarify that I will state that I think the
development of  guidelines is really  an  attempt to make the contractual agreement
with the industry you are requiring to do the test clear, so that they understand what
you will accept  as positive  or negative evidence of safety at a given point in time.
This really has nothing to do with retarding methods development, in fact I would see
it as a stimulus for methods development.  In short, you are asking them to develop
the information  in such a way that you  feel that you are going to have a good chance
of understanding the results. Am I making that clear.

Dr. Page,  EPA:  I think you are basically right, but I can tell you one of the results of
this  guideline and standards activities  is helping to define the uncertain  areas, the
highly controversial  areas so that 1 think it will lead to additional  or new research -
to try to resolve some of the prevailing issues.  It is likely going to be years before
the research can yield the results we would like to have.

Dr. Morris, EPA: I agree with you 100 percent on the activities and the usefulness of
the IRLG  exercise.  I would, also, like  to make the group here aware of a document
we  have  developed through another international  effort with  the Organization for
Economic Cooperation and Development that Dr. Page, Dr. D'Aguanno, and I have
participated, representing the United States in developing short as  well as long-term
guidelines including a carcinogenicity guideline.   That document is available for
review and comment  within this country in  all of the regional  offices of EPA and in
our Headquarters Office.  We would certainly appreciate comments on them.  This
was another attempt to try to harmonize at  even a 2^-nation level our approaches.

Dr. O'Conor, NCI: Everybody is trying  to coordinate and harmonize. There is a new
program started at WHO, called the International Program for Chemical Safety in
which the WHO, the UNEP and ILO and maybe  the FAO are all involved,  and this is
sort of, again, a  grand scheme for coordinating information and activities in  the area
of toxicology and chemical safety throughout the world.

There is one question here that  relates to several things, one of which  we talked
about,  and that is  the predictability of short-term tests and the  suitability of the
animal models, but then it goes on to talk about approaching the problem of threshold
in a heterogeneous  population and  interspecies extrapolation.  That is  a subject,
again, that relates to some of  the things we have  talked about, and it probably has
relevance, considerable relevance to the program.

I guess what I want to get back to is the focus to help Dr. Kraybill  from the NCI and
the others who are responsible  for management of this program to ensure that we are
really spending these funds in the most useful and meaningful way.

Initially, and to  a certain extent it is still going on a little bit, I think the different
agencies are using the funds to  do projects that they would have done anyway and this
is  not  to  say that  they are  not useful, but I think  we really  ought  to make a
concentrated effort in trying to define the program a little more carefully.

Certainly, in epidemiology this is the subject of another session, and we will have the
plenary session  tomorrow.   That clearly  is an  area where the  NTP is  minimally
involved and where this type of program, I would think,  has a very  large role to play,


                                      575

-------
because all three of the agencies have an epidemiology component, and some of the
other agencies don't really have that component.  I think we can  be  real leaders in
that area if we continue to enhance the degree of cooperation that  has been initiated.
In this field of toxicology and methodology, though, there really is a lot  of overlap
with NTP and other organizations, and I think I am not quite sure how to handle that.
I  guess there is so much to be done that we need not worry too  much about overlap,
just make sure that we are supporting the best work and the highest quality research.

Dr. Kraybill,  NCI: I would like to make some practical comments regarding this.  For
FY 1980, the funds are just about all committed. For FY 1981 we have almost a full
commitment  of funds. We had looked upon this sort of effort to  get new ideas and
stimulus for new  projects  for the ensuing years. That would be  helpful.  It would be
very difficult, let us say, to cut off a good project. Oh, you could do that,  but I think
it would be chaos  for some of the projects committed for three or four years.

We would like to get good  projects and the kind of advice today is very useful to us.

I  don't quite  see that this collaborative program interfaces or impinges much on the
NTP.  From what my understanding is,  maybe  some of the papers we heard on the
NIOSH  side  this  morning  may  have been, but  on the NCI/EPA  side we are dealing
more with mechanisms and surveys, like, for instance, the Gulf Breeze, Florida study.
Certainly we are  looking into mechansims and things of this sort that are really not,
as I understand it, necessarily the mission of the NTP alone.

Dr. O'Conor, NCI:  I guess primarily I am referring to the field of development and
validation of  short-term tests.

Dr. Kelsey, NCI:  We hear a lot about the development of the short-term tests which
obviously are going to be very important, and they are  getting better and better;
particularly  one area that has not really been talked  about, the  area of metabolic
activation. However, with regard  to the animal tests we don't hear much on how to
make the bioassay a more economical model, if that can be done.  I don't  know what
research efforts are being done, but I am sure people here are aware  of it, and it
would be good to know what is being done in the area of in vivo animal  models, in
terms of carcinogenesis. I know some people are interested in developing, maybe,  a
small animal population or a different species and that sort of thing, and it would be
good to get some  information in those areas.

I  would like to hear some discussion about what is being done more with the in  vivo
models in terms of either streamlining the tests or improving the  animal bioassay?

Dr. Bull,  EPA:   There are  several  things that can be  done,  and they are being
explored.  One in  particular that we are interested in in Cincinnati  has to do with Dr.
Farber's preneoplastic foci which can be found in the liver following treatment  with
carcinogens.  We  can develop this system as  an initiation  promotion  assay  and  that
would considerably shorten the time needed  to assess carcinogenic potential.  Of
course, it needs to be validated in the sense that you want to see how  predictive it is
of actual tumor development and so on and so forth.

The mouse skin system, if you believe  that the papilloma development is somehow
predictive of the  development of a malignant neoplasia, can be  used as  a short-term
assay. Perhaps Dr. Whitmire would like to say something  about the Strain  A mouse
which we are also quite interested in.


                                    576

-------
Dr. Saffiotti, NCI:  In addition to these somewhat new uses of these in vivo systems,
particularly for the  intermediate  range of  duration, what I think  is really the
important  area that is  still  developing  is  the utilization of  in-depth  research
methodologies for definition of tissue  susceptibility,  metabolic activation, organo-
tropism and specificity, interactions and so on, to improve our ability to interpret, to
evaluate, and to improve the methodologies for in vivo bioassay studies.

The  animal bioassay, work which has  been sort of  codified  and  started  as  a
large-scale effort about the last 10 years or  so, was based on  a  lot of research on
animal models on induction of major forms of cancer reproduced in animal systems by
a variety of carcinogens that was done in the 10 or 20 years before.  We were able in
the early  seventies to crystallize  some of  the animal test protocols that are still
largely  being used because  we had the  10  or 20 years  of experience behind us on
which to draw.  We have now a lot of other information that was not there 10 years
ago, especially in mechanisms, metabolic activation specificity, markers for interac-
tions, and markers for binding of carcinogens. That is the kind of work that can be
eventually brought to bear  on  the  interpretation  of the bioassays, on the design of
specific subsets of bioassays.  For example, instead of  just doing simple general
toxicity, one can do and is beginning to  see work done that addresses  these problems;
studies  on distribution, binding,  interactions, repair, all these things which will
qualify the in vivo response.

The other thing is in the development of the in vitro models we still need to draw on
the experience  of the in vivo models, especially when we are trying  to correlate
organotropism and target effects in different tissues, epithelial tissues in particular.
Again, I see that as an area in which the work  being done now and  that has been done
in the last few years  will bear fruit in the next few years in the development of  more
specific methods and  more specific protocols.

So, I would hope that we would continue to work in a collaborative fashion in a way
that would be  research oriented in that  sense and  then provide  almost a resource
from which those who are particularly concerned with developing protocols, including
NTP, will draw to develop their additional and more modern protocols.

In terms of our relation to  NTP, let me make a  sort of general comment on this.  I
have been expressing this to several of our colleagues in the other institutes and in
our own institute. I see a certain amount of  polarization. There is NTP going towards
the fairly straightforward testing  approaches with pretty  much existing  methods.
Some of the areas of  major  emphasis in the research program of EPA and NIOSH are
addressed  to  very important areas of  methodology and data  and  research  on
monitoring, on  environmental definition of exposure, all  these  aspects which are
somewhat at one extreme of the spectrum.

NCI, in the context of the NIH basic research orientation, and because of the  rapid
development of very exciting models  in the molecular mechanisms of cancer,  is
putting  most of its emphasis  on the more basic molecular and more basic research
approaches, and there is  this area which really links  the  two which  is somewhat,  I
think,  undersupported  at the  present time.    I personally  think  that  the  best
opportunity we  have  is this one of an  interagency commitment in  which we really
share a strong  interest  in  this middle area  of  studies of mechanisms of toxicity,
mechanisms of disease related to specific problems that  affect human exposure from
both sides.
                                     577

-------
We can, in fact, both contribute specific scientific methodologies and  experience
from both sides, and this program could well become a unique program and therefore
perhaps become unassailable from the fiscal point of view and all the rest of it, if it
is, as Dr. O'Conor said,  more than just more of what each agency would  be  doing
anyway.

Dr. Whitmire, NCI:  I have  heard a lot here today regarding our studies at NTP, so I
thought I had better stand up.

Dr. O'Conor, NCI:  You are still NCI.

Dr. Whitmire, NTP:  I appreciate that.  Sometimes sitting on the fence gets kind of
wobbly.

We are  indeed, trying to  improve the protocols.  We are trying to include parameters
which will assist in evaluating toxic signs.  If we know that certain aspects of clinical
chemistry can  assist us  in  setting the doses for the chronic study, we will include
these.   Behavioral studies may be included.  Any type of study which may give us
some indication as to the organ involved may be included. These studies are included
in the initial experimental design or may be added if results of the subchronic studies
indicate their usefulness.  I think you will see more and more improvements in the
experimental design because each chemical is assigned to a manager, known as the
chemical manager,  and they are the ones that are the pivot point between  NTP and
the other agencies.  I plead for a lot of help from the other agencies in letting us
know who in their agency is responsible for each chemical.  We have more trouble
finding  out who to  contact so that  we can  take advantage of  their expertise in
designing the most useful studies than any other aspect of our job.  Anything that we
can obtain from the other agencies in this area of cooperation would be most helpful,
but our  job  is toxicology,  not just  carcinogenicity.   People are nominating  these
chemicals now for our use, and when we cannot find out who nominated or why they
nominated  them we have a difficult time designing the best  experimental design, and
so this is important.

I might  say, also, that we are indeed  trying to develop both  in vitro and in vivo
short-term tests.  We have at present two contracts out to  verify and determine if
two different laboratories can do the "A" strain lung carcinogenicity assay, and when
this is done blind, then we will see  how  much  use this  test will be.  We  are also
interested  in promoter studies and combinations of  chemical, but  these studies are
very difficult to design and until we have  an  organization,  people, and laboratories
these will have to wait.

Dr. Page, EPA:  We were just  wondering what would be a logical follow on.  We have
got a meeting session tomorrow in which we are going to have to come to grips  wth a
few suggestions as to where this program should be going.

Dr. O'Conor, NCI:  Since  there are only a few papers tomorrow in the morning, maybe
we ought to think  about moving pretty  quickly into the plenary session and then
allowing plenty of time for any general discussion.

Dr. Page, EPA: Unless there are other viewpoints, we will adjourn for today.
                                     578

-------
         FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

            PROGRESS ON 3OINT ENVIRONMENTAL AND

                OCCUPATIONAL CANCER STUDIES
                   Wednesday Afternoon, May 7
                          SESSION C
WORKING GROUP ON THE IMPORTANCE OF INTERAGENCY PROGRAMS;

    DEVELOPMENT OF FUTURE COLLABORATIVE PROGRAMS AND

          MEETING THE NEEDS OF REGULATORY AGENCIES
                    SESSION CHAIRPERSONS
                       Dr. Richard Marland
                  Environmental Protection Agency

                        Dr. Nelson Leidel
          National Institute for Occupational Safety and Health
                                579

-------
SESSION  C  - WORKING GROUP  ON THE  IMPORTANCE  OF  INTERAGENCY
PROGRAMS;  DEVELOPMENT OF FUTURE COLLABORATIVE  PROGRAMS AND
MEETING THE NEEDS OF REGULATORY AGENCIES

Dr. Leidel,  NIOSH:   The  intent of the session is  to  discuss  the importance  of
interagency  programs, development of future collaborative programs and meeting the
needs of the regulatory  agencies.  Probably the last is a very key issue.  We do not
have  a  structured format.  Hopefully,  there will  be free-flowing  discussion.  Are
there any particular questions for the first discussion topic?

Dr. Cameron, NCI:  I would like to mention a couple of  things.  Do  all of the people
here know some of the background?  Dr. Herman Kraybill gave  some of the opening
remarks  and he laid the groundwork. Is there anyone that did  not  hear, who would
like some amplification of what he was talking about?

Dr. Leidel,  NIOSH:   Maybe it would be  a good idea to go over  it again  real quickly
because I was not able to make it yesterday.

Dr.  Cameron, NCI:  I will  try  to keep  it brief.   Simply,  both  of these are
congressionally mandated situations. They arose out  of  the need or  the desire  of
Congress  to have the interagency collaboration they felt was missing.  Congress felt
that there was a  non- meshing of gears  so far as expertise, and since the  National
Cancer Act  of 1971  passed to the NCI considerable  funds, Congress said that the NCI
would pass money to other agencies, figuring where  money goes, interest will follow.

That is-not a bad philosophy, and it has  seemed to have worked out. I am  the third
coordinator  for the  NIOSH/NCI program  so there is a  lot of history  that I am really
not familiar with.

Dr. Kraybill mentioned this morning that initially, it was strictly a transfer of money
from NCI to NIOSH, and they really did  with it what they wanted.  That was in the
days of  Dr.  Finklea who allocated the funds amongst his staff and amongst his own
projects.  But we are now coming back  into  an era where there is truly much more
collaboration. Within the last year and a half that  I have been active in this program
there has  been a lot more input by NCI.  Within the last 2 months, 2  out of the three
projects recently authorized were NCI-initiated.

We are not trying to change the ballgarhe by any means, but I just hope that  this type
of situation  continues,  and  that  NCI  will  participate   in the  NIOSH  interagency
agreement with some new ideas.

The NCI/EPA interagency activity is quite different.  Dr. Kraybill has been involved
in that since its inception and that has truly been  a collaborative effort and all the
projects were pretty well split up the middle.  Some  have come from NCI, and a lot
have come from EPA, and it has been a  much more organized activity, only because
it has been in one  man's hands from the beginning  and he has done a good job with it.

With that, I will stop.  I would mention one thing;  there are ceilings on the amount  of
funding to the extent of $4 million dollars for each program. The funds come out  of
DCCP/NCI.

I  don't know  whether Dr. Kraybill  mentioned that  in  the NIOSH agreement since  it
started in fiscal year 1976, there have been, counting the  three we  passed this month,


                                     580

-------
71 projects. One of them is a duplicate.  This project, developed last year, was never
funded. It just did not get anywhere. They killed it, and then they reactivated it this
year at a higher level.  Nineteen projects have either been completed or terminated
for due reason.  So we have about 50 ongoing. In the EPA Agreement - I do not think
they have completed any projects - they have something like 30 projects underway.

Mr. Harris, NIOSH: I have been involved with the NIOSH/NCI Interagency agreement
for the past 2  months.  It has become apparent  that  most of the projects are
concerned with identifying problems.  What about resolving some of  these problems,
resolving some of the issues in terms of, for  example, control technology?  Most
NIOSH/NCI efforts are  concerned with basic science in terms  of trying to identify
the problems, but, what  about  the  other  side of  the  issue,  trying to  resolve the
problems with practical  solutions?

Dr. Cameron, NCI: That was alluded to  in the third part of the charge to us today:
How can we be more responsive to regulatory activities?

Dr. Leidel, NIOSH: Yes. I think, in terms of problem  solving.  However, the way I
think of it, regulation, of course, is one tool, but even when you  issue regulations, the
companies  have to have some  mechanism  for controlling the exposure,  and I think
that is what we are getting at.  In other  words the control technology or  control
techniques, I would probably envision in the general realm of public health.

Dr. Cameron, NCI: The only ones that come to  mind are the projects concerned with
protective  clothing and respirators.  Is that what you had in mind?

Mr. Harris, NIOSH: Yes.

Dr. Cameron, NCI: Then I would certainly agree that there are a few projects in that
realm but not too many, certainly not the bulk of them.

Dr. Marland, EPA: May I inquire if any  of the projects which have reached you and
Dr.  Kraybill for  consideration  have been the result  of  the Division of  Research
Grants or,  say, Dr.  King's  extramural program references? Have they  referred to
any of these?

Dr. Cameron, NCI: I don't believe so. I cannot speak for Dr. Kraybill.

Dr. Marland, EPA:  What is the nature, then of  the  peer review that these are
provided with?

Dr. Cameron, NCI: Well, it is basically an ad hoc situation between the senior staffs
of the  concerned agencies.  In fact, Dr. Cooper is Associate Director for  Extramural
Affairs in our division and he happens to be on the committee we have set up for that
purpose.  We  have Dr.  Saffiotti, Dr. Weisburger, Dr. Fraumeni,  who  ^fairly  well
represent the senior people of our division.

Dr. Marland, EPA: So  the review is intramural rather  than including any personnel
outside?

Dr. Cameron, NCI: There is no outside review.  In fact, up until last year, anything
coming to us from NIOSH was presumed to have gone through their  peer review; we
were only to consider their proposals on the basis of relevance to cancer and priority.
That has changed a bit.   There  is now a tendency to look at the science, too, which

                                        58J

-------
makes it sort of difficult because obviously the cancer people look at it from slightly
different perspectives than NIOSH, and it caused the non-funding of several  projects.
But  I  think  we  can  overcome  that philosophy problem  with  a  mechanism  for
presentation, and instead of having a very skimpy  written proposal handed to  the
Cancer  Institute, we are now going to go into more of a combined staff review with
an oral presentation by the proposers.

Dr. Marland,  EPA:  Would the principal investigator make the presentation himself or
herself?

Dr. Cameron, NCI:  We have just done that several times, and it seems to work a lot
better than simply sending a piece of paper to us. Dr. Cooper and I are very familiar
with that system.  We are comfortable with it because that is the way Dr.  Saffiotti
started  the carcinogenesis program, and we were both members of it for a very long
time. It is not  an adversarial position when you come before the group, but after you
get your teeth kicked in the first time, you do your homework the second time.

The one that  comes to mind, for instance, that we just had was Dr. Brian Hardin who
came with a styrene inhalation study. That was a tricky concept, and we discussed it
in the  group  for about an  hour and  a half with Dr. Hardin there. It did  pass, but I
think if he had not  been there, it might  have  been  in jeopardy.  We had  legitimate
questions of  the rationale  for the protocol and some other aspects  of the  study.  I
think Dr. Bridbord is geting the same feeling, that  it is probably the way we should
approach  it  for our agreement  and probably for  the NCI/EPA  program, I would
encourage such a mechanism.   It  also  has  the advantage of  getting some  more
interaction between the staff.   When our people approve a  proposal, they have a
chance to talk to the Principle Investigator, to interact with him.  I know in  one case
Dr.  Weisburger said, "I like that project  and I understand what he is trying to do; I
would like to  be a co-project officer  with him." So there are a lot  of spin-offs that I
think should  be encouraged, but it would  require in  our case with NIOSH  that  these
people would come  in  throughout the year and make presentations.  Now, I can  see
one problem.  Secretary Harris is clamping down  and is essentially telling  us that we
are now working on a three-quarter year. She  does not want anything  funded in  the
last quarter.  So essentially you are going to have to fund in the first three quarters,
which sort of makes it a skimpy year.  Last year, NIOSH met in the summer and  got
all their projects in line and worked it out staff wise between themselves,  and then
they come in during the balance of  the  year  with  other proposals.  This new  edict
from the  Secretary's office will  compress that whole process. What I am  saying is
that the NIOSH people, for new  projects, are  going to have to start coming in  the
first of the fiscal year so we can get them  approved and start funding them right
away.  I cannot speak for the EPA agreement, but I would check into it.   We just do
not have  the luxury  of 12 months  any more to get things  approved and funded.
They've  got  to push  them  in  the  first of   the  year.   The philosophy  of  oral
presentations by prospective Principle Investigators  will make for a busy beginning of
the year but I would encourage it.

Dr. Leidel, OSHA:  Of  course, there  is the problem with travel funds constantly being
decreased. Every year the cost goes up 10 percent  and  the amount of travel  funds
goes down by 10 percent.

Dr. Cameron, NCI:   I  appreciate that.   I think it can be justified on the basis that
we've got a  $4 million pot, and the only way we  can  spend it  wisely  is to have
face-to-face discussions.  I think some provision has got to be made for  it.
                                    58?

-------
Dr. Leidel, OSHA: I do not think there is any basic quarrel in NIOSH that it is a good
idea.  You've got to have a good idea and  the person has to be able to defend it.

Dr. Marland, EPA:  What concerned me was my recognition that those projects which
came in from the EPA staff had indeed not received an external peer  review. They
were  the  proposals  that  had  been received from certain investigators around  the
country that appeared to be quite cogent and appropriate to the regulatory needs of
the agency.  I see how you are finessing it.  I think that is a very satisfactory way of
doing it.  I have no  qualms  about people like Dr. Armstrong  or, for that matter,
certainly Drs. Fraumeni and Kraybill sitting as peer reviewers of research.  I am  not
questioning  that, but I  am questioning the absence of any of this.  Sometimes  the
enthusiasm of a non-current bench researcher carries him away when he is maybe 5
years out of step with  current practices, current literature.   I would not consider
myself a satisfactory peer reviewer because I have not kept up in my field.

If I have gained a sense that my fears  are put  to rest, that is what I was looking for,
and you have pretty  well done that.  I know  the stuff coming  out of EPA now will
have had at least an  opportunity and probably will have received  a vastly improved
review by both  extramural and intramural peers than it used to.

Dr. Leidel, OSHA: I would like to pursue  Mr. Harris' question.  I think it bears on  the
matter  that NIOSH,  as I  was in it,  always prided itself on  both having research
interests in terms of basic and applied research for problem identification such as in
the area of  epidemiology.  Of  course, we have always had a mandate and a  role and a
very strong interest in developing the solutions to the problems,  the applied research;
everything,  of .course, from monitoring  methods to control technology, both in  the
form  of  engineering  controls  and personal  protective equipments, particularly  the
area of respirators.  In other words, once you have found that a  chemical is a suspect
carcinogen or it caused a particular health effect, we sense  a strong public health
role to  do something about it and make  contributions in that area.  I think what he
was trying to get at is the question, could we be devoting an increasing share of  the
funds in the interagency programs to that kind of an effort, which related to OSHA's
needs, because when  we go  forward  on  a health  standard, we  both  have  to
demonstrate, of course, that there is a problem, but a very strong input we have to
consider is the  regulatory analysis, which considers both the feasibility of even doing
something about the  problem, other than simply  putting a respirator  on the worker,
and then the cost of these controls.

Dr. Cooper,  NCI:  I  think one of the problems in this area is that  although there  are
some NCI activities that are directed toward these goals - for example, how one  can
accomplish degradation of carcinogenic substances in the laboratory environment  - it
is a little difficult to see how the  Division of Cancer Cause  and Prevention of  the
NCI can  contribute  much in the  way  of expertise  to the other sorts of control
activities that you are talking about.  Perhaps other areas of the Institute might be
able to contribute but I do not see very much intellectual input that we can  provide in
this area. Perhaps NCI  cannot really collaborate in any real sense in such efforts.

Mr. Harris,  NIOSH:  You mentioned protective clothing, and one  or two others.  Areas
such as these are meaningful  to both agencies.  I believe this is what Dr. Leidel is
referring to in his statement.

Dr. Cameron, NCI:   There is a constraint, albeit minor, that I understand in at least
the NIOSH/NCI agreement.  There  must be a connotation of  occupational cancer.
NCI funding has to have  some relevance to, hopefully, cancer  prevention  or cancer
                                     583

-------
identification.  I  am not negating the missions of some of you people, with all  the
other parameters of toxicology, but when  our people look at these proposals, they
have got to keep that in mind.  I am not sure if we can justify NCI partial funding or
full funding of subject matter that  does  not have some relevance to the cancer
problem.

Dr. Leidel, OSHA:  I think NIOSH would probably look at it from the standpoint that
the way you prevent workers in an occupational environment from contracting cancer
from a  carcinogen  is to keep the chemical from reaching the worker, which  means
either through engineering controls  to prevent its  release into there workplace or
personal protective  equipment once it is in the workplace air or the environment.

Dr. Marland, EPA:  Is  there any reason why such research - if I could use the word
"applied" as opposed to a more basic development of the identity of the cause and  the
solution - should not be the function of the regulatory agency as opposed to a more
pure research function that would be the major concern of the NCI?

A prototype that  we are beginning to develop, which Mr. Costle last week described
in  rather strong terms to the Congress which made our group quite happy, is that  we
were instructed  by the Congress some 2 years  ago to allocate 15 percent of  our
research costs to long  range,  more fundamental, research. In other words, to identify
causes and effect relationships, perhaps, or mechanisms, or whatever else would be
called a more basic or a fundamental or a longer range project. The other 85 percent
therefore,  presumably, would be  dedicated  to  the application  of  some of  these
principles through  a  regulatory mode.   Now, whether  the 15 and 85 is a rational
percentage, of course, if  a  subjective entity and is not necessarily a  good or  bad
figure, but at least  it is a recognition that there are two  functions involved here, both
of which are called research.  I think that  you were trying to state  that the talents
that NCI could apply were perhaps more nearly into identifying the basic mechanisms
for the  NIOSH  research  group,  and then when these are  identified,  the NIOSH
research group can understand  how  we can apply certain preventive mechanisms to
this, which appears  to  be a rational way to go about it.  This  is what we  are indeed
attempting to do at EPA.

Dr. Cameron, NCI:   There is another interesting  feature that you may not be  aware
of. The funding from  NCI to NIOSH is not necessarily  all contractors.  They use a
varied mix and this impressed me. A project might  be either completely staffed and
operated in-house in NIOSH, in which case the funds allocated are used  internally to
pay the salaries of their staff people while they are on the project, or they are  a mix.
They have staff time allocated and paid for and they go out and get a contractor  for
support, usually  a  resource  support, and then there are  some  which  are entirely
contract operated.

Dr. Leidel, OSHA: Well, of course,  the problem  as I see it at  OSHA is that they  are
dependent upon the research  agencies, and they  do  not  have quite the luxury  of  the
control over what needs to be done.

I know that when I was on the NIOSH end, there was a perpetual problem about OSHA
needing both long and short term research and trying to get that into the mechanism.
Of course, people come and go and all sorts of things and new ideas keep cropping up
in  terms of  ways of improving  communications and  getting the long range planning
between the two agencies, particularly in light of the fact  that we were set up under
the same Act.

-------
Mr. Turner, OSHA:  Of course, there is the formal mechanism of the NIOSH Planning
Group which was set up a couple of  years ago to develop the long range needs that
OSHA and other Department of Labor agencies  could  foresee in both safety and
health.

I really  cannot say  ~ you or somebody else  from NIOSH can probably tell me how
that is working from the NIOSH end.  NPG has only been effective for a year, so we
really do not know what results we are going to get.

Dr. Leidel, OSHA: I could not speak to that.  I was not involved in that group when I
was  at NIOSH.  The predecessor groups seemed to work.  It goes back to the  basic
question of how do research projects get proposed,  and, of course, there could be two
routes.  Down at the working group level, that is at the branch level, you can either
have the ideas come from the researchers or they can come down from the top.  It
comes over from OSHA and folks say this is an important  problem and it goes down to
the  division  and branch  chief  and  then  to the section  chief  and down  to the
researcher, who writes up a proposal trying to answer  the problem.  So these are
some of the realities and problems that have to be faced. Of course, how well these
working groups pan  out many times would  depend on the personalities and the kind of
people that are sent to them. So, I really cannot answer that.

Dr. Marland, EPA:  There is another similar function that  is now going on at EPA, and
that is the research committees. There are people here who are probably more aware
of problems than I am, but  it is a  research planning function in which the actual
research budget for the agency is developed by the research committee, co-chaired
by a senior  official whose responsbilities  are regulatory, and the other co-chairman
being a responsible research director. Committees sometimes run 50 persons in size.
They are dealing with a  budget that runs close  to $300 million divided among 13
committees.  So, it is a very substantial undertaking, and the  agency is pretty well
bound by their portion of the resources into various kinds of projects.

This is not an old, enduring function.  This is a function which is still evolving.  It is in
its second year.  But it equates to that planning function which you have at OSHA.

Mr. Turner,  OSHA:  The same kind of thing, yes.  This may be too basic a  question,
but it is one I presume both you and I and other OSHA people would like to ask.  What
research functions  does  EPA have  and  how do  they work with  their regulatory
division?

It was mentioned that NIOSH is our sister  agency and does our research for us.  Is all
your research done within EPA itself?

Dr. Marland, EPA:  About one-third.  If you are talking  about allocation of dollars,
about one-third of our research budget is dedicated to intramural research within our
own staff.  Close  to a third is  interagency agreements with the Departments of
Energy and Agriculture, but a substantial  amount  of that is with DOE, and another
third, but slightly  more  than a  third, is  extramural, which ranges from  a grants
program,  which is  quite  small  but  is substantially a "hands  off" program,  to  a
cooperative  agreement activity  which is  like a grant but is "hands on," through a
contract  which is hard and fast.

Each of these  elements is approximately 100  million, so they are somewhat substan-
tial. In other  words, you  can influence the course of  events with  that much of an
investment.  All  of this money is subject to the dictates, if you will, and the very


                                      585

-------
strong recommendations coming from research committees which are not composed
only of researchers. The most influential elements in the research committees are
those elements which  describe the needs of  the  regulatory aspects of the agency,
which  we call program  offices instead of sister agency. We use the term "program
office" instead of "research office."

The  program offices have an absolutely critical and important, an almost dictatorial
role, in relation to describing the problem that must be solved.  The research people
are expected to say, well, if that is the problem, here are our recommended solutions,
and here is how we should do this.  Ideally, the committees plot that kind of approach
and come up with a funding pattern for the entire research budget.

As I say, this is a system which is still evolving.  Everyone recognizes that it  still has
imperfections. It is in its second year, and we are making rather significant  changes
in it as we go along.

Dr.  Murray, EPA:   Interestingly, even though you have used the  term "program
office," it might be just as useful for this  group to think  of  it in terms of client
offices.   They are  the people who  receive  and  use  the  research results  in their
regulatory role.

Dr. Marland, EPA:  That also describes an attitude which reflects, in a sense, that we
are indeed a service agency.

Dr. Cameron, NCI:   Is there a list of projects that OSHA has suggested to NIOSH that
has been deferred because of funding or staff? Is anybody keeping a tally?

Dr. Leidel, OSHA:  Since I, again, personally  have not been involved, I  really cannot
answer that question. I know there have been lists of research needs sent over.  I can
speak  to one small  area where I have  been involved to a limited degree, and  like any
list of research concerns, of course, they come all the way from ideas  of particular
individual's concerns up to very major  issues.

Mr.  Harris, OSHA:  Well, the new cancer  policy may help address some  of  these
specific issues as we start identifying some of the materials to be regulated.

Dr. Marland, EPA:   Are people troubled within that cancer policy by the allegation
that there is a measurable  risk associated with a given environment and a  form of
cancer or maybe total cancer production and risk assessment techniques?  Any are of
you troubled by the problems associated with risk assessment?

Dr. Byrd, EPA: Well, I would not say I was troubled by them, since I am a consumer
of them, but it is interesting to me where things stop. The area of risk assessment
technology is one question that is  being addressed currently, but it is getting on the
table fairly late, and it is, as you describe it, in a two-year evolution.

Some omissions still exist in EPA's research funding strategy. As a matter of fact, I
would  really be interested in running  an experiment within this contract to  see how
the individuals in this  room think that new chemicals  are important in relation to
existing chemicals.   What proportion of the importance of the regulatory actions that
we  take  relates to  dealing  with the existing  problems now and what proportion of
importance would they  put upon preventing new situations from arising in the future
like  the  case  with  which we have to  deal today, particularly with respect to new
chemicals  coming on-stream?   I  would really be curious about what that sort of


                                     586

-------
allocation of resources would be like or what that allocation of importance would be
like.

One of the curious things is that the kinds of algorithms that are devised to deal with
chemical  problems are technologies which  are appropriate to deal with  existing
chemical problems.  They really are not very helpful in trying to cope with how you
assess  new  chemicals  which have  never existed  before.   It  is a  strange  kind of
omission.

I was examining the research that was presented at the morning session, which deals
with the three inputs into our decisionmaking.  We have a risk algorithm which says
we need to  know something about the biological effects, we need to know something
about the exposures, which is a concern that has not been here,  and we need to know
something about the control technologies which are available. Getting an "A" in two
out of the three and knowing nothing of the third will not get you a passing grade.

We have to use this algorithm in  shorthand for new chemicals.   There is no  data
analysis. It is all a hypothetical world.  It is all a future scenario world.  I personally
am of  the  opinion that it is possible to do research  in how to  get at those sorts of
things.  I do not know that it is appropriate in this particular context, but I can think
of some work the  NCI's Division of Cancer Treatment is doing that would really be
fascinating to translate over into the interagency cooperative research program.

The Division of Cancer Treatment  at NCI has been conducting computer assessment
of input chemicals. They have a large number of input chemicals in the program and
they have  to decide how  to choose which ones will be  bioassayed for  anti-tumor
effects.   DCT  has become skilled in assessing chemical structure with respect to
biological  end point.

Now, given  the nature of the kinds of  decisions you have to make  in regulating new
chemicals, an analogous kind of problem arises. I do not think research in this area is
coming through the  machinery right now.  EPA is not interested in addressing new
chemicals with respect to their anti-cancer potency, but, for example, with respect
to potential carcinogenicity. Any sort of research in an area that  will supply us with
tools to fulfill  these kinds  of needs will be extremely valuable.  I  do not know how
people would vote -on it.  I have my own personal view, which  is  obviously a biased
one, but it is a strange thing that thinking about things that might happen to us in the
future,  that do not exist  right  now,  is  something that tends to  slip  through the
decisionmaking process very frequently in determining research needs.
                                                             (
Dr. Marland, EPA:  Do you  believe that the area that you are describing, which is
extremely close to that which, quite frankly, troubles  me when I see rather expensive
kinds of decisions being made on  the basis  of  risk assessment technology which is
extremely fancy arithmetic — some of the finest statistics and the application of
statistical methods that are knbwn  are used -- but if  you look at it simplistically, the
existence  of 2ft  deaths from  arsenic-induced  cancer each year, is ..that a  good
number?

In the first place, are 2ft deaths significant?  Yes, if you are one of them, and yes, if
you are downwind of a zinc smeltering  plant.  But when  the  government indicates
that we will regulate arsenic on the basis that there will be 2ft lives per year saved,  I
sure hope  that  we are right on that 2ft value. You know,  if it  is  2,500  or if it  is a
negative and we find that there is a  potential error of  10 , that 2ft does not look
awfully good to me when you figure that the chances are that you are off by an order


                                     587

-------
of six magnitudes.  It could be 25 million, which we know it is not because there are
not that many people  dying with arsenic-induced cancer.  But the research, which
could be  collaborative, designed  to improve  methodologies,  I  would  find  quite
intriguing, at least I would find the results intriguing.  I would not find the research
intriguing at all because I would not know how to go about it.

You hit a point again that is very much a part of this, and  that is exposure. I think
that perhaps due to the efforts of NCI, the world has done a better job of identifying
health relationships to various substances, particularly cancer health effects, but the
rest of  the world, meaning  probably EPA and NIOSH, have not done an equally good
job identifying risks due to exposure and then putting these together to constitute a
risk assessment.

You say you  are  dealing with five equations and three unknowns.  It is rather poor
arithmetic on which to spend billions of dollars. This is why I would like to see  the
collaborative  program coming  into  some  kind of  methodology development or
improving the degree of confidence that our risk assessment technology is good.  I do
not want  to fault it.  Mind you, I have no intention of faulting it, because I do  not
have a better idea.  If I do not have a better idea, I do not  intend to fault something
that is  in place.  But I am worried.  It troubles me, because we are spending billions
of dollars as a nation in correcting problems where there is a degree of uncertainty in
the order of 10 .

Dr. Cooper, NCI:  Well, I think one has to say  that the National Cancer Institute is
also extremely concerned about both of the issues raised. They ultimately go to  the
question of risk assessment. Fortunately, or  unfortunately,  what society is concerned
about is the risk assessment in people, not risk assessment in mice.  That has created
some serious problems in the development of quantitative structure activity relation-
ships. Data bases are extremely weak in this area and if we are talking about human
hazard, we are dealing with a very small subset of chemicals which can be clearly
identified as members  of this class.

Now, when we talk about risk assessment, there is frequently a concept in the back of
people's minds that there are a few critical experiments which could be done, which,
if done well, would permit the  unambiguous extrapolation of  animal data to  the
human situation.   Personally, I do not believe that  is true.  Partially as a  result of
certain  questions from regulatory agencies  within  the last few months,  a  group of
people  were  brought together to consider what studies could be  done that would
facilitate that kind of extrapolation.   The kinds of people  that were involved were
those whom we all recognize as having outstanding expertise in this area.

The consensus of that meeting  was  that we  really  are not ready to identify  any
individual definitive studies that  will let us reach that goal. There are general areas
of basic research that  can be identified that in a 10-year time frame may permit us to
get at these kinds of definitive studies, but we cannot identify them now.

As a result of that meeting, there will be .a sizable program developed over the  two
years, largely based on RFA solicitation,  requesting  the attention  of  the  basic
researcher to specific  directed areas which may help resolve the problem of differing
carcinogenic response among animal systems.

Beyond  that I cannot go, but I really believe they are right.  There is not any simple
set of studies that can be  done today that are going to provide us with the answers
that the regulators so desperately need.  We can get clues, but that is all.


                                     583

-------
You are talking about arsenic. Let's face it, if you looked just at the animal data and
you did not know anything about effects on people, you would say that there is not a
cancer risk for man.  So, it is really  kind of dicey as to how far you want to go in
basing your risk estimations for man on single or even several determinations in non-
human models.

Dr. Byrd, EPA:  This is fascinating.  You just gave me half of the paradox that I
wanted to demonstrate.

Let me ask you what relative importance you would put as a percentage in the total.
How would you spend your time on new  versus existing chemicals?  Five percent, 10
percent, 20 percent, 2 percent? Oust pull a number out.

Dr. Cooper, NCI:  I think this could only be determined  on the basis of prevalence
information, and our prevalence information now says that there are something like
5,000 chemicals  that  are in major commercial usage; there are something like 500
new chemicals that are introduced each year.  I have no hope of catching up with the
backlog ever, but I would tend to distribute my resources roughly in a 10-1 ratio.

It may be a lousy answer, but it is the best one I can come up with.

Dr. Byrd, EPA: Nobody can come up with an answer; it is all guesswork.  The paradox
is this:  It is fascinating to me that you would feel that I, as a regulator looking  at
new chemicals, would only be able to look  at proven human carcinogens.  In fact, a
computer program that tells me whether there is a prediction of carcinogenic risk in
a mouse would make me deliriously happy.

I have no problem  regulating with the  shorthand information, and having to do it now
is really  based on  flipping a coin.  You  look at chemical  structures and you make a
guess.  A computer program which  gives  me something better than a random  or
educated guess, that of all  the  chemicals out  of the  universe, this  has  a higher
probability of causing cancer in a mouse, would be a great answer for me.

I construe that kind of computer work as a  very basic field of science.  I am aware
that NCI does research  in some areas.  I am not aware that it has  been plugged into
the carcinogenesis program, for example.

Dr. Cooper, NCI:  Can I ask a question, because I want to know if I am clear.

Dr. Byrd, EPA: Sure.

Dr. Cooper, NCI:  When you say a computer  program, are you referring to a data base
resource, or are you referring to a de novo calculation of probability?

Dr. Byrd, EPA: I am referring to a de  novo calculation of probability.

Dr. Cooper, NCI:  That is probably beyond the realm of existing technology.

Dr. Byrd, EPA:  Well, why would it be beyond the realm of existing technology for
carcinogenic potential when it is not for anti-cancer potential?

Dr. Cooper, NCI:  I do not know enough about the anti-cancer  area to even make a
sensible comment.  But  in terms of a de novo system for  carinogens  you must  be
prepared to predict metabolic pathways within jthe system.  I think that is the hang-


                                      589

-------
up.   In  terms of  direct  acting  carcinogens,  I think you and I could  build  that
tomorrow, in a simplistic sort of way. The system is not going to tell you very much.
It  is going to say that if the compound has a nitroso  group hanging on it, you had
better watch it, and that is probably the level of sophistication we would get.

Dr. Byrd, EPA: Sure, but even if you built me one, just  for fun and games, that would
give me  predictions of proximal carcinogens.  We  have in our  employ  a  number of
sophisticated chemists who could predict metabolic patterns, and I could  feed those
in  independently,  if I had such a computer program.

Dr. Cooper, NCI:  I hope you are right.

Dr. Byrd, EPA:  What I am saying is there is an assumption that whatever applies to
existing  chemicals must also apply to  dealing  with  new chemicals.  You made an
assumption.  I think that it is true that in  the  world of having to deal with passing
regulation for arsenic, that is already out there, you really are confronting a very
difficult  problem - is  it going to cause  cancer  in  humans.   But new chemical
regulation is a different world.  It is qualitatively different. The kinds  of decisions,
the kinds of thinking that you have to make in order to  make regulatory decisions are
just fundamentally different.

Dr. Cameron, NCI: Isn't there a  structure activity tree?

Dr. Cooper, NCI:  Yes. That is probably worth pursuing because we may find an area
of commonality that we can really do something with.  Quite a number of years ago,
the NCI  established, as part of a contract at Stanford Research  Institute, a chemical
hazard ranking index.  One side of it was an attempt to develop a structure activity
tree  which  would provide a  questioner  with  a  P-value  for  the   probability of
carcinogenicity of  compounds in the tree.   The  P-value  was  basically derived by
acquiring the view of experts on what particular structural  modifications in a  class
would mean in terms of  that probability. We effectively killed further  development
of that aspect of the program  a  few years  ago.   The basic reason was that the
contractor had drifted away from  his use of expert groups and begun  to introduce P-
values which were  derived  from  sources that  we  felt  were  questionable, so we
abandoned further development. But there is no reason that this aspect could not be
further  developed.  There are  some  programs in   the area  of synthetic  organic
chemistry which  allow you  to  predict  reaction products and  side  reactions,  and
attempts have been made to  plug them into this sort  of program in reverse to see  if
they  could predict the  metabolic pathway.   It is my  impression that these efforts
have  had limited  success, but that may well be because we have  not invested enough
resources.  Certainly, it is an area that could be further investigated. It would be of
interest to the Cancer Institute.  Anything we could do to  model the  situation better
would be enthusiastically greeted.

Dr. Mar land, EPA:  The tree  problem, I think, has already been looked at by EPA, and
unless I am mistaken, they have chosen to go the route of more traditional  testing for
demonstrable carcinogenicity in  their  Section  5  regulation requiring  industry to
pretest.  Because the affected parties — which  is, of course, industry — found quite
logical and perhaps, from their perspective, quite reasonable objections to classifying
a substance toxic on the basis of its structure, there  are substantial exceptions to
that,  and so, as in  most  technical and even scientific problems, it is not necessarily
the science or the technology that deters  the solution to  the  problem.   It is the
administrative process.
                                        590

-------
Dr. Cooper, NCI:  Well, clearly, what has been suggested would be a major research
effort, but it would have only limited value for regulation.  Basically, it comes down
to the  fact that the reason you have said, "X is more likely to be carcinogenic than
Y," is because someone, who is an expert in the area, said it was going to be that
way, and that is not a sufficient basis for regulation.

Dr. Byrd, EPA:  I make those decisions every day.

Dr. Cooper, NCI:  Well, I am glad I do not have to do that.

Dr. Byrd, EPA:  That is the difference.  The difference is, when we go to pull arsenic
out of  the environment, you are talking about something which is economically very
painful.  People are using existing  compounds because  they meet economic needs.
You are talking about high cost, you are talking about putting people out of work.
When the  decision is  made  to  regulate  a new chemical, you  are  talking  about
something that  does not exist right now.  The economic cost, except for the company
which has come forward with the compound, is extremely low.  The decision is most
often not to ban it or to let it be made with no restrictions, but rather a decision to
require more testing.  In actuality,  it is not that simple.  I am making things sound
easy when it is  not.  What we do is we regulate pending the  development of test data,
so we have to have a control option available also.  Again, my answer is that new
chemical regulation is  different, and that computer-based predictions would be very
useful in a new chemical regulatory environment.

Dr. Cooper, NCI: I am sorry, I feel  like we are monopolizing the conversation. I just
would be very concerned about trying to regulate on the basis of a de novo sort of
calculation for such compounds.  This is not a reason to say we should not investigate
it and  drive the technology further.  If you drive it far enough, it  may be  a very
useful system. But right now, I do not think it is.

Dr. Leidel, OSHA:   Well, to attempt  to  answer your question about  allocation  of
resources on  new chemicals, keep in mind my tunnel vision of just the occupational
environment as opposed to the fact that the chemical, of course, might be released
into the  general environment  via the  air, the water, or the  land.  We look at, of
course, the problem of where  to focus  our resources  in  terms  of, generally,  the
number of people  that might  be exposed, which then becomes a function, and the
hazard involved, which, of course, depends on the potency of the  effect, and
sometimes  the  amount of the chemical produced; but generally, it is the number of
people  exposed  because, of course, there  are many industrial chemicals made in this
country in billions of pounds to which there are relatively few exposures because they
are in closed systems or this sort of thing.  And  so, it does not bother me personally
when I hear about all the new compounds that are being  produced  each year, because
to  me  usually  they are  rather  limited  in  the nature of  their distribution and
poundages.  The thing that professionally has frustrated me for  many years is the
problem we have in attacking the known problems. I feel a sense of frustration from
the knowledge we have gained  from the NIOSH survey  back in 1973-7^.  The NOHS
survey  which we are now talking about is NOSH  II and it will be discussed tomorrow
morning by Mr.  Dave Sundin from  NIOSH.

Ms. Spadafor, EPA: I agree with you on the need to examine new chemicals carefully
because to prevent a hazardous chemical from being made is a first step. Why let the
chemical get to market only to later have to go through  the regulatory process to fry
to pull  it.
                                      593

-------
Dr. Marland, EPA: You get a much better bang for your buck.

Ms. Spadafor,  EPA:  I agree, but it is important to  know how the  premanufacturing
process wouid  assist us.   Could you  get  the information you need  to  prevent  a
hazardous chemical from being manufactured?  Could you get the information from
what is asked for in the notice requirements?

Dr. Byrd, EPA: Well, it is a question of what information comes in. What  EPA has a
legal right to in premanufacturing notices is very, very skimpy information. What we
actually get, in many notices, happily, is a lot better than, legally, we have a right to
have.  But we have to make decisions in some instances on very skimpy information
indeed.  Legally,  we do not  have a  right to very  much  information other than
chemical structure, for example, simplified use, and whatever health and safety data
is  available to the manufacturer, but he does not have to run out  and  get any data
that is not  "reasonably ascertainable."  Depending on which lawyer you  talk to,
"reasonably ascertainable" can be construed twelve different ways.

So, it is decisionmaking based on very skimpy information in some circumstances.  We
confront the  decision tree  problem again.  Industry told EPA before the  premanu-
facturing program came into existence that there was no such thing as a chemical
going into manufacture without an acute toxicity test.   Then, you know, to turn
around and tell us that  they  would feel  unhappy being regulated on the  basis  of
chemical structure sounds a little funny to me.

Dr. Marland, EPA: Kind of skeptical, too.

Dr. Byrd, EPA: Yes.

Dr. Cooper, NCI:  But this  brings us to another area in terms of old versus new.  We
heard a paper today on the effects of feeding of a benzidine based dye.  Last year, we
heard about an NCI study which produced tumors in rodents after 6  months  of feeding
of this dye.   But  Dr. Troll published his initial paper in 1971 on  the  excretion  of
benzidine in monkeys which were fed such dyes.  It was no surprise when we got the
results of the NCI study. We knew everything we  had  to know in terms of  what was
going to happen if humans were exposed to that dye in 1971. Where have we been ever
since?

Ms. Spadafor,  EPA:  We knew what was going to happen with asbestos in, what, 1940?
1930?

Dr. Cooper, NCI:  I am not sure of the specific date at which  we had adequate
information.

Ms. Spandafor, EPA: Facts about hazards of exposure to asbestos, as far as lung
pathology is concerned, go back approximately 40 years.

Dr. Cooper, NCI:  Let's accept that as an appropriate date.

Ms. Spadafor,  EPA:  I don't know what the government did,  or  if  they were equipped
to do  anything at  that time, but when the regulatory agencies were established, little
was done.

Dr. Cooper, NCI:   Yes, and your argument is clearly quite valid with  regard  to
asbestos, at least since the fifties.   Before  that, although we had some information,
we did not have very much in  the  way of a strong organization to  do anything about

                                 592

-------
it.  In addition, we had a war, and that war involved a lot of shipping, and shipping
required a lot of asbestos work, and it just was not seen as being feasible to regulate
at that time.  I am concerned with why it is that we want to focus our attention on
new substances when we already know all we need  to know about a lot of old ones,
and we are doing nothing about them, or doing very little about them.

Dr. Byrd, EPA:  It is a very reasonable question, one  which I cannot answer here, but I
think that one reason to look at the new ones is simply that you get the most bang for
the least  bucks  in  dealing  with new  chemicals.   It is a relatively  painless way.
Fifteen years from now, our successors will not have to deal with a whole new set of
aggravating  problems.   The benzidine dyes that you mentioned  are a fascinating
example of  the kind of difficult problems that you confront as  a regulator in new
chemicals.  The dyestuff industry  will, on demand,  make  a small batch of a totally
new dye.  The profit margins are small. The dye industry says, and I have no reason to
disbelieve them, that if we require them to do  even  minimal testing on a new dye lot
of that  sort, that they will simply choose not  to  make  it.   Now, we have a legal
mandate exactly not to do that.  We are told by Congress that we should  not  unduly
inhibit innovation.  It is very hard to justify knocking that guy out of the box on that
dye strictly  on an exposure  basis,  because he is not making many pounds.  The dye
industry has been particularly cooperative with EPA and what  we would like to get is
testing  on selected structures  within the benzidine dye  class  together  with an
agreement that says that they will agree to abide  by a structure activity  analysis.
The problem is that the research base is deficient; the research base which says how
would I tell  them which structures to do and how, exactly, would I make a convincing
structure  activity argument on a new  benzidine dye.  It is clear to me at this point
that not all of  them are  going to prove  to be carcinogens.   It is going to be some
subset.

Dr. Cameron, NCI:  If your structure tree was big enough, strong enough, it  would be
hard to come out with a new compound that you could not attach to part of the tree
and get some idea.  SRI is equipped to do that.

Dr. Cooper,  NCI:  Well, let's not make  any arguments for that  particular tree.  It has
a lot of holes and a lot of weaknesses. When you say benzidine dyes, I presume you
mean benzidine derived dyes.  If you have a modified  benzidine structure, you might
have serious questions about making a statement that  the compound is going to be
carcinogenic.  But  if you have  a  simple  benzidine structure with those two  diazo
links, I  do  not think you have  any reservation  at  all  today  about saying  that the
compound is going to be carcinogenic.  You are perfectly right in terms of the broad
class, but  once you  have gone beyond something like the IARC monographs,  you have
run out of data base.

Dr. Bellin, EPA:  I am sorfy, but I  think that the dyes are a perfect example of why it
does not pay to say the old versus the new, because while we are concentrating on the
"old" benzidine family of dyes, we  are forgetting about, let's say, thiazine dyes, which
have terrible problems, triphenylamine dyes, some of which are suspect.  $o it would
be very convenient  and kind of appealing to say, as in  hazardous waste, you know, the
old hazardous waste or the new hazardous waste, but I do not think that it is going to
work that easily, and I do not agree that  you  can simply  say we will  get the biggest
bang out of  a buck by  that  approach.  I think it  may well turn out to be somewhat
simplistic. I wish I had a better handle on it, but I do not.

Dr. Byrd,  EPA: Well, I had  not meant  to debate  that issue.  It is just curious to me
that the sorts of things that are appropriate towards new chemical regulation tend to



                                      593

-------
get completely lost in the research priorities.  I do not quite understand that, but it
seems to work out that way.  I can see plenty of things that can be done, and I would
anticipate that there are people within the basic research community that would find
those problems exciting to pursue.

Dr. Cooper, NCI:  What  are you thinking about when you talk about whole areas of
research that would be useful for the regulator that are falling through the cracks?  I
am not arguing that that does not exist, but what are some illustrations?

Dr. Byrd, EPA: Well, I am just particularly sensitized to the new chemical program,
because that is the area where I work.  My problems lie in those areas.

Dr. Marland, EPA:  I would  be surprised  if the research community  had the faintest
idea about  working in  that direction.  Perhaps the function that the EPA in general,
and your office, perhaps in particular, could undertake would be to direct some kind
of  fundamental research approach  in providing  you with tools,  whether  it  is a
structure tree — I am scared to death of  a structure tree. I am really afraid of it.  I
would hate to go  to court and try and climb the tree, if you will, and be successful.
You might get hung on a branch.  But really, the reason I brought up this whole thing
is that I am personally dissatisfied with the extent to which the research community
has  addressed  the  problems  of  you  people  who are  regulating  pesticides, toxic
chemicals,   and hazardous  wastes.   The  reason  I  brought up  this issue on risk
assessment is that we  are taking the little bit of known information  we have and we
are  putting this  through some  extremely fancy  statistics and coming  up  with a
number that may  appear  rational, but at  least it confuses  people so they do  not dare
attack it — well, they do attack it — but we can put all kinds of hand blessings on
these things and  give this figure  some  kind of authority which it  really does  not
possess.  Of course, the basic problem is a lack of functional knowledge about what
we are  talking about,  and  that  comes only from research.  That is why I tried  to
stimulate you,  since I  was aware, not of  your particular relationship, but I knew that
you were concerned with  the toxic substances program, and I am very much aware of
your lack of research.  Our job is to try and provide it but we are not giving you any.
By the way, you are not asking for any, either, but that is a small point.

Ms.  Spadafor,  EPA:   We  ought  to  be discussing developing  future cooperative
programs. I do not see how we are fulfilling our purpose at  this meeting - discussing
the  development  of  inter agency  programs   - unless we consider  the problem  of
coordinating our  activities  in risk assessment.  I was involved  in  the Interagency
Regulatory Liaison  Group, Regulatory Development Work Group for OSHA (Occupa-
tional  Safety  and Health  Administration), and I am presently a  member  of  the
asbestos work group for EPA. I notice that there is a serious lack of communication
among the IRLG member agencies. The  working group discovered that two  research
projects  being  done by two different agencies were identical.  There should be a
greater sharing of information and conservation of federal funds.  Better communica-
tion on  what projects  are being undertaken by each agency is needed so we  do not
have duplication of research.

Dr. Marland, EPA:  Some people would say that two researchers working on the same
topic is not redundancy.

Dr. Plotnick,  NIOSH:  On this  duplication, I think you  will see less of that in the
future,  and what you probably saw  was completion or soon completion  of two
identical studies  that  were  started three or four years ago in  most cases, not more
recent.

                                        594

-------
Dr. Spadafor, EPA:  Not in this case.

Dr. Plotnick, NIOSH:  Well, it depends upon which agencies, but generally, or at least
with the NTP program and some of these collaborative programs now, the duplication
has been cut down considerably.

Dr. Marland, EPA:  We have a new mechanism in  EPA now on our grants research
that, prior to  our  funding  any grant,  the National Institutes of  Health and the
National Science  Foundation, the Department of Agriculture and the Department of
Energy are canvassed  if the grant we are talking about is appropriate to those areas,
to see whether or not they  are funding that or a similar project.  Interestingly, we
have  found that  we  have changed our intentions  to  fund  or  not fund  based on
communications with  some  of  those agencies, because the same investigators, of
course, are sending their proposals to  all of these groups, therefore, it  is a very
important communication that  we  make.   That has begun within the  last 3 months
because that is how old our grants program is.  It is 3 months old.

Dr. Leidel, OSHA:  Well, just out of curiosity, do you happen to know, Dr. Plotnick,
how NIOSH avoids this? I assume that there are some sorts of lists of research being
done that  go into common data bases, and you plug in the chemical you  are interested
in and you see who is working on it or what has been in the reported literature.

Dr. Plotnick, NIOSH:  It is not  that mechanized. It is still a manual search system.
Generally, the  compounds go into the National Toxicology Program  (NTP),  at least,
but that is only a portion of NIOSH's total toxicology  budget.  But those go  in and
obviously are compared with all the other projects going in from the other agencies,
and  if there is duplication, it is going to be  noted and some modification will be
recommended,  and generally, being the  smallest, we are probably the  ones that will
have  to modify ours.   But  that is not the important  thing.   I  think that that  is
something that really bothered me when I started here 5 years ago because I saw two
or three studies that were identical.  EPA was doing one, NCI was doing one, NIOSH
was doing  one.   If the routes  were  different, let's say, and there were different
reasons for support of regulatory action based upon different routes and they were
not otherwise duplicated and they would in the end, support each  other if  they came
out with essentially the same or similar results, then I would see nothing wrong with
them.  But I agree with you that some of it was absolutely duplicative.

Dr.  Plotnick,  NIOSH:   Without being  critical of  EPA, Congress put all kinds of
responsibilities on one single agency which has to  act  as a  research  agency and a
regulatory agency, and  again, this  is the  only time that, without being facetious, I
will say that Congress in its  wisdom separated  research and regulation with NIOSH
and OSHA, and I  think it works  better.  You know, there is a lot  of cooperation but
there is  a total  separation, administratively.  Our  research  is not dictated by
somebody's idea of what standards  should be set.  We look at priorities on our own.
We obviously get suggestions from  the interagency group, but it does  not absolutely
dictate our work and our research, whereas, I think in other areas it does.

Dr. Marland, EPA:  We get a little schizophrenic in EPA, but I think that I would
rather have the relationship with the regulators that we have in R&D  than you folks
have in NIOSH.  So, I guess  that  each  one of  us likes  the activity in which he is
engaged.
                                  595

-------
Dr. Plotnick, NIOSH:  Were you making some comments earlier about the fact that
some things were difficult to support in rulemaking hearings because of the way that
the reserch is supported?

Dr. Marland, EPA:   No, because the basic scientific evidence that the regulation is
justified sometimes is sparse.

Dr. Plotnick, NIOSH: We can give you examples where there is more than sufficient
justification for regulation, and yet it is overturned.

Dr. Marland, EPA: Well, of course.  Dr. Bellin pointed that out effectively.  That is
because the system  on which  government operates and  in which the economic system
of the country operates is at play in whether or not a regulatory action can indeed
become effective. I am not talking about that.

Dr.  Bellin,  EPA:   What  I am talking  about is  that  combination of statement  of
problem  associated  with  a quantification of  that  problem  and the  science  that
permits you to do  that.  In  other  words, this man's decision to regulate or not  to
regulate has to be based on something that has to be  a scientific fact,  and I allege
that we are not doing a terribly good job at providing that scientific fact, whether it
is a new chemical  or an  old chemical or anything else.   Now,  however well you
document the facts, you then go into a series of procedures where scientists no longer
are  in command  of what happends,  but  as long  as we  are  in  command of what
happens, providing science and a recommendation to  the regulator, I think that we
are not terribly proud of  the extent to which  we have good science going into our
regulatory actions.  This is where I had hoped our collaborative program could bring a
stronger influence into an expression of confidence on the part of such as the NCI and
NIOSH, who are better known and established and have  reputations for good scientific
integrity. If our resaearch can be pointed in that direction, I think that we would get
a better utility out of a collaboration than if we all find that there are really some
worthwhile projects here that we ought to solve. What  I am trying to say is, focus on
those issues that make our regulatory life miserable, which is lack of good science,
and good scientists saying  that it is good science.

Dr. Plotnick, NIOSH: No, I disagree.  I am not sure  that it is a lack of good sciences
as much as it  is a lack of understanding on the part of the  scientist who  has  to
support this in an advocacy manner, and a scientist is not to be an  advocate.

Dr. Marland, EPA: Oh, that is very much a part of it.

Dr. Plotnick, NIOSH:  But I know it is not the quality of science or the lack of data
many times; it is professional judgement in interpretation of the data.  You turn it
over to somebody who is writing a regulation, who may or may not have a great deal
of scientific background, and then, all of a sudden,  he says a hearing is going to be
held on this day, we need you, and you, and you to support us at a  ruiemaking hearing.
I  can tell you that there is absolutely no briefing or  understanding of the rulemaking
on the record and  the  Administrative  Procedure Act  and  what  the purpose of the
advocates for the government and those opposing the government  is and how to come
across in the right  way without losing  your  perspective.   I think it is  more of that
than it is of quality of science.

Dr. Marland, EPA:  I had originated my comments on  the need to address science to
risk-making quantification, risk assessment quantification.



                                     596

-------
Dr. Plotnick, NIOSH:  I have less difficulty  than  you with indicating that a given
compound is very highly suspect as a potential human carcinogen and it should be
regulated as such once there has been a determination in several species, let us say,
that this compound is a carcinogen.   If we have supporting solid evidence in animals
and possibly, at least in our case, we have very often at least got some suggestive
epidemiologic evidence to show that it is producing an effect in man because man is
our own experimental animal in a lot of these places.

Dr. Marland, EPA:   I am talking about  benzene, I  am talking about arsenic, I am
talking about those real tough ones where we have what any  scientist will say  is a
good scientific basis for regulating, and yet  you  are not regulating, you are  not
regulating effectively because we do not have  credibility, we do not have the ability
to translate a series of mouse deaths into something that will convince a jury.  Now, I
allege that that is a scientific effort that calls for that.

Dr. Plotnick, NIOSH: Okey, but the benzene standard was not being challenged on
scientific basis as much as it was on a cost benefit-analysis.

Dr. Marland, EPA: Your description of risk did not warrant the  economic cost.

Dr. Plotnick, NIOSH: Within the terms of the Occupational Safety and Health Act,
there are differences, because it says, "so that no worker will suffer impaired health
or decreased life expectancy as a result of his work experience."  The Clean Air and
Clean Water Act  are not quite as specific,  and yet the court  decided that that
requires a cost-benefit analysis and that the Secretary had not done it. I still say it is
a  lack of  good  interface  between  the  lawyers on  the  line  and  the scientists in
presenting  the data, and I  am both.  I am a lawyer  as well.  I have been involved in
rulemaking hearings just  as a scientist, and the  other side is not that potent, even the
high priced Washington lawyers.  They are not  that good and effective.  It is just that
we are not that good in presenting our side because we do not have the training.

Dr. Leidel, OSHA: I will second Dr. Plotnick on that  one.

Dr. Plotnick, NIOSH: Dr.  Leidel and I were involved in  one case  where we ran  all
over the people on acrylonitryl.  We had solid evidence. We had all gone over how our
presentation would be made.  You know, we  were not faking  any data or anything
else.  It was our interpretation.  But we know what information we had.  We did not
have to look at our notes.  We could answer any questions and we could respond to
cross-examination.   If you are well  prepared ~ you know, most of the time I would
bet you that you bring  your experts in to testify the same day that the hearing is
being held. They do not know anything else  about it other than maybe they have
prepared a written  statement.  They have no  idea of what to  anticipate.  I do not
think that is science.

Dr. Leidel, OSHA: There has been poor communication in  the past.

Dr. Marland, EPA: Risk assessment is not science.

Dr. Plotnick, NIOSH:  I agree.

Dr. Marland, EPA:   Risk assessment is the application of science to the regulatory
process,  and I say  we  have provided  very,  very  poor scientific basis for  risk
assessment. It is not that as  a scientist I say that acrylonitryl is good or bad.  I do
not have to convince myself.   I think that there is a scientific  component to the

                                   597

-------
calculation which convinced even an attorney from Fargo, North Dakota, who does
not even have to be a high priced Washington lawyer, but you have to convince him
that the simple arithmetic  which takes your confidence  of  toxicity and translates
that into the risk of a  person in Fargo, North  Dakota, that   has to be convincing
evidence and we have not done that.  I allege that that is not a job for an attorney.
That is not a job for anyone but  a scientist, and this  is what I have  been trying to
drive at, that I have not seen the  dedication of a single research dollar from the
Cancer Institute or from  NIOSH  or  from  EPA,  a  single  research  dollar,  the
improvement of the technique of presenting scientific data in the form of hazard of
that environmental  element.  There is not any research being  done on  this, and I just
do not know why because it is a most important stumbling block, and that is the basis
for my complaint.

Dr. Rivkin, EPA: So you see the problem, then, perhaps, as presenting information,
packaging it?

Dr. Marland, EPA:  I think it is more fundamental than that. A scientist talking with
an  attorney is a difficult  thing  in  itself,  but  putting  it in  writing  is  even more
difficult.  Getting a convincing story presented in writing where you have calcula-
tions that are simple and clear and understandable by an intelligent person, this is not
being done.

Dr. Leidel, OSHA:   I will agree with you to a point, but to me there are two things we
have to talk about, one is the convincing argument, but in many cases I have found
nothing will convince industry. You can come up with a convincing model, but then
the second thing is,  is it a "defensible model," and that to me is the real tough  part in
the regulatory arena as  I have experienced it.  The problem is that you can go out on
any subject in this country, and, for a price, get some expert witness to come in and
say the government's model or the government  research was just full  of hot air, and
here are all  the weak points.  That is what  is so disturbing to me on  this  advocacy
thing.  People say it would sure be great if we did not have it, and OSHA has  caused
all  the problems, and, if we would just be nice  to industry, we would not have this.
Whether we like it  or not, we are never going to go back to the good old days,  as I
personally see it. We are just going  to have to recognize that it is here. It reminds
me of the situation in NIOSH with some of  our  epidemiologists  who  would get
involved in very controversial issues  and find their scientific credibility attacked and
all this sort of thing by  other, company-paid, scientists simply  because the companies
realize that the easiest way to cut  off new government regulation is to go  to the
roots of the  problem,  which  is  the research, and attack  the  credibility  of  the
research.  What happened to our scientists, of  course, is  they  were just not prepared
for this kind of  advocacy science.  The thing  that disturbs me, and I agree, we have
essentially nothing in the area of risk assessment, is how good are those estimates
and  what is  the credibility  of the  data that  gets into them.  I think  what has
happened, as I interpret  the position of the government regulators that say we  cannot
get into  the area because we cannot put a price on human life,  is maybe to avoid
getting dragged into that whole arena.

Dr. Marland, EPA:  The  cancer policy calls for  you to do it.

Dr. Bellin, EPA: You cannot avoid it. The executive order says every regulation we
have to do involves that kind of thing.

Dr. Marland, EPA:  Yes, absolutely, every one.



                                    598

-------
Dr. Leidel, OSHA:  Well, it will be interesting because I think that probably what I
would anticipate, then, is in the decades of the Eighties they would test chemicals
and the companies  would come in  and say how good are  your estimates and then
attack the credibility of the data that went into them. Once they force you into that
position, which they have, of course, then they will find something else to go at.

Dr. Plotnick, NIOSH: To give you an idea of this, the Office of the Solicitor in Dallas
sent me a copy of an opinion from an administrative law judge in an OSHA situation
where OSHA's expert witnesses were not nearly as good as the ones brought in by, in
this  case, Texaco.   The administrative law  judge,  in the  findings of fact,  said,
"Benzene is  a leukemogen; benzene is not a carcinogen."  OSHA was attcking it for
its carcinogenicity, but he made that finding  because they  did such a poor job of
presenting their side.

Dr. Cameron, NCI:  I think that is just something  the regulatory agencies have to
concede to industry that, by and large, we will never have on  the government side an
expert who  can  match  in depth the people brought in by  industry,  just because of
their long association with a  process or a compound or a class of compounds.  I am
sympathetic to the  regulatory agencies.  I  do not think they have the staff to afford
that specialization.  I have  seen it in the bioassay program when we brought in
industrial  people to help us  in the chemical selection process  and/or  the protocol
development.  It is  uncanny how much they know.   When you talked about the dye
people, that is a good example. You come up against a man who has spent 40 years
with one class of dyes.   There is just  nothing  you  can debate with him about.  He
knows every reference,  every test.  He knows every company in the country that is
using it, how they are using it, probably how much they are using, and they just tear
you up.  I  think,  if you get into an adverserial relationship with them, you are doomed
to failure.

Dr. Holland, EPA: Everything we regulate  is adversarial.  As I say, we are coming in
with pretty poor tools.  I want to improve our tools.

Dr. Cameron, NCI:  We  in NCI have been fairly fortunate.  They understand we are a
research operation.  They do not like the way some of our research findings are used,
but that is another experience.  We are going to  have to coordinate  that a little
better.  It is not  the subject here.

Dr. Plotnick, NIOSH: OSHA taps you all the time for expert witnesses.

Dr. Leidel, OSHA: I would disagree to the  extent that I do not think we always have
to concede.  I think there are areas where, obviously, industry is up on us.  I helped
out OSHA when I was at NIOSH as an expert  witness many times,  and I think the
cases the  government loses are just due to the tremendous workload that  they have.
The government  will not give the resources to the regulatory agency for the regional
solicitor's  offices.  Many times a  typical attorney will be  carrying 60 cases.  The
average attorney that I work with carries 60 cases.

Dr. Cameron, NCI:  No, I was not questioning their competency. I am talking about
the  aspect of diffusion.   You  cannot carry  60 case  briefs  at  one  time.   It is
impossible.

Dr. Leidel, OSHA:  Where the government is shorted is on time  and resources, not, I
do not think  necessarily, the technical expertise. That is the impression I have gotten.
Many times it is tough to get the right expert  witness, especially a fellow that works



                                     599

-------
as a consultant to companies, say, designing equipment.  He is not going to ruin his
professional career by coming in and  testifying for the government,  and that has
become more and more a problem.

Dr. Plotnick,  NIOSH:   All right, but  you agree that when  we support regulatory
agencies in rulemaking, we overwhelm the other side with people,  we take six or
eight people.   There  will be epidemiologists, statisticians, industrial hygienists,
toxicologists, chemists, etc.

Dr.  Leidel,  OSHA:   Well, the resources are available.   It  is just  a matter  of
coordinating the efforts in putting together a good case.  I think that is an area where
we can always improve in.

Dr. Cameron, NCI:  It seems that with some of the regulatory agencies there is an
eagerness to  move  ahead  very rapidly, perhaps too rapidly,  when looking at  other
agencies,  they may  be notorious for  missing milestones,  but  it does not seem  to
bother them, and probably they avoid a  lot of problems by doing that. They just take
their time.  The  major case  that comes to mind was Reserpine.  We agonized over
that and we gave out the  results, and the report was delayed 10 months by industry
for a pathology review. Finally, when it became a known carcinogen, it was given to
an FDA advisory board, and they said, "Thank you. We hear you, but we  are not going
to do anything about that  drug just yet."

Dr. Mar land,  EPA:  I guess maybe the courts have gotten tired  of trying to get that
agency to move. However, EPA is still  trying to react quickly.

Dr. Plotnick,  NIOSH:  EPA has a different constituency that is filing the actions
compared to  FDA.   The drug companies are  ultraconservative because they do not
want to get burnt the next time they submit something, either.  I presume, just about
with anybody, you remember  when  somebody has challenged you and  pushed you  a
little bit,  and New Drug Applications are slow enough as it is, and you don't want it to
take you an extra three years.

Dr. Marland,  EPA:  You can  figure  that 30 days after EPA has set a deadline there
will be a lawsuit instituted by one of the environmental  groups  requiring compliance
with a statute, and the judge, of course, is inclined to say, well, the statute says you
shall report, you have not  reported here, you are thirty days late.  You  tell me when
you will make  a report, and  EPA may  say it will be 5  years  from  now.  Forget it.
When?  Is it  60  days or  90  days?  Take your choice.  These  deadlines are  not
welcomed by  EPA staff, I can assure you.   They are thrust on  us and I  think
conscientious  people will do the  very best  they  can  to  do a reasonable job  of
complying with them. That is what EPA is struggling to do.

Dr. Plotnick, NIOSH: Remember, you  recommend the legislation that actually sets
those standards, too.

Dr. Marland, EPA: I do not want to allege  that that is a correct statement.

Dr. Plotnick, NIOSH:  No, I  think in some cases EPA  obviously recommends  what
should be  done and what time.

Dr. Marland,  EPA:   I do not  say that  that is correct, but what may be true is that
EPA does  not protest enough that they are wrong, that the deadlines are wrong.
                                   GOO

-------
Dr. Plotnick, NIOSH:  I did not say that  EPA invites  the short time.  They should
make a rational projection of how long it is going to  take and indicate during the
committee hearings the period of time; give  us a little  more legislative leeway. You
know, just a little foresight.

Dr. Marland, EPA: I agree.

Dr. Cameron,  NCI:  You did not solve my problem. Where are we going next year
with projects.  I think I got a glimmering of one, that structure activity tree.

Dr. Marland, EPA:  That is very, very close to what I see is badly needed, and that is
to give the regulatory agencies some help.

Dr. Cooper, NCI: It seems to me that we are not addressing in this meeting all of the
interagency agreements between  EPA, NIOSH, AND NCI.  That suggests to me that
when we say NCI/EPA/NIOSH collaborative program, that "collaborative" must mean
something other than just things that are of mutual interest.   That  is what I was
trying to get at when the first question came up about why don't we do more in terms
of physical, primary prevention kinds of activities. I felt that  NCI does not  have  a
great deal of expertise in this area.  Perhaps it would be useful  to try and understand
what we mean by a collaborative program. It does not  mean everything we do under
an interagency agreement.

Mr. Harris, NIOSH: NCI has an awareness of problems, and together with the people
who know how to solve them is  where the collaboration begins. And together with
the problem solvers, such  as the engineers and other technologists, the problems can
be solved.

Dr. Cooper, NCI: You are describing, it seems to me, an interagency agreement.

Mr. Harris, NIOSH: A collaborative effort.

Dr. Cooper, NCI:   But we are not talking about all interagency agreements; all the
things in which two agencies have mutual interest.  We are now talking about a subset
which we have defined as a  collaborative program. I am not  at all sure what that
word "collaborative" is meant to imply, and I feel we ought to think about that.

Dr. Cameron, NCI:  I think I would define  it  as meaning a blending of expertise.

Dr. Cooper, NCI: I think that is the case, too, but that  implies that there ought to be
expertise on both sides in areas undertaken under this collaborative program.

Mr. Harris, NIOSH:  Let   me give you an example. The synthetic fuel technologies
are in the early stages of development.  If the engineers  work together with the
health professionals and the biologists, we may  be able to understand how to solve
many of the problems. This is what we are talking about here.  It is an intermeshing
of different types of expertise to work together in a common bond.

Dr. Cameron, NCI: You are giving me a hypothetical case when you are talking about
DOE  and synthetic fuels, and so on.  It  is  obvious there is  going to be a lot of
toxicology done on the products, the by-products, the combination.   There  will be
bioassays attempted by DOE. They should not place contract one for a bioassay until
they have consulted with our group, or NCI, specifically in bioassay. We have  had our
experiences and  our problems.   We  know  all the mistakes you can make  with  a


                                     60J

-------
bioassay.  We should be asked to freely offer  our advice.  Dr. Cooper  and I were
offended,  candidly, at that monkey experimental design. NIOSH should not fund  any
bioassay until  there is  an input.  Our people should not attempt any epidemiological
studies in  the work place until they have talked to you people at NIOSH.

Dr. Cooper,  NCI: That is a fiat.

Dr. Cameron, NCI: That is just the way it has got to be.

Dr. Cooper, NCI:  Maybe that is a good working definition of collaborative program,
— one in which neither of the agencies has sufficient expertise  to undertake the task
alone.  In  that case we ought to consider as candidates for the collaborative program
only those activities in which the expertise of both the agencies must come together
to do the  job.  That is  fair enough, and I am not holding a brief for any definition.  I
am just suggesting that we ought to agree on what is suitable  for this collaborative
program.

Dr. Leidel,  OSHA:  I am just wondering, I kind  of balk at  the idea of bureaucracies
and review groups, but it reminds me of a problem  we had in NIOSH that in the early
seventies  we had studies done with, say, poor  statistical protocol, both  in experi-
mental  design  and then the techniques used  to  analyze  the  data  and  draw  the
conclusions.  Well, out of  that, some of our people got sufficiently embarrassed so
that they set up within the Statistical  Services  Branch  what they  called  SPRG,
Statistical Program Review Group, and basically it comes  down, you  know, after all
these fancy  acronyms, that all interagency agreements and research contracts and in-
house research programs have  to go through this group  and be reviewed by some
statisticians. Theoretically, that could occur here. Anything that involved epidemio-
logy could  go  through  an industrial hygiene review group that  we  could provide
scientists  from NIOSH on, or anything that involved bioassays, we could set up a
bioassay group and that sort of thing.  That is just tentative thinking.

Mr. Harris,  NIOSH: We could maybe start off the meeting tomorrow with some of
these ideas.
                                     602

-------
FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

    PROGRESS ON JOINT ENVIRONMENTAL AND

        OCCUPATIONAL CANCER STUDIES
             Thursday Morning, May 8
            CONCURRENT SESSION I:
          RADIATION CARCINOGENESIS
             SESSION CHAIRPERSON


              Dr. Wayne Galbraith
         Environmental Protection Agency

-------
        Laboratory and Field Trial Evaluations of the Cost/Effectiveness of
                 Two Types of Personal Ultraviolet B Dosimeters
     Dr. Arthur 3. Sober, Massachusetts General Hospital
     Dr. George Goldsmith, Boston College

Project Officers:
     Dr. Thomas W. Orme, NCI
     Dr. Herbert Wiser, EPA

I  am  glad  to  see Dr. Joseph Scotto in the  audience  because the first  talk I am
presenting  may be of interest to him. I am  presenting this material for Dr. Herbert
Wiser, the  EPA Project Officer. The subject matter deals with the development of a
personal UV dosimeter which may be useful ultimately in epidemiology studies.

The  key  personnel  in  this project who  are doing  both the  thinking and  the
experimenting are Dr. Arthur 3. Sober in the Department of Dermatology,  Massachu-
setts  General  Hospital, and Dr. George Goldsmith in the Department  of  Physics at
Boston College.   The project  is  supported  by an  EPA  grant  funded  through  the
NCI/EPA  Collaborative  Program.   The material I will present  has been abstracted
from progress  reports submitted.


A major  source  of uncertainty in the correlation of skin cancer incidence  and
exposure to solar ultraviolet radiation is the difficulty of relating the actual exposure
received  by individuals to the measured incident solar radiation at  given geographical
locations.   Geographically  fixed  meters  record  maximal  incident  radiation  and
seasonal  variation.   Although  data obtained  from fixed meters have contributed
significantly to our understanding  of the relationship between skin cancer incidence
and UV, we still lack data about variation in UV exposure related  to lifestyle and, in
particular, about the relationship  of pulses  of exposure,  which an individual might
receive in the  summer during a trip to the beach, to overall individual exposure.

Two types of UV dosimeters are being developed:  an electronic meter based upon the
prinicples of the Robertson-Berger  meter  and a  film badge based on  photographic
film or photochromic materials. Such instruments have many potential uses.  A few
of the  questions which these dosimeters could be used to answer are the following:

      1.    Which activities are associated with greatest UVB doses?
     Conceivably some activities may  be more dangerous per unit time than others.
     For example, beach activites involving UVB  intensified by  reflection  from sand
     may  result  in  much  greater exposure  doses  of  UVB than  golfing  on grass.
     Measurements  with a personal  dosimeter could lead to a rational  evaluation of
     the hazards associated with various types of recreational activity.

     2.  How  much  acute UVB in comparison to total annual dose does an individual
     receive on a vacation to a sunny climate? Conceivably a substantial fraction of
     the total annual dose may be experienced in a relatively short period of time.
     A vacation UVB to annual UVB exposure  ratio could be calculated.
                                       604

-------
     3.   How  much UVB penetrates clothing?  From earlier  studies a  tentative
     conclusion has been drawn that about  20 percent of  incident UVB penetrates
     white clothing.  Since the most frequent site of occurrence for melanoma in the
     males is the back, the use of UVB meters beneath clothing could help quantify
     the  dose  through clothing  and  evaluate a potential  solar role for  melanoma
     arising in  these so-called covered sites.

     4.  What are the dose differentials by anatomic site?  For example, basal cell
     carcinomas are frequent in sun-exposed areas; yet, the vast majority are on the
     face and  neck rather than hands.  Do these areas differ intrinsically in their
     response to UVB? Or does the dose differ at these sites?

     5. Which  occupations are associated with highest UVB exposure?

Clearly, the ability to quantitate UVB exposure in the above examples would allow
classification of exposure by type of activity, recreational and/or occupational, which
in turn, would lead to some quantitative estimate of environmental hazard for each
type of exposure.

Before any such device can be employed for extensive short or
long term studies, each type of device must be carefully studied to determine that it
actually measures the desired UVB wavelengths under a wide range of situations. The
following must be determined:

     — UVB spectral sensitivity.
     — quantitative abilities under high and low flux rates.
     — positional sensitivity.
     — temperature and climate sensitivity.
     — shock resistance.
     — reproducibility.
     — response to non-UVB wavelengths.
     — cost per unit.

     Dr.  Sober has  received  prototype  personal  UV  dosimeters  from the  Boston
College group and is  beginning field evaluations to determine whether they will  be
useful  in  large scale personnel studies.  Some judgment has to be  made  as to the
feasibility of producing large quantities of personal dosimeters and conducting large
scale clinical trials. Two dosimeters are now ready for clinical evaluation.  One is an
electronic monitor  with  a spectral response  similar to the  Robertson-Berger meter,
employing a magnesium-tungstenate fluor as a sensor.  This has been developed by
Dr.  Goldsmith   and Dr.   Davidson  of Photometries, Incorporated in  Lexington,
Massachusetts.

   The  second device is  a UV  sensitive film badge similar to  the polysulfone film
badges which have been previously employed.

     Trials will be of two types:  laboratory evaluations and field evaluations.  All
laboratory testing will be completed on both types of units before field trials will be
undertaken.   Laboratory evaluation will consist of the following testing procedures.
For each type of dosimeter, spectral response, spatial response, reciprocity with the
badges and linearity with the electronic system, dynamic range, and variation  due to
temperature and stability will be determined.
                                       605

-------
     The field trials will consist of the following testing procedures to be conducted
during two different seasons.  In early March, the low intensity solar radiation series
of experiments was to have begun.  This summer Dr. Sober's group will conduct a set
of experiments under conditions in Boston which supposedly will mimic high intensity
solar  radiation. Phase  I  will consist of  static  outdoor testing  for reproducibility,
specificity, climate, overload.  Phase II  will consist of dynamic outdoor  testing to
determine the effect  of motion, inclination, and responses to graded increases in
solar exposure, dynamic range, and temperature.

     In  the  NCI/EPA  annual report,  which has  just  come  out (yellow book) the
physics of these devices is described more fully.

     At  this  time,  it  might  be  worthwhile  to  mention some  of  the  expected
advantages and disadvantages of both types  of dosimeters. The electronic meter will
probably be quantitatively more precise,  be reusable and will  have a better spectral
match to the DNA action spectrum than the film badges. It will also have a response
that is  roughly similar to the  Robertson-Berger  meter so  that  comparison with
geographical data already available will be  facilitated.  Its disadvantages are higher
costs,  which will limit the  number  of devices available, and a slight deviation from
the DNA action spectrum.

     The film badges, for  their part, are  inexpensive, easily produced and easy to
quantify with a densitometer. There are problems with reciprocity in the film badges
and there is a major deviation from the DNA action spectrum.

     As I said, this project is at an early stage with respect to the evaluation of the
devices.   Dr. Sober sent  me  a photograph  of the electronic meter (Figure 1).  It is
about half the size of a package of cigarettes. It looks like a very convenient device.
I am still not convinced that it is going to be the type of device you can ask people to
carry around for long periods of time, but I think there are problems  for  which Dr.
Sober will be able to enlist the help of Bostonians, student population probably, and to
look at the specifics of how useful these  devices can be. I expect that by the end of
the year Dr.  Sober will provide  some useful data on  the practicality  of using these
meters in any large population project where personal monitoring would be preferred
to geographical monitoring.
                                          60b

-------
Fjoure 1.
         607

-------
                                    DISCUSSION
     DR SCOTTO:   You mentioned that the personal  dosimeters  are going
to be measured simultaneously with the global  measurements  in a particular
location.   Is that true?  In other words,  are  they also  going to have
Robertson-Berger meters measuring the global response in Boston?  Because
I did not know that there was a meter there.

     DR. ORME:  That is not what I implied.  I implied that  the spectrum
of the electronic dosimeter is specifically  designed  to  match that
of the RB meter.  I do not know whether they are  going to set up an
RB meter in Boston.  That might be a good  idea.

     DR. SCOTTO:  My suggestion would be to  either set one  up there
or go to a place where we do have the other  measurements.   At the same
time of the day when you are measuring the individual, you  are getting
the global measurement.  In this way, I would  know how to evaluate
the global responses that we have been getting.

     DR. ORME:  That is a very good suggestion.   I think Dr.  Sober
is committed to evaluate these meters in the Boston area.

     DR. SCOTTO:  Then you should get one  of the  other meters.   Because
how are you going to quantify it?  As you  said, you know how much is
reaching the earth's surface, but you do not know how much  is reaching
the individual's skin.

     I do not think you mentioned the price, although you said the
electronic meter was more expensive than the film badge.  Also what
is the size of the electronic device?

     DR. ORME:  I do not have specific numbers on price.  Figure! .
gives the size of the device in relationship to a package of cigarettes.

     DR. SCOTTO:  When I was talking with  Davidson, who  is  developing
this, I thought the meter was to be no larger  than a  25£ coin as originally
planned.  So, it got bigger and it was in  the  neighborhood  of $500
or $600 back then.

     DR. ORME:  I do not know how much it  will cost.

     DR. SCOTTO:  So the differential in cost  in  terms of $500 or $600
for one of these units as opposed to the film  badge is to be considered.

     DR. ORME:  Yes, there is appreciable  difference  expected.

     DR. KELSEY:  How are they going to wear this thing? It is like
a cigarette pack.  They are going to wear  it at different times of
the year and when they go to the beach? How are  they going  to wear
it?

     DR. SCOTTO:  That is part of the problem, if I might answer that.
I was going to mention that they wanted us to  be  available  to field
test these in an epidemiological way, as we  do the incidence surveys.
All of these things have to be worked out.  Should you put  it on the
shoulder?  Should you wear it on your lapel?   What is convenient and


                                          608

-------
what is practical?  This all  goes into the problem of  which  device
to select.  Nobody has really answered that.

     DR. ORME:   I was thinking of one particular  problem  that  you alluded
to.  To indicate the relative exposure that you  get on the forehead
and on the back of the neck as opposed to the hand one must  actually
place the meter on those positions.   I think Dr.  Sober is going  to
have to hire people to run around with the meters on to measure  the
exposure received by skin.  How to relate this type of exposure  to
that received by cancer patients is  an important  question.   It is not
going to be a direct correlation, but I think that it  is  possible to
enlist people to obtain exposure data when they  are playing  tennis,
lying on the beach and so forth.  That does not  have to be done  over
long periods of time.
                                           609

-------
                               Joseph Scotto
                     Field Studies and Statistics Branch
                          National Cancer Institute
                     Skin Cancer Epidemiological  Studies
     I will  begin this presentation by explaining that both projects listed in
the program refer to the same basic mission - that is, to provide epidemiologic
information relative to the potential human health effects of stratospheric
ozone depletion.   The NCI/EPA program provided support in two waves.  The first
was for a small amount of funds ($60,000) to supplement our initial, short term
project entitled, Special Skin Cancer Epidemiologic Studies.   The second, also
a small amount ($200,000) was to initialize the long-term effort, the National
Nonmelanoma Skin Cancer Study.
     At the opening session, Dr. Kraybill reviewed the brief history of the
NCI/EPA program.   I believe it was around 1978 when funding was  actually
provided under this cooperative effort.   But just before this program
materialized the EPA and NCI were already engaged in an interagency collaborative
agreement on skin cancer epidemiology.  The NCI was asked to utilize its ongoing
Surveillance, Epidemiology and End Results Program, usually referred to as the
SEER Program, to obtain information, as  soon as possible, which  would reduce
the degree of uncertainty in the dose-response estimates of UV related skin
cancer in our country.  It was recognized that the SEER locations were not
necessarily the best or only places where these studies should be done, and
that to monitor the trends in skin cancer incidence as well as ozone depletion,
a longer term project was needed.  In addition NCI was asked to prepare for
field studies which would provide new measurements of solar ratiation exposure
utilizing personal  dosimeters, which were currently being developed by the EPA.
The project presently labeled the "National Nonmelanoma Skin Cancer Study", is
essentially an extension of the Special  SEER study.  To start us off on this

                                      610

-------
          long-term effort, funding was provided to initiate studies in two new locations,
          San Diego, California and the combined states of New Hampshire-Vermont.  The
          data collection phase in San Diego is just being completed, and the
          New Hampshire-Vermont study has just gotten underway this winter.  This
          presentation will now deal with the progress, early findings and first analysis
          of the current surveys just being completed.
StLde. 1         The first slide shows the locations where incidence data and UV-B
          measurements were obtained.  Before looking at the preliminary report, a brief
          review of the recent history of events leading to the urgent need for skin
          cancer data may put this project into proper perspective.  As an adjunct to
          NCI's Third National Cancer Survey, 1969-1971, which provided incidence data on all
          cancers, except nonmelanoma skin cancer, a special survey of skin cancer was
          conducted during the later part of 1971 and the early part of 1972.  Four
          locations were able to participate in this study:  Dallas-Ft. Worth,
          San Francisco-Oakland, Iowa, and Minneapolis-St. Paul.  In 1973 while we were
          editing and reviewing the results from this study, the Department of
          Transportation was becoming quite concerned about the potential danger to the
          protective stratospheric ozone layer which may result from the excessive use
          of supersonic aircraft (the SST's).  The DOT developed a multifaceted research
          program called the Climatic Impact Assessment Program (CIAP) to study the
          effects of the nitrogen oxides which were being emitted as exhaust gases from
          the SST's.  Ozone depletion results in increases of solar ultraviolet radiation
          reaching the earth's surface, and consequently potentially greater risk for
          skin cancer among humans.  In addition to the incidence data for these four
          locations, NCI collected and reported to the CIAP Program measurements of
          solar ultraviolet radiation reaching the earth's surface at these and other
          locations in the United States.  By 1975, other man-made, pollutants,
                                                 611

-------
chlorofluoromethane gases (CFM's) which we know as "freons11 used in aerosol
spray cans and as refrigerants in air conditioners were discovered to be
potentially much more devastating to the ozone layer than the nitrogen oxides.
Soon afterward federal  regulatory agencies were in great need of information
on both the biological  effects to plants and animals as well  as the human
health effects of ozone depletion.  The CIAP Program had only begun to scratch
the surface.
     The epidemiologic information which the NCI provided from its early
surveys supported the hypothesis that UV may cause skin cancer and that
greater amounts of UV exposure which result from ozone depletion may lead to
increased risk to skin cancer.  However, most researchers agreed that much
more information was needed.  Not only more geographic locations but also
more epidemiologic information on host factors (such as skin  color and
ethnicity) and environmental factors (such as lifestyle and outdoor exposure
habits) would be needed to estimate the potential hazards of increased doses
of solar ultraviolet radiation with greater precision.  In the mid 1970's it
was estimated that an eventual ozone depletion of 7 percent may be expected  to
occur sometime in the 21st century.  Today, National Academy of Science sources
indicate that a 16.5 percent ozone depletion may be expected  from the continued
release of chlorofluoromethanes at 1977 levels.  It was also  noted that a one
percent decrease in ozone translates to a two percent, or a twofold increase,
in solar ultraviolet radiation reaching the earth's surface.   This is usually
denoted as the physical amplification factor.  And this factor may be greater
than 2 for relative decreases in ozone greater than 10 percent.
     Turning back to the map which displays the locations where UV and
incidence data are available, in addition to the locations depicted on this  map,
we will include New Hampshire/Vermont, representing the Northeast, and San Diego,
California, representing the Southwest Pacific Coast.
                                       n 12

-------
Slide,  2        The next slide shows a schematic diagram of the electromagnetic spectrum.
         We are most concerned with the invisible solar ultraviolet, called UV-B.
         Stratospheric ozone shields the earth from high intensity wavelengths shorter
         than 290 nm.  However, UV-B between 290 nm and 320 nm, which does reach the
         earth's surface  in small amounts,  fs known to cause sktn cancer tn experimental animals
         and erythema, or sunburn, in man and is suspected of causing skin cancer in man.
Stide.  3        Measurements of the amount of UV-B reaching the earth's surface are
         provided by Robertson-Berger meters.  A count of 400 to 440 units of UV-B
         will produce a reddening of the skin in a typical, untanned Caucasian.  The
         next slide shows that, in general, as latitude decreases, UV-B increases.
Slide.  4        The next slide shows the added SEER locations where new estimates of
         annual amounts of UV-B were obtained.  The open circles represent the original
         10  locations obtained in 1974.  The new 1977-78 UV locations are depicted by
         the asterisk (*) in the graph.  It can be seen that the relationship between
         UV and latitude  remains, as we have seen before.  In addition to latitude
         dependence, we should consider altitude and sky cover as well.  That is why
         some of the locations may not fall in line.
              We will now turn to the epidemiological information on our recently
         collected studies dealing with basal cell and squamous cell skin cancers from
         these eight locations.  The eight  locations are in the order of increasing
         latitudes:  New  Orleans; Atlanta;  Albuquerque, New Mexico; San Francisco/Oakland;
         Salt Lake City,  Utah; Detroit; Minneapolis-St. Paul; and Seattle.
Slide.  5        This slide  shows the dramatic difference in the latitude dependence of skin
         cancer morbidity compared to all other cancers.  Incidence rates for the White  race
         only are given,  since this disease is rare in other race groups.  The broken
         line indicates a limited amount of variability in cancer risk by geographic
         location for "all other cancers" combined.  The solid line shows that as
         latitude decreases, skin cancer incidence increases.

                                                   615

-------
Stide. 6
  7-B
Stidu
  70-77
     The next slide ranks the age-adjusted skin cancer incidence rates by sex
and geographic area according to recent estimates of the annual  amounts of UV-B
reaching the specified locations.  In Utah the Robertson-Berger meter was placed
at Salt Lake City, and in New Mexico it was placed at Albuquerque.   The Salt
Lake City rates appear to be comparable to those for Utah State as  a whole.
In Albuquerque an additional adjustment was made for ethnic group.   The "Anglo"
rates for Albuquerque refer to Caucasians other than Latin.  It should be noted
that Albuquerque, while not the southernmost point in the survey, had the
highest UV-B index.  It is clear that the risk for males is approximately twice
that for females.  Utilizing these new rates we now estimate that as many as
400,000 Caucasians will develop new skin cancers each year in the United States.
Compared with data from the earlier NCI survey, incidence rates  appear to have
increased by 15 to 20 percent over a six year period.
     The next two slides show the age-specific incidence rates by geographic
area for males and females.  In the southern locales, the male rates appear  to
diverge from the female rates and show increased risk as early as age 30 (see
Albuquerque, Anglo).  In the Northern and Central  regions (next  slide) the male
rates begin to depart from the female rates by age 45.   This difference in
age-specific risk by geographic area should be remembered when applying
mathematical models to these data.
     The next slide shows age-specific incidence by grouped anatomical site,
for all geographic areas combined.  Basal cell  and squamous cell  cancers occur
most frequently on the face, head and neck.  Exposed areas of the body account
for about 80 percent of the malignant lesions for both  men and women.  The
incidence for lower extremities among females is equal  to or greater than that
observed for males.
     The next two slides summarize the most important findings to date.  All
available information on the annual UV-B levels, and the age-adjusted skin
                                                       614

-------
cancer incidence rates are graphically displayed.   The solid  squares  represent
the results from the most recent 8-area survey and the empty  squares  represent
results from the earlier 4-area survey.  Two locations, Minneapolis-St.  Paul
and San Francisco-Oakland, were involved in  both surveys.   The  UV-B indices
for the 10 locations vary from a low of 101  for Seattle to a  high  of  197  for
Albuquerque.  The incidence rates for males  vary from a low of  172 for Detroit
to a high of 752 for Albuquerque Anglos.
     An exponential, or log-linear model, was applied to the  data  to  estimate
the change in skin cancer risk due to small  relative  increases  in  ultraviolet
radiation.  In locales of relatively low insolation a 1  percent increase  in
UV-B (290nm-320nm) may result in about 1^ percent increase in skin cancer
incidence (e.g., Seattle, White males); while in locales of relatively high
insolation levels, skin cancer incidence may be expected to increase  by more
than 2 percent if UV-B levels are increased  by 1  percent (e.g., Albuquerque,
Anglo males).  Estimates for females were somewhat (next slide) lower than
those for males.  At this juncture the results appear to be consistent with
earlier NCI estimates of the biological amplification factor  (roughly 2 to 1).
The degree of uncertainty in the estimates,  however,  has substantially been
reduced.  Should these relationships hold, a one percent decrease  in  ozone may
result in an eventual four percent increase  in skin cancer incidence.  A
preliminary report on the nonmelanoma studies will  be available for distribution,
perhaps by next week.  Please leave your name and address  if  you would like a copy
Interview Studies
     In addition to the incidence studies, we conducted telephone  interview
surveys designed to obtain information on host factors and environmental  factors
which may be associated with skin cancer incidence.  The information  obtained
from these studies will soon be incorporated into the incidence and UV exposure
analyses.  This should further decrease the  degree of uncertainty  in  the
dose/response estimates.
                                          615

-------
SLida 72
SLLde.  73
Slidu
  74-22
     The next slide shows the instrument which was used.   Individuals received

a copy of the questionnaire in the mail, prior to responding to the telephone

interview.  In the patient sample, 500 patients were computer-selected for

interview.  Before any contact was made, the dermatologist or attending

physician granted permission to make contact with the patient.  The patient's

free and informed consent was obtained prior to conducting the interview.   In

the general population sample, at least 500 Caucasian households in each

location were selected through the telephone random-digit-dialing technique.

Adults 20 years of age and over were selected for interviews in these households.

The instrument was mailed to cooperating households and again, free and

informed consent was obtained prior to conducting the telephone interview.

     The next slide shows the number of individuals responding to the telephone

interview.  The overall  general population response rate  was between 75 and 80

percent.  The patient response rates vary widely among geographic areas.   In

fact, the success of the patient surveys in San Francisco and New Orleans  remain

questionable.  In New Orleans, physician cooperation was  the big problem,  only

a 50 percent response rate was obtained.  It should be mentioned, however,

that once contact was made with the patient, the response rate was well over

90 percent.  As you can  see, there are over 10,000 interviews to evaluate.

     The next series of slides will highlight preliminary findings for several

host and environmental factors which have historically been  associated with skin

cancer morbidity.  This  slide (14) shows the proportions  of  respondents who

claimed to have "fair" complexions.  As expected, the patient group had a

greater proportion of "fair complexioned" individuals than the general  population

group.  Also, women apparently admitted to be more "fair" than men.  We were

concerned that this type of question may produce only a subjective response,

and we therefore attempted to provide a more objective measure of determining
                                                    eie

-------
          skin color by developing  a  skin  complexion  chart, which you noticed on the
          bottom of the instrument.
S-tt.de 75       The next slide shows the  proportions of  respondents who matched the
          inside of their upper arms  to  the  lighter colored skin swatches, color
          numbers 7 through  10.  It is the inside  of  the  upper  arm which  is  usually
          untanned.  Here again, it appears  that the  women may  indeed be  the fairer
          sex.  At each location, the female proportion with  light skin matches was
          greater than the male proportion.
S£tde6         The next two  slides  show  the  response  to questions on eye  color and
 16-17
          hair color.  Blue  eyes and  blond or red  hair  predominate among  the patient
          groups for both sexes.
S&cdei         The next three slides  deal  with ancestry or ethnic categories.  More
 n-u
          Scottish (18) and  Irish (19) people are  found among the patient groups, as
          expected.  Responses to Scandinavian ancestry were  somewhat surprising.  In
          Minneapolis-St. Paul, where the  concentration of Scandinavian decents is high,
          the proportions of Scandinavians were lower in  the  patient group for both sexes.
Slide. 21        The^next slide shows  the proportions  of individuals who held outdoor
          jobs.  The differences in proportions are clearly in  the expected  direction,
          except for New Orleans females.
S-ti.de 22       Finally, the  last slide shows the proportions  of individuals  who are able
          to develop a deep  tan. There  is no question  that the patient group cannot tan
          as easily as the general  population group.
               To summarize  our progress to  date,  we  are  winding down on  the data collection
          phases of this project and  we  are  beginning to  get  into the thick  of the analyses.
          We plan to provide two monographs  displaying  complete details and  descriptions
          of the data probably by the end  of this  fiscal  year.   It has  taken us a great
          deal of time to edit the  information which  we ha.ve  received.  Unlike some of the
          other studies that go on  in the  National Cancer Institute, we had  the

                                                     617

-------
responsibility for all  of the editing procedures  and  developing  the  programs  for
the analysis, doing the resolution checks and actually working with  the  physical
documents and making all  kinds of comparisons by  hand as  well as by  computer.
It is very time consuming and we are glad to be getting out of this  phase  and
getting into the thick  of the analysis.
     With respect to future research, more information is needed on  personal
dosimetry measurements, as Dr. Orme has  already mentioned.   But  perhaps  even
more importantly, we should look to epidemiologic studies of skin melanoma.
Most of the general relationships relative to UV-B exposure and  skin cancer
are also found for skin melanoma.  But skin melanoma  is a much more  serious
skin malignancy than the nonmelanomas.  The nonmelanomas  are 95  to 99 percent
curable, whereas the malignant melanomas have a survival  rate equal  to that
which is found for breast cancer (about  70%).  The process  by which  UV may
be involved in either the induction or promotion  of skin  melanoma is complex.
Some of the reasons, which Dr. Orme also mentioned, are the distribution of the
anatomical sites on skin melanoma patients, the trunk in  the males,  for
example.  We strongly suggest that if this long term  effort is to continue,
that we get the skin melanoma studies under way very  soon.
                                            616

-------
                                                                                                                 o.
                                                                                                                 (0
                                           SKtN CANCER-ULTRAVIOLET
                                      MEASUREMENT LOCATIONS IN THE U.S.
                                                      MINNEAPOLIS-ST. PAUL
                                                               3
                                                               MINNEAPOLIS
                                               DALLAS-FT. WORTH
                                           FT. WORTH /C
                                                                                  ATLANTA
                                                                      E5IRMINGHAM  X
                                                                                TALLAHASSEE
 CpGEND
g^ri TNCS (NCNMELANOIViA) LOCATIONS
r~7T7 TNCS (MELANOMA) LOCATIONS
r—i SEER LOCATIONS
CU3 TNCS AND SEER
 O  ULTIuVviOLtT METfcf: LOCATIONS

-------
                                           THE ELECTROMAGNETIC SPECTRUM
      X-rays and gamma rays
{, llltravjnlpt ,
Extreme i Far • Middle 1 Near
XUV ' FUV ' MUV J NUV


UV-C

,

CO
>
J
,


UV-A


- V/irihlo *.



"5
^


(D
_D


II
> (U
c / S O1
£ -5 2
0 > 0
II.

•a
0)
CC

,
,^r 	 , 	 , 	 ,._



Infrared

i i i
CD
ro
O
100        200         300        400        500        600

                                    Wavelength in Nanometers
                                                                                700
800
900
                                                                                                               1,000

-------
Slide 3
               3.5
               2.5
            o
            u
               1.5
            1   1
                 15
                           FIGURE 2.1. ANNUAL UV COUNT BY LATITUDE
                             Maura Loa
                                                      El Paso
                                                             Albuquerque
                                          Tallahassee 0
                                                        I Fort Worth
                                     • Oakland





                                     Des Moines




                                 Philadelphia O
                                                  I
                                           I
20
25         30         35        40



     DEGREES NORTH LATITUDE
                                              Bismarck


                                          e   *

                                          Minneapolis
45
50

-------
JOU

200
o
O5 °
ro *>
TO O
X
m
100
1
HNNUHL UVH (.(JUKI BY LS1I J IUUC.
0
k
0
0
*
0 o*
0 g
h 0
0 * o °,
• ft I 1 1 1 1
15 20 25 30 35 40 45 50
DECREES NORTH LATITUDE
-O



JL

-------
Slide  5
                              AGE ADJUSTED INCIDENCE (U.S. 1970)

                                 AMONG WHITES BY LATITUDE
        500
        400
                                                                         D
     O
     0.
     O
     a.
     §
     T-

     GC
     UJ
     D.

     UJ
        300
200
         150
         100
                 30
                     34
      a	
                         38          42

                        DEGREES NORTH

NONMELANOMA SKIN CANCE.l (1977-78)


ALL OTHER MALIGNANCY (1973-76)
                                                                 46
                                                                    50
                                            623

-------
                   NOIMMELANOMA SKIN CANCER INCIDENCE AMONG WHITES BY

                                GEOGRAPHIC REGION AND SEX, 1977-78
     tr

     <
     LU
     >
     -^
     z
     o
     Q.

     O
     a.
     8
     <
     DC
     LU
     a

     o

05   ?


£   *
     a
700
600
         500
         400
        300
     Q

     "J  200

     CO
     D
     •>
     O
        100
      II Male

      B Female
          a
          ai
          CO
o
a_
o
«•*
o>
O
9)
*••
a
4-1
to




S
^-


O

a>
                                              S
o
u
to
'5


I?
ul jo
                                          (/) O
a>
c.
ID
_£


6

5
                                                                               o> o
                                                                               3 U
                                                                  3 U>
                                                                  V) a
                                                                          > t:   cr— o
                                                                          Is   = 5-5
                                                                          z «   £ o c

-------
Cfc
ro
en
                            AGE-SPECIFIC NONMELANOMA SKIN CANCER INCIDENCE AMONG WHITES BY

                                                REGIONS OF THE UiMiTED STATES
   8  '«>

   r

   K


     10
        ATLANTA
                                     SOUTHERN REGION (LATITUDES 30-35 DEGREES NORTH)


                                                                                       10.000 p-
                                 _	MALES
         OS  15-24 25-34 3544 45-54 55-64 (5-74 75-M 85*


                     AGE GROUP
                                                 : NEfJ ORLEANS
                                                          _J	I	I	I    I   1
<1G 15-24 25-34 35-44 45-54 5544 «S-74 75-M

            AGE Gq
-------
                          AGE-SPECIFIC NONMELANOMA SKIN CANCER INCIDENCE AMONG WHITES BY
                                               REGIONS OF THE UNITED STATES
10.000 r
     SEATTLE
                                    NORTHERN REGION (LATITUDES 40-50 DEGREES NORTH)
         1H4 J634 3544 4554 &S 84 86 74
                  AGE GROUP
                                              : MINNEAPOLIS-ST. PAUL
115 1574 2534 35-44 45 54 5564  9574 75 84 85.
            AGE GROUP
                                                                                        DETROIT
<15 1524 2534 35-44 4554 5564 6574  75W  85 «
            AGE GROUP
                                                                                                                               —
                                                                                                                               .i"
  ro
  05
                                       MID REGION (LATITUDES 35-40 DEGREES NORTH)
                                                                10.000,.	
                          SALT LAKE CITY
                           J	L.
                                          i	1	1	1	L.
                           <15  1U-J4  K-14 15-44 4S-64 5S-64 8S-74 Tt-U IS*
                                       AOE OHOUP
                                                                     SAN FRANCISCO-OAKLAND
                      <1B  1B-24 2S-M J6-44 4S-&4 65-»4 66-74 75-§4  «&.
                                   AGE OHOUP

-------
Slide  9
TRENDS IN ANNUAL AGE SPECIFIC. NONMELANOMA
    S.ilN CANCER RATES AMONG WHITE MALES
                       sana
                       1000
                                                                                             FACE, HEAD
                                                                                             OH NECK
                   O
                   o.
                   O
                   a.
                        100
                   o
                   a
                   o
                         10
                                                                                             UPPER
                                                                                             EXTREMITIES
                                                                                             LOWER
                                                                                             EX'TREMIVIES
                                        1524
                                                2544
                                                       35-44
                                                               45-54
                                                                       55-64
                                                                              65-74
                                                                                      7544
                                                                                               85 +
                                                            AGE GROUP
                                      TRENDS IN ANNUAL AGE-SPECIFIC. NONMELANOMA
                                         SKIN CANCER RATES AMONG WHITE FEMALES
                      6000
                      1000
                  O
                  a.
                  O
                  a.
                  OC
                  ut
                  O
                  ui
                  O
                  U
                       1CO
                        10
                                                                                            FACE. HEAD
                                                                                            OR NECK
                                                    UPPER
                                                    EXTREMITIES
                                                                                            LOWER
                                                                                            EXTREMITIES
                                                                               I
                                              _L
                                       1624
                                               2S34
                                                       3S44
                                                              4S64
                                                                      b5M
                                                                             6574
                                                                                     7584
                                                                                              85 +
                           627
                    AGt GROUP

-------
                              NONMELANOMA SKIN CANCER INCIDENCE AMONG WHITE MALES

                            IN THE UNITED STATES BY GEOGRAPHIC AREA AND UV-B EXPOSURE
        o
        o
        o

        o"
        o
        cc
        UJ
        Q_
OD

?0
00
o

01



Q
UJ
I-
ISI


-»

D
     800



     700



     600




     500





     400







     300










     200
             100
                    Modi'i   In R-• = 4. • + ti (UV-) + t :


                           (Assumes common slope)
                                                                       •  WHITE MALES-  197778'

                                                                       Q  WHITE MALES  197172.






5
1?
o
(/>


in

O
Q _
M
i w






s 1
D , 5
i 111
r
re
D
t:
u

*

X f
Q
1 ? .
£
O
5


to a-

JO (Q
l.°


c

£

O
?
. ! i
0
O)
c
<
o
3
O"


3
CT
3
A
<







J
               100
                      120             140             160             180             200



                       SOLAR ULTRAVIOLET RADIATION INDEX  (ANNUAL UV- B COUNT X 10 4)

-------
                            NONMELANOMA SKIN CANCER INCIDENCE AMONG WHITE FEMALES
                            IN THE UNITED STATES BY GEOGRAPHIC AREA AND UV-B EXPOSURE
             800

             700


             600
             Model
                    (Assumes common slope)
        o
        o
        o
        o"
        o
       oc
       UJ
       Q.
            500
      400
     300
• WHITE FEMALES  1977-78 '

O WHITE FEMALES  1971-72
                                                                   O
       3
       o
     200
ro
CO
Q
UJ
t-
V)
       UJ
       O
            100 -

             90

             80
                             O
                   5 a> O
                                I
                                          I
                                                           I
I
             100
                        120               140               160               180              200


                        SOLAR ULTRAVIOLET RADIATION INDEX   (ANNUAL UV- B COUNT X 10" 4)

-------
        Slide  12         COMMUNITY HEALTH SURVEY-TELEPHONE QUESTIONNAIRE

When the Survey's interviewer telephones, the following questions will be asked:
   I'm going to ask some questions about the amount of time you have
   spent outdoors during the summer.

   1  In your early adult life  (20's and  30's) during atypical summer
      week, how many hours per week did you spend outdoors during
      daylight hours on weekdays?  What about during your 40"s and
      50's7  What about since you  have been 60?
   2. In your early adult life  (20's and  30's) during a typical summer
      week, how many hours per week did you spend outdoors during
      daylight hours on weekends? What about during your 40's and
      50V  What about since you have been 60?


   3. How many weeks per year do you usually  vacation?


   4. How  many  hours per week  do you usually spend in the sun
      when  you are on vacation?
   5. Since age  20, during a typical  summer, did you sunbathe fre-
      quently, occasionally, rarely or never?
   6. When you  are out in the sun  do you use suntan lotions fre-
      quently, occasionally, rarely or  never? What about sun screens?
      What about protective clothing such as long sleeve shirts or
      hats?
  Now the next two questions will deal with your reaction to the sun
  without the use of suntan lotions.

    7. In the summer, once you have already been in the sun several
      times, what  reaction will your skin have the next time you  go
      out in the sun for two or more hours on a bright day?  Would
      you  say  you get no reaction, some  redness  only, a burn, or a
      painful burn?
   8. After repeated sun exposures, for example, a two-week vacation
      outdoors,  what kind of  a tan will you  have:  Will  you  have
      practically  none,  a  light  tan, an average tan  or  a  deep tan?
   9. Do you use a sun lamp frequently, occasionally, rarely or never?

  10. Have you ever worked with or been routinely exposed to oils,
      coal tar, pitch,  radiation or  radiation  therapy,  industrial
      chemicals, dusts,  fumes,  or arsenic?  If yes, to which one(s)
      of these were you exposed?
11.  Have you ever been treated by a doctor for any of the following
    skin conditions?
        Dry skin
        Oily skin
        Acne or pimples
        Moles/birthmarks
         Eczema
         Psoriasis
         Warts
         Hives
         Unusual  loss of hair
12. What is the color of your eyes?
13. Do you have freckles?

14. What was your natural hair color when you were 15 years old?


Thinking back over your working lifetime:

15. What is the occupation in which you were employed the longest?

    In what kind of business or industry  was that? For how long?

    Were you outdoors on this job frequently, occasionally, rarely,
    or never? How many hours was that per week?
Now I would like to ask you about any jobs you have held for more
than one year at a time, since age 20, that required you to be out-
doors for two or more hours per day.

16. Would you start by  telliipg me about those jobs you had during
    your 20's?  How many years did you hold that job?  How many
    hours per day were you outdoors on that job?
17. Have  you lived in this State most  of  your lifetime?   If no,
    where did you live most of your lifetime?
                                                                   18. In what countries were your four grandparents born?
19. To  which of the following ancestral groups do you consider
    yourself to belong? You may answer more than one:
    English/Welsh
    Scot
    German
    Irish
    Scandinavian
    Polish
Russian
Other Slavic
French
Italian
Spanish
Mexican
Greek
American Indian
Asian
African
Middle Eastern
Other
    20.  Please look at the color chart on  the bottom of the questionnaire and tell me which color matches your skin complexion best.
        Match the chart against  the inside of your upper arm, (the portion that  is not exposed to the sun).  Please give me the number
        above the color.  How closely does your choice match your skin color?  (exactly, fairly closely, not very closely) Is the color chart
        lighter or darker?  What do you consider your complexion to be? (fair, medium, dark)
                                                   SKIN COMPLEXION CHART
       1
                                                                                                 8
                                                                                                                           10
                                                                       630

-------
S11de 13              NCI/EPA  Skin  Cancer Sample Survey
          No. of Individuals  Responding to Telephone Questionnaire

                                                      GENERAL
                                  PATIENTS           POPULATION
         Seattle                    343                 743
         Minneapolis-St.  Paul       443                1143
         Detroit                    374                 829
         Utah                       347                 899
         San Francisco-Oakland       274                1075
         Atlanta                    399                 793
         New Orleans                251                 778
         New Mexico                 421                1219
                 TOTAL             2852                7479
                                           631

-------
Slide  14
         SKIN CANCER  EPIDEMIOLOGY -  White Hales * All  Ages




                                             Complexion

UV-B Count
xlO'1*
101
106
110
147
151
160
176
197



Proportion
"Fair"
i
PATIENT GENERAL POPULATION
Seattle
Minneapolis-St. Paul
Detroi t
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
SKIN CANCER EPIDEMIOLOGY
Prop.
.662
.611
.656
.604
.688
.613
.690
.631
S.D.
(.035)
(.032)
(.033)
(.035)
(.036)
( ..031 )
(.041)
(.032)
- White Females * Al
Prop.
.415
.423
.355
.357
.416
.332
.376
.341
1 Ages
S.D.
(.028)
(.022)
(.026)
(.023)
(.024)
(.023)
(.028)
(.024)

Complexion

UV-B Count
xlO-*
101
106
110
147
151
160
176
197


Proportion
"Fair
II
PATIENT GENERAL POPULATION
Seattle
Minneapolis-3t. Paul
Detroit
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
Prop .
.666
.574
.568
.628
.659
.610
.633
.650
S.D.
(.040)
(.035)
(.040)
(.041)
J.049)
(.040)
(.049)
(.037)
Prop.
.562
.514
.525
.482
.518
.474
.501
.418
S.D.
(.028)
(.019)
(.023)
(.023)
(.020)
(.026)
(.029)
(.021)
                                     632

-------
   Slide 15
             WHITE MALES
                                         Skin Color No.  & Meter-Reading
101
106
no
147
151
160
176
197
 101
 106
 110
 147
 151
 160
 176
 197
Seattle
Minneapolis-St.  Paul
Detroit
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
               WHITE FIMALES
 Seattle
 Minneapolis-St.  Paul
 Detroit
 Utah
 San  Francisco-Oaklant'
 Atlanta
 New  Orleans
 New  Mexico
Color
PATIENT
Prop^ S.D.
.848 (.027)
.831 (.025)
.843 (.025)
.803 (.029)
.773 (.033)
.841 (.023;
.774 (.036)
.845 (.024)
Skin Color
Color
PATIENT
Prop. S.D.
.893 (.026)
.851 (.026)
.871 (.028)
.918 (.022)
.891 (.031)
.839 (.030)
.820 (.039)
.864 (.027)
Number 7-10
GENERAL POPULATION
Prop. S.D.
.691 (.031)
.669 (.022)
.650 (.022)
.642 (.025)
.664 (.021)
.594 (.033)
.535 (.030)
.549 (.027)
No. & Meter Reading
Number 7-10
GENERAL POPULATION
Prop. S.D.
.835 (.020)
.792 (.019)
.835 (.020)
.747 (.022)
.771 (.021)
.727 (.022)
.681 (.025)
.642 (.022)
                                           633

-------
Slide 16
           SKIN CANCER EPIDEMIOLOGY -  White  Males  * All  Ages
                                                Eve  Color
UV-B Count
101
106
no
147
151
160
176
197


Proportion BLUE EYES
PATIENT GENERAL POPULATION
Seattle
Minneapolis-St. Paul
Detroit
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
SKIN CANCER EPIDEMIOLOGY

Prop^
.562
.523
.510
.491
.530
.463
.353
.417
S.D.
(.036)
(.033)
(.034)
(.036)
(.039)
(.032)
(.041)
(.032)
- White Femalas *

Eye
Prop
.462
.441
.365
.464
.352
.423
.293
.304
S.D.
(.029)
(.023)
(.023)
(.026)
(.022)
(.028)
(.024)
(.020)
All Ages
Color
Proportion BLUE
UV-B Count
101
106
110
147
151
160
176
197

PATIENT GENERAL
Seattle
Minneapolis-St. Paul
Detroit
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico

Prop .
.519
.413
.304
.388
.435
.448
.371
.434

S.D.
(.042>
(.035)
(.040)
(.041)
(.051)
(.041)
(.047)
(.038)
634
Prop.
.395
.429
.331
.336
.297
.365
.271
.250

EYES
POPULATION
S.D.
(.027)
(.021)
(.024)
(.022)
(.023)
(.026)
(.022)
(.020)


-------
Slide  17
           SKIN CANCER EPIDEMIOLOGY  - White Male-,  * All Ages
                                              Hair Color

UV-B Count
x.10-*
101
106
110
147
151
160
176
197


UV-B Count
xlO-*
101
106
no
147
151
160
176
1P7,

Seattle
Minneapolis-St. Paul
Detroit
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
SKIN CANCER EPIDEMIOLOGY

Seattle
Minneapolis-ft. Paul
Detroit
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
Proportion
PATIENT
Prop. S.D.
.313 (.034)
.299 (.030)
.346 (.033)
.329 (.034)
.326 (.036)
.296 (.030)
.382 (.041)
.303 (.030)
Red or Blond
GENERAL POPULATION
Prop. S.D.
.235 (.027)
.272 (.020)
.174 (.020)
.276 (.022)
.238 (.021)
.218 (.024)
.226 (.022)
.188 (.019)
- White Females * All Ages
Hair
Proportion
PATIENT
Prop. S.D.
.378 (.040)
.392 (.035)
.368 (.040)
.416 (.041)
.309 (.048)
.436 (.041)
.399 (.048)
.413 (.038)
Color
Red or Blond
GENERAL POPULATION
Prop. S.D.
.350 (.026)
.312 (.023)
.316 (.024)
.310 (.021)
.299 (.021)
.294 (.022)
.313 (.021)
.260 (.020)
                                       635

-------
   Slide 18   WHITE MATES
                                                     Scotch
 101
 106
 110
 147
 151
 160
 176
 197
101
106
110
147
151
160
176
197
 Seattle
 Minneapolis-St. Paul
 Detroit
 Utah
 San Francisco-Oakland
 Atlanta
 New Orleans
 New Mexico
              WHITE FEMALES
Seattle
Minneapolis-St.  Paul
Detroit
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
PATIENT
Prop.
.328
.130
.202
.309
.296
.370
.183
.329

S.D.
(.035)
(.022)
(.028)
(.033)
(.035)
(.032)
(.033)
(.031)

PATIENT
Prop.
.280
.191
.223
.316
.372
.355
.196
.379
S.D.
(.037)
(.028)
(.034)
(.039)
(.050)
(.040)
(.039)
(.038)
GENERAL
Prop
.202
.112
.139
.220
.187
.223
.105
.176
Scotch
GENERAL
Prop.
.233
.087
.142
.198
.199
.244
.133
.157
POPULATION
. S.D.
(.024)
(.014)
(.022)
(.021)
(.016)
(.025)
(.017)
(.018)

POPULATION
S.D.
(.024)
(.on)
(.018)
(.021)
(.020)
(.022)
(.018)
(.015)
                                                 63 a

-------
            WHITE MALES
                               Slide 19
                                                   Irish
101
106
110
147
151
160
176
197
101
106
110
147
151
160
176
197
Seattle
Minneapolis-St. Paul
Detroit
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
             •WHITE FeMATKR
Seattle
Minneapolis-St.  Paul
Detroit
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
PATIENT
Prop.
.419
.298
.338
.198
.509
.462
.449
.538

S.D.
(.036)
(.030)
(.032)
(.029)
(.039)
(.032)
(.043)
(.033)

PATIENT
Prop.
.478
.307
.422
.291
.433
.578
.486
.594
S.D.
(.041)
(.03'3)
(.040)
(.033)
(.051)
(.041)
(.049)
(.038)
GENERAL
Prop
.334
.245
.258
.202
.316
.393
.321
.325
Irish
GENERAL
Prop
.363
.270
.326
.229
.350
.478
.390
.370
POPULATION
. S.D.
(.025)
(.020)
(.021)
(.021)
(.021)
(.030)
(.024)
(.027)

POPULATION
. S.D.
(.027)
(.019)
(.026)
('.021)
(.023)
(.026)
(.025)
(.023)
                                            637

-------
Slide 20
           SKIN CANCER EPIDEMIOLOGY  -  White  Males * All Ages
 UV-B Count

   101
   106
   no
   147
   151
   160
   176
   197
UV-B Count

   101
   106
   110
   147
   151
   160
   176
   197
                                              Scandinavian
                                      PATIENT        GENERAL  POPULATION
       Utah
      Utah

ttle
neapolis-St. Paul
roit
h
Francisco-Oakland
anta
Orleans
Mexico
SKIN CANCER EPIDEMIOLOGY
Prop. S.D.
.179 (.028)
.348 (.031)
.072 (.018)
.322 (.034)
.165 (.029)
.035 (.012)
.035 (.016)
.080 (.018)
- White Females
Prop.
.281
.424
.050
.273
.125
.043
.029
.050
* All Aaes
S.D.
(.025)
(.024)
(.013)
(.024)
CJ015)
(.on)
(.009)
(.010)

Scandinavian


rtle
leapolis-St. Paul
•01 1
i
Francisco-Oakland
mta
Orleans
Mexico

PATIENT
Prop. S.D.
.215 (.034)
.372 (.035)
.056 (.018)
.339 (.040)
.005 (.030)
.026 (.013)
.020 (.014)
.074 (.020)
638
GENERAL POPULATION
Prop .
.279
.401
.052
.341
.149
.040
.045
.069

S.D.
(.025)
(.018)
(.011)
(.021)
(.017)
(.010)
(.010)
(.on)


-------
Slide  21
          SKIN CANCER EPIDEMIOLOGY - White Males  * All Ages
                                         Held fin Outdoor Job
Proportion "Yes"
UV-B Count
xlO-*
101
106
110
147
151
160
176
197


PATIENT
Seattle
Minneapolis-St. Paul
Detroi t
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
SKIN CANCER EPIDEMIOLOGY

Prop.
.702
.666
.597
.833
.745
.664
.567
.773
- White

S.D.
(.033)
(.031)
(.034)
(.026)
(.034)
(.030)
(.043)
(.027)
Females
Held an
GENERAL POPULATION
Prop.
.495
.459
.478
.578
.519
.478
.554
.593
* All Ages
Outdoor Job
S.D.
(.027)
(.025)
(.026)
(.024)
(.025)
(.027)
(.027)
(.029)


Proportion "Yes"
UV-B Count
xlO-*
101
106
110
147
151
160
176
197
PATIENT
Seattle
Minneapolis-St. Paul
Detroi t
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
Prop.
.241 (
.167 (
.297 (
.285 (
.182 (
.164 (
.076 (
.298 (
S.D.
.036)
.027)
.037)
.038)
.040)
.031)
.026)
.035)
GENERAL POPULATION
Prop.
.165 (
.087 (
.133 (
.188 (
.141 (
.085 (
.124 (
.166 (
S.D.
.018)
.013)
.018)
.020)
.017)
.014)
.017)
.020)
                                          639

-------
      Slide 22
  UV-B Count

    101
    105
    110
    147
    151
    160
    176
    197
                  SKIN CANCER EPIDEMIOLOGY -  White  Males  *  All  Age<
                                                          of Tan
  Seattle
  Minneapolis-St. Paul
  Detroit
  Utah
  San Francisco-Oakland
  Atlanta
  New Orleans
  New Mexico
Proportion Deep Tan
PATIENT GENERAL POPULATION'
Prop.
.196
.223
.173
.211
.179
.197
.155
.214
S.D.
(.029)
(.027)
(.026)
(.030)
(.030)
(.026)
(.031)
(.027)
Prop.
.352
.359
.404
.362
.391
.391
.387
.410
S.D.
(.028)
(.022)
(.026)
(.027)
(.019)
(.030)
(.026)
(.027)
UV-B Count
   101
   106
   no
   147
   151
   160
   176
   197
                SKIN CANCER EPIDEMIOLOGY - White Females  * All  Ages
                                                        Type  of  Tan
Seattle
Minneapolis-St.  Paul
Detroi t
Utah
San Francisco-Oakland
Atlanta
New Orleans
New Mexico
Proportion Deep Tan
PATIENT GENERAL POPULATION
Prop.
.135
.190
.178
^?Q
.183
.138
.129
.156
S.D.
(.028)
(.028^
(.031)
(.028)
(.040)
(.028)
(.035)
(.028)
Prop.
.235
.219
.260
.231
.275
.271
.213
.278
S.D.
(.023)
(.017)
(.024)
(.021)
(.019)
(.021)
(.022)
(.019)
                                                640

-------
                                   Discussion

Dr. Kelsey, NCI:  Have you included any data on people who use sunscreens, for
example?

Mr. Scotto, NCI:  Yes. We ask the question whether they use a sunscreen, as you may
have seen in the slide.  We have gotten very little information on that.  I do not think
that the  general population understood what a sunscreen was, but the patient group,
as expected, had a higher proportion.  They did admit to using or even knowing about
sunscreens.

Dr. Cameron, NCI:  I  had  two  questions, but I think you have already answered the
first one just in the last few moments. The reason for  rationale for breaking out the
melanoma from the other skin cancers is the fact that  it does not necessarily appear
in the exposed portions of the skin, is that correct?

Mr. Scotto, NCI:  The reason for breaking out?

Dr. Cameron, NCI:  Yes, for separating the melanomas from other skin cancer.

Mr.  Scotto, NCI:   One reason for separating  these studies  is  that melanoma is a
malignancy which is routinely  reported  to the  SEER  program,  which the NCI also
conducts and monitors.  But SEER does  not uniformly  collect incidence information
on non-melanoma.   The reason for this is that the  basal cell and squamous cell
carcinomas  of the  skin are  usually treated  in the  physicians's  office  or as  an
out-patient.   We have to  canvass doctor's office to  access their  records,  a  more
tedious kind  of study.  The  information on the other malignancies is pretty  much
complete and available  in the hospital chart  records.   Another  reason  is, as I
indicated earlier, that  the process  by which UV relates to either the induction or the
promotion of skin melanoma appears to be different from  the skin cancer.  I think Dr.
Orme  mentioned  that the reasons  why we  want to  get  at personal  dosimetry
information is because we  want to measure something about a short-period, and to
see if  we could measure the effects of various modes of exposure.  Mathematical
models applied to the various skin malignancy data indicate that the  process involving
UV may be different for skin melanoma and skin cancer.

Dr. Kelsey, NCI:   My second question is has anybody approached the reason for the
difference or variance  in physician cooperation?

Mr. Scotto, NCI:  There is usually a variance of physician  cooperation in most studies.
Epidemiological studies are usually difficult in the South, where the tendency has
been to  not  get  involved  with federal  projects.   Our  contractors in each of the
locations were local universities, health  groups and cancer registries.  That was the
beauty of  attaching to an  existing program.   The  SEER  program  had  already
established  the cooperation from the  medical community.   Physicians are not
reluctant to provide medical records.  However, obtaining permission to contact the
patient for additional epidemiological information was difficult in some locations.

Dr. Orme, NCI:  I was not aware that the  personal  dosimeter was tied into your
program.  I think that is a major incentive to prod the Boston group.

Mr. Scotto, NCI:  We have  been waiting. The information on the personal dosimeter
was supposed to come to us eventually.  Drs.  Forziatti and DeFabo, who  had earlier
                                    641

-------
represented the EPA on this NCI/EPA project, had hoped that we could set up some
field tests for personal dosimeters. I have talked to Dr. Davidson and the people who
are developing the personal dosimeters and one of the reasons we were getting  into
the new locations was not only to obtain more needed epidemiologic information from
northern and southern locations and to explore some of the leads on these epidemiolo-
gical factors, but also to be able and ready to conduct the field studies.  From what
you said, it sounds like when Boston is finished developing and evaluating the physical
measuring device, we will probably  be out of funds and out of the new locations
where studies have recently been implemented.

Dr. Orme, NCI: Right. That is what I am asking.

Mr. Scotto, NCI:  We are going to run out of funds by the end of this year.

Dr. Orme, NCI: Would the film badge type of thing, even in the developmental stage,
be useful to you now?

Mr. Scotto,  NCI:  Yes. I would recommend that whatever you do on it, first of all you
should, before we do anything as you indicated, make sure we make all the  laboratory
tests to see what kind of variability we are stuck  with and to see how useful such a
thing would be, before we conduct field studies. I suggest and recommend that you
do these in locations where we  already have epidemiological information on  skin
cancer and  where we already  have UV measurements such  as from  the Robertson-
Berger meter, especially if you are going to use the personal dosimeter device which
was calibrated to the R-B meter.

Dr. Orme, NCI:  Well, I am more optimistic about a continuation of this than perhaps
you are  at this stage.

Mr. Scotto, NCI:   Right now,  by  the way,  is a good time. The study is going on in
New Hampshire/Vermont, which is real close to Boston.

Dr. Orme, NCI:  Well, I will definitely get back to  Herb Wiser about this to see if we
can coordinate it  a little more closely. The other question I had was, you  mentioned
that the incidence rate has gone up from 300,000 in an earlier estimate to  400,000.
Now, I was not sure that you were suggesting that that was real change in incidence
or is that an improvement in your methodology? Are  you saying that that  is actually
correlated with real decreases  in ozone?

Mr. Scotto, NCI:  No, I cannot say that that  is correlated with real decreases in
ozone.  With respect to the measurements  of the  ultraviolet radiation reaching the
earth's surface over time, we  hardly see any trends during the short period we have
been obtaining measurements.  So, I cannot say that there has been a substantial, or
any notable, increase in UV, or decrease in  ozone.   The  estimate of the biological
amplification  factor  is better  because of  the added  locations.   After  making
adjustments for  the time  of the  year in which the studies were conducted in San
Francisco and Minneapolis-St. Paul the indications are that there has been a 15 to
20% increase  in  skin cancer  over  the six year period from  1972  to 1978.  These
increases are mainly  observed for basal cell carcinomas of the skin.  Hardly  any
increase was noted for squamous cell carcinomas.

Dr. Orme, NCI:  If in fact the Robertson-Berger meters  over this same  period are
giving us generally a steady reading, I am just wondering whether we should take  into
consideration the  possibility of a chemical UV interaction in some of these  areas.


                                    642

-------
Mr. Scotto, NCI:  I thought some of you were doing that.

Dr. Orme, NCI: We are doing it experimentally.

Mr. Scotto, NCI:  I have not gotten that far into the human studies.

Dr. Orme, NCI:  The third question I had was the relationship between susceptibility
to skin cancer and fair skin, which we  have toyed with in a lot of ways.  This is
obviously an over-simplification  of things. I was just going to point out some of these
things.  We have looked at a number of different strains of albino mice, for instance,
and measured the susceptibility.  These were hairless  albino mice and they  still
showed a  wide spectrum  of  ranges  of susceptibility.  So there are obviously many
factors contributing to that variation.
                                      643

-------
FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

   PROGRESS ON 3OINT ENVIRONMENTAL AND

       OCCUPATIONAL CANCER STUDIES
            Thursday Morning, May 8
      CONCURRENT SESSION I (CONTINUED)
         RADIATION CARCINOGENESIS
            SESSION CHAIRPERSON


             Dr. Thomas P. Cameron
            National Cancer Institute
                         644

-------
                     Hairless Mice for Carcinogenesis Studies

                                Thomas W. Orme
                            National Cancer Institute

Key Personnel:
     Dr. P. Donald Forbes, Temple University Skin and Cancer Hospital

Project Officers:
     Dr. Thomas W. Orme, NCI
     Dr. Herbert Wiser, EPA


     First,  I am going to talk about  non-haired mice  in general and work sponsored
by the  NCI/EPA agreement at Temple University directed by Dr. P. Donald Forbes.  I
am going  to describe an outbred strain that is very common, the Skh-l line, and an
inbred  derivative of that line, which I  think will be very useful in photocarcinogenesis
work in the future.  I will talk about the  photoresponses  of  these lines  and of a
number of other non-haired mouse lines. Then, I am going to talk about a contract at
Emory University with Dr. Issac Willis, who is attempting to dissect the UV  spectrum
to  show whether there are synergistic biological effects attributable  to  UVA and
UVB light,  again  using the  Skh-l mouse.  Lastly,  I will talk about a postulated
mechanism of UV Carcinogenesis involving sterol derivatives.

     Non-haired mice are not to be  confused with nude (nu/nu) mice.  I really want
to emphasize this, because 90 percent of the people to whom I  talked about hairless
mice really think I am talking about the nude mice which are used commonly in
immunological studies.  The mice which I will be referring to  are  not  nude (nu/nu)
mice.  They are not immunologically incompetent and they  do not have the athymic
condition  that the nude mouse has.  Furthermore, the  term hairless is not always
reserved  for the genotype  (hr_/hr_) which refers to a specific  mutant, but  is used
interchangeably  with the term non-haired.

     Non-haired mice have been used in photocarcinogenesis  studies as a conven-
ience.  In  haired strains, tumors induced by UV irradiation are often confined to ears
or non-haired  extremities and are often of mesodermal rather than epidermal origin.
Shaving or  enzymatically  removing  hair  from  haired mice  in  photocarcinogenesis
experiments is possible but is  generally considered laborious.  There are,  however,
some people who contend that  the haired strains  are  intrinsically better models for
photocarcinogenesis  than the hairless  mice or non-haired varieties.

     Several mutations lead to the non-haired phenotype.  The recessive gene hr  is
encountered most frequently.  Non-haired mice, although convenient sources of bare
skin  for  irradiation, have  some  peculiar  problems  on  their  own.   Since  mouse
pigmentation  is closely associated with the hair follicle  and is  most noticeably
expressed  as coat color or hair color rather than skin color, hairless mice  are only
weakly pigmented due to  the disruption of hair  follicles.   A hairless mouse  skin tyoe
comparable to negroid skin does not exist, as far as I know. Non-albino hairless mice
are capable of a tanning reaction, however.

     Another common  problem  with non-haired mice  is that the young  will not
always accept a non-haired mother.  This  necessitates a  heterozygote   breeding
program, which cuts the efficiency of animal production in half.
                                     645

-------
     It was  also suspected, and  to  some  extent  shown experimentally, that non-
haired  mice  suffer from a variety of immunodeficiencies, not  as  striking as the
athymic  condition in nude (nu/njj) mice but, nevertheless, important in considering
photocarcinogenesis.  Photobiologists have addressed these problems in  a variety of
ways.  The result has been the establishment of numerous colonies of outbred and
inbred  non-haired mice with fundamentally different  characteristics.  In the 1960's
and  1970's, it was becoming more and more  obvious that different responses to UV
irradiation or to chemical treatment coupled with UV  irradiation could be attributed
in part to differences in the strains of mice that were being used.

     I would like  to digress  here for a  minute.   I  first became interested in
photobiology about  four  years  ago when I was  given responsibility for monitoring a
contract at Temple University.   I was coming from  the field of immunology.  It
appeared  to  me when I first reviewed some projects  in photobiology as if the field
were in a situation similar to that of histocompatibility antigen research prior to the
discovery of inbred  strains.   There were  two major variables that had not been
standardized in photocarcinogenesis research.  I think  these two variables accounted
for most  of  the discrepancies in  data and for most  of the disagreements  between
photobiologists.  One was the differences that were prevalent in the  mice that were
being used. Everyone had his own non-haired strain of mouse reared  in the basement
breeding colony.  Everyone assumed  that he  had the  same mouse that  others were
working with.  This was  not the case at all.  The other variable that was extremely
significant to the outcome of photocarcinogenesis experiments  was the  light source
used to  irradiate  animals.  Different light  sources  had widely  different  spectral
characteristics.  They were roughly classified as UV-B emitters  or UV-A emitters and
were employed with and without filters.  The spectra of light they emitted were quite
different.  As we shall see, this has a major effect on the outcome of a radiation
experiment.

In a number  of respects  UV induced skin cancer as  it  is being studied in animals and
as it is observed clinically is better understood than any other  type  of cancer.  The
relationship  between  UV  exposure  (dose  or  irradiance)  and tumor incidence is
understood in general terms.   Molecular mechanisms  of initiation of carcinogenesis
have been proposed  and  various hypotheses are amenable to experimentation. Skin
cancers, because they are surface cancers, can  be observed readily, can be biopsied,
and  for these reasons, progression or regression can  be  scored very simply in both
animals and  in man.  There is good correlation between the information  we have
about humans and the information we are generating  in animal experiments.  Hence
the subject area, skin cancer,  could serve well as  a proving ground  for  ideas about
human risk assessment based on an interpretation of animal data.  Futhermore, there
has  been  considerable progress in analyzing the various genetic factors that are
responsible for susceptibility or resistance to skin cancer.

     Fig. 1 shows the spectrum from a solar simulator. It is pertinent  to all of the
papers on photocarcinogenesis.   I am going  to  use  it  here  to  indicate what the
atmosphere does to  protect the surface of the earth from ultraviolet radiation.

     The top line in  Fig. 1 represents the actual solar spectrum as it passes through
space before reaching the earths atmosphere.  UV light is divided  arbitrarily into
three categories, A,R and C.   It  is generally assumed that the UV-B component of
ultraviolet light is the most active biologically and  the most important type of light
that reaches the surface of the earth. UV-C  is also biologically important radiation,
but it is effectively filtered by the ozone in the atmosphere.  The biological role of
                                       646

-------
UVA is controversial.  The bottom curve is the spectrum of the solar radiation (in this
particular case experimentally simulated) as it reaches the earth.

     In Figure 2, we can see that ozone does not act exclusively as a neutral density
filter.   When the effective  thickness of the  ozone layer changes, and this can be
simulated in the laboratory by a series of Schott glass  filters, there is a qualitative,
as well  as a quantitative  shift in the UV spectrum. Filtration effectively eliminates
most of  the  UVB that is biologically  active.  As the  filtration  decreases, either
naturally (hypothetically) by  cataclysmic loss of ozone from the atmosphere or, in the
laboratory setting by decreasing the thickness of Schott glass filters,  there  is a
disproportionate increase in incident irradiation in  the biologically active region.  For
that reason, the potential changes in the atmosphere associated with ozone  destruc-
tion are considered significant,  because  not only would they let in more UV light, but
they would let in light at particularly active wavelengths.

     Figure 2 gave an indication of the spectral quality of a solar simulator, which
mimics very nicely what is called an effective ozone concentration.  One of the very
tedious jobs that was performed at the Temple University laboratory was to construct
real ozone filters, and to measure filtration as a function of O, concentration.  Dr.
Forbes was able to show that the Schott glass filter system mimicked very nicely true
ozone filtration with respect to  the quality of the spectrum in the UV-B region.

     The contract, which I am  going to describe now, is one that was awarded to the
Temple University, School of Medicine, Skin and Cancer Hospital in  Philadelphia.
The principal investigator is  P. Donald Forbes. The contract was initiated to  examine
the extent of strain variation in response to irradiation with a solar simulator and to
develop criteria for selecting one strain of  hairless mouse for large scale production.
The initial objective  was not to look at the significance of  strain variation, but to
find the ideal hairless mouse to put into production for all photbbiology work.  That
concept proved somewhat naive, although Dr. Forbes has obtained information that
has resulted in a special interest in the inbred strain designated HRA/Skh.  This strain
has good breeding characteristics and high UV sensitivity.

     Table I outlines the strains that are being examined by the Temple University
group and gives designations  of their genotypes, their UV sensitivity, that is,  whether
they are highly susceptible to UV-3  induced tumors  or solar  simulator  induced
tumors, their acceptance of a non-haired mother, which is a parameter of importance
in developing a colony because it gives the production efficiency, and the breeding
schedule for these strains. I am using inbred here  to imply brother/sister mating and
I am using the word  outbred loosely to indicate any schedule that does not involve
specifically brother/sister mating.

     Figure  ?> illustrates the  production problems that are encountered with the
animals that will not accept a non-haired mother.  The  figure shows rhino mice.  One
parent mouse  is a homozygous male.   It  is the male,  which is mated with the
heterozygous female.  Since 'lie rhino gene is recessive, the heterozygous female will
be haired. The offspring are either haired (hrr /+) or non-haired (hrr /hrr  ).  This
gives you an example  a forced heterozygosis breeding schedule.
                                         647

-------
                                      Table 1
Name
Mouse Strain or Line Designations  and  Characteristics

        Genotype
     UV
Sensitivity
Non-haired Breeding Schedule
 Mother
Accepted
Skh:hairless-l


Skh:hairless-2


Skhrcrh



Balb/cSkh-ab




HR/De/Hflcr



C3H/HeN-hr



HRS/J



HRS/Anl

HRA/Skh

Skh:(HRxRH)F1
        select hr/hr        H
        and c/c

        select hr/hr        M
        and non-albino

        crh, c/c, a/a       L
        ab, c/c, b/b,
        hr, br/br, p/p
        hr                  M
        hr,  c/c
        b/b, d/d
        hr/hr,c/c           M

        hr/hr, c/c          H

        hr/hrrh             ?
                         Outbred,  segregating
                         c,  b and  a

                         Outbred,  segregating
                         c,  b and  a

                         Outbred,  forced
                         heterozygosis
                         for crh

                         Inbred, forced
                         heterozygosis
                         for ab to maintain
                         haired counterpart

                         Inbred, forced
                         heterozygosis
                         for hr

                         Inbred, forced
                         heterozygosis
                         for hr

                         Inbred, forced
                         heterozygosis
                         for hr

                         Outbred

                         Inbred

                         Hybrid
H,M,L mean high, medium and low susceptibility to carcinogenesis induced
by UV irradiation with a solar simulator.

Inbred implies a brother-sister mating schedule which may not have reached
a 20th generation; outbred includes various schedules not specifying
brother-sister mating.
                                          648

-------
      Two of the  offspring in Figure  3 are homozygous (hrr /hrr )  rhinos that still
have their juvenile hair coat.  The mutation does not affect the juvenile hair coat; it
only affects the adult hair coat.  These mice are born haired and  then they gradually
lost their hair, starting from  the face proceeding all the way down  the back of  the
animal.  Another one of the young in figure  3 is a heterozygote; it will be  used in the
subsequent breeding  schedules if female.  Most  of them  are  discarded  and this
decreases the efficiency of the production of these mice.

Most  of the mice  in Table 1 which require a forced heterozygosis breeding schedule
are bred homozygous male to heterozygous female.  Fortunately,  there are some
strains that accept a  non-haired mother.  They can be bred directly by brother/sister
mating.  The  most  vigorous  line is the HRA/Skh, which is bred as a homozygous
(hr/hr) breeding pair on a brother/sister mating schedule.  The pedigreed line is now
approaching the F20  generation.  It then  will be designated an  inbred strain.  This
strain has  many useful characteristics, and has been  designated for  a large scale
production and for use  in the bioassay  of psoralen  derivatives.  V7e suspect that this
strain, the HRA/Skh, will replace the  frequently used outbred strain  Skh-hairless-1
from  which it was derived.

      DR. CAMERON:  Can  you give  us the reasons for picking an inbred  over an
outbred?

      DR. OR ME:  We  want eventually to  examine questions related to the genetic
control of  susceptibility to UV carci no genesis.  To  do that, we needed  the inbred
lines.  I think  that in conducting a wide variety of  biochemical experiments in which
you are specifically looking for genetic control over various phenomena, there is no
choice but to  go the  inbred route.   I also  do not see any advantage  in continuing to
use an outbred line when the breeding characteristics of a comparable inbred line  are
good.  The only legitimate reason for using outbred lines is production ease.  I think
we have in the  HRA/Skh  line, an inbred line that has good breeding characteristics.
They  are not quite as good as those of the outbred Skh-1 line, but Dr. Stanley Mann
who directs the breeding operations  at Temple is getting satisfactory litter sizes and
satisfactory viability at weaning.  My general preference is  to design carcinogenesis
and toxicological  experiments exclusively  with the inbred  line, because it allows
repetition of  experiments and the  use of genetic tools  in analyzing the observed
phenomena.

      The idea  that  outbred   mice  . are better  models of a  heterogeneous  human
population does not lead to any worthwhile experiments. If, for instance, only 5 of 50
outbred mice irradiated with  UVB were to develop tumors, it would  be impossible to
prove that these 5 are  unusually susceptible for genetic  reasons and impossible to
reproduce their genotypes.   If,  in  fact, only those 5 were  susceptible because of
genotype, the statistically meaningful group size for experimental design, would have
been  reduced  from  50 to  5, and   100% of  the  mice  of that  particular genotype
receiving UV treatment would have had to respond with tumors before an effect
could be detected. The topic  of genotype specific responses is important as we shall
see.  But meaningful experiments cannot be  conducted  with outbred  mice.  Artifical
heterogeneity  is the only approach,  and this  artificial heterogeneity in  a mouse
population can be constructed  by using defined numbers of inbred  mice selected from
a variety of strains.  The conclusion  is simple. If genetic tools are  to  be used to
analyze iri vivo  carcinogenesis, use  inbred lines.  A corollary  might be that hi vivo
experiments not amendable to genetic analysis are primitive in conception.
                                        649

-------
      I would now like to describe the solar simulator.  The solar simulator system
and its relationship to  other models for solar light and ozone filtration has been
described by Or. Forbes (Figure 4).  This unit contains a xenon arc lamp.  Each of the
panels contains a system of filters.  These can be either  neutral density filters, cut-
off filters or the  Schott glass filters, which are used  to simulate various effective
thicknesses of  ozone.  With this set-up, Or. Forbes  can simultaneously irradiate
different  racks of mice with qualitatively different light spectra depending upon the
system of filters put in at the various windows.

      Figure 5 is another picture showing the xenon arc lamp in the middle of a bank
of racks.  Each one of these racks is getting a qualitatively different type of UV, but
they can be irradiated for the same period of time and reirradiated simultaneously.

      Figure 6 shows what  is called  the  Mouse Sheraton in contrast to some simpler
arrangements which are called the Vlouse Holiday Inns. You can see what is actually
happening in the  cages.  In  one  cage, a mouse is trying to hide.  But mice cannot
escape irradiation.  They do preferentially turn while  they are being irradiated, so
that most of the radiation falls on the back.  There are left-handed and right-handed
mice, as oertains to the side they prefer to present to the  irradiation apparatus.

      The most common response during  the irradiation period  is for  mice to curl up
and to go to sleep.  I imagine that is  part of their simulated nocturnal/diurnal cycle.

      OR.  CAMERON:  Tom,  do  you  know if there is any degree  of blindness or
retinal degeneration?

      OR. ORME:   These mice are albino.  I do not know  the answer  to that, Tom.  I
do not know of any specific  physical changes in  the eye.  Eye tumors, for instance,
are not common.  So, I would have to ask about any effects leading to blindness.

      Figure 7 was made before we  introduced  good laboratory  practices. It  shows
the position and the multiplicity of the lesions observed.  As I said, the entire back of
this animal is a susceptible  target, rather than just the ears as might be the case with
a haired mouse.  v'aos or diagrams  of the actual position of the various tumors  are
maintained  on a 'vee'-dy basis,  so that the progression, regression or  coalescence of
various tumors  can be followed precisely. One of the things I mentioned earlier is the
uniqueness of this  system for measuring time to tumor  and following the progression
of lesions.  That coupled with the fact that these tumors arise in 2*f weeks makes it a
very useful experimental model.

      figure 8 shows some advanced  tumors.

      I think it is important  to present Figure  9 to  explain one of  the  major
experimental variables.  One  of the reasons why plots of tumor incidence versus time
differ  from laboratory  to  laboratory  so significantly  is that  people  have  not
standardized the tumor scoring procedure.  Figure 9 pertains  to the same group of
mice, but scoring  is dependent on different diameters of tumor. If in fact you start
the scoring with 0.5 millimeter tumors, which are barely perceptible red dots, you
record an early incidence.  Or.  Forbes thinks the tumors that can be scored with some
degree of certainty are tumors  one millimeter in diameter or larger. The time course
changes very significantly  according to what  you consider a tumor of scorable size.
Since most  of the  publications in experimental  photocarcinogenesis  do not specify
this parameter, you can understand why discrepancies in published data exist.
                                    650

-------
     The first major experimental variable that Dr. Forbes wanted to investigate in
these studies was immunocompetence.  Since  there was a prevalent notion that non-
haired mice  were immunologically deficient, he  set  up a screen of immunological
parameters looking  at both  cell mediated immunity  and antibody formation.  The
conclusion is described in  a  publication that  has just  been submitted by Drs.  Sharon
Smith and Don Forbes, and it is that all the lines, as  far as major responses such as
the ability to form antibody and react with specific antigens and such as the ability
to mount a T  cell mediated immune response, are  immunocompetent.  There  are
quantitative differences  in the various strains, but there was no deficiency that could
be called a major immunodeficiency which could be responsible for differences in
strain susceptibility to UV carcinogenesis.  That does not mean that there are not
specific subsets of the various basic immunological tests that are  completely lacking.
For instance, Dr. Smith did not break  down  T and B cell mediated functions into
subfunctions that cpuld be  tested independently.


     Then comes the meat of the matter.  I find this fascinating.

     Figure  10 shows the response of the various non-haired lines at 24 weeks.  The
names differ somewhat  from  those shown in Table  1.  For instance, the HRS./3 is
called JAX in Figure 10, and the  HRS/Argonne is called Anl.  This is the response of
the various lines to  exactly the same irradiation conditions.  The  mice have been
exposed  five times  a week  to the solar simulator for a specified  period  of  time.
Those irradiation conditions are spelled out in the abstract.

     Cryptothrix and absebia  are the  most resistant.  They are mutations distinct
from hairless (hr).  They map at different sites and have quite a different physiology.
The other mutations are all  hairless (hr).  Still, we can see a wide  variation in the
susceptibility with different backgrounds carrying the same mutation.

     There is no  correlation between susceptibility and the  albino gene.   For
instance, 3ax  is an albino  strain as  is  HRA/Skh.    Their responses  to  the solar
simulator differ significantly.  There are major differences in the susceptibility even
when the animals carry the albino gene. There is no association of susceptibility with
the hairless (hr) gene.

     The bar graph  in the upper half of  Figure 10 presents the percentage  of the
animals that have at least one tumor a millimeter in diameter at 24 weeks.

     Not only  is there a wide discrepancy in the number of mice or the percentage of
mice affected, but the lower  bar graph of Figure 10 shows the  multiplicity  of  the
tumors  on  the affected  mice.   Again,  pronounced  strain specific variation  is
indicated.  The Skh-1 and the HRA  showed  multiplicities of tumors  40  and above
within 24 weeks.  What is being scored in Figure 10 are mainly papillomas, but a high
percentage of them will progress to frank carcinomas.


     That is really all I  wanted to say about  Dr. Forbes1 work.  Temole will exercise
its option in a three-year incrementally funded contract to revise  its work statement.
Rather than screen  more strains for susceptibility, they  are going to start using the
various differences that they have observed to analyze the mechanism of  photocar-
cinogenesis.

-------
      A possible line of experimentation is to actually start making crosses between
the inbred  strains  that are available to find  out  whether the susceptibility relation-
ships are dominant or recessive. This could be done, for instance, with absebia, which
is on a Balb/c inbred background. It could be  crossed very easily with the inbred HRA
line to look for different relationships there.

      What I like about this series of experiments is that it is telling us something
that we have known for a long time about humans but which we continue to ignore.
Different people have different susceptibilities  to irradiation.  Probably  the  same
could be said about chemical  carcinogens. I use this to  illustrate what I think is a
great mistake  in our approaches to human risk assessment.

      We could have taken the data from any one of these lines, turned it  over to a
statistician and said give me a dose response curve that we can use to extrapolate by
various fudge  factors to the human population.   Literally,  we would  have the full
range of possibilities.  What  we really  want to know  when  we make a human risk
assessment is  not  the dose-response curve for or the probable susceptibility of the
whole population;  what we really want to know is  the size  of the population at high
risk and why it is at high risk.  I hope that this tyoe of experiment is going to give us
some insights  into how to start looking  at the human risk  assessment problem with
those considerations in mind.

      I personally  feel that this type of data shows the  futility of applying  mathe-
matical modeling  to the data that are obtained from one  species  or  one  strain of
animal.  It just may be totally atypical of the population  at large.  For this reason, I
think we have to  start building into our experimental designs for risk assessment
some consideration of the variation in responses that we get.

      DR. KELSEY:  I just wanted to say that I am very excited about this area.  I
have  been  following  the  hairless  mice story from a different point  of view,  from
cholesterol  metabolism.   But  showing   the  genetic  susceptibility  seems like an
interesting way, as you have just mentioned.  I am wondering how widespread would
this type of analysis be in chemical carcinogenesis.

      OR. ORME:   Well, I have submitted a question for this afternoon's discussion
which  treats that.  Vly  feeling is  that  if you  were to  do the same thing with a
chemical carcinogenesis screen, let's say test a carcinogen at the maximum tolerated
dose, which I  think overrides genetic resistance  in  some  cases,  in  ten  different
strains, some might develop tumors and others not.  I give you an example:  C57  black
mice  are known to be resistant to most  of the chlorinated hydrocarbons.   They are
not totally resistant, but  you  have to go  a full two years before you start  detecting
carcinogenesis  in  the C57 black.   On the other hand,  with the C3H  line,  similar
exposure conditions induce liver tumors rapidly.

With 2-AAp, the same pattern prevails:the C57 is resistant and will get tumors only
very late, while the C3H gets tumors  right away.  In fact, if  you give 2-AAF in the
drinking  water, a  dose that kills C3H immediately is tolerated by  C57  black for its
full lifetime.  So you see that there are major metabolic differences in these lines.

      The problem, and this is the matter that I  want to bring up this  afternoon, is
that we  are not gearing our resources to multistrain testing.  Current systems for
production  of  animals are designed to make large numbers  of Fischer  3^4- rats and
        mice available and the use of a restricted number of strains is justified when
                                      652

-------
one considers animal health problems. The non-haired mice I have discussed are now
in  conventional facilities and they are suscentible to and are carrying a variety of
undesirable viruses like Sendai and mouse hepatitis virus. We hope that those viruses
are not influencing experimental  results.  But if we were  to rederive and  put into
production all of  these  animals, the cost would  be  enormous.  As  1 said,  we have
decided to do this  with a single mouse line only so we are not getting much closer to
multistrain testing.

      OR. KELSEY:  I  guess  my question would be particularly where vou know have
positive  carcinogenic  response  from  classic bioassays.   Would  it  not then  be  a
reasonable step to look at these effects in various other strains.

      DR. ORME:  Yes, I think that that ought to be done.

      DR. KELSEY:  How do you determine the metabolism':'  I realize that  you may
not be able to do a full blown bioassay.

      DR. ORME:  Yes,  I think we get into a variety of problems by not doing more
work in  metabolism and pharmacokinetics.   There  are  big  loopholes  in the basic
theory of the bioassay. A good example pertains to the  benzidine dyes that have been
tested with the Fischer  344 rat.   The benzidine dyes are  very  carcinogenic  and
metabolism  studies have been done  in the Fischer 344  rat showing that benzidine is
excreted in the urine along with 4-amino-biphenyl. Unfortunately for the theory, 4-
amino-biphenyl and  benzidine have  never  been  tested for  carcinogenicity in the
Fischer 344 rat but only in other strains of rats.  So those people who are arguing that
the  measurement  of urinary metabolites is presumptive evidence for  the  carcino-
genicity of the parent benzidine dyes have one more experiment to do. And after this
is  done we will still wonder why the mouse is resistant to the benzidine dyes.
                                   653

-------
 .01
4.00. .    UV-C
1.00.
 .70.

 .40.
 .10
 .07.

 .04.
                        UV-B
                                                    UV-A
                                               is?
•tt
   250
            Figure 1

            Solar Spectrum:   Upper Line, in space; lower
            line;  at sea level on Earth's Surface.
                                    654

-------
4.00
1.00.

 .70.

 .40 .
 .10,

 .07.

 .04.
       pW/CM SQ
 .01.
                                              Schott WG 32O
                                              Filter thickness
    O.64 MM
    1.OOMM
    1.3O MM
    1.7OMM
    2.OOMM
    aOOMM
    4.OOMM
   280
             290
                        300
                                  310
                                        NM
320
                                                       33O
                                                                 340
                                                                           350
              Figure 2

              Modification of  Solar Simulator Output by Schott glass
              filters of decreasing thickness.  Simulation of Ozone
              filtration of solar irradiation.
                                         655

-------

Figure 3

Heterozygous Breeding of Rhino Mice.   (See text
for explanation)
                     656

-------
Figure 4



Solar Simulator
                      657

-------
Figure 5



Solar Simulator in bank of racks.
                        658

-------
Figure 6



Irradiation Rack
                    659

-------

                       •
Figure 7



Early tumors
                    6GO

-------
Figure 8
Advanced Tumors
                    661

-------
  1.0-
   .8-
LU

8  -6-1
UJ
a
o
2  .4
   .2-
EXPMT.76-85

 TUMOR DIAMETER

 • 5 .BMM
 • 5  1MM
 A 5  2MM
 a >  4 MM
    0- i.
                                                                12341-2
              20
                   30
    40

WEEKS
50
60
            Figure 9

            Tumor Incidence vs. Minimal Size for Scoring.
                                   662

-------
                         Figure 10
1 n n
AFFECTED
^ Ol O) ^J 03 «D <
O O O O O O (
I I I I I
& 30-
20-
1 0-
l
228<
S
K
H
-1
?-]
\-\
R
A

C
R a
hi b
I

I
C
R
78-82 (2)
51 mm TUMORS
24 WKS.

C
3
H

J
A
X

"A
N
L"
S
K
H
-2

               12289-2
50-
45
UJ
to 4 n
*T V —
o
5 35 -
5 30-
Q.
2 5
y) ^ ° -
tr
O 20
5
1 0-
5 -






















S
K
H
-1












H
R
A












78-82 (2)
ALL TUMORS

24 WKS.








C
R a
H b
I
	 H







I
C
R








C
3
H





J
A
X



"A
N
L"




S
K
H
-2



Figure 10.  Strain Variation in Tumor Incidence
                                663

-------
                                                  Willis:Grant # R806392010
     Project:  "Effects of Varying Doses of UV on Mammalian Skin:
                Simulation of Decreasing Stratospheric Ozone."
A.   Introduction:

     The malignancies and premalignancies unequivocally associated with sun-
exposure include solar keratoses, basal cell epitheliomas, squamous cell
carcinomas and keratoacanthomas.  The evidence that such tumors might be
associated with solar radiation includes their increased prevalence on parts
of the body habitually exposed to sunlight, in lightly pigmented and, thereby,
less protected individuals in areas of the world which have the greatest
insolation, and in individuals who spend more time out of doors.

     The problem of ultraviolet carcinogenesis is complex.  Among the factors
which must be considered are  (a) relative effectiveness of different wave-
lengths in carcinogenesis, (b) dose-response relationships, (c) time-dose
effects,  (d) repair processes and (e) immunologic considerations.  Closely
related to the above are the environmental effects on UV carcinogenesis,
which involve geographic, temporal, anatomical, and life-style-related conside-
rations (e.g. modification of the stratospheric ozone layer by human activities)
This work deals with factors a-c and the ozone depletion problem explained
below.

     Although it is generally agreed that the UVB portion of the solar
spectrum  (i.e. 290-320 nm) is most effective in causing skin cancer, knowl-
edge of the relative effectiveness of these wavelengths is not well known
(i.e., reliable action spectra do not exist).  Moreover, the UVA region
(320-400 nm) , which was long thought to be unimportant in UV carcinogenesis
is now known to augment UVB sunburn response and also appeals to augment the UVB
carcinoqenesis.   in preliminary experiments, we observed that the photo-
augmentation effect for sunburn response is time-dependent; the time interval
between exposure must not exceed six hours for the effect to be observed.
The observation of UVA involvement is of great practical (as well as mech-
anistic) importance because of the predominance of UVA in the total UV
solar spectrum.

     Because of the complexities brought on by the above-mentioned photo-
augmentation phenomenon, as well as by competing repair-, biochemical, and
immunologic mechanisms, the dose-response or time-dose effects may not be
straight forward.  Indeed, time-dose reciprocity does not exist for UV-
induced tumors (Urbach et al, 1979; our own data to be published). The
detailed knowledge of time-dose-response characteristics are not known,
but are required for explanation and prediction of observed carcinogenic
effects.

     In recent years, concerns have been raised that environmental pollution
may affect skin cancer incidence by affecting the stratospheric ozone shield.
The ozone shield removes more than 99% of ultraviolet radiation of wave-
lengths less than 320 nm from the incoming sun's rays.  (Emmett, 1979). The
ozone is formed by the interaction of oxygen and wavelengths shorter than
240 nm.  The resultant ozone then screens out longer ultraviolet wavelengths
up to about 320 nm. The ozone is naturally destroyed by various mechanisms
                                    664

-------
including combination with OH ions from water and destruction catalyzed by
certain gases such as NO.  The stratosphere  differs from the lower atmosphere
(trophosphere) in that it has a very low turnover rate so that any pollution
accumulates and is very slowly removed.

     Man could, theoretically, reduce the stratospheric ozone concentration by
a variety of ways including thermonuclear explosions, exhaust gases from
SST's, and halomethanes used as spray propellants. Predictions of the degree
to which these activities will increase the actual frequency are difficult to
make because they require estimation of several relationships which are not
at present well known, namely the degree to which these man-made activities
will increase the amount of ambient uv radiation, and the degree to which these
increases will be effective in producing cancer.   (We already have a glimpse of
the problems posed in elucidating the degree to which UVB increases will result
in skin cancer, for an appreciation of the formidable problems associated with
assessing the degree of ozone depletion, see J. L. Fox, Chem and Eng. News,
Oct 15, 1979, pp 25-35).  One estimate  (T. H. Maugh II, Science, 1980, 207:
394-395) projects a 16% reduction in stratospheric ozone and consequent 44%
increase in UVB radiation reaching the earth's surface.

B.   Objectives of this Work;

     The overall objectives addressed by these studies are of a two-pronged
nature:  they are concerned with (1) the effects of superimposing a constant
or varying band of UVA  radiation on the action of UVB light ("photoaugmentation
effect") and  (2) the effect of depleting the ozone layer.  To accomplish this
goal, we irradiated albino hairless mice with "solar-simulating" ultraviolet
light and with "monochromatic" light at selected wavelengths in the UVA and
UVB regions.  Solar-simulating UV was obtained from a 1600 watt   ozone free
Xenon arc with its emission passed through a 45° dichroic mirror and a 2 mm
Corning #9863 filter.  This combination limits the spectral output to the
range 290-400 nm.  The #9863 filter does not significantly affect the short
wavelength cutoff, which is limited to   290 nm by the impure quartz used for
ozone-free emission.  "Monochromatic" light (band width 10 nm)  is isolated
from a 200 watt Xenon-mercury arc (Canrad Hanovia) by means of a Bausch and
Lomb high intensity grating monochromator.  To simulate  varying thicknesses
of stratospheric ozone, we vary the amount of UVB while keeping that of UVA
essentially constant.  This is accomplished by filtering the solar simulating
(290-400 nm) radiation through a series of Schott WG-320 filters ranging
from 0.5 to 4.0 nm.< These are sharp cut-off filters whose characteristics
are given in Table 1.  UVA light is obtained by means of filtering 290-400 nm
solar simulating light through a Schott WG-345 filter, which cuts off essen-
tially all light below 324 nm (Table 1).  Absorption data was obtained using
a GCA-McPherson Absorption Spectrophotometer.
                                    665

-------
                                   TABLE 1

                 Characteristics of Schott Filters used to Vary
Amount of
Filter thickness (mm)
WG-320 0.5
1.0
2.0
3.0
4.0
WG-345 2.0
UVB Output from Solar Simulating Lamp.
fy cut
310
313
318
321
323
346
^ 50%
305
308
313
316
318
341
7go%
296
300
305
307
309
330
^ 99%
292
295
300
303
304
324
*                       <\
 All wavelengths in nm. ^ cut refers to approximate wavelength where filter
 starts to cut off. ^ 50%,*700% and'"X99% refer to wavelengths where 50,
 90, and 99% of the incident light is absorbed.

C.   Experimental:

     I. Relative Carcinogenicity of "Monochromatic" Bands of UV Radiation

        Experimental: Groups of five animals were exposed to "monochromatic"
light  (10 nm bandpass) at 280, 301, 307, 313, and 366 nm.  For the first
three wavelengths, the dose was 8.9 J-cm~2  for 5 days,  and then decreased
to 4.5 J'crrT2.  For the 366 nm irradiation, the dose was 75 J-cm~2.  The
same test sites were exposed to the same light for five consecutive days per
week.  Observations were made on a daily basis, and responses were graded on
a 6-point scale as follows:

        EI - Mild to moderate macular erythema

        £2 - Intense macular erythema

        1+ - Light scaling with or without accompanying erythema

        2+ - Firm, scaling, palpable keratosis

        3+ - Raised, palpable keratotic plaque, corresponding to early malignant
changes as defined by Epstein et al (1969)

        4+ - A papilloma or tumor corresponding to extensive malignant develop-
ment.
                                      666

-------
     In a companion experiment,  irradiation at the  above wavelengths was promptly
followed by 6.5 J.cnr2 UVA light.   For the 280 nm and 313 nm experiments, the
fluence was 0.2 J.cm~2 day"I,  for both augmentative and non-augmentative condi-
tions.;  At 301 and at 307  nm,  where the action is much greater, the fluence was
0.02 J.cm~2 day"1 and 0.01 J.cm~2, dayl for 301 and 307 nm radiation respectively.
In the latter cases, the output  from the monochromator  was attenuated by means
of wire screens which had  been calibrated with the  thermopile.  Precancerous
responses were graded as described above.  In cases where the mice burned the
amount of energy required  to produce each of these  two types of "action" were
recorded; their reciprocals were plotted as a function of excitation wavelength
to yield preliminary "action spectra" for precancerous changes (minimal erythema)
and burning (figures 1 and 2 respectively).  In cases where burning resulted,
no further irradiation was carried out.  It is doubtful that either heat generated
from the lamp or straight  infrared radiation are the primary causes of burning
since we observe a pronounced  wavelength effect for the latter phenomenon (see  below)

     At selected intervals, mice were sacrificed and skin biopsies of the ir-
radiated areas are taken.   Histological preparations (see below)  were monitored
for morphologic and biochemical  changes.

     Results:
     Precancerous lesions were induced at 280, 301, 307, and 313 nm.  These same wave-
  lengths also produced burning. No precancerous lesions were seen at 366 nm despite
  the much higher irradiation dosage.  Both burning and 5- responses appear to peak at
  307.  The shape of the two action spectra appear to differ, especially on the short wave-
  length side where the effects seem to fall off more rapidly in the case of the burning.
  Similar effects have been noted by Urbach,  1969. These results also suggest that UVA
  can photoaugment the effect of UVB, even though it is a poor carcinogen by itself.

     Preliminary investigations reveal the following:  H and E preparations show that irradia-
  tion produces marked epidermal thickening and increase keratosis.  These changes are
  wavelength-dependent and they seem to parallel the clinical responses. However, there
  were some ambiguities, which probably arose from difficulties  in obtaining good  biopsies
  due to the small size of the monochromator-induced tumor.  This was especially in the
  case where UVA radiation from the solar-simulator was used in conjunction with the
  monochromator.

  Comment;  These data, though preliminary in nature and, therefore, subject to some
  modification, nevertheless illustrate some important points. First, they suggest  that
  UV carcinogenesis may be preferentially produced by light near 307 nm. It is difficult
  to assess the chromophore(s) responsible for precancerous lesions and/or burning from
  this data, especially as the observed action spectrum may be considerably distorted from
  the true chromophore absorption, in the short wavelength region by internal scattering
  and absorption by non-active chromophores. By the same token, it is difficult at this
  juncture to ascertain whether the apparent differences between the action spectrum
  for skin cancer and the action spectrum for  burning arise from differential action of
  endogenous chromophores or whether the changes arise from early  chemical modifications
  which affect the subsequent absorption properties of the skin for action which occurs
  at a later stage (burning).  Further mechanistic data is needed  to help resolve these questions.
                                        667

-------
     II. Solar-Simulator Experiments

         Experimental:   While the results of the preceding experiment indicate
that shorter wavelengths do not necessarily increase carcinogenesis, we feel
that data using monochromatic light is "artificial" and cannot be extrapolated
to environmental conditions, since sunlight emission is a polychromatic
continuum.  We, therefore, conducted experiments with the 1600 watt solar-
simulator as described above as well as with broad band "monochromatic" radia-
tion of UVB centered at 300 nm (half value band width = 20 nm).  In order to
enable a meaningful comparison between the two sources and to gauge the
magnitude of UVA in augmenting the UVB effects, an "effective" dose was cal-
culated for both sources by convoluting the action spectrum determined above with
the spectral distribution of the emitting source (Willis et al, in preparation).
Experiments involving both sources were carried out under conditions of a)
constant irradiation (  f" 1 MED) and b) a regimen whereby the dose was increased
at 20% increments (of the standing dose i.e. 1 MED) after every five days of
irradiation.  Additional experiments were carried out with broad band UVA
(solar-simulating radiation filtered through a Schott WG 345 cut-off filter).
Clinical and histological responses were graded as previously described.  Twenty
animals were used in each experiment.

     Results:
                                                           2
     a) UVB radiation.   After 30 days (total dose 1.62 J/cm , effective dose
1.44 J/cm2) 75% of the animals had 1+ response, 20% had 2+, and 5% had 3+
responses.  At the 3+ stage, histopathological changes were compatible with
early squamous cell carcinoma, or carcinoma in situ.  There were cells with
large and bizarre nuclei, as well as "rounding off" of cells, with apparent
loss of  desmosomal  attachments.  In many (but not all regions, the base-
ment membrane appeared to be intact.  Since the epidermis becomes thickened
on exposure to UV radiation, we felt that it should be possible to incre-
mentally increase the dose without burning the skin.  This turns out to be
the case.  For equivalent effective doses, the latter regimen results in
increased clinical and histological severity of response.  An example is given
in Table II.

         Table II - Comparison of Constant vs. Incrementally Increased
         Doses of UVB Radiation.

                   Eff. Dose  (J/cm2)     1+	2+	3+
Constant Dose
Increased Dose
1.24
1.23
15
4
4
12
1
4
      (See "Experimental" for detail of dosage, source, etc.)

     b) UVA Radiation:  Daily exposures of 62 J/cm2 were given for 30 days
at 5 days/week.  By day 10, 65% of the animals exhibited 1+ responses, but
this regressed until, by day 30, only EI(minimal erythema) was present.


                                     668

-------
     c)  Effect of UVA 4- UVBs  Twenty mice were irradiated with a constant daily
dose of 9.0 J/cm2 C^—-1 MED) effective dose =» 0.113 J/cm2).  At the end of 30
days (effective 3.40 J/cm2, 80% of the mice had developed 3+ reactions, as
compared to 5% for UVB alone.  The effect of UVA in augmenting precancerous
and carcinogenic effects of UVB is further indicated in Table III.

        Table III - Photoaugmentation of UVB by UVA Radiation at
        Constant Dosa

               Eff.  Dose  (J/cm2)        l-t-           2+            3+
UVB
UVB -i
1- UVA
1.44
1.59
15
2
4
4
1
14
Histopathologic changes paralleled those occurring for UVB radiation alone,
but were more pronounced.

     As was the case for UVB radiation, when the doses were increased at 20%
increments every fifth day, the response at a given dose was more severe than
for the constant regimen.  After 18 days, three animals  (15) had developed
4+ responses  (advanced tumors) after 18 days, and the number of 4+ reactions
continued to increase up to day 30.  At equivalent effective doses of UVB, no
animals had developed 4+ reactions.

     Histopathologic changes in the 4+ mice showed atypical mitotic figures,
hyperplasia and hyperchromasia of cellular nuclei, disintegration of inter-
cellular bridges, and increasing variability in all sizes.  Several specimens
obtained 4-6 weeks after irradiation revealed so-called  ''spindle-celled"
squamous cell carcinomas.  This type of tumor closely rsembled a fibrosarcoma
with spindle-shaped cells extending from the epidermis to deep into the dermis.
This unexpected histopathological finding is of extreme  interest, since it is
the typ© reported to occur in areas of radiodermatitis in humans, and is
regarded as a relatively rare Grade-4 highly malignant metastatizing form of
squamous cell cancer.

     Comment;   These results indicate two major points: Firstly, time-dose
reciprocity does not exist, as evidenced by comparison of responses at equi-
valent effective doses of either UVB or UVA + UVB radiation (Table II).
Secondly, despite the relatively poor carcinogenic effectiveness of UVA ra-
diation, it can augment the carcinogenic effects of UVB, as evidenced by the
comparison in Table III.  This demonstrates that UVA effects must be explicitly
considered when effects such as ozone depletion  (see below) are considered.

     III.  Simulation of Qaone Depletion

     Experimental:  Schott WG-320 filters (Table I) were used in conjunction
with the solar-simulator, as described above, to produce conditions aimed at
mimicing th© effect of varying ozone layer thickness.  The spectral distribu-
tion of th© solar-simulator under the various conditions was measured.
Th© wavelength at which the output is down to 1% of its  value at
                                     669

-------
340 (plateau region) is given in Table IV.
bitrarily defined as a cutoff wavelength.
                          This wavelength is somewhat ar-
Filter  (mm)
                                  TABLE IV

              Cutoff Wavelength of Solar-Simulator/W5 320 Filter

              Combination - Minimal Erythemal Dose  (MED) for Each

              System.
Cutoff Wavelength (run)
MED (J/cm2)
None
0.5
1.0
2.0
301.0
303.5
305.5
307.5
7.05
13.4
17.3
22.4
    3.0
         308.5
 25.7
 Wavelength where output is 1% of that at 340 nm.

    Groups of 10 animals were exposed to the light from the solar-simulator
which had been filtered with the various Schott filters (including no filter).
Minimal erythemal doses  (MED) were determined for each set of conditions.
Following MED determination  (Table IV) two sets of experiments were carried
out, using the technique of incremental increases described above.  Firstly,
equal doses of radiation (0.9 MED for the "no filter" condition) were used
for each filter combination.  In the second set of experiments, doses equi-
valent to equal responses   (i.e. 0.9 MED for each filter combination) were
used.  Responses were evaluated as previously described.

    Results:

    a) Constant Dose: Only the "no filter" and the 0.5 mm filter combinations
produced lasting precancerous changes beyond the £2 stage.  These changes
were along the lines of those described previously.  The effects due to 0,
2.0 and 3.0 mm combinations eventually regressed back to the E^ stage or
they disappeared altogether.

    b) Dose Equivalent to Constant Biologic Effect: Much different results
were obtained for two groups (of 10 and 8 mice respectively per filter
combination) which were irradiated with light equivalent to 0.9 MED at each
filter combination.  In this case, the "no filter" group clearly progressed
to the 3+ (early cancer) stage after 30 irradiation days, whereas the remain-
ing mice all exhibited effects which were difficult to distinguish from one
another (mostly 2+ responses, some 3+ with or without moderate burning).
                                670

-------
     Comment;  These experiments disclose several salient features.  Firstly,
they again attest to the superiority of the UVB component in causing pre-
cancerous and cancerous changes.  The results again point out the non-
reciprocal time-dose relationship.  Indeed, it appears that in choosing our
experimental conditions, we may have chosen two extreme cases - one (equal
dose) in which little or no net response is observed and the other (equal
response) in which the precancerous response in each system (obscured some-
what by burning)  are relatively acute and difficult to tell apart.  In order
to obtain data which would allow a more quantitative distribution of these
wavelength effects, it appears that some intermediate condition(s) for irra-
diation will have to be carried out.  Conversely, these results also imply
that observed increases in skin cancer may not be related to ozone depletion
in a straight forward way: the effects of dose, angle, environmental and other
factors previously mentioned will all influence the results inasmuch as they
are determinants  of solar dose and spectral distribution as functions of
time.

     General Prognosis; The above results indicate the complex nature of the
problems of explaining and predicting the effect of broad band solar UV light
in producing carcinogenesis, and of predicting the effects of ozone depletion
on these processes.  The combined effect of component wavelengths in poly-
chromatic UVA + UVB light is different from the sum of its individual components,
and the augmentation process must be better understood.  Hence, reliable action
spectra for UVB carcinogenesis in the presence and absence of UVA are needed,
and such work is  projected for the future.  Since time-dose reciprocity does
not hold, the provision of a much more detailed picture of the dose-response
characteristics under different filtering conditions (i.e. "different ozone
concentrations")  is also projected.  The overall aim is to eventually explain
and predict the observed results on a more basic level.  Therefore, future
studies will involve the expansion of the scope of histological,  biochemical
experiments and to conduct mechanistic photochemical studies on the molecular
level. /-
                                    671

-------
                               REFERENCES
1. Emmett, E.A. (1979):  "Global Environmental and Genetic Aspects of Skin
   Cancer Including Melanomas". In Photobiology Course Syllabus, 1979
   Amer Acad Derm Mtg.

2. Epstein, J.H, Fukuyama K, and Dobson R.L. (1969):  "Ultraviolet Light
   Carcinogenesis" in "The Biologic Effects of Ultraviolet Radiation" F.
   Urbach, Ed. Pergammon Press 551-568.

3. Fox, J.L.  (1979) "Atmospheric Ozone  Issue Looms Again" (1979) Chem and
   Eng News, Oct.  15, 1979 25-35.

4. Maugh II, T.H.   (1980) "Ozone Depletion Would Have Dire Effects".  Science
   207:394-395.

5. Forbes, P.O., Davies  R.E., and Urbach, F (1979). "Aging, Environmental
   Influences and Photocarcinogenesis. J. Invest Derm 73_: 131-134.

6. Willis I, Kligman A.M. and Epstein J.H. (1973): "The Effects of Long
   Ultraviolet Light on  Skin - Phot©protective or Photoaugmentative?"
   J. Invest. Derm. 59:416-420.
                                672

-------
                  EPA/NCI Grant # R806066010
                  "In vivo Analysis of UV-B Induced Photooxidations"
                  Norman I. Krinsky, Ph.D., Principal Investigator
                  Period: 8/28/78 - 4/28/80

    The basic objective of this project is an attempt to elucidate the
mechanism of UV-B induced photooxidations  occuring in vivo, utilizing
cholesterol synthesized in vivo  as a substrate for photooxidative
reactions.  By analyzing the cholesterol oxidation products formed when
mice are irradiated with UV-B, it is our intention to be able to describe
the types of photooxidations initiated by UV-B in vivo.

    During the first phase of this project (8/28/78 - 8/27/79) we concen-
trated on the development of a suitable HPLC system for the rapid and
quantitative resolution of cholesterol and of some of the expected oxida-
tion products.  We have used an Altex 310/50 HPLC with a variable wavelength
detector coupled to a reverse-phase column for the separation and analysis.
The following compounds have been prepared from cholestrol to serve as
chromatographic standards:                      Retention time (min)
         cholesterol                                    11.5
         cholesterol-5a, 6 a-epoxide
         cholesterol-SB, 6 B-epoxide
         3B-hydroxycholest-6-ene-5a-hydroperoxide        7 _ Q
         7-ct-hydroperoxy cholesterol                     6.0
         7 a-hydroxy cholesterol
         7 B-hydroxy cholesterol
         7-B-hydroperoxy cholesterol                     9^
         3B, 5a, 6B-trihydroxycholesterol
         7-keto cholesterol
    In addition, we have also prepared several of these compounds from
l^C-cholesterol to serve as internal standards and, in particular, to
locate some of these steroid derivatives that do not have an adequate
absorption at 205 nm to permit detection in our variable wave length
detector.  The l^C-steroids synthesized in our laboratory  from  ^C-
cholesterol include:  !^C-cholesterol-5a, 6a-epoxide, ^C-cholesterol-SB,
6B-epoxide and .33, 5a, 6B-trihydroxy cholesterol

     The retention times for several of these standards are listed above,
using acetonitrile as the developing solvent.

     For the purpose of providing a radioactive substrate for the UV-B
photooxidations, we have utilized RS-(5- H)-mevalonic and (*MVA) injected
intraperitoneally into mice.  Although there are no reports in mice,  *MVA  has been
utilized in rats and found to give significant incorporation into skin
steroids within 30 min (J. Lip. Res. 2:344, 1961).  We have now been routinely
injecting 0.25 u curies of *MVA into each experimental animal in order to
obtain an adequate quantity of  cholesterol in the skin steroids.  Our
present protocol consists of injecting *MVA and incubating in vivo for 60 min,
either in darkness or exposed to 2 GTE Sylvania fluorescent lamps, F40T12/2021
at a distance of 15 cm.

     The experimental animals are nude mice, Skh :hr-l, obtained from Temple
University Skin and Cancer Animal Hospital, Philadelphia, PA.   Following in
vivo incubation, the skin is removed and the lipids are extracted into
chloroform/methanol using a Brinkmann Polytron Homogenizer.  The entire


                                 673

-------
process is carried out in a nitrogen-filled chamber under dim light to avoid
spurious oxidation of the  cholesterol.   After suitable preparation, the
steroids are separated by preparative thin-layer chromatography to yield a
"steroid" fraction.              To remove polar lipids the TLC plate is
irrigated 3 times.  This material is then subjected to HPLC analysis, with
the effluent collected directly in scintillation vials via a Gilson FL-100
fractionator.

     Our results are still preliminary,  but do indicate that the major radio-
active skin steroid observed 1 hr after  IP administration of *MVA is choles-
terol.  There are also smaller amounts of two other radioactive compounds,
one more polar and the other less polar  than cholesterol.  The more polar
compound has been tentatively identified as the 33,5a,6g triol.  As yet,
we have not been able to identify the less polar compound, but we may be
dealing with a precursor of cholesterol  such as 7-dehydro cholesterol.

    In the extracts of animals exposed to UV-B, we find increased amounts and
types of more polar radioactive peaks.  The precise characterization of these
peaks is difficult,  due to the very low  levels of radioactivity observed and
a problem in reproducibility.   We are quite concerned that some of the changes
we are observing may be due to the UV-B  catalyzed formation of vitamin 03, via
pre-vitamin D~, from 7-dehydrocholesterol, as recently described by Holick
et al. (Biochem. 18, 1003-1979).   We are in the process of obtaining some radio-
active standards of the compounds described by Holick et al. to compare to our
UV-B induced radioactive peaks.

    Our remaining effort in this project period (8/28/79 - 8/27/80) will be
to characterize these products of UV-B photooxidation of *cholesterol.
                                674

-------
                                       Discussion


Dr. Cameron, NCI:  I think this question has been raised before, but how much work
has been done with skin painting of these varoius strains? I also have another naive
question. What is a sencar mouse? Is that a hairless, too?

Dr. Orme, NCI:  No.  The sencar mouse is one that was developed by Torn Slaga at
Oak Ridge.   What they did over a  number of generations was  select  mice that
developed tumors after treatment  with DMBA-TPA.   Since the  tumors were not
lethal, they could breed tumorigen bearing animals, and, in that manner, developed a
population with high susceptibility to DMBA-TPA. They also developed another line,
which was supposedly the resistant counterpart.

I may not have that story completely straight because I have obtained it verbally and
have never seen any  publications on the sencar mice, but Tom Slaga has described it
as a very useful system, not only because of its high susceptibility for screening of
skin carcinogens and looking at the promotion phenomenon, but also because he has
explanted skin cells from the sencar mouse and has an in vitro transformation  system
which he says is remarkably similar to the in vivo system as far as its  response to a
spectrum of chemicals is concerned.

It  has always been puzzling to me why hairless mice were not picked up for skin
painting studies, but to my knowledge they have never been used extensively for that.

Dr. Cameron, NCI: Would you propose that  that be an avenue for further research?

Dr. Orme, NCI:  Yes.  Dr. Fred Urbach at Temple has proposed  to do some screening
studies  with  hairless  mice.    He  wants  very  badly  to start screening for skin
carcinogens using the HRA strain. That is a possible use.

Dr. Cameron, NCI:   That  would seem  to  me to have  an occupational smattering,
would it not? There would be different exposures for different occupations.

Dr. Gass, OSHA:  I have one question - a kind of fun question. Do juvenile hair coats
in susceptible strains offer any protection against the UV range?

Dr. Orme, NCI: I do not know if anybody has done that experiment.

Dr. Gass, OSHA:  It would be a neat control.

Dr. Orme, NCI:  It would be a  hard one to  do, because the juvenile hair coat  is gone
so fast.

Dr. Gass, OSHA:  It lasts about six days?

Dr. Cameron, NCI: Their juvenile hair coat lasts about ten days, doesn't it?

Dr. Orme, NCI: I am not really sure.

Dr. Gass, OSHA:  They do  not get  a hair  coat; that  is the problem.  If hair does
protect against UV, would the juvenile hair coat protect?  That is the question.
                                    675

-------
Dr. Orme, NCI:   I  think the  protection with  the  hair is just  a neutral density
protection. It is a neutral density filter protection.

Dr. Gass, OSHA: That is why I would use it for a control.

Dr. Orme, NCI: There is an  interesting phenomenon which occurs in the C3H mouse,
which is being  used  by  Margaret Kripke in  her  immunology studies, where she has
shown that there is a UV-B induced antigen on skin tumors which develop after UVB
irradiation.  These are all fibrosarcomas.  I  do not know, and I do not think anybody
knows, why irradiation of some mice leads to fibrosarcomas and irradiaiton of others
leads  uniformly to squamous  cell carcinomas.  Hairless mice, for  the  most part,
respond to UV irradiation  with squamous cell  carcinomas.   Haired mice yield a
variety of tumors.
                                     676

-------
             SUPPORT SERVICES FOR RADIATION RISK ESTIMATION
             FOLLOWING RADIOTHERAPY FOR CERVICAL CANCER

                            John D. Boice, Jr., Sc. D.
                       Field Studies and Statistics Program
                            National Cancer Institute

There is a lot of public, political  and scientific interest in evaluating the adverse
health effects of low-level radiation exposures.   This study is designed  to  quantify
the risks of low-level radiation exposures in a population of cervical cancer patients
that have received radium implants and external x-ray therapy.

The study has been misinterpreted as being clinically oriented in the sense of evalua-
ting the efficacy of the radiation therapy, but in fact the study goal is to determine
the radiation risks. The low-level exposure comes from  the radium that is implanted
in the uterus; the emitted gamma rays result in whole body exposures.  So, although
the doses to the pelvis  are very large, the doses to the stomach, lung and breast are
below or on the order of 100 rads of radiation.

This study  is the largest human radiation study ever conducted.  It is larger than the
study of  ankylosing spondylitics; it is larger than the  atomic bomb survivor study.  It
is complex in design.  What I am going to try to do this morning is give a broad sweep
of the aspects involved in the study.

First, we are starting off with what I call the clinical sample.  It involves 30 different
clinics in nine countries in the world and approximately 30,000 women have currently
been evaluated from  1960 to 1970.   The clinical sample is being continued with an
evaluation  from 1970 to 1980.  What is rather unique about these 30,000 women  is
that they have been monitored with blood studies every year for ten years. So they
have Seen  closely screened for hematopoietic effects, particularly leukemia. There
are eight clinics in the  United  States and 22 clinics in Europe at this time; three  in
England, five in Germany, four in Denmark, four in France, two in Austria, one  in
Czechoslovakia, one in Greece, and two in Italy.

To enhance the population size in order to be able to evaluate effects of low-level
radiation,  we have included cohort follow-up studies  in  population-based cancer
registries across  the world.   These include Connecticut in the  United States  and
cancer registries in Denmark, Sweden, Finland, Norway, five in England, Yugoslavia
and  several Canadian  registries.   The addition  of these  cancer  registries to  the
clinical  sample results in a population size  of several hundred thousand women that
received radiation therapy for cervical cancer. It is important  to note that cervical
cancer is a disease with a good survival rate. The radium implants were used mainly
for patients  with low stage disease, who  have  very  good survival,  and thus  the
probability of detecting late radiation effects is enhanced.

To obtain more information on  the characteristics of the patients, that is on medical
history  and  other cancer  risk  factors,  we  have  two  approaches.   One' is  by
case-control studies, which have been initiated already in the  United States. In the
SEER cancer registries in  the  United States and  in the cancer registries in Europe,
and Canada, we  are going to go back to the medical records and obtain information
about medical radiation exposures and other  cancer risk factors, such as reproductive
histories, smoking histories and alcohol consumption.  We are also planning to send
mail questionnaires to those persons who are still alive today.
                                    677

-------
As mentioned, the clinical study was initiated in 1960 by the World Health Organiza-
tion. We are continuing it and have completed a ten year analysis.  In 1960, radiation
was not as conclusive a human carcinogen as it is today, particularly with respect to
leukemia.  Only the atomic bomb survivors and  the study of patients with ankylosing
spondylitis in England had shown risks of leukemia that could be quantified; although
radiologists were reported to be at increased risk in the 1940s, the radiation  doses
could not be determined.

The WHO study  was initially set up to clearly show that radiation caused leukemia
and to quantify  the risks.  These  women received very  large average  doses to  the
bone marrow, 500 to 1,000  rads.  It was such  a large sample size  that a leukemia
excess should have been readily detectable.

The same advisory committee has remained with the  study since its initiation in  1960.
They include Dr. Brian  MacMahon, Chairman of the  Department of  Epidemiology at
Harvard, Dr. George Hutchinson, Professor of Epidemiology at Harvard, Dr. Herman
Lisco,  pathologist at Harvard Medical  School and formerly Secretary of the U.   N.
Scientific Committee in 1960 when the study  was  designed, and Dr. Jim Nixon, a
radiotherapist in  Memphis.   I am the epidemiologist  at NCI  coordinating  the
investigation.

I am going to just briefly present  the results of the  ten year follow-up.  These were
patients who were followed from 1960 to 1970, and no excess leukemia was observed.
This is one of the most surprising  things about  the study. There were 30,000 women
studied;  the doses were  enormous; there were actually  fourteen  leukemia  cases
observed and 15  were expected among these who received radiation. Essentially,  the
risk was  zero with regard to radiation-induced  leukemia.  This was startling at  the
time, particularly since 40 to 80 excess cases would have been expected based on  the
studies of the A-bomb survivors and the spondylitics.

It is hard to imagine the types of  biases that would  have gone into this study,  since
there was essentially a  10096 follow-up  and each patient  was clinically examined  for
up to ten years with blood studies. The average follow-up was about five years and
the person-years at risk was about 148,000.  Overall, there were 15 leukemias and
about 16 expected.  There did not seem to be any  unusual variation by any of  the
countries reporting leukemia cases.

Once  again, the  dose from external x-ray or from radium  to the total bone marrow
was on the order  of between  300 to 1,500 rads.   This is a very large dose.  There  did
not seem to be any variation by the type of treatment, whether they received radium
alone, or external beam therapy, or radium and external, or any radiation.

As  mentioned, what was extremely unusual was  that  based on the studies of  the
A-bomb survivors and the ankylosing spondylitics, there would have been expected to
be an excess of between 40 and 80  leukemias and there was none observed.

The study is consistent, however,  with  other studies or cervical cancer patients.  In
several studies, women  receiving large doses were not found to be at increased risk of
leukemia.

I might add that the historical study dealt with leukemia and lymphoma, whereas  the
expanded study is to look at the risk of  cancer  in organs  outside the pelvis.  That is,
to look for low-dose radiation effects.
                                    678

-------
There have been several possible explanations given as to why no radiation effect was
observed. One involves radiation factors. It is possible that the doses were so large
that the cells were killed and not transformed.  In other words, cell sterilization
occurred as opposed to cell  transformation.  There  is some  evidence from animal
experiments that this may be the case.  It  is  also possible that the women may not
have been followed long enough, although from the studies of A-bomb survivors and
the spondylitics the peak incidence of radiogenic leukemia is about four to six years.
The minimal latent period for leukemia is about two  years; the  incidence  peaks
around six years and then returns to normal levels around 20  to 25 years.  But it is
possible that older women may have a different latent period or temporal pattern  of
excess risk than these other two studies which involved men predominantly. There is
also a suggestion  that females may be  at  slightly lower risk of radiation induced
leukemia than males.

A recent animal study by Major and Mole at Harwell in England, shows a downturn in
the incidence of myeloid leukemia in mice at high radiation doses.  Dr. Upton has also
done similar experiments in the 1960s and showed similar effects. There seems to be
an effective dose  to cause leukemia of  around 200 to 300 rads.  At  the very high
doses, there seems to be a decrease in incidence with increasing dose.  This trend has
been used by some to explain the findings in the cervical cancer studies to suggest
that the dose may be so large that the cells  were killed and transformation could thus
not occur.

What is unusual about this analogy is that the cervical cancer patients received whole
body exposures, and although it is true that  the doses in  the pelvis were quite large,
on the order of 1,000 rads to the  bone marrow, the bone  marrow in the spine and the
chest cavity and the femur  and  other parts  of the  body also  received  substantial
radiation exposures that should have been leukemogenic.  It is difficult to explain why
these doses to the  non-pelvic bone marrow did not produce any excess leukemia.

Data from the A-bomb survivor study, suggests that latent period is a function of age
at exposure. If we follow the cervical cancer patients for another ten years or  so, we
may be  able to  find  a  slight excess  if this pattern is a realistic representation  of
latent periods for older women.  The  data  are particularly weak, but the A-bomb
survivor studies do suggest that the temporal pattern of excess incidence seems to be
related to age at the time of exposure. Cervical cancer patients have an average age
of around 50 to  55 years.  Perhaps if we follow them another ten years and have  20
years of follow up, we would be in the range where a leukemogenic effect might be
detectable.

Before  embarking  on such a very large study to  quantify the  risks of low-level
radiation, we felt  the need for as much  guidance as possible.  We convened a study
group consisting  of scientists in the  United States and in Europe.  They met  in
October 1979 and  people attending included  scientists from the International Agency
for Research on  Cancer in Lyon, from England, the United States, EURATOM, which
is the Atomic Energy Commission of the Common Market in Europe, epidemiologists
from  Italy, Germany, Finland and Norway.   They evaluated the study protocol and
design and concluded that it was scientifically valuable to conduct  such a  study.
Although we do not need any  more studies to show that radiation causes cancer, there
is a need to be able to quantify the risks at  various low levels and also to determine,
if possible, why various organs vary in sensitivity and why age may be an important
factor in determining subsequent risk.
                                 679

-------
The study group did raise  certain concerns though.  Can  we measure the radiation
doses in  these patients?  Can we follow patients in European  countries?  Are the
sample sizes large enough to detect low-level effects?  Can we obtain information on
other cancer  risk factors, such as smoking or breast cancer risk factors, in order to
control them  in the analysis?

The  study  group recommended three things.  One  is that  we conduct  feasibility
studies, getting information from hospital records regarding risk factors.  Two, that
we perform dosimetry studies.  Three, that we expand the study to include  the cancer
registries.  For the last six to  eight months, we  have been  embarking on these
feasibility studies in the United States and in Europe.

Regarding radiation dosimetry, we have a subcommittee consisting of physicists at
the M.  D. Anderson  Hospital in Houston, the Harvard Joint Center for Radiation
Therapy, and the Bureau of Radiological Health. Other  European physicists have also
been contacted.

The dosimetry  is complex.   It involves external  beam  therapy and internal radium.
We have made measurements now  on  patients  who are  today  being treated with
radium  and external beam  therapy.  We  have  made dosimetry  measurements  on
phantoms.  We are trying to characterize neutron exposures,  which come from high
energy betatron machines and very high energy accelerators.  We are also working on
computer simulations and questionnaire  designs to contact the physicists at all the
clinics and the cancer registries to characterize all the radiation exposures.

Our  preliminary  measurements  indicate that  we can  accurately  characterize  the
organ doses that specific  patients received.   The work was  done mainly at  M.  D.
Anderson.  For typical treatment modalities, the  range  of doses is as described. The
ovaries are in the pelvic area and they would receive very large doses, 1,400 to 1,700
rads.  The stomach  is much lower, 90 to 300; the pancreas a little lower; the lung still
lower, and the breast,  15 to 40 rads. This is in the dose  range of current interest,
under 100 rads for the organs outside the pelvic range.  The thyroid received 6 to  20
rads, which is also a low dose; the brain also receives a very low dose.

To refine these estimates  will probably  take us  another few years,  but  we do feel
confident that we can accurately characterize the doses.

Of the eight clinics in the United States,  we conducted  a study  of a ten percent
sample patients to  learn what information is  available  in the  medical records. This
included a sample of about  600 records.   Most of the important information is in fact
available in the records.  The birthdates are usually available. Even occupations for
these women is usually available.  Marital status  is always available.  The  number of
children  is there.  Age at first birth  is not recorded usually. Family history of  breast
cancer was given  70 percent of the  time.  The  smoking habits  and drinking  habits
were  recorded  in the medical records  66  to 56 percent  of the  time.   Regarding
information based on medical histories, the medical records had places for whether
second cancers occurred,  which was entered about 60 percent of the time.  This
indicates that we will have to do additional follow-up to characterize the population,
and this  is planned. The additional information was reasonable for many  of the risk
factors  relating to  breast cancer. The radium dose was available  99 percent  of the
time and external radiation therapy was  available 98 percent of the time.  Forty-two
percent of the sample had already died.
                                    680

-------
The  other aspect of  the  study concerns our  conducting cohort studies in tumor
registries across  the world.  Next week the study group is  meeting again and the
tumor  registries  from  the  nine or  ten  different countries  will be combining their
analyses of the risk of second  tumors  among cervical cancer patients that were
treated with radiation.

We have also  completed analyses using  data from the Connecticut Tumor Registry.
This is a particularly unique registry because it is the earliest in the world, starting in
1935.  It has the potential for long term follow-up, at least for about  40 years. This
is  the analysis of  about 6,000 patients  in  Connecticut and  shows  observed and
expected incidence of various cancer sites.  By itself, I do not think that any of the
numbers are exceptionally meaningful.  There is a lot  of variation in them.  A lot  of
factors are not controlled.  However,  when they are combined with the other  15
registries, certain patterns and  clues about radiation effects might be indicated. We
also plan to go back into the records in Connecticut and the tumor registries across
the  world  to  get detailed information  about  radiation  exposures  and  other risk
factors.  But  there  was a  40 percent risk overall  of  a second cancer among all the
patients who received radiation therapy. There were  339  second tumors observed  in
about 5,000 patients and 244 would have been expected based on the Connecticut age
and calendar year incidence rates.

Rectal cancer was in excess.  This might have been expected since it receives very
large doses.  The bladder was in excess,  and  the kidney. The bladder and the  kidney
are  particularly  interesting.   This excess  occurred  15 years after  exposure, and
therefore is more aligned  with the  expected latent period for solid tumors.  There
was a  very slight excess of stomach cancer, not very much.  This would be at the
lower dose. Ovaries had a slight excess.   There was also an excess in the body of the
uterus, the corpus.

Consistent with all the other studies, there  is no excess of leukemia, or only a very
slight one, eight observed  versus six expected.  There was an excess of lymphoma,
actually non-Hodgkin's lymphomas.  Lung cancer  was substantially in excess. This
was actually one of  the most startling findings, 48 lung cancers were observed and
nine expected.  It is questionable whether this is  a radiation effect since the great
bulk of the excess occurred within the first five or ten  years and this is not consistent
with what is  known  from most other  radiaton experiences.   In other studies  excess
risk due to lung  cancers occurs  10 to  15 years after exposure.   Cervical cancer
patients, perhaps, can be characterized by a low socio-economic status, perhaps they
smoke more than the general population, and these factors may be related in part  to
some of  the excess.

Pancreatic cancer was not in excess.  Interestingly, breast cancer was in significant
deficit.  This  is consistent with studies  of women  who have been treated for  benign
menstrual diseases with castrating doses of radiation.  It seems that in human studies,
if the ovaries  are ablated, the subsequent risk of breast cancer is reduced because  of
the cessation of ovarian activity.  This finding is consistent with this hypothesis.

There  were two bone  cancers.  They  were osteosarcomas and  both occurred in the
pelvic area. There was no excess of brain cancer.

Thus, I have presented a quick overview of the entire study.  There will be  another
three years of data collection  and another five years  before the  final results are
completed.
                                   681

-------
                                   Discussion
Dr. Gass, OSHA:  Do you have any evidence of leukopenia or thrombocytopenia from
the blood studies that would indicate that you  might have an overdose of radiation to
the hematopoietic system?  We know for example that we  can detect acute dose
delivery  at 25 rad and at  100 rads.  Surely, with these doses you might see  it.  A
related  question to that is  what was the average dose delivery internally  and then
externally? How did you relate that to your data?

Dr. Boice, NCI:  Details on the blood studies are not available.  The clinicians  in the
thirty clinics took the blood samples and  when something particularly  unusual was
found, the slide  was sent  to the  hematologist, Dr.  Moloney, at the  Peter  Brent
Brigham  Hospital in Boston. So,  the data were not systematically collected regarding
the blood studies, except  that  they were referred to  us when there  were severe
abnormalities.

Dr. Gass, OSHA:  There  is no  biochemical  dosimetry either, such as  uric acid or
purine or pyrimidine  metabolites?

Dr. Boice, NCI: No.

The second  question was  about the radiation doses.   Since there was no excess
leukemia, we did not go into exquisite detail  regarding dosimetry.  We did a first
approximation and the range in doses of milligram hours  of radium was perhaps 4,000
to 8,000  milligram hours.  The external doses were 2,000 to 10,000 rads to the pelvic
area.

The exposures varied by stage.  This was the  other comlexity.  If you had low stage
disease,  you received radium  only.   If you had higher stage disease, you  received
mainly external with some radium.  Currently we are looking into the dosimetry
problems in great detail.  But from the former study, there was not the need because
there was nothing to quantify.  There was no excess risk.

Dr. Cameron,  NCI:  I have  a political question. Why did the Japanese and Russians
not participate?   Where  they asked?  Especially with  Japan,  because  we  had  the
U.S.-Japanese cancer collaborative effort.

Dr. Boice, NCI:  Dr. MacMahon, who is on our steering committee, is also chairman
of an NCI/Russian Scientific Committee.  He  is going to Russia shortly  and  plans to
quiz them on whether they might be interested in collaborating with us.  I am not too
keen on the idea, however,  because one of the scientific requirments for  this study is
that good follow-up  on patients  is available as well as good medical records. This is
not generally the case, but if large  hospitals  with large numbers and good follow-up
can be found, we would welcome them into  the study.

The Japanese have not been approached with regard to this study.

The original population had  been selected from the  "Annual Report on the Results of
Treatment in Carcinoma of the  Uterus," Vol.  13, 1963.  All  hospitals treating large
numbers  of cervical cancer patients were invited to  participate.
                                   682

-------
Dr. Orme,  NCI:  I have a couple of naive questions to ask.  I presume that none of
these women are in a childbearing state. Were they sterilized for the treatment?

Dr. Boice,  NCI:  They are definitely sterilized after treatment  because of the high
radiation dose ablating the ovaries. Some women had their uterus removed prior to
radiation.  The average age was about  55 for cervical cancer patients in this study.
There was, however, a small percentage of young women with  cervical cancer.  I
think the youngest age was about 25.  There was a small percentage of women 25 to
40.

Dr. Gass, OSHA:  Did they use any thermoluminescent dosimetry? Did they use any
lithium or glass dosimetry on these people?

Dr. Boice,  NCI:  We  are  doing that now.  In the M. D.  Anderson dosimetry study
thermal luminescent dosimeters and  lithium fluoride were used.  These TLDs were
placed on patient's breasts and thyroid to estimate doses  from radium implants and
from external  beam therapy from a betatron and from cobalt-60. For units that  are
not used anymore, like orthovoltage  and  other  low energy x-ray  machines,  we
conducted  phantom studies.  We are also conducting computer simulation studies
regarding patient  treatment.

Dr. Orme, NCI:  I have one more question.  In the studies that you showed where the
dosimetry seems to have a maximum at the lower end and then it  goes off at the end,
what is the survival of the animals that get the high dose radiaiton? Is that sufficient
time for them  to develop leukemia?

Dr. Boice,  NCI:  Actually  that  is a very important question. It is the same question
that I asked Dr. Mole during a recent seminar.  Adjustment for competing causes of
risk was made in  their percentages. Although in several strains of mice 600 rads is a
lethal dose, in the particular strain used,  850 rads was the  critical lethal dose.  Thus
the decrease in leukemia risk in mice that received over 600 rads does not appear to
be related to mice killing but rather to cell killing.

Dr. Orme, NCI: Isn't it true that over 1,000 rads will kill the mice within two weeks?

Dr. Cameron, NCI:  I remember years back when I was at  Roswell Park I think they
used 600 rads  to  wipe out the spleen. They had to replant from  donor mice. I think
the dose they used there was 600.

Dr. Boice,  NCI: Six hundred rads is a lethal dose for most mice strains although 850
rads appears to be the lethal dose in the Major-Mole experiments.

Dr. Orme, NCI:  I do not even see how they can correct for that, if it is what I think
it is. Those mice  will not live more than two weeks.
Dr. Boice, NCI:  I think there is some survival. I think they call it the LD^Q, which is
probably around 400 or  500 in most strains.
Dr. Cameron, NCI;  This gentleman said the LD^g /•,« is about 480 to 600 depending
on the strain.

Dr. Gass, OSHA: That is right.  I say it is about 480 to 600 depending on the strain.
                                    683

-------
Dr. Boice,  NCI:   The LD^Q /™ means that half  would live longer than  30 days, and
then it would be the percent of those survivors that would develop leukemia.

Dr. Orme:  I am very suspicious about that.  Even if they survive more than 30 days,
they may all be dead at 50 or 90 days.  We are  still outside of the latent period for
leukmeia.

Dr. Cameron, NCI:  I would agree with you. The upper limit seems to be amazing.

Dr. Boice, NCI: I agree, but the authors appear to have a strain of special mice. It  is
also hard for me  to believe, however, that "cell-killing" totally  explains our lack of
an effect in the cervical cancer study.  We think further follow-up just in terms of
leukemia alone will be very interesting and informative in terms of radiation biology
and car ci no genesis in general.
                                     684

-------
FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:
    PROGRESS ON JOINT ENVIRONMENTAL AND
        OCCUPATIONAL CANCER STUDIES
            Thursday Morning, May 8
           CONCURRENT SESSION II:
           DATA BASES/MONITORING
            SESSION CHAIRPERSON

               Dr. George Simon
         Environmental Protection Agency
                         685

-------
             NCI/EPA/NIOSH Collaborative Workshop
       Progress on Joint Environmental  and  Occupational
                        Cancer  Studies

                          May 8, 1980


    CHEMICALS  FOUND  IN HUMAN  BIOLOGICAL MEDIA,  A DATA BASE
               Cindy Stroup, EPA Project Officer
           Office of Pesticides and Toxic Substances
                 Survey and Analysis  Division
                 Design and Development Branch


     For those not  familiar with this program,  the
establishment of a  data base on chemicals that have been
measured in human tissues  and body fluids, a little history may
be in order.

     In 1975, it was suggested by members of the Interagency
Collaborative Group on Environmental Carcinogenesis (ICGEC) and
staff at the EPA, that a more relevant exposure assessment to
toxic agents and environmental carcinogens could be achieved by
the acquisition of  data on human body burden.  Residue levels
of inorganic and organic compounds are reflections of exposures
to food, air,- and water contaminants and Pharmaceuticals.
These data are needed:

     in the identification of industrial chemicals of concern;

     to place chemicals in a priority order for agency
     attention;

     to complement  data already on hand such as environmental,
     human exposure, and materials balance information;

     in the setting of appropriate regulatory limitations;

     in the National Cancer Institute nomination and selection
     process for Carcinogenesis bioassay;

and for a number of other purposes.

     In 1976, the Task Group on Chemicals in Human Tissues was
established under the aegis of the ICGFC to 'investigate
potential activities in the area of human body burden.  Members
of the Task Group represent a broad spectrum of federal
agencies:
                                   680

-------
     Armed Forces Institute of Pathology
     Center for Disease Control
     Department of Agriculture
     Environmental Protection Agency
     Food and Drug Administration
     National Bureau of Standards
     National Cancer Institute
     National Center for Toxicological Research
     National Institute for Environmental Health Sciences
     National Institute for Occupational Safety and Health
     national Library of Medicine

     One of the first objectives identified by the Task Group
was the establishment of a body-burden data base.  To that end,
the Oak Ridge National Laboratory's Information Center Complex,
under contract to the National Library of Medicine, was
identified to perform a feasability study for computerized
searching of appropriate on-line data files for a selected list
of chemicals.  The retrieval of a large number of inappropriate
articles as well as the omission of many pertinent ones
revealed that available descriptors apparently were not
adequate for the efficient automated retrieval of body-burden
information from existing computer files.

     The consensus of the Task Group was that a manual
literature search might yield better results than a computer
search and so a second feasibility study for manual searching
was undertaken, this time by Tracor-Jitco, Inc., under contract
to the Office of Toxic Substances.  The successful completion
of the twenty-one month (January 1977 through September 1978)
search of forty-two selected journals confirmed the Task
Group's hypothesis and the manual approach was adopted.

     During this same time period, the NCI/EPA Cooperative
Program was evolving.  The establishment of a body-burden data
base was ultimately identified as an activity to be a part of
that interagency effort.  The level of funding for the first
four years of this six-year program is as follows:

     FY 79: $181,272 expended

     FY 80: $164,946 expended of planned $263,000;

     FY 81: $303,000 planned; and

     FY 82: $303,000 planned.

     This program is being accomplished for us by Oak Ridge
National Laboratory through an interagency agreement involving
the EPA and the Department of Energy.  Oak Ridge is under
contract to DOE.  The work began in September 1978 at the Oak
                                  687

-------
nidge national Laboratory's Information Center Complex Health
and Environmental Studies Program.

     In the first year and a half of this program, our efforts
have focused on:

     the identification of a core list of journals for routine
     manual searching retrospective to 1974;

     identification of alternate sources of body-burden
     information, such as US federal data bases and
     international contacts;

     efficient procedures for the identification of source
     documents, i.e., pertinent body-burden articles;

     data extraction techniques including appropriate keywords;

     efficient data storage and retrieval systems;

     a format for the tabular display of data in an annual
     publication; and

     preparation of the first annual publication of the data
     base.

     As appropriate data sources are identified through manual
searches or otherwise, a wide variety of pertinent information
is extracted and entered into the data base.  The major fields
used for data extraction are:

     type of data source (report, journal, data base);

     language (other than English);

     chemical name, as used in the data source;

     CAS preferred name and registry number;

     chemical formula;

     major use(s);

     immediate source or possible source of chemical;

     number of cases in the research;

     levels measured in biological media, mean and range;

     chemical analytical technique employed;

     route  of exposure;

-------
     half-life;

     pertinent demographic information;

     toxicity and health effects;

     pathology and morphology information;

     pertinent explanatory comments or caveats;  and

     keywords.

Data in these fields were entered  into the data  base only  if
available in the source document itself.

     At this time, we have completed the  retrospective  search
of our core list of approximately  84 journals.   Journals
continue to be added or deleted as appropriate.  So far, body-
burden data has been extracted from over  760 of  the 2,000-plus
documents produced by the manual search.  As of  May 1st, the
data base contained approximately  2,730 records  of information
on 432 chemicals that have been reported  in human tissues  and
body fluids.

     In the published version of the data base,  body-burden
information is presented in tabular format arranged
alphabetically by chemical, using  CAS preferred  names.
Generally, the CAS names are from  the 9th Collective Index, and
in some cases, names from both the 8th and 9th Collective
Indices^ are given.

     Physical and chemical data are generally not given in the
source documents but are obtained  from a  number  of standard
references and supplied in the published  version of the data
base for the convenience of the user.

     Within an individual chemical, records are  arranged in
alphabetical order by tissue.  A GENERAL  INFORMATION column
contains a variety of pertinent information such as
experimental design, demography, health effects, pathology,
morphology, toxicity, source, half-life,  and use.  Obviously,
it is impossible to include all details of the research in the
table.  In general, supporting information deemed important for
understanding the data presented will appear in  this column.
KEYWORDS also appear in this column.

     The first annual publication  of the  data base is now
available.  It has been our goal to make  this unique reference
tool as easy to use as we possibly can and toward that  end have
regular review and comments from members  of the  Task Group on
Chemicals in Human Tissues have been obtained.
                                    689

-------
      A number of indices allow the user to access  the  data
base in a variety of ways in addition to by chemical name.   The
table can be searched by author, corporate author,  tissue  and
keywords.  Cross-referenced chemical lists provide  assistance
in finding CAS preferred names from common names, or vice
versa, as needed.

     Future editions of the data base will contain  data
collected subsequent to this report.  However, the  indices will
be cumulative and so will refer to the entire series of tables
published to date.

     Preparation of the first publication was started last
October.  Emphasis on inputting recent literature and
significant research documents has resulted in a chronological
mix of articles from 1974 to the present.  The 1100-plus page
report contains 1,580 data records from approximately 440
source documents which reflects 244 chemicals measured  in human
biological media.

     Just a word about the distribution of this document.
Since, to our knowledge, this is the only comprehensive  body-
burden data collection of its kind, and due to the  keen
interest that has been expressed in human body burden,  we are
providing wide distribution of this first annual report.
Copies have been
sent to:

     all US medical schools;

     all US schools of public health;

     DHEW Committee to Coordinate Environmental and Related
     Programs;

     NCI Interagency Collaborative Group on Environmental
     Carcinogenesis; and

     ICGEC Task Group on Chemicals in Human Tissues.

     In addition, a number of government and private
investigators, both domestic and foreign, were provided
copies.  All recipients have been invited to, and I hope will,
respond with constructive suggestions that can serve to  make
future publications of the data base a more useful  and  valuable
resource.

     We have been overwhelmed with the unexpected amount of
human body-burden data and the mammoth job of retrieval,
extraction, input and publication.  This is the first such
effort, and good start, but there is certainly room for
improvement.
                                   600

-------
     We are looking  to  the  future with  some  ideas  or goals in
mind.  With routine  procedures  for data acquisition  now fairly
well established, the data  base has  been expanded  to include
pertinent supporting information on  food contaminants and  on
feral populations.

     Such supporting information is  available  from a number of
sources including existing  data bases at various  federal
agencies such as the US Department of Agriculture, the Food and
Drug Administration, and the Fish and Wildlife  Service.  The
pertinence of such information  as indicators of environmental
contamination and subsequent potential  human body  burden is
indisputable and the appropriate mechanisms with which to
identify, collect, and  incorporate supporting  data into the
human data base are  being actively examined.

     The investigation  of existing on-line data systems for
placement of the data base  has  been  underway for several
months.  In order to achieve maximum accessibility for
potential users, it  is  desirable to  place the  data base in as
many on-line systems as possible.  An offer by  the National
Library of Medicine  has been received.   It is  also anticipated
that the data base will ultimately become part  of  the Chemical
Substances Information  Network  currently under  development.   It
is expected that later  this year the data base  will  have
evolved to a point to allow it  to be placed on-line  at several
locations.

     A potential problem area  is the requests  for  computerized
searches of the data base.  If  the demand to date  for this
document, and it has just been  released, is any indication of
the level of interest in a  centralized  source  of body-burden
data, we are going to be swamped with these requests.

     The current mechanism  for  obtaining computerized searches
has been to contact  the EPA project  officer.   Requests for
computerized searches of the data base  have been received  from
Congressional and state government offices, and federal
scientists who usually  need the information immediately.

     Ideally, the development  of appropriate mechanisms to
allow computerized scanning of  the literature  for  efficient and
cost effective identification  of body-burden data  will
eventually eliminate the need  for manual searching.   As those
in the business of publishing  body-burden data  (e.g.,  journal
editors and abstractors) become sensitized to  the  need to
facilitate the identification  of such information, appropriate
techniques will be developed and implemented.

     Finally, concern has been  expressed about  the fact that
the data base contains  unvalidated data.  The  intention of this
program has never been  to provide a  validated  data base nor to
                                891

-------
obviate the need for users to ultimately refer to the original
source documents and to exercise good scientific judgment  in
the use of others'  research. The need that was recognized  by
the Task Group was  for a mechanism to readily identify a
collection of body-burden data, to easily find out what sort of
information exists  about a particular chemical or tissue.

     A suggestion has been offered that a mechanism be
identified for the  review and evaluation of data in the data
base.  Preliminary suggestions include the formation of a
scientific panel to rank the research data against a number of
criteria that may include quality control/quality assurance
measures, precision and accuracy of techniques, and statistical
design and evaluation of resultant data.

     Names and addresses of all persons interested in receiving
copies of the report may be submitted to the EPA project
officer.  Everyone  using the document or interested in this
program is encouraged to provide constructive criticisms,
suggestions for improvements, or any appropriate comments.
                                  692

-------
                     I  C G E C
  TASK   GROUP   ON   CHEMICALS
          IN   HUMAN   TISSUES
        ARMED FORCES INSTITUTE OF  PATHOLOGY
            CENTER FOR DISEASE CONTROL
             DEPARTMENT OF AGRICULTURE
          ENVIRONMENTAL PROTECTION AGENCY
           FOOD AND DRUG ADMINISTRATION
           NATIONAL BUREAU OF STANDARDS
             NATIONAL CANCER INSTITUTE
    NATIONAL CENTER FOR TOXICOLOGICAL RESEARCH
NATIONAL INSTITUTE FOR ENVIRONMENTAL HEALTH  SCIENCES
NATIONAL INSTITUTE FOR OCCUPATIONAL SAFETY AND  HEALTH
           NATIONAL LIBRARY OF MEDICINE
                               693

-------
                     BACKGROUND AND FINANCIAL DATA

o    FEASIBILITY STUDY FOR COMPUTERIZED APPROACH
          OAK RIDGE NATIONAL LABORATORY INFORMATION CENTER COMPLEX
          1976

o    FEASIBILITY STUDY FOR MANUAL APPROACH
          TRACOR-JITCO,  INC,,  1977-78

o    NATIONAL CANCER INSTITUTE/ENVIRONMENTAL PROTECTION AGENCY
          INTERAGENCY AGREEMENT FOR THE COLLABORATIVE PROGRAM, 1978

o    ENVIRONMENTAL PROTECTION  AGENCY/DEPARTMENT OF ENERGY
          INTERAGENCY AGREEMENT,  SEPTEMBER 1978
               OAK RIDGE NATIONAL LABORATORY
               INFORMATION CENTER COMPLEX
               HEALTH AND ENVIRONMENTAL STUDIES PROGRAM
               FY 79:     $181,272 EXPENDED
               FY 80:     $263,000 PLANNED
               FY 81:     $303,000 PLANNED
               FY 82:     $303,000 PLANNED
                                      694

-------
                     ACTIVITIES
0    CORE LIST OF JOURNALS FOR MANUAL  SEARCHING
0    ALTERNATE SOURCES OF BODY-BURDEN  INFORMATION
0    IDENTIFICATION OF SOURCE DOCUMENTS
0    DATA EXTRACTION
0    DATA STORAGE AND RETRIEVAL SYSTEMS
0    TABULAR DISPLAY OF DATA
0    PUBLICATION OF FIRST ANNUAL REPORT
                               695

-------
    DATA   EXTRACTION   FIELDS
0    TYPE OF DATA SOURCE (REPORT,  JOURNAL,  LETTER)



0    LANGUAGE (OTHER THAN ENGLISH)



0    CHEMICAL NAME,  AS USED IN THE DATA SOURCE



0    CAS PREFERRED NAME AND REGISTRY NUMBER



0    CHEMICAL FORMULA



o    MAJOR USE(S)



0    SOURCE



0    NUMBER OF CASES



0    LEVELS MEASURED IN BIOLOGICAL MEDIA,  MEAN,  RANGE



0    CHEMICAL ANALYTICAL TECHNIQUE EMPLOYED



0    ROUTE OF EXPOSURE



0    HALF-LIFE



0    PERTINENT DEMOGRAPHIC INFORMATION



0    TOXICITY, HEALTH EFFECTS



0    PATHOLOGY,  MORPHOLOGY INFORMATION



0    PERTINENT EXPLANATORY COMMENTS OR CAVEATS



0    KEYWORDS
                                 696

-------
            CURRENT   STATUS   OF   DATA   BASE
            0    RETROSPECTIVE SEARCH TO 1974 COMPLETE


            0    84 JOURNALS


            0    COLLECTED OVER 2,000 DOCUMENTS
CD

            0    EXTRACTED DATA FROM OVER 760 DOCUMENTS


            0    OVER 2,730 DATA RECORDS IN DATA BASE


            0    432 CHEMICALS REPRESENTED

-------
 ARSENIC
 7440-38-2
 As
 AiW 74,9216, MR 817  C AT 28 ATM, BP 613 C  (SUBLIMES), VP 1 MM HG AT 380 C, 10 MM HG AT
         EXPOSURE   ANALYTICAL   NUMBER
 TISSUE   ROUTE       METHOD    OF CASES
                     RANGE
MEAN
  GENERAL
INFORMATION
REFERENCE
 80     INHALATION
 URINE
CD
oo
ES     A) 41     A)  NOT GIVEN  A)  1,3  UG/L
       B) 30     B)  NOT GIVEN  B)  2,2  UG/L
       C) 23     C)  NOT GIVEN  C)  4,8  UG/L
       D) 30     D)  NOT GIVEN  D)  8,6  UG/L
                               GEOMETRIC
          A) CONTROLS       SMITH, ET AL
          B) LOW EXPOSURE
          C) MED EXPOSURE
          D) HIGH EXPOSURE
          CONCENTRATIONS
          EXPRESSED AS ELEMENTAL
          ARSENIC (III),
          COPPER SMELTER WORKERS
          AGES FROM 26-65, MEAN
          OF 46 YRS,
          ARSENIC:  METHYL RADICALS:
          URINE: COPPER: SMELTERS:
          OCCUPATIONAL HAZARDS:
          PARTICULATES

-------
  FIRST   AN-NUAL   REPORT
    TABLE   OF   CONTENTS

            ACKNOWLEDGEMENTS
              INTRODUCTION
              USERS'  GUIDE
  ABBREVIATIONS FOR ANALYTICAL METHODS
        REFERENCES FOR DATA BASE
REFERENCES FOR PHYSICAL AND CHEMICAL DATA

               APPENDICES:
            JOURNALS  SEARCHED
         CHEMICALS IN DATA BASE
     CROSS-REFERENCED CHEMICAL LISTS
  TISSUES AND BODY FLUIDS IN DATA  BASE

                INDICES:

                 AUTHOR
            CORPORATE AUTHOR
                 TISSUE
                 KEYWORD
         DIRECTORY OF CHEMICALS

            DATA BASE (TABLE)

                   699

-------
                  CONTENTS   OF   FIRST   ANNUAL   REPORT
                          o    DATA INPUT THROUGH OCTOBER 1979
                          o    DATA EXTRACTED FROM 440 DOCUMENTS

-vl
o

                          0    1,580 DATA RECORDS INPUT
                          0    244 CHEMICALS REPRESENTED

-------
      ANNUAL   REPORT   DISTRIBUTION
0    US MEDICAL SCHOOLS

0    US SCHOOLS OF PUBLIC HEALTH

o    DHEVI COMMITTEE TO COORDINATE ENVIRONMENTAL AND
          RELATED PROGRAMS

o    NCI" INTERAGENCY COLLABORATIVE GROUP ON  ENVIRONMENTAL
          CARCINOGENESIS

o    ICGEC TASK GROUP ON CHEMICALS IN HUMAN  TISSUES

0    PRIVATE INVESTIGATORS,  DOMESTIC AND FOREIGN
                              701

-------
                           IDEAS   AND   PLANS
           o    EXPANSION TO INCLUDE PERTINENT SUPPORTING DATA
           o    PLACING THE DATA BASE ON-LINE IN 1980


-------
                       CINDY   S T R 0 U P
                       DESIGN AND DEVELOPMENT BRANCH
                       SURVEY AND ANALYSIS DIVISION (TS-793)
                       OFFICE OF PESTICIDES AND Toxic SUBSTANCES
•si                     ENVIRONMENTAL PROTECTION AGENCY
S                     401 M STREET SW
                       WASHINGTON,  DC 20460
                       755-8294

-------
                                Discussion
Dr.  Weisburger (NCI):   On what basis was the decision made to include
certain chemicals within the list; for example, I find a whole page of
ascorbic acids and citric acids and compounds of that sort which are,
one would think,  usually ocurring in the well-nourished human body.  I
just wonder on what basis it was decided that a certain chemical should
be selected for this list.

Ms.  Stroup (EPA):  At this stage of the game, we made the decision not
to exclude any substance.  There is interest in the compounds you
mentioned by different factions of the scientific community.  For
instance, the American Institute on Nutrition has exhibited a lot of
interest in the data base, and in particular, in the dietary and nutri-
tional elements that we at EPA are not traditionally interested in.

This is the first go-round, and if we get many requests to exclude
certain substances, then perhaps we will.  At this point, yours is the
first such comment.

Dr. Weisburger (NCI):   The original question was on what basis the
compounds were selected for inclusion in this data base because there is
a whole page of listings on ascorbic acid and citric acid, as mentioned,
and some other compounds that are found ordinarily in humans.

My next comment is that there are really thousands of chemicals running
around in the human body, and if you include all of these, you know, is
that going to be something considered physiological, or is it supposed
to be a listing of abnormal chemicals found in the human body?

Ms. Stroup (EPA):  The objective of this program is to identify all
chemicals that have been measured in human biological media.  It is
conceivable that the scope may be narrowed in the future, but at this
time, there doesn't appear to be the need to do so.

Dr. Simon (EPA):   Are there any other questions?  Dr. Kraybill?

Dr. Kraybill (NCI):  Do you suppose that the development of this data
base could lead,  ultimately, to some projects — I do not say that you
necessarily support, or anyone will -- on a tissue bank, where actual
measurements will be made on autopsy tissue or surgical biopsy tissue?
I know this was brought up a couple of years ago.  What are your
thoughts on it?

Ms. Stroup (EPA):  I believe such an effort is ongoing out of George
Goldstein's shop at EPA and also  involves the National Bureau of
Standards.  I do no know the present state of that project, but there
certainly is a lot of interest in this area.
                                          704

-------
Mr. Feinstein:  Are you also collecting data on what subpopulations the
samples came from, like, for example, if it might have been an
agricultural worker?

Ms. Stroup (EPA):  It usually is.  Our data extractors have been
instructed to pull out all information deemed pertinent to interpreting
the analytic results.  This is, of course, dependent on the information
provided in the source document.

Mr. Feinstein:  Thank you.

Dr. Kraybill (NCI):  I am sorry, again.  I think you alluded to the
other data base, and maybe the rest of you in here are not familiar with
it.  The Department of Agriculture, by regulation and by law, is
required to annually make a surveillance of all the meat animals, and
you should be aware that they record data on bovine, sheep, swine, et
cetera.  This would include pesticides, drugs, environmental chemicals,
what have you.

The beauty of this is that these animals are just like people; they are
a monitor.  Years ago, at the Perrine, Florida lab in the pesticide
program down there, they were going out and getting wildlife, and I
guess even the armadillos, and recording what the body burden was for
these chemicals.

Another historical event in this whole thing was about 5 years ago.
They said, well, we are looking at chemicals in air and water, food and
diet, and this was brought up in the NCI testing program that .we might
as well be aware of what was in people; that is, at the cell level, like
we all know about DDT and DDE and PCB and PBB's and things of that sort.
That is, a direct insult right there at the cellular level.

So these monitors, I think, are very important because it should tie in
with the multi-media exposure, and if we can see what is in animals and
man, that will be most useful to us.

Dr. Simon (EPA):  Okay, one more question.

Mr. Cooper:  I was curious as to what the indexing criteria was for a
particular chemical.  Is it the name which was used by the author or is
it the correct name?  I noticed that you have benzene hexachloride, and
you use as a synonym the 6-chlorinated cyclohexane.  They are clearly
different chemicals, and yet they are listed as though they were a
single chemical.

Ms. Stroup (EPA):  I am sorry, I am not sure I heard all of your question,
Are you asking how we decided what to name a given chemical?

Dr. Cooper:  That is correct.

Ms. Stroup:  We use the compound name provided in the source document.
Do you find a conflict?
                                      705

-------
Dr.  Cooper (NCI):  Well, yes.  Our experience in PHS-149 is that very
frequently the person who writes the articles uses the wrong name.

Ms.  Stroup (EPA):  Yes, I know.

Dr.  Cooper (NCI):  Do you attempt to correct this?

Ms.  Stroup (EPA):  We do not change what is represented in the source
document.  We do have a chemist involved in data extraction and she may
translate when she feels certain that what the author is calling a given
chemical is incorrect.  Then she will translate into the CAS preferred
name.  Otherwise, there are no changes from the source document.

Dr.  Kraybill  (NCI):  I think Dr. Cooper raised a very pertinent question
because there are many synonyms — what is the other word, acronyms,
synonyms?

Speaker:  Antonyms.

Dr.  Kraybill  (NCI):  Antonyms?  No, synonyms.  But if you look up a lot
of these chemicals, they may have eight or ten.   One safe way to do it
is to list it by chemical abstract name, right?  That would be ideal.

Ms.  Stroup (EPA):  That is what we do.

Dr.  Kraybill  (NCI):  Well, the other day I looked up a drug, Flagyl, and
I could not find it.  If I had known that Flagyl was a metronidazole,  I
would have found it very quickly, so I had to trace back and that is how
I came to get it.  So you need to sometimes list several names.

If these people that are abstracting the literature could go back and
list a couple of other names, that would be helpful.  I think that is
what you are getting at, aren't you John, having several recordings.

Dr.  Simon (EPA):  This is going to have to be the last comment because
we have a couple of other papers to get to.

Ms.  Stroup (EPA):  Apparently there is disagreement over the use of the
6-chlorinated cyclohexane as a synonym for benzene hexachloride.  If
that usage is incorrect, then it is also incorrect in the National
Library of Medicine's CHEMLINE which is the source of all  synonyms in
the data base.

Thank you.
                                       706

-------
  COMPLETION OF 1972 MORTALITY DATA, NCI/NCHS COOPERATIVE STUDY


                            Thomas Mason, Ph. D.
                          Field Studies and Statistics
                    Division of Cancer Cause and Prevention
                           National Cancer Institute
                             Bethesda, Maryland


!n reporting the vital statistics for 1972, the National  Center for Health
Statistics (NCHS) included information from a 50% sample  of certificates.
While this sample is argued to be representative of the total  United States,
it is not representative of individual  states or counties of the U.S.   Due
to the historic interest of the NCI in investigating cancer mortality at
the county-level, I became involved in attempting to convince  the NCHS to
code the missing 50% for 1972.

The NCHS adopted this coding procedure for 1972 in an attempt  to catch up,
and, thus, be able to present summary vital statistics data in a more timely
manner.  Due to the fact that the sample was not stratified by cause of
death, age, sex, and race, one has no way of knowing what percent of the
total number of events one has for a specific cause of death at the local
level.

With support from the Office of the Secretary, DHEW, we at NCI were charged
with supporting the coding of the missing 50%.  This translates to  983,108
death certificates.

At the present time, 73% of that 50% have all of the demographic data coded.
Included in this category are sex, race, age, county of death, state of
birth, residence, both state and county, and information  with  regard to
whether or not an autopsy was performed.

For the 983,108 certificates which require the assignment of an underlying
cause of death, 38% are now completed.  The project will  be completed in
another year and a half.  At that time these data will be made publicly
available.
                                    707

-------
                             Discussion

Unidentified Speaker:   Has occupation been included in this?

Dr.  Mason, NCI:   No.

Unidentified Speaker:   Why not?

Dr.  Mason, NCI:   They  do not code occupation,  and perhaps  that is a blessing
in disguise.  Several  states are trying it now with the National  Center for
Health Statistics'  data.  North Carolina i»s one.   I believe Michigan and
Minnesota are others  that are attempting it.

I believe more thought needs to be given to this  aspect.   We are  lobbying
for several additional states where we believe it would be desirable to
look at it.  It has been our experience, in taking the information from
the death certificate  and attempting to code occupation from it,  in densely
populated urban places, it is bad to be retired and it is  bad to  be a house-
wife, and beyond that  you do not learn much.

We are going to see if we cannot do something  about that,  and I think the
way to proceed is through additional education of the  persons who put the
information on the certificate.  Once it is there, it  is there, and there
is not much you can do about it.  I think that it is only  going to be through
some of these pilot studies and through some of this that  we will  be able  to
see it.

Yes we think it is important, but if you just  take it  the  way it  is right
at the moment, I do not believe you are going  to  get much  from it.   In some
senses, in some places, in areas with smaller  populations,  where  it is more
likely that the individual who signs off the certificate actually knew the
individual, you have  a better opportunity.   I  mean, we are fortunate to have,
still, a number of places in the United States that are relatively rural,
that are one-industry  places, where you can indeed get some information from
the certificate with  regard to occupation and  pursue it.   In many instances,
however, you cannot just simply take it from it.   It breaks down; you do not
get any good information from it.
                                     708

-------
                      National Mortality  Databases
                     for Environmental  Epidemiology
                                   by
                         Carol G.  Graves,  Ph.D.
                            Jacob Thomas,  M.A.
                         Charles   Poole,  M.P.H.
                               INTRODUCTION

     The subject of this talk is national mortality databases which are
being used for studies in environmental epidemiology.  Before these
databases are discussed, UPGRADE will be described.  UPGRADE is the
system with which the mortality databases are analyzed.   A brief
overview of the data available in the UPGRADE system will be presented
with the mortality data being described  in more detail.   The paper will
conclude with a glimpse at the data and  some specific questions and some
general analyses suggested by the data.
                                         709

-------
                                 UPGRADE

     UPGRADE is an analysis system developed for the  President's  Council
on Environmental Quality (CEQ).  The UPGRADE system has been  under
development for five years and has been funded by a number  of  federal
agencies including the National Cancer  Institute (NCI) and  the
Environmental Protection Agency (EPA).  An UPGRADE User's Manual  became
available in January, facilitating increased use of the system

     UPGRADE is interactive and English-language prompted.  The analyst
sits at a computer terminal and works with the data by responding to
straight-forward prompts.  Many prompts can be answered by  "yes"  or
"no".  Others require the analyst to select an option.  If  the proper
response is not obvious, available choices can be displayed by typing
"help".  The system  is designed to be used directly by the  analyst  and
is essentially self-teaching.

     A wide variety  of procedures are available in UPGRADE.   First  are
the necessary, but not so glamorous, data manipulation procedures such
as filtering, partitioning, listing, sorting, and transforming.   Then
there are graphics procedures such as bar charting, scatter plotting,
polygon plotting and regression plotting.  A wide variety of  plot
options are available.  These allow the user to tailor a graph and
produce a version suitable for presentation.  Another UPGRADE  procedure
is mapping.  Five different types of national and regional  maps can be
drawn.  Finally, UPGRADE contains statistical procedures such  as  a  basic
statistics report and regression.  Other statistical  procedures are
available tnrough UPGRADE'S interface with SAS, a well-known  and
sophisticated statistical package developed by SAS Institute,  Raleigh,
North Carolina.  The SAS job  is set up  interactively  in UPGRADE and runs
independently from but concurrently with UPGRADE.  A  SAS job  is usually
ready in a matter of seconds.
                        DATA  IN  THE  UPGRADE  SYSTEM

      Along  with  the  development of  the  analysis  system  has  been  the
 development of databases.  Because  UPGRADE  was developed  for  CEQ and
 because  CEQ uses UPGRADE for  analyses which go  into  the CEQ Annual
 Report on  Environmental Quality,  a  great  deal of environmental  data has
 been  accumulated for use with the system.   There are site-specific,
 time-series water  quality  data  from EPA's  STORET system and USGS's
 NASQAN database  along  with air  quality  data from EPA's  SAROAD system.

      The data most often used in epidemiological studies  are  contained
 in  the UPGRADE  Integrated  Database  (IDB).   These data are geographically
 defined, e.g., on  the  county- or state-level.   Other geographic  units
 proposed for  inclusion are SMSA's and census tracts.  The IDB has been
 designed so that data  from a  variety of sources  using a common access
 code, such  as the  FIPS state-county code,  can  be put into one dataset
 and analyzed  using UPGRADE.   In this way morbidity or mortality data
 from  the National  Center for  Health Statistics,  demographic and socio-
 economic data from the Bureau of the Census, and environmental data froi;,
                                      710

-------
EPA can be put together and analyzed.  For example, mortality could  be
graphed versus population, education, or an environmental factor.  Or a
regression could be run with the mortality rate as the dependent
variable and the other factors as independent variables.
                   NATIONAL MORTALITY DATA IN UPGRADE

     These are two types of mortality data available for analysis  in
UPGRADE.  The first consists of mortality rates for grouped causes of
death; the other consists of the number of deaths by individual four-
digit ICDA codes  in a number of age, race, and sex categories.

     The mortality rates for grouped causes were tabulated for  NCHS by
Herbert Sauer of the University of Missouri at Columbia.  For each
county mere are mortality rates for 70 causes of death.  Of these, 25
are rates for cancer such as stomach cancer, breast cancer, or
respiratory cancer.  Eight additional rates are broader categories of
cancer such as cancer of the genital organs  or non-respiratory
cancer.  These broader categories are sums of the more specific rates.

     The mortality rates are for two time periods.  There is one set of
rates calculated on deaths occurring from 1968 to 1972.  An earlier set
of rates is based on deaths occurring from 1959 to 1961.  Because  deaths
for these two time periods were classified according to different
versions of the  ICDA, comparability ratios calculated by NCHS are
available so that these two sets of rates can be compared.  A later set
of rates for the years 1973 to 1976 may be obtained for the system.

     These rates have been calculated separately for four race-sex
groups -- white  female, white male, black female, ancj black male.  Tne
rates are age-adjusted to both tne 1940 and the 1970 United States
population.  Hence, the set of mortality rates currently in UPGRADE is
age-adjusted and race-sex specific.  Also on hand, but not yet  included
in tne IDB, are  age-specific rates for eleven age groups (under 1, 1-4,
5-14, 15-24, 25-34, 35-44, 45-54, 55-64, 65-74, 75-84, and 85 plus).

     The second  type of mortality data available for use by UPGRADE has
been tabulated from NCHS detail mortality tapes.  These data consist of
counts of deaths rather than mortality rates.  For each county, 1970
population and 1970 deaths by  individual four-digit  ICDA code are
classified according to the following factors:
     Sex  - female, male
     Race - white, black, other
     Age  - five-year age groups
            (under 1, 1-4, 5-9, 10-14,  ... 80-84, 85+)

     These mortality data are  being expanded and in  another year  (by
June, 1981) should include data for  individual years from 1965  to  1976
(with the exception of 1972),  two sex groups, nine race classifications,
21 five-year age groups, and two autopsy categories.
                                      711

-------
                           EXAMPLE  UF  ANALYSIS

     In order to demonstrate a few UPGRADE capabilities and to
illustrate the mortality databases, two analyses will be  suggested.   The
firut dealing with cancer of the liver, gall bladder, etc. and  cirrhosis
of the liver in white males illustrates use of the mortality rates.   A
look at two causes of death for which the number of deaths are  few  --
malignant neoplasms of the eye and the thyroid gland -- illustrate  the
second type of mortality data, the raw death counts.

     Using the age-adjusted, race-sex specific mortality  rates  described
previously, the geographic distribution of cancer of the  liver, gall
bladder, etc. and cirrhosis of the liver in white males will be compared
using UPGRADE'S mapping capability.  Figure 1 shows the distribution  of
mortality rates for cancer of the liver in white males.   Included  in
this mortality category are all sections under ICDA codes 155 and  156.
These  include  cancer of the gall bladder and bile ducts  as well as
cancer of the liver.  Rates are per 1,000,000 and are calculated on
deatns occurring from 1968 to 1972.  For this map the range of  mortality
rates for all U.S. counties was divided into four equal parts.  Because
tne distribution of these rates is skewed with more rates falling on  the
lower end of the range, there are not an equal number of  counties  in
eech interval.  More counties fall in the lower shading intervals and
fewer fall in the higher intervals.  Figure 1 shows that  the high rates
for this cause are concentrated in the Rockies and plains states.

     Figure  2 presents the distribution of white male rates for
cirrnosis of the liver.  This cause includes all sections of ICOA code
571.  Again  these are 1968-72 rates per million population.  For this
map the range of the cirrhosis rates has been divided so  that one-third
of the counties fall into each shading interval.  Fran the map
concentrations of high rates occur in the west coast, the southwest and
tl ' northeast.  In Figure 3, this same cause  is mapped with a different
selection of shading intervals.   In this figure only ten  percent of
counties are shaded  in the darkest pattern and the concentration of
counties with the highest rates is in California,

     The maps suggest a number of questions each of which might  lead  to
further analyses.  Figure 1 suggests that cancer of the liver should  be
mapped separately from liver of the gall bladder and bile ducts to  see
it these causes have the same or different geographic distributions.lt
might  be expected that as the treatment for cirrhosis improves, the
mortality  from cirrhosis would decrease and the mortality from  liver
cancer might then increase because cancer would have the  latent period
to develop.  This could be  investigated by  looking  at mortality rates
for these causes from different time periods.  Perhaps since these  two
maps show  different  pictures, there are two different risk factors  for
these  diseases which are geographically distributed.  An  investigator
would  want to consider other factors such as  demographic  and
sociueconomic variables and alcohol consumption.  This type of  analysis
i^ possible  using UPGRADE'S Integrated Database which  allows the  user to
access variables from a variety of sources.
                                        712

-------
     Using the second type of mortality data  -- the  number  of  deaths  per
county by individual four-digit  ICDA code  —  deaths  from  all cancer
causes were tabulated by age, sex,  and race,  and  two rare causes  of
cancer were selected for investigation.  There were  352 deaths  due to
malignant neoplasms of the eye in the United  States  in 1970.   There were
1067 deaths from malignant neoplasms of the thyroid  gland in this
year.  Sex and race distributions of these deaths  are shown  in  Table
1.  Looking at a list of counties in which deaths  from these two  cancers
occurred, the deaths appeared to be concentrated  in  urban counties.   For
20 of the largest SMSA's there existed air quality data in  the  form of
the Pollution Standard Index  (PSI)  developed  by EPA.  An  approximate
annual average was calculated for 1973 for each SMAS.  These annual
averages were plotted versus  the number of deaths  in each SMSA.   These
plots are shown in Figures 4  and 5.   In both  cases the correlation
coefficient between the number of deaths and  the  PSI was  0.57-

     As  in the case of the pervious maps,  these graphs suggest  a  number
of interesting questions. Other  urban factors than air quality  should be
considered.   It would be nice to map deaths from  rare causes using spot
maps rather than shaded maps.   In spot maps a symbol  is plotted which is
proportional  in size to the variable being plotted.   This capability  is
being added to the UPGRADE system.  As  is  always  the case in descriptive
epidemiology, analysis of this type are best  suited  to generating
hypotheses for further research  and cannot by themselves  provide
evidence of increased risk.
                                CONCLUSION

      This  purpose  of  this  presentation  was  not  to  present  scientific
 studies  on cancer,  but  to  illustrate the  UPGRADE system  and  the
 associated national mortality databases.   UPGRADE  has  been developed  as
 an  exploratory  analysis tool.  The  Integrated  Database brings  together
 information  from  a  variety of sources.   It  is  hoped  that the analyses
 presented  here  show how the system  can  be used  to  investigate  the
 relationships between environmental  factors and health indices.
                                  713

-------
Table 1.   Sex and race distribution of  1970 deaths  from cancer of the
          eye and from cancer  of  the thyroid gland.
                   White
                   Black
                   Other

                   Total
Cancer of the Eye

            Female

             164
              10
               1

             175
Male

163
 14
  0

177
 Total

  327
   24
    1

  352
                       Cancer of the Thyroid Gland

                                       Female      Male
                   White
                   alack
                   Other

                   Total
             649
              54
               3

             706
338
 22
  1

361
 Total

  987
   76
    4

1,067
                                714

-------
Figure 1.  Geographic distribution of age-adjusted mortality rates for
wnite males from cancer of the liver, gall bladder, and bile ducts (ICDA
codes 155 and 156).  Rates are per 1,000,000 population and are
calculated on deaths occurring 1968 to 1972.
Figure 2.  Geographic distribution of age-adjusted mortality rates for
white males from currhosis of the liver (ICDA code 571).  Rates are per
1,000,000 population and are calculated on deaths occuring from 1968 -
1972.
Figure 3.  Geographic distribution of cirrhosis of the liver with the
ten percent of all counties having the highest rates shaded the darkest
                                     715

-------
AGE-ADJUSTED MORTALITY FROM CANCER OF  THE  LIVER,
    \6ALL BLADDER,  AND DUCfl ilCOA '55,!56I  PER 1,000,000
          POPULATION AT 1ISK, »H[TE MALFS, !965-l9r2
                                                                                         NATIONAL CENTER FOR HEALTH STATISTICS
                                                                                     U.S.OEPARTMENT OF HEALTH,  EDUCATION AND WELFARE
                                                                                       COUNCIL  ON  ENVIRONMENTAL QUALITY

-------
AGE-ADJUSTED MORTALITY  FROM CIRRHOSIS  OF THE  LIVER
           (IfDA S7:  ^ES :(?C.;.;0?;- nOrbu.Ai;ON Al  TT3S
                                                                                        NATIONAL  CfNIER FOR HEALIr SIAlISIIt^

                                                                                            MENT OF nEHLTH. EDUCATION ANE «CLFASE
                                                                                     COUNCIL ON ENVIRONMENTAL QUALITY
                            'fTfi

-------
AGE-ADJUSTED  MORTALITY  FROM CIRRHOSIS  OF THE  LIVER
          IICOA 5711  PER 1,000,000 POPULATION AT RISK
                   WHITE MALFS:  1968-1972
                                                                                        NATIONAL CENTER COR HEALTH STATISTICS
                                                                                       DEPARTMENT OF  HEALTH, EDUCATION AND WELFARE
                                                                                               C37

                                                                                      COUNCIL ON  ENVIRONMENTAL  QUALITY

-------
C
A

E
Y
E

D
E
A
T
H

C
0
u
N


1

7
  IS
15
13
11
           IYE CANCER UfftfUS TIC POLLUTION STANDARD INKX

                        M UMNEIT MM'f

                    
-------
ro
o
                                   TMVROID CANCCT US THE POLLUTION STANDARD INDEX

                                                a LAMEST SRSA'S
                                           (1ST OftOEft LEAST SQUARES FIT)
                                                                                       1 ( •) --15.134
                                                                                       1 ( 1 > • .326
                                                                                       R2 •   .33*
                                                                                       F •  8.877
                                                                                       PR>F •    .W8
                                          fSI
                       Figure 5.   Regression  -plot  of deaths from cancer of  the
                                    thyroid gland  (1970) versus  the average annual
                                    PSI  for 1973.

-------
                               Discussion


Dr. Mason (NCI):  I think it would be helpful for those who are not as
familiar with the system as several of us are, if you would share with
the audience where it is resident, how one goes about getting access to
it, what charges are incurred, does one have to be a sponsor, et cetera.

If you would do that, then I would like to interact with you about a few
of these things.

Dr. Graves:   That is a good question.  It lets me do a little marketing.

The system was developed on NIH computers and has recently been moved to
the Boeing Computer Services computer facility in McLean, Virginia.
Boeing has a computer network so that access to the system is a local
call from most places around the country.  The system continues to be
available to government agencies who have sponsored its development.  In
addition, it is now more easily available to government agencies whether
or not they are interested in sponsoring the system.  Now an agency can
just pay for off-the-shelf use.

Because of the public access to Boeing, the system will also be avail-
able to other government agencies, i.e., to state agencies, educational
institutions, consulting firms, et cetera.  In summary^ there are two
ways to access the system.  Federal agencies for the most part will
access the system through agreements with CEQ.  Non-federal access will
be through Sigma Data Computing Corporation.

The current cost is the Boeing computer charge plus a percentage for the
maintenance and administration of the accounts.

Dr. Mason (NCI):  Okay, so that if someone from NIOSH or from EPA or
Industrial Health or wherever were interested in getting access to
UPGRADE -- and by this I mean really creating a subfile of data, say
mortality rates for a particular disease that they are interested in --
they would do the following.  They would contact CEQ and say, "These are
the things that I am interested in.  Can you provide them?"  All they
would have to do is reimburse CEQ for the creation of that file.

Dr. Graves:   Yes. charges for creating a file would include both
computer and personnel costs.  After the file is created the only charge
would be for that person's or agency's computer usage.

Dr. Mason (NCI):  Okay, and one last point.  I would encourage you, when
you give this presentation and show these particular maps, not to make
too much out of tiny numbers in any one year, and rather lean on the
strength of the system.  The system is interactive.  The system gives
the research-oriented person an opportunity to sit down at a CRT and look
at the data to see how rich they are, to look at how many measurements
                                       721

-------
are there, to look at the range, look at characteristics of the under-
lying distribution, look at factors as they relate to climatic conditions,
and things such as that.  And I think the strength of the system is that
an analyst has an opportunity to sit down in a cost-effective manner and
look at, in whatever way is deemed appropriate, these particular data.
The analyst can then branch over to appropriate analysis procedures.

I would encourage you to perhaps make this point a little stronger
rather than stressing a regression equation fitted to tiny numbers in
one year; rather argue that you have the full  complement as far as staff
and facilities to do whatever analysis is deemed appropriate.

Dr. Graves:  Yes, I agree with you.  These slides were put together, as
I said, more to illustrate the system than to  illustrate the science.

However there are several points I'd like to make.  The rates  which were
mapped were not calculated from deaths in a single year but for the
years 1968 to 1972.  Because several years are included, the rates are
based on a larger number of deaths.

The second point I want to make is that in looking at these rare causes
of cancer, we are interested in their geographic distribution.   These
maps are not the most suitable for this purpose.  For this reason a spot
mapping capability is being added to UPGRADE.   In spot mapping  a symbol
appears wherever a death or other event has occurred rather than shading
an entire geographic area such as a county. That is another tool  and it
should be very useful.

Dr. Kraybill (NCI):  Looking at that map and not being an epidemiologist,
I am going to ask a simple question.  The map  of liver cancer  rates
shows high rates in the Midwest.  What is your explanation, or  does any
epidemiologist here have any explanation for that?  Is it real?

Dr. Mason (NCI):  No, and that is why it looks that way.  What  has been
done in order to get numbers is fine.  Here cancers of liver and gall-
bladder and bile ducts have been combined, when in fact, the determinants
of liver cancer and the determinants of gallbladder cancer are  totally
distinct.  If you look at liver cancer by itself, you see aggregations
in Texas and other such places.   This distribution of cases has prompted
a number of follow-up studies in which we are  looking at everything from
the hepatitis B carrier state to the potential for dietary aflatoxin to
the potential for specific classes of compounds which are known to be
carcinogenic to the liver.

If you look at gallbladder cancer by itself, you see a midwestern
concentration of cases.  This is more consistent with the dietary
practices of the ethnic groups who have settled there.  The ethnic diet
leads to the development of gallstones which is the primary risk factor,
the first and most important risk factor, for  gallbladder cancer.   So
the maps do not show liver cancer alone.  They show liver and  gallbladder
combined.
                                          722

-------
It is interesting,  if you pick up on the liver-cirrhosis  type of  thing,
you do indeed see disparate distributions.   Some of our  earlier work
never got into press because we were not convinced  that  it was not an
artifact -- that perhaps what was being called cancer in certain places
was really advanced cirrhosis and vice versa.   So we have  gone into the
field to attempt to sort this thing out in a much more analytic way.

Dr. Graves:  The fact that people are interested in just liver cancer or
just gallbladder cancer and not a combination is one reason we are so
enthusiastic about going to the detailed mortality  data  that  I mentioned
With the detailed data, we can break it down, if you want, to the  fourth
digit ICDA code.  But this first set of rates gives people some ideas,
and as you have done, they can go on from there.
                                      723

-------
MAPPING CHEMICAL EXPOSURES
         BY
    Kenneth D. Kreitel
                  724

-------
                                ABSTRACT











The Hazard Section of the Surveillance Branch is actively pursuing




several new techniques for focusing on potential worker exposures to




chemicals in use in industry.  This paper reports on a technique to




display the geographic concentration of potential worker exposures




through the linking of large computer files and data bases.






The computer files from NIOSH's National Occupational Hazard Survey




are searched for specific instances where a chemical material of interest




was identified during the site visit phase of the survey.  Those companies




in which the material was observed are classified by their four-digit




Standard Industrial Classification (SIC) code.  The names and addresses




of similar companies are then extracted from the Dun & Bradstreet




computer files.  The resultant computer file is then analyzed statis-




tically and a cartography system at The National Oceanic and Atmos-




pheric Administration (NQAA) produces a map of the continental United




States with each county shaded appropriately.






For the purpose of providing immediate visual impact, areas of the




United States which contain large numbers of companies similar to those




which were observed using the materials in question are shaded very dark.




Areas less likely to contain significant quantities of such chemicals are




shaded in lighter half-tones.
                                         725

-------
The system is  capable of producing maps for each of  the  8,000 poten-




tially hazardous materials encountered during the National Occupational




Hazard Survey, either individually, or in combination.
                                         726

-------
I.   INTRODUCTION




     The purpose of this paper is to give a preliminary report  on




     a new technique that the Hazard Section of NIOSH is actively




     pursuing.







     The Hazard Surveillance Section of NIOSH is charged with the




     responsibility to develop, compile, and analyze information




     on the number and distribution of workers exposed to potential




     occupational hazards to enhance the preventive aspects  of




     occupational health.







     It is natural, therefore, for the Hazard Section to be  interested




     about the geographical disperson of various industrial  materials




     throughout the United States.  Our interest in this stems  from




     the often expressed need for greater awareness of potential




     occupational exposures to hazardous materials.







     This awareness is very difficult to achieve because of  two




     built-in confounding factors.  Chief among these factors is




     the practice of tradenaming products.  That factor only




     slightly overshadows the other; which is inadequate labelling




     requirements.  Taken together, these two constitute a rather




     large impediment to the kind of awareness we feel it is




     necessary to build.







     The Hazard Section does, however, have access to a unique




     resource which is useful in penetrating the mystique surrounding
                                         727

-------
     the question of who is potentially exposed to what in the




     work place.   This report is an attempt to briefly describe




     that unique  resource,  and to introduce an intriguing new use




     of its data.







II.   GOAL




     The goal of  the mapping project is to develop maps of the




     continental  United States showing suspected locations and




     concentrations (if possible) of potentially hazardous exposure




     agents.   Furthermore,  insofar as practical, we would like to




     compare these maps and the underlying data with other data




     sources such as NCI's  "Cancer Mortality by County: 1950-1959" (1)




     and their "Atlas of Cancer Mortality for U. S.  Counties:




     1950-1969."(2)







III. RESOURCES







     The principal resource used for the mapping project was  NIOSH's




     National Occupational  Hazard Survey Data Base.   The National




     Occupational Hazard Survey (NOHS) was conducted during the




     period 1972  through 1974.  It was a nationwide effort to  gather




     information  on potential workplace exposures to hazardous




     material through the use of 20 field surveyors who actually




     visited over 4,500 different plant sites throughout the




     United States.
                                       728

-------
The surveyor's job, to simplify it greatly, was to first




interview the plant management about current practices within




the plant, and then to conduct a detailed walk-through survey of




the plant, noting occupational exposures to potential chemical,




physical, and biological hazards.  The surveyors also noted




the conditions under which the exposures were occurring, and




the control measures that were being applied.






The plants that were surveyed represented a national probability




sample of selected industries.  The result of that effort is




a computerized data base which contains almost five million




records, and which is useful for describing potential occupa-




tional exposures by industry, by occupation, and by exposure




agent. (3)






NIOSH's experience in compiling this data base indicated that




most of the workers' exposures were to products that were trade-




named as opposed to being in pure chemical form with the




chemical adequately labelled.  Some 70% of all exposures noted




during the survey were, in fact, to tradenamed products.  NIOSH




then began a program of follow-up by writing to the manufacturers




of the tradenamed products to obtain the ingredients and the




formulation of the product.  This auxiliary effort, dubbed




"TNIC" or Trade Name Ingredient Clarification yielded information




which has proven to be invaluable for the mapping project.  Now




integrated into the main NOHS data base, the TNIC data provides
                                     729

-------
     very valuable insights  into  the potential  for  occupational  exposures




     to hazardous  materials  that  were formerly  obscured  or  disguised




     due to the twin problems  of  tradenaming and  inadequate labelling.




     Of the approximately 80,000  different  tradenamed  products encountered




     in the course of the NOHS survey,  about 64,000 or 80%  have  been




     resolved into components  through the cooperation  of the manufacturers.







     The secondary resource  that  the mapping project draws  upon  is  the




     Dun & Bradstreet file.  This computerized  file contains information




     on 4.3 million companies  throughout the United States.   Each




     company record includes the  company name and address,  its size




     in terms of the number  of people employed  there,  and its Standard




     Industrial Classification (SIC) code.







IV.   METHODOLOGY




     As a first step, a single important industrial material suspected




     of being in widespread  usage throughout the  United  States was




     chosen as an appropriate  vehicle for developing the methodology.




     This material was chosen  because it was suspected of being




     incorporated into a wide  range  of products which  enjoyed a




     wide variety of uses within  industry.







     This material, asbestos,  was used as the basis for  a computerized




     search of the entire NOHS data  base.   The  result  was a compilation




     of all the plants in which NOHS surveyors  had  noted at least  one




     worker potentially exposed to the material by  virtue of his or




     her job, during the period 1972 through 1974.   Any  worker who
                                      730

-------
indicated he or she used this material (or a tradenamed product




which, upon resolution was found to contain this material) for




periods totaling more than one-half hour per week in the aggregate




served to nominate an entire industry for further consideration.






The list of industries nominated through this process by the




NOHS data base was lengthy.  It included one-hundred and forty-




seven distinctly different industries, as delineated by the four-




digit Standard Industrial Classification (SIC) code.






The length of the list was due, in part, to the ability of the




integrated NOHS and tradenames data bases to penetrate the




tradename barriers and detect obscure or unrecognizable exposures




to the material.  This is expected to become a regular occurrence




in future attempts at mapping.






The list of nominated industries was then analyzed a number of




ways, to determine the extent to which the NOHS study accurately




reflected each of the industries in question.  A set of decision




rules was then developed which was capable of separating the list




into two smaller lists of industries; one qualified for further




consideration, and one unqualified.  The decision rules were




carefully applied to each of the industries on the candidate list.
                                     731

-------
The first rule specified that NIOSH surveyors must have observed




the material in question at least twice in an industry during




1972-1974.  The second rule specified that in addition to Rule #1,




the NIOSH surveyors must have noted exposures to the material in




question in at least 25% of the plants of that industry type that




were visited.  The two tests were designed to eliminate from




further consideration those industries in which the NOHS data




was too limited to provide a good case for continuing.






The fully qualified list of industries (see Table 1) then formed




the basis for estracting the records of similar business establish-




ments throughout the United States from the Dun & Bradstreet file.




The entire Dun & Bradstreet file was searched for companies whose




industry codes matched the twenty-five (25) on the "fully qualified




industries" list.  Records of approximately 60,000 business establish-




ments were extracted using the matching procedure.






These records were then organized by county and analyzed with the aid




of a widely-available computerized statistical analysis system.




Results were tabulated and displayed on a county-by-county basis




as a means of providing the researcher with some preliminary insight




prior to readying the data for the cartography system.






The cartography system was county-based.  It contained X and Y




coordinates of all the county lines, and required only the proper




county code and a code to indicate the desired shade of darkness for




each of the counties. (2)  The proper country identification code was not







                                   732

-------
     the one extracted from the Dun &  Bradstreet  file, however.  Therefore, a code-




     conversion table was  built and software  developed which was capable of auto-




     matically converting  the Dun & Bradstreet  code  to the  preferred code.






     The shade code was assigned to each  county on the basis of the




     number of qualified industrial facilities  within the county.  To




     provide clarity and to achieve the greatest  visual  impact, the counties




     with the highest number of qualified facilities were shaded the




     darkest.  Nine different shading  protocols were investigated as a




     means of becoming familiar with the  variation in subsequent outcomes




     that the different protocols afforded.   As the  maps show, it is




     possible to prepare shading protocol that  becomes more selective




     until only the very,  very high interest  counties remain shaded on




     the map.
V.   RESULTS
     The maps appear to corroborate the conventional wisdom that asbestos




     and asbestos-contining tradenamed products conceivably were used




     in industrial settings across most of the face of  America.






     The industries that were rated "fully qualified" tend  to be found




     in conjunction with large population centers.  There are, however,




     some notable exceptions.  Fargo,  North Dakota, and Sioux Falls,




     South Dakota, for example, with populations less than  100,000




     persons, cannot be considered major population centers, yet each




     contains several "fully qualified" industries. Two large popula-




     tion centers in particular, Cook County, Illinois, and Los Angeles
                                          733

-------
     County, California, contain very large numbers of businesses that




     fit the "fully qualified" description.







     These maps are not "rate" maps.  That is, they are independent




     of population considerations and thus they serve only to locate




     geographical areas with large numbers of fully qualified industries,




     They do not attempt to depict nor to predict high incidence rates




     of asbestos-related illnesses.







VI.  LIMITATIONS







     There are two principal limitations inherent in this technique




     which must be understood for correct interpretation of the




     results.  First, the NOHS survey was not designed to be statisti-




     cally representative of industries at the four-digit SIC code level.




     Some industries represented by four-digit codes, in fact were not




     visited during the survey.  Those industries are not represented




     anywhere in the data.  In addition, more four-digit industries in




     general were not sampled with enough frequency to assume that the




     sample that was drawn was representative.  The decision rules




     detailed in the methodology section above formed the sole basis




     for qualifying industries to the list of highly interesting




     industries.







     The second principal limitation upon the interpretation of the




     results is more mechanical than statistical.  The NOHS data was




     described in terms of 1967 SIC codes.  The Dun & Bradstreet
                                       734

-------
file uses the 1972 version of the same publication. (4)  There




were some industry classification changes between the two




versions of the publication.  The changes were minor in nature.




The major problem involved in the methodology above is the




great "leap of faith" that was made in assuming that an




industry as delineated by a four-digit SIC code in 1972 is




essentially the same as the industry typified by the same SIC




code in 1979.  No attempt was made to account for possible




changes in technology, methods of production, changes in




regulations, or geographic shifts in industry in the years




between 1972 and 1979.






The maps simply represent the distribution of industries that




qualified as being of "high interest" through the methodology




above at a single point in time.






Asbestos was chosen only as a first attempt to map chemical




exposures.  The Hazard Section is continuing to develop this




technique, and will map other industrial materials in response




to the Institute's Surveillance needs.
                                735

-------
                               TABLE 1



               STANDARD INDUSTRIAL CLASSIFICATION CODES  OF

                 INDUSTRIES CONSIDERED  "FULLY QUALIFIED"
1967 SIC CODE
       INDUSTRY DESCRIPTION
    1742
    1752
    1761
    2011
    2821
    2851
    2911
    2952
    3241
    3291
    3292
    3312
    3352
    3433
    3443
    3519
    3661
    3711
    3713
    3721
    3731
    3742
    3791
    3843
    3996
Plastering, Drywall, and Insulation
Floor Laying & Floor Work N.E.C.
Roofing and Sheetmetal Work
Meat Packing Plants
Plastics Materials and Resins
Paints and Allied Products
Petroleum Refining
Asphalt Felts and Coatings
Cement, Hydraulic
Abrasive Products
Asbestos Products
Blast Furnaces and Steel Mills
Aluminum Rolling & Drawing
Heating Equipment, Except Electric
Fabricated Platework (Boiler Shops)
Internal Combustion Engines, N.E.C.
Telephone and Telegraph Apparatus
Motor Vehicles and Car Bodies
Truck and Bus Bodies
Aircraft
Ship Building and Repairing
Railroad Equipment
Trailer Coaches
Dental Equipment and Supplies
Hard Surface Floor Coverings
                                       736

-------
BIBLIOGRAPHY






1.  Mason, T.J., McKay, F.W.,  U.  S.  Cancer Mortality by County




    1950-1969, National Cancer Institute,  National  Institutes




    of Health, DHEW, 1974






2.  Mason, T.J., McKay, F.W.,  Hoover,  R.,  Blot,  W.J.,




    Fraumeni, J.R. Jr., Atlas  of  Cancer Mortality by County




    for U. S. Countries; 1950-1969,  National Cancer Institute,




    National Institutes of Health, DHEW,  1975






3.  National Occupational Hazard  Survey,  Volume  II, Data




    Editing and Data Base Development, Division  of  Surveillance,




    Hazard Evaluations, and Field Studies, National Institute




    for Occupational Safety and Health, Center for  Disease




    Control, DHEW






4.  Standard Industrial Classification Manual 1972, Statistical




    Policy Division, Office of Management and Budget,  Executive




    Office of the President, 1972
                                     737

-------
                              Discussion
Dr. Jenkins (EPA):  How are you planning to utilize the data you are now
collecting from both the labor unions and industry on specific chemicals
and specific plants, and how is this going to interface in the future
with updating this mapping system?

Mr. Kreitel (NIOSH):  It is actually a little too soon to tell that. The
information that we are collecting from the unions that relates specific
chemicals to specific plants is not coming in on anything that resembles
a nationwide basis at all, so it really would have to have much more
complete coverage before we could do anything like this with it, so it
probably will not interface for a large number of years.

Dr. Bell in (EPA):  I am just curious, as an ancillary question, what is
your experience with trade name products?  How often does their com-
position change?  Can we assume that something that had a certain
composition in 1972 has the same composition now?

Mr. Kreitel (NIOSH):  As a matter of fact, that is probably a very
dangerous assumption to make.  We believe that there is a product life
cycle out there and that products in certain industries turn over much
faster than products in other industries and other uses.

There are a few more presentations that will be made today on our plans
to update that data base and on a brief description of that data base
itself.
                                      738

-------
FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

    PROGRESS ON 3OINT ENVIRONMENTAL AND

        OCCUPATIONAL CANCER STUDIES
            Thursday Morning, May 8
           CONCURRENT SESSION II

     DATA BASES/MONITORING (CONTINUED)
            SESSION CHAIRPERSON
            Dr. Elizabeth Weisburger
            National Cancer Institute
                       739

-------
                     THE NATIONAL OCCUPATIONAL HAZARD  SURVEY
                            (NOHS II)—PROGRESS REPORT

                                 David S.  Sundin
                   Hazard Section, Surveillance Branch,  DSHEFS
                             NIOSH - Cincinnati, Ohio

     In order to make intelligent decisions  concerning priorities  for action,
NIOSH requires a current and continuing supply of information on nationwide
patterns of occupational exposure to health  hazards.   The National Occupational
Hazard Survey of 1972-1974 (NOHS  I)  which  developed  such information  among a
probability sample of nearly 5,000 plants, has had important  applications  in a
wide range of NIOSH research activities.   It will become increasingly necessary
to update the results of that survey in order  to reflect current conditions.

     The foundation of experience which was  developed  during  NOHS  I will be
used to support and guide the planning and implementation of  a  second national
survey.  Although methodologies used in the  new survey will closely .resemble
those of NOHS I, the previous survey forms,  data processing systems,  programs,
and procedures will be critically examined to  identify areas  where improvements
can be made.

     A sample of approximately 5,000 facilities will be  selected to cover  a wide
range of industry types, facility sizes and  geographical locations, and will be
constructed so as to permit the development  of national  exposure estimates.  This
sample will be selected in the fourth quarter  of fiscal  year  1980.  Twenty-one
industrial hygienists will be recruited and  trained  intensively in observational
survey techniques during the fourth quarter.   The surveyors will then be sub-
divided into teams and deployed into the selected facilities  beginning in  fiscal
year 1981 in order to gather data on the nature and extent of potential occupa-
tional exposures to chemical and  physical  agents, including trade  name products.
The field phase of the survey will last approximately  two years, and  preliminary
data should be available during the year immediately following  the survey.
                                         740

-------
JUSTIFICATION AND BACKGROUND
An intelligent and informed sense of priorities  for action  requires a
current and continuing supply of information  on  nationwide  patterns of
occupational exposures to health hazards.   It 1s important  to  know what
hazards are associated with certain industries and  occupations,  and to
have some idea of the relative importance  of  those  hazards,  both in terms
of total number of workers at risk, and the toxic properties of  the sub-
stances.  An effective program of hazard surveillance  requires a diverse
array of data collection activities.   It must identify and  seek  out net-
works which supply intelligence on occupational  hazards, and must be
sensitive enough to detect early indications  of  potentially  hazardous
situations in time for effective intervention on behalf of  those affected.
Hazard surveillance aims to provide the information necessary  to prevent
occupational disease, so that the in vivo  experiment need not  be carried
to its obvious conclusion. . It is a goal which challenges us for a number
of reasons.  Chief among these is the fact that  new chemicals  are intro-
duced into the workplace in numbers which  largely overwhelm  the  capacity
of toxicological testing facilities.   A second factor  involves the
difficulty in extrapolating animal  data to humans.   Another  confounding
factor is the extensive practice of masking chemicals  with  trade name
designations, thus frustrating attempts to even  identify the chemicals
to which workers are exposed.
Recognizing these factors, NIOSH has constituted a"  program of  hazard sur-
veillance which (1) attempts to monitor a  broad  range  of current informa-
tion on chemical toxicology, (2) conducts  periodic  special surveys to
develop intelligence on the distribution patterns of health  hazards at
the workplace, and (3) persistently seeks .to  gather information  on the
chemical ingredients of industrial  trade name products.
                                    741

-------
NIOSH's first nationwide  special survey of occupational  hazards was
conducted from 1972-1974  in a  probability  sample of  nearly  5,000  facilities.
A team of twenty-surveyors conducted wall-to-wall observational surveys  in
a broad sample of workplaces, inventorying  chemical and  physical agents,
Including trade name products.  A  total of more than 9,000  different
potential hazards were discovered, and more  than 85,000 trade  name products
were listed.  After an extensive and time-consuming  period  during which
over 10,500 manufacturers were contacted and asked to supply ingredient
information on their products, the data base was available  for producing
estimates of the number of people  exposed  to selected hazards.
The foundation of experience which was developed during the original National
Occupational Hazard Survey is  being used to  support  and guide  the planning
and implementation of a second national survey.
OBJECTIVES
The broad objectives of MOHS II can be simply stated:
(a)  To develop a data base capable of producing reliable estimates of
     the total number of  workers in the target population exposed to
     the health hazards which  are  observed during the survey.
(b)  To develop data which is  capable of reliably describing each industry
     type .comprising the  target population in terms  of  the  nature and ex-
     tent of exposures to health hazards,  and the degree to which facilities
     have implemented programs to  reduce occupational health problems.
(c)  To compile the data  in such a way that  analyses of industrial hazard
     exposure trends can  be made by comparison with  NOHS I  information.
There are a number of important parallel activities  which must be conducted
to initiating the field phase  of the survey.  The ultimate  utility of the
data and tha speed with which  it can be made available  for  dissemination
and use depend on the successful timing and  execution of these tasks.
                                     742

-------
FORMS DESIGN
One task which is a necessary  prerequisite  to many  others  is  the  drafting
of forms for data collection and  guidelines to  enable  the  surveyors  to  use
the forms.  NOHS I used a 3-part  form, and  the  same basic  structure  will
be employed in the second survey.   Part  I of the  form  is filled out  during
an Initial interview with plant management. It consists of a series of
approximately 50 questions which  elicit  information about  the health and
safety program at the facility, and the  degree  to which the program  has
been effective in reducing illnesses and injuries.   We have encouraged  wide-
spread input of candidate questions for  inclusion in Part  I.   Part 2 of the
form is filled out during the  walk-through  survey,  and captures data on the
types of workers {sex and occupational title) observed, and their exposures.
Additional data on the duration"of the exposures, the  physical  form  of  the
substance, and the control.-measures being used  are  also collected at this
time.  Part 3 contains information on the amount  of surveyor  time spent
travelling to a plant, walking through the  facility, and writing  up  the
results.
SAMPLE FRAME
A second major task which must be completed prior to initiating the  field
phase of the survey is the construction  of  a sample frame  or  comprehensive
listing of facilities from which  a probability  sample  can  be  drawn.  Since
there are more than 4,000,000  facilities in :he United States,  a  certain
portion of which enter or leave the sceyie each  year, the assembly of a
current and accurate frame requires sone effort.  A NOHS II sample frame
will be built from existing data  sources, and modified where  necessary.
The construction industry, which  includes general contractors and special,
trade contractors, will create special problems due to the fact that the
majority of exposures probably occur at  the job site rather than  at  the
location of the business headquarters.   Separate  listings  of  construction
projects in progress will likely  be required to adequately describe  this
important activity.
                                    743

-------
The target population for NOHS II will  be slightly  different than that
chosen for NOHS I.  Extremely small  facilities  (7 employees  or less) will
still be excluded from the frame, but an  attempt will  be made to achieve
a larger, more complete sampling of the manufacturing  industries.  This
will mean that fewer facilities in industries  like  wholesale and retail
trade, financial institutions, and educational  services  will be surveyed.
After the target population has been described, a sample strategy will be
developed which maximizes the use of the  surveyor resources  and meets the
objectives of the survey.  It is anticipated that some form  of stratifi-
cation by Standard Industrial Classification (SIC)  and size  of facility
will be used, and that geographical  clustering  will  be required.   Part of
the sampling strategy will involve developing exigency procedures to deal
with situations where a facility is  unsurveyable.   While NIOSH has been
granted statutory authority-to enter workplaces during the conduct of its
research activities, there may be occasions  where this authority will be
challenged.  In such cases a decision will be made  on  whether to undertake
the legal action required to effect  entry, or to draw  another facility as
a replacement.
The final stage in design of a sampling strategy is to produce a list of
facilities which constitutes the sample.   It will probably be necessary to
validate the list of facilities by verifying the critical data through
phone contact with management personnel at each candidate facility.
A specific task which is closely related  to  the development  of a sampling
strategy Is the design of analytical  methods which  will  accept the survey
data as irput and produce national  estimates of numbers  of workers exposed
to selected hazards.  This projection algorithm must be  computationally
efficient and capable of identifying those estimates for which measures of
variance can be computed.
                                  744

-------
RECRUITING AMD TRAINING SURVEYORS;
It will  be extremely important  to  select and  train  the  surveyors  properly.
All surveyors will  be required  to  have  a bachelor's degree  in  a scientific
discipline such as  chemistry, chemical  engineering, or  health  sciences.   A
minimum level of training in organic  chemistry will be  necessary.   Six to
eight weeks of classroom training  in  general  principles  of  industrial
hygiene with special  emphasis on  recognition  of  hazards  is  planned.  A
two week module tailored to the specific goals and  methods  of  the  survey
will follow, after  which a period  of  two to three weeks  will be spent  in
1n-p>ant instruction in the use of the  forms.
CLARIFYING TRADE NAME INGREDIENTS
The entire process  of acquiring'ingredient information  on trade name products
will be critically  examined to  identify areas where improvements  can be  made.
It will be important to check the  product names  from NOHS II against those
Identified in NOHS  I to avoid requesting the  same information  from a manu-
facturer a second time.   It may also  be possible to access  other  sources of
Ingredient information which provides a satisfactory level  of  detail.  The
strategy of offering the option of comprehensive reporting  on  an entire
product line to manufacturers is being  considered as one means of  minimizing
the need for repeated contact.  Current procedures  for safeguarding confi-
dential data will be evaluated  for possible improvements.
DATA PROCESSING
It 1s hoped that the extensive  collection of 'systems and software  developed
during anc' following NOHS I can be used as the basis for the development of
the data automating system which will be required for NOHS  II.  If the survey
data is automated and edited as soon  as it is collected, it will be possible
to spot errors in time to communicate this information to the  surveyors  for
correction while they are still in the  field, and before it becomes a  wide-
spread problem.
                                 745

-------
FIELD PHASE

The operational aspects of the field phase of the survey will  be modified
somewhat to improve the quality and quantity of data collected.   The
survey force will be divided into several  smaller teams, and a team leader
for each group will be identified.   It will be the responsibility of this
Individual to coordinate the activities of the group, schedule and assign
facility surveys within a region, desk check and organize the data forms
for submission to headquarters, and serve  as a conduit for information be-
tween the surveyors and headquarters staff.  The surveyors will  spend more
time disseminating general information at  the plant site, including infor-
mation about NIOSH services such as the Health Hazard Evaluation program.
They will also be on the alert for any special  problems that arise, in-
cluding previously undiscovered health problems associated with  new chemicals
or new uses of common chemicals.  They will thus function as a continuing
source of current intelligence on emerging problems.   All such information
will be relayed to headquarters for appropriate action.
DISSEMINATION PLANS
The ultimate worth of such a large-scale survey effort is judged on the
basis of how the data is disseminated and  used.   Many of the changes in
survey design from NOHS I to NOHS II are designed to  reduce the  interval
between termination of the field phase-of  the survey  and availability of
survey data.  Rether than publishing a single volume  of summary  statistics,
a series of interim, special topic  reports is contemplated.- These special
topic reports could take the form of profiles of selected industries or
occupations, or could be designed to deal  with  classes of chemicals.
Efforts wiII be devoted to loading  the data under a data base  management
system to facilitate specialized retrievals soon after automation is
completed.
In a survey of this magnitude and complexity there will always be unantici-
pated developments which necessitate modifications to the original plan.
Realization of this fact increases  the ability of the survey managers to
accommodate the unexpected.   NOHS I provided many essential lessons which
will not be lost on the architects  of NOHS II.   It is with considerable
enthusiastic anticipation that we look forward to this challenge.
                                    746

-------
                              Discussion
Dr. Green (NIOSH):  I have two questions.   You mentioned that when you
get some trade name information, you will  go back and check and see if
it overlaps with requests to manufacturers from NOHS I, and there was a
point brought up that the ingredients may have changed in the interim,
so perhaps you might consider that, and a second contact might be
worthwhile.

Mr. Sundin (NIOSH):  Yes.  I should clarify my statements here.  We are
undertaking at the current moment a research effort which analyzes our
existing trade name file.  One component of that research effort is to
gain some sense of the rate of change in production formula.

A similar study was conducted on CPSC's file of trade name ingredients,
and we are looking with some interest at the results of that study.

In any case, I think it will be important not to simply go out and re-
request information on a product for which we have even antiquated data.
Rather, an intelligent strategy would seem to be to at least give the
manufacturer the opportunity to indicate whether the product formula as
we have it has changed or is the same.  I think that the people that are
looking at respondent burden would want us to operate in such a fashion.
We are sensitive to the fact that we cannot just use old information
that we have on file, but we will at least establish whether or not we
have any information on that particular product and then attempt to
validate with the manufacturer whether it is current information.

Dr. Green (NIOSH):  The other thing I wanted to bring up, in the first
survey there was a certain percentage of the exposed population whose
estimate was based on what was called the generic exposure.  Can you
discuss what that means and what are the criteria for selecting people
as falling into that group?

Mr. Sundin (NIOSH):  A "generic procedure" was used to derive data on
exposures for those people that were exposed to trade name products for
which we did not have ingredient information from the manufacturers.

The whole process of clarifying trade name products has been an
extremely difficult one for a number of different reasons, which is why
we are directing a lot of our attention during the current planning
stages to improve that effort.

Any kind of response you could possibly imagine occurring, did in fact
occur at some time during the course of contacting these 10,000 manu-
facturers.  Our file of ingredient information grew rather slowly, and
the pressures to examine survey data began almost immediately after the
field phase was completed, as you might expect.
                                      747

-------
Everytime a trade name product was noted in the field, an accompanying
record was input by the surveyor concerning what that product was
doing, what type of product it was.   Was it a surfactant, a cleaning
compound, and so forth?  Under contract, we then researched the con-
temporary chemical  literature and assembled a list of probable
ingredients for product formulas that we have not yet received from
manufacturers.  At such time as we receive actual ingredient information
from manufacturers, we then replace the generic resolution with the
actual data.

With a view toward NOHS II and some of the problems that we have noticed
with using generic resolution, we are attempting to compare the set of
ingredient information which we actually got in specific product classes
from manufacturers with that generic list and see how good the generic
list actually was and modify it based on actual data received.  We
realize we will still be faced with that kind of a situation, since to
get trade name ingredient information requires some time, but we hope
our generic resolution process will  be a little more accurate the
second time.

Dr. Weisburger (NCI):  There have been some comments made that some
times the NOHS overestimates the number of workers that are exposed to
a particular chemical.  Could you comment on that, please.

Mr. Sundin (NIOSH):  In this business, nobody is ever satisfied with
your estimates.  Either they are too high or they are too low.  But in
most instances where we have researched what seemed to us to be a
serious challenge to the estimates,  what we find is that there are so
many non-standard, secondary, widely dispersed uses of certain products
which the manufacturers or the processors or the distributors of the
chemical are totally unaware of, that we indeed find that the NOHS
estimates derived, as they are, from a plant level sort of survey
procedure, tend to represent potential exposures fairly well.

Realizing that we are not talking about exposures that were defined by
taking environmental measurements, but rather about a compound that is
in the workplace of a worker, being used at something more than 30
minutes per week, the other process that we sometimes use to answer
these sorts of questions is to look at only full time; i.e., more than
4 hours a day, or part time, and an amazing number of exposures occur
in very short bursts.  The numbers drop drastically when you talk about
only full time exposures, and that tends to be closer to the classic
definition of exposure in many people's minds.  We realize however that
even part time exposures for certain chemicals are important to
consider.

So in most cases when we examine apparent discrepancies, it turns out
that there is some validity to our estimates.
                                   748

-------
Dr. Bellin (EPA):  Down to what level  do manufacturers identify their
products, and is there any attempt to gain information on hazardous
contaminants that we have some idea might exist?

Mr. Sundin (NIOSH):  That is something we will  probably be examining
closely, with an eye to potentially modifying during the second survey.
We asked the manufacturers to report anything that was there at greater
than one percent by weight or volume, except in the cases of carcinogens
which should be reported regardless of the percentage -- but the
definition of carcinogen was not presented to manufacturers, and I am
sure they interpreted it fairly liberally.

We have seen cases where a pesticide, for example, will come back as 100
percent petroleum distillate; the active ingredient being in the product
at less than one percent, and therefore the manufacturer is choosing to
ignore it.  We are therefore contemplating a change in the wording of
our request form to include a request for total reporting of any active
ingredients and impurities where the manufacturers have such knowledge.

We permitted also a plus or minus 5 percent margin on the percentage of
composition reporting.  That does not affect the way in which we use the
data back in NOHS.  I think it is probably an adequate level of
reporting and gives us fewer problems with certain manufacturers if they
are permitted to use a plus or minus 5 percent tolerance on individual
percentages of the components.
                                   749

-------
        "Extent of  Industrial Exposure to Epichlorohydrin,
               Vinyl Fluoride, Vinyl Bromide and
                      Ethylene Dibromide"
                        Talk Presented at
            First NCI/EPA/NIOSH Collaborative Workshop
                       Rockville, Maryland
                           May 8, 1980
                               By
                          James L. Oser
                       Industrial Hygienist
                    Industrial Hygiene Section
                   Industry-wide  Studies  Branch
     National Institute  for  Occupational  Safety and Health
Division of Surveillance,  Harard  Evaluations, and Field Studies
                        Cincinnati,  Ohio
                                  750

-------
                               ABSTRACT
Industrial hygiene studies were conducted in several industries to evaluate
present levels of worker exposure to hazardous chemicals and to determine
the feasibility of worker mortality studies.

The greatest exposure to epichlorohydrin in the chemical and resin manufac-
turing processes occurs with chemical operators.  In the chemical  manufac-
turing processes, operator exposures were found to be less than  2.1 ppm for
20 TWA personal samples and in the resin manufacturing processes, operator
exposures were found to be  less than 0.8 ppm for 39 TWA personal samples.

Chemical operator exposure levels for vinyl fluoride in the manufacture and
polymer processes were generally found to be less than 5 ppm for  18 TWA person-
al and general area samples. One sample was found to be 21 ppm  during the
initial start-up  shift. Operator exposure to vinyl bromide in the manufacturing
process was found to be less than 0.4 ppm for 4 TWA personal samples.  Tank
car loading crewmen were found  to have the greatest exposure,  and 1.2 and
6.3 ppm VBr, respectively for two TWA personal samples taken.

Ethylene dibromide median  exposures by similar job types in the manufacturing
processes ranged 10 to  500 ppb for 35 TWA personal samples; while, median
exposures by similar job types in  antiknock blending operations ranged 0.2
- 54 ppb for 39  TWA personal samples.  EDB ceiling levels for quality control
sampling and for loading or  unloading tank cars ranged 0.04 to 23 ppm for
7 samples and 0.09 to 2.4 ppm for 4 samples,  respectively.

The reported levels are generally within permissible exposure limits.  Many
of the processes are closed  chemical processing which minimize worker exposure.
However, continued attention needs to be given to the open system processes
and equipment maintenance to prevent worker exposure.
                                    751

-------
I. INTRODUCTION

The purpose of this paper is to present and summarize the pertinent
findings of industrial hygiene studies of epichlorohydrin, vinyl
halides (vinyl fluoride and vinyl bromide) and ethylene dibromide.
Study of these substance agents have been conducted by NIOSH under
contract with Tracer Jitco, Inc. and SRI, International. More com-
plete information concerning the findings of these studies is con-
tained in the published NIOSH technical reports: (1) Epichlorohydrin
Manufacture and Use - Industrial Hygiene Survey, (2) Vinyl Fluoride
and Vinyl Bromide Industrial Hygiene Survey Report, and (3) An In-
dustrywide Industrial Hygiene Study of Ethylene Dibromide.

II. STUDY PROCEDURE

These chemical agents were selected by NIOSH for industrial hygiene
study because they are suspect cancer agents and limited information
exists on exposure levels in industry.

Walk-through surveys were conducted for each facility to obtain pre-
liminary information about the process, production activity, poten-
tial exposures and worker exposure. Based on the walk-through survey,
in-depth survey requirements are defined for each plant. Essentially,
the in-depth survey procedures are directed toward determining 8-hour
TWA personal exposure and characterizing work practices and controls.
Sampling is conducted for all shifts (usually three) and potentially
exposed workers. Job descriptions are characterized for potentially
exposed workers. Appropriate survey sampling and analytical methods
are determined. The NIOSH recommended sampling and analytical methods
formed the basis for methodologies employed in the study. Recommended
vinyl fluoride sampling methods were not available at the time of  the
Study. Personal sampling for vinyl fluoride was conducted using a  7.7
liter Teflon bag attached to the sampling pump operating at flow
rates of 14 to 100 cc/min. The samples were analyzed by gas chroraato-
graphy within two days after,sampling. Laboratory decay tests indica-
                                   752

-------
ted 107. loss in four days and* 507. loss in two weeks for the Teflon
bag used.

All other samples were collected on the standard 150 mg charcoal  tube
and analyzed by gas chromatographic techniques.

III. Epichlorohydrin

Studies were -conducted at two epichlorohydrin manufacturing and user
facilities and at three additional resin manufacturing facilities in
1976.

A. HEALTH CONCERN

The current Federal Standard  for epichlorohydrin is 5 ppm  (19 mg/m  )
as an 8-hour time-weighted average permissible exposure level. This
is based en the known acute (short term) health effects to humans
from over exposure, i.e., respiratory tract  irritation and systemic
poisoning. Skin contact can result in dermatosis and systemic
effects. Exposures occur principally by inhalation and direct skin
contact and to a less extent  by ingestion. After a comprehensive
literature review, NIOSH concluded that exposure risks may include
carcinogenasis, mutagenesis,  and sterility in humans and recommended
a time-weighted average occupational exposure limit of 2 mg/m  of air
(0.5 ppm) and a ceiling limit of 19 mg/m   (5 ppm) as reported in  the
Criteria Document (Sept. 1976). More recent  information on human  expo-
sure data lias prompted NIOSH  to issue a Current Intelligence Bul-
letin-30 recommending exposures be reduced to the extent feasi-
ble. (Oct. 1978) The 1978 TLV for epichlorohydrin has been revised
downward to 2 ppm for the 8-hour TWA and a short term exposure level
of 5 ppm.

B. SURVEY RESULTS

   1.Epichlorohydrin Manufacturing Process

                                    753

-------
Manufacturing plants for epichlorohydrin are of conventional open
structure chemical process design. Both facilities manufacture
epichlorohydrin for shipment and for use in their own manufacture of
glycerine and epoxy resins.

Epichlorohydrin is made by the chlorination of allyl chloride
yielding a mixture of dichlorohydrin. These products are washed with
a cold dilute alkali solution to remove .hydrochloric acid and yield
impure epichlorohydrin. The epichlorohydrin is then further refined
by distillation processes.

Survey Data

Plant A

The  results of 8-hour TWA personal sampling at the first facility
ranged from not detected (less than 0.05 ppm) to 0.4 pprn for 8 chemi-
cal  opera'.or samples (median of 0.3 ppm). Lower exposures were re-
reported for the shift foreman, drumming operator and maintenance per-
sonnel. An exposure level of 0.3 ppm was reported for the tank car
loader.

Plant B

Similar survey data was collected at the second manufacturing
facility. Personal samples ranged from not detected (less than 0.05
ppm) to 2.1 ppm for 12 chemical operator samples (median of 0.1 ppm).
However, two of these samples were considered   significant levels,
i.e., 1.9 ppra and 2.1 ppm. Lower exposures were reported for the
shift and maintenance foreman. An exposure of 0.3 ppm was reported
for  the tank car loader. No drumming operations were performed during
the  survey of Plant B.

Controls
The  production of opichlorohydriii  operations  arc  located out of doors
                                  754

-------
with automated operations monitored from a control room. For normal
Operation, operators are seldom in the process areas. They are in  the
production areas Cor routine inspection of equipment, product
sampling and occasional on-stream maintenance. Maintenance work  is
supervised by the shift foreman with work practices in effect. Pumps
and pipe flange seals arc maintained to prevent leakage.

Tank cars and trucks are cleaned and inspected before loading by out-
side contractors. Loading involves connecting supply lines and vent
lines, loading tank car in which the operator is  located at a control
site some distance away and the disconnecting once the tanks are
filled. The most significant exposure to the loader is during the  dis-
connect procedure. The loading operation takes about 1\ hours with
about \ hour for connect and disconnect time. Drumming operations  em-
ploy local exhaust ventilation systems.

2. Epoxy Resin Manufacture

Five epoxy"resin facilities were surveyed for epichlorohydrin expo-
sure. The first two are also manufacturers of epichlorohydrin.

Process

The basic process involves reacting epichlorohydrin with bisphenol A
under alkaline conditions. Depending on the particular resin being
produced, reactants and/or solvents may be introduced to modify  the
resin properties and the viscosity of the liquified resin  products.
The reactants, depending on resin specifications  may. include bi-
sphenol A, tetrabromo bisphenol A, o-cresol, paraformaldchyde,
caustic soda, oxalic acid, and p-tertiary butyl phenol. The  solvents
may include methyl ethyl ketone, methyl isobutyl  ketone, acetone,
toluene, and xylene. Epoxy resins may be produced as thcrmpset
(cured) resins, containing no free or unreacted epichlorohydrin.
Survey Data

Plant A & B                          '55

Exposure to epichlorohydrin were  found.to be less  for the epoxy resin

-------
production operations.  Resin process chemical operator exposures were
found to be less than 0.8 ppm for 22 TWA personal samples taken in
the two epichlorohydrin manufacturing facilities. Lower exposures are
reported for foremen and maintenance.

Plants C, D and E

At three additional epoxy manufacturing facilities, chemical operator
exposures were less than 0.2 for 17 TWA samples. The highest level re-
ported (1.5 ppm) is that for a general area sample taken in the
Isolated pump room at Plant C. This demonstrates the potential for ex-
posure from equipment leakage.

Control

Epoxy resin processes are located both indoors and outside, with a
variety of natural, general and local types of ventilation control.
Reaction kettles are equipped with vent lines. Epichlorohydrin was
fully reacted in processes studied so that packaging and shipping of
finished products did not present epichlorohydrin exposure to workers.

IV. Vinyl Halides

Studies were conducted for vinyl fluoride exposure at a manufacturing
facility and at a polymerization facility. Vinyl bromide was studied
at a manufacturing facility.

Health Concern

The vinyl halides including vinyl fluoride and vinyl bromide have
gained prominent health concern as a result of the recently reported
cancer associated with vinyl chloride workers, in the absence of de-
finitive Federal Standards for vinyl fluoride and vinyl bromide,
NIOSH recommends tliat worker exposures be minimized and that levels
be maintained below the established Federal Standard for vinyl

                                   756

-------
chloride. The Federal Standard for vinyl chloride is one ppm as an 8-
hour TWA exposure limit. The 1978 ACGIH Threshold Limit Value for
vinyl bromide has been revised downward for the 8-hour TWA exposure
limit.

B. Survey Results

   1. Vinyl Fluoride Manufacturer Process

Plant A  is a manufacturer of vinyl fluoride. The basic process in-
volves a pressurized reaction of hydrofluoric acid and acetylene. Di-
fluoroethane is formed as an intermediate product that is cracked to
yield ethylene fluoride and hydrogen fluoride. The reaction products
are refined by distillation and the off-products are recycled to the
process  stream. Liquid vinyl fluoride is piped to insulated storage
tanks and from there to insulated tank cars for shipment.

Survey Results and Control

Sampling results at Plant A were generally not detected due to inter-
ference  from difluoroethane. Difluoroethane was measured to be 5 ppra
and less. The vinyl fluoride was estimated to be less than 2 ppm for
7 personal and general area samples. One TWA personal operator sample.
was 21 ppra for the start-up process on  ;he first shift. This demon-
strates  that during a work shift with abnormal or unusual work opera-
tions, the potential.for greater exposure exists. With continued
caution  on the part of the operators and the use of respiratory pro-
tection] protective clothing and face shields during these opera-
tions, the actual worker exposure is minimized.

2. Vinyl Fluoride Polmerization Process

Plant B  is a manufacturer of polyvinyl  fluoride. Vinyl fluoride  is  re-
ceived,  transferred to storage tanks and piped to the vinyl fluoride
polymerization building. The monomer is continuously pumped to a

                                    757

-------
supply tank at the process building and from there injected into
water and pumped to the reactor to which an aqueous solution of the
reaction initiator is simultaneously added. The reaccor is barricaded
from the remainder of the processing area since the reaction is con-
ducted under high pressure carefully controlled. The reacted vinyl
fluoride, a finely divided precipitate, is separated from the reactor
aqueous liquor. Unreacted vinyl fluoride is recycled to the supply
tank. The polymer is stored as a. 5% aqueous slurry and fed to a
rotary filter. The process is a closed system until the slurry
(completely polymerized) is fed to the rotary filter.

The resulting filter cake (white odorless product) is further dried,
collected in bag filters, classified and stored for further pro-
cessing.

Survey Results

Sampling results for the vinyl fluoride polymer plant, Plant B,
          indicate TWA exposure levels below 5 ppra vinyl fluoride.
Samples ranged from 1 to A ppm with a median of 2 ppm for 7 TWA per-
sonal samples of polymer operators. One general area sample in the
pump room was 5 ppm vinyl fluoride.

3. Vinyl tromide Manufacturing Process

Plant C is a manufacturer of vinyl, bromide. Ethylene dibromide is con-
tinuously fed to a reactor where caustic reacts with the dibromide
and yields vinyl bromide. The unreacted ethylene dibromide is removed
by distillation and recycled. The vinyl bromide is pumped to storage
tanks and transferred to railroad cars for shipment. The process is
enclosed and the plant structure is open and outside. The process con-
trol instrumentation is located in a separate building, where opera-
tors monitor the process.

Survey Data

                                   758

-------
Operator exposures Co vinyl bromide in Plant C range from O.I to 0.4
ppm with a median of 0.3 for 4 samples. Higher levels were associated
with the laboratory analysis and loading operations. A level of 6.3
ppm for a one hour sample during loading vinyl bromide tank cars was
reported. A level of 1.2 ppm was reported for the 8-hour TWA sample.

Control

The primary control in these processes involve enclosed processes and
process equipment located out of doors. The integrity of process
systems are maintained to present minimal exposure hazards to plant
workers.
Certain potential exposure situations such as sample collection,
coupling or decoupling tank lines and maintenance of process equip-
ment require specialized controls and work practices.

V.'Ethylene Dibroraide

Two manufacturing and two blending operations were studied for expo-
sure to ethylene dibromide.

A. Health Concern

The current Federal Standard for ethylene dibromide is 20 ppm deter-
mined as an 8-hour time-weighted average occupational exposure level.
This is based on the acute health effects to the respiratory tract
and systemic poisoning. Skin contact may result in skin irritation
and systemic effects. Exposures occur principally by inhalation and
skin contact and to a less extent by ingestion. A comprehensive re-
view of the literature by N10S1I of animal studies indicate reproduc-
tive effects, carcinogcnicity, mutagenicity and teratogenicity. Based
on these adverse effects possibly associated with human exposure,
                                 759

-------
N10SI1 recommended an employee exposure ceiling of 0.13 ppm
(1.0 mg/m ) in the Criteria Document of August, 1977. Based on the
preliminary results of animal studies of a toxic interaction between
disulfiram and ethylene dibromide, NIOSH published a Current Intelli-
gence Bulletin-23, April 11, 1978.   NIOSH recommended that workers
should not be exposed to ethylene dibromide during the course of di-
sulfiram therapy, used for treatment of alcoholics in industry. The
TLV committee recognizes the carcinogenic potential of EDB as re-
ported in animal studies and is considering a reduced recommended ex-
posure limit.

B. Survey Results
   1. EDB Manufacturing Process

Plants A and B are manufacturers of ethylene dibromide in continuous
flow, closed system operations. Ethylenc dibromide is produced by the
exothermic reaction of bromine and ethylene in a countercurrent flow
reactor. With bromine entering the top and gaseous ethylene entering
the bottom, the  reaction occurs in the upper portion of a column
packed with ceramic chips. The crude liquid product is further re-
fined, stored and loaded into tank cars for shipment.

Survey Data

The results of 8-hour TWA personal sampling at two manufacturing oper-
ations range from not detected (less than 0.02 PPB) to 1600 PPB. The
higher levels were reported for the surveillance technician and lab
technician at Plant A and the crew leader and product leader at Plant
B. For normal plant operations these higher levels of exposure to
ethylene dibromide result from open system operations.

Control

These-processes  are operated and monitored from remote control rooms.

                                   760

-------
Potential exposures to operating personnel occur for quality control
sampling and analysis, loading and maintenance operations.

2. EDB Blending Process

Plants C and D produce antiknock blends using ethylene dibromide.
Antiknock blends consist of homogeneous mixtures of ethyiene dibro-
mide, ethylene dichloride, tecraalkyl lead and may contain toluene
and a dye. Raw materials are generally received by tank car, stored
and pumped to the process area. Blending is performed in outdoor
blending tanks. The batch-type process is activated and monitored
from a control room. Manual operations involving potential worker ex-
posure include loading and unloading tank cars, quality control sam-
pling and analysis and drum loading.

Survey Data

The results of sampling for various operations for antiknock blending
operations indicate lower overall exposures than occur in manufac-
turing. The results of 8-hour TWA personal sampling for two plants
ranged from 0.1 to 8.2 ppb for 39 samples. Again, the higher expo-
sures result from open system operations such as loading and un-
loading,  Jrum cleaning, and sample collection. Drum loading is con-
ducted using enclosed local exhaust hoods. The drums are reusable and
undergo cleaning prior to reuse. Drum cleaning is conducted in an en-
closed exnaust hood and the operators wearing respiratory protection.

3. Ceiling Levels

The final slide presents  short term exposure data for quality control
sampling, loading and unloading of tank cars at the various plants.
These levels are expressed in ppm rather than ppb for the two prior
slides.

The sample collection process varies from 5 to 30 minutes depending-

                                    761

-------
on the number and location of sampling. Process scream sampling  is
conducted generally at the beginning of each shift. Tank car  loading
requires about one to two hours and may be sporadic depending on ship-
ment schedule.

Most of these levels are above the NIOSH Criteria Document  recom-
•mended ceiling level of 0.13 ppm for a 15 minute period.

VI. Conclusion

The reported  exposure levels are considered typical of the  industries
represented.  The primary control for each operation is the  closed pro-
cessing of chemicals. This is      typical of  the chemical  processing
industry in general. Particular attention needs to be given to those
operation and processes which can not be conducted as closed  pro-
cesses and routine maintenance to detect and correct leaks  or acciden-
tal chemical  releases. The potential for health effects associated
with the materials studied is significant to warrant application of
ventilation systems, personal protection and safe work practices.
                No discussion followed this paper.
                                 762

-------
               EPICHLOROHYDRIN
       STANDARD             5 PPM
NIOSH CRITERIA DX,         0.5 ?m         5 PPM
1978 T.L.V,                 2 Pffl         SPPf^l
                       763

-------
   PLANT /  JOB
PLANT  A:
 CHEMICAL OPERATORS
 SHIFT  FOREMEN
 DRUMMING OPERATOR
 TANK  CAR LOADER
 PIPEFITTER
 (MAINTENANCE)
              TABLE 1.1
EPICHLOROHYDRIN SAMPLING DATA SUMMARY
     EPICHLOROHYDRIN  MANUFACTURERS
      8-HR. TWA PERSONAL SAMPLES
   SAMPLE SIZE        RANGE (PPM)
         8
         3
         2
         I
         3
N.D. -  0.4
O.I  -  0.2
0.06 -  0.08
     0.3
     N.D.
MEDIAN (PPM)

   0.3
   O.I
   0.07
   0.3
   N.D.
    N.D. - NOT DETECTED BASED ON THE SAMPLING AND ANALYTICAL METHOD.

-------
                                     TABLE 1.2
                       EPICHLOROHYDRJN  SAMPLING  DATA SUMMARY
                            EPICHLOROHYORIM MANUFACTURERS
                             8-HR.  TWA PERSONAL  SAMPLES
       PLANT/JOB        SAMPLE SIZE        RANGE (PPM)        MEDIAN (PPM)
   PLANT  B:
     CHEMICAL OPERATORS           12           N.D. -  2.. I            O.I
     SHIFT  FOREMEN              3            N.D  -  0.3            N.D.
     TANK  CAR LOADER            I                0.3               0.3
     MAINTENANCE'FOREMAN         I                0.08              0.08
01
        N.O. -  NOT DETECTED  BASED ON THE  SAMPLING AND ANALYTICAL. METHOD.

-------
                                     TABLE  2.1
                     EPICHLOROHYDRIN  SAMPLING  DATA SUMMARY
                                RESIN  MANUFACTURERS
                           8-HR. TWA PERSONAL  SAMPLES
     PLANT/JOB        SAMPLE  SIZE         RANGE (PPM)        MEDIAN (PPM)
   PLANT  A:
     CHEMICAL OPERATORS           6            N.O.  -0.4           N.O.
   PLANT  B:
     CHEMICAL  OPERATORS          16           N.D.  - 0.8           0.04
S   OPERATING FOREMEN           5            N.D   - 0.6           N.D.
     MAINTENANCE  FOREMEN         I                N.D.              N.D.

      N.D. -  NOT DETECTED  BASED  ON  THE SAMPLING AND ANALYTICAL METHOD.

-------
    PLANT/ JOB
  PLANT  C:
    CHEMICAL OPERATORS
    G.A.IN  PUMP ROOM
  PLANT  D:
^  CHEMICAL OPERATORS
o
""PLANT  E:
    CHEMICAL OPERATORS
    OPERATING FOREMEN
    RESIN  FINISHING FLARER
                TABLE 2.2
EPICHLOROHYDRIN  SAMPLING  DATA SUMMARY
          RESIN  MANUFACTURERS
      8-HR.  THA  PERSONAL  SAMPLES
   SAMPLE SIZE        RANGE (PPM)
           I
          2

         14
          I
          2
    0.09
    1.5

0.05 -  0.15

    N.D.
    N.O.
    M.O.
MEDIAN (PPH)

    0.09
    1.5

    O.I

    N.D.
    N.D.
    N.D.
      N.D. - NOT DETECTED  BASED  ON  THE SAMPLING AND ANALYTICAL METHOD.

-------
                 VINYL HALIDES
VINYL FLUORIDE:                      TWA
     FEDERAL STA'JDARD
     NIOSH CRITERIA DOC,              1 PPM
     1978 T.LV.

VINYL BROMIDE:
     FEDERAL STANDARD
     NIOSH CRITERIA DOC,              1 PPM
     1978 T.LV.                     5 PPM
VINYL CHLORIDE:
     FEDERAL STANDARD                1 PPM
     NIOSH CRITERIA DOC.              1 PPM
     1978 T.LV,                     SPffl
                         768

-------
                                     TABLE 3.1
                       VINYL  HALIDE SAMPLING DATA SUMMARY
                                  VINYL  FLUORIDE
                               8 HR .   TWA  SAMPLES
      PLANT/JOB/LOCATION  SAMPLE  SIZE         RANGE (PPM)      .  MEDIAN (PPM)
   PLANT  A:
    PLANT  OPERATOR             4                 N.D.              N.D.
    PLANT  OPERATOR             I                 21               21
   (START-UP PROCESS)
^  G.A. IN  CONTROL ROOM         3                 N.D.              N.D.
* PLANT B:
    POLYMER OPERATOR            7                 H              .2
    G.A. IN  SUPERVISOR'S OFFICE     3                 1-2                2
    G.A. IN  PUMP ROOM             I                 5                 5

       G.A.-GENERAL AREA  SAMPLES.
       N.D.-NOT DETECTED BASED-OH SAMPLING AND ANALYTICAL METHOD.

-------
                                  TABLE 3.2
                   VINYL HALIDE  SAMPLING DATA  SUMMARY
                               VINYL  BROMIDE

   PLANT/JOB/LOCATION   SAtiPLE SIZE'       RANGE (PPM)       MEDIAN (PPM)
PLANT  C:
  PLANT OPERATOR             4             0.1-0.4           0.3
  LA8. TECHNICIAN             2             0.3-0.5           0.4
  LOADING CREWMAN            I                1.2              1.2
  LOADING'CREWMAN            I                6.3              63


          RESULTS ARE 8-HR. SAMPLES UNLESS INDICATED.

-------
                 ETOYLBE DIBRQMTOE
                            M        £E1UML
       STANDARD           20 PPM            -
NIOSH CRITERIA DOC,           -             0,13 PPM
1978 T.LY.              -GO).* PPM

* SUSPECT CARCINOGEN /WAITING REASSIGNMENT
                        771

-------
                                 TABLE 4.1
                  ETHYLENE DIBROMIDE SAMPLING  DATA SUMMARY
                     8HR. T¥/A  PERSONAL SAMPLE  RESULTS
                            EDO  MANUFACTURING
   PLANT/JOB        SAMPLE  SIZE        RANGE (PPB)       MEDIAN (PPB)
PLANT  A:
  CONTROL ROOM  OPERATOR      4             20-140           80
  SURVEILLANCE TECHNICIAN      8            N.D.  - 1600          370
  LAB. TECHNICIAN             4            N.D.  -  570          140
  BRIIJE FIELD TECHNICIAN      4            N.D.  -   30           10
PLANT  B:
  CONTROL ROOM OPERATOR      7              3-160           40
  CREW LEADER               2             40  -  950          495
  PRODUCT LOADER             4             50  -  620          360
  LABORATORY TECHNICIAN       2             10-80           45

     N.D. - NOT DETECTED  BASED  ON SAMPLING AND  ANALYTICAL METHOD.

-------
                                   TABLE  4.2
                   ETHYLENE DIBROHIDE  SAMPLING  DATA SUMMARY
                      8HR. TWA  PERSONAL  SAMPLE  RESULTS
                          EDB    BLENDING     OPERATIONS
-si

Co
  PLANT /  JOB        	
PLANT  C:
  BLEND OPERATOR
  LAB. TECHNICIAN
  SHIFT  SUPERINTENDAMT
PLANT  D:
  BLEND OPERATOR
  RELIEF  OPERATOR
  REACTOR OPERATOR
  DRUM  LOADER
  DRUM  PROCESSING
  RAW MATERIAL  HANDLER
  LAB. TECHNICIAN
  COMPOUND  BULK OPERATOR
SAMPLE  SIZE
       5
       6
       3

       3
       2
       2
       4
       3
       2
       4
       2
                                           RANGE (PPB)
4 -
0.2 -
O.I -
1 -
0.5 -
1 -
8 -
12 -
.27 -
58
12
0.4
9
7
3
18
36
82
                                            0.1-0.5
                                              1-8
MEDIAN (PPB)

     22
      4
     0.2

      6
      4
      2
     14
     16
     54
     0.4
      4
       N.D.  - NOT  DETECTED BASED ON SAMPLING AMD ANALYTICAL METHOD.

-------
                                  TABLE 5
                     EDB CEILING LEVEL  PERSONAL SAMPLING
                      (SAMPLE TIME VARIES  1 TO 20 WIN.)

    TASK/PLANT         SAMPLE SIZE        RANGE  (PPM)       MEDIAN (PPM)
QUALITY CONTROL SAMPLING:
       PLANT  A              3            5.3  -  23.4          12.0

       PLANT  B              2            0.3  -  0.5           0.4

3     PLANT  C              I                 1.5               1.5

       PLANT  0              2            0.04 -  0.7           0.4

LOADING TANK  CAR:
       PLANT  D              I                  O.I               O.I

UNLOADING TANK CAR:
       PLANT  D              I                  1.6               1.6

-------
           ENVIRONMENTAL LEVELS AND URINE
       CONTENT OF WORKERS EXPOSED TO AZO DYES
                  Mark F. Boeniger
                   Larry K. Lowry
                  William P. Tolos
    Department of Health, Education, and Welfare
National Institute for Occupational  Safety  and  Health
             Robert  A. Taft  Laboratories
               Cincinnati, Ohio 45226
                   Charles R.  Nony
                   Malcolm Bowman
     Department of Health, Education and Welfare
            Food  and Drug Administration
     National Center for Toxicological Research
              Jefferson,  Arkansas 72079
                           775

-------
                              ABSTRACT










Benzidine has been agreed to by both industry and government as being




a proven human bladder carcinogen. Henceforth, the use of benzidine and




the handling of it has either been curtailed or its exposure to workers




greatly minimized. However, the azo dye products of benzidine have re-




ceived little serious concern and have for  at least  75 years been con-




sidered as being innocuous. Chemically and biologically, however, the




azo bond is quite labile to reductive cleavage.










Because of recent suggestive evidence that these dyes may be broken down




to their component amines in the body, NIOSH initiated field studies




into the dye manufacturing and dye consuming industries where potential




exposure to benzidine derived dyes were suspected. Both environmental




and biological urine sampling was performed in order to evaluate actual




exposure and the excretion of benzidine and its metabolites, hypoth-




esizing that the dyes are metabolized to benzidine in vivo.  The findings




of this study are presented for each of the six field surveys performed




and the results are discussed briefly.










Based on this and other evidence, NIOSH has recently recommended that




benzidine derived azo dye be treated as carcinogens and their manufac-




ture and use be discontinued.
                                    776

-------
INTRODUCTION










During the relatively brief history of the synthetic dye industry, ben-




zidine has undoubtedly played a major role. Much of what we now know




about dye chemistry was laid down between 1870 to 1910, a period which




is sometimes called the classical period of dye chemistry. In 1884,




P. Boettiger discovered Congo Red. This was the first direct cotton dye




derived from benzidine (1). Since then more than 200 benzidine dyes have




been listed in the Colour Index (2). In 1948, for instance, some four




million pounds of benzidine and 31 million pounds of benzidine derived




dyes were produced (3). This accounted for 21% of all dyes reported to




be manufactured and almost all of the direct class dyes on the market




in that year. While there is still a good demand for these dyes by




industries, most manufacturing plants have stopped producing dyes made




from benzidine, primarily because of environmental and health concerns.




Today, there is only one domestic producer of these dyes; however, im-




ports have risen appreciably. Benzidine derived dyes now used in the




U.S. represent less than 1% of the at least 1200 different dyes made




in the United States, and the additional 800 dyes imported (4,5).




However,  benzidine based Direct Black 38 has remained the single largest




dye produced among all dyes.










During 1973 the proportional use of benzidine derived dyes, by industry,




was estimated to be: 40% used to color paper, 25% to color textiles,




15% for leather, and 20% for diverse applications (6). They may be used
                                   77
                                      t-w

-------
by crafts,artists,  and the general public (7).









The object of the present study was to characterize the industrial en-




vironment in terms of worker exposure to benzidine derived dyes and to




monitor their urinary excretion of benzidine and its metabolites.









A thorough literature review of the pertinent information on benzidine




derived dyes has been recently published by NIOSH as a Special Hazard




Review (8). In addition, the composite report for this study is to be




published shortly (7).
STUDY DESIGN









Procedures had been developed for determining the concentrations of




airborne dye exposure and for quantitating the urinary excretion of




benzidine and its metabolites. It was desirable to locate industrial




facilities where the probable exposure was relatively pure, e.g. where




the primary chemical exposures during the work day were to benzidine




dyes. With the cooperation of the American Textile Manufacturers Associa-




tion and the Tanners Council of America, as well as other various sources




of information, prospective facilities using benzidine dyes were queried




and selected. In all, two textile dyeing and finishing facilities, a




leather tanning facility and a specialty paper company were surveyed.




The two benzidine dye manufacturing facilities operating at that time




were also surveyed.






                                     778

-------
Employees in each facility were selected with the aid of the managements.




Each employee who participated was monitored for personal airborne ex-




posure and urinary excretion of benzidine.
ENVIRONMENTAL SAMPLING AND ANALYSIS










Personal airborne dyestuffs exposure was determined by sampling a known




volume of air through a pre-weighed closed face 37-mm glass fiber filter




in a three piece cassette. The sampled air was drawn through the filter




by a calibrated personal sampling pump at approximately 1.8 liters per




minute.










Bulk samples of the benzidine dyes used by the workers during the surveys




were obtained for determination of residual-free benzidine present as




both the base and salt.










Air filter samples were analyzed gravimetrically, which is a routine




procedure designed to measure total gross particulates in the air. As




a supplementary procedure for a quasi-specific method for identifying




the proportional quantity of a benzidine dye on the same filter sample,




NIOSH method P&CAM 234 was used (9). The principle of this method is




that a sample filter is extracted with an appropriate solvent, and a




spectrographic scan of the solution is performed in the 400-700 nm range.




The absorbance maxima are compared to the absorbance maxima of standard
                                      779

-------
solutions prepared from the bulk samples of the azo dyes. This procedure




was of use in the present study since only one to four different color




dyes were used during most of the surveys over any one day.










Bulk samples of the dyes were analyzed for residual free benzidine and




its salts using a liquid chromatographic procedure using a 280 nm UV




detector. Studies have shown that this method yielded a recovery of




approximately 100% for benzidine and its salts. The detection limit for




benzidine is 1 ppm (w/w) from a one gram sample (7).
BIOLOGICAL SAMPLING AND ANALYSIS










Urine samples were collected in 180 rt& polyethylene screw top bottles.




The period of collection usually began during the beginning of the first




shift that was monitored and continued (excluding the non-work period)




into the middle of the next day. However, monitoring periods were ex-




tended to include up to a five day period, depending on the dye usage




period. The time was recorded on each sample submitted, which was frozen




immediately on dry ice and remained frozen until analyzed. Many samples




were split so that they could be analyzed by both NIOSH at the Clinical




and Biochemical Support Section, Division of Behavioral and Biomedical




Sciences, and the Chemistry Division, National Center for Toxicological




Research (NCTR).
                               780

-------
Urines were analyzed by a screening colorimetric method adapted  for use




on human urines. The procedure was based on the pH5 extraction of




aromatic amines from urine with chloroform, back extraction  into HCL




and reaction with 2,4,6-trinitrobenzene sulfonic acid  (TNBS) to  produce




a yellow chromophore absorbing at 400 nm. The sensitivity of the method




was 1 ppb based on a 100 ml urine sample. The presence of benzidine could




be confirmed by thin layer chromatography if concentrations  of benzidine




exceeded 3 ppb. Complete details of the method have been published in




Volume 5 of the NIOSH Manual of Analytical Methods  (10).










Selected specimens were also analyzed by the electron-capture gas chro-




matograph method described earlier at this workshop in a presentation




by Lowry and reported by Nony et. al. (11,12). This method was used to




confirm the presence of specific metabolites and benzidine.  The  lower




limit of detection for benzidine (Bzd) was 1.4 ppb  and monoacetylben-




zidine (AcBzd) 5.8 ppb. The method was also used to detect 3,3-dimethyl-




benzidine (DiMeBzd) and 3,3-dimethoxybenzidine (DiMxBzd) at  a lower




detection Ir'r.jt cf 3 ar.d 3.6 ppb, respectively. The identity of




metabolites was confirmed by chemical synthesis of  metabolites and deriv-




atives followed by gas chromatography-mass spectrometry (GC-MS)  analysis.




Confirmation was also obtained to establish that these amines were in




fact metabolites and not a result of chemical reduction of the intact




dye by the analytical procedures.
                                        781

-------
RESULTS










Air filter samples were analyzed and reported as milligrams of total




airborne particulate per cubic meter of air sampled. Detailed results




of the environmental and spectrophotometric analysis on the samples are




presented elsewhere (7).










Urine samples from 23 NIOSH office workers were submitted for analysis




with the colormetric/thin layer chromatography procedure used to screen




workers urine. Table I indicates that fewer than 3570 excreted one or




more ppb aromatic amine. No benzidine was detected. These results from




this non-exposed group were useful for comparison purposes.










 In the first dyestuff plant that was visited only spot urine samples




were collected during a walk-through survey. Eight of thirty four dye-




stuff workers who were potentially exposed to the finished dye submitted




urine samples. Using the screening method, less than 1 ppb aromatic amine




was found in eight of ten workers. No benzidine was detected; however,




3 and 7 ppb monoacetylbenzidine was reported in two workers as shown




in Table II. A return survey was not possible due to the discontinuance




of benzidine dye manufacture at that facility.










The low level of aromatic amine excretion among these workers might be




expected since cartridge filter respirators and local ventilation were




commonly employed in this process.
                                    782

-------
                              Table II
              Urinary  Excretion  in Dye Manufacturer  I

Urine       Benzidine or                Aromatic Amines    Thin-Layer
Specimen    Monoacetylbenzidine (MAB)   (ng/100 mL)	Chromatography
1
2
3
4
5
6
7
8
3 ppb MAB
N.D.
N.D.
N.D.
N.D.
N.D.
7 ppb MAB
N.D.
120
80
100
80
90
80
90
60'
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.—not detected


The second survey to the other dye manufacturing facility discovered

the highest worker exposures of the entire study. A small company with

minimal engineering control of the process, no formal respirator program,

and no industrial hygiene or medical programs were implemented. Of about

55 production workers, about 34 were likely to be potentially exposed.

While environmental monitoring data was collected in all dye finishing

departments, employee cooperation was poor and only four provided inter-

mittent urine samples. All urine samples contained benzidine and/or mono-

acetylbenzidine. Two workers also excreted 3,3'-dimethylbenzidine

(o-tolidine) which was also used to make dyes. Environmental and urine

monitoring results are tabulated for the four workers in Table III. Bulk

samples of eleven dyes known or suspected to be benzidine derived were

collected for residual benzidine analysis. All contained less than 20

ppm benzidine as the amine or salt.



The following results are from two dye finishing areas at two textile

manufacturing facilities.
                                  783

-------
In the first facility surveyed, personal exposure concentrations were




determined while the urine samples were analyzed by both the screening




procedure and the EC-GC method. Benzidine and/or monoacetylbenzidine




were found in three of seven workers who were monitored. Urinary concen-




tration of non-specific aromatic amines were all above 1 ppb and gen-




erally considerably above the NIOSH comparison data. Table IV summarizes




these results. Two bulk samples of the dyes used were found to contain




1 and 4 ppm residual total benzidine.










In the second textile facility visited, ten workers were monitored. None




of the urine samples analyzed contain benzidine; however, one contained




4 ppb monoacetylbenzidine. The screening method indicated that 40% of




the workers excreted one ppb or more aromatic amine in their urine. En-




vironmental samples indicate that daily worker inhalation exposure in




the dye room and otherwheres was less than 1.5 mg/m . Bulk samples of




the dye contained up to 20 ppm (w/w) total benzidine. Table V summarizes




the results of the monitoring data.










Employees at a leather facility were also monitored. Only Direct Brown 95




was used at the time of the survey. Three employees with the highest




likely exposure were monitored as in the other facilities. None of the




urine samples collected from the dye weigher or the two dye drum opera-




tors contained any detectable benzidine or monoacetylbenzidine. It is




probable that appropriate work practices including wearing of respir-




ators, eating in clean lunch facilities, using shower and wash facilities
                                    784

-------
after work, and changing out of work clothes after work probably con-




tributed to these results. Sampling results are summarized in Table VI.










The last facility surveyed produced colored specialty paper. Black paper




using 2500 pounds of Direct Black 38 was produced over a three day




period. Urine samples were collected for up to five days from employees




over all three work shifts each day. In all, 47 urine samples were sub-




mitted. Seven production workers participated from this facility.




Environmental and urine monitoring data is summarized in Table VII.




Workers I, III, and IV are dyestuff weighers, while the other four




workers are operators of the pulp beaters. Like other dye using indus-




tries, these seven workers are probably the only ones on a regular basis




to be directly exposed to dyes.










In addition to the analysis of benzidine and its metabolites in the urine




of these workers, a recently discovered contaminant of Direct Black 38,




diaminoazobenzene, was detected. Diaminoazobenzene (DAAB), also known




as Basic Orange 2 or Chrysoidine, is an animal carcinogen. The dye used




in the paper dyeing facility was not analyzed for DAAB. However, DAAB




was found in the urine of four of the seven workers. Benzidine and/or




monoacetylbenzidine was also found in four of seven of the workers,




though concentrations were generally near the lower limits of detection.




In addition, 57% of the urine samples submitted contained 1 ppb or




greater non-specific aromatic amines. This is a higher percentage than




was found in the NIOSH comparison group.
                                      785

-------
It should be noted that the above workers who weighed dyes did so while


wearing a half face NIOSH approved respirator. Exhaust ventilation near


the dye weighing scales was also utilized to  lower exposures. Airborne

                                          3
concentrations were generally below 5 mg/m , while actual exposure would


have been less when using a respirator.
CONCLUSIONS




In total, urine samples were collected over varying lengths of time from


38 industrial employees who were regarded as potentially exposed to


benzidine derived dyes. Of that number^ benzidine or monoacetylbenzidine


in quantities ranging from one part per billion to 112 ppb benzidine


and 590 ppb MAB were found in 12 of the 38 workers monitored. Environ-


mental exposures and work practices were recorded in an attempt to asso-


ciate these factors with biological excretion.




Evidently, benzidine derived dyes can be used in the work place without


detecting benzidine or its metabolites in the urine. By comparing the


results from the six facilities surveyed it appears that total airborne


particulate concentrations above 3-5 mg/m  frequently resulted in de-


tecting benzidine, its metabolites, or elevated, non-specific aromatic


amines in the workers urine. Worker exposure to airborne particulates


less than 3 mg/m  are less often associated with finding aromatic amines


in the urine. However, full shift airborne exposures as low as


1.1 mg/m  resulted in considerable benzidine in some workers urine, thus


suggesting the difficulty in providing sufficient controls.


                               780

-------
Piotrowski (13) has provided evidence that urinary benzidine concentra-




tions of 100 ppb or greater are associated with an elevated risk of




bladder cancer in man. In addition, the National Cancer Institutes recent




testing of three benzidine derived dyes Direct Black 38, Direct Brown 95




and Direct Blue 6 indicate that the benzidine derived dyes may be more




carcinogenic in the rat than benzidine alone (14). In view of this and




other recently published information on these dyes, it is difficult to




see how they can be used in a sufficiently controlled and safe manner.




Therefore, NIOSH has recommended that benzidine derived dyes be treated




as carcinogens and that steps be taken to substitute or minimize employee




exposure as much as possible.
                              787

-------
                            REFERENCES

 1.  Venkataraman,  K. ,  The  Chemistry  of  Synthetic  Dyes,  Academic  Press,
    New York,  1952,  pp.  2-10.

 2.  Colour  Index,  Ed.  3  Rev.,  Research  Triangle Park,  N.C.,  The
    American  Association of  Textile  Chemists  and  Colorists,  1971,  1975,
    Vol.  1-6.

 3.  Synthetic Organic  Chemicals:  Report  on  Synthetic Organic Dyes,
    Series  6-2,  Washington,  D.C.,  U.S.  Tariff Commission,  1949.

 4.  Synthetic Organic  Chemical—United  States Production  and Sales,
    1977, U.S.  ITC Publication 920,  U.S.  International  Trade Commission,
    1978, pp.  87-132.

 5.  Imports of  Benzenoid Chemicals and  Products,  1978,  U.S.  ITC
    Publication  990, U.S.  International  Trade Commission,  1979,
    pp. 40-72.

 6.  Environmental  Protection Agency,  40  CFR Part  129,  Benzidine:
    Proposed  Toxic Pollutant Effluent Standards,  Vol.  41,  No.  127, June
    30, 1973.

 7.  Boeniger,  M.,  The  Carcinogenicity and Metabolism of Azo  Dyes:
    Especially  Those Derived From Benzidine,  Cincinnati,  National
    Institute for  Occupational Safety and Health,  1980, 160  pp.
    (in press ) .

 8.  Special Occupational Hazard Review  for  Benzidine-Based Dyes.
    DHEW (NIOSH) Publication No.  80-109,  January,  1980.

 9.  Diazonium Salts  and  Azo  Dyes  in  Air;  NIOSH Manual  of  Analytical
    Methods,  2nd Ed. Vol.  1, DHEW (NIOSH) Publication  No.  77-157-A,  April
    1977.

10.  Benzidine in Urine (screening test),  P&CAM 315, NIOSH Manual  of
    Analytical  Methods,  2nd  Ed.,  Vol. 5., DHEW (NIOSH)  Publication No.
    79-141, 1979.

11.  Nony, C.R and  Bowman,  M.C., Carcinogens and Analogs:  Trace Analysis
    of Thirteen  Compounds  in Admixture  in Washwater and Human Urine,
    Int.  J. Environ. Anal.  Chem.,  Vol.  5, pp. 203,  1978.

12.  Nony, C.R.,  Bowman,  M.C. Trace Analysis of Potentially Carcinogenic
    Metabolites  of an  Azo  Dye  and Pigment in  Hamster and  Human Drive
    as Determined  by Two Chromatographic Procedures, J. Chromographic
    Science (accepted).
                               78ft

-------
13,  Piotrowski,  J.,  Benzidine  In: Exposure Tests for Organic Compounds
    in Industrial  Toxicolgy, DREW (NIOSH) Pub. No. 77-144, Cincinnati,
    Ohio,  1977,  pp.  81-85.

14.  13-Week Subchronic  Toxicity  Studies of Direct Blue 6, Direct Black
    38,  and Direct Brown  95 Dyes, NCI-CG-TR-108, DREW Publication No.
    (NIH)  78-1358, 1978,  127 pp.
                               789

-------
              Table  I.  NIOSH control urine results.
 Individual
 2
 3
 4
 5
 6++
 7
 8
 9 +
 10
 11
 12
 13
 14
 15++
 16
 17
 18
 19
 20
 21
 22
 23
 Standard 1 (Bzd)
 Standard 2 (Bzd)
 Standard 3 (Bzd)
Aromatic Amines*
   (ng/IOOml)

     205
     ND
     ND
     NO
     100
     ND
     NO
     120
     300
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     145
     100
     100
     200
     ND
     ND
     100
     250
     350
Approximate
    ppb

     2
   < 1
   < 1
   < 1
     1
   < 1
   < 1
     1.2
     3.0
   < 1
   < 1
   < 1
   < 1
   < 1
   < 1
   < 1
   < 1
     1.4
     1.0
     1.0
     2.0
   < 1
   < 1
     1.0
     2.5
     3.5
    TLC**
Confirmation

     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
     ND
  Positive
  Positive
  Positive
 * Expressed as benzidine.
**Thin-layer chrcmotography
 +Allergy medication taken
++Pipe  smokers

 Note:  Lower  lirr.it of detection is 100 ng/100 ml of urine.
                                790

-------
               Table  III.   Environmental  and biological  sampling data from Dye Manufacturer—Facility B.
Job Description
Dally Personal  Exposure
Urinary Excretion
                                       Notes
Pulverizer 1
Spray dry
Pulverizer 2
Tray Oven
Day  I) 12 mg/m -total
     2; 5.9 mg/m:; total
        2.5 mg/m  respivaole
     1) 17.4 mg/m  -  area
        near chute
     1) 6.2 mg/m -total
     2) 14.l mg/m  total
     1) 7.0 mg/m^ total
     2) 6.5 mg/m  total
52 ppb Bzd;
18 ppb Bzd
248 MAB
112 ppb Bzd;  590 ppb
MAB:  50 ppb  DIMeBzd
10 ppb aromatic amines expressed
as Bzd;  Bzd confirmed on TLC
5 ppb aromatic amine  as benzidine

11 ppb Bzd; 22 ppb MAB;
15 ppb, DiMeBzd
Wore cartridge respirator.
Occasional exposure to very high
levels during adjustments


Wore no respirator - most  of day  spent
outside bull ding.  The presence of
DiMeBzd would indicate previous exposure
as no DiMeBzd dyes were being used on
day of sampling.


Wore cotton gauze respirator; became
very dirty from dyestuffs  worked  mostly
with a non Benzidine dye.


Wore cartridge dust respirator; emptied
dried oven trays into drums
+ Abbreviations:
  Bzd - benzidine
  MAB - monoacetylbenzidine
  DiMeBzd - 3,3' - dimethylbenzioine
                  * Limit  of Detection (ppb)
                    Bzd  -  1.4
                    MAB  -  5.8
                    DiMeBzd - 3.0

-------
Table IV.  Environmental  and urinary excretion in Textile Facility C
Workers Personal Exposure
1) 1.39 mg/m3


2) 1.06 mg/m3




3) 1.68 mg/m3


4) 2.06 mg/m3


5) 1.07 ng/m3







3
6) 1.12 mg/m



7) 1.98m;j/m3






urinary Excretion*
"Aromatic
Amines
3.2 ppb
4.4 ppb

8 ppb



5.8 ppb
5 ppb


3 ppb


4.5 ppb








9 ppb


13 ppb
3.4 ppb
3.0 ppb
4.0 ppb
3.2 ppb



Benzidine
ND


benzidine
confirmed
by TLC

39 ppb
1 ppb


ND


ND








benzidine
confirmed
by TLC
16 ppb


ND
ND



Monoacetyl-
benzidine
ND






5 ppb
7 ppb


ND


ND











38 ppb


ND
ND



Notes
Dye weigher wore no
respirator. General
ventilation only.
Dye weigher wore no
respirator. General
ventilation only.
Boiled up dye by hand.

Dye weigher wore no
respirator. Beginning
of shift-dustiest
Dy_e weigher wore no
respirator. Beginning
of shift dustiest.
Pad 
-------
      Table V.  Results  of  urine and environmental monitoring  at Facility
                a textile dyer.
Personal worker Exposure*
1 ) 1 . 54 mg/m,
1.31 mg/m.
1.45 mg/mJTWA









2) 1.15 mg/m.
1.11 mg/nT
1.13 TWA
Area sample ever scales
0.55 mg/m
3) 5.31 mg/m3 (void)












4) 0.90 mg/m3
5) 1.58 mg/m3
3
6j 0.67 mg/m
7) 0.63 mg/m3
8) 0.20 mg/m3


9) 0.60 mg/m3
10) 0.48 mg/m3





Urinary Concentrations
Aromatic
Amines
4 ppb
4.8 ppb
3.6 ppb









ND
ND
1.3 ppb


ND
ND











ND
3.2 ppb
ND
3.2 ppb
ND
ND
ND
ND
ND





Benziaine
ND
ND










ND






























Monoacetyl-
benzidine
4 ppb
ND










ND






























Notes
D.Y£ Weigher
Ueighed dyes in Drug
Koom before dissolving
in boil-up tubs. Work-
er sometimes wore a
ha If- face pad type
respirator and gloves
when weighing dyes.
General ventilation
from roof exiaust fans
only.
Same as above




Dye Tub Operator
Worker loads ana un-
loads cloth from
rol 1 s to and from
dye tubs. Tubs were
ventilated by top
nood exnaust and had
front hood moveable
doors. No respirators
worn. RubDer gloves
worn sometimes.
Worker is splashed by
dye liquor during work.
Same as above
Same as above
Same as above
Same as above
Same as -above


Same as aoove
Roll -up Machine Operator
worker operates steam-
press roll -up machine.
Considerable heat and
resultant steam evolved.
No respirator worn.
•Environmental concentrations expressed as total airborne particulates per cubic meter
 of samples air.
                                         793

-------
Table VI.  Dye Exposure among  three  workers at a Leather Tannery--
           Facility E.
'ersonal Worker Exposure
1) 12.05 mg/m3
2) 12.95 mg/rn
3) 14.72 mg/m3
4) 1.27 mg/ni
10.65 mg/m3 TWA*
5) 1.42 mg/m3,
6) 0.44 mg/m
7) 0.00
0.69 mg/m3 TWA
8) 1.12 mg/m3
9; 1.65 mg/m
10} 16.79 mg/m3
->
5.79 mq/m" TWA
Urinary Excretion
Day 1
ND+
ND
Day 2
ND
ND
Day 1
ND
ND
Day 2
ND
ND
Day 1
ND
ND
Day 2
ND
ND
Notes
Dyestuff Weigher
Only benzidioe-derived
C.I. Directo Black 38
and C.I. Direct Brown
95 were used. Dye
weigher spent 80%
of time weighing tnese
dyes into paper bags.
Half-face cartridge
respriator was worn
during weighing.
Dye Drum Operator I
Operator picks up
bagged dyes from
weigher and empties bag
into solvating tub.
Potential exposure
would only occur during
emptying. Both operators
wore half-face car-
tridge respirators during
dye handling.
Other responsibilities
include loading and un-
loading dyed hides from
dye bin.
Same as above

  Time Weighted Average
 +Benzidine not detectable
                                    794

-------
                     Table VII.   Results  of environmental and urine monitoring at Paper Facility F


Worker
1
Jorker
II
Worker
III
Worker
IV
Worker
V
Worker
VI
Worker
VII

Env. Conc.^
lmg/m3)

.
-
-



3ay 1
Urine
iPPb)
N.D+
(N.D)++
N.D
N.D
11. 0)





Env.
img/







Day
Cone.
m3)

-
-
-

"

2
Urine
IPPb)
N.D
IN.D)
N.D
IN.D)
N.D
(2.0)
N.D
(N.D)
N.D
(1.0)
N.D
(1.3)

Da
Env. Cone.
(mq/mj)
3.3
2.3
1.6
3.7
2.9
2.6

y 3
Urine
(ppb)
N.D
(1.4)
N.D
N.D.
2.2
N.D
(1.3)
N.D
(N.D)
N.D
11.3)

Ua>
Env. Cone.
(mq/m )
3.4
5.1
2.5
Void
2.3
Void

4
Urine
IPDb)
3 MAB
ID
N.D
N.D
N.D
N.D
2 MAE
3MAB
3DAAB
Da>
Env. Cone.
(mq/m3)
Void

-

-
-

M
Urine
(ppb)
3 MAB
l Bzd
32 DAAB
(<|9)
1 Bzd
5 DAAB
(2.9)
N.D
(2.6)
(N.D)
IN.D)
2 DAAB
(N.D)
2 MAB
2MAB
Day
Env. Cone.
lmq/mj)







6
"Urine
(ppb)
8 MAB
2 DAAB
N.D



2 MAB

CD
    FOOTNOTES:
        * environmental concentrations expressed as milligrams total
          airborne participates per cubic meter air.
        + concentration of specific aromatic amines in ppb with tne following
          detection limits:  benzidine l.Oppb; MAB, 1.8ppb; DAAB, O.Sppb; diacetylbenzidine, 0.2ppb

       ++ concentration of non-specific colorimetric procedure with the limit of detection at Ippb.

    NOTE:  Area within heavy black line signifies period of Direct Black 38 useage.

-------
                             Discussion
Dr.  Weisburger (NCI):   We have time for one or two questions.   Yes?

Dr.  Landrigan (CDC):   Did you analyze bulk samples of the dye or samples
of the airborne particulates that you captured on your filters to see
whether or not there was benzidine in that material, or are you con-
vinced that the benzidine which was detectable in the urine arose
entirely by metabolism within the bodies of workers?

Mr.  Boeniger (NIOSH):   This was an important step in the research
protocol.   In every case where benzidine-derived dyes were used in the
work place, the dyes were collected and analyzed for residual  free
benzidine content.

We have summarized considerable data on domestically produced benzidine
dyes and imported derived dyes for their content of residual  benzidine.

At the levels of residual free benzidine in these dyes, we tried to
calculate from historical information what the likely excretion from
simply the residual quantity of benzidine was, not accounting for
possible metabolism of the dyes themselves, and in all  cases,  even with
leniency,  the levels of excretion would have been below the minimal
level of detection of our procedures.

So while we have not been able to prove the metabolism of the Direct
Black 38 or any of the other benzidine-derived dyes, it is highly
suggestive that this work is in agreement with the animal experiments
which have demonstrated metabolism of the benzidine dyes.

Dr.  Lanrigan (CDC):  Have OSHA or the Department of Commerce picked up
on the NIOSH recommendation to limit or ban the use of this material,
and are they working on regulations either to restrict industrial use or
to forbid the importation?

Mr.  Boeniger (NIOSH):   At this time, my understanding is that they have
not taken the position of eliminating them or banning the benzidine-
derived dyes, but simply to minimize the exposure to lowest feasible
levels.

Dr.  Saffotti (NCI):  I wanted to ask you something about the presence of
several aromatic amines concurrently.  You have indicated, and I am
sorry I missed the paper given by Dr. Lowry on the previous session,
that you find benzidine and 4-aminobiphenyl and some other compounds as
a breakdown product of the dyes.

We have been doing some work in in vitro systems on combined effect of
various carcinogens, and one of the series we have recently studied is
a series of aromatic amines, for which we have gotten a number of data
in the Ames test, which again I would like to eventually compare with
those that were reported by Dr. Lowry.
                                   79H

-------
In studying the combinations of various aromatic amines together, we
have found a number of them inhibiting each other in the Ames test and
on occasion some synergism.  One of the combinations that seems to be
repeatedly postive as synergistic is benzidine and 4-aminobiphenyl.

In comparing the mutagenic activity of the Direct Black 38 with that of
benzidine or 4-aminobiphenyl alone, one certainly finds a much higher
level of activity of Direct Black 38 than the individual compounds,
although I would like to have eventually some discussion about the
solvent systems used and other things like that.

In relation to the monitoring in the human, do you have evidence of  4-
aminobiphenyl or other compounds being detectable in the urine and
possibly their concurrent presence with benzidine?

Mr. Boeniger (NIOSH):  Initially, our methodologies were rather simple,
but as we progressed into the last survey, and only in the last survey
were we actually looking for other metabolites or other aromatic amines
other than benezidine or its metabolites -- for instance, 4-amino-
biphenyl -- and in the last facility we did not find 4-aminobiphenyl in
any of the workers' urines there.  So that was the only indication where
they were actually looked for.

We did find it using Direct Black 38 in animal experiments and just  do
not know how to correlate that right now.

Dr. Jenkins (EPA):  What other facilities have you now monitored over
several days trying to establish some of the kinetics of the excretion
patterns of benzidine besides that one paper facility?

Mr. Boeniger (NIOSH):  Well, that was the last one in our study.  We had
at that time run out of both time and funds anticipated for the com-
pletion of this study.

There were some surveys previous to that where we looked at workers
during the day of exposure and the following day of exposure, so there
was a 2-day follow-up period there.

Dr. Jenkins (EPA):  But no other long exposures like that paper?

Mr. Boeniger (NIOSH):  No, I am afraid not.
                                      797

-------
                                             DR.- STEPHEN BERARDNELLI

           1H CHAKALIbKlZAIlUP!  Uh  LUAL  Li'-'UMU 1UM hAClLllHiS.

     COAL LIQUEFACTION REPRESENTS A TECHNOLOGY THAT MAY DECREASE OUR
DEPENDENCE ON IMPORTED PETROLEUM,  RESEARCH IN THIS AREA HAS BEEN
CONDUCTED WITHIN THE UNITED STATES SINCE 1934,  HOWEVER, IN SPITE
OF THE TIME AND EFFORT EXPANDED ON THESE PROCESSES MANY QUESTIONS
STILL REMAIN TO BE ANSWERED BEFORE LIQUEFACTION BECOMES AN ESTAB-
LISHED TECHNOLOGY,  ONE OF THE QUESTIONABLE AREAS NOW UNDER INVESTI-
GATION IS THAT OF OCCUPATIONAL HEALTH,   WORK IN THIS AREA VMS INITIATED
SEVERAL YEARS AGO BY THE NATIONAL INSTITUTE FOR OCCUPATIONAL.SAFETY
AND HEALTH (NIOSH) THROUGH A SERIES OF STUDIES DESIGNED TO IDENTIFY
POTENTIAL HEALTH HAZARDS IN THE COAL LIQUEFACTION ENVIRONMENT,
     THIS PAPER IS A REVIEW OF THE RESULTS OF ONF OF THFSF N
STUDIES,  THIS STUDY ENTITLED "A STUDY OF COAL LIQUEFACTION PROCESSES"
IS DESIGNMD TO CHARACTERIZE THE OCCUPATIONAL ENVIRONMENT OF DIFFERENT
LIQUEFACTION PROCESSES IN TWO STAGES;
     1,    Th'E IDENTIFICATION OF CONTAMINANT CLASSES, AND
     2,    THE DETERMINATION OF EXPOSURE LEVELS FOR SELECTED
           CONTAMINANTS.

     THE RESULTS REPORTED HEKF. ARE BARED ON COMPLETED SURVEYS IN THREE
OF THE FIVE FACILITIES UNDER STUDY,   ALTHOUGH THE PHYSICAL AGENTS AND
PARTICULATFS WLRE INVESTIGATED,  THE FINDINGS FOR THE ORGANIC CONTAMI-
NANTS ARK LMPHASiI!:D IN' THIS PAPER,
                                       798

-------
 PROCESS

     THE FACILITIES INVOLVED IN THIS STUDY UTILIZED THE DIRECT LIQUE-
 FACTION PROCESS IN WHICH COAL IS CONVERTED DIRECTLY INTO A LIQUID
 PRODUCT AS OPPOSED TO THE INDIRECT PROCESS WHERE COAL IS FIRST
 GASIFIED AND THEN CATALYTICALLY CONVERTED TO A LIQUID FUEL,  BRIEFLY,
 THE DIRECT PROCESS INVOLVES THE FORMATION OF A SLURRY OF FINE-MESHED
 COAL I.M  A COAL-DERIVED LIQUID,   THIS SLURRY IS THEN SUBJECTED TO
 ELEVATED TEMPERATURES AND PRESSURES TO .DISSOLVE, DEPOLYMERIZE, AND
 HYDROGENATE THE CARBONACEOUS MATERIAL IN COAL TO FORM THE LIQUID
 PRODUCT,  GASES AND SOLIDS ARE BY-PRODUCTS OF THE PROCESS,

     THE STUDY WAS DIVIDED INTO TWO STAGES,   THE FIRST STAGE IS
 QUALITATIVE ANALYSIS; THE SECOND IS QUALITATIVE ANALYSIS OF THOSE
 COMPOUNDS IDENTIFIED IN STAGE 1.
             GIVES THE OPERATING SIZE AND OPERATING PARAMETERS FOR
THE DISSOLVER/REACTOR STAGE OF THE PROCESS FOR THE IHREE FACILITIES
DISCUSSED IN THIS PAPER,   FACILITY II DIFFERS FROM THE OTHER FACILITIES
BY HAVING A CATALYTIC HYDROGEHATION STEP,   As CAN BE SEEN FROM THE
COAL PROCESSING RATE, ALL THE PLANTS STUDIED WERE PILOT PLANT FACILITIES,

PROTOCOL AND -RESULTS/STAGE 1

     THE CONTAMINANT CLASSES PRESENT WITHIN THE COAL LIQUEFACTION
WORKING ENVJRONMr/NT OF THE TWO FACILITIES  WFRF IDENTIFIED THROUGH
AN ANALYSIS OF  8-HOUR AREA AIR SAMPLES,   SAMPLES WERE COLLECTED ON
                                      799

-------
                            SLIDE 1
           OPERATING PARAMETERS AT THREE FACILITIES


FACILITY              SIZE               PRESSURE         TEMPERATURE
                 (TONS COAL/DAY)
   i                  50 (150 BBL/D)    1500-2000           800-375

  II                  20                150-400            500-750

 III                   6 (18 BBL/D)     1400-2500           800-875
                                  800

-------
             CRITICAL ORIFICE
                a ADAPTER
Oo
o
                          CELLULOSE SUPPORT PAD
                                                                   SSmrr,
                                      CHROMOSOR3 102-
     -SiLVER MEMBRANE


-STAINLESS STEEL SCREEN
                               Slide  2.   HIGH-VOLUME SAMPLING DEVICE FOR PNA

-------
600MG CHARCOAL TUBES AND 875MG SILICA GEL TUBES FOR THE IDENTIFICATION



OF HYDROCARBONS AND AROMATIC COMPOUNDS,  E,G,  AMINES, PHENOLS,  THESE



SAMPLES WERE COLLECTED AT A FLOW RATE OF 100 ML/MIN,







     A SPECIAL SAMPLING CASSETTE CONSISTING OF A SILVER MEMBRANE



FILTER AND CHROMOSORB 102 WAS DESIGNED FOR COLLECTING POLYNUCLEAR



AROMATICS (PNAs),   THIS CASSETTE SHOWN IN SLIDE 2 INCLUDES:




            •  A SILVER MEMBRANE FILTER



             *  STAINLESS-STEEL SCREEN



             o  CHROMOSORB 102 - 3 GRAMS



             ,  SUPPORT PAD
     A GLASS FILTER WAS NOT USED IN THIS SAMPLING DEVICE BECAUSE



PARTICIPATE LEVELS AT THE SURVEYED FACILITIES WERE NOT EXPECTED TO



ClOG THE SILVER MEMBRANE FILTi-R.   IN OPERATION,  THIS ASSUMPTION



PROVED CORRECT,  THE CHROMOSORB ADSORBENT WAS ADDED TO THE SYSTEM



TO CAPTURE VAPOR PHASE PNAs AND SPECIES EVAPORATING FROM THE FILTER



COLLECTED PARTICUL.ATES,  AREA SAMPLES WERE COLLECTED AT A FLOW RATE



OF 9,2 LPM,








     A GENERAL SCAN OF CHARCOAL AND SILICA GEL TUBE AREA SAMPLES



FROM VARIOUS PROCESSING UNITS WAS CONDUCTED FOR MORE THAN 25,000



COMPOUNDS USING A GAS CHROMATOGRAPH/MASS SPECTROMETER (GC/fv]S) WITH



QUALITATIVE IDENTIFICATION BY A COMPUTER DATA J3ASE.  CHARCOAL AND



SILICA GEL SAMPLES WERE ALSO ANALYZED BY GAS CHROMATOGRAPHY FOR



BENZENE, TOLUENE, AND XYLENE, AND THE AROMATIC AMINES AND AROMATIC
                                    802

-------
ALCOHOLS LISTED IN SLIDE 3.  THE COMPOUNDS IN-SlJLnL-2 ARE REPRESEN-
TATIVE OF TWO CLASSES OF CHEMICAL WHICH HAVE BEEN ASSOCIATED WITH
THE LIQUEFACTION PROCESS,   THE SAMPLE CASSETTES WERE QUANTITATIVELY
ANALYZED FOR THE CYCLOHEXANE-SOLUBLE FRACTION WHICH IS IN LIEU OF
THE BENZENE-SOLUBLE FRACTION,   ALL ANALYSES WERE PERFORMED ACCORDING
TO THE APPROPRIATE MIOSH ANALYTICAL METHOD, WHERE AVAILABLE.

     THE PNA SAMPLE CASSETTES  WERE ALSO QUANTITATIVELY ANALYZED FOR
THE 30 PNAs LISTED IN Su_DE_Jl FOR WHICH ANALYTICAL STANDARDS WERE
AVAILABLE,   ANALYSES FOLLOWED  THE GC/MS AND HIGH PERFORMANCE LIQUID
CHROMATOGRAPHY METHODOLOGY DEVELOPED BY THE UNIVERSITY OF IOWA
HYGIENIC LABORATORY, IOWA CITY/ IOWA.   IN THIS METHOD THE SILVER
MEMBRANE FILTER IS ULTRASONICLY EXTRACTED WITH CYCLOHEXANEj  THE
CHROMOSORB  102 is EXTRACTED IN A SOXHLET EXTRACTOR USING A 50:50
MIXTURE OF  METHYLENE CH1.QR1DE-METHANOL.   SENSITIVITY OF THE METHOD
FOR EACH OF 30 PNAs IS IN THE  LOW NANAOGRAKi RANGE PER SAMPLE,

STAGE 1 RESULTS AND DISCUSSION

     A TOTAL OF 85 COMPOUNDS WERE IDENTIFIED IN THE 23 SAMP! ES TAKEN
AT FACILITY I•  AND 68 WERE IDENTIFIED IN 21 SAMPLES TAKEN AT FACILITY
II (S_LlPf,Ji).   FACILITY III WAS NOT INVOLVED IN THE STAGE 1 SURVEY,
FROM THE QUALITATIVE ANALYSIS,  MORE DIFFERENT KINDS OF AROMATIC THAN
ALIPHATIC WERE  FOUND AMONG THE COMPOUNDS WHICH COULD BE DETECTED,
                                 803

-------
                              SLIDE 5
                     LIST OF SELECTED COMPOUNDS
AROi'iATIC AMINES
ANILINE
ft [f-BlKiETHYLANILINE
2,  MlMETHYLANILINE
P-illTROANILINE
0-ToLUIDINE
Q-ANISIDINE
P-ANJSIDINE
AROMATIC ALCOHOLS
PHENOL
P-PHENOL
O-ETHYLPHENOL
P-ETHYLPHEMOL
O-CRESOL
M-CRESOL
P-CRESOL
2,. 3-XYLENOL
2/ 5-XYLENOL
                                                     5-XYi.ENOL
                                      804

-------
                              SLIDE
                    POLYNUCLEAR AROMATICS  (PNAs)
NAPHTHALENE
QUINOLINE
2-liETHYLNAPHTHALENE
1-rlETH YLN A PHTHA LE N E
ACENAPHTKELENE
ACENAPHTHENE
FLUORENE
PHENANTHRENE
ANTHRACENE
ACRIDINE
CAREOZOLE
rLUORAri liiiT-KE
PYRE N't
BENZO(A)FLI!°;'.
-------
                             SLIDE 5



      BREAKDOWN OF COMPOUNDS IDENTIFIED AT FACILITIES  I AND  II







FACILITY               ALIPHATIC           AROMATIC            TOTAL



   I                       5                  81                 36



  II                       1                  67                 68
                                806

-------
     THE  PREDOMINANCE OF THE AROMATICS  AMONG THE IDENTIFIED COMPOUNDS
IS BELIEVED  TO  BE DUE TO THE GREATER STABILITY OF THE AROMATICS
RELATIVE  TO  THE ALIPHATICS  AT THE  OPERATING  TEMPERATURES AND PRESSURES
OF THE PROCESS,   0)   JHIS PATTERN  OF RELATIVE STABILITY IS ALSO
APPARENT  AMONG  THE AROMATICS AS  CAN  BE  SEEN  IN Su_DE_6. WHICH GIVES
A BREAKDOWN  OF  THE AROMATICS BY  THE  NUMBER OF AROMATIC RINGS WITHIN
THE COMPOUND,   IT IS  APPARENT THAT THE  MORE  STABLE CLASS OF AROMATICS.,
THE SINGLE-RING AND  FUSED TWO-RING COMPOUNDS, ACCOUNT FOR A MAJORITY
OF THE COMPOUNDS IDENTIFIED AT THE TWO  FACILITIES WITH AN AVERAGE
DISTRIBUTION OF 52.3% AND 66,8%,  RESPECTIVELY,  FOR FACILITIES I AND II,

     THESE FINDINGS  SUGGEST THAT COMPOUND STABILITY is A FACTOR IN
DETERMINING  THE TYPES OF COMPOUNDS BEING  FORMED UNDER THE LIQUEFACTION
PROCESS,  OTHER FACTORS  WOULD BE THE TYPE OF PROCESS, OPERATING PARA-
METERS, CATALYST.,  ETC,

     THE  BENZENE-SOLUBLE FRACTION  HAS BEEN USED AS AN INDIRECT MEASURE
OF PNA LEVELS  IN THE  WORKING ENVIRONMENT,  -'  FOR EACH SAMPLE CASSETTE
THE CYCLOHEXANE SOLUBLE  FRACTION WAS DETERMINED SEPARATELY FOR THE
FILTER AND ADSORBENT  TO  EVALUATE THE APPLICABILITY OF THIS METHOD TO
THE COAL  LIQUEFACTION ENVIRONMENT,   THESE RESULTS ARE GIVEN IN SLIDE 7
ALONG WITH MEASURED  PNA  LEVELS FOR THESE  SAMPLES,   THE PNA VALUES
REPRESENT THE SUM OF  THE CONCENTRATION  OF THE 30 PNAs UNDER STUDY,
AS CAN BE SEEN  SOMF.  CONSISTENCY  IS SUGGESTED AMONG THE SOLUBLE FRACTION
VALUES BUT A HIGH VARIABILITY WAS  NOTED WITHIN THE PNA VALUES,  Mo
APPARENT  RELATIONSHIP WAS NOTED  BETWEEN THE  CYCLOMEXANE-SOLUBLE FRACTION
                               80?

-------
STRUCTURE


NONFUSED RINGS


   ONE RING
   Two RINGS
       .E RINGS
              SUE 6

BREAKDOWN OF AROMATIC COT-TOU-IDS AT

         FACILITY I ATID II

   (No,  OF  COMPOUNDS)

FACILITY I       FACILITY II
  37,2
  (32)

  31.4
     ,6
     )
                        (i)
                   2,9
                  (20)

                  26,1
                     ?
                      0
                                 REPRESENTATIVE COMPOUNDS
INDENE
 OLLIIDINE
  ISIDINE
 HIOPHENE
SUBSTITUTED BENZENES
TOLUENE
XYLENE
ANILINE
SUBSTITUTED ANILINE

BlPHENYL
SlBSTITUTEfl BlPHENYLS
                                   TRJ^JCNYL ESTER
y
                        6.9
   Two RINGS
     I^EI: RINGS
   FOLK RINGS
   Fivr RINGS
20,9
(13)
W
.12,8
(Jl)
                        00
                    69,6
                    40,7
                    (2&)
                    13,0
                    (9)
                     7,2
                     G)
                     87
                      •y


                       808
NAPHTHAU:NE
SUBSTITUTED NAI^HTI IALENES
QuINCLINE
ACENAPHil-IALENE
ACENAPIITMENE
FLUOREKI-
AZULENE

PHENANTHRENE
SUBSTITUTED Pl-IENANTI IRENES
ANTH^ACriC

CARMZOLE

BENZANlll.RACENfZ
IRIPMENYI.ENE
Cl^YSEN!.:

Bs-NXI'Yfx'i-NF
PLRYLf.NE

-------
                                SUEZ

              CYOJOHEXRiE-SOUBLE FRACTION vs,  TOTAL PNfts
SAMPLE         CYCLOHEXANE-SOLIBLE FRACTION (MG/M^)           TOTAL PNAs U/M^)
              FILTER      ADSORBENT        TOTAL       FILTER  ADSORBENT  TOTAL
001
002
0)3
(0}
015
021
l-fiK-RFi I
0,2
0,4
0,3
0,1
0,3
0,07
iTTfii'j i.OFFF. i ~\
1,6
0,6
0,9.
0,7
1,2
0,8
r, n-r-r, fl OP.
M i r K \j i L/IJ
1,8
1,0
1,2
0,8
1.5
0,9

0,9
0,4
19,4
13,6
0,07
67V/I

8,3
45,0
1727
260,1
30,0
127,
N = R
9,2
45,4
1746,
273,7
30,1
m

                 ABSORBENT H- 0,01

                      TOTAL - 0,12
                                    808

-------
VALUES AND THE MEASURED PNA LEVELS INDICATING THAT THE SOLUBLE
FRACTION MAY NOT BE AN EFFECTIVE TOOL FOR MEASURING WORKER EXPOSURE
TO PNAS,

     ON THE BASIS OF THESE QUALITATIVE RESULTS, STAGE 2, QUANITATIVE
SAMPLING AT THE THREE FACILITIES WAS LIMITED TO BENZENE, TOLUENE,
XYLENE, AROMATIC AMINES,  AND PNASj COMPOUNDS AND COMPOUND CLASSES
IDENTIFIED IN STAGE 1.  A PANEL OF NIOSH RESEARCHERS HAD PREVIOUSLY
PRIORITIZED THE LIST OF CHEMICALS,  THE ONES LISTED ABOVE WERE IN
THE HIGHEST PRIORITY GROUPING,

STAGE 2  SAMPLING PROTOCOL

     THREE MAJOR JOB CATEGORIES WERE IDENTIFIED IN WHICH WORKERS MAY
BE EXPOSED TO PROCESS CONTAMINANTS BECAUSE OF ASSIGNED DUTIES,  THESE
CATEGORIES INCLUDE THE PLANT .OPERATORS,  LABORATORY TECHNICIANS, AND
THF MAINTENANCE STAFF,  THE SAMPLING PROGRAM EMPHASIZED SAMPLING TO
DETERMINE EXPOSURE OF THE PLANT OPERATORS BECAUSE THEIR EXPOSURE IS
BEL!EY'M;> TO BF MORE REPRESENTATIVE OF THE PROCESS THAN THE OTHER JO)'.
CATEGORIES,   Tl-iL RESULTS  REPORTED HERE FOR THE THREE FACILITIES ARE
THEREFORE PRIMARILY CONCFRNFID WITH THE FIFLD OPERATORS,

     IN STAGE ?. OPERATOR  EXPSOURF WAS DFTFRMINLD FOR EACH OF THE
CONTAMINANTS IDENTIFIED IN STAGE 1,   THESE CONTAMINANTS ARE LISTED
IN SLIDE_O.   SAMPLES TAKEN WERU OF 8--HOUR DURATION WITH FLOW RATES
OF 100 f'iL/M)K! FOR CHARCOAL AND SILICA GEL. TUBE SAMPLES AND 1,5
                                  810

-------
                             SLIDE 8
                CONTAMINANTS  STUDIED  IN  STAGE II
BENZENE
TOLUENE
XYLENE
ANILINE
N,N-DlMETHYLANILINE
2,'J-DlMETHYLANILINE
P-I'llTROANILINE
0-TOLUIDINE
O-ANISIDINE
P-ANISIDINE
ALPHA-HAPHTHYLAMINE
NAPHTHALENE
QUIMOLINE
2-^ETHYLNAPHTHALENE
I-METHYLNAPHTHALENE
ACENAPHTHALENE
ACENAPHTHENE
FLUORENE
PHENANTHRENE
ANTHRACENE
ACRIDINE
CARBAZOLE
FLUORANTHENE
PYRENE
BENZO(A)FLUORENE
BENZO(B)FLUORENE
BENZ(A)ANTHRALENE
CHRYSENE/TRIPHENYLENE
DIMETHYLBENZ(A)ANTHRACENE
BENZO(E)PYRENE
BENZO(A)PYRENE
PERYLENE
DIBEIMZ(A,J)ACRIDINE
DibENz(Ayi)CARBAZOLE
INDENO(1,2,3-CD)PYRENE
DlBENZANTHRACENE
BENZO(G,H,I)PERYLENE
ANTHANTHRENE
CoRONENE
DlBENZPYRENE
                                   811

-------
LPM FOR THE PERSONAL PNA SAMPLE CASSETTE,   PERSONAL MONITORING
SAMPLES WERE SUPPLEMENTED AT EACH  FACILITY  WITH 8-HOUR AREA SAMPLES,
THE AREA SAMPLES ARE UNIT EXPOSURES,

     A DIFFERENT SAMPLING DEVICE HAS BEEN DESIGNED TO MONITOR WORKERS
FOR PERSONAL PNA EXPOSURE,  THIS SAMPLING DEVICE IS SHOWN IN SLIPJLJL
IT CONSISTS OF A 3/MM CASSETTE WITH A  SILVER  MEMBRANE FILTER SAND-
WICHED BETWEEN TWO GASKETS AND A /MM INNER  DIAMETER GLASS TUBING
(BORON SILICATE) CONTAINING 150 MG OF  CHROMOSORB 102,   THE TWO
SECTIONS WERE JOINED WITH INERT ^-INCH TYGON  TUBING,   THIS CHANGE
WAS NECESSARY DUE TO PUMP WEIGHT RESTRICTIONS,

     SAMPLES TAKEN IN STAGE 2 WERE ANALYZED USING THE PROCEDURES
DESCRIBED FOR STAGE Ij THAT IS GAS CIIROMATOGRAPH FOR  CHARCOAL AND
SIITCA CM  TUB!:. .WU-'LL'S AND GC/HS AN.U  HIGH  PERFORMANCE LIQUID
CIIROMATOOUAPHY USING A FLUORESCENT DETECTOR FOR THE PNA SAMPLE
CASSETTES,

STAOI; 2  RLSUI.TS AND DISCUSSION

     TllE RESULTS FOR BLNZEh'E, TOl.UIZNL, XYLENIE AND TUT. HI GUT AROMATTCS
STUDIED ARE SUMMARIZED IN SjJJlFjlQ,  Till7 RANGES GIVEN REPRESENT A
COMPOUTH OF THE RESULTS FOR TIM! THREE MAJOR  JOB CATLiiOR J F.S ,   As CAN
BL SEEN MLACUREABLE QUANTITITKS WI£R!f OBTAINED PRIMARILY AT FACILITY  1
WITH TUT. HIGHER END OF Till: KANC-K ]U-lNi", ASSOCiATf-J) WITH 'I HE PLRPORMANCI
OF MAINTENANCE ACTIVITIES,  /\T F/ACILJTY  II  ONLY TOLUCNL! V/AS FOUND AT
                                    812

-------
oo
M
CO
                                                   70 mm
                 — 30 mm	

                 GLASS WOOL
                   CHROMOSORB 102
                                              TYGON TUBING
-1/4" O.D. GLASS TUBING
                                                                                CASSETTE
                                                       CELLULOSE GASKET

                                                                                                        FLOW
                                                   -CELLULOSE
                                                    GASKET
                                                                            SILVER MEMBRANE
                                      Slide 9..   PERSONAL MONITORING  DEVICE FOR PNA

-------
                                              SLIDE 10
oo
                RESULTS  IN  PPM FOR BENZENE, TOLUENE, XYLEtiE, AND EIGHT AROMATIC AMINES
                                  AT THREE LIQUEFACTION FACILITIES
COMPOUND

BENZENE
TOLUENE
XYLENE
ANILINE
N, N-DlMETHYLAN!LINE
2^-DlMETHYLANILINE
P-iv'ilTROANALINE
O-TOLUIDINE
O-ANISIDINE
P-ANISIDINE
I-NAPHTHYLAMINE
                                  FACiLITY  I
                                                          -ACILITY II
*-1 Y " T
f  i-i
No, OF
SAMPLES
15
15
16
21
21
21
2
21
21
21


0,
0,
0:


0,

0,



RANGE
01 - O.iS
01 - O.fc
01 - 0,07
0,02
0,01
01 - 0,0^
-
01 - 0,02
0,02
0,02
—
NO: 0?
SAMPLES
3
3
3
27
27
27
27
27
27
27

RA
0.02
0,02
0,
ND
ND
ND
m
MB



NGE
- 0,0-
- 0,03
G'J.
- 0,05
- 0,03
- 0,05
- 0,05
- 0,05
ND
NB
—
S^'n n-
1 \ I) i Or
SAMPLES
in
_LO
10-
10
12
12
12
12
12
12
12
12
RANC
ND
ND
.\ ' *n
['; J
•vr>
(\LJ
ND
HD
i'\'~>
ND
ND
ND
ND

-------
MEASUREABLE LEVELS  WHILE NONE OF THESE CONTAMINANTS WERE DETECTABLE
AT FACILITY III,  THE DETECTION LIMIT FOR  BENZENE,  TOLUENE AND  XYLENE
WAS 0,91 PPM AND  FOR  THE AROMATIC AMINES RANGED FROM 0,02 TO 0,05  PPM
DEPENDING ON THE  COMPOUND AND ON SAMPLE VOLUME,   THE HIGHEST VALUE
RECORDED WAS FOR  TOLUENE AT 0,4 PPM AT FACILITY I INDICATING THAT
WORKER EXPOSURE TO  THESE CONTAMINANTS WERE WELL BELOW CURRENT
OCCUPATIONAL HEALTH STANDARDS.
     WORKER EXPOSURE  TO  THE PNAs is SUMMARIZED  IN  8uDF,_ll.  VALUES
GIVEN ARE FOR TOTAL PNAs AND REPRESENT THE SUM  OF  THE MEASURED
CONCENTRATION OF THE  30  PNAs STUDIED,  As CAN BE SEEN WORKER
EXPOSURE TO THE PNAs  WAS           IN THE MICkOGRAM"PER-Clil?IC~METER
RANGE, A COMPARISON OF THE' THRPE FACILITIES SHOWS  THAT FACILITY 1
HAD HIGHER EXPOSURE LEVELS THAN THE O'il-IP.R FAC1LJ 7 i L't t-  IT IS ANTIC]
PATED THAT USING APPROPRIATE INDUSTRIAL. HYGIENE •''l:.t:HMltiU!- s i\w
CONTROL TECHNOLOGY, THESE VALUES WILL LP. QUITE  !.OV!,
     AN EVALUATION OF  THE  REf.ULTS JNDICAiEb THAT TilL-  TWO- AKD TI-iRi.:!?-
RING.PNAS ACCOUNTED  FOR  MORE THAN 97 PERCENT BY WEIGHT Or TH;: TOTAL
PNAs IDENTIFIED AT THE THREE FACILITIES A" SHOW!! iii  o{ .T.]\;-.. 1 ?,   Ti!!-S!-:
INCLUDED NAPHTHALENE^  QUINOLINE.,  2~METIiYLI\lAP!-n;!/\L'.:;NLJ  l~Fif-i'liVLNA:-:!--
THALENE, ACENAPHTIIENi;, FLUORENE.-  PHENANTiiUEK'E Ai\')J  AMTiir:/,'-1:^1!-:,   Ti!E
REMAINING THREE PERCENT  WAS  COMPOSED PRlMARI'i.Y 0!"  I:OUH-Ri ;;G  f'MA.r.t
                                 815

-------
                            SLIDE  11
         FIELD OPERATOR  EXPOSURE TO TOTAL  PNAs  IN
                                FACILITY  I    FACILITY  II    FACILITY III
NUMBER OF SAMPLES                   12            12              4
AVERAGE                           62,8         0,2             9,9
RANGE                            3,5-254,3    0,02-0,3       0,05-21,.'I
STANDARD DEV,                      72,2         0,1             8,8
GEOMETRIC MEAN,                    35,6         0,1             3,1
GEOMETRIC STD,  DEV,                 3,3        22,8           16,0
                                     816

-------
                  BREAr
                   SLIDE 12
BY WEIGHT OF PNAs FOUND AT THREE LIQUEFACTION FACILITIES
                                   FACILITY
                                FACILITY II
FACILITY III
co
iwo RING COMPOUNDS
PiEAN (UG/M-)
(PERCENT)
RANGE (UG/M3)
(PERCENT RANGE)
NO, OF SAMPLES
THREE RING COMPOUNDS
{:;Qp( j - V. UG/ M /
(PERCENT)
— . r. • ! r- T- /• / } \
- ' » i .. i -*- [ i {r / v ]
i \j i • \ -J '_ •„ L C/i i /
(PERCENT RANGE)
-, T o or* o {\ * K n T ?~ ""*
i'U = Or SAnPL^
FOUR PLUS RING
COMPOUNDS
njlnh vJG/M /
(PERCENT)
r, r • • /-• •" / / ' %
Mi-iGt (UG/M")
(PERCENT RANGE)
i-JO, OF SAMPLES
AREA
27,2
(49,3)
0,12 - 72,7
(20 - 78,4)
8

26,5
(46,0)
•0,4 - 119,
(18,7 - 73,3)
8


0,7
(2,1)
C,Q2 - 2,4
(0,7 - 4,7)
8
PERSONAL
52,8
(72,8)
0 - 244,5
(0 - 94=5)
19

7,8
(24,8)
0,6 - 19,8
(6,2 - 100,0)
19


0'.'9
(3,3)
0 - 4,5
(0 - 25,4)
19
AREA
52,9
(97,5)
14,8 - 72,2
(89,3 - 99,8)
12

0,6
(1,7)
0,03 - 1,5
(0,3 - 7,2)
12


0,4
(0,9)
0,01 -1.0
(0,02 - 3,5)
12
PERSONAL
0,2
(100)
0,02 - 0,5
(100)
25

0
-
-

25


0
25
AREA
7,2
(68,8)
0 - 18,4
(0 - 94,1)
13

1,0
(28,1)
0.01 - 4.6
(2,9 - 100)
13


0.1
(2.6)
0 - 0,8
(0 - 16.7)
13
PERSONAL
14,2
(86,1)
0 - 42,0
(0 - 97,1)
14

0,8
(13,5)
0.05 - 1.5
(2.6 - IOC
14


0.04
(0.4)
0 - 0,99
(0 - 1.1)
14

-------
CONCLUSION

     BASED ON THE RESULTS OF SURVEYS CONDUCTED AT THE THREE FACILITIES,
A PRELIMINARY ASSESSMENT CAN BE MADE WITH REGARD TO THE COAL LIQUE-
FACTION WORKPLACE ENVIRONMENT,   THE STAGE 1 RESULTS INDICATE THAT THE
SPECTRUM OF ORGANIC CONTAMINANTS FOUND IN THE LIQUEFACTION WORKPLACE
ENVIRONMENT FOR THAT PLANT ON THOSE DAYS  UNDER THOSE CONDITIONS IS
NOT AS DIVERSE AS WOULD BE EXPECTED, CONSISTING OF THE LOW-MOLECULAR
WEIGHT AROMATICS WITH ONE TO THREE RINGS  APPARENTLY PREDOMINATE IN
THIS ENVIRONMENT,  HOWEVER,  MOT ALL LOW-MOLECULAR WEIGHT SPECIES
WERE PRESENT,  NOTABLY ABSENT WERE THE PHENOLS AND CRESOLS WHICH,
ALTHOUGH BELIEVED TO BE PRESENT IN THE LIQUID EFFLUENTS WERE NOT
DETECTED IN ANY OF THE AIRBORNE SAMPLES,

     STAGE 2 PERSONAL MONITORING RESULTS  INDICATE THAT EXPOSURE TO
THESE LOW-MOLECULAR WEIGHT AROMATICS VMS  IN THE PARTS PER  MILLION
RANGE AND FOR THE PNAs,  IN THE MICROGRAM~PER"CUBIC~METER RANGE.   AT
THE PPM LEVEL, COMPARISON OF BENZENE,  TOLUENE,  XYLENE AND  THE AROMATIC
AMINES WITH CURRENT OCCUPATIONAL HEALTH STANDARDS INDICATE THAT THESE
COMPOUNDS ARE WELL BELOW THE OSHA LIMITS.

     THERE is AN ABSENCE OF  TOXICOLOGICAL DATA TO ASSESS THE HEALTH
HAZARD OF PROLONGED PNA EXPOSURES AT THE  MICROGRAM-PER-CUBIOMETER
RANGE.   HOWEVER, TOXICOLOGIC STUDIES HAVE SHOWN THAT PROCESS STREAM
EXTRACTS OBTAINED FROM THE PROCESS STREAMS OF DIFFERENT COAL CONVER-
SION PROCESSES EXHIBITED CARCINOGENIC  PROPERTIES WHICH WERE ATTRIBUTED
TO THE PNA CONSTITUENTS (2-7).   Sl-NCE  A NUMBER OF PMAs HAVE EXHIBITED
                                   818

-------
CARCINOGENIC PROPERTIES IN ANIMAL TOXICOLOGY STUDIES (3,9), IT COULD
BE ASSUMED THAT THE PNA CONSTITUENTS FOUND IN THESE PROCESSES MAY BE
POTENTIAL CARCINOGENIC HAZARDS AS WELL,   THIS SUGGESTS THAT WORKERS
WITHIN THESE FACILITIES MAY HAVE AN ADDED RISK OF DEVELOPING CANCER
BECAUSE OF PNA-INDUCED CANCER RELATIVE TO WORKERS WITHOUT SUCH
EXPOSURE,   WHAT REMAINS UNDEFINED is THE QUANTIFICATION OF THE RISK
FACTOR,

COMPARISON OF NIOSH CONTRACTOR AND DOE INDUSTRIAL HYGIENE DATA

     THE DATA COLLECTED BY OUR CONTRACTOR AT THE COMPREHENSIVE SURVEY
WAS OBTAINED DURING OPERATION OF THE SRC-II  PROCESS,  FOR MEANINGFUL
EVALUATION THIS DATA IS BEING COMPARED WITH  DOE RESULTS OBTAINED DURING
SRC-II OPERATIONS,   BECAUSE OF DIFFERENCES IN SAMPLING AND ANALYTICAL
PROCtbUKhS,  THE COMPARISON OF DATA FROM THE  TWO SURVEYS IS CONCERNED
WITH MAGNITUDE OF THE OBSERVED CONCENTRATIONS FOR EACH PROCESS ARF.A,

BENZENE, TOLUENE, AND XYLFNE

     A COMPARISON OF THE LEVELS FOR BENZENE, TOLUENE,  AN.O XYLENE IN
THE COAL PREPARATION, MINERAL SEPARATION, AND SOLVENT RECOVERY AREAS
SHOWED SIMILAR LEVELS FOR BOTH AREA SAMPLES  AND PERSONAL SAMPLES,
HOWEVER, IT SHOULD BE NOTED THAT ONLY ONE SAMPLE WAS TAKEN IN EACH
AREA BY OUR CONTRACTOR.
                                81!

-------
     THE DOE DATA TAKEN OVER A PERIOD OF 12 MONTHS SHOWED LITTLE
VARIATION AMONG SAMPLES WITHIN PROCESS AREAS TESTED:  MORE THAN 90
PERCENT OF THE SAMPLES HAD VALUES BETWEEN 0,01 AND 0,05 PPM FOR
BENZENE, TOLUENE, AND XYLENE.   OUR CONTRACTOR'S DATA TAKEN 8 MONTHS
LATER FALLS WITHIN THIS RANGE, WHICH SUGGESTS A UNIFORMITY IN PROCESS
EMISSIONS FOR THESE COMPOUNDS IN THE SRC-11 PROCESS,
PNAs
     WITH REGARD TO THE PNAs,  DOE SAMPLES WERE ANALYZED ONLY FOR
BENZO(A)PYRENE (BAP),   AN EVALUATION OF CONTRACTOR'PERSONAL SAMPLES
SHOWED SIMILAR LEVELS  FOR BAP  IN ALL. TESTED PROCESS AREAS EXCEPT
THE PRODUCT SOLIDIFICATION AREA,  FOR THE TWO SURVEYS,  LEVELS RANGED
FROM NONDETECTABLEL TO  LESS THAN 9,01 UG/.V, WHEREAS DOE HAD LEVELS
ONE TO TWO OiiDERS OF MAGNITUDE HIGHER (0,13 - 0,99 UG/1V) ,   THIS
DISCREPANCY IN THE PRODUCT SOLIDIFICATION AREA CANNOT BE PROPERLY
L'VALUATED BECAUSE PLANT OPERATING STATUS INFORMATION DURING THE DOE
SURVEY WAS NOT AVAILABLE.

     THE DATA ARE COMPARABLE,   BOTH DOE AND OUR CONTRACTOR RESULTS
AGREE, OR AT LEAST ERR IN THE  SAME DIRECTION,
                                 82C

-------
                                 REFERENCES
 1.  Biologic  Effects of Atmospheric Pollutants - Participate
     Polycyclic Organic Matter, Committee on Biologic Effects of
     Atmospheric Pollutants, Division of Medical Sciences, National
     Research  Council, National Academy of Sciences, Washington, DC. 1972

 2.  Epler  JL, JA Young, AA Hardigree, TK Rao, MR Guerin, IB Rubin,
     CM Ho  and BR Clark; Analytical and Biological Analyses of Test
     Materials from the Synthetic Fuels Technologies-, Part I;
     Mutation  Research 57:265-276; 1978.

 3.  Hueper WC, Experimental Studies on Carcinogenesis of Synthetic
     Liquid Fuels and Petroleum Substitute's, Industrial  Hygiene and
    •Occupational Medecine 8:307-327, 1953.

 4.  Hueper WC, Experimental Carcinogenic Studies on Hydrogenated' Coal
     Oils.  Industrial Medecine and Surgery.,  pp. 51-55, February ]956.

 5.  Rao TK, JA Young., AA Hardigree, W Winton and JL Epler:  Analytical
     and Biological Analyses of Test Materials from 'the  Synthetic
     Fuel Technologies Part II; Mutation Research  54:185-191,  1978.

 6.  Rubin  IB  and MR Guerin; Fractionation. cf Synthetic  Crude Oils  from
     Coal for  Biological Testing, Environmental Research 1?:  358-365.  1976.

 7.  Weil CS and Ml Condra, Hazards to Health in the HydroCjenaticn  'of
     Coal - Part II Archives Environment?!  Health, 1:  187-193.,. 1960.

 8.  Weisburger Jli, Chemical Carcinogenesis, Toxicology _-  the Basic Science:
     of_ Poisons.' LJ Casarett and J Doull e-d. , •Macniil'Uin  Publishing  Company
     pp. "337-343, 1975.

 9.  IARC Monographs on the Evaluation of Carcinogenic Risk  of  the
     Chemical  to Man:  Certain Polycyclic Aromatic Hydrocarbons  and
     Heterocyclic Compounds - Volume 3, International  Ag.'ncy  for Research
     on  Cancer,. World Health Organization, . 1973.
10. Roberts,  JD} and MG Caserio;  ^todjrn_^Orjj;cnT:l.c;_
    WA  Benjamin, Inc.,  N.Y.  (1967)" J ~   '   ~~ ~

3.1. NIOSH Manual of  Analytic Methods,  2nd. Ed.,  Vol.  I.
    U.S.  Departinont  of Health,  Education and  Welfare  Publication
    No.  77-157A.
                                         821

-------
                               Discussion
Dr. Weisburger (NCI):  I have one or two comments.  First, what was this
coal-derived liquid that was used for the process you mentioned?

Dr. Berardinelli:   It is a mixture of aromatics, i.e., anthracene oil in
which the coal is  then slurried.  At facility  number 2, I think,
methyl-naphthalene is probably one of the biggest constituents.

Dr. Weisburger (NCI):  Also, you have compared the extent of exposure to
the polycyclics with -- I mean, between these people and people, say,
who are out standing on the street corner in some busy city, you know,
where there are lots of automobiles going by and lots of combustion
taking place; what is the relative extent of exposure?

Dr. Berardinelli (NIOSH):  Okay, that is a very valid question.  This is
something that I think we will  pursue further.  The big problem here is
analysis.  I think EPA was the first one to point out that we cannot use
just a silver membrane filter.   If you look at the NIOSH methods of
sampling analysis  for PNA's, a silver membrane filter is recommended.
Well, we know that this is not entirely suitable because we are missing
the whole vapor phase.  A sorbent resin is needed to catch the vapor
phase PNA's.

But what I am saying is that some of the old data really cannot be
compared to our data because the techniques are different.   In addition,
we have some problems as to analytical  capabilities.  Most people have
used the cyclohexane-soluble fraction and tried to use that as a PNA
indicator.  EPA has done work where they have looked extensively at BAP-

I would say that BAP is in the high nanogram per cubic meter range for
ambient levels, I  think in the Pittsburgh area.   I believe more
important would be to compare PNA's to other work places.   One thing we
can do there is to look at some Scandinavian data, which I  do not have
with me.  However, they looked at coke oven emissions using an
analytical scheme  which analyzed 40 PNA's, and found total  PNA's in the
milligram per cubic meter range.

So to answer your question fully, I really have to kibitz  a bit, because
I will say that PNA's seem to be worse than the rural environment.

-------
                             TRADE NAME  INGREDIENT
                           DATA BASE—PROGRESS  REPORT

                               Herbert L.  Venable
                      Hazard Section, Surveillance Branch,  DSIIEFS
                               NIOSH - Cincinnati, Ohio

     To assist in carrying out  its assigned  mission in the  field  of occupational
health, NIOSH is developing a trade name ingredient data base.  This data base
is intended to provide NIOSH with a more realistic perspective  of the problems
in the workplace, particularly  hazards.   As  a result, effective programs  and
policies are designed and implemented.

     This data base grew out of the first National Occupational Hazard Survey
(NOBS I).  Ingredient information on approximately 60,000 products has been
compiled through the trade name ingredient clarification (TNIC) process which
involves contacting manufacturers, editing incoming information,  automating the
data, and safeguarding confidential formulas.

     To fully appreciate the complexities encountered in obtaining this data, a
brief description of the TNIC process will be presented  with  special emphasis
on obtaining information through administrative subpoena action *   This action
and authority for such action,  i.e. Section  15  OSHAct 1970  and  the McGee
decision, will be described briefly. The involvement of foreign  manufacturers
provides a special case in the  authority to  obtain this  information.   This  and
ether problems unique to the development of  this data base  will be presented.
The current status of the data  base will be  included in  the summary of this
report.
                                    82

-------
                             WORKSHOP PAPER
In carrying out its assigned mission in occupational safety and health,
NIOSH requires a reliable data base on the nature and extent of potentially
hazardous exposures in the workplace.   The National Occupational Hazard
Survey (NOHS) of 1972-1974 developed potential exposure information in a
sample of nearly 5,000 plants.  Almost 70% of the exposures noted by the
surveyors occurred in the form of trade name products.   In an effort to
clarify those exposures, NIOSH is developing a trade name data base from
information gathered in the National Occupational Hazard Survey-  To date,
ingredient information on approximately 60,000 products has been compiled
through the Trade Name Ingredient Clarification (TNIC)  process.   This
data base will be enlarged by the addition of resolved  trade name data to
be gathered in a second survey.  This  survey is tentatively scheduled to
begin early in fiscal year 1981.

The uses of the NOHS data base are numerous and varied  and include the
development of estimates of numbers of workers potentially exposed to
hazards, cohort location, and development  of lists of potential  hazards
associated with industry types.  The utility of the NOHS data base is
directly affected by the accuracy and  completeness of the trade  name
product ingredient information.

The trade name resolution process currently being used  consists  of the
following steps:

     1.  Identify the manufacturer or  distributor of the product.

     2.  Contact the manufacturer by certified mail requesting the
         information.

     3.  Edit responses to ensure accuracy and conformance with  the
         required xormat

     A.  Code and automate the data.

     5.  Input the data to the NOHS data base.

In practice, however, the process of obtaining this information  from
product manufacturers is not a simple  matter of going through the
steps described.  Numerous problems requiring clarification or further
assistance are encountered.  Briefly,  some of the mora  common problems
are:

     1.  The manufacturer is unable to identify a product by the name
         or number on file.  Custom formulations, especially in  paint
         and ink manufacturing can be particularly difficult.  Several
         sources of error are possible here:
                                         824

-------
         a.   The product  name  or manufacturer was  incorrectly  recorded.

         b.   There was  an error in keypunching the information.

         c.   The product  name  has been  changed.

         d.   Insufficient descriptive information  was recorded e.g. need
             a batch  or customer order  number was  needed.
     2.   A second  major problem  occurs when we  cannot  readily  identify  the
         correct manufacturer  of the product; usually  for  one  of  the
         following reasons:

         a.   The manufacturer  is no longer in business,  at least  not by the
             name  recorded.

         b.   The original manufacturer has merged with another company,
             usually involving a name change.

         c.   Product formulation rights have been sold to  another company,
             usually involving a product name change.

     3.   There have been  cases where the manufacturer  chooses  to  give us no
         response  or refuses to  accept certified mail.

     4.   If a request is  sent  to the wrong division  or department within a
         large corporation, the  recipient of the forms may not be able  to
         respond.

     5.   In some c.ises,  there  is no definite formula for an ingredient  in
         a product,  Thi.; often  occurs in pigments,  dyes,  oils, fats, and
         pol}Tiiers.  In some instances boiling point  ranges or  length of
         carbon chains nu-.y be  of some help.  This problem  is further
         complicated when ingredient composition varies  because of varia-
         tion in f'ied stuck used in its manufacture.   Efforts  to  resolve
         petroleum based  products is a good example  of this problem.

     6.   Manufacturers occasionally refuse to respond  because  the
         information requested is considered a  trade secret.

Approximately 30%  of the  trade name products in the  current data  base have been
claimed  as proprietary formulas  by manufacturers.  Obtaining and  handling
trade secret information  presents unique problems.   There  is considerable
reluctance by some manufacturers to provide trade secret information to NIOSH
because  of fear of accidental  release of such information  to competitors.  This
is particularly true of small  chemical manufacturers who market one or  a very
few products and may also be true with larger manufacturers who market
relatively unique  products.
                                      825

-------
In a number of cases the impact on manufacturers providing this information is
considerable in terms of time, money,  and manpower.   This  is especially true
for those manufacturers who market hundreds  or even  thousands of products
for which information is requested.

Unfortunately this impact may be largely duplicated  when information similar
to that requested has already been provided  to another Federal agency,
e.g. Food and Drug Administration, Environmental Protection Agency,
Consumer Product Safety Commission, but  cannot be accessed by NIOSH for
the purpose of the development of the  trade  name data base.ning

Trade secrecy also becomes an issue when one or more of the ingredients
of a product are manufactured by another company which consideres their
ingredients to be proprietary information.   Nondisclosure  agreements
between manufactureTS are honored by NIOSH by re-routir.g the request for
information directly to the primary manufacturer.

Under Section 15 of the OSHAct of 1970 NIOSH is given the  authority to obtain
trade secret and other confidential information.  This section also requires
NIOSH to maintain and safeguard this information.  This legislated authority,
as described in the OSHAct, is explained to  manufacturers  when necessary.

This authority was challenged in the case of the United States of America
versus McGee Industries, Inc. in January 1977.  In this cases NIOSH was
seeking the enforcement of an administrative subpoena (duces tecum) issued
by NIOSH to McGee Indue tries, Inc. to  obtain trade secret  information that
McGee Industries had refused to provide.   The decision in  this case, and
a subsequent appeal, was ruled in favor  of NIOSH.  This case has since
served as a legal precedent for NIOSH  in obtaining trade secret information.

The use of administrative, subpoena to  obtain information from noncooperative
manufacturers is uj.ed when other measures fail.  Although  this is not a
court action, it does require the manufacturer to either comply with the
request for information cr show just cause as to why it should not have to
comply.  If this sibpoen? is not observed, court action is required.  To
date, only the McGee case has gone to  court  and a decision rendered.

We have attempted to minimize some of  the problems involved in obtaining
ingredient information through the careful design of the materials which
are sent to the marufacturers.  The initial  letter and directions for
completing the response torms must be  carefully worded.  The letter must
be explicit and sufficiently comprehensive in describing precisely what
information is requested and why without confusing or intimidating the
recipient.  Similarly, the directions  for completing the forms must be
simple and direct, without being perceived as threatening.  Clarity of
the response form is also important.  In some instances, copies of the
relevant portions of the OSE(\.ct and/or the McGee decision are sent along
with the letter to provide information.   It  has become apparent to us
                                     826

-------
that communicating with the manufacturers of trade name products requires
a well-designed package of basic letters and forms, with considerable
flexibility to accommodate the unforeseen.

In addition, NIOSH has developed an unpublished Trade Name Product Security
Policy for the TNIC process which includes background information of the
National Occupational Hazard Survey, applicable leg-al statutes, physical
security measures, data access procedures,  and restrictions on data use.
Copies of this policy are made available to manufacturers upon request.
Despite these measures, several manufacturers have drafted up 'their own
security agreements.  These often take the form of an affidavit requiring
the signature of a NIOSH representative as  a precondition for compliance with
NIOSH requests.  NIOSH has taken the position that such agreements are not
necessary, and has further stated that the Trade Name Product Security Policy
provides appropriate protection for trade secret data.   The penalty for
disclosing trade secret information by a federal officer or employee is
spelled out in 18 United States Code, Section 1905.

NIOSH's authority to obtain ingredient information on products manufactured
by a foregin manufacturer rests with the fact that foreign-made products
do in some cases involve exposures to American workers.   Compliance to our
request by foreign manufacturers has so far been on a voluntary basis.
Nondisclosure- agreements between foreign and domestic companies, or between
foreign parent companies and American subsidiaries are  honored.

The current status of the trade name data base (as of February 1980) is  given
in the accompaning cable.  Note that various code letters are assigned, to a
category of products depending on the status of resolution.  These symbols
are as follows:

     E  =  trade nc.ine entered into system,  to be resolved

     1  =  first request letter sent out

     2  =  second request letter sent out

     N  =  referred to NIOSH surveillance staff by contracted
           clerical personnel for assistance

     S  =  satisfactorily completed response

     H  =  holding for additional information or decision

     R  =  response received containing chemical information

     0  =  mail returned by post office
                     Ho discussion followed this paper.
                                        827

-------
                                TABLE
             CURRENT  STATUS OF THE RESOLUTION OF PRODUCT TRADE
             NA>ES  IN THE NIOSH PRODUCT TRADE NAME DATA BASE
             CODE                         NUMBER OF PRODUCTS

              E                                    553

              1                                  1,069

              2                                  2,683

              N                                  6,/:31

              S                                 61,599

              H                                  2,145

              R                                  4,882

              0                                    174
In closing,  I would like  to  reiterate that the validity and reliability
of the trade name data  base  and  the  larger NOHS data base is directly
related to the accuracy and  completeness of the trade name resolution
effort.  Subsequent use of this  data base by NIOSH and other users will
only be as effective as the  data base itself.  Although the file on trade
name ingredients  x^as initially compiled to complete the NOUS data base,
it has emerged as an important resource in its own right.  Our staff
receives 600-700  requests for product ingredient information each year
from field industrial hygienists,  researchers, and other members of the
occupational safety and health community.  We continue to regard the
identification of trade name products, which are potential occupational
exposure agents,  and the  subsequent  clarification of those products into
their respective  ingredients to  be an important function of hazard
surveillance.
                                       828

-------
FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:

    PROGRESS ON 3OINT ENVIRONMENTAL AND

        OCCUPATIONAL CANCER STUDIES
            Thursday Afternoon, May 8
               PLENARY SESSION
             SESSION CHAIRPERSON
             Dr. Gregory T. O'Conor
             National Cancer Institute
               PANEL MEMBERS
 Chairpersons from Working Group Sessions A, B and C

            Dr. Joseph Fraumeni, NCI
            Dr. Richard Marland, EPA
              Dr. Norbert Page, EPA
           Dr. Kenneth Bridbord, NIOSH
            Dr. Nelson Leidel, NIOSH
                      82S

-------
                             PLENARY  SESSION

                         Thursday, May  8,  1980
Dr. O'Conor  (NCI):   Because we have  lost  a number of participants, the
organizing  committee  as well  as most  of  the  chairmen  of  the working
groups agreed that instead of repeating the working groups of yesterday,
we would  be  better  to proceed directly to  the  plenary session.  But  if
there are specific questions relating to some of the papers this morning
which anybody  wants  to bring  up,  I think  there  will- be  ample time  to
cover virtually every subject that any individual would like to mention.

This  is   the  windup   of  what  I  think  has  been  a very  successful and
productive  meeting.   Instead  of  saying  it at the  end,   I  will  say  it
now.  I think on behalf of all of us we would like to thank Dr. Kraybill
and  the   other members  of  the  organizing  committee for  taking the
initiative  to  plan  and organize  this meeting  at this  time.    It  is  a
first and has  become  a  very effective program.   I think we are all glad
that this meeting has taken  place  and hope that it will be the first  of
many more to come, each one more effective  and productive  than  the  last.

The way  I thought we might  start this  is  to  have  one member of each  of
the working  groups to summarize  the  events of yesterday and to make any
further  comments  he  might  like to  make.    We can  open  that  for limited
discussion  and  then  have the  reports discussion after that.   So may  I
start  by asking  Dr.  Bridbord to  report  on  the  working  group  and his
reflections on the meeting in relation to epidemiologic studies.

Dr. Bridbord (NIOSH):  Thank you.  I think  we had a very  interesting and
interactive  session  dealing  with  epidemiology needs  and opportunities
for  collaboration that span both  the  working  as  well  as  the general
environment.   I  tried to  condense  the discussion  into  roughly a  dozen
areas  and perhaps Dr.  Fraumeni might  also  come  up  here and  add any
additional points that come to mind that I  may not have touched on.

One of the  most  intriguing  areas for  discussion,  which relates to both
the workplace  as  well as  the  general  environment,  is follow-up studies
secondary  to  the  cancer  maps  that  NCI  has  developed,  which  have
generated  many hypotheses  and  suggestions for new  avenues  of studies,
both  looking at  communities  around specific  point  sources as well   as
particular  working  groups.   These opportunities  include, for example,
case control studies  and additional studies using hospital records.  But
certainly  I think  there  was  a  general  consensus that  some   follow-up
studies should be given a high priority.

In  this  regard,  we  spent  some time talking  about the  need  to pay more
attention  linking  indices  of exposure  to  particular  substances,   to
                                    830

-------
specific  end  health  effects,  be  they  based  on   either   indices   of
mortality  or  morbidity for  that matter.   Here we  had  a fairly  lively
discussion  about the  role  of people  in  the area of chemistry and  how
important  it  is for the  chemist to be  interacting  with the people  who
are  more  health  oriented  and  for  the  physical   scientist   and   the
biological  scientist to  understand  the strengths of each approach  and
some  of  the   limitations that  go  along  with those  approaches.    We
collectively  felt  that the  National Death  Index  offers a particularly
important  opportunity  for  studies  looking  at the  risks that  may  be
associated with community exposures  as well  as  workplace  exposure.  Here
the existence of the  National  Death Index  might  actually allow a much
more  aggressive posture  with  respect  to  prospective studies,  at least
forming  registries  of exposed  populations  that might be  assessed  a
number of years from now.  These registries  might  help to clarify either
some  existing situations  or  may help to clarify risks in newly  emerging
technologies,  particularly where we can  start  to define groups exposed
to specific materials  at  the onset.

Some  of  the  examples  that were  suggested  included  the possibility  of
looking  at the population  in  Duluth,  Minnesota  exposed  to  asbestos  in
drinking water  and using  the National  Death  Index  to  follow a very large
number of people exposed  to that common  source.

Three other examples of emerging technologies that might  be particularly
important for occupational safety  and  health, as well  as  perhaps in some
cases for  populations  residing in communities,  include the emergence  of
coal  gasification  and  liquification  as  an  additional  source of petroleum
products,  the emergence of the recombinant DNA industry  which is  likely
to  grow  at a very  rapid  rate, and some of the  dramatic changes that
occur  in the  electronics  industry.   These  are just three examples,  in
addition to Duluth,  of  situations  that might be looked at from the point
of view  of registries  and  more  prospective studies.

While the  main point of our discussion centered around  risks of cancer,
there was  also a fair recognition that there was  an  opportunity to gain
some  insights in the  area of reproductive  effects, effects that  might be
mediated  through  exposure of men  as well  as women to toxic  agents,  and
the need  to  look at that area in  its  own  right and  also  to consider  how
that  related  to the whole problem of cancer  risk. Here  it was felt that
we  should  be  a bit  more  flexible  in the future to consider those types
of studies  in this  collaborative program.   I know  that has importance in
the case  of  OSHA and in the case  of EPA  as  well.   In the case of  NIOSH,
we frequently get  questions  about  those  kinds of  risk.

Speaking  of  OSHA and  EPA, there  was also a  recognition  of the  need  for
the health researchers to' be  as  responsive as  possible  to the  needs of
the  regulatory  agencies, in  this  case OSHA  and EPA,   but  also other
regulatory agencies  as well.   The issue of  asbestos  was  raised  again as
one  which  is perhaps  worthy  of  some  additional  study.  The  one  very
concrete  suggestion was  to  consider  the  possibility of  a mesothelioma
                                831

-------
registry and that this may give us some insights in terms of populations
exposed  that may  be  at  risk  though  we  do  not  currently understand
precisely what that risk status might be.   From the EPA perspective,  the
issue  of  indoor  air  pollution  was  raised.   There  is  concern  about
exposures   to   many  toxic   substances  in   the   indoor  environment,
particularly as we  move toward energy  conservation  in cutting down  the
air turnover rates  within  buildings.   This would tend to accentuate  and
exacerbate  any potential indoor exposure  situation.

A number of points were raised, including the  problem  of asbestos inside
buildings,  as  well  as some  of  the  problems with  urea formaldehye
insulation, particularly in trailers.

The  area of control  technology  was  also discussed.   It was recognized
that there  needs to be  some consideration given to  looking  for  solutions
to problems  as  well  as  identifying problems.  Up until now, we have  not
had major  emphasis  in the control technology  area.   This  is one  that  I
think  is particularly important.   I  know it is important to NIOSH and  I
know  it  is  important  to OSHA.   It applies  to  the general environment as
well as  to  the working  environment.

A  very specific suggestion which  would be  particularly  helpful   in  the
case of  cohort mortality studies  looking  at risks in the workplace would
be  to  spend  some  of  our  energies   developing  a  fairly  comprehensive
control  population with which to compare  specific working populations in
order  to  assess  potential risks  associated  with  cancer  or  deaths  of
certain  types  in  specific  industries.   Here,  in  general,  in  workplace
studies  we  tend to  use  the general population of the  United States.   To
the  extent that workplace hazards represent  an  important  part  of  the
overall   occupational  and   environmental   cancer  problem,   then   we
underestimate what  those  true  risks  might be by  looking at the United
States population instead  of a more comparable control group, that is,  a
more  healthy working population  but  not a  population exposed  to  toxic
agents.  This is something that would have  potential long-term  benefits,
if we  could  develop that.

Other  ideas for studies that  were mentioned,  and these do  not  foreclose
other  opportunities  either,  include  the point that farmers appear  to be
an  increased  risk for leukemia and multiple mye-loma.  This  has recently
received  some attention.   Another idea  is in the  area of nitrosamine
exposure,  where we  have a  fairly  good amount of data from animal  studies
but  not a  lot  of  corroborating  evidence  in  terms  of specific  exposed
human  populations,   largely because  I  do  not  think  we  have  had  the
opportunity to  do  additional  follow-up  studies  and to try to  correlate
some  of  the animal  data with human observations.

I  think  there was a  strong sentiment raised that it  would  be  important
for  the  federal  agencies  that  depend  upon  the  data  from the  Social
Security  Administration   to  conduct   epidemiology  studies  to   be   as
consistent   as   possible   in    requests   to   the   Social    Security
                                 832

-------
Administration,  but also to make  sure  that our collective  requests  get
across  the  importance  of  having  further  access  to  those particular types
of  data for  epidemiological  studies.

Additional  opportunities for  study were  suggested  in terms  of  foreign
countries,   particularly  the  Scandinavian  countries.    Sweden  is  as
outstanding  an  example of that as any country, where  many registries of
exposed populations exist and  provide an opportunity  to  link  health  and
exposure data that  do not  exist  in  most countries.   Here it was  also
mentioned that  in the  area of occupational safety and health, there  has
been  a  start to have better collaboration  between the United  States  and
the Swedish  researchers.

Finally, we  did  note some  words and  suggestions on  where to  go  from
here.   One  of the suggestions would be  that we have a briefing for each
other  on our proposed  new  starts for FY  1981.   This  is  among  EPA,  NCI
and NIOSH.   We should make  sure that in  looking  to  future  fiscal  years
we  do not  inadvertently  duplicate some  of our  work,  but as important, we
should  also  look  for opportunities for  future  collaboration.   We should
try to  get  scientists  from  each  of the  agencies to work closely together
and to  help  solve some of these  important problems.

Dr. O'Conor  (NCI):  Thank you,  Dr. Bridbord.   Dr. Fraumeni, do you have
anything you would like to add  to Dr.  Bridbord's remarks?

Dr. Fraumeni (NCI):  I think Dr. Bridbord has  summarized very nicely the
meeting that took place  yesterday.  I  do not  have  very much more to add,
except  that  I think we should  be mindful that  there  are federal  agencies
other than  NCI,  EPA and  NIOSH  with experiences and resources that may be
 indispensable for  collaborative  projects.   For  example,  Dr.  Caldwell
from  CDC  is  part  of a  very  active   group  working  in the field  of
environmental cancer.   He  and  others  should  be brought  into the  total
picture when we talk about  federal interagency efforts.

Dr. Bridbord  talked  about  a special  priority being given  to  certain
types   of  studies.   These   should  also  include efforts to clarify  the
effects of  widespread  population exposures,  to help  resolve  matters of
public   health,   public   policy,   or  regulatory  concern;   and  to  help
contribute  toward a better  understanding of carcinogenic mechanisms.   It
 is  important,  for example, not  only to  know that  nitrosamines  in  the
workplace  may  give   rise  to  cancer,   but if  we  were  to  discover  a
relationship,  it  would   have   much   wider   ramifications  toward   an
understanding of  the role of nitrosamines in  human cancer generally.

Dr. Bridbord has  discussed  the obvious  importance  of analytic studies in
cancer   epidemiology,   particularly  case-control   and  cohort  studies.
Perhaps more emphasis  should be  given to the  utilization and  integration
of  routinely collected  data at  the  federal  level.    This includes data
resources  of the  Social  Security Administration,  the National  Center for
Health   Statistics,  the  Bureau  of  Census   and  the   Internal  Revenue
                                833

-------
Service,  toe  need  to collaborate with  investigators  at these agencies.
For instance, we need  a system of occupational  mortality statistics in
selectea states, which would be carried-out with the National Center for
Health  Statistics.    We should  make  special  efforts  to  preserve  and
utilize the  data  from private industry,  labor  unions,  insurance plans,
nealth  care delivery systems,  federal  employee  records, and, of course,
death certificates.

breat  emphasis  should  be  given  to  multidisciplinary  studies  between
epidemiologists and  experimentalists.    This will  help resolve several
issues.  More efforts should be given to measuring chemical  and physical
agents  in the environment  in  the  context of an  epidemiologic evaluation
of cancer risk.  we should strive to identify the subclinical effects of
environmental agents  in  Dody  tissues and fluids  and  attempt to measure
nost   susceptibility   through  laboratory   markers  in   the  hope  of
elucidating  environmental   interactions.    Here  again,   the effective
conduct  of  these  types of  studies will   require  the  coordination  of
resources and talents in the various agencies.

coordination, of course,  is  critical in these  interagency efforts.   At
the very simplest  level, this  may involve,  as suggested by  Dr. Spirtas,
an  exchange  of  organizational   listings   of  epidemiologists  and  the
routine exchange  of information  that  may  help  stimulate collaborative
research  to  fill  gaps  and  also  prevent   any  wasteful  duplication  of
effort.   We  might  consider   periodic   interagency  meetings  to  review
fastoreaking  developments  in  epidemiology  and  to  uncover  special
opportunities for further collaboration.

Although our focus has oeen environmental and occupational chemicals, we
snould  taxe account of pressure that is mounting to evaluate  the effects
of  radiation.    'we  need  studies  of  sufficient  power   to  evaluate
populations  exposed  to  ionizing  radiation  at various  dose   levels that
can be  reasonaoly quantified in order to provide risk estimates that are
required in  setting  radiation protection  guidelines  for occupationally
exposed  groups  and  for  persons  exposed   to  medical  or environmental
radiation.

vie  also need  to evaluate the  potential  hazards  of the depletion of the
ozone   layer  in special   surveys  to  monitor  trends  in  skin  cancer,
including melanoma, along with analytic  studies to measure the impact of
ultraviolet radiation and other risk factors.

ur.   dridbora   mentioned  the  importance   of   international  studies,
particularly   in   countries   that   have   unique   data  resources,  or
environmental exposures  that  are  unusually heavy  or  early.   An example
would  be the  exposure of workers  in coal  gasification plants in Europe
ana  South  Africa  that  started many years  ago.    Of  course, the Atomic
bomb  Casualty Commission  in  Japan may  be  the most  valuable radiation
resource in tne world.
                                 834

-------
We  need  more epidemiologic  inputs into preventive  measures.   Further,
epidemiologic  and biometric  research  is  needed  on programs  aimed  at
primary  and  secondary  prevention  of  cancer,   including   intervention
studies in high-risk groups.  This includes, for example, the assessment
of  work  practice and industrial process  changes  and screening programs
that  may enable  the early  detection  of environmental  cancer.    Here
again, the utility of  laboratory indicators  of preneoplastic states  and
the  effectiveness of  chemoprevention on  the  natural  history  of these
lesions should receive more emphasis.

Also  needed  is the further development of biostatistical methodology  to
assist the area  of risk  assessment,  including methods for extrapolating
carcinogenic  resonse in  laboratory  animals  to man,  especially at  low-
dose  levels.

Finally,  in  studies  that are primarily  designed  to  study cancer, it  is
crucial  that we   not limit  our  attention  to cancer  but also search for
other    potential   effects   of    environmental    exposures.       The
interrelationships between cancer and  other effects may  shed  light  on
important biological mechanisms.

Dr. 0'Conor  (NCI):  Thank you.

I  think   it  is  clear  from those  two summaries  that the epidemiologic
studies  will continue  to  play  a very large and  important  role in  this
program.   I was  interested  to  note that  in the  last few  areas  Dr.
Fraumeni  mentioned there was  interest among  the epidemiologists in  some
of  the  very  specific  areas that  were discussed  in  the second working
group.  So I think this  is a healthy  sign  of the recognition of the  need
for  interdisciplinary studies.

Before we go on  to  the  next summary, I wonder if  there  are  any specific
questions that any  of  the participants  have  now  relating to either the
epidemiology papers  or  the  epidemiology  program.    Let's   keep  the
discussion  at this  time fairly  specific, because  we  will get  into a
broader  general  discussion later.   Are there any specific  points  that
anybody would like to clarify or raise at  this time?

Dr.  Morris  (EPA):    I   have  a  question.   In  the IRLG   (Interagency
Regulatory Liasion Group), there  is  a working  group on data formatting
and  standardizing data  presentations at  least  in  the testing area.   Is
there  any thought  or  is  it even  reasonable  to  consider  this  in  the
epidemiology area?   Because you  are  getting  a variety of data from a lot
of  different sources,  is there  any  need for us to consider any kind  of
standardization  in this  area?

Dr.  0'Conor  (NCI):   Do  either  Drs.  Bridbord or Fraumeni want to  answer
that?  It sounds  like something that  is  going on  in  NTP.  I  believe  that
Dr.  Chu was  associated with that effect  but  I do  not  see him here.
                                  835

-------
Dr.  Orme  (NCI):   I am not  sure  if that was  the  question,  which really
has  to do with epidemiology  standards.   As you know, this is a somewhat
controversial  area,  and   I  have  not   been  involved   in  the  IRLG
discussions.    However,  I  think   that  the  major  emphasis  of  the
interagency  program  that we  are  discussing  here should be  in pushing
oack the  frontiers  of knowledge  in  the field  of environmental cancer,
rather  than   trying   to  rehash  what   are   acceptable  practices  of
epiaemiologic research.

Dr.   Morris   (NCI):    I  do  not believe  I  was  suggesting  standards  in
epiaemiology.   I  was referring more to the  reports that are  presented
ana the development  of  data.  It  should be put into common format that
can be utilized by a number of agencies  and sources.  That is what I was
addressing.

ur.   Fraumeni  iNCI):   we  did not  discuss  this, but my  feeling is that
this particular issue should not be a major priority for our group.

Dr.   baloraith  (EPA):   I would  like  to address  the  comment  regarding
standards,  because  regulatory  agencies  run  into  problems   regarding
standards  for  epidemiologic  studies,  particularly  those  sponsored  by
non-regulatory agencies.  There have been  instances  within  the last six
months  in  which  I have  discussed  epidemiologic studies  with various
regulatory  groups within the EPA  and found that they did not want to be
involved  in the  planning of studies  because  they felt that involvement
in  tne planning  of  studies  to  be  conducted in  foreign  countries  may
create a  situation  in which they might  have  to endorse the conclusions
of  the study.    It was  feared the   study  might  lack  proper  controls.
Would you have any comment to make in that regard?

ur. Fraurneni (NCI):  Each new study  can  be evaluated on its own merit by
epidemiologists  at  each of  the  agencies  or  collectively.    Already at
hand  are  standard  methods   for  designing, conducting  and  interpreting
epidemiologic studies.  Each study can be  examined separately regardless
of  any new  standards that are promulgated.  We  do this all the  time.

Dr. Galbraith (EPAj:  Should there be minimum standards for epidemiology
studies?

Dr.  Fraumeni  (NCI):     I  am  rather  flexible  on  how  epidemiology  is
conducted,  and  favor a  wide variety  of approaches.   Therefore,  it is
important to ensure that any guideline or  standards  do not  inhibit fresh
approaches  in  etiologic research.   I am  not  sure  anyone is opposed in
trie  general sense to standards  of epidemiologic practice,  so this may
simply be a sematic problem.

ur.  U'Conor (NCI):    One  can consume  a  tremendous  amount of energy and
time  in  standardization and development of methodology.   I think what
Dr.  Fraumeni  is  saying,  and  I  certainly would agree  with  him, that
altnougn  a  certain amount of that  is  important, the  major thrust of this
                                   836

-------
program  should  be innovative and  should be concerned  with research  in
the broadest context in terms of etiology and prevention.

Are there any other specific points?

Dr. Alavanja (NIOSH):  Just as a point of information,  it seems  that  the
discussion  did  not get  around  to it, but  there is  a  working group  in
epidemiology for  the IRLG.   It  does have  a set of guidelines, not  for
standardizing epidemiology, but  to  sort  of  set forth guidelines for  the
documentation  of  epidemiological  studies.     It  was  not   felt  that
standards for epidemiology  or methodology should be created.   It should
not be  standardized  in  any fashion,  but the  full  documentation of  the
study may be  important.   In cases  where studies simply do not have  all
the information, where information may not  have  been  incorporated, there
may be   design  features  that  have simply  not been  reported  which  are
vital  to the interpretation of the studies.  So  this  is  an  activity that
is ongoing.  In fact, it was out for  public comment and  now  it  is at  the
stage of revision.  So these guidelines  for documentation are  out and it
is going to De something  which, if  it  holds  to its current  timetable,
will be  PUDlished as a  series of  guidelines  -- not standards  -- before
the end  of  the year.

Dr. 0'Conor (NCI):  Thank you.   Any other points?

If  not,  Dr. Page  will  present  the  summary for  Working Group 8, which
dealt with  toxicology ana methodoloy.

Dr. Page (EPA):  Thank you, Dr.  0'Conor.  My report will be  a  little  bit
shorter  than what I  anticipated  since many, of the areas that  have oeen
described already  in the Epidemiology Working Group were identified also
Dy the Toxicology  Working Group  as high  priority areas.  However, I will
go back  into those  shortly to  explain  where  we feel  maybe the animal
toxicology  effort  can correlate  well  with the  epidemiological  efforts.

As an overview, however, it was  felt  that this  collaborative program  has
gotten off  to  a pretty  good start.   The  impression  I received,  however,
is that  it can even  be  better focused  towards  areas that  may be truly
important to the  regulatory agencies, and t,o the research agencies.   The
term  technology transfer has  been  used quite  a bit.   I  think the  EPA
sees the role  of  research  organizations as helping to  identify fruitful
research  findings to  be  translated into  aspects of  applied  research
which the regulatory agencies can use to go about their business.

It  is my   observation  that there  may  be  a  need  for  a  little closer
coordination  perhaps  between  the   agencies  and  between   the  project
officers within the  various  programs.    In some cases,  I  think it  has
oeen  very  close  and  in  some cases  it  may not  be  quite as close as we
might desire.

Again, I am going to give  some of my own personal  observations,  since I
                                 837

-------
have not  been  that involved with the  program.   But it also seems to  me
that we  are  talking about  a  program which  is  rather  small compared  to
the total  biological  program  in cancer research.   It occurs to me  that
there  is  a great  amount of research going  on  in the same areas that  we
are doing just a little bit of in this program.   I  think  it  is essential
that we be aware of all the research under way.   I  am  not saying that  we
are not,  but  one  of the difficulties  in  presenting a program, at  least
as  I saw  it,  is that  we are talking about small  pieces of  research  in a
particular area.

In  some  cases,  the other related research  is  not brought in so that  we
know the  scope of this  activity in relation  to  a  particular  area.

Another  observation  is that we  are a small  closed group.   I  think  it
would  benefit the future meetings to at least invite observers from  some
of  the other  program  areas  and  other agencies, perhaps even some of the
people in the contracts  that  are  doing  this  work.   This  has probably
already  been  thought  of, so I  am sure that I  am not  providing any new
thoughts  for you,  but  there is  a fair amount of  research under way  of a
more  applied  nature  by the Army,   some  under  the National Toxicology
Program  and the  Food  and Drug Administration.   I  would  recommend that,
as  you advance in this program and at future  meetings  of this nature,
you  bring in  some  of  these observers from the  other organizaitons.  I
think  that would  help  to  cement  coordination  with the  other research
programs.

As  far as the  identification of research  areas  in toxicology, there  is a
great  overlap with those that have been presented.   It seemed to me  that
the  number one concern expressed by  at  least  those  having regulatory
responsibility, and  I  am including  NIOSH  in this  category because  much
of  the work  that  NIOSH does is  of a regulatory  nature even though  they
are not  a regulatory agency, was the issue of assessment  of  carcinogenic
hazard or carcinogenic risk.   How  do  you quantitate  potency?  This was
asked.  Here was  an  area  where we often do not talk  the  same language
when   we  talk  about  potency.    Some  talk  about  strictly  the chemical
reactivity, the  biological activity at the  molecular level, and others
are  talking   about  the  summation  of  the  total  events  that take  place
which  include the  metabolism,  the  kinetics of  the  transport  and the
disposition of the chemical.

It  was   identified  as  a   crucial   area  for  the  regulatory  agencies.
However,  it is one that the research agencies  like NCI can  contribute  to
in  a  very meaningful  way.  So,  it  is  felt that  there is a  crucial  need
to  improve our methods  of  carcinogen risk  assessment.

Another  area  that was  identified  as  high priority was that  of research
on  methods to  reduce  the  carcinogenic  hazard  to  populations  that are
exposed  to "known" carcinogens, to chemicals which have  been  identified
by   an  animal  test   or   even   in  vitro  tests   as being   potential
carcinogens.   There  are a  number of suggestions  representing monitoring
                                  838

-------
aspects  including  diagnostic tests.  We  heard  of some immunodiagnostic
tests  that may  be  a  fruitful  area  for research  in  a collaborative
effort.   The methods  of reducing  hazards  and  control  technology were
identified as high priority for this collaborative effort.

Another  major  area which the  Toxicology  Group  felt  was  very important
was the  further  improvement and development of  the  shorter term tests,
the in  vitro  screening tests.   I  am  singling  out the in vitro tests at
this  time  even though  there was  interest  in improving  the routine in
vivo bioassay systems.  But inasmuch as this seems to be  now more in the
purview of the National  Toxicology Program  or  other efforts at the NCI,
we did  not  deal  much with  the  in  vivo area.  However,  in  the in vitro
systems,  the  emphasis  was  placed  on  the cell  transformation systems,
ueing as they represent a unique system for bridging the  gap between the
bacterial systems  and then the  longer term in vivo tests  in a battery or
a  tier  scheme.    There  was  considerable  interest  in  seeing  further
research  on  the  validation or  improvement  of  the  cell  transformation
systems.

However, along the line of  the in  vitro tests,  I think it also came out
that  we  really  need to take  a  close  look at  all the in vitro tests and
prove  the  predictability of  these tests.   We heard  at this meeting of
the  variability  in  the  S-9 fraction  and  how  this can  influence  the
results  from some of the  bacterial  tests.    I  think  for  a regulatory
agency  to  utilized  these   test  systems,  we   have  got  to  have  some
assurance  that they are predictable and  that  there  are   laboratories
whicn  can  perform these  tests  in  a reliable manner  and, of course, we
want  to  know  that the  results  are of relevance  to the human situation.
I will put it that way.   We  are  always on line to justify actions in the
regulatory agencies.   Therefore, test systems  have to be first utilized
as a  basis of  predicting the human effects.   In vitro  systems will play
a major  role, certainly for  the  EPA,  in an assessment scheme.  They will
nave  a role in the  tier scheme or a  cattery  approach  to assessment of
potential carcinogens.

A fourth area  that was  identified as  important  and perhaps underfunded
was that of  assessing effects  of  mixtures  or  additive effects.   A good
example  was presented also  at this meeting  on the co-carcinogen effects
of  disulfuram  and  ethylene  dibromide.    That  was  just   one  example
presented,  but   I  think  there   are  probably  many  that  could  be
mentioned.  In the real work  situation, exposure is  usually  to a mixture
of chemicals  and   not  to  a  single  pure  chemical.    So  I  think there is
certainly a priority that should be given to research  on the effects of
multiple exposures rather than  pure exposure.

There  was  a  little thought  given  to  the  need for further research  into
the structure/activity  relationships.   It was not discussed  to any great
extent,  except I believe  that  Dr.  Fraumeni has already  mentioned this as
a high priority area  also for  epidemiological work.
                                   839

-------
As you  see,  the main research  areas  identified in the Toxicology Group
are  very similar  to those  which have  already been  identified  in the
Epidemiological Work Group.

Dr. 0'Conor  (NCI):   I would only make one comment  to supplement what Dr.
Page  has said  about  our meeting,  since we have  already exchanged our
ideas  and  agreed  on  what  we  thought was  important  at  the   working
session,  and that relates  to  the in vitro  tests.  It  was pointed out
that the development and the study of in vitro  methodology  is  one of the
most  important  tools  that  we  have  in   studying  the  mechanisms   of
transformation  and  the  mechanisms  of  carcinogenesis  generally.   I think
a  major  part of our time was  devoted to the whole question of in vitro
tests and their use  and  the different purposes that they serve,  and Dr.
Page  mentioned  their  use   in  screening  and   tier  systems.    So  this
particular  area and methodology  automatically  becomes a bridge  between
research  of  interest to  the fundamental  basic researchers  and  to the
regulatory  agencies.   It  is  really  a  natural and  convenient   meeting
ground.

Are there any  specific questions or comments in relation to  this  general
area, which  is  a very broad one?

Dr.  Saffiotti  (NCI):  A point  that was somewhat referred to  during the
discussion of  the Toxicology Subgroup might be  useful  to discuss  here  as
a  plenary group.  I  am referring to the  growing applicability  of  many  of
these new methods of research,  especially the in vitro methods and those
applying  in  vitro techniques  to human tissues  and metabolic studies,  to
problems  that  have   a   counterpart   in   epidemiology  and  clinical
medicine.  So  one can actually  introduce laboratory methods  into  a study
that  has a  basis in epidemiology  in  terms  of  selecting populations for
study;   in  terms of correlating  levels,  which   was  mentioned   in  the
Epidemiology Report; in terms of determinations of levels of interaction
or metabolites  in human tissues and body fluids.   Take these back to the
laboratory  site  and utilize  dynamically  the  give  and  take of human
observations  and  laboratory observations,  in a sense,  in a more  tightly
planned  fashion than we  have  perhaps  done  so  far.    I  guess  that this
type  of interaction   of  the  various  agencies  with  their   various
backgrounds  can  be very helpful and stimulating. So, I would like to see
a  greater   cooperation   between   laboratories  and  the   clinical  and
epidemiological  approaches.

Dr. O1Conor  (NCI):   Thank you.  Are there other comments?

Dr. Orme  (NCI):   I am sorry that  I missed the session  yesterday,  because
I  had  a question that I  wanted to pose.  If  I had  been there,  I would
have  advocated the  further  development  of in vivo methodology.   I think
we have  tremendous obstacles to overcome before we have  adequate  in vivo
methodology  for  identifying  carcinogens.

One  of   the  facts which  is  generally  ignored  is the  variation  we see
                                   840

-------
experimentally  in  response to exposure  to  carcinogens.  Drs.  Griesemer
and Cueto  have recently  put  together a summary  paper on 192  bioassays
that were  conducted  by the bioassay  program.   Of these 192 tests  which
were conducted at the  maximum tolerated  dose,  there  were  35 cases  of
strong   carcinogenicity  detected  in   one  species  and  no  evidence
whatsoever  of carcinogenicity in  the other  species.    There  were  also
another  ten tests  where  sufficient evidence  for  carcinogenicity in  one
species  was  obtained,  but  no evidence  in  the  other  species.    I  am
talking mainly about the  mouse and the rat  in our bioassays.

Not only do we not know  at  this  time how to extrapolate from  rodent  to
man, we  also do not know how to extrapolate from  mouse to  rat or  vice
versa.  Even within a species, in reference to one  of  the phenomena  that
I was discussing this  morning, we find that certain strains of mice are
almost totally  resistant  to UV irradiation  as  far  as carcinogenesis  is
concerned and others get  tumors in high multiplicity within 24  weeks.

I think the significance  of this, as  far as various mathematical models
that are posed for assessing  human risk, should be  investigated.  We may
actually require  an  additional experimentation step which  looks at  the
strain  specific difference  or the species-specific differences.    This
could  also  be  used   as  a  tool   which  gets  at   the  mechanism  of
carcinogenesis.  We  should get out of this game of assessing  potential
risk  and  get  into  the  much  more   interesting   game of  identifying
populations  and the size  of populations  at  high risk.

I think  one of the biggest problems we are now faced  with in going  from
a  methodology of  testing based  on  the Fischer  344 rat or  the B6C3F1
hybrid mouse is that  our entire resources program is developed toward
producing  these  animals  specifically  and that we  do not have  resources
to  get  into  multiple  strain  testing at  this time.   It is  a serious
logistical  problem to  produce  the  mice  required   for  mutliple strain
testing.

The second  thing that  I would  like to mention concerns  prioritization  of
work.    This  EPA/NCI  agreement  is   the  last  resort for  people  in
experimental  photocarcinogenesis.  The Department of Transportation,  the
EPA Bacer Program, and  NCI  to  a large extent, have  all  pulled out of the
photocarcinogenesis area.  We were  very fortunate that Dr. Kraybill  and
the other  members  of  the organization of  this  agreement  allowed us  to
support the  residual activities in photocarcinogenesis.

I  think  it  is  a  very  important  area  There are  400,000  cases of  skin
cancer diagnosed each year.   They are easy  to treat, fortunately, but it
is  still   an  amazing   number   of   cancers   to   receive  such  little
attention.   There are  several experimental  features  about  skin cancer
which  I  think  could  be  pursued quite  easily.   It is one of the  few
cancers that  is easy to follow, as far as progression  is concerned.   The
mechanism,  the time-to-apearance  of tumor,  and the  dosing with  UV can be
measured quantitatively with ease.   I think it is  a very good  system to
                                  84J

-------
pursue.   I  am very  happy that we  have received  some  support from  the
NCI/EPA  Program  for  it.   If it  is  diminished,  I  am  afraid that  there
will be  very little photocarcinogenesis  research  conducted anywhere  in
the country.

Dr. O'Conor (NCI):  That  is  a slant that we were not fully  aware  of,  but
I think  it is an  important one.

Dr.  Kraybill  (NCI):    I  would like  to  ask Dr. O'Conor  and perhaps  Dr.
Domanski  a  question.    I  quite  agree with Dr. Orme on the  first  point.
We   need  to   know   more  about  strain  sensitivity  and   interspecies
sensitivity.   But will   that  type  of  effort not  be  encompassed to  a
degree  under  your  interspecies   comparison  program?   Am  I correct  in
assuming this?

Dr.  O'Conor  (NCI):     Dr.  Kraybill  is   referring to   what   might   be
considered a  new  programmatic  approach  to this general problem which  we
are  taking  at  NCI  in  cooperation   and  under the  encouragement  of  the
EPA.   Some  of the  officials at  EPA were  strongly  encouraging  us  to
conduct    more   research   related    to    extrapolation    involving
pharmacokinetics  and  biochemical  parameters  in  different  species   of
animals  so  that, hopefully,  this  could  lead to  better  development  of
regulatory guides.

Under  this  stimulus,  we  considered  various  approaches.   A  month  or  two
ago, we  had  a workshop on this subject.   It was an  interesting day-long
meeting,  but  it  certainly was not one where  we   arrived  at  a  general
consensus.   Everybody  recognized that  this  was  an extremely  important
but  extremely difficult  area.    As we  mentioned   in  our working  group
yesterday,  we really  do  not have  enough  information  at  hand.    So  the
objective is  to encourage more research and get additional  information.

So we  are going to issue  some RFA's  within the next  few weeks  which will
cover  various aspects  of  the approach  to  interspecies comparisons.   We
will see that  all of the  participating  agencies here get  copies of these
RFA's.   It  is a  modest  start  with  some commitment of funds,  but  if  the
response  is  good from  the scientific  community,  and  the  types   of
proposals  seem to be  directed  to  this area  of  research, then we would
hope to  expand it.  There is  so much work  to  be done in this area that I
do   not  think  that  what  we  are   doing  precludes  the  collaborative
interagency  program  from also getting  into  this  field.   As has  been
mentioned  earlier today  by  Dr.  Bridbord,  I think  there  is room  for
interagency  discussion of the on-going work  in  each agency which might
be  related  to the program, whether it is  supported  under the  program or
not.

Dr.  Weisburger (NCI):    I have  a  qualm  about some of  the work  on  the
benzidine-based  dyes,  where I think  we  need  another  approach   to  the
metabolism  of these  compounds.   All the work  has  achieved  is  to  look at
the  metabolic products of the chemicals as if those are  responsible  for
                                      842

-------
all of  the  activities seen in animals.   Nobody seems to be thinking of
the fact that the whole molecule, even though it is very complicated and
could  be  the  focal  point  of  a  tough  problem   indeed,  should  be
considered.

Of course,  when  one  looks  at  the result  in  animals, benzidine itself,
which is the  split  product,  is  not  very effective in rats, much less so
than the benzidine-based  dyes.   I think  a new approach should be taken
to this problem.

Dr. Galbraith (tPi\):  EPA is sponsoring an aromatic amine program at the
National Center  for  Toxicology  Research.   We  currently  have underway
three  carcinogenicity  bioassays  and  a  fourth  is  scheduled  to  be
initiated.   This program  was  initiated several  years ago as a result of
concerns regarding  benzidine  and  aromatic amines in  water.   We are very
interested  in continuing  with the program and  expanding  it in order to
learn  more  about   structure/activity  relationships.    We  are in  the
process  of determining  the  program  direction.   We will   welcome  any
suggestions in this  regard.   I  am sorry that Dr. Hart, the new Director
of NCTk, was  not able to  be  here,  however,   I  believe  that he would be
happy to entertain  any suggestions or  discuss this program with you.

Dr.  Jammer INIOSH):   I  would  like to  strongly support the  work  that
needs to De done in  the area of structure/activity relationships.  It is
apparent   from   even  a   casual   reading  of  the  relevant  scientific
literature  that  only  a   very  few  of  some  highly  sophisticated  and
computerized  techniques  identifying  structure/activity  relationships
are  used  in the  area of  mutagenesis  or carcinogenesis.  These methods,
wnich are  based  on  N dimensional  vector spaces  and based on information
tneory,  have  enormous numbers of variables  simultaneously,  such as you
are faced  with in  species extrapolation.   The power  is there.  They are
used  primarily   by  drug  companies  to  synthesize  new molecules and new
moities.

The  second use for  such  studies, beyond  the  point  of trying to figure
out which  neighboring congeners will be active  in biological systems, is
their   usefulness   in  pointing   out   which   compounds  will  be  useful
substitutes  for  carcinogens  in  terms  of  their physical  properties.
Identification  of  substitutes  could  be  made  much   more  rapidly  using
structure/activity  relationships  than  the  trial  and  error methods.

Dr. u1Conor (NCI):   Thank you.

A  question was  submitted  yesterday to our  group.    It was  of a fairly
general  nature  and  we thought it was  probably more  appropriate for the
plenary session.    I  think, we were  right, because  the question in fact
covers  a  number  of  the areas that  have already been discussed  in terms
of   epidemiology  recommendations  and  the  Toxicology  Working  Group
recommendations,  in that   it  refers  not   to  just  interspecies  but
intraspecies, particularly  as  it  relates to human beings.   It  is really
                                  843

-------
more  of  a  comment  than  a question  and  it might  serve as  a  point  of
Discussion at this point  in the meeting.

The time is overdue to consider the  significance of  individual  variation
in  response  to  carcinogens.    Current  methodology favors  the yes/no
approach to  carcinogenicity.    Not  only  is  this  the wrong methodology,
but  it  implies that  we  are asking  the wrong  question.  The individual
who  is  posing the question  would like to  suggest that the  appropriate
question  in  carcinogenicity  testing  is  what  percentage  of the  human
population  is in  a   high  risk  category  with  respect  to  exposure  to  a
chemical  and  why?  He  goes on to  say that to  answer this  question  we
must  develop  resources not  currently available.

I  think one  could  not   argue  with  the  need  and  the desirability  of
Knowing  what  percent of  the  human  population  is,  and who these people
are,  wno  are  at a high risk  to exposure  to a chemical  and why they  are
at  a risk.   This really comes back  to  the  whole problem of the etiology
and  prevention of cancer.

Would anybody like to comment on  that statement or  proposition?

Dr.  Fraumeni   (NCI):   Well,  this is  a very challenging  issue.   Case-
control  studies  of various  cancers  should collect  detailed  information
not  only to  identify environmental  hazards,  but  also to help  clarify
interactions  oetween  multiple risk  factors, including  cigarette smoking
and    occupational   exposures,     environmental     agents    and    host
susceptibility,  and  so forth.

For   a  more  complete,   comprehensive  and  precise  evaluation  of  the
interplay  between these  risks  factors,  we are  relying  increasingly  on
the   experimentalist  to  come up  with  markers  of  susceptibility.    The
epidemiologist  is  waiting for these  new  developments  to  emerge from  the
 laboratory  that  will permit us  to  characterize  high-risk  states  with
greater  precision.

Dr.   Bridbord  (NIOSH):     One   other  point  to   note,  again   from  an
epidemiologic perspective,  is  that  in  the existing cohort  mortality
studies  most  of those populations  represent  a diversity of  length  of
exposure and  latency with the  vast majority  of  people  not  having been
followed for  a  long  enough period of time  to  really have a true idea of
what the  quantitative  impact  might  be  on  that   population.   We  also
probably miss some  effects that may otherwise be  on the margins.   An
 additional  recommendation might  be  the need to selectively  go back  and
update  previous conort  mortality  studies perhaps  every five or  ten
years;  to  at  least  expend a certain  amount of  effort to do that, so that
eventually  we follow enough  cohorts  pretty much  to  extinction  that we
 really do  understand  what that  lifetime impact might be.

ur.   O'Conor  INClj:   I  think it  is  good  that we  got  that  statement and
those comments into  the  record.
                                  844

-------
Dr. Orme  (NCI):    I  would  like to make a  very  specific suggestion with
respect to  the  resource problem that we  are  facing.   Everytime we have
attempted to do a multi-strain test for carcinogenicity, we have come up
with  dramatically  different results  in  the  various  strains.   I  agree
neartily with the epidemiologists.  The epidemiologists are way ahead of
the experimentalists as far as identifying high risk populations.  I had
the honor  of organizing  the  bioassay of  8-methoxy psoralen,  which  is
used  in  the  treatment of  psoriasis.   There had  been  an epidemiology
study  done  on  the patients who  had  received this.   It was quite clear
from these studies that people who had already received x-ray treatment,
poeple who  were  in fair skin categories,  people  who  had previously had
an excised  skin  cancer were  in  a much higher  risk category than other
populations.

The specific proposal  that  I would  like to make is  that  the NCI/EPA
group consider funding  a feasibility trial for introduction of the mouse
embryo freezing technique  as  a  potential  way to obtain large numbers of
different  mice  simultaneously for  multistrain  testing  approaches  to
assessment  of  risk.    As  I said,  this would solve the  one big problem
that  we  are faced  with now;  that  is,  getting enough  of  the different
animals  together  at  one  time  to   do   a  coordinated  bioassay  using
different biological material.

Dr. KrayDill  (NCI):    I would  like to ask the  experimentalists who are
here,   Dr.   Saffiotti,   Dr.  Weisburger and the  rest   of  you,  is  there
association,  for  example,  between aryl   hydroxylases   as  a biochemical
marker,  and  the proclivity  to develop a  carcinoma?   Does this enzyme
system  come into  play?   If  it  does, could  one   look  for  this  kind  of
parameter  and  then  say that these kinds  of people may be at less risk;
tney   can   smoke  cigarettes  and   not    come  down   as   readily  with
oronchiogenic carcinoma?

Dr. O'Conor  (NCI):   I  think you  have oversimplified the problem.

Dr. Weisourger  (NCI):   Well,  the study from Texas which was reported in
trie  Dook  by  Griffin  and  Shaw,  "Carcinogens:    Identification  and
Mechanisms of Action"  seemed to indicate that there was not any clearcut
correlation  between  these  two factors.   Others  may  show  a different
aspect, out  that is what I have heard so far.

Dr. O'Conor (NCI):    I  think the  question  is  whether they  come into
play.   The answer  is yes.

Dr. Saffiotti  (NCI):   Yes, Dr.  Kraybill  has put the question in sort of
photographic  form  oy  picking  an  example  of  one particular  enzyme.
Obviously,  the biological  responses  are  controlled  by a  much broader
spectrum of  systems, including a battery of enzymes.
                                 845

-------
The  point  is  an  important  one.   I  think that  with the  progress  of
research in these  areas,  we will gradually learn  to recognize at  least
certain high  risk  groups  or categories or  conditions,  and then we  will
have  to  learn  to  see how  permanent  they  are,  whether  they  vary  with
environmental  factors,  with  time,  with  age,  and  so on.   I  mean  the
approach is  right; I do  not think we  are  yet at  the  time when we  can
pick  one  of  these  as  an  index  of  susceptibility.   The  point that  Dr.
Fraumem was  making earlier  is a  point  that  we  also  share very much,
that  is,  the  epidemiologists  are  waiting for the  laboratory  people  to
come up with markers that could  be  used.    I think  that types  of markers
ana  methods  for studying them,  and so  on,  are in  fact,  coming out.   I
indicated this  yesterday  in the other  group.   I  am optimistic that  in
this new decade this whole area  will really come to  some  fruition.

Dr.  G'Conor (NCI):   dell,  I  think that now might be  a good time to  turn
to  the  third  working group, which  is  really  the  subject of the plenary
session and which  relates  to tne importance of interagency programs.   I
think  we  all   agree  on  that.   Fundamentally,  we  are interested  in  the
development of  the future.   L)r. Marland will give the  report and  open
the  discussion.

Ur.  Marland  (EPA;:  Thank  you.  Dr. O'Conor.   Let  me first explain  the
absence  of  my  colleague.    He  was called  to Buffalo,  New York  as  an
expert witness.

It  is  probably  fair  to  say that there was generated  yesterday  afternoon
in  workshop  C nothing new,  nothing exciting,  nothing  that is probably
not  pretty  clear to all of  us.   Therefore, let me indicate that it  is
likely  that  the comments  that I would make would  be more for emphasis
than for any startling revelation that may  be there.

In  the  first  place,  it  was clear that the collaboration  between NCI  and
NlUSh  is a different form of collaboration  than between NCI  and the  EPA,
even tnough the  program  is somewhat similar in terms of  its description
under  the  statutory and other developmental  and  programmatic  language.
That  can  become  important  and  perhaps  I will get back to that  a  bit
later.  But I think that the fundamental  difference  pervades some of the
comments tnat were made.

One  of  the   earlier   comments  which   I  think   showed   an   important
recognition  of the  value  of  collaboration came  from one  of  the  NIOSH
representatives,   in  which   he  expressed   hope  that  the   nature   of
cooperation betwen the  NCI and  NIOSh  in  undertaking some basic studies
had  provided  some insight  into some type  of intervention  process  to
prevent  a  cancerous  development within  an occupational  enviroment  and
tnat some of the more applied  studies could also be continued  on a  joint
oasis.    This  indeed  stimulated   some  discussion  and   I  think  it  is
indicative  of  the high  value  placed  by  NIOSH  on  the  involvement  of
personnel from  the NCI  and from other  kinds  of  researchers involved  in
assisting  in  the  application  of some  of  the  results  of   the   more
fundamental studies.
                                 846

-------
There was  a continual  commentary that the conversations, communications
ana  exchanges  of  information  among  the  agencies  involved,  i.e., NCI,
tPA, NIUSH  and  OSHA are indeed  too  sparse.   There is a quick and broad
recognition among  all  of us  that we need  to collaborate better and that
this one episode that we are celebrating here with  a workshop is perhaps
something which should be repeated, extended, or at least enlarged upon.

I think there is no question that the kind of collaboration that EPA and
NCI  have  had  has  been  a  source  of  major  benefit  and  has  prevented
probably  some  unfortunate  glitches.    Most  important,  however,  it has
brought to  bear  the kind of input,  the kind of collaboration,  the kind
of communication  and  conversation which has  benefited  EPA and,  I hope,
it has also benefited the NCI.

Now, where  are  some specific areas, in addition to the  one to which the
NIUSH representative  referred?   What  are  some other  ones?   One which
attracted rather significant discussion was  advice  as to where one would
proceed to  deal with  chemicals as  potential  carcinogens.   Do we as a
regulatory  agency  or  as  a  nation  or  as  government  intend to  use our
resources   and  devote  our  research  to   the  existing "baddies",  the
arsenic,  the  benzenes and  the  dyes,  or  do  we  concentrate on  those
substances  which may be entering the marketplace for which use patterns,
distribution, manufacturing  indeed  have not been established, but which
are  indicated as being on the threshold of being manufactured?

This is an  important question.   It  is a question which  EPA has to  answer
every  day.   Tnat  in  itself  is enough  to make  it  important  to EPA.
However,  I  think  it has  an important  impact  on  the  nature of research
activities  that could  help EPA  solve  its problems.

For  instance, quite clearly, if one is going to address a major part of
his  work  control   activity  to  the new  chemical prior  to its entry into
the  marketplace,  the dependency on  18 month rat tests  as a minimum kind
of  classification  for risk is  not  possible.    You  then  get  to  the
structure/activity tree  as  a model  of decision-making, as to whether or
not  that  particular  compound,   as  yet a  secret  material  in  which the
structure  is  known and very little  else,  should  be further tested.  By
virtue  of  its   structure,  EPA could  indeed  require  a • substantial
testing.    But  in  the absence  of  structural  kinds  of  indications  of
toxicity,  such  expensive  testing  on  the part  of industry may  riot  be
warranted.

I  think that  there was  a  fairly  strong sense  coming from  the group
yesterday  that reliable  structure/activity  prediction models  are not
presently  sufficiently reliable for us to win court cases.  Research is
needed to  improve  the  kind of quality  data that one gets from, the  use of
models.   So there is  a  specific  suggestion  which does represent some
degree of consensus among some  of the participants.
                                 847

-------
Another one, one which I have a personal feeling  about  and  apparently so
did the participants in Dr. Bridbord's  and Dr.  Page's groups,  deals  with
the ability to better quantify  risk.   Risk  assessment by a  great  many
names  is  a  terribly  hazardous  undertaking.    Unfortunately,  it is  the
assignment  of  a quantified  risk  that  puts  EPA into court  and it  is  a
feature of  our regulation which  is mandated.  It  is  necessary.  We  do
not have  the luxury of saying,  well, our scientists assure  us that  this
compound  is bad  under almost all  circumstances,  and then being  able  to
regulate  on the  basis  of what  the  consensus  among  scientists would
indicate.   It  is  necessary  that  we quantify.   It  is   necessary  that  we
assign  a risk  factor.    None  of  us,  apparently  in  any  of  the three
groups,  feels  satisfied  that the  state of  the art  here  is where  it
should be.

Parenthetically,  I would  like  to  say  something  for at  least a few  of
us.   I  think that  Dr. Kraybill  and Dr. Saffiotti at least were  present
at  a  recent meeting of the Environmental Carcinogenesis  Subcommittee  of
the  National  Cancer Advisory Board in  which  this  issue was  discussed
loudly, long,  and  for  days.   The  consensus there  was very clear that the
state  of  the art  is  in  rather  poor shape.   The  issue  under  debate was
not whether more  research was  needed, but  whether  or  not the  National
Cancer Institute should come  out  with  a  rather  strong statement pointing
out the weakness of this.

However,  that  paper  has  now been  converted into  a description of the
nature of research which  would  be desirable, which  is  to strengthen the
science base on  which  risk assessment is made.   I  believe  that at  least
in  EPA,  and judging  from the  other groups  also in NCI and  NIOSH,  we
should begin to  pay serious  attention  to  recommendati-ons such  as that.
Even  though it  is not  an offical  report  of  the  Cancer  Board, it  is
certainly  an  extremely  well  thought  out document  which does  describe
research  needs  in the   field   of risk  assessment.     I   think   that
constitutes another fairly sincere  recommendation from  our  group.

The next  point is  probably the  most important  one  and  yet  it  is  the one
which  is  most difficult  to  describe.   I  do not know  how anyone would
approach  a research  project  on  it.    It  can  be  described  as   the
institutional  problems   inherent  in   controlling   carcinogens  in   the
environment.   The  institutional  problems are  those which  do not  deal
with  the  development  of  scientific data.   They do not  even  deal  with
risk  assessment,  which  I  allege^  is  not science,  risk assessment  is
policy -- but  that is  a debate.

When  you  have  a   fairly  clearly   established  relationship  between  a
chemical  compound  or  an  environmental  condition  and cancer  and you  find
that  society is  still unwilling  to make any changes so  as  to deal  with
the threat, then you  find a  very strong sense of  discouragement on the
part  of  the regulator.   I am  sure the scientist  feels  that  he  is not
being  heard or appreciated.
                                  848

-------
The institutional  problems  brought about by the  courts,  the law suits,
the interminable  delays in getting commitments from the affected parts
of  society  to change  the environment,  and  prevent the  access  of this
compound to  people are problems that   need  to be addressed.  They need
to be attended to.   I  do not know how one goes about it.  But if anyone
needs to  be told  further  that smoking  is  related to cancer,  I  do not
know  how  it  could  be said.   Why  is  it  that  we still  have  a  very
significant population  of  smokers?  The same  is  true with the exposure
to a great many other  materials.   There is the issue of  saccharin.  Why
is  it that  suddenly not only  the  Congress  but Americans by the tens of
thousands rise  up in  indignation  and  say that we scientists are crazy
because they do  not have  cancer  and  they  have  been  using saccharin?
This is in effect what they are saying.

In  the  PHS,  we call the institutional  problems  of translating science
ana  regulation  into meaningful  action  and  into an  agreeable  thing,
education.   I think it  is  an impossible chore to assign simply to the
word education.    I  think it  has  to  be a more  highly  refined  kind of
problem.  Whether  it can De researched or not, whether it is a function
of collaborative research or not poses  a real  problem.

Tnat is tne  nature of the  comments which  we think might be  substantive
and whicn point  to tne  areas  in which future collaboration could be of
use Doth to NIOSH  and to the EPA participants.

I  would  like  to  add  a  personal   commentary.    I   feel  that  the
representation from EPA should  be stated  and that is that  the  EPA is
registering  a strong degree  of  satisfaction with  the  way in which the
National  Cancer  Institute,  and  particularly Dr.  Kaybill  and his staff
and those whom he  has  called in  to  help  us, the way  in which you are
helping us  identify superior  research  activities  and  projects  to help
meet our  regulatory aims.  He think  the system is working well in that
regard.   If  there is  a way in which  it  can be  improved,  it  would go
along  the lines  that  Dr.  Bridbord pointed  out.   There should  be an
additional monitoring  of the  projects  in place.   In other words, there
should  be a continuity of effort  as  these projects develop, mature and
begin to  produce  reports.    If there  can be an improvement,  it could be
in  that direction.

There roust  be expressions  of sympathy  when  the  regulator is in court.
he  nas  been sues.   He does  not  have  the luxury of telling the judge,
"well,  judge,  just cool it for  awhile; we are going to  turn loose some
people »rho  are going to get these numbers and we  are going  to come here
in  about  two years  and  we  are going to impress  the devil of out you."
That  does not work when the  suit  is  being brought by a  group who feels
as  though the>  have been harmed because they have right to  due process
and  they  are \eepin9  EPA or  QSHA's feet  to the  fire.   Therefore, the
Kind of uata which we  need  are those that are not  currently available.
I continue to point  out  that  the twain  probably will never meet.
                                   848

-------
There  is  the   regulator/scientist  who  says  that  these  data  are  of
sufficient value  and  validity to recommend  to  our lawyers that  they  go
to  court  with  them.   The more  traditional  and,   if you  will, the  more
appropriate view  of  the researcher is one  who  says, "These data may  be
all  right  for  what  they  represent,  but  they  do   not  represent   a
sufficient picture  to  oe  absolutely   confident  of what  we  are  doing."
Now  these  two  will   never  meet.    But   I  think  that  through  the
collaborative  work  that  we  have here we  can  at least  bring about  an
appreciation of each  other's  problems, so that we tend not  to embarrass
each other.   If we can  do that, we have come  a  long  way.  I  hope  that
can serve as possibly a  challenge to us for  the future.

Dr.  O'Conor  (NCI):    Thank you,  Dr. .Marland,  for those  very  thoughtful
remarks  which  reflect  the  deliberations of  the  committee.    It almost
leaves  little   to say.    Maybe  that  is  a good  time to close.   Dr.
Krayoill, would you like to make some  remarks?

Dr. Kraybi'll  (NCI):   In all  humility, I  appreciate the  fine  remarks  by
Dr. O'Conor and Dr. Marland.  However, I would  like  to  deflect attention
away from  myself.   If you would look  in  this program,  you will  see the
people who organized  this  meeting  and  this program.  Great  thanks go  to
Dr.  Bridbord   ana  Dr.  Leidel, who  played  a big  role  in getting   this
agenda  together.   Thanks  also  go  to  Drs.  Cameron,  Burton, Lee, Morris
and  Galbraith  and the  advisory  levels with the NCI/NIOSH program,   Drs.
baffiotti, Weisburger,  Cooper,   Fraumeni, and Ms.  Blackwood.   These are
the people who made  this program possible.  Above all  of this,  if we  do
not  have the   support  of Dr.  O'Conor, and  Dr.  DeVita,  and Mr.  Jellinek
and  Dr.  Gage  and  Dr.  Kobbins, we are  ineffective.   I was listening  very
closely  to  what Dr.  DeVita  said the   other day  and  I  gather  that he  is
supportive of  this sort  of effort.  We understand  that  Dr.  Richmond, the
Surgeon  General,  is very keen on collaborative  program  work.

I  would  like to make  another  comment that  is  not  quite  as complimentary,
uecause  it comes  deep from the   heart  and I  am  really concerned.  I  hope
Dr.  Galoraith  and Dr.  Lee support  me.  If they do not,  well  get up and
disagree with  me.  Dr.  Marland touched on  it.

This  is  a collaborative  program.   We are  not yet  fully collaborating;
oelieve  me.    When  we organized the  NCI/EPA projects,  we had a lot  of
coordination and  we  had collaboration.  What does collaboration  mean  to
me?   One day  I asked Dr.  Mason, "You have  that  other  chap from EPA  on
that  project  with you.  Are  you really collaborating?   Are you  talking
to him?"  Dr.  Mason  said to me,  "Oh, yes,   I  know what is  going  on."
Now,  in  some of our projects  this  is happening,  I  can point  out one  as a
shining  example.   I  think the project of body  burden of  chemicals  Cindy
Stroup  directs reflects the  interplay of  many people  and  I believe the
interaction  here  is   excellent.   But  for some  of the projects  that
prevail,  the  direction  and supervision  is  not  collaborative.    Tom  Orme
mentioned  that this  was  a learning curve  for  him in  his  connection  or
collaboration  with Dr.  Wiser.
                                  850

-------
You have  to be really  interested  in wanting to  work  together.   You can
drive a horse to water, but you cannot necessarily make him drink.

As I mentioned  in  the  overview statement the other day, and  I guess  I  am
referring to  NIOSH here,  and Dr. Marland you  touched  on it too; I would
certainly like  to  see  our NCI/NIOSH program represent more in the spirit
of true collaooration.  By that, I mean  that half of our project officers
be NCI  ana  the other half be  people from NIOSH.   When we give a report,
we should show  that it is mutually  shared  and  directed and we are truly
cooperative,

I Know that we  have made  some  great  strides, Dr.  Bridbord, but I think  we
ought to go a  little further in that direction to show a little more NCI
participation into the NIOSH program.

Dr. O'Conor (N~ ):  Does  anybody want to challenge Dr. Kraybill?

Dr. Marland  (EPA]:   Dr.  Kraybill  and  I used  to share an  office years
ago.  I learned then not  to challenge him, but you can disagree with him.

It was  Dr.  Page  who made  a very  specific  and  useful  commentary.   Dr.
KraybiH was  so busy  making notes  on  it,  that  he  did not  fully catch
it.  An improved collaborative monitoring would be one of the things that
I noted on  my sheet.  That  is why I did not mention  it.   But I believe
that EPA  at least, represented by  Dr.  Page  and me,  would like to see  it
improved and we would recommend it  to NIOSH for their consideration.  Dr.
Page, did I speak  fairly?

Dr. Page (EPA): I  agree.

Dr. sridbord  (NIOSHJ:   I  think we too would like to see this move toward
much more  true collaboration  than  has  existed in the past.   I  think  we
would all agree that this meeting  and hopefully future meetings will help
achieve that.

Dr. Galoraith  (EPAj:   I  have  a couple  of comments.   Being involved with
several collaoorative efforts  for  the Office of Research and  Development,
I would like  to say tnat  I think this one is operating the best of all  of
them.   We  seem to be  accomplishing a great deal more in a collaborative
fashion,  even  though   tnere  is   a  great  deal   of  improvement  that  is
necessary.

One thing that has not been  said  and I would hate to leave here without
saying  it,   is that  no  one  has  commented  on   how  these  collaborative
agreements  or  mechanisms  come  about.    I think  this  is addressed in the
preface to the  report by  Dr. Kraybill.  This mechanism did come about as
                              851

-------
a result of an initiative by the Office of Management and Budget.   Other
collaborative  agreements  or  mechanisms  are  developed  as  a  result  of
initiatives in the Office of Science and Technology or Congress.

1 seems  to me tnat  every  one of  them  that I know  anything  about is  a
result of  a need.   Mayoe  there are times when we should be anticipating
tnese needs.   Mechanisms  that are already  set  up,  like the one that  we
currently  have in place,  should  be anticipating future needs so that  we
ao  not  have  other  mecnanisms   developed  which  may  be  wasteful   of
resources  and manpower.

in tnat  regard,  I  had a call  last week  from  a  staffer in the Office  of
Science  and  Technology.   He  was  very  interested  in  getting  some EPA
people and some  HEW  people  together  to  set  up a collaborative mechanism
to  work  on  what   he Deli eves  is  an   area  that  requires  additional
collaborative   work   between   the   agencies.      In   this   telephone
conversation,  I  told  nim  about  this  mechanism, the  one we  have been
discussing  during  the  last  three  days,  and I  suggested that  he look
toward this and  other mechanisms  that  are in  place in trying to resolve
some of  the issues concerning  him.

Ur.  u1 Conor  (NCI):    In terms  of  how the program came  about,  as far  as
NCI  is concerned, the program  was more or less urged upon us.  I can say
that when  it was urged upon us, it was not received with the greatest  of
enthusiasm.   I  think that  is the good  thing  about  it.   At  the  upper
 levels you can  lower the  boom, as Dr.  Kraybill  said, but I do not  think
that  works very well.   I  think  what makes  a  program  like  this  work,
particularly  among   scientists,  is  that  there  is  some  benefit  to   be
gained.   You  cannot make people collaborate  unless  they feel  that they
are  getting  some scientific benefit and  enjoyment  and productivity out
of it.   The people  involved  have  recognized  that it is advantageous  to
work together and it  can be fun and useful.

I think  this  has become  a model  program in collaboration.  I anticipate
tnat  it  will  expand.  As Dr.  Saffiotti  pointed out yesterday,  a  lot  of
the  areas that  are  covered  under this  program really  are  relatively
underfunded  in  our   interest  and  in   support  of  basic  fundamental
research.  We  all recognize  the  importance  of that, but there is  also a
big  need  for  applied research.    I  hesitate  to use  the  word "applied"
because  it is really more  than applied  research,  but it  is  a certain
type of  researcn that  is clearly needed.

From  tne  NCI   point  of   view,   I   think   this  program  will   become
increasingly important because we  are in  the process of developing  a new
format   and  structure  to  what  was  formerly  the   Division  of   Cancer
control.   1 think this Division, which Dr. Sloan is representing at this
meeting,  will  be very much  concerned  with the  more  applied aspects  of
tne  projects  and   the  subjects  of  interest  to  this  collaborative
program.   I  look forward to  a lot of  participation from the members  of
that  Division  and  more participation  from  the  members  of  the Division
                                     852

-------
which  I  heaa,  and  hopefully  more  participation  from  the  various
components  of  EPA and  NIOSH.   As has been mentioned,  we will probably
look for ways to bring  in appropriate components of other agencies.

Finally, I can speak with a lot of confidence when I say that  I know Dr.
DeVita  is  very  supportive  of  this  type of  effort.    The  Board  of
Scientific  Counselors  for the  Division  of Cancer Cause  and  Prevention
was extremely  impressed when  Dr. Kraybill presented his  program at the
last meeting.   They aid not realize  that  it  existed  and  they felt that
we has made  a mistake  in not  publicizing more widely the fact that this
interesting, extensive  and collaborative program was ongoing.

Dr. bridbord (NIOSH):    I  would just like  to point out  that there has
been  some   steps  within  HEW   through   the   Committee  to   Coordinate
Environmental and Related Programs.  A subcommittee of that committee is
examining the waste  dump issue.  NIOSH  has  already  interacted directly
with OSHA and EPA to facilitate collaboration on that specific problem.

Dr. Morris  (EPA):   I have  a  comment  to  make.   In our discussion here at
the end and  certainly during  the summary meetings,  we have concentrated
on intergovernmental relations  and  looking at mechanisms  on  how we can
communicate  and  collaborate  better.    I  certainly  am  a  supporter  of
everything that has been said.

One thing  that I  think we have not gotten  into,  and I  would  like to
throw out for  you  to think about in  the future,  is where do we go from
here  in  terms  of  the  products we  produce?   There  have  been  so many
reports in  government where we  compliment ourselves  on what a great job
we  nave  done  then  we   put  this on  the shelf  with  another  volume  of
accomplishments that the Federal  Government  has  achieved.  The reason I
bring this  up  is  because  last  year, as a part  of  the Toxic  Substances
Control activities, we  were making a very conscious  effort to work with
puolic participation  activities, in  particular  with  states and smaller
segments of  states.   The New Jersey group is a good  example.  I think
that  was  mentioned here  the  other  day.   There,  the  Environmental
Protection  Agency  has  been  one  of  the  foremost  leaders  at  the state
level.

i  would  like  to  see   some  of   these  products   that  we  have developed
transmitted  through some  interagency  mechanisms or  intergovernmental
mechanism  to  these  individuals,  so   that  they  can  appreciate  the
importance  of  this interaction.  I  know at  least from the applications
we 'receive  in our  agency that there were a number of activities relating
to epidemiology studies that  are being  done at  the state  level.  So the
kinds of things that you do in this area for NIOSH is important to these
people.  They  are  getting  involved  in short term test methodologies.  I
worry a  little  bit about who is  going to be  doing  them,  what are their
particular  qualifications,  how  they  are going  to assess  the data, and
how they  are going  to  use that data.   That concerns me  as  well as it
concerns all of us.
                                 85S

-------
So, to  make a long  story short, I  think  we ought  to think about  some
intergovernmental mechanism  by which our  products can  get  down to  the
working level in the state and local governments for their use.   Perhaps
that kind of effort can be done through EPA's regional or state  programs
or through the information programs within NCI.

Dr. Sloan (NCI):  Following  up on Dr.  Morris1 comment,  I think  it might
be  helpful  if I  gave  a  brief outline of  what the  Division of Cancer
Control,  as  part of the  new division,  may be  able  to do to supplement
and complement  what you  are doing.  We  are on the  edge. of technology
transfer, taking  the  developments from research and  trying  to  put  them
into  application  as  rapidly  as  possible  in   the  field  of cancer
control.  So whatever comes  out of the research areas  with which you  are
dealing in the field of cancer should be grist for our mill.

We do have two interagency agreements, one with NIOSH  and one with OSHA,
which I  would  like  Dr. Galbraith to know  about.   The one with  NIOSH  is
to support  demonstration  and education  grants on  proper methods  for  the
removal  and  containment   of  asbestos  in  schools.    We  also  have  an
educational  program  through  NIOSH working with the  American College  of
Radiology  to  develop  educational  materials  for  radiologists about
asbestos-related  disease.    Then, through  an interagency agreement  with
OSHA, we  are working  to  support those new  directions grants which  are
concerned  with  teaching   cancer control  measures  to  current   workers
throughout the country.

As  Dr.  DeVita  said  at   the  beginning   of this  meeting,  through   a
reorganization we are  about  to add  to  the Cancer  Control activities  the
work  of the Cancer  Centers   and  their  research programs  and the organ
site programs, which have a  very  heavy interest in the environmental  and
occupational causes  of the  particular cancer with which they deal.   So
we  feel  that in  the future  we will  have  many more reasons to work  very
closely with your interagency  group.

Dr. 0'Conor  (NCI):   Unless  anybody  has  any other pressing comments,  we
have reached a period when we  can entertain  a motion for  adjournment.

Again,  on behalf  of the National  Cancer Institute, I would like  to thank
everyone for their  participation  at this  meeting.   I  would also like  to
thank everyone who contributed so effectively to it.
                                    854

-------
              FIRST NCI/EPA/NIOSH COLLABORATIVE WORKSHOP:
                PROGRESS ON JOINT ENVIRONMENTAL AND
                    OCCUPATIONAL CANCER STUDIES

                              MAY 6-8, 1980

                 SHERATON/POTOMAC, ROCKVILLE, MARYLAND

                         PARTICIPANTS LIST
Dr. Michael Alavanja
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 847
5600 Fishers Lane
Rockville, MD  20857

Charles M. Auer
Chemist
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. Irwin Baumel
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 847
5600 Fishers Lane
Rockville, MD  20857

Dr. Judy Bellin
Environmental Protection Agency
401 M Street, S.W.
TS-794
Washington, D.C.  20460

Dr. Stephan Berardinelli
Industrial Hygience Chemist
Environmental Investigations
  Branch
NIOSH
944 Chestnut Ridge Road
Morgantown, WV  44226

Ms. Ingeborg C. Blackwood
Office of Scientific Coordinator
  for Environmental Cancer, DCCP
National Cancer Institute
Landow Bldg., 3C37
Bethesda, MD  20205
Dr. Aaron Blair
Environmental Epidemiology
  Branch, DCCP
National Cancer Institute
Landow Bldg., 3C07
Bethesda, MD  20205

Dr. William Blot
Environmental Epidemiology
  Branch, DCCP
National Cancer Institute
Landow Building., 3C07
Bethesda, MD  20205

Mr. Mark Boeniger
Industrial Hygienist
Industrial Hygiene Section,
  Industry-Wide Studies Branch
DSHEFS, NIOSH
4676 Columbia Parkway
Cincinnati, OH  45226

Dr. John D. Boice, Jr.
Field Studies and Statistics
National Cancer Institute
Landow Building, Room 3C07
Bethesda, MD  20205

Dr. Kenneth Bridbord
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 847
5600,Fishers Lane
Rockville, MD  20857

Mr. David Brown
Epidemiologist
Biometry Section
Industry-Wide Studies Branch
DSHEFS, NIOSH
4776 Columbia Parkway
Cincinnati, OH  45226
                                          856

-------
PARTICIPANTS LIST
Ms. Edna Brown
OSHA, Room N-3700
200 Constitution Avenue, N.W.
Washington, D.C.  20210

Dr. Richard Bull
Health Effects Research Lab.
Environmental Protection Agency
Cincinnati, OH  45268

Dr. George Burton
Field Studies and Statistics
National Cancer Institute
Landow Bldg., Room 4C03
Bethesda, MD  20205

Daniel Byrd, Ph.D.
Science Advisor
EPA-OPTS-OCC-PRD
490 M Street, S.W.
TS-794
Washington, D.C.  20460

Glyn G. Caldwell, M.D.
Chief, Cancer Branch
CDC
Atlanta, GA  30333

Miss Faye Calhoun
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 847
5600 Fishers Lane
Rockville, MD  20857

Dr. Thomas P- Cameron
Assistant Coordinator for
  Environmental Cancer, DCCP
National Cancer Institute
Landow Bldg., 3C39
Bethesda, MD  20205

Jerry R. Chandler
Scientific Advisor, DCDSD
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 847
5600 Fishers Lane
Rockville, MD  20857
Kenneth Chu
Special Assistant
OSHA
200 Constitution Avenue, N.W.
Washington, D.C.  20210

Margaret Chu, Ph.D.
Chemist
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. John Cooper
Director for Extramural Activities
National Cancer Institute
Division of Cancer Cause and
  Prevention
Landow Bldg., 3C41
Bethesda, MD  20205

Mr. Roger Cortesi
Environmental Protection Agency
401 M Street, S.W.
RD-783
Washington, D.C.  20460

Dr. John Couch
Environmental Protection Agency
Gulf Breeze Environmental Research
  Laboratory
Gulf Breeze, FL  32561

Elbert L. Dage
Environmental Scientist
Environmental Protection Agency
Office of Testing and Evaluation
Assessment Division - TS-792
Washington, D.C.  20460

Thaddeus J. Domanski
Chief, Chemical and Physical
  Carcinogenesis Branch
National Cancer Institute
Bethesda, MD  20205
                                          85G

-------
PARTICIPANTS LIST
Dr. Carol Edwards
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Mr. John Fajen
Industrial Hygienist
Industry-Wide Studies Branch
DSHEFS, NIOSH
4676 Columbia Parkway
Cincinnati, OH  45226

Jerry Feinstein
Regulatory Analyst
Environmental Protection Agency
6826 Dean Drive
McLean, VA  22101

Mr. Roy Fleming
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 847
5600 Fishers Lane
Rockville, MD  20857

Barrett N. Fountos
Epidemiologist
OSHA-Office of Carcinogen
  Identification and
  Classification
U.S. Department of Labor
200 Constitution Avenue, N.W.
Washington, D.C.  20210

Dr. Joseph Fraumeni
Environmental Epidemiology
  Branch, DCCP
National Cancer Institute
Landow Bldg., 3C07
Bethesda, MD  20205

Dr. Howard Fribush
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. Wayne Galbraith
Office of Health Research
RD-683
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460
John Gamble, Ph.D.
Chief, Epidemiology and Statistics
  Section
NIOSH
944 Chestnut Ridge Road
Morgantown, WV  26505

Mr. Art Gass
OSHA, Room N-3700
200 Constitution Avenue, N.W.
Washington, D.C.  20210

Harvey Geller
Acting Program Director for
  Epidemiology
National Cancer Institute/
  National Institutes of Health
7910 Woodmont Avenue
Bethesda, MD  20205

Ms. Carol Graves
S.D.C.C.
2021 K Street, N.W.
Suite 207
Washington, D.C.  20006

Dr. Mark Green
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 8A53
5600 Fishers Lane
Rockville, MD  20857

Arthur Gregory
Pharmacologist
CPSC
5401 Westbard Avenue
Bethesda, MD  20207

Ms. Alice L. Griefe
Industrial Hygienist
Environmental Investigations Branch
NIOSH
944 Chestnut Ridge Road
Morgantown, WV  26505

Dr. Helene N. Guttman
Deputy Director, Science Advisory
  Board
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460
                                           857

-------
PARTICIPANTS LIST
Lynne R. Harris
Environment and Energy Science
  Advisor
NIOSH
MS-8-53
Rockville, MD  20857

Dr. Steve Hassur
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Clark W. Heath, Jr., M.D.
Director, Chronic Disease
  Division
Bureau of Epidemiology
Center for Disease Control
Atlanta, GA  30333

Dr. Andrew L Hegyeli
Program Director, Carcinogenesis
DCCR/NCI
Blair Bldg. , #616
8300 Colesville Road
Silver Spring, MD  20910

Dr. Ronald Herberman
Laboratory of Immunodiagnosis,
  DCBD
National Cancer Institute
Bldg. 10, 8B11
Bethesda, MD  20205

Doreen Hill
Health Research Scientist
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. Robert Hoover, Chief
Environmental Studies Section,
  EEB
National Cancer Institute
Landow Bldg., 3C06
Bethesda, MD  20205

Dr. Vilma Hunt
Deputy Assistant Administrator
Office of Health Research
Environmental Protection Agency
RD-683
401 M Street, S.W.
Washington, D.C.  20460
Dr. Gate Jenkins
Environmental Protection Agency
401 M Street, S.W.
PS-794
Washington, D.C.  20460

Dr. Larry Keefer, Chief
Analytical Chemistry Section, DCCP
National Cancer Institute
Bldg. 32, 1E22
Bethesda, MD  20205

Dr. Morris Kelsey
Landow Building, Room 8C25
7910 Woodmont Avenue
Bethesda, MD  20014

Harriet Kennedy
Science Writer
National Cancer Institute
Bldg. 31, 10A21
Bethesda, MD  20205

Ms. Kay Kennedy
Administrative Assistant
DCCP
National Cancer Institute
Bldg. 31/11A11
Bethesda, MD  20205

David S. Klander
Environmental Scientist
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Diann M. Kraft
Industrial Hygienist
DOL-MSHA-Metal and Monument Mines
4015 Wilson Boulevard
Arlington, VA  22203

Dr. H. F. Kraybill
Scientific Coordinator for
  Environmental Cancer
National Cancer Institute, DCCP
Landow Bldg., 3C37
Bethesda, MD  20205

Mr. Kenneth Krietel
NIOSH
Robert Taft Laboratories
4676 Columbia Parkway
Cincinnati, OH  45226

-------
PARTICIPANTS LIST
Dr. Phillip Landrigan
Director, DSHEFS
NIOSH
Robert A. Taft Laboratories
4676 Columbia Parkway
Cincinnati, OH  45226

Dr. C. C. Lee
Environmental Protection Agency
401 M Street, S.W.
TS-792
Washington, D.C.  20460

Ms. Sharon Leeney
Office of the Scientific
  Coordinator for Environmental
  Cancer
National Cancer Institute
Landow Bldg., 3C39
Bethesda, MD  20205

Carol Leibee
Environmental Protection Specialist
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. Nelson Leidel
Occupational Safety & Health
  Administration
Health Standards, Room N3718
Department of Labor Building
200 Constitution Avenue, N.W.
Washington, D.C.  20210

Carolyn Lingeman
National Cancer Institute
Bethesda, MD  20205

Edwin Lisiecki
Expert, Biometry Branch
National Institutes of Health
National Cancer Institute
Bethesda, MD  20205

Bob Liss
Environmental Protection Agency
TS-794
401 M Street, S.W.
Washington, D.C.  20460
Frances Loebenstein
Microbiologist
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. Larry K. Lowry
Robert A. Taft Laboratories
NIOSH
4676 Columbia Parkway
Cincinnati, OH  45226

Donald H. Luecke, M.D.
Chief, Special Programs Branch,
  DCCP
National Cancer Institute
Bethesda, MD  20205

Richard E. Marland
Director, Office of Research
  Grants and Centers
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. Thomas Mason
Environmental Epidemiology Branch,
  DCCP
National Cancer Institute
Landow Bldg., 3C07
Bethesda, MD  20205

Genevieve M. Matanoski, M.D.
Professor of Epidemiology
Johns Hopkins University
615 N. Wolfe Street
Baltimore, MD  21205

Dr. Linda Mendelsohn
Field Studies and Statistics
National Cancer Institute
Landow Bldg., Room 3C15
Bethesda, MD  20205

Ms. Sherri Milan
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 8A53
5600 Fishers Lane
Rockville, MD  20857
                                         859

-------
PARTICIPANTS LIST
Dr. Carl Morris
Office of Toxic Substance
Health Review Division
Environmental Protection Agency
TS-792
401 M Street, S.W.
Washington, D.C.  20460

Niren L. Nagda
Project Manager
Environmental Protection Agency
15 Firstfield Road
Gaithersburg, MD  20760

Larry Newsome
Environmental Protegtion Seps.
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. Gregory 0'Conor
Director
Division of Cancer Cause and
  Prevention
National Cancer Institute
Bldg. 31/11A03
Bethesda, MD  20205

Dr. Tong-Man Ong
NIOSH
944 Chestnut Ridge Road
Morgantown, WV  45226

Dr. Thomas Orme
Toxicology Branch, DCCP
National Cancer Institute
Landow Bldg., 3C25
Bethesda, MD  20205

James L. Oser
Industrial Hygienist
Industry-Wide Studies Branch,
  DSHEFS
NIOSH
4676 Columbia Parkway
Cincinnati, OH  45226
Dr. Norbert Page
Health Review Division
Environmental Protection Agency
TS-792
Mail Stop 460
401 M Street, S.W.
Washington, D.C.  20460

Mr. David Pedersen
Industrial Hygienist
Surviellance Branch, DSHEFS
NIOSH
4676 Columbia Parkway
Cincinnati, OH  45226

Dr. Linda Pickle
Environmental Epidemiology
  Branch, DCCP
National Cancer Institute
Landow Bldg., C307
Bethesda, MD  20205

Mr. Stanley F. Platek
Robert A. Taft Laboratories
NIOSH
4676 Columbia Parkway
Cincinnati, OH  45226

Dr. Harry B. Plotnick
NIOSH, DBBS, ETB, MIDS
Robert A. Taft Laboratories
4676 Columbia Parkway
Cincinnati, OH  45226

Dr. Gerald Rausa
Environmental Protection Agency
401 M Street, S.W.
RD-683
Washington, D.C.  20460

Dr. Richard Rhoden
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 8A53
5600 Fishers Lane
Rockville, MD  20857
                                      860

-------
PARTICIPANTS LIST
Dr. Wilson Riggan
Health Effects Research Lab.
Environmental Research Center
Environmental Protection Agency
Research Triangle Park, NC  27711

Kenneth Rivkin
Environmental Scientist
Environmental Protection Agency
401 M Street, S.W.
TS-794
Washington, D.C.  20460

Dr. Sheila Rosenthal
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. Umberto Saffiotti, Chief
Laboratory of Experimental Pathology
National Cancer Institute
Bldg. 37/3A17
Bethesda, MD  20205

Mr. Joseph Scotto
Biometry Branch, FSS
National Cancer Institute
Landow Bldg., B504
Bethesda, MD  20205

Joseph Schechter
Program Analyst
U.S. Dept. of Labor-OSHA
200 Constitution Avenue
Washington, D.C.  20210

Dr. Peter Schoor
Environmental Protection Agency
Gulf Breeze Environmental
  Research Laboratory
Gulf Breeze, FL  32561

Janet L. Scudiero
Criteria Manager
NIOSH/DCDSD
5600 Fishers Lane, 8A45
Rockville, MD  20857
Dr. George Simon
Acting Chief, Deputy,
  Epidemiology, Health Effects
RD-674
Environmental Protection Agency
Waterside Mall
401 M Street, S.W.
Washington, D.C.  20460

Dr. Margaret H. Sloan
Special Assistant to Director
DCCR, NCI
722 Blair, 8300 Colesville Rd.
Silver Spring, MD  20910

Christine Spadafor
Environmental Protection Agency
401 M Street, S.W.
TS-794
Washington, D.C.  20460

Dr. Robert Spirtas, P.H.
Biostatistician
National Cancer Institute
Room 3C07, Landow Bldg.
Bethesda, MD  20205

Dr. Janet Springer
Environmental Protection Agency
401 M Street, S.W.
TS-792
Washington, D.C.  20460

Mr. John Steelnack
OSHA, Room N-3700
200 Constitution Avenue, N.W.
Washington, D.C.  20210

Ms. Mariyln Stone
Environmental Protection Agency
401 M Street, S.W.
TS-794
Washington, D.C.  20460

Sandra Strassman-Sundy
Project Officer, National Hum.
  Monitoring Program
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460
                                       861

-------
PARTICIPANTS LIST
Ms. Cindy Stroup
Survey & Analysis Division
TS-793
Office of Toxic Substance
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Mr. David Sundin
Chief, Hazard Section
NIOSH
4676 Columbia Parkway
Cincinnati, OH  45226

Ms. Latidia Tahan
National Institute for
   Occupational Safety and Health
Parklawn Bldg., Room 8A53
5600 Fishers Lane
Rockville, MD  20857

Mr. Wade Talbot
Environmental Protection Agency
401 M Street, S.W.
RD-683
Washington, D.C.  20460

Scott D. Thayer
Senior Research Scientist
Environmental Protection Agency
15 Firstfield Road
Gaithersburg, MD  20760

Jacob Thomas
Statistician
Sigma Data
2021 K Street, N.W.
Washington, D.C.  20006

E. June Thompson
EPS
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

M. W. Townsend
Environmental Health Science
Environmental Protection Agency
401 M Street, S.W.
TS-792
Washington, D.C.  20460
Dr. Charles Trichilo
Environmental Protection Agency
Office of Toxic Substance
TS-792
401 M Stree't, S.W.
Washington, D.C.  20460

Jack Turer
Chemical Industry Specialist
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Michael Turner
Program Analyst
OSHA
756 9th Street, S.E.
Washington, D.C.  20003

Dr. V. Vallyathan
Pathology Section
NIOSH
ALOSH Building
P.O. Box 4257
Morgantown, WV  26505

Dr. Herb L. Venable
Robert A. Taft Laboratories
NIOSH, F-3
4676 Columbia Parkway
Cincinnati, OH  45226

John R. Ward
Senior Scientist
GEOMET,'lnc.
15 Firstfield Road
Gaithersburg, MD  20760

Dr. Michael Waters
Health Effects Research Lab.
Environmental Protection Agency
Research Triangle Park, NC  27111

Dr. Elizabeth Weisburger, Chief
Carcinogen Metabolism & Toxicology,
  DCCP
National Cancer Institute
Bldg. 37/3B25
Bethesda, MD  20205
                                      862

-------
PARTICIPANTS LIST
Mary C. White
Health Scientist
U.S. DOL-OSHA
200 Constitution Avenue, N.W.
Room N3718
Washington, D.C.  20210

Dr. Carrie Whitmire
Toxicology Branch, DCCP
National Cancer Institute
Landow Bldg., 3C25
Bethesda, MD  20205

Dr. Kin Fai Wong
Chemical Engineer
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Dr. Catherine Woodbury
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 8A53
5600 Fishers Lane
Rockville, MD  20857

Rhoda Yarkin
Biologist
National Institute for
  Occupational Safety and Health
Parklawn Bldg., Room 847
5600 Fishers Lane
Rockville, MD  20857

Peggy L. Young
Research Scientist
GTI/Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460

Mr. Dennis D. Zaebst
Industrial Hygienist
Industry-Wide Studies Branch
DSHEFS, NIOSH
4676 Columbia Parkway
Cincinnati, OH  45226
Mr. Matt Zenkowich
Environmental Protection Agency
401 M Street, S.W.
Washington, D.C.  20460
                                <1U.S. GOVERNMENT PRINTING OKHUL: I 93 I -3 4 I - 1 3 2 / 3 52 2

-------