. '•"'••:'
UNJTEO STATeS;£Nyj*ONMeNTAL''PROTECTION AGENCY
. . '.WASHINGTON. D c >04«Q. .•••.'
J.rov«Bb«r 28,, 1989
Mr. William Eeilly ; ' ' '
Administrator
US Environmental Protection Agency
Washington, D.C. 20410
Dear Mr. Reillys
In 1988 the Agency asked the Science Advisory Board to
review two document*; "A cancer Risk-specific Cose Estimate for
2,3,7,a~TCDDtt and "Estimating Exposue to 2,3,f,S-TCDO". In
response, the SAS Executive' Committee- (1C); appointed a "Dioxin"
Panel, co-chaired by Dr. , Bernard Goldstein of the Robert wood
Johnson Medical School' and Dr. Nancy Kim of the New York stare
Department of Health,. The Panel conducted a public meeting in
November, 1938, to review the document*. On 'October 24, 1989, Dr.
Kim presented the attached report to the EC, • which was approved by
the EC. . • _ . '
The EC believes 'that.the "Dioxin" Panel has done an
outstanding job of reviewing two very fine.Agency documents.
Both the Panel and the Agency are to be .congratulated on their
efforts, neither' of which was easy, given the lack of scientific
consensus and the heightened public concern surrounding the
topic, - •-fl-
it should b* noted that th« "Dioxin" Panel was not asked.
nor did it choose, to address d.i-'
-------
, Th«\ EC .'.conclude* .that- 'the - existinf t! •• U«S-ls4*«d;;.i'riaH:- •;•. •'/.-. - .' ; ."
assessment foir 2,3,7, 8-TCOO lacks a fina- scientific foundation. '
How* v«r, until th« n*w approach** ar* fully d«v«lop«d and ; -
peer-previewed, estimates ftamed on othar mcd«l» ar« «qually ,
quest ion'ablev-- -unfortunately^- thf •ji'ir«ction '«iid"«xt of ;.«iy:
change from th« LMS-bas«d rislc'estimat* that might r«ault from
the application of mor* appropriat* models cannot b« dfttenain«d
at this time."' ' '•• • • ' ';•.-"'• . '':• , "". "••'.^ • • •'
The Panal was favorably impressed by th* exposure document.
They" found that '.indirect, routes of exposure, via food 'Chain
contamination,' result in higher exposures to the general
population than 'do' direct environmental exposures.' 'This
exposure model-based conclusion 'has- 'important implications and
should, be tested -by comparing predictions • with environmental
measurements , • . "
The SAB appreciates the 'opportunity of addressing these
inportant, but '-difficult, questions. -We look forward to the
Agency's formal response to1 this review,
Sincerely,
Ph . D .
Chair, Executive Committee
Science Advisory Board
-------
•A)
\^
November 28, 1989
, . .
Dr. Raymond C. • LoeHr- ••' . .-'-',
Chairman ' ''••'•.' • ' '
Executive Committee • •. • ' ••
Science Advisory loard (A-loiF) .
U.S. Environmental Protection Agency '
401 -M Street, S.W. • "
Washington, D.C, 20460
Subject: Science Advisory loard'a Dioxin Panel review o>f
documents from the Office of.Research and'Development relating to
the risk and exposure' assessment of. 2,3,7,s~TCOD,
Dear Dr. Loehr, . "•.'.. .
The Ad Hoc Panel on Dioxin of the science. Advisory Board's
Executive Committee met in Washington, DC on November 29-30, :;-f •
to review two draft documents on 2,3»?,8-TGDD: "A Cancer Risk-
Specific Dose Estimate for 2,3,7,S-1TCDD11' (including appendices^
and "Estimating Exposure to 2,3»"7»a-fCDD", .
The Panel concluded that both documents were carefully
constructed/well written, and represented a significant effcr1:
on the part of EPA to document and explain how the hazard and
exposure scientific data have been analyzed. The Panel in
particular commends the EPA Oioxin Working Group fothaving
alertly initiated an analysis of the extent to whish new
scientific information concerning 2,3,7,8-TCDD can b« factorea
into risk assessment.' . .^»-v . -,
The Panel's review focused on the scientific validity of. *-»
statements and judgments in the documents as they, were writte-.
The Panel has not:reviewed the process by which EPA arrived 31
the judgments expressed in these documental nor -has it concer-*J
itself with the degre« of unanimity of the EPA working Group. ;
specific written chatge from EPA or the' EPA science Advisory
Board was received by the Panel.
The'Panel has taken into account the oral comments made -,
the public during the course of th« committee meeting. It r.im.
-------
t- •.'..'•• , ..^•.•••••?-^\4>.<<-lW-V-*^»^^^ 'TVC;.",:•*« Vf.< • '>* "2*
the Novemiser.:^9-3Q^;iij^Hi^ • ••.
siibaission* were^tece'iy^^ •" '.- •
Panel member*. •'.•-• •• •%1^;:-^:^ .-,'•..'" •
' : Understanding the biological effects of the dioxin« presents
one of the most intriguing challenges in modern biology. ; \ -
Research' in this' area' has been particularly exciting for 'the. • .' .
insights,it• has•given into.exposure assessment, including the *
role' of bioavailability, and for its potential for linking •
receptor interactions with modern toxicology, including the
mechanism of carcinogenesis. While there was extensive
discussion of these active research areas, the Panel did not lose
sight of the fact that it had ;a relatively prosaic task: the
provision of advice concerning the scientific adequacy of
documents. • .*•'•.••• • . •. • . '. • ' -
The document "A cancer Risk specific Estimate for 2,3,7,8-
TCDD" contains the background and rationale for the EPA Working
Group's new risk specific 'dose for 2,3,7,8-TCDD, in order to
help focus its review, the. Panel -chose to summarize -the cancer
Risk document as containing five key points* These points, and
the Panel's response, are as follows:
1. The Panel generally agrees with the EPA Working Group1*
criticism of the Linearized Multistage Model* While
there are promising alternative models which may be
expected to more accurately reflect the biological
basis of 2,3,7,8-TCDD carcinogenesis, such newer models
need to be further developed and validated.
2. The Panel agrees that since the determination by EPA of
a risk specific dose of 0.006 pg/kg/day in 1985, no n**
'information has.appeared to permit reevaluation of tfie
risk specific dose through the use of the standard
• Linearized Multistage Model approach .{i.e.. there are
•• • neither new. long -term animal studies nor epidemiologic
'studies which appropriately could be used to •. ^
recalculate the risk specific dose using the previous
• model). . , ,;,
3. The Panel fully agrees that a series of important%nd
exciting mechanistically oriented studies have
much new insight into the toxicological effects of
2,3,7,8-TCDD and related compounds, and that such
information is likely to be of major significance to
regulatoryDecisions concerning 2*3,7,8-TCDD. There
has been a parallel increase in the understanding of
cancer biology. The Panel commends.EPA for reviewir
-------
'.; ' ^•r^ms^^^'^i^^^li^^ci^ai^^
-' '• '•-V.". '';.'"•.;;!'-'": .•i'i.l'-;-'.' '>:'"-•'•';-•.•.!. ;';^.-V,;-':; •''';-;J:^v..;';r;""'tj?:^':-^v'^'>V.^-":-:;1"v:-.•:='',>'
4 .
. .
. conrtrflijig.v2V3V7fjHF^ 1985
.Hs.k^spe.eific;;do's*:,of ^i^idf'-'pg/itg/aiiy; -TUB panel did -
^ '•
'''' ''''' '' ' ' '
- , ..-5.. ' -..'/AS7' there; is '^lio;y«as6Il'^^vn%c4ssariiy^fceliev«•.ti^at a new'
-- ' nechanistt model would lead to a relaxation of -the risX
" . ' v ;sp!ec.if ic , dose •. for. ;2> 3^r7.f'8-TCbO-indue«d ' cancer, ' in th«
\ \ absence of a validated model suitable to recalculat* a '
• risk specific .dose,,; ''the -'Panel docs not concur with th«
EPA Workiftf Group's "science policy" decision to. change
' , • : the risX specific dose. Further, the Panel agrees with
" ' • the EPA document that there is no specific scientific
.pathway that would allow determination of the extent' to
''which such 'a change 'might occur. 'The 'Panel 'therefore
finds no scientific basis at -this time for the proposed
change 'in 'risk, specific dose for 'the causation of •
, ' cancer by 2,3>7,8-TCDD.- ' ' "
The Panel thus concluded that at the present time the
important new scientific evidence ' about 2,3,7, 8-TCDD does not
compel a change . in ' the current" -assessment of the carcinogenic
risk of 2, 3, 7, 8-TCDD to humans. EPA may for policy 'reasons' set a
different 'risk. specific: dose number 'for. the cancer risk of
2,3,7 , 8-TCDD, but the Panel . finds no scientific basis for .such a
change at this- time. The Panel does not exclude the possibility
that the actual risks of dioxin-induced cancer for humans may b«
less than or greater than those currently -estimated by the Agency
'using a linear extrapolation approach.. '
' EPA is strongly urged to 'build upon their excellent review
of the new scientific data ''relevant to, i,3,7f 8-TCDD
carcinogenesis. by 'moving rapidly to develop and -validate a new
risk model capable of more accurately estimating the risk of
human cancer caused' by the dioxins and "related compounds . Other
recommendations by the Panel, including a need for. further focus
on reproductive effects, and for enhanced use of exposure data in ^
epidemiologic studies, are detailed in the Panel report which is -I
attached . . -
The Panel also urges EPA to remain abreast of advances in
our knowledge of 2 ,3, 7, 8-TCDD, particularly at the mechanistic
level, and to continue oversight of relevant epidemiological
findings as these 'become .available in the peer-reviewed
literature. '
-------
....- . , • . . ...-. , ,
..r
-------
>. . . ..
Exposure and -for • cancer 'and -non-'
cancf r :'riisk:dv6htsy'*can 'and should 'be accomplished In the
>rela'tiv*ly':';niear^;f«tiire-lsy-; pursuing an. active 'research program in
this ''•''' '' " ' •' ''
We ' appreciate • the -opportunity to conduct this particular
' ' ' ' ' ' " '
Sincerely
Bernard D. Goldstein
Chairman •
Ad-Hoc Dioxin Panel
Nancy Kfm
Co Chair
Ad Hoc Dioxin Panel
-------
D 'SrATESENVlRON'MENTAL PROTECTION AGENCY
".'•.'; .VASHING'TCV D'C •^^•jO "'''.
-• !390
THE
Raymond C. Loehr, Ph.D,
Chair, Executive Committee
Science Advisory Board
U.S. Environmental Protection Agency
Washington, D,C. 204SQ
Dear Ray:
Thank you for your November 28, 1989, letter which forwarded
the^Seience Advisory Board (SAB) Dioxin Panel's review of two
documents from the Office of Research and Development (QRD)
relating to risk and exposure assessment of 2,3,7,8-tetrachloro-
dibenzo-fi-dioxin (2,3,7,8-TCDD). The Agency appreciates the -crk
of both the Dioxin Panel and the Executive Committee in carrying
out their respective reviews and providing their insightful
comments. ' •
I was very pleased to' see the extent to which 'the SAB
reviewers agreed with the EPA working group on many of the
difficult,issues presented by the 2,3,7,i-TCDD data base. The
discussion by the SAB of'the science supporting the lisk Spec;:;r
Dose (RsD) for 2,3,7,8-TCDD was especially helpful. The cor.c;-^j
views of the EPA working group and the SAB reviewers have cleir;/
defined for the agency the areas of scientific consensus and --•?
areas where such a consensus has not yet "been achieved.
The comments of the Panel on the document entitled, "A
Cancer Risk Specific Estimate for 2,3,7,8-1000," 'have been
forwarded to th* authors of that document. In addition,
discussions'have already begun within the Agency's Risk
Assessment Council both to. address work necessary to develop
alternative approaches to modeling potential cancer risk
associated wit* 2,3,7,8-TCDD exposure, as suggested by the FI--?.
and to consider the Panel's comments on the science policy
approach, to selecting a risk-specific dose for 2»3,7f8-TCDD i.
presented in the Agency's draft document.
-------
The comments of the Panel on the document entitled,
"Estimating Exposure to 2,3,7,3-TC0D,M have been-', forwarded
to the Office of Health and Environmental Assessment' in ORD
for consideration' daring -revision- of1 .that'document. The comple-
mentary nature'-and' important substance of these comments will
encourage Agency staff to maJce'significant improvements'in this
already veil-received, document. '• - .<'••••
I will provide a more detailed response on the.Agency's
disposition of your comments in the next few months as the Risk
Assessment Co.uncil and ORD bring their deliberations on these
documents to a Close*' Thank you again for your efforts in
reviewing these two documents, *
Sin
William K. Reilly
< *
\ m,
-------
CD A 0;S. Eiivifemriiintal Wtthlngton, DC?
CUM- i : Prot«ctkm Ag»ncy ' ••; " EPA-SAB-EC-90-003
Report of the ad hoc Dioxin Panel
of the Science Advisory Board
Review of Draft Documents: "A Cancer
Risk-Specific Dose Estimate for 2,3,7,8
TCDD" and "Estimating Exposure to
2,3,7,8 tCOD"
A SCIENCE ADVISORY BOARD REPORT NOVEMBER Tili
-------
....
Science Advisory Board
Ad Hoc Dioxin Panel '
REVIEW'- OF DRAFT; ,
"A CANCER RISK-SFECIPIC''ix}Sl^'ESTiMAfE''FOR
AMD • " '
"ESTIMATING EXtOSTOS TO 2 , 3 , 7 , 8-TC0D"
-------
"• ••
3.2
. . .
3.1. 3 Mechanistically Or ient«d Re»*arch •
3*1.4 Relaxation of the Rimk Specif ic Dose for
. TCDD Carcinogenesis . ' • ' . .
3.1.5 Chang« of the Riak Specific Dose for
TCDO catcinog«n«sis , ,'' _ "•••'' -
Comments on -Specif ic- Issues ' >. ' .
S.2-,1 Mechanisms •.•'•. • • . •
2.2 Research' Needs; ' . •
2.3 Development and Selection of a Model
toxicology of Related Compounds
'''
2
2
4.0
2.6' Pharnacokinetics : • • ;
2*7 Keproduetion and .Immuno-Toxicity
Conunents on EPA's Draft Document "'Estimating
Exposure to 2,3,7, 8-fGDD"
6
8
8
10
10
11
11
12
4.1
4.2
overall comments .'-on IPA's Draft Exposure' Document
Comment's on Specific .Issues
4.2.1' Use. of " the Best. 'Data • .
4.2.2 Pathways of 'Exposure
4,2.3 Indirect Exposure ' • ,
' 4.2.4 Scenarios
4.2.S Exposure and Epidemiology
4 ..2 ; 6 . Estimating Exposure in the Absence
• 'of Empirical Data ' '
4.2.7 , Rely .More on . Field ' Sampling . .Data and '
•• ' Less on Dispersion Modeling Results
4.2.8 . Verifiable' Exposure "Estimates
4.2.9 Incorporation , of the. 'Most Reiable .
1 Exposure Assessment Information
4.2.10 Rigorot**.; characterization of
Uncertainty:' . ' •
4 . 2 . 11 Phanucokinetics
12
13
13
13
14
14
IS
16
1?
IS
It
'IS
It
-------
...... ....... ^.....-.,;"I988'':;a'n;:A«I;''lielc^finiel^ief;(iici«ati»t8-was:
•convened,by' the'EPA Science Advisory Board to review'the•June, .
1988 'external\reyieVdz»fts"of Vm^*tries:'':'Of'.i^^^ocuments." ''' • ' •
concerning 2,3,7,p-Tptrachloro^ *A Cancer
.
•Appendices)' prepare'd'/'fcy:;-'aii::drit^^^ by
Dr Peter Preuss of th« Office of Research and Development; and
"Estimating Exposure ;to 2t3t'f,B-'TCQV« prepared toy the Exposure
Assessment Group under the direction of Dr. Michael Callahan;
The Panel, chaired by Dr. Bernard Goldstein, functioned in,
part as two separate, subpanels, one of ''which /focused on the
cancer risk document and the other, chaired by Dr* Nancy Kim, on
the exposure document* During part of the time the subpanels met
separately. At other times a plenary session was held at which
each subpanel had the opportunity to offer comments on both of
the documents.' However, responsibility for-the-written critique
of the two'documents.was independently assigned to each of the
subpanels. '
The Panel's review focused on the scientific validity of the
statements and judgments within the documents.. ' To a lesser
extent, the Panel also reviewed the appendices as free standing
scientific documents.' The review of 'these appendices took into
account EPA statements' that, the appendices served as background
documents'and did not-necessarily represent the EPA'Working
Group's collective judgment as to the 'science.
Understanding the biological effects of 'the dioxins presents
one of the most intriguing challenges in modern biology.
Research In this area has been .particularly'exciting for the
•insights it has given into exposure assessment! including the
role of bioavailability, and in its potential for linking
receptor interactions with modern toxicology, including the
mechanism of carcinogenesis. The /anel strongly commends EPA Cor
initiating, a review of this info .-jiwtion and' for considering the
implications of this'new data to .'issues central to the regulation
of TCDD. We recommend that EPA's^ Valuation'of the*.risks of-. • •
dioxin be an' ongoing process as new 'data on hazard and exposure
are published.• • ' '
3,0 comments on EPA•» Draft Document "A cancer Risfc-Sp«cific
Dose Estimate f**** f, fr. 7. S-TCDP"
The-'Subpanel reviewing the cancer risk document consist«d of
* •* ' ' 3
-------
w^^"^^-^:^^^^
!£?:'2,f~''''t"•''"^:' "•:"**. i, •"'•'. , • ' -rH''/^jJ>"^^
!•«••"'*:•!'''.'.i.i* ' •".• '..-I-. "-'-"^ •.."'•.• , ••.. * . •
iDjr||l«j^sd!«^ .
.ME^I^^^^
"••y|rv:| s document makes it clear that there is no
specific scientific pathway that will allow
determination of the extent to which such * change
should occur.
-------
.•agrees- :
th«r* ar« many probl*MB inherent in using th* U4S model for
TCDD, agr««« that thera is no currently availabl* 'alternate- •-
validated •»eKlil/agr««s that there ar«' no new studiea which. could
'be used to recalculat* tho risk sp«citic dose with the IMS model,
-and agrees that there in exciting new data about the mechanism of
action, of; TCDD that pbinti toward; .the.. development of a ' new model
suitable for estimating the risk specific dose for TCDD
carcinogenesis. However, the Panel does net agree that this'new
information necessarily leads to a belief, that the 1985 RsD of
p.006 "pg/kg/d'ay is t?>o gtfinfent. This is the crucial point of
disagreement with the EPA Working Group. Iks the Panel does not
concur with the assumption; .that a new mechanism model of TCDD
carcinogenesis would necessarily result in a relaxation of the
risk specific dose, it can not possibly concur with the proposed
change from the isSSSsD of 0.006 pf/kg/day to Q.I pg/kg/day.
^; ' *' • *. ' * • .
The Panel's response to each of these points is described in
more detail below: •''•• ' ;
3.1.1 The Adequacy; of j:heLinearized Multistage m.odel
The Panel agrees with the general thrust of the EPA working
Group's critique of the Linearized Multistage Model (see page
21), but would make the point more.forcefully. In particular, the
Committee believes that there'are a number of models, in addition
to the Linearized Multistage Model which incorporate the concept
of low dose linearity. Some of these can also account for TCDD's
promoting effects on the selective clonal expansion of
preneoplastic foci. Further development and validation of these
alternatives to the" IMS model are needed and encouraged.
The Panel does.agree that no 'validated model is available
to'day to replace the LHS for calculation of an RsD for TCDD
carcinogenesis. '
3.1.2' The Availability, of ffev Infomatien For Use in the LM$
Modal '•''.-'.
The Panel concurs that no new epidemiological or long ten
animal studies aril available that would appropriately, allow
recalculation of the RsD using the linearized multistage model.
However, the Panel wishes to emphasize the likelihood that
epidemiological data, may well become available in tty& near •
future, particularly if appropriate attention is p&M to exposure
and body burden,
3.1,3 Mechanist.ieg^ly_.0riented Research
. The Panel agreef*that there is-much exciting new information
-------
^^^9^^^^^ji^^^i^^^}2-tS;^9^cxa^ carcinogenieity • which'
•• ,;;p;e*|&ni^ It "-commends, the Working - •'•
Group for »ddr«ssing this issue and 'for its. careful and well*
- . writteipr^Af&i.Qf V*h*;.'pta> jipetV ; ,; In ' agreement • with'-' th*'.- EPA •
' .• Working ':-*Sv••..' ; • • '; '•• ' •' : '.
>;;3U1.4 Relaxation >of the aisle' Specific Dose .for "TCDD
••' i , . Carcinoaeneais •.'.' ' . • •.'•;. •
A highly significant area of disagreement with the SPA
Working Group is the Panel1* failure to concur with the
contention that .a. new nodal of TCDD carcinogenesis would
necessarily lead to a relaxation of the existing RsD. Some
member's of the Committee were of- the opinion that a -receptor-
based model, could eventually support EPA's contention that the
Linearized Multistage model overestimates the upper confidence
limit for the carcinogenicity of TCDD (i.e, .006 is too
stringent) . other committee members' felt that it was not
possible at present to predict whether the receptor-based model
would lead to a less or more stringent RsD.
3.1.5 hange of the is • Spcife Pose ............ fjor.. JC-PD Carcineqepes is
. The Panel agrees with the EPA Working Group that there is
currently no specific scientific pathway -that would' allow
determination of the. extent of change in the existing RsD for
TCDD carcinogenesis. Accordingly,, the Panel believes that at
present there is no firm scientific basis for a change. In
particular, as the' t Panel does not support the contention that a
new model would necessarily result in a reduction in the
stringency of the RsD, the. Panel thus concludes that at the
present time the important new scientific evidence about TCOD
does not compel a change in the current assessment of the
carcinogenic risk of dioxin to humans 'by the existing approach.
EPA may for policy reasons set a different RsD for TCDD
carcinogenesis, but the Panel finds 'no scientific basis for such
a change at this time* ' The Panel does not exclude the *•*••
possibility that the actual risks of dioxin- induced camsr for
humans may be. less than or ' greater than those currently vestiaaced
by the. Agency using, a linear extrapolation approach. 'V-
3. 2. Comments on Ssgcig,ie |.sjsueg
3-2.1 Mechanisms.
While there Is still considerable uncertainty as to the
6
-------
: invpl v*;/ ; • • '?
-the :J>ahelT = '• '; ^T:>'';'
;:6oncli(i^s^tliSli .,£t 'is., likely 'that 'receptor- mediated, mechanism* '*'
pl1iy;'ii.''Wl*-.;in:'.-tlii* process as,in''''many;"of th* biological effoust*-
';o£ ;'TC13l?$';:>:lii'ta3R(l. r«ebgiiitibn» reeeptor-1'if and"' binding,' •• ••••''• ''
,sBtabii;ia[atio».'.of'-..th*- r*ceptp^ligand-''eofl^>l fCDD increases the. effective concentration of Ah
receptor '
o Estradiol increases the concentration of estrogen
. receptor •.
o Glucoeorticoids increase the concentration of '
glucocorticoid receptor
d) The biological activity of hormone receptors is
activated with only partial occupancy of the hormone
responsive element on DNA (about 20%) and this ia
similar to the Ah receptor
e) TCDD exposure may decrease 'the incidence of hormone-
related tumors in rodents, which further supports m
'mediated role for earcinogenesis.
f) TCDD modulates a variety of growth regulatory genes,
including the estrogen and glucocorticoid receptors.
Modeling of .receptor mediated event (s) follows the classic
. ' 7
-------
.
'
:Xf-,'thi*' type ,of modeling i* /conducted^ it 'iiaay • i*ad .to ' . •"'
cdnclusloi* by *ome that }a v threshold; :*icist»V'rIow*ver,' a 'receptor
based model will be linear at low do»«, when receptor* are not :•
saturated* A« dose increases, receptor binding will be
saturated, . thus' describing a curvilinear dose-response ' . '
relationship,' . ••• ..-.;,„• , " --1 .'''•"•"'"'..••.'.
• , It is also important to note that if TCDO is acting •
similarly'' (an agonist or antagonist) at the Ah receptor then
there, 'is likely to be partial occupancy by an "endogenous ligand"
.which means that receptor-mediated effects of TCDD may be
additive to background.
3.2.2 Research Needs. ,
The Subpanel recommends the following research:
a) more work 'understanding the biophysics of the
interaction between dioxin and. its receptor, and
specific DNA recognition sequences - rate constants,
equilibrium analyses, thermodynamics 'etc.
b) role of receptor in carcinogenesis
c) • role' of gene transcription in careinogenesis
' d) -relationship between receptor occupancy and gene
products ' . . -
e) identification and function ,of the endogenous ligand
f) • role of steroid hormones in carcinogenesis
oncogenes and 'tumor suppressor genes (effects of
glucocorticoids, estrogen and Ah receptor binding
to DMA)
growth' factors (oncogenes) and Ah receptor action
g) A 'number of models,' in addition to the LMS model
incorporate the concept of low-dose 'linearity. Some of
these can also account explicitly for TCDP's promoting
effects on the selective clonal expansion of
preneoplastic foci. -'Further development and validation
of these alternatives to the IMS model are needed and
encouraged.
3.2.3 Development anff Selection- of a Model.
Many of the toartc effects of TCDO, including tumor
promotion, appear to be receptor-mediated. Thejiioxin receptor
-------
which is: mostly tv related;:.%o; w^^^^S^iS^^m^^'^'^^
grcwtl*.:ebnttptfi(^
IGF rece^te-ty" -ipctfar' -t»y*pidriiotmon«'-ric^
receptor; .^t^*rvmppMJ^vn«tfewi!Mxy;' iifa"':ri^;i^t&ij^t':'tGr .t&V' •=-•'"'
responses^-''"fiMuei-specific* factersr'l^-a^i^cp;^-'-:.•'"'<' ;":>. '.;:'-;"
«peci«»-*i»«ei;^ics'faetorf;'"control.-the 'hEti^r^lid;-;,iai» *xt*nt-bf "•'•:•
these .receptors-mediated'1 responses. \Thu's,Vaoi|*-.:,,', ' •'
Mechanistically,, all data to date, are consistent with the
concept that most, if not all, of the biological events,
including tuaor promotion, elicited by TCOD are'mediated
initially through the stereospecific binding of TCDO to a
receptor protein and the subsequent modulation of'gene
expression. Evidence for this concept comes from 'a number of
studies; genetic, 'Structure- activity.' relations, and those
examining the molecular interaction of the TCDD-receptor'complex
with specific responsive elements upstream from identified genes.
This model is remarkably'similar'to that-described for a number
of hormonal'systems, and in fact'the Mi receptor has been
suggested to be a member of a supergene family including the
glucocorticoid, estrogen, and thyroid hormone receptors .
Based on the hypothesis that the binding of TCDD to the
receptor"may be the initial' event in the biological responses to
this compound, it may be possible to utilize receptor binding
theory, and/or actual binding kinetic data to derive a model that
estimates & dose of TCDD necessary to produce a particular
response. However, it should be recognized that there are likely
to be a number of events subsequent to receptor binding which
eventually lead to the final biological response, in'this case
tumor development or tumor suppression.' Some of these events
include binding of the ligand:receptor complex to specific DNA
sequences as well as gene transcription and translation.
Furthermore, there are' likely''to be a number of tissue* and
developmentally specific'regulatory factors, both positive and
negative, that modulate these events. Some evidence for this has
been presented, at least for the P450 gene . Receptor binding
may or may not. be the rate-limiting event in the processes
leading to tumor development.
Classical receptor theory suggests that at low doses
(concentrations) there will be a linear relationship between
ligand concentration and binding. Thus receptor binding theory
does not itself allow us to disassociate risk assessment froa a
low dose-linear model, nor does it allow us, .at this time, to
determine the'slope of a dose-response curve for TCDO.
-------
•".•-•• ....-,•••;'.••.•--..•;• *",
-•"
this tinnr'to'. 'rijle^put^tK*; jp^^
.other • mechanismsT*; •-'• rurtheraio^*,^''ithf^'
-------
. been ;:pbse;rved ;lj^t&«3^i£i*£i^^
___ .. . .
-
' ' '
except ions - •• of ' -; 2 O'^aiid}; 4$:;:j^t£^ *;' .
• Recogniz ing ' .the"' iiaportanc;* ' of ^ exposure &easur«m«nt» in human
' studies , ,' th« Subpanel suggested that . some - existing cpideroiologic
studies-, such';as':-,the Ai:r."'Fo;rs!B::^n^ ' ..;'-
useful information en health outcomes if the existing data arc ' '
reanalyzed' according to new exposure information. The NIOSH • -
mortality study of 7000 dioxin-exposed U.S. Chemical Workers and
' the- medical study of a subset of these workers should provide '
useful information for the next EPA dioxin risk assessment .
Because the evidence is that substantial exposure to Agent Orange
did not occur' on' average '.for ground troops in Vietnam, studies of
.this group probably, will 'not Contribute useful information to
dioxin risk assessments.-".' • ' . . - '
The subpanel suggested that there may be value in
considering whether -existing studies might be analyzed together
•in a met a- ana lysis. The; International Agency- for .Research on
cancer (I ARC) has organized an International Dioxin Registry to
compile data from , studies of dioxin-exposed workers conducted in
several countries in, hopes of increasing 'statistical power for
evaluation of 'rare outcomes 'in merged analyses. . Data from this
effort may contribute useful information in future risk analyses.
The Subpanel suggested, that EPA consider whether the
epidemiologic data from the Yusbo and Yucheng populations might
provide useful information for future. risk assessments, since the
effect of furans in animals are mechanistically like those' of
dioxins. " '
3.2.6 Pfaaramaeekinet ics .
, The Subpanel commended the- EPA initial .exploratory efforts
in the area of pharmaeokinetic modeling and 'encourages further
work in this area. -'..',..''''
3.2.7 Reproduction. and Immno-TOxicitv*
in the- appendices to the document, EPA has provided short
r; views of 2 other endpoints associated with TCDD, reproductive
And immunotoxicity* By definition, however, thes* endpoints . are
ttxcluded from the final discussion which is exclusively focused
.-JSP ^cancer. The Subpanel .has two comments on. -this general
strategy! (1) the mechanisms of TCDD- induced reproductive and
immunotoxicity may be relevant to understanding TCDD-induced
carcinogenesis (for instance, oncogene activation, growth.
dysregulation, disrupted cell; cell communication, inhibition of
immune surveillance) , and (2) if the science eventually supports
• 11 *
-------
• ' ' ' . • r '• .' '• . l "'••"' "'-A' •'• ..'• J."'[rt!* I Jjj%*''*:V'^ '"•> '
^^'^^y^^^^^^Ly^^^--^^^ ft. , • :,.^f-.,.. .. r^^i-^,-^
:\,de^reas^:inj^s|c|f^|^^|||^r^
bh';/c4^cer;Q(ai^ii^|;pf^SgjeiBSife^feased'.'pii:-rr'educirig:::V- ':'••'"•
. cancer.' .-eisfe.. *lon«);.'^dpei!^^tAfis|||; •.which. > might, • permit:
human' .«epP^ur««v.tb-^TCPOii:imy^^ ;•. :;,,
effect*;.' on. r«pr^u<^.ioifi^;op(:';p08i;ibi*';>iMim ' /r -.;>. .'" • ' -
'. The Subpanel r«comm«nd» that, th.« appendices on reproductiv* •
and ' ironunotoxicity be expanded to include relevant information on
related compounds (dibenzofurana and PCBs) and- that- 'the •
mechanistic information be drawn out More' specifically in' the
appendices1' ' '
-4*0 Conmenfi on SPA * s 'raf t 'Document. "Est.ffian Exosure to
Dr. Nancy Kin chaired the Dioxin Exposure Subpanel which al«o
included;' Dr. Edward Calabrese,* Or* Warren Grummet; Or. Michael
Gochfeld; Dr. Raymond Klicius; and Dr. Stephen Rappaport. Dr.
Dennis Paustenbach, a member of the Subpanel who was unable to
attend the' meeting 'also .contributed written comments on exposure
as did Dr James Falco, a scientific advisor 'to the subpanal, and
Or Linda Birnbaum, .a member of the cancer risk subpanel.
The Dioxin Exposure Subpanel 's comments on EPA's draft
"Estimating Exposure to 2,3,7,8-TCDD" appear below, where the
document is referred to as the "exposure document." Page numbers
and references are from the' exposure document, except where
otherwise noted, • " , ' '' ,
4,1 overall Comments on EPA's Draft Exposure Document,
. This update of 2,3,7,8-TCDD exposure document' is an
improvement over the earlier' assessment because it includes new
data, 'expands consideration of- pathways, and .addresses additional
phenomena. It is a credible document and is one of the better
exposure assessments. The important section on the- uncertainty
evaluation .fives the reader, an understanding about the magnitud«
•of this factor. . ' . • • • ,
•The, .-exposure document showed .that indirect routes of
exposure, that' is dietary- impacts, will usually predominate the
level, .^f; risk for -the general, population, this hypothesis «hou Id
"be valMff'£ed using 'environmental modeling and _ field measuroents.
•EPA TCDD" exposure 'assessments should focus on' indirect sources cf
TCDD.
The exposure document's .Executive Summary is excellent,
Several points from that Executive Summary should be emphasi2«i
and reiterated throughout the report. For example, when
assessing actual -sites, whatever monitoring data, are available
should be used and are preferable to the estimation methods m
. 12
-------
•'.
iio^mf ivd^^ tliafe:^|*^ • throughout " *
tho doc^iinent whenever appropriate; T^ al*6 recommends •
thit^tht'^oi!^ measuring. .'
human ''lniaj<&.:v>is. to;' monitor ;/ ;peopifc::1iiid/^^ -monitoring
should bft us«d when possible. These data will «stai3lish g«n«i a 1
•xposure and identify subsets of ,th« population (i.«. , fish
' eaters., • 'small ch ildren , :• people with pica) '.who.; -nay • have •• elevated
exposures!" .above; ''"''' ' "
4 . 2 .
: ' The Subpanel recommends that' the document distinguish
between excellent versus poor experimental work and thorough
studies versus 'cursory' examination. Not all data should be given
the same weight.- Foe example, the plant uptake data should be
reviewed with this in mind, The 'conflict between data which are'
discussed may 'be because the' radiolabelled TCOO used by cicucci •
and Sacchi (pp. 44-45} may be only 60-80% pure as reported by
Marple (p. 13). The Wipf data (p. 43) did not suffer from this
shortcoming and the 'Subpanel' recommends that these data be
reviewed to consider weighing the Wipf data ttore heavily and
revising the presentation of the Wipf data.
' The data supporting appreciable uptake of TCDD by plants are
weak. The soil-to-plant "ratio data must.be interpreted
.carefully, since this phenomenon, is dependent. on soil type and
aging. Further, there may be a dramatic reduction in the
relationship at very low 'levels such that, as the concentration
decreases, lesser amounts 'are available. to be absorbed by the
plant. The Subpanel recommends that the published work on the
eye lodienes be reviewed to evaluate if -plant uptake is likely to
be less than. estimated. • . ' .
The uptake of TCDD by grazing animals (page 5) is perhaps
the weakest discussion in the document. , Dr. George Fries, us DA,
has described an approach which may provide a good . estimate of
the magnitude of the 'exposure; due to ingestion of food from
grazing' animals. The Subpanel recommends that this approach b«
included in the i'.:,iument. • .• " . .
s - .
4,'2v2 . . Pat-nvrays....of.-.ExpQsnre. . • .
• ••
The"- Subpanel recommends that a number of specific question*
be- addressed in the final report. '
a) Are TCDD concentrations higher on small particular**
which may be poorly captured by air pollution control
' . ' devices? «^
b) What is the impact of declining performance or
13
-------
_ removal .of
^ • . • . . '' ;.
of- TCDD in vet
' •
^
pi-bportiori of the population with. high exposure (i.«. ,
'the ^limited nuabiir of children with pica eating 7g of
! "
e) Are photolysis and vapor phase diffusion, rather than
runoff , the major processes for mobilization from soil?
f) What is the impact of the environmental fate of TCDD in
soil and -fly 'ash? ' • '. .
, '• ' r • '
4 * 2 .3 . \ indirect; Exposure. ' _ • *
The modeling activities in the exposure document clearly
imply that indirect routes of .exposure (dietary impacts) appear
to predominate'.'- ''This; can be seen in 'Tables 6-6- and €-16 which
show that, in all scenarios which consider dietary sources, the
ingest ion of beef, fish, and dairy products contributed ' more than
82% of the total exposure.
The Subpanel .' makes two recommendations regarding this
important implication* . First/ the Agency should validate the
modeling' outcomes with' environmental' monitoring. Sensitivity
analyses should be conducted to' prioritize the various data gaps
prior to undertaking such monitoring studies* To improve
understanding of indirect exposure to 2,3,7,8-TCDD, it (and
perhaps its _ congeners) 'should -be '-added: to the market-basket
survey. Second, the. revised document should conclude that
exposure' assessments' should particularly focus upon indirect
sources of TCDD. •
The exposure scenarios generally concentrated on average
• estimates , ' ' The ' Subpanel recommends that the revised document
discuss the potential for two different types of situations which
'could lead to much highec^e-xposures than the average. One
situation is illustrate l;by the pica child or adult. Average
soil ingcstion level's ven-e used in estimating soil exposures, but
this subpopulation may hj^r@'^exposures '.which are,- more. than one-
hundred times greater. than the average. The second situation
would involve a dramatic increase in exposure. For example, an
individual who ingests a highly contaminated fish may have a
pulse exposure which would be orders of magnitude higher than th*
average daily exposure. The size of each population should be
estimated. **
4.2.4; Scenarios. - •
14
-------
. 16 to; 19, should be ••
rftcvaluat«d using other data because neither pase represents an
'average :*xiltirif:'''plmte';vrtor.r;a-;new 'modern stata-of-the-art
facility. :"-;;fh
-------
. .-that volatiization
acc&unted-for ,-in\ off-site as well' 'as. on-site- analysis;. ' This • will
avoid bia« which could inflate dairy and befif ingestion related
' eatposuresv • 'It would also insure consistency of analysis'across. '
exposure estimates.,-,.' "• ' • • •' . • ' ', • , •' • ''•••' ' •
Partitioning phenomena between vapor phase and particulat*-
bound TCDD in air and dissolved and particulate-bound TcDD are '
not; defined.-.'. Since the environmental behavior is dependent on
the extent of partitioning, it is recommended that development of
•'descriptions of these phenomena be undertaken* '•-' ' .-'- ,
Hie dilution of deposited TCDD-contaminated particulate with
uncontaminated soil is not calculated. A mixing depth for
particulates of 1 cm top soil (stated on p. 239) is presented
without justification* The revised exposure document should
include basis for the -selection of this specific depth. Research
to develop a dilution model -analogous to the mixing model
developed for dilution of land-eroded particulates' should be
carried out so that parallel phenomena are treated In like
manner. ,
4.2.5 fixposure and Epidemiology.
Epidemiology studies of human populations have yielded
little information on health effects, but are a fruitful source
of information on human exposure. Improved understanding of
exposure is essential for designing studies of human health
affects. This 'is very clear in the targeting of herbicide
manufacturing workers. It is also clear that studies of Vietnam
veterans as a group, have failed because of lack of good exposure
data. Epidemiological studies, of veterans or'other occupational
groups with known and adequate exposure data should'be pursued.
However,, the subpanel recommends that the'exposure document
emphasize that without food exposure data, the epideniologic
studies are meaningless. '
4.2.6 'Estimating Exposures, in the 'Absence of Empirical
pata
•.**. ' '
The document provided exposure estimates'.in the absence of
reliable empirical data. In .the absenei, of data, any approach
that is proposed cannot be refuted or validated. The Subpanel
recommends'that the reader be alerted when'an estimate from a
given pathway has a particularly large uncertainty associated
with it and that the document note that an exposure estimate nay
be too uncertain to be reliable* The exposure document only
^rea'ches this latter conclusion for plant uptake of 2»3,7,a-TCDD
(pp. 45, 205} after «apining that the plant uptake data were
conflicting. {The problems with the plant uptake data are
discussed in detail under section 2.2.2.)
16
-------
The data-" on- the-, "division: of" 2,3,7,8-1
the vapor and, particulitervphase' is-also.; weak",./-'' TJh«-'S^paniel:i--';^;:';i-^'
'recommends that the /reader" 'ber"-s'pecifIcaily' al*r^ed:''.to''::ti&«:;:''; vi;':-v:v'V ;'
particular ..weakness, "in /scenarlos^usi'ng- these'" data »•.'_•• •'•;".••-'•','•.••.':• ..•'"• ' _
on page 220 the: exposure document states, "There are
virtually no data concerning the division of 2,3,7,8-TCDD
emissions between the vapor and- particulate .-phase for stack;. '';"
emissions in. the U.s.A."' , From' 'the*" foregoing quotes,' it is ' ': .
apparent that the authors''of this exposure document are net sure
what, percentage of. .'.2,3 f'7r8-fCBD .from ''incinerator stack-..emissions-
is in the vapor or particulate phase. Nevertheless,\the exposure
document states on page 220: "For our scenarios we .assumed 63%
to be in the vapor phase and 37% particulate, since these values
are reported in two studies (Hagenmeier et al«? 1985; Scheidl et
al., 1985)." In Hagenmeier,. et al, various sampling methods for
FCDD and PCDFs in stack gases'were compared, . It specifically
stated that "the distribution of PCDD/PCDF between filter dust,
condensate and iropinger does not'allow to distinguish between
particle bound and gaseous PCDD and PCDF in the stack gas."
It is also impossible to determine the distribution'of
2,3,7,8-TCDD.released from incinerators in the vapor and
particulate phases based'on information on the'release of'total
CODs in the vapor and particulate phases since 2,3,7,8-TCDD makes
up such a small fraction, of all CDDs released (cf. Table 6*27).
Other exposure -estimates.' which are based on little or no
data include; (1) vapor inhalation from landfill or. contaminated
soil scenarios, and- (2) all pathways from incinerator 'fly'ash
stack emissions. '
4.2.7 Rely More on Field Sampling Dataand Less on
Dispersion Modeling_Jtesults • • . • •
The Executive Summary of,the exposure document (p, 1)
states: "It should be emphasized that when assessing particular
sites, monitoring data may be available, or measurements, may'be
made, that would preclude the necessity-of part or all of the
estimation methods described 'here.11 This essential principle
receives very little'attention throughout the development of this
exposure document.. While it is mentioned several C^mes in the
exposure document, the'emphasis is clearly on developing an
approach to estimating exposures regardless of how much date are '
available. . • '..*:.••
An example of a particular problem is the exposure estimate
to 2,3,7,8-TCDD from incinerator fly ash, which was deposited on
cattle fodder, which was 'consumed by beef cattle which were
subsequently ingested* If EPA were presented with such an
estimate, it would ppobably instruct the submitter to "measure"
17
-------
'sUl
the "level that:is-'prese;nt^tov$^ so :many''"..;
assumptions must be made to estimate human exposure 2,3,7,8-
from beef derived from cattle ingesting fodder contaminated with
is seriously questioned and the Subpanel recommends that this be
highlighted in the document. A preferable alternative approach
to exposure estimation would be to measure the level of 2f3,7/8-
TCDD in the matrix of concern (in the above example, in beef) and
then to proceed with the exposure assessment. ' . '•'. • , "
4*2.8 . Vcrif iafole ^xpQ§yre'':Eistimatp3* ' •"•''." ' ,'•
The exposure estimates rely heavily on dispersion models and
exposure parameter values that are highly dependent on site-
specific characteristics. Further, a critical factor like
deposition velocity of particles receives little attention as
does estimating the concentration of 2,3,7,8-TCDD in a,particular
matrix. If field sampling data are available and if they
correspond to the modeled values, one does not know whether this
is because all .the attendant 'uncertainties canceled each other
out or whether the modeled value's were indeed "accurate;** Th«
Subpanel recommends that research be undertaken to verify and
improve the modelling estimates.
4,2.9 Incorporation. ;_eif': the'Most Reliable Exposure
4g s e s spen t | ri f o'rm a t i o_fl . " .
The exposure document should include information on the low
bioavailability of 2,3,7,8-TCDD on fly ash as compared to 'soil in
those scenarios which assess exposure to fly ash* The work of
van de Berg, et al {1S83, 1985}, is cited which suggests a lower
bioavailability of 2,3,7,8-TCDD from fly ash than from soil.
4.1,10 , R i go rous Ch a rae t e rri za t ion of Uncer t a in ty.
The exposure document presents its uncertainty analysis in
Chapter 7, The Subpanel commends the Agency for the detailed;
presentation on uncertainty. , This characterization is an
important element in the risk assessment process and the Agency
.has done a commendable job in this particular document. jc-
In Tables 7-1 to 7-17, the uncertainty analysis of aome
parameter variables for .various/exposure pathways is given-,,- Far
example, Table 7-1 on soil dilution .factor presents 'uncertainties
for three parameters* . "quantity of erosion from contaminated
area", "quantity of'eroded soil deposited on adjacent field",, ar.j
"mathematical model to assess the rate of 2,3,7,8-TCDD in soil
placed on adjacent field"., The Subpanel recommends that
uncertainties surroupiing other variables, such as soil
characteristics, land use, elimatological conditions, and hew tn*
landfill is contaminated with 2,3,7,8-TCDD, be addressed.
18
-------
• Th* exposure doeimeht1?.
made kore quantitative, -instead^of ''li^ieal^ng;;,th^>^ \tua'4tity<'
of «ro«ion from the contaminated area ranges b«tween 0*'$ and 306
tons '"par/ ie're-yewr and • that. thelvaiueVfcsed ''wasr f 5 :tons per: acre~;;
year, th« Subpanel r«comncnds that th« «xposur« docxment stater/- ''
the uncertainty in .terms' of a 'deviation.,:'••,:'• By , stating the ,
uncertainty of each parameter value in thi» manner, the magnitude
of the uncertainty for the entire exposure assessment can be
determined by multiplying the uncertainty factor of each
parameter'.- In addition, the parameters that "drive" the exposure-
assessment could be identified and prioritised for verification
with field sampling results. .''••-'-.•..'
4.2.11 Pharrnacekinetics. •
The exposure document outlines two pharmacokinetic
approaches for relating exposure with tissue levels pf TCDD. The
first,, referred to as the "Commoner Approach,11 is described on
pp. 128-130. It refers to the use of a single-compartment open
model with'first-order elimination to describe TCDD disposition
in the body; this model was presented'by Commoner, "et al, at two
symposia in 198§-86, Although the model itself is
straightforward, several errors in developing equations (5-1) to
(5-4) detract from the description.
The Subpanel recommends that the document be corrected
either to be consistent with the treatment of Commoner,'et al, or
to follow a similar approach which employs other constants which
may be desired. For example, the agency may wish to include an
absorption factor to account for fractional bioavailability.
The second pharmacokinetic approach outlined in pp. 130-137
describes development-of a physiologically based pharmacokinetic
model (PBPK) where none currently exists.- The'document suggests
that one such model developed for 2,3,7,8-TCDF by King, et al
(1983), may be adapted'to TCDD. -'While the tissue compartments,
volumes, and perfusion rates of the King's model may be
appropriate for TCDD, the binding and metabolism of TCDD may be
sufficiently different from that of TCDF to require a somewhat
different structure. 'Furthermore, because the half life of TCDD
in humans is greater than that predicted from observations in the
rat, it may not be"possible to rely entirely upon-rodent species '
in developing such a PBPK model. The Subpanel agrees with the
Agency that development of a PBPK model for TCDD is an important
task which should be pursued with high priority. However, the
Subpanel recommends that these potential -limitations be included -
in the document.
19
-------
Chairman'- '':
Dr. 'BernaTd'-rCipldstein.' ' .;• ..i.-^-.', .-.- ..•-.;: •' '. t/./ .., ' ,. • ' _ '.
' ;.. Department ' of '- Environmental and Community , Medicine
UMDNJ -Robert Hood Johnson M*dical School ' •
'. 675 Hoes Lane '. • '- -'' ' V • ', '•'"'''•'' ;" ../
Pisd«ta«ayf KJ 08854 '. -^ ;'-;v -;'•""
Dr> Nancy K. Kim ,
. .Dirfteto,rf division of -Environmental' Health-
New Yorlc Department of Healtli
2; University Plaza '.-'•-'- . ' . -
Albany, NY 12203
DIOXIN RISK StJBPANEL
hairman
Dr. Bernard Goldstein . •
Department of Environmental and Community Medicin*
UMDNJ-Robert Wood Johnson Medical • School
67S Hoes Lane . . ' • '
Piscataway, NJ 08SS4
|f ember s /Consultants . " ; '
Dr» Linda S, Birnbaum
National Institute of Environmental Health Science*
MD C3-02
P.O. Box 12233 • .
Research Triangle Park, NC 27709
Dr. Marilyn A. Fingerhut
NIOSH ' '
5555 .Ridge Avenue
Cincinnati, OH 45213
Dr. Tom Gasiewicz
Department of Biophysics
Box EKSC , . • '
University of Rochester
. 575 Elmwood Avenue
' RochesterT"NY 14642
20
-------
National Center for Toxicological R*»«arch
' ' '
Mail Prop A3-02 •• "4 '.•'•'- .
F.p..;.pp.^;i.22».r.. .••••. .• . , ' . •
RescarclH triangle ParX, NC 27?0i
Or.
Environftental Health Center
' Roon '-120 •,..-.' ' ' • '•
Tunney. •''*.- 'Pasture - ' •.
Ottawa, Ontario, Canada K1A OL2
Dr. Ellen Silbergeld /
Environmental Defense Fund
161i P Street, H.W.
Washington, D.C. 20036
Dr. Thomas Starr •-
CUT " •
P.O. Bqx 12137
Research Triangle Park, NC 2770§
•DZOXIK EXPOSURE PANEL
Chairman
Dr. Nancy Kim . • . •
Director, Division of Environmental Health
2 University- Plaza • • • .
Albany, NY 12203
Members/consultants ' ' ' • '
Dr. Edward Calabrese
N-344 Merrill Science center
University of Massachusetts
Aoherst, Kh 01003
Dr. Warren ..Czrumiett
DOW chemical, U.S,A.
. 1897 Building
Midland, Ml 48667
-------
. Dr.'
•'••'" UMDNJ-Robort Wood 'Johnson Medical School
''•' Environnental and CoBuaunity Medicine -
''• '' "'' ' '' '
,Dr* Raymond
Enviroxua«nt Canada . ,
Hull, Quebec, Canada K1A OH3
.Dr. Dennis Paustenbach
CheroRisk, McLaren •
Environmental Enginnaering
" ''
980 Atlantic Avenue ,
Alameda, CA 94501
Dr. Stephen H. Eappaport
Department of Bioroedical and Environmental
Health Sciences
Earl Warren Hall i Room 317
School of Public Health
University of California
Berkeley, CA 94720
Secretary
Dr. C. Richard
science Advisory Board (A-loiF)
'U.S. Environmental Protection Agency
Washington, D.c. 20460
g t a f ,f _ • S ecr .e t a rv " . .
Ms. Mary L. Winston
Science Advisory Board (A-101F)
U.S. Environmental Protection Agency
Washington, D.C. 20460
Acting __ Sta_f f .Director ;
Mrs, Kathleen Convay
Science Advisory Board (A-1Q1F)
U.S. Environmental' Protection Agency
Washington, D.C. 20460
------- |