PPA U.S. Environmental Washington. DC
Protsction Ag«ney EPA-SAB-EHC-ao-001
Report of The Joint Study
Group on Lead
Review of Lead Carcinogenicity
And EPA Scientific Policy
on Lead
A SCIENCE ADVISORY BOARD REPCfl' 06CEVSS-
-------
f • • 4
(mj
UNITED STATES gNV IRQNMESTAU PROTECTION AGENCY
£?A-SA3-£HC-?Q-001
November 21, 198§
Honorable William X. Reilly
Administrator
U.S. Environmental Protection Agency
401 M Street, S.w. 20460
Subjects science Advisory Board's Review of the offie* of S
and Development's (QRD) Draft Assessment Document "Review cf .
Carcinogenic Potential of Lead Associated With Oral Exposure,"
(QH1A-C267, July 1981}
Dear Mr. Re illy,
At its November 10, 1988 meeting, the Science Advisory 2cari
(SAB) Executive committee discussed an inquiry members had receives
from representatives of the Lead Industries Association, :-~.
regarding the ?7rs"C*d classification of lead and lead cor,±=_r;3
as "B2" carcinogens in a draft assessment document noted * = ;-••>.
The 32 designation identifies a substance as a probable - --i-
carcinogen, a term off art generally indicating the presence ::
sufficient evidence of aniaal carcinogenicity, and inadeq-.ite
evidence of human carcinogenicity,
As a result of its discussion, the Executive Cc—:-_-.--
requested from then EPA Administrator Thomas the opportunity t-
review the basis for the'12 classification of lead, as 'well -;
examine related issues concerninf the Agency's scientific PCS.- .:.-
on lead and lead compounds* An ad liflfi Joint Study Grc,= -is
created, including members of the Executive Committee, t.-.e :-.3
Environmental Heal til Committee, and til* Clean Mr Scier, z,:.;
Advisory Committee* Meetings were held in Washington, D,:. :r,
March 31, Iftf, and Research Triangle Par*, N.c. on April . :
28, 1989.
-. After ext*-»«iv« review, the Study Group agrees wi-r. --a
proposed B2 classification for lead and lead compounds, pri-ir,./
on the basis of the animal tumor data summarized in the asses3----
docuaent. It must be stressed however, that given the
state of our understanding or' the mechanisms of lead-:-:.
tumorigenesis and the data gaps involved, and cdiiftidftsriitg "•*
high levels of exposure in' the critical studies, --•.-
classification is not considered to provide a sufficient ^25 ,
risk, aitetiment-
-------
The Study Group also reviewed the manner in which scientific
data on lead were utilized by the research, air, and drinking water
programs; information from other SPA units was not available. The
Group found a generally sound, but not fully consistent, approach
Cor the three program area* noted, but there were some significant
issues—-target blood lead levels and definition of populations at
risk, asong others—-in which difference* were noted. The study
Group recommends that SPA develop a national policy on blood lead
level reductions (particularly for children), and that EPA should
address environmental exposure to lead on the basis of preventing
adverse neurological effects in children* Basing regulatory
strategy on the observable neurobehavioral effects in a sensitive
huaan population can avoid problems of rodent*to-nait and high-dosa-
to-environmentally realistic level extrapolations? this apprsacr,
appears likely to provide an acceptable degree of prot,ec.-±".
against other adverse effects in the entire population.
Lastly, the Study Group urges the Administrator to undertake
positive action to assure the uniform application, of scientific
data on lead to regulatory decisions by all organizations witr.i-
tha Agency. The mechanisms to accomplish this are beyond t-e
purview of the Joint" Study Group, but we regard it as a highly
important goal if the Agency is to make sound and consistent
decisions concerning lead in the environment.
Me appreciate the opportunity to carry out this review, 2-^
look forward to your response.
Sincerely,
•DrT-Raymond c* Loehr, Chairman
Executive committee
Science'Advisory Board
Dr. Arthur Upton, chairman
Joint Study Group on
-------
V* 3. ENVIRONMENTAL VftOTlGTXOK XGJNCY
NQflCl
This report has been written as a part off the activities of
the Science Advisory Board, a public advisory group providing
extramural scientific information and advice to the Adair.istra-c:
and ather officials of the Environmental Protection Ager.cy. r:-a
Board is structured to provide balanced, expert assessment -t
scientific matters related to proolems facing the Agency. T!-,_5
repuct has not Seen reviewed for approval 'by the Agency and,
hence, the contents of this report do not necessarily reprsser-.i
the views and policies of the Environmental Protection Agency,
r.or of other agencies in the Executive Branch of the Federal
government, nor does mention of trade names or commercial prc-
ducts constitute a recommendation for use.
-------
This report presents the conclusion* and recommendations of
the U.S. Environmental Protection Agency's Science Advisory Board
summarizing a review of th* Of fie* of teseareh and Development's
Draft Assessment ' Document "Review of th* Carcinogenetic Potential
of Lead Associatitd With Oral Exposure,1* Th* Board's major
conclusion is that the proposed 82 classification is apprcpnata,
but that there is not sufficient basis for a quantitative r^s.<
assessment. The Board recommends that £FA establish a natic-al
blood lead policy, create internal mechanisms to assure '-?.- al-
organizations within th* Agency deal with lead-related scier.-i:i-
.issues in a sound and consistent manner, and that regulate ry
i--rategy b* based upon preventing lead-induced neurological „-.
children — a sensitive population.
Lead; care inogene tic? 12,' blood lead levels;
tumor igenesis
-------
O. 8. IMTXAOMtmTAL PROTICTIOM A0ENCY
SCIIHCZ ADVISORY IOAJIO/CL2AJI All ieilHTI»IC ADVISORY COHKItTfE
JOINT LEAD STUDY
Chairmaa
Or. Arthur Uptan
New York University
Medical Canter
Institute of -
Environmental Medicine
550 First Avenue
Room M-S-8 213
New YorK, New York
10016
MeaJa_ers ant Consultanta
Or. Gary Carlson
Department of
Pharmacology
and Toxicology
School of Pharmacy
Purdue University
West Lafayette, IN
47907
Dr. Paul Diesler
11215 Wilding Lane
Houston, Texas 78712
Dr. Richard Grieseoer
P.O. Box 12233
Research Triangle Park,
NC 27709
Or. Paul Hammond
Department of
Environmental Health
University of Cincinnati
Medical Center
Cincinnati, Ohio 4526?-
OQ5«
Dr. Rolf Hartung :
School of Medicine
University of Michigan
3125 Fernvood Avenue
Ann,Arbor, Michigan 48108
Dr
Dr.
Morton Lippmann
Institute of
Environmental Medic i.-.e
M«v ¥orK University,
Lan«a Laboratory
Lang Meadow load
Tuxedo, New
Thomas Mack
PM1-I105
2025 Zonal Avenue
Los Angles, CA
Or. Kathryn ft. Mabaffey
National Institute =f
Environmental
Health Sciences
University of Cir.cir.r.a
Medical Center
Mt-S«, 3223 Eden Av«-
Cincinnati, Ohio --::
Dr. Marc SchenJcer
Director , Occupat i = r i ,
and Environmental
Health Unit
University of calif—-
Davis, California
Dr. Ellen K. silbergeid
, Environmental Defense
Fund
I61i P straet, N.w.
Washington, 0. C.
Dr. Bernard Weiss
Division of Toxics 1 : *
P.O. Bex IBB
Ofnivarsity o£ Roche sr *
School of Medicine
Rochester, New Ycr< .
ILL
-------
Samuel
Executive Secretary
Science Advisory Board (A101F)
U.S. Environmental Protection Agency
Washington, D»C* 20460
Staff
Ms. Mary L. Winston
staff S«cr*tary
Sci«nc« Advisory Board (A101F)
U.S. Snvironn*ntal Protection Agency
Washington, D.G. 20460
Director, Seimnem Advisory Board
Dr. Donald G. Barnes
Scienca Advisory Board
Scienc* Advisory Board (A101)
U.S. Environmental Protection Agency
Washington, D.C. 20460
LV
-------
o>
1.0 EXECUTIVE SUMMARY ...... . . . ! .........
2.0 APPROACH , ...... . ...... . ......... 2
3.0 REVIEW OF THE WEIGHT OF EVIDENCE CLASS I FI CATION AS A 32
CARCINOGEN » ......... .......... 3
3.1 DETAILED FINDINGS ..............
3.1.1 Animal Tumor Data ........
3.1.2 Epidemiology . » . , .
3.1,3 Potency .............
3.1.4 Mechanism of Lead Carcinogenicity
4.0 REVIEW OF OTHER ASPECTS OF THE STUDY GROUP'S CHASGE
4.1 Review of ORO and OAQPS Documents .......
4.2 Lead Toxicity ....,.- 9 •
4.2.1 Definition and comparison of Different
Adverse Effects .............. 9
4.2.2 Mechanisms and Dose-Response Models .... i:
4.3 Risk and Exposure .........
4.3,1 High-Risfc Subpopulations . ,
4.3.2 Exposure-Dose Relationships
4.4 cross-Agency consistency on Lead Exposures and
Their Health Effects—QA§iS»QOW» And OHEA . .
4.4.1 Health Effects Data Used to Support
Regulatory Actions ......
4.4.2 Target Values for Hood Lead
4.4.3 Population at Risk .
4.4.4 Percentage of Population Exceeding Blood
IMtA Limits ..............
4.4*5 Exposure Models .......
4.4.6 Risfc-reduction strategies .......
5.0 RECOXKENDATIONS
-------
1.0 BJ1C9TXVB
At its November J.Q, ifig meeting, the Science Advisory
(SAB) Executive Committee discussed an inquiry members had
received from representatives of the Lead Industries Association,
Inc., regarding the proposed classification of lead and lead
compounds as "B2" carcinogens in a draft, assessment document
(Review of the Carcinogenic Potential of Lead Associated With
oral Exposure, OHEA-C267, July 1918) produced by the office of
Research and Develop aent ' s (ORD) Office of Health and
Environmental Assessment (OHEA) . The B2- designation identifies a
substance as a probable human carcinogen, a term of art generally
indicating t.i* presence of sufficient evidence of aniaal
carcinogenicity, and inadequate evidence of human
carcinogenicity.
As a result of its discussion, the Executive Committee T.cv=d
to request from the EPA Administrator the opportunity to review
the basis far the B2 classification of lead. It also recede.- ded
that the review take into consideration related issues csr.cerr.ir.g
the Agency's scientific position on lead and lead compounds.
Since many of the same issues were already being addressed by --e
Office of Air Quality and Standards (QAQPS) , the latter office
requested that the SAB postpone its review of the QAQFS position.
on lead until after the Board had resolved the carcinogenicity
question. An §£ hog Joint Study Group was created, including
members of the Executive Committee, the SAB Environmental Heali-
Comnitte*, and the Clean Air Scientific Advisory Committee,
Meetings were1' held in Washington, D.C. on March 31, 1989, ar.d
Research Triangle Park, N.c. on April 27 and 28, 1939.
As a result of its review, the Study Group supports
proposed §2 classification for lead and lead compounds, primarily
on the basis of the animal tumor data, summarized in the review
document. Although there are some uncertainties and anomalies ir,
the. interspeeies data, the high incidence of Kidney tumors
resulting from I'&vg* cumulative doses of lead in ruts and raica -:
bath sexes in many independent experiments provides » sufficient"
evidence of til* carcinogenicity of lead in laboratory animals. :.--
addition, it is appropriate to conclude that lead is genotoxir.
although we do not have a full understanding of how this p roper- y
may relate to its turner igenic effect. It is also noteworthy --.ii
the available epidemiological data are inadequate for evaluate- 3
the potential carcinogenicity of lead.
It must be stressed moreover, that given the unknowns nctei
above, and cansidsrinf ths very hifM Isvsls of exposure in the
critical studies, ths B2 classification may not provide a
sufficient basis for Quantitative risk assessment.
This' report also provides a detailed analysis of issues
-------
raised by the OH£A document, as well as recommendations for
further improvenents. The study Group also reviewed the manner
in which scientific data on lead were utilized by the research,
air, and drinking water programs; information on other EPA units
was not availably. The Group found a generally sound, but less
than fully consistent, approach for the three program areas
noted, i.e., there were some significant issues«target blcod
lead levels and definition of populations at risk, among others
in which difference* were noted. These issues are addressed in
detail below.
The study Group recommends that EPA develop a national
policy an blood lead level reductions (particularly for
children), and that EPA should address environmental exposure 13
lead on the basis of preventing adverse neurological effects ir,
children. Basing regulatory strategy on preventing the
observable, well documented neurobehavioral effects in a
sensitive human population does not raise the unresolved
methodological problems of rodent-to-man and high-dose-to-
environmentally realistic level extrapolations; this approach
can also be expected to provide an acceptable degree of
protection against other adverse effects in the entire
population. In order to achieve these policy goals, the Agency
should conduct further research to obtain mass-balance data en
lead in the environment and its fluxes to routes of exposure to
the human population.
Lastly, the study Group urges the Administrator to under::2.--.
positive action to assure the uniform application of scientific
data on lead to regulatory decisions by all organizations witr.i.-
the Agency. The mechanisms to accomplish this are beyond the
purview of the Joint study Group, but we regard it as an highly
important goal if the Agency is to make sound and consistent
decisions concerning lead in the environment.
2.0
To accomplish the desired review, the Executive Committee
recommended the formation of a Joint Study Group on Lead, the
members of which would be drawn from the Executive committee, t.-s
Clean Air scientific Advisory Committee (CASAC), the SAB Environ-
mental Health Committee, and other source* of expertise, as
appropriate)*
-. The charge to the Joint study Group' directed it to:
'*.... perform e 'breed speetvu** review of lead-related
health effect* end exposure) issue* whioh cut across
organisational line*. To accomplish tnis, the Group
will assess the scientific information concerning
' now available to the i^euey* «a& determine if it is
being applied in a sound and consistent rnana*? to
-------
standard setting and other regulatory
throughout the Agency.1*
Specifically, the Joint Study Group was requested to r
1 * The weight of evidence classification off lead and
lead compounds as carcinogens, as discussed ia the ORD
Office of Health and Environmental Assessment document
entitled 'tvalustloa of the Potential carcinogenic!ty
'of Lead And taad Compounds
2. Tls Supplaasat to th« 198« Add«adua to ta* If3«
Quality Crit»ria Oocumaat for L«ad, »ad th« doeuvtnt
•atitlcd 'R«vi«v of Tl* N4tioa4l AaHicat Air Qualir/
st4Bd4rd» for L«4d: Ms«*sa*at of specific And
Taeaaieal Inforaation' (OAQPB draft itaff pap«r)
3. fh« coaaiat«aey of tha proposed regulations for
lead in drinking water vita praviouily reviewed and
approved data, studies, aad aaalyiaa.
4* Other relevant issues which may be identified by
the Joint Study aroup, CA3AC, or the SAB Executive
Coaaittee.
The Joint Study Group met twice— first on March 30, ;s=?
with the primary purpose of reviewing the oHEA <|ocumant en
eareinogenicityj and again on April 27-28, 1985, sitting ^si.-.-:.
with the CASAC, to address the broader aspects of the Execu.:.-
committee charge.
3.0 mrrriw or rai wgianr ay EVTDEMCH CLAaaiyicATioN xa A 82
With the availability in March, ISiS, of an updated
from OHEA on the carcinogen i city of lead and lead compounds,
Joint Study Group met in Washington, D.C. on March 31 to revi
the weight of evidence classification of the carcinogenic ity
lead.
After detailed briefings from Staff Officials of the €32,
comment frov a representative of the Lead Industries Asscci^i
and extensive) discussions, the study Group broke into sul9gr:u
to focus on specific aspects of the relevant issues—animal '.
data, epidemiology, potency, and mechanisms of lead care in =3
icity, Subsequsntly, the study Group mat again in plenary
session, at which time there vas further discussion of aach
issue, with details! suggestions to Agency staff for improve-
of the document.
The most significant suggestions for improvement relate i
the interpretation of certain critical studies, the omisaicr
-------
few relevant publications, and the failure to exploit the
available human data for deriving upper bound estimates of
carcinogenic potency. These issue* not withstanding, the
consensus of the Study Group was that the main conclusion of the
document—1*4*, that the weight of evidence was sufficient to
support a B2 classification for the carcinogenicity of lead — is
correct.
3.1 DETJULZD
3.1.1 Animal, T^mog-_0ata. in more than 20 separate studies, isa
acetate or lead subaeetate administered chronically at relativel
high concentrations in the diet has been observed to cause an
elevated incidence of kidney tumors in rats and mice* In a :=v
experimenta, the repeated administration of lead phosphate by
intrapesritoneal injection, subcutaneous inject ion , or skin
painting also has been observed to cause comparable tumor iger.iz
effects in mice and rats, on the other hand, experiments in
which lead acetate or other lead compounds have been administeze
similarly to hamsters, rabbits, monkeys, and dogs have not
demonstrated tumor igenic effects in these species, but the
results are inconclusive in view of the small numbers of anirais
exposed and/or the limited duration of exposure and/or follow—,;?
The animal studies are definitive in demonstrating the c*r-
cinogenicity of soluble lead salts at relatively high dose L-av^i
for rodents. However, most of the studies were conducted ncr?
than 10 years ago. in the large majority of these studies, -r.=
group sizes were small, and the majority of the results were
under-reported in respect to the strain of animal used, the
survival of test and control groups, and the presence or acse-za
of competing risks. Also, the absence of age-adjusted analyse*
complicates the interpretation of non-renal tumors (such as
glioma* or reticulum cell, sarcoma*).
Partially unresolved questions include: (1) tumor igen i c.t.-
of lead for species other than the rat and the mouse j (2) t-e
forms of lead that produce renal tumors (lead itself and orgs-.i
lead compound* have not been studied adequately)! (3) the
effectiveness of th« inhalation route of exposure, which has -;-.
been studied as yet in animals; (4) the susceptibility of
prenatal and postnatal age group*, which have not been stud;*:
adwquately; (9) the possible effects of dietary variations : •
(6) , the Nineties of turnover of lead in trabecular and c=r- .. j,
bone. ' ;
In addition to the above points, it also should be
emphasized that we currently do not understand the mechanis-: -.:
kidney tumor induction by lead, :r, despite the caveat* acc-«
can we explain the general absence of such tumors at lower >*.*.
of exposure —albeit levels which ar* several tia**. higher -- i-
-------
humans are typically exposed to in the environment. '
consequently* it is not clear how one would extrapolate the
observed effects in rats to estimate the carcinogenetic effects
of low-level exposures to humans,
Also noteworthy is the fact that the renal tubular
carcinomas associated witll lead administration differ from those
produced by ether substances and associated with hyaline droplet
formation (alpha 2u globulin-related) . Rats and mice of both
sexes are affected by lead, and focal droplet formation, celluia;
degeneration, and cast formation are not reported. The lead-
related lesions differ, too, from those typically produced by
analgesic drugs, which include papillary necrosis and tuners -f
the renal pelvis rather than of tubular epithelium,
In spite of the above uncertainties, the high incidence ;:
kidney tumors resulting from (admittedly) large cumulative dcses
of lead in rats and mice of both sexes in many independent
experiments provides "sufficient11 evidence of the carcinogenic*':-;
of lead in laboratory animals. Thus the data justify the 32
classification of lead as a carcinogen but are limited uitn
respect to support for a quantitative risk assessment.
3.1.2 EpJ^mloj. oay. Results of the published human studies -r,
lead fall short of satisfying even the ** limited evidence"
criteria for careinogenicity, Relatively few epidemiological
studies of quality have dealt with this ubiquitous and familiar
agent, the Agency should make a major effort to support
innovative studies of long-term outcomes of human exposure -r
lead. Such efforts should include population-based, as well is
industry-based, studies and should be designed to assess
environmentally distributed lead compounds in relation to car. 13
risks.
Host of the U.S. .population was exposed to increasing
concentrations of lead in air and food during the period 192 r-
1975, primarily owing to the us* of lead in fuel additives; . .-
then, the input* of lead have diminished dramatically. Ouriri
the same period, millions of workers have had significant
occupational exposure*. Xn addition, many thousand* of chi.;—
living in deteriorated housing infested large quantities of i-^
particularly from lead-based paints and high- lead content =r.:-
in the plumbing system, as well as from lead particulate r.a---r
associated with automobile traffic. Those surviving the acui-s
effects retain elevated body burdens. As a result of these
exposures, there- is an extensive literature on blood lead L*--*.
in both occupationally exposed workers and children examined .-
community screening programs. Furthermore, there isanexte-,,
data base on blood lead levels in the populations of selects:
U.S. cities sines the 1950s, as well as a large population-^ . •
sample in the NHJUilS IX (National Health and Nutrition Sur-.e
study performed in the period 19T«-1980. TH» latter sxtands - •
-------
the time when the usage of lead in gasoline peaked to a tiae when
it had sine* dropped by more than 75%. Because blood lead is a
marker of lead intake, although not of the body burden-
especially in population studies of on-going exposure*— an
analysis of kidney cancer trends during the past six decades in
relation to changes in population mean blood lead levels mights
have some ability to show whether upper-bound estimates of
carcinogenic potency extrapolated from animal bioassay' data were
realistic. Analyses of bone lead may increase the precision of
dosimetry. Other points along the exposure- response continuum
might also be established fros those occupational cohorts for
whom exposure histories" ware available. Such analyses would r.st:
be capable of determining whether kidney cancer was caused by
lead, but might be capable of establishing an upper bound far
plavsitle risk estimates. Thus, the data available in the span
literature on blood lead in various populations could be ar.alyzsi
jar trends in relation to trends in kidney cancer incidence.
In addition, EPJfc should not reject out of hand the
quantitative risk assessment approach based on animal data,
although this approach involves many uncertain assumptions
concerning species differences in lead adsorption and organ
distribution, i.e., that people and rats have similar responses
to comparable dose levels. It also assumes that high-level
exposure can be extrapolated to low-level situations.
Other quantitative risk assessments have been made in the
past, using rodent databases no better than, nor even as good =s,
the existing database for lead.
3.1.3 PotanqY- Assessment of the potency of lead as a potential
human carcinogen involves a series of extrapolations, for whicn a
number of methodologies can be used. The Agency has often
selected the upper confidence bound of the linearized multi-
stag* model as th* default model for potency estimations
applicable to lower levels of potential risks. This default
model assumes that there is no threshold to the response. The
evidence that ft threshold model would be more appropriate for
lead is considered to b* ambiguous by most of th* Joint Study
Group, but ther* is consensus that th* linearized multi-stage
model fails to take into account much, relevant information,
including pharmacokinetic data and measures such as time-to*
tumor.
Ideally, the potency extrapolation model should employ best;
estimates of biologically available lead dose, based upon
pharaacoXinetic modeling. A question that may not be readily
answerable at this time, however, is whether blood lead levels.
kidney lead levels, cancer-related bone lead levels, or renal
lend clearance rates would be' the preferred measures of
dose. Nevertheless, th* wealth of ph*rmaco kinetic data on
available 'from studies in humans and rats should be. used,,
-------
as possible, in defining the carcinogenic dose in rats and in
naJcing comparisons with humans, When possible, moreover, it
would appear prudent to compare studies on the basis of
roj/kg/day, i.e., the rate of exposure, rather than on the basis
of total cumulative exposure, divided by an arbitrary net body
weight and time for exposure. Physiologically-based
pharraaeofcinstic models should be useful in such comparisons.
If the carcinogenicity of lead is assumed to be associated
with the lead ion itself/ the question of the comparative
carcinogenic potency of the various inorganic lead salts should
be addressed in relation to their respective bioavailability and
delivered (internal) doses. Also, as pointed out in the OHHA
. document, appropriate allowances for age and nutritional status
need to be included.
Although numerous physico-chemical properties of lead
influence its biokinetics during absorption via the respiratory
or gagtro- intestinal routes, the anion to which lead is teour.i ir.
the external environment does not necessarily remain associated
with the ligand that transports it to the target organ. Divalen
lead binds reversibly to various inorganic or organic ions, &. T^
proteins, which transport it to target tissues. It is Pb**,
rather than the molecular complex in which it is found in the
external environment, that is responsible for the toxic effects
of lead. Since the mechanism through which lead acts as a
carcinogen is not resolved, the way in which its biodistributian
may influence its carcinogenicity remains to be determined.
Also/ the ORD report needs to recognize more clearly that-
daily exposures are cumulative and, for humans, must be
integrated over decades. This is particularly important because
internal stores of lead can be mobilized and redistributed, far _
example, from bone to other target organs during pregnancy and
lactation, or because of osteoporosis.
3.1-4 t^eetianisiii ot_. Lead Carcinoqenieitv. Although the anizsal
data indicate lead to be carcinogenic, they raise a number of
questions about the underlying mechanisms and their generality.
The primary cjuestion, perhaps, is why the Kidney is ths most
susceptible rodent target organ* Indeed, the bulk of the
evidence implicating lead as a potential human carcinogen deper.is
upon this site of action in rodent species, h subsidiary
question that arises as a result is; do the tumors in rodents
develop in part from systemic toxicity? Such a question is pc
by the large doses typical of those causing tumors in the lead
carcinogenicity studies. For example, 1% lead acetate in the
diet corresponds to a dose as high as 800 Bf/k? daily. This
value contrasts markedly with ths amount demonstrated to induce
naurobehavioral effects ~ a fraction of I
Unfortunately, lead tissue levels are typically absent :r.-
?
-------
the report;* addressing eareinogenicity, so that comparisons are
largely speculative, especially given the many variables
determining fcioavailability and toxicokineties. The kidney,
however, is one of the documented targets o.f lead toxicity. It
remains to be determined, therefore, if renal damage, or impaired
renal function is a necessary precursor to tumor expression. Far
example, kidney tumors night be postulated to arise from a
combination of cellular damage and impaired function, leading
ultimately to elevated lead concentrations in kidney tissue that
might b* sufficient to damage DMA. Reports on eareinogenicity
have also largely ignored measures of heme synthesis, an
important endpoint for lead toxicity. These processes could be
related to tumor production.
Another potential contribution of systemic effects ecuId
also result from the joint consequences of lead kinetics and
aging. With advanced age, bone mass declines-, and lead
previously stored in bone, is released. Concurrently, nephrens
ars lost, so that advanced age may offer an especially sensitive
period for systemic toxicity to b« translated into tumprigenesis.
Although such proposed toxicological mechanisms are highly
speculative, they help to emphasize the uncertainties in
converting'the available animal data into a coherent estimate =f
human hazard, and, finally, into human risk.
In addition, there is substantial evidence for the
genotoxicity,of lead, while the data do not provide a
mechanistic understanding of carcinogenesis, they add weight -.:
the assessment of lead as a potential carcinogen. The ORD
document contains a discussion of genotoxic mechanisms; houevsr
it tends to over-emphasize relatively preliminary data on the
induction of protein kinase C activity as a mechanism of lead
carcinogenicity. Effects of lead added in v|jprQ on the actv-'.-.
of this enzyme in endothelial cells may be relevant, but t.-,.=,
is not yet known* It cannot be assumed that all inducers of -.-.e
enzyme are carcinogenic--the enzyme is also implicated in re-.n,
transmission.
The discussion alee omits two other possible mechanises;
th* ability of lead at relatively high concentrations (mM^ --
cause breeJcn in DMA, and the potential displacement by Pb" ~:
2n~ on so-called DMA finger loop* (ONX-binding proteins wh^r-
appear to regulate gene expression through their conformatic--
specific binding to specific genes, including some proto
oncogenes).
-------
j while it is appropriate to conclude that lead ia
genotoxic , it i» not presently known how this property aay
relate mechanistically to tuaorigenasis in th* kidney or any
other organ. Th* data do not in themselves provid* a scientific
basis for any specific mod* I for risk ass*ssment» or for a
judgment that lead is a promoter rath*r than an initiator of
carcinogenesis.
4.0 RBVIBW Of OTTTBH ASP1CT1 Of TCT gTCDY Saefgtj gPUUil
On April 27-2S, 1999, th* Study Group m*t jointly with ti-.e
CASAC at Research Triangle Park, N.C. to address other issues
driving from its charge-- particularly thos* issues which cut
across EPA organizational lines. On April 27, th* Study Gr=u?
participated in th* CASAC diacua«ion o£ two aajor lead-related
documents ("Review of th* National Anfciant Air Quality Standards
for Lead; Assessment of Scientific and Technical Information"
(OAQPS Draft Staff Pap*r» March, 1989), and th* ORD document,
"Supplement to th* Idas SFA Air Quality Criteria for Laad-Voiu.-a
1 Addendum (pages A1-A.67) , Harch 1989) , as identified in item ;2)
of th* specific charge. On April 28, th* CASAC and th* Joint
Study Group received briefings from representatives of EPA
regulatory- program offices (th* Of fie* of Toxic Subttances (CTS)
and the Of fie* of Drinking Water (ODW)) and discussed issues cf
lead toxicity, exposure/ ris)c, and cross-Agency consistency in
technical approach** to lead regulatory questions, with
particular attention to th* ODW's recently proposed regulate- :
permissible I*al l^v
Detailed discussion of each issue follows.
4.1 Revjjsv of_ QI^P §nd QftQPS ^eemaent* Th* d*tail*d findi*"s
recommendations on th*s* documents are contained in th* CASAC
Closure Letter (EPA-SAB-CASAO9Q-002, Kovember, 1989). It ;s
relevant to not* h*r* however that th* CASAC concurred with tr.
general modeling framework: presented in the OAQPS report and
endorsed th* us* of th* biolcinetic mod«l in children under s . <
years of age, and th* us* of th* disaggregate approach in ai- .
Th* CASAC cautioned that these modeling predictions were noc
valid for pregnant women and their fetuses du* to a lack of
information on this subpopulation. Th* us* of th* biofcineti-
model for metals oth*r than lead was not recommended.
4,2- Lead Toxieitv
j»
4,2.1 Detin^i^s -?t^A- Cftttgar^sen f f _ P|f ff*rf fit Advfg-a*' Affects
Adverse effects of toxicants ar* typically scored by the
intensity or severity of effects, or by th* numb*r of
'Zelilcoff et al, rtGenetic toxicology of Lead cccpc-- 1
Careinaqenieitv. Vol 9, p<| .1727-1732, USf,
-------
affected. For example, it is customary to estimate the nuaUsers
of cases of cancer that may be induced in the population by a
given exposure to a carcinogenic agent.. With neurotoxicants,
however, a different problem is posed? e.g.. the health risks
resulting from low-level exposure to lead are probably reflected
most clearly by a shift in the distribution of 1Q test scores,
for example, a blood lead level of 10 M^/dl (compared, say to 3
ug/dl) may shift th* mean intelligence test score by 5% (from a
score of 10o to a score of 95); such a shift in a population 3f
* 100 million would reduce the number of individuals scoring above
130 from 2.3 million to 0.99 million, and, in parallel, would
increase the number of individuals scoring below 70 corresper.i-
ingly. Furthermore, everyone would be affected, so that tr.e n = /.
should be expressed in terms of a shift in the distribution sf :;
scores in the population. Analyses based on the median
underestimate the total impact*
Hence, it would be informative in comparing the risks cf
the tvo end-effects to estimate the percentage o.f children at -r.e
extremes of the distribution-of intelligence test scores (beirv,
say, 70 and above 130), who would be affected by the projects!
shifts, in the mean. Such an approach would indicate conseque.-r?d
stemming from any specific choice of a targeted blood level. :.
would also avoid the use of such term* as "level of concern,"
which, to' some readers, might imply support for a threshold ev-±-
though no such position is proposed.
f
Similarly, the total public health impact of relatively
'small displacement* of blood pressure levels by lead might als=
be expressed mere clearly if stated in terms of shifts in
distribution within the whole population.
4,2.2 Efachanisas and Pose-Response Models. The rich source ::'
information that, exist* on the experimental neurotoxicity ?f .-;;
is highly relevant ins (1) supporting the human studies
remarkably well, providing confirmation of observations in ---
absence of th* confounding variables inevitably present in -..-<-
populations, and (2) providing the mechanistic basis for
developing health-based foals for reducing texicity.
Particularly relevant are the recent studies on
neurobehavioral dysfunctions associated with low-dose lead
exposure in primates* Also relevant are recent mechanist;r
studies, based on Inferences from neuroscience, on the crif.: >.,
events in neuroqenesis within th* late fatal and early n«orj- :.
periods, during which lead may have irreversible effects. ' -
together, these experimental data are th* basis for propos:-:
1 Rie«, D. C. and Karpinski, K. P. "Ufatim*
Exposure Produces Deficits in Delayed Alternation
Monkeys," Heugoteacieoi. Teratal.. 1Q1207-214, 1911
-------
that there is likely to be no threshold for lead neurotoxicity,
at least witliin the contemporary rang* of blood lead levels
(i.e., 1-19 tff/dl).
The use of mechanistic data in risk assessment is a long
established practice for carcinogens, The mechanistic data on
lead call for an approach in assessing neurobehavioral'effects
that goes beyond the customary practice of applying standard
uncertainty factors to LOAELs (Lowest observed Adverse Effects
Leval) or NQAELs (No observed Adverse Effects Level) in
developing a health-based goal for neurotoxins. Thus,
mechanistic models for risk assessment deserve to be explored :.-
this context.
4*3 Risk and .Exposure
4.3*1 Highj-Sisk SubigQBUlatj,onjjjt. The most vulnerable populations
may differ with respect to common sources of lead. Kith regard
to exposure to lead in dust and drinking water during early life,
for example, sensitive populations vary for several reasons.
Mobility and normal hand-to-mouth activity make the 12-36 month
old most sensitive to lead in dust. ly contrast, water
requirements are highest from birth to three months of age,
primarily because of high growth rate, high metabolic rate, and
age-dependent differences in renal function* Therefore, three
months should be selected as the critical age for a drinking
water lead standard. At the geriatric end of the age spectru-
other considerations apply. For these reasons careful
consideration of age differences oust be given in assessing t^e
'risk of toxicity of lead from each source,
4.3.2 ^xpejurf-Pqfe-_Ji.e\antiLonsftipf. ' T&« mother-fetal unit
represents a particularly -sensitive population; however,' one
should not focus exclusively on this population. There are
data sets on fetal tissue that could be used to address this
question, particularly the work reported by Barlthrop, et al ,
In addition, research Is now in progress on Australian
populations near lead smelters which should provide some of the
data that are needed.
.The multi-media model should b* of such a nature as to
recognize that the age of the sensitive group may change with -r.s
medium. That is, water may be the most important route of
exposure at three months; later on, dust. The amounts of ti-a
spent indoor* and 'outdoors also vary greatly with age and should,
therefore, be considered also.
Transfer of 'Lead to The Human Fetus. Barlthrop, D.,
L. f Ed, in Minersl_ntffe£abqlifm yn fi^^iyi^tt^ pg» 135*151. l
Publishing, Philadelphia PA, 1969.
11
-------
speeiation i» important in distinguishing organic lead from
inorganic lead and fram a«taliie laad in terms of kinetics. it
is suspected, how«v*r, that subspeciation within th* inorganic
group »ay not be relevant. The influence of particle aiz* and
absorption characteristics also must b« considered, as w«ll as
the nature of the salt.
In further studies to correlate exposure with blood levels,
efforts should be mado to obtain as much information as possible
regarding the exposure condition* t by whatever route (s) , in =rde
to define th« relative source contributions more adequately.
.4.4 (^rojtf^Aqenev Cons latency on Lead Exposures and Their HeaL*n
E£fects*«-cJftQPS,.QDig. And OHEA. The study Group received a
comprehensive briefing, as wall as extensive background
documentation, from the ORD Environmental Criteria And Assess-e.-.
Office (ECAO) and OAQPS staff on issues relating to exposure by
all routes, toxicofcinetics, and health effects including
neurobehavioral , developmental , cardiovascular, and carcinogen is
effects. It also received a brief oral presentation on drinHir.^
water issues froa ODW staff, and had an opportunity to ask
questions about consistency and differences between the
scientific and regulatory considerations affecting approaches to
health protection used for ambient air and drinking water. 3y
contrast, there was neither background documentation nor an-site
representation froa the offices dealing with lead in pesticides
and toxic substances, solid waste, super fund* or emergency
response. Thus, the extent of consistency and coordination ar.cr.
these offices and the air, drinking water, and research programs
within the agency could not be determined fully. This was
'particularly unfortunate, since human exposures to lead from
sources associated with the program*, not' reviewed may rise. *
Th* following summary, while incomplete for the; reasons
stated above, indicates some important differences in the
procedure* followed by OHEA, OAQPS, and oow.
4*4.1 Health Effaets Pata Us«d to gupport Regulatory Actions.
All three groups consider neurob«havioral effects in infants and
young children to be the most important basis for regulatory
action • OKIE and ODW giv* more prominence to potential
carcinog«nicity than OAQPS. on tti* other Hand, OAQPS and ocw
give consideration to cardiovascular effects, while OHEA did ~ci
One cause for differences may lie in the data base utilized.
it was also noted that many of the papers used by OHIA to suppor
their analysis of carcinogen ic potential would not meet the
acceptance criteria for standard-setting for the NAAQS. All
three agreed on the use of blood lead as an indicator of expcs-r
and as a surrogate index of health effects.
y
4.4.2 Target Valuta fay Blood Laad. Both OAQPS and OOW have
identified 10-15 ng/d& as the area in which, adverse health
12
-------
effects hava been obaerved. However, no threshold naa been
demonstrated-.
4.4.3 Population at i^ak. OAQPS, ODW and QHZA agree that:
children are the population of primary concern (for
neurobehavioral effects) . However, there appear to b* seme
differences ~in the age rang* for primary 'concern.
4.4.4 ___ _ _
specifies protection" of 95% of the population from exposure to
lead in water > 2Qj*g/L. The Study Group estimated that 20 -g/L
would correspond to a blood lead level of 8 wg/dL in children
-with no other source* of lead exposure. QAQPS is considering 3 •
HAAQS in the range of OtS to l.S j*g/ar maximum monthly average,
A monthly NAAQS between 0.5 and l.S ug/m would keep more than
99,9% of the non-pica children, living without" lead-paint hazards
near the Dallas and E. Helena smelters, below a blood lead level
of 15 P9/&L, while a NAAQS of 0.5 »q/w would protect 99.97% of
affected children from reaching a blood lead of is iig/dL and keep
98,8-91.01 below 10 pg/dL (QAQPS Staff taper p» IV-24) , with rr.cs::
of the blood lead attributable to. non- inhalation sources. An
exact comparison is not possible, but clearly a NAAQS at or be_=w
l.S Mg/n is considerably more conservative than the proposed MCL
for drinking water.
4.4*5 Expo aiigp t-fa d e I a . The exposure model developed by OAQ?s .$
being used by ODW and OHEA, What is not clear is how an
acceptable total blood lead of 10 or 15 wg/dL is allocated t?
each source of exposure. For «xample, as lead dispersion frcr.
the operation of incinerators and from the leaching of landf^i.s
increases,, will the allocations allowed for drinking water
decrease? (Systematic evaluation of the distribution of blcci
lead levels by census tract sight be useful in defining the
contributions by various source*.) It is also not clear vhetr.er
the models are being used in a. uniform fashion. For example, ire
there differences in the coefficients for bioavailability? "--er
question* that need to be pursued further are the various
programs approaches to sampling and analysis issues, statist in,
analyses* time-activity-patterns, dietary factors, etc.
4.4*6 Ri«teT«du<±tion s^a^ecrifs. It i* recognized that
different strategies are needed to reduce ambient air and
drinking watar exposures, and that the more site-specific
exposures associated vita waste disposal and transportation
spill* need still different approaches. The development of
Agency -wide risfc reduction strategy for l«*d would benefit, f
study of the cost-ef fectiv«n«3s of various means of reducing
population blood lead levels,
s.O
The Study Group strongly recomswRd* t&ftt EPA d«ve>lop 3
13
-------
national policy on the reduction of blood lead levels in children
specifically, and all people in general, since current general
population blood lead levels are with the range in which adverse
health effects have been observed. They are attributable to
numerous sources and exposures by multiple routes, and are, for
many people, within a range associated with demonstrated and
substantial health effects. It is also strongly race-amended that
the Agency obtain mass-balance data on lead in environmental
^edia and people for past, current, and future exposures. These
data would provide information on source strengths, pathways,
fluxes, bodily uptake and retention, etc*, to guide the
implementation of a cost-effective risk reduction strategy.
EPA should assess environmental exposure to lead in terns cf
the risk of adverse neurological effects in children, with tnxs
approach, no extrapolations from animal to man, or from extrenely
high to environmentally realistic exposure levels are required.
Well documented neurobehavioral effects are observed in husar.s,
and at exposure levels close to those of regulatory concern."
Effective regulation of such exposures should insure that bleed
lead levels remain low enough to'limit most risks to most of -~e
adult population for these and other adverse effects.
. finally, the Joint Study Group recommends that the EPA
Administrator take action to assure that all components withxr.
The Agency are employing consistent scientific approaches to i^ii
with lead-relates *ssaes, and working from equivalent, up-to-
date, information bases. We believe that our' review contric--^;
to this goal, but, because several organizations did not
participate in the effort, we are unable to draw any conelusi — s
about the over-all level of integration, to the best of our
Knowledge, no, mechanism to bring about such coordination now
existst nor is it within the purview of the Joint Study Group —
suggest one, but we regard this need a* vital if the Agency ;s -3
carry out its regulatory responsibilities' via-a-via lead in a
technically sound and consistent manner.
*M.A. Smith, L.D. arant, and JUX. Sors, *t**4 Exposure *-
•child Development—An International A*»e*saent," Kluwer AC.*;*?.
Press, 1989, Boston, Mass.
14
------- |