UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                             WASHINGTON, O-C. 20460


                                   July 11, 1986
         M.
Administrator
U.S. Environmental Prelection
   Agency
Washington, D,c.  204SO

Dear Mr. Thonast

     The Science Advisory Board's Environmental Health Ccmnittee has co^leted
its review of the Office of Research and Development's Health Assessment
Document for Nickel.  The Committee carried out its review through its Metals"
Subcommittee'which net on March 24-25, 1986.  The Subcoamittee's report is
attached.

     The document appropriately characterizes the current scientific literature
on the carcinocjenicity of nickel compounds.  Hi is current revised document
is much improved in a number of ways over the previous draft.  The Subcamittee
identifies sane remaining revisers that the Agency staff should incorporate
into the final document before its final publication.

     The Board thanks you- for the opportunity to present its scientific
ccmnents on this issue and requests that a formal Agency reply be prepared
in response to the attached report.
                                        Sincerely,
                                               K
                                        Norton Nelson
                                        Chairman
                                        Science Advisory Board
                                        Richard A. Grieseiner
                                        Chainman
                                        Eiwirontnental Health Ccnmittee

-------
             UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

                           WASHINGTON, D,C. 204SO
                                June 9, 1986               SAB-EHC-86-Q26
Dr. Richard A. Griesemer
Chair, Environmental Health Committee                               OFFICE or
Science Advisory Board                                          THE ADMINISTR
U.S. Environmental Protection Agency
401 M street, sw
Washington, DC  20460

Dear Dr. Griesemer:

The Metals1 Subcommittee of the Environmental Health Committee met on March
24-25, 1986 in Farmington, Connecticut, to review the draft final Health
Assessment Document for nickel (EPA/600/8-83/01 2Py September, 1985)*
Subsequently^ members of the Subcommittee prepared individual comments.
This letter summarizes the Subcommittee's major conclusions.

The Environmental Health Committee reviewed a previous version of the
document in September, 1983»  The Subcommittee agrees that the current
draft is responsive to our earlier comments.  It now is clearer and more
comprehensive, and with the corrections suggested by the Subcommittee,
should be a scientifically accurate document*  Since staff indicated that
further editing is underway, our detailed comments have been transmitted
directly to the Office of Research and Development (ORD).  In addition,
the Subcommittee recommends that ORD resolve some technical problems that
remain in the three areas described below before the document is released
in final form.

 (1) As the document notes repeatedly, different forms of nickel exhibit
 different profiles of toxicity.  Nickel subsulfide is appropriately ae~
 knowledged as presenting the most serious carcinogenic hazard, but the
 document lacks clarity beyond that conclusion.  The scientific terminology
 from section to section is not always consistent, the chemistry of nickel
 is, at times, presented inaccurately, and the understanding of nickel
 manufacturing processes is incomplete.  Moreover, the nickel-ion hypothe-
 sis, which asserts that the proximal carcinogenic form is divalent nickel
 ion in solution, seems overstated, especially on page 8-210.

 (2) The discussion of nickel absorption, distribution, metabolism and
 elimination- -lacks a firm grasp of terminology and principles because of
 numerous instances of confusing and imprecise phrasing.  For example,
 "clearance" and "elimination" are used interchangeably, but these are
 different processes.  The term, "retention time," is used in a confusing
 way.  The definitions and roles of metal binding proteins are presented
 inaccurately.  The understanding of the principles of inhalation toxicology
 and of relevant empirical data seem limited, which leads in turn to incor-
 rect assumptions and .conclusions about the role of respiratory mechanics
 and function.  AS with other assessment documents, body surface area is
 used to correlate delivered doses between animals and humans.  This is an
 incorrect assumption for substances delivered by inhalation.

-------
                                    -2-
 (3) Heither the animal nor the European epidemiology data seem to b* used
 properly for quantitative risk assessment.  The animal data are seriously-
 flawed and have limited utility for either extracting definitive conclu-
 sions or assessing human risk.  For example, little confidence can be
 placed in a data set in which control animals exhibit 31* survival,  The
 epidemiology data lack reliable exposure estimates and do not reveal much
 familiarity with the complexities of the manufacturing process.  The
 quantitative risk assessment based on European epidemiology relies on U.S.
 cancer mortality rates for unexposed persons, despite the differences in
 these rates for U.K. and Norwegian population studies.  This inappropriate
 adjustment could seriously bias the unit risk estimates derived from these
 studies.

We appreciate the opportunity to comment on this public health issue and
hope that our comments are useful to the Science Advisory Board and the
Agency.
                                 Sincerely yours,
                                    Bernard Weiss, Ph.D*
                                    Chair, Metals' Subcommittee
                                                   .'-' _.^  f 7
                                    Ronald Wyzga, Sc.Dj/" / '/'
                                    Vice-chair, Metals' Subcommittee

-------
                  U.S Environmental Protection Agency
                          Science advisory- Board
                      Environmental Health Committee
                            Metals' Subcommittee
                              March 25, 1986
Dr. Bernard Weiss [Chair], Professor,  Division of "toxicology,  P.O.  Box
RBB, University of Rochester, School of Medicine, -Rochester, NY  14642
(716)275-3791

Dr* Bonald Wyzga [Vice-chair], Electric Power Research Institute, 3412
Hillview Avenue, P.O. Box 1041, Palo Alto, California 94303  (415)855-2577

Dr. Thomas Clarkson, Professor and Head, Division of Toxicology,  University
of Rochester, School of Medicine,  Pest Office Box RBB,, Rochester, New York
14642 (716)275-3911                                                '   ' -

Dr. Gary Diamond, Assistant Professor  of Pharmacology, University of
Rochester School-of Medicine, Rochester, New York 14642 (716)275-5250

Dr* Edward F. Ferrand, Assistant Cotmiss ioner for Science and  Technology,
New York City Department of Environmental Protection, 51 Astor Place, New
York, New York 10003 (212)566-2717

Dr. Itobert Coyer, Deputy Director, NIEHS, P.O. Box 12233, RTF, North
Carolina 27709 8-629-7620

Dr, Marvin Kuschner, Dean, School of Medicine, Health Science  Center,
Level 4, State University of New York, Stony Brook,  New York 11794
(516)444-2080

Dr. Brooke T. Mcssroan, Department of Pathology, The University of Vermont,
Medical Alumni Building, Burlington, Vermont 05405-0068 (802)  656-2210

Dr, Gunter Oberdoerster, Associate Professor, Radiation Biology and Biophysics
Division, University of Rochester, School of Medicine, 400 EJUwood  Avenue,
Rochester, N.Y. 14642 (716) 275-3804

Dr. P. William Sunderman, Professor of laboratory Medicine and Pharmacology
and Head, Department of Laboratory Medicine, University of Connecticut
Health Center, Room C 2021, Farraington, Connecticut 06032 (203)674-2328

Executive Secretary

Dr. Daniel Byrd, III, Ejsecutive Secretary, Science Advisory  Board [A-1Q1F],
U.S. Environmental Protection Agency,  Washington, D.C. 20460 (202)382-2552

-------
UNIVERSITY OF  ROCHESTER
School of Medicine and Dentistry

Department of Radiation Biology & Biophysics
                                                      DIVISION OF TOXICOLOGY
                                                      Toxicology Training Program
                                                      Environmental Health Sciences Center
                                                       April 28, 1986
          Dr. Daniel M.  Byrd III
          Executive Secretary
          Science Advisory Board
          USEPA
          401 M  Street,  S.W.
          Washington,  D.C. 20460

          Dear Dr. Byrd:

               Please  find enclosed roy post-meeting comments on HftD - Nickel
          Speciation.   I am also forwarding a copy to Bernie Weiss.

                                                       Sincerely,
                                                       Tom Clarkson
                                                       Professor
           tw/cmkb
           enc.
                        Mailing address: University of Rochester School of Medicine,
                              P.O. Box RBB • Rochester, New York

-------
                           HAD - NICKEL - SPECIATION

                                 TOM eiARKSON




                             POST MEETING COMMENTS
INTRODUCTION

The HAD Is well written  and addresses the  key  issues raised in  previous SAP
reviews.   These post-meeting comments are  directed towards the speciation of
nickel  and  its  compounds - the physical  and chemical  species  that  contain
nickel,

Speciation  is  a theme that  permeates this document.  Indeed the  previous SAB
review  stressed its importance in  all  aspects of  the HAD,   One  of  the major
issues  -  the  "nickel  ion" hypothesis for  careinogenesis devolves around the
speciation of nickel.  This brief post-meeting review will be presented under a
number of sub-headings.

TERMINOLOGY

The public  comments  from  INCO indicated  ambiguities in the HAD.   Sometimes the
word  "nickel"  is used to indicate the element, sometimes to refer generically
to all  forms of nickel,   clearly this  problem is common to all HAD's dealing
with metals.  In the case of mercury, the problem was solved by using  the terms
metallic mercury or  elemental mercury when  referring to  the element.   I do not
regard  this as an important issue  but "down  the  line"  it would  be  useful to
develop consistent terms for all the metal HMJS.

Public  comments at  the meeting drew attention to  the many physical  forms of
nickel oxide.   One of the public written comments (INCO) used the term "oxidic
Nickel",    unfortunately  some  of  the  original  publications  do  not  always
identify  the specific physical  form used in  the  study.  Questions  were also
raised  at the meeting  on the use  of the designation "chemical compound* for
nickel  oxides  or complex  oxides  of nickel  and other metals such  as  copper as
the atomic  proportions are variable.   It was suggested that the word substance
be used,    I  do not  think this  issue is sufficiently important  to  justify a
complete check  of all the publications  on "nickel oxide" used in the HftD.  The
on-going "Nickel Speciation Research Project" in which the EPA is participating
would help clear up this ambiguity for any future documents on nickel.

THE MANUFACTURING PROCESS

A thorough  knowledge of the manufacturing process  is  essential  to identifying
each  species of nickel of importance to human exposure.   The public comments
from  one  of  the  manufacturers  emphasizes this  point  -  "The  retrospective
estimation of environmental conditions can be done only by people familiar with
the plants  and processes  and with access to industries records and personnel".
Clearly  the  authors  of   the  HAD  are  at   a  disadvantage  in having to  use

-------
"second-hand"   and  frequently   incomplete  information   in  the   published
literature.   It is obvious  from the written comments and from those  given at
the meeting that the HAD has not identified all the species of nickel and other
substances involved  in human exposure  at  various stages  of  the  manufacturing
process,  in  this  respect,  the  comments from the two principal manufacturers -
INCO  and Falconbridge -  are  specially important.   Their  detailed  written
comments on  nickel species  that exist  at stages  of  the  manufacturing process
should be used to improve both the accuracy and scope of  the HAD's treatment of
speciation during  manufacture,  e.g., Table 8-1 is clearly incomplete.   INCO's
comments on speciation are too numerous to be detailed here.

ATMOSPHERIC  NICKEL

The public comments agree with  the HAD that currently available methods do not
allow  speciation of  nickel  at  concentrations of total  nickel  found  in the
ambient  atmosphere.    Calculations made at the  meeting  indicate that direct
speciation will  remain unattainable  in the foreseeable  future.   This lack of
information is a critical gap in our knowledge in view of the controversy over
which species of nickel may be regarded as carcinogenic.

HAD has attempted through one of its contractors to calculate  the most probable
species from  themodynamic  and  other considerations   - in all, 19 species were
identified.   The public comments urge  that  a more practical  approach could be
made by measuring  the  forms of  nickel in the undiluted emission sources to the
atmosphere.  Techniques are  available  to  speciate nickel into water soluble
forms,  Ni,S2, metallic nickel  and  "nickel  oxide".    in  addition, it  will be
necessary  tb  know the relative  contributions  from the different  emission
sources  and  the  residence  time of  each species  in the atmosphere.   These
estimates clearly  cannot  be  included in the current HAD  but should remain an
important objective for the future.

TOXICITY AND CAECINOGENICITY

The HAD adequately covers the importance of speciation in determining toxicity
and carcinogenicity.   Some  of  the earlier  papers may not have satisfactorily
identified  the precise  species of  nickel,  e.g.  the specific fora  of nickel
oxide.

Theories on mechanisms of carcinogenesis are appropriately mentioned in the HAD
but it  remains  for future  research to further elucidate these mechanisms.  The
so-called "nickel-ion"  hypothesis is at best vague and interpreted differently
by different  people  as at the meeting.  One interpretation is that the nickel
ion reacts directly with DMA. to  initiate the carcinogenic process.  However, as
noted  by Nieboer,  many other divalent  metal  ions react  with DNA but  do not
produce  cancer.    So  what  is  special  about  nickel  ions?    Theories  on
bioaccumulation  must  be  added  to  account for  carcinogenesis.   To further
complicate  the picture, Nieboer has presented evidence  that a  soluble small
molecular  complex  of  nickel can  cause  the  cellular  machinery to  produce
"di-oxygen  free radicals"  that  have mutagenic properties.   The  HAD correctly
describes  this  putative  mechanism  as a   hypothesis  in  the main  text  but
unfortunately mentions it in the important summary paragraph on page 8 - 210 in
an attempt to justify classifying  "all compounds of nickel" as "potential human
carcinogens".  Instead, it is suggested that the first paragraph of section 8-5
should  be  rephrased.    The text  starting "However,  there  is  a  reasonable

-------
probability,..to the end of  the  paragraph..."is  not well understood" should be
deleted and  replaced by "The  question of the carcinogenicity of  other nickel
compounds  remains open  and  is  therefore  an  important  subject  for  further
investigations."

-------
Chapter 3.  Nickel Background  Information


General Comments

The background information should be directed more strongly
towards those nickel compounds and  their special properties which
may have some relevance  to health effects.  Thus, the physical
and chemical properties  of nickel subsulfide, nickel carbonyl and
nickel oxides should be  featured.
The electronic structures of nickel and nickel  ions and  the
effects which they have on  the properties of  the metal and the
compounds and complexes of  nickel should be explained,


The use of nickel metal in  fi.nely divided form  as a catalyst  in
the hydrogenation of oils should be mentioned.  The possibility
of nickel entering the food chain via  this route should  be
evaluated, -Nickel's ability to serve  as a catalyst in these
reactions may be important  evidence in understanding some of  its
biological interactions,


Nickel metal atoms have two 4s electrons and  eight 3d electrons,
two of which are unpaired.  Formation  of the  nickel II ion
involves the loss of two 4s electrons.  The two unpaired
electrons in the 3d shell of the metal and nickelous ion give
rise to paramagnetism.  There are indications that catalytic
hydrogenation by hydrogen in the presence of  nickel metal in
finely divided form entails the dissociation  of hydrogen
molecules into atoms within the nickel metal  luttice.  The
decrease in paramagnetism as hydrogen  gas is  absorbed by the
metal is evidence that the  unpaired elections in the nickel atoms
are being paired with those in the dissociated  hydrogen  atoms.
The formation by Ntll of stable complexes with  cyanide ion also
occurs with loss of paramagnetism*


Ground rules should be established for use of the word nickel so
that it is always clear whether the reference is to the  elemental
state or a compound or ion of the element.


The information should be presented in a more direct manner - the
writing style needs to be changed to accomplish this.  For
example, the first two paragraphs on p.3-13 could be rewritten in
a more concise manner.
"Brief has described several methods which range  in sensitivity
from 0.008 to 0.10 g for the determination of nickel carbonyl
(Brief et al, 1965).  A chemiluminesience method  base.d  upon  the
reaction between nickel carbonyl and ozone is faster and

-------
sensitive down to parts per billion  (Stedman et al.,  1979}."


There is a considerable amount Of material included which out of
context of the source documents provides little useful
information and is sometimes intelligible only to readers who are
already thoroughly familiar with the subject.


P.3-1

Comments on Specific Material in Chapter 3

     "... nickel content of some nickel-containing minerals" is
     redundant.

     Has or has not native metallic  nickel in a pure  form been
     observed?  (What does rarely, if ever, imply?)

     Elemental nickel dissolves in diulte acids, not  only dilute
     oxidizing acids.  The statement that "even oxidizing salts
     do not "corrode nickel because the metal is made  passive, or
     incapable of displacing hydrogen, by formation of a
     surficial oxide filin" is taken out of context and requires
     further explanation.

     The existence of a true -1 state of nickel, that is, nickel
     acting as a nonmetal by adding  electrons does not make
     chemical sense.
P»3-2 Table 3-1

     Nickel Arsenite should be Nickel Arsenate

     According to the text in p.3-3, the basic salt
     2NiCO33Ni(OH)24HzO "is the roost important form11 yet it does
     not appear in the table (Is it the most important commercial
     form?)

     The melting point and boiling points given for nickel
     nitrate hexahydrate are highly questionable; decomposition
     rather than melting and boiling are involved.
P.3-3
     Greater attention should be given to the formation of nickel
     oxide which is an important process because of its relevance
     to the nickel emitted into the environment as a result of
     combustion.

     The nickel-ammonia complex is Ni(NH3)64"1" in solution - the
     hydroxide ion is not part of the complex in solution.

     "When dissolved ... and is rendered soluble" is redundant.

-------
     Cations form complexes with ligands
P. 3,8 Section 3.2

     This section would be easier to write and easier to read if
     it were organized in a different manner.

        1.  Sampling methods
        2.  Preparation of samples for instrumental or,
            colorimetric analysis
        3,  instrumental analysis


     Analysis of most air, water or soil samples are currently
     performed by atomic absorption with the use of inductively
     coupled plasma spectroscopy expanding.  After the sample
     preparation step, the measurement of the nickel
     concentration in the prepared solutions is essentially the
     same regardless of the type of environmental sample.

     Particulate matter may be solid or liquid.  Nickel compounds
     occur in the atmosphere as particulate matter - they do not
     necessarily have to be associated with other atmospheric
     pollutants.  Referring to nickel as having to be associated
     with particulate emissions is confusing.
   3.9

     The melting and boiling points-of nickel are 1455*C and
     2920*C; referring to nickel as a volatile trace element in
     streams at temperatures up to 500'C is inappropriate.

     High volume samplers are used for ambient air monitoring.
     Because of instability it is not likely to find nickel
     carbonyl in the ambient atmosphere.
3.2.2
     The phrase "in its elemental state" implies presence of
     nickel as the metal which is not the intent of the
     statement.

     The meaning of the third sentence is not clear as stated.

     Identification of the nickel compounds present in an ambient
     air sample is very difficult because of the very small
     amounts usually present mixed in with a complex matrix of
     other particulate matter and the changes that attempts to
     separate trace amounts often have on the nature of the
     compounds present.

-------
P. 3.13
     The discussion on sampling should either be expanded by
     providing a better description of the sampling methods or
     shortened by providing more detailed references.  I suggest
     the latter.

     The statement;  "Any of the following three methods are
     recommended;  "is misleading,  The state:  "The sample is
     removed by a valve regulating flow from a clean Teflon line
     inserted into the sampling bottle.  " is awkward - valves do
     not remove samples and samples are not removed from the
     sampling bottle.
P.3.14
     The discussion of preeoncentration does not belong under
     sampling,

     Under 3,2.2 (Air) the detection limit of the AAF method is
     given is 0.005 ug/ral? under 3*2.4 the detection limit is
     given as 0.05 mg/1  (U.S. EPA 1979}

     "Standard Methods for the Examination of Water and
     Wastewater (1985)" gives a detection limit of 0.02 ug/ml by
     flame and 0.001 ug/ml by graphite furnace.

     Because the detection part of the analysis of either air
     samples or water samples by atomic absorption involves the
     same procedure, the discussion of atomic absorption should
     be given in one place with these obvious discrepancies
     resolved.
P. 3-15
     The first two sentences in 3.2,7 say very little and should
     be eliminated.
P. 3-16
     The heading of Section 3.3 should be "Sources of Atmospheric
     Nickel" rather than "Nickel in Ambient Air".  The text  is
     concerned predominantly with sources and not ambient air.


     The information in Section 3.3 should be summarized in
     tabular form with references.  The five main groups of
     sources together with the nickel species emitted and
     emission factors.  (Amounts of nickel emitted per unit
     activity), where available, would be the components of  the
     table.  The discussion would focus on the table and could be
     considerably shortened.

-------
     The availability of emission factors for different types of
     operations is of great importance to environmental control
     work because they provide major guidance in the allocation
     of the resources dedicated to the elimination of a problem.
     See the discussion in Section 3.6.1*  Note the estimate of
     up to 80% of nickel from human activities as coming from
     fossil fuel combustion.
P. 3-24
     Table 3-2 would be more appropriately placed in section
     3.3.2 than under analytical procedures.

     The statement is made that neutron activiation analysis is
     not performed on atmospheric nickel samples "because no
     suitable states exist in the nuclei of nickel isotopes".
     How is this statement reconciled with the discussion of NAA
     on P. 3-11?
P. 3-25
     The return to another discussion of the same five source
     groups and the separation of "Nickel Species in Water" from
     "Concentration of Nickel in Ambient Waters" repeats the
     organization under air.  This argues for a restructuring of
     chapter 3 which unifies discussion of sources that release
     nickel into the environment, their impact on air, water and
     soil and the pathways into the human body then represent.
3-26/3-34
     The author continues to have problems identifying nickel
     species in the ionic state.  Soluble nickel salts,
     especially in highly dilute solutions exist as ions which
     are relatively independent of each other*  Statements such
     as "This nickel is likely to be discharged as the Ni+2 ion
     or as dissolved nickel salt {sulfate chloride, etc.)" imply
     incorrectly that Ni+^ can exist either as the independent
     ion or in solution as
     The map legend (p. 3-33) does not indicate the significance
     of the counties which are shown without coloring.

     in the text the Southeast basin is said to have means
     ranging from 85.1  to 754 ug/1 while in Table 3-3, the range
     of means is 45.4 to 77,6.  The Northeast had a maximum of
     9,140 ug/1 in 1980 but its 85th percentile was 105 compared
     to one of 173 for the southeast which had a maximum of only
     900.  This shows the desireability of frequency distribution
     statistics for such data sets.

-------
     The text  is again at variance with  the  table.   "...  for  the
     Southeast Basin in  1980 where the maximum  reported value was
     1500 ug/1 but 85 percent of  the  remaining  samples contained
     less than 130 ug/l"»  Table  3-3  shows 900  and  173 for  the
     same statistics.
3-34/3-38
     Oil and coal used for space and hot water heating  should  be
     included with power utilities as sources of nickel  in  soil.
     Space and hot water heating emissions usually occur  at  lower
     altitudes with less opportunity for dispersion  that  those
     from tall stacks,  incineration of urban wastes also should
     be cited as a source of soil contamination.

     in Section 3-6.1 it is reported that combustion of oil  alone
     accounts for 83% of atmorpheric nickel  from human
     activities.  Space and hot water heating are major
     contributors.
P. 3-40
     Nickel in Food - Should restrict the discussion  to  the
     nickel content of different foods.  Section 4.1,2,
     "Gastrointestual Absorption of Hickel", is the proper place
     to discuss the relationship between intaJce and fecal
     excretions of nickel.
P. 3-42 Section 3.6

     The last sentence in the first paragraph should be
     rewritten.

-------
                           F, WILLIAM SUNOERMAN, JR.. M.D.,
        Qtpartments of Labaracofy Medicine and Pharmacology, University of Connecticut School of Medicine
             283 Fgrmington Avenue, Farmington, Connecticut. 06032. Telephone 203/674-2328
10 March 1986
Mr, Daniel Byrd
Executive Secretary
Science Advisory  Board  (A-101F)
U.S. Environmental  Protection Agency
401 M Street,  S.W.
Washington, DC 20460

Dear Dan:

In preparation for  the  meeting on 25 March 1986 of the  EPA Metals Subcommittee
to consider the Health  Assessment Document for Nickel,  I  have prepared the
following line-by-line  citations of some points in the  document  that may need
clarification  or  correction;

p1v, 1(2, line 11.    The word "brain" should be omitted,  unless  the statement
                      is modified to specify nickel carbonyl.

p xiv, line 23.       My initial "F." should be placed before  "William",

p 2-4, fl.            Soya beans and soya products contain an  average of 5,5 mg
                      Ni/kg (range 1.1 to 7.8) and cocoa contains 9.8 mg Ni/kg
                      (range 8.2 to 12), according to Nielsen  and Flyvholm
                      (1984).

p2-4, |2.            Volcanic emissions may be cited as an additional natural
                      source of atmospheric nickel.

p 2-4, 1(3.            Human parenteral exposures to nickel  are of importance in
                      relationship to iatrogenic sources,  such as implanted
                      orthopedic pros-theses, hemodialysis  treatment, and
                      injections of nickel-contaminated  drugs  and X-ray contrast
                      media.

p 2*9, 13, line 5    The statements on post-partum hypernickelemia should be
and p 2-12» 1J6,       deleted  from the summary.  The original  observations of
line 5.Rubanyi  et_aK (1982) are dubious, owing to analytical
                      prob 1 emsy~post-partum hypernickelemia has not been
                      observed in a follow-up study by Nomoto  et  al. (1983).

p 2-13, Jlt line  1.   Is there evidence that family history is predictive of
                      susceptibility to nickel sensitization,  on  the basis of
                      hereditary predisposition?

-------
                                                                            (2)
p 2-13, |2»1ast     The measurements by Sutenmann etal.  (1982)  of  nickel  in
2 lines.             cigarette smoke do not  agree withearlier  measurements
                     (Menden et al. 1972, Wescott and  Spincer,  1974, Perinelll
                     and Carcigno, 1978),  This matter  Is  unsettled  and  further
                     research is needed.  Studies by Alexander  et al.  (1983)  do
                     indicate that the nickel  in mainstream  smoke is not
                     present as nickel carbonyl.  The  statement that "nickel  in
                     mainstream smoke is "minimal" is  imprecise and
                     controversial and should  be deleted.

p 3-11, If2, line 6.  The unit should be 'ii 1" rather than "ml".

p 3-15, |4, lines    The statement that "neutron activation  analysis and
Z-3'.  "colorimetric procedures are also  used"  is  incorrect and
                     should be deleted.  Instead, a statement can be inserted
                     that "anodic-stripping  voltametry  and isotope-dilution
                     mass spectrometry are also used".

p 3-16, |1, line 14. The reference to Nomoto and Sunderman (1970)  is
                     out-of-date.  More up to  date references are Stoeppler
                     (1981,1984) and Sunderman (1984).
p 3-23. 1f2, line 3.
p 3-4Q, |4-6, and
1able "3-8".
p 3-42.


p 3-42, 14.
p 4-1, fl, line 6,
p 4-9. HI.
p 4-18. Table 4-2.
The importance of nickel exposures from mold-making  in
glass bottle factories should be mentioned  (Raithel  et  al.
1981,1985).                                             ~~

Except for the study by Myron ^ a1.  (1978)>  the  cited
studies on nickel concentratioffs"irf foods are out-of-date.
More recent references are Ellen et al. (1978), Nielsen
and Flyvholm (1984), and Flyvholm et  al.  (1984).

A paragraph on nickel in bacteria and  other microorganisms
might be appropriate at the end of section  3,5.

The citation of Weast (1980) (CRC Handbook  on Chemistry
and Physics) can be deleted; the paper by Hassler (1983)
refers to an unpublished report; the  paper  by Sutenmann et^
al. (1982) refers to nickel in smoke  from tobacco grown on
municipal sludge-amended soil.  Additional  references that
may be cited are given above,

Parenteral exposures of humans to nickel from prostheses,
medications, hemodialysis, etc,, should be mentioned, as
discussed by Sunderman et al. (1986).

The discussion of nickel in cigarette  smoke should be
modified, as indicated for p 3-42, f4.

The dosages given for the mouse experiment  of Oskarsson
and Tjalve (1979) are erroneous (4.6yg Ni/kg).   Moreover,
the footnotes of the NAS table have been omitted.  The
footnotes, which specify the intervals between last
injection of SSnnc^ and death, are necessary for
interpretation of these experimental  data.

-------
                                                                            (3)
p 4-20 to 4-22,      Nickel concentrations in human milk might be  discussed
                     (Mingorance and Lachica, 1985; Feeley et al.,  1983).

p 4-24, 1(4.          Evident analytical problems in the study by  Rubanyi et  al
                     (1982) should be mentioned, and the contrary  findings  of
                     Nomoto st._tlt (1983) should be discussed.

p 5-8, 12, Jine_!_,_   DL-alanine (not alaline).

p 5-13, f4.          Additional cases of cancers at the sites of  implanted
                     nickel-containing prostheses have been reported  (see
                     references cited by Linden et al., 1985).

p 5-23, |4.          The findings of Hopfer et al. (1985), showing  that the
                     erythrocytosis is mediated by enhanced erythropoietin
                     production should be mentioned,

p 8-116, |2.         The word "tested" should be changed to "found".
line 22,

p 8-157, footnote.   Change "cm3" to "m3".


     In general, the present document fs substantially improved,  in comparison
to earlier drafts.  Unless you disagree, I do not believe that the  specific
points that are mentioned in this letter need detailed consideration  by the
Metals Subcommittee.  I suspect that our discussions in Farmington  will
be centered upon the controversial proposition that "there is a reasonable
probability that the ultimate carcinogenic form of nickel is nickel ion" and
that "on this basis, all compounds of nickel might be regarded as  potential
human carcinogens , . ." (p 8-210).

Looking forward to seeing you on 24 March, and with cordial regards,  I am,

Sincerely yours.
F. William Sunderman Jr., M.D.
Professor of Laboratory Medicine
and Pharmacology

/ms

-------
                                                                            (4)
References that are not cited In the Document

Ellen, G., van den Bosch-Tibbesma, S.» & Douma, F.F. (1978) Nickel content  of
various Dutch foodstuffs.  Z. Lebensin. Unters. Forsch, 166, 145-147,

Feeley, R.M., Eitenmiller, R.R., Jones, J.B., Jr., & Barnhart, H.  (1983)
Manganese, cobalt, nickel, silicon and aluminum in human railk during early
lactation.  Fed. Proc., 42, 931.

Flyvholm, M.A., Nielsen, G.D., & Andersen, A. (1984)  Nickel content of food
and estimation of dietary intake.  Z> Lebensro. Unters. Forsch., 179, 427-431.

Hopfer, S.M., Sunderman, F.W., Jr., & Goldwasser, E. (1985b)  Effects of
unilateral intrarenal administration of nickel subsulfide to rats on
erythropoietin concentrations in serum and in extracts of both kidneys.  In:
Brown, S.S. & Sunderman, F.W., Jr., eds,» Progressin Nickel Toxicology,
Oxford, Blaekwell, pp 97-100.                ""

Linden, J.V., Hopfer, S.M., Gossling, H.R., & Sunderman, F.W., Jr. (1985)
Blood nickel concentrations in patients with stainless-steel hip prostheses.
Ann. Clin. Lab. Sci., 15, 459-463.

Menden, E.E., Ella, V.J., Michael, L.W. and Petering, H.G. (1972)  Distribution
of cadmium and nickel of tobacco during cigarette smoking.  Environ. Sci.
Techno1., £:83Q-832.

Mingorance, M,D. & Lachica, M. (1985)  Direct determination of some trace
elements in milk by electrothermal atomic absorption spectrometry.  Anal.
Lett., 18. 1519-1531.

Nielsen, G.D. & Flyvholm, M. (1984)  Risks of high nickel intake with diet.
In:  Sunderman, F.W., Jr., ed.-in-chief, Nickel in the Human Environment, Lyon,
Intern. Agency Res. Cancer, pp 333-338.

Perinelli, M.A., and Carugno, N, (1978) Determination of trace metal in
cigarette smoke by fTameless atomic absorption spectrometry.  Bertr. Tabakfors.
Intern. 9:214-216.                                            ~~         "~^

Nomoto, $,, Hirabayashi, T., & Fukuda, T. (1983)  Serum nickel concentrations
in women during pregnancy, parturition, and post-partum.  In:  Brown, S.S.  &
Savory, J., eds., ChemicalToxicology and Clinical Chemistry of Metals, London,
Academic Press, pp 351-352.

Raithel, H.J., Mayer, P., Schaller, K.H., Mohrmann, W., Weltle, D.f and
Valentin, H.  (1981) Untersuchungen zur Nickel-Exposition bei Beschaftigten in
der Glas-Industrie.  Zbl. Arbeltsmed., 31_;332-339.

Raithel, H.J., Kress, W.t Schaller, K.H., Weltle, D.t & Valentin, H. (1985)
Toxicological and occupational medical investigations concerning nickel-
exposure in different industrial areas in the FRfi.  In:  Brown, S.S. &
Sunderman, F.W., Jr., eds., Progress in Nickel Toxicology, Oxford, Blackwell,
pp 219-222.

Stoeppler, M. (1981)  General analytical aspects of the determination of lead,
cadmium, and nickel in biological materials.  In:  Facchetti, S., eds.,
Analytical Techniques for Heavy Metals in Biological _F1 inds, Amsterdam,
Elsevier,pp 133-154.

-------
                                                                           (5)
Stoeppler, M. (1984)  Analytical chemistry of nickel.  In:  Sunderman, F.W.,
Jr., ed.-in-chief. Nickel In the Human Environment, Lyon, Intern, Agency Res.
Cancer, pp 459-468.

Sunderman, F.W., Jr. (1984)  Nickel,  In:  Vercruysse, A., ed., Hazardous
Metals in Human Toxicology, Amsterdam, Elsevier, pp 279-306.

Sunderman, F.W,, Jr., A1t1o» A., Morgan, L.G., & Norseths T. (1986)  Biological
monitoring of nickel.  Toxicology and Indust. Health (1n press).

Wescott, D.T. and Splncer, 0. (1974)  The cadmium, nickel and lead content of
cigarette smoke.  Belt. Tabakfors. 7:217-221.

-------
EPRI
Electrie Power
Research Institute
April 14, 1986
Dr. Daniel Byrd
Science Advisory  Board
US/EPA A-1Q1F, Rm 508
Washington, DC  20460

Dear Dan:

I have reviewed the  Nickel  Health Assessment Document  and
offer the following  comments about the quantitative risk
assessment.

Animal Studies.
The data from the Ottolenghi et al. study were used to
undertake a quantitative  risk assessment for nickel,   r have
great concern about  using these data for two reasons.  First
of allr one-half  of  the control and treated- anamals were
injected intravenously with hexachlorotetra-flourobutane,  an
agent used to induce pulmonary infarction.  There is no
available information to  indicate whether there may be any
synergism or antagonism resulting from the simultaneous
exposure of this  drug and nickel subsulfide* which also
largely affected  the lungs.  At a minimum the Ottolenghi  et
al. data set should  be subdivided into two groups for  risk
assessment purposes.  Those animals injected with
hexachlorotetrafluorobutane and those which received no
injection,  if the risk assessment results are similar, the
data could be concerned.

I am also very uncomfortable with the very high mortality
rate in the nickel-exposed  and control groups in this
study.  Apparently only 5%  of the exposed group and 31% of
the control group survived.  It would be useful to have some
guidelines about  minimally  acceptable survival rates before
a data set is subjected to  a quantitative risk assessment*
    3412 Hillview Avenue, Post Office Box 10412, Palo Ate, CA 94303 Telephone (415) 855-2000
    Washington Office: i860 Massachusetts AM., NW, Suits 700, Washington, DC 20036 (202) 872-9222

-------
Dr. Daniel Byrd, 4/14/86
Page 2


    In addition, clearly-articulated caveats should be attached
    to any quantitative results using this study — even if the
    results are only used to compare the "best" animal study
    with epidemiology results.   The comparison may not be
    meaningful.

    Epidemiology Studies
    Four data sets were used for these analyses.  The
    Huntington, W.Va. data are  taken from a study by Enterline
    and Marsh, an apparently carefully undertaken study.  The
    analyses derived from this  data set are reasonable and
    clearly articulated.

    Data from Copper Cliff, Ontario were the second data set
    analyzed.  I have two concerns with the analyses of these
    data.  First of all, exposure is poorly characterized and
    rough exposure estimates are used in the analyses.  Since
    the selection of exposure values can influence the results,
    I suggest that there be some type of sensitivity analysis
    undertaken for this study showing how risk (potency)
    estimates vary as exposure  estimates vary.  Secondly, the
    analysis used a background  lung cancer rate of .036, derived
    from U.S. white male data.   There is some difference between
    the U.S. and Canada as seen from the attached figure; but of
    greater concern is the increase over time in lung cancer
    rates.  A recent EPA publication (U.S.Cancer Mortality,
    Rates and Trends, EPA600-1-83-015A) shows that the lung
    cancer mortality rate for white males in the U.S. has
    increased from 29.6 (1950-59) to 46.8 (1960-69) to 64.0
    (1970-79) per 100,000.  Given this volatility, it is
    important that the correct  background rate Pa be used in the
    risk assessment calculation for the Copper Cliff data.  The
    use of .036 may be incorrect.

    The third study used in the analysis was from Clydach,
    Wales.  The analysis of these data suffer from similar
    problems as the Ontario data except the extrapolation of a
    U.S. background rate to Wales may be more questionable than
    extrapolation to Canada. Figure 1 demonstrates how much
    higher male lung cancer mortality rates are in England and
    Wales over the U.S.  This study may also suffer from some
    confounding of exposure as  practically all of the lung
    cancer deaths apparently were exposed to arsenic.  The
    analysis should show how any arsenic exposure would impact
    the risk estimates.

-------
Dr. Daniel Byrd, 4/14/86
Page 3
    The final study analyzed is of a Kristiansand, Norway
    population.   Concerns about this analysis are similar to
    those for the Copper Cliffs analysis.   The attached figure
    demonstrates the extrapolation of U.S*  background lung
    cancer mortality rates to Norway may be dangerous, as
    Norwegian rates appear to be considerably lower than U.S.
    rates.  The  analysis for this study is  also not clearly
    articulated.  A table analagous to Table 8-47 would be
    helpful here so that the assumptions made in the analysis
    would be clearer.

    Overall Comment
    I am somewhat troubled by the use of "median of the range"
    risk estimate.  Since this apparently turns out to be the
    average of the highest and lowest risk  estimates, it is
    dominated by the larger number.  A better approach might be
    to derive a  median estimate for each study and take the
    median of the medians.

    I hope the above are useful.  Obviouslyf if any
    clarifications are desired, please let  me know.
    Sincerely,
    Dr»  Ronald Wyzga
    Technical Manager
    Environmental Risk Assessment

    Enclosure

    cc;   Bernie Weiss

    HEW/acc

-------
Dr.  Daniel  Byrd,  4/14/86
Enclosure
                                      FIGORE 1

         - :GURE !6. AGE-ADJUSTED MORTALITY RATES FOR MALIGNANT NEOPLASMS
                  OF THE LUNG, BRONCHUS, AND TRACHEA IN VARIOUS COUNTRIES,-
                  1966-1967,
                    MALE
                        t
Scotland
England i Wales
        Netherlands
        Austria
        Belgium
        U. S., Non-wnit
        North. Ireland
        Germany, F.H,
        U. S>. Whits
        New Zealand
        Australia
        South Africa
        Denmark
        Switzerland
        Ireland
        Italy
        France
        Chile
        Norway
        Jaoan
                             33       40       SO
                            RATE PSR 100.000 POPULATION
                                                         Seoiland
                                                         England & Wales
                                                         Ireland
                                                 Denmark
                                                 U.S.. NOn-whrte
                                                 North, trelarnj
                                                 Sou in Africa
                                                 U.S,.Wnile
                                                 Aystfta
                                                 Chile
                                                 Canada
                                                 New Zealand
                                                 GarfrWrty, F.R.
                                                 Japan
                                                 Australia
                                                 Italy
                                                 Belgium
                                                 Finland
                                                 France
                                                 Netherlands
                                                 Switzerland
                                                 Norway
                                                 Portugal
                                              80
                                                            0   10
                                                      RATE PER 1QO.Q00
                                                         POPULATION
 Source:    Levin,  D.   (1975)   Cancer Rates  and  Risksr
 2nd.  edition,  U.S.H.E.W.  NIH  75-691.

-------
             THE UNIVERSITY  OF ROCHESTER
                                                                601 ElMWQQO AVENUE
             MEDICAL CENTER
                                                                             ™BK 14M2
             SCHOOL OF MEDICINE AND  DENTISTRY * SCHOOL OF NURSING
                              STRONG MEMORIAL HOSPITAL

DEPARTMENT OP PH AKMACOI.OGV
                                               March 27, 1986
         Dr. Daniel Byrd
         Executive Secretary
         Science Advisory Board (A-101F)
         U.S. Environmental Protection Agency
         401 M Street, S.W,
         Washington, D.C,

         Dear Dr. Byrd:

              I have prepared a list of comments on the Health Assessment Document  for
         Nickel.  These comments relate specifically to sections 4,2.1 - 4.4  of the
         document.  Evaluation of the remaining portion of section 4 is being
         conducted by Dr, Qberdoerstei:.


         General Comments on Sections 4.2.1 ^ 4.4

              1. In view of the controversy over the "Ni+  hypothesis", it seems
         appropriate to include a discussion of the jpetabolism of nickel compounds  to
         Ni   in tissues,  is there evidence for Ni   being a common metabolite of
         nickel compounds?  Reference is made to this possibility in section  4,2.2.
         (P3) where the conversion of nickel carbonyl to Ni   is mentioned.

              2. when possible, the elimination half times should be stated precisely,
         as these values are very useful,

              3. The expressions "clearance", "retention" and "intake" are used
         incorrectly in places.

              4. A uniform set of units should be selected to express concentrations
         of nickel.
         Specific Comments OT Sections  4.2.1 - 4.4 :

         4.2.1 Nickel in Blood i

              1. Pl, 1.1 s  The meaning of the statement "Blood is the main vehicle
         for transport of absorbed nickel." is not at all clear.  How is the  term
         "absorbed nickel" defined?  Is nickel that enters the epithelial cells of the
         intestine by crossing the mucosal weiabrane considered to have been absorbed?
         Is it meant that blood is the  predominant route by which nickel is
         distributed to other tissues (e.g. rather than lywph)?  Can the relationship

-------
between the concentration of nickel in blood or serum and the body burden of
nickel be stated in precise terms?

     2- ?2j_ LJ, :  Change "no" to "not".

     3- p3> 1.-2 :  Change "clearance" to "elimination".  The term "clearance"
has a very specific meaning.  It refers to the volume of a compartment from
which a substance is removed over a given length of time.  Thus, the units of
clearance are volume per unit of time (e.g. ml/tain).  It should not be
confused with rates of elimination, rate constants for elimination processes
or with half-times for elimination.  There are several places in the text
where this discrepancy occurs.  It is a minor point, however, for
consistency, a uniform terminology should be used,

     3. P4, 1.1 ;  Change "clearance of nickel" to "kinetics of elimination
of nickel".

     4*« H/ iil '  Can the elimination half-times be stated precisely here?

     5. P6, 1,3 ;.  Change "may be the factor..." to "may be an important
factor in the transfer of nickel from blood to other tissues.".  Blood is a
tissue.

     6. P7, 1.7 :  Change "histidine may serve to..." to "histidine may
facilitate the movement of nickel from the serum into other tissues.".

     7• Mi Iii '  Change "labeled nickel" to "radiolabeled nickel".

     8« PS? 1-3 :  What is alpha-2-nickolplasmin?  Is this the same as
nickelplasmiiFTp. 4-13, 1.30)?

     9. P9, 1.2 j  Change "which reflect" to " that my reflect".  Is this
range of TbTnding of nickel to serum albumin the result of differences in
binding characteristics of serum albumin or does it reflect differences in
the binding characteristics or amounts of other binding ligands in the serum
(e.g. nickelplasnin, histidine)?
                            !

4.2.1*1. Tissue Distribution of Nickel;  Human Studies

     1, P3, 1.11 ;  Change "that the element..." to "that the accumulation of
nickel does not increase with increasing age.".

     2* E3, 1^.18 *  Change "ppm" to "ug/g".  It would help if similar units
for concentration were used throughout the document.

     3- ?6,_ 1.5 ?  Does this mean that accumulation of nickel in lung, but
not other tissues, increases with increasing age?  If so why not state this
precisely?


4.2.2.2 Animal Studies ;

-------
     1* ffv 1.3  :  What is meant by the expression "elevated, rapidly cleared
levels of nickel"?  is it meant that the levels of nickel in these tissues
were elevated in animals that inhaled nickel carbonyl vapor and that the
accumulated nickel was eliminated quickly?

     2. P3, 1.5  t  The basis for this presumption is not clear,

     3. P3, 1.9  :  is this pathway of metabolism of nickel carbonyl a    ,
hypothesIs~oFTias it been established?  it is highly relevant to the "Ni
hypothesis" that nickel compounds can undergo conversion to the divalent ion
in tissues.  Also change "erythrocytes and tissues" to "erythrocytes and
other tissues".

     4- P6.A 1,.2  :  Change "metal transport protein" to "metal binding
protein",  A role for metallothionein in the "transport" of metals has not
been established.  This is to distinguish metallothionein from other proteins
(e.g. transferrin) that have been shown to function in the transport of
metals in the extracellular or intracellular compartments.

     5. P6, 1,4  :  Convert the units "mmol/Kg" to "ug/Kg" for consistency
with other notations.

     6. P7, 1,10,12,13 :  Convert these values to a consistent set of units.

     7. BB :  What mechanism could regulate nickel intake?  Is it meant that
absorption of dietary nickel Erott the gastrointestinal tract is regulated or
that the levels of nickel in tissue are maintained constant as the amount of
nickel in the diet is increased?
4.JU3 Subcellular distribution of Nickel :

     PS, 1.9 ;  it is not at all clear why endocytosis would deliver
insoluble nickel adjacent to the nucleus.  What is meant by this statement?


4.3 Retention and Excretion of Nickel in Kan and Animals :

     P2  j  Is the term "retention half-time" synonyaous with "elimination
half-time"?  What is meant by the term "retention rate"? Does this mean that
30% of the ingested nickel (400 ug Ni/day) is absorbed from the
gastrointestinal tract?  If so, the elimination half-time o£ 1200 days seems
to be very high.  What is menat by the term "retention time"?

     F3» JUB :  What is meant by the term "daily intake retention figure"?
The information that is discussed in P2 and P3 are very important and need to
be presented very clearly by using precise terminology to describe the
elimination Kinetics.

     p§  :  Change "major clearance route" (1.1) to "major excretory route1**
&lso,~Tt is-not clear why variations in the concentration of nickel in urine

-------
are the result of analytical limitations.  The "urine  flow rate  vries 50 to
100 fold as function of body water content and diet.   If  the  rate of urinary
excretion of nickel was relatively constant  in humans of  agiven sise or age,
the concentration of nickel in urine would also vary  by this  amount.

     P6, 1»1 j  Change "clearance route" to  "excretory route".

     P6, 1*8 :  Change "clearance" to  "excretion".

     P8, 1.2 :  Is the measurement of  nickel in hair  a useful indicator of
the body burden of nickel or is it not?

     j?9_r 1,1 ;  Change "clearance" to  "excretion" or  "eliaiination'1.

     pip, 1,2 :  Change "clearance" to "elimination".

     Pll, 1.7 !  Can the elimination half times be stated precisely?

     P12,_ 1.1 :  What is meant by  the expression "the pattern  of
labeleBPnicTtel urinary excretion"?  Also, what is i»eant by a  "ligating
moiety"?

                                       Eespeetfully,
                                          .•"? *•
                                         ^__X_   S',/'
                                       ^•t^-y
                                       Gary  L» Diamond, Ph.D.

GliD/lm

-------
              THE  UNIVERSITY OF ROCHESTER                   „

              n j. I-r*t I ,-* A •   X-xr-lk i-r-i-i-fc                 ROCHESTER, NEW YORK 14642
              MEDICAL  CENTER                 AREA CODE 716

              SCHOOL  OF MEDICINE AND  DENTISTRY • SCHOOL OF NURSING
                              STRONG  MEMORIAL HOSPITAL
DEPARTMENT OF RADIATION BIOLOGY
       AND BIOPHYSICS
                                                   May 28, 1986
        Dr. Daniel M.  Byrd,  III
        Executive secretary
        Science Advisory Board
        US EPA
        401 M Street,  S*W.
        Washington, DC   20460

        Dear Dan:

            Thank you  for your letter of May 21 with the draft committee and sub-
        committee letter reports  on  the review of the nickel IAD*  Unfortunately, I
        had overlooked an error on page two of my comments:  The formula should read:
                                    At - i (1 - .-<*)


        I apologize for this oversight and I would like to ask you to correct it
        before submitting it to the  SAB,  I have no further comments on the draft
        letter.  Your  memo and letters did reach me on fuesday, May 27,

            With kind  regards,

                                                   Sincerely yours,
                                                   GOnter Oberdocsc&rr D.V.M.,  Ph.D.
                                                   Associate Professor of Toxicology

-------
             THE UNIVERSITY  OF  ROCHESTER                  ^ ELMWOOO
             .._.—^.j^..   A i— n .-r-r-r-i                 ROCHESTER, NEW YORK 14642
             MEDICAL  CENTER                 AREA CODE 7i6

             SCHOOL OF MEDICINE  AND DENTISTRY  • SCHOOL OF NURSING
                              STRONG MEMORIAL  HOSPITAL
DEPARTMENT OF RADIATION SiOLOGY
       AND BIOPHYSICS
                                                April 23, 1986
     Dr. Daniel Byrd III
     Executive Secretary
     Science Advisory Board
     US - EPA
     410 M Street, S.W.
     Washington, D.C.  20460

     Dear Dan:    -  •_

         Please accept my apologies  for  sending these comments on the Ni  document  -
     as requested by you and Bernie  Weiss - so late.  I hope the people working on
     the document can use my comments, if they or you need additional
     clarification,  please contact me.
                                                Sincerely yours.
                                                Gunter OberdSrster,  D.V.H.,  Ph.D.
                                                Associate Professor  of Toxicology
     GO/jh

     Attachment

-------
                   COMMENTS ON NICKEL HAD (<32nter Oberdorster)
Chapter 4;  Metabolism.

    It may be useful to include a general diagram of the metabolic model of
Ni-kinetics.  I have attempted a draft of such a model, and if possible, EPA
could put some number of transfer rates into this model.  It gives the  reader
a quick and general idea of major metabolic pathways and points out the target
sites.

Specific comments:

    p. 4-1:  relative solubilities of Ni-compounds in biologically relevant
media are discussed.  While these are useful to have, they may not as easily
be useable for predicting in vivo, elimination rates as it is stated on  page
4-1.  It is known, for example, that insoluble CdO is rapidly solubilized  in
the lung, and probably the most important mechanism for this is the
solubilization in the lysosoraes of the alveolar macrophages.  This was
recently shown by Lundborg _et_ _al_. (1984, 1985) for Mn02f and the  low pa of
the lysosome (pB4) may account for this.  Thus, more emphasis should be placed
on the in vivo solubility; in addition, particle size plays an important role
in this process, for 2 reasons:  (1) the phagocytosis is dependent on the
particle size, (2) the solubilization rate is slower for larger particles.

    p. 4-4:  The conclusion drawn from the Wehner and Craig study (1972) -
that absorption of NiO in the period of 45 days was negligible -  should be
judged under the experimental conditions:  the concentrations being used were
2-160 mg/m3, and this may lead to a decrease in lung clearance of particles
and possibly also of HiO particles.  For example, we showed that  exposure  for
several weeks to 50 pg/»3 Of Nio - a very low concentration - led to a
highly significant decrease of the lung clearance of particles in the rat
(OberdSrster et_ _ajU, 1980, ins  Nickel Toxicology, p. 125, Academic Press}.
We also showed that lung retention of inhaled NiO has a half time of 36 days.
this could possibly be included in the document.

    p. 4-5s  The term "dose-lung deposition relationship* (line 1) is
unclear.  Is it exposure - lung deposition relationship?

    On the following pages, the terms clearance rate - which is the amount
being eliminated per unit time - and retention half-time - which  is the time
during which 50 percent of the initial amount is eliminated using a
monoexpoential model -are used indiscriminately such as "clearance half-time*
or even "clearance rate half-time.*  This should be changed.

    p. 4-7:  Clearance rate is given as 72 hours  (last paragraph); it should
be retention half time.  This number was derived from urinary excretion, based
on the Corvaiio and aiemer study (1982).  However, the Ni compounds used  in
the 2 studies were different, and it can be expected that intratracheally
instilled "moderately* soluble Ni-carbonate will be cleared to a  large  extent
into the Gl-tract and is then excreted through the feces.  Therefore, the  T
1/2 of 72 hours is in all likelihood lower.

-------
Page Two

    p. 4-8:  First paragraph;  It can be expected that accumulation of NiO in
the lymph-nodes of the lung will be seen when the lung is overloaded.  This is
not spegifio foe Nio, but for any particle, as has been shown for Ti02 by
Ferin et al. (1980) when the lung was burdened with too much TiQj.  What is
"retro-ciliary" removal?  {second paragraph) -probably meant is mueo-eiliary.
I don't see that a half time of 34 and 21 days for dissolution of Ki
subsulf ide in vitro is "roughly equivalent* to an in vivo retention T 1/2 of
12 days,  it is not, and it shows - as mentioned earlier - that in vit.ro.
dissolution rates cannot simply be extrapolated to in vivo dissolution rates.
In this context, it might be mentioned again that the rate of solubiliiation
is of importance in addition to the solubility per se.

Chapter 8.3.  Risk estimates based on animal studies.

    p. 8-156;  I am not certain - and it needs some better justification - how
the duration of the experiment (Le) should be included in the formula at the
bottom of the page.  The dose retained in the lung over an exposure period
depends very much on the biological half time of the compound; Cor example, if
f 1/2 is longer* than the total dose (expressed here by total exposure) it is
very much different than it would be if the T 1/2 is shorter.  So, I believe,
the retention of Ni in the lung should be included in the formula.  This could
be done according to
where At is the amount retained at time t, a is the amount being deposited
each day and b is the elimination rate (b =   . ,.).


    p. 8-157!  first paragraph deals with a completely water-soluble gas or an
aerosol and a poorly water-soluble gas.  Ni carbonyl is insoluble in water,
its T 1/2 in air is only about 100 seconds.  To which category does it belong
here?  Aerosols are not absorbed proportionally to the amount of air breathed
in, factors like particle size will affect deposition and subsequent
absorption.  Therefore, Nij 32 cannot reasonably be expected to be
absorbed (?)  (probably meant - deposited) proportionally to breathing rate*
The distinction between "absorption" (-leading to an effective dose) and
"deposition* (resulting in a deposited dose) of an aerosol should clearly be
made.  (I attach a copy of pages 8-157 and 8-158, from which some handwritten
suggestions for changes can be taken).
    p. 8-161;  Line 2 - it should be 970 yg/JG^ nickel subsulfide.  Likewise,
in the table on this page the compound is nickel subsalf ide.

    p. 8-162:  Calculation of equivalent lifetime continuous exposure  (top  of
page) for calculation of equivalent human dosage cannot be done without taking
into account and knowing retention of inhaled Ni in rats and humans  {see
comments for p. 8-156).

-------
Page Three


    A few general remarks on the extrapolation of experimental animal data to
humans seem in place. In this and other Health Assessment Documents, body
surface area is used to correlate delivered doses in humans and animals.  It
would be more appropriate to use lung surface area, more specifically
bronchial and alveolar surface area in rats and humans for calculating a
"surface area dose,1  Enough information on lung morphometry is available to
attempt those calculations.  There is probably a difference in tumor sites
between the animals and humans, namely bronchial cancer in Hi-exposed workers
and possibly more peripheral tumors in the rats.  The significance of this
difference for extrapolating from animal studies to human could be pointed
out, i.e., stating that such extrapolations are not well supported and will
limit an estimation of tumor risk for humans derived from animal studies.  The
clearance of inhaled dust in the bronchial region is normally very fast,
unless bronchial clearance mechanisms are impaired.  This could happen during
a high exposure situation or - at lower exposure concentrations - when there
is a toxic effect on bronchial clearance mechanisms,  it is also possible to
calculate the deposited alveolar dose in humans and compare it to a calculated
bronchial dose.*  This then could be compared to the respective doses in
animals for the same particle size*  The result will show that bronchial and
alveolar doses are different in man and animals after inhalation of the same
particle size.  Prom such calculations a particle size for animal studies can
be estimated that would approximate the human lung dose and that should be
used in planned animal studies.

-------
         NICKEL METABOLISM
Skin Deposition
Inhalation
Inqestion
                     Lung
                  Gl-tract
                    Blood
              Liver
                          jr
                              System I
            Urine
                                       Feces
           Critical or target organs
                                    4,

-------