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
------- |