v>EPA
United States
Environmental Protection
Agency
Office of Health and
Environmental Assessment
Washington DC 20460
EPA/600/6-87/006
August 1987
Research and Development
Carcinogenicity
Assessment of
Aldrin and
Dieldrin
-------
-------
EPA/600/6-87/006
August 1987
CARCINOGENICITY ASSESSMENT OF
ALDRIN AND DIEtDRIN
Carcinogen Assessment Group
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC
-------
DISCLAIMER
This document has been reviewed in accordance with U.S. Environmental
Protection Agency policy,and approved for publication. Mention of trade names
or commercial products does not constitute endorsement or recommendation for
use.
-------
CONTENTS
Tables. . . . . . . . . . . ...... . . ... .... . . . ... . . . vii
Figures x
Preface ...... , . . . . . .,. . ............ xi
Abstract xii
Authors, Contributors, and Reviewers xiii
1. SUMMARY AND CONCLUSIONS 1-1
1.1. SUMMARY 1-1
1.1.1. Qualitative Assessment. 1-1
1.1.1.1. Human Studies . 1-1
1.1.1.2. Animal Studies 1-2
1.1.1.3. Mutagenicity 1-4
1.1.1.4. Structural Relationship 1-5
1.1.2. Quantitative Analysis 1-5
1.1.2.1. Aldrin 1-5
1.1.2.2. Dieldrin 1-6
1.2. CONCLUSIONS 1-7
1.2.1. Aldrin 1-7
1.2.2. Dieldrin . 1-8
2. INTRODUCTION 2-1
3. GENERAL BACKGROUND INFORMATION 3-1
3.1. CHEMICAL AND PHYSICAL PROPERTIES 3-1
3.1.1. Identification. . 3-1
3.1.2. Synonyms 3-1
3.2. USES. 3-3
sis! ROUTES'AND'PATTERNS' OF EXPOSURE .' .* .' .* .* .' .' . .' .' .* .* .* .* i 3-4
4. HAZARD IDENTIFICATION 4-1
4.1. METABOLISM AND PHARMACOKINETICS 4-1
4.1.1. Absorption 4-1
4.1.1.1. Aldrin 4-1
4.1.1.2. Dieldrin 4-4
iii
-------
CONTENTS (continued)
4.1.2. Tissue Distribution 4-5
4.1.2.1. Aldrin 4-5
4.1.2.2. Dieldrin 4-6
4.1.3. Metabolism 4-13
4.1.3.1. Aldrin 4-13
4.1.3.2. Dieldrin 4-15
4.1.4. Excretion 4-21
4.1.4.1. Aldrin . 4-21
4.1.4.2. Dieldrin 4-24
4.2. TOXIC EFFECTS . 4-27
4.2.1. Human Studies 4-27
4.2.1.1. Aldrin 4-27
4.2.1.2. Dieldrin 4-28
4.2.2. Laboratory Animal Studies .... 4-30
4.2.2.1. Acute Studies 4-30
4.2.2.2. Subchronic Studies 4-30
4.2.2.3. Chronic Studies 4-32
4.2.2.4. Teratology and Reproduction Studies .... 4-39
4.3. MUTAGENICITY 4-43
4.3.1. Aldrin 4-44
4.3.2. Dieldrin 4-45
4.4. CARCINOGENICITY 4-46
4.4.1,. Animal Studies 4-47
4.4.1.1. Aldrin 4-47
4.4.1.2. Dieldrin 4-66
4.4.2. Epidemiologic Studies ..... 4-103
4.4.2.1. Van Raalte (1977) 4-103
4.4.2.2. Ditraglia et al. (1981) ..... 4-104
4.5. STRUCTURE-ACTIVITY RELATIONSHIPS. .............. 4-106
iv
-------
CONTENTS (continued)
4.5.1. Chlordane 4-106
4.5.1.1. Evidence for Carcinogenicity - Human
Exposure (IARC, 1979a) 4-106
4.5.1.2. Evidence for Carcinogenicity - Animal
Studies (IARC, 1979a) ...... 4-106
4.5.1.3. Evidence for Activity in Short-Term
Tests (IARC, 1979a) 4-108
4.5.2. Chlorendic Acid . . . 4-108
4.5.2.1. Evidence for Carcinogenicity - Human
Exposure (NTP, 1985b) 4-108
4.5.2.2. Evidence for Carcinogenicity - Animal
Studies (NTP, 1985b) 4-108
4.5.2.3. Evidence for Activity in Short-term
Tests (NTP, 1985b). 4-108
4.5.3. Endrin. 4-109
4.5.3.1. Evidence for Carcinogenicity - Human
Exposure (IARC, 1974c). ,,,.,..... 4-109
4.5.3.2. Evidence for Carcinogenicity - Animal
Studies (NTP, 1985b) 4-110
4.5.3.3. Evidence for Activity in Short-term
Tests (NTP, 1985b). 4-110
4.5.4. Endosulfan 4-110
4.5.4.1. Evidence for Carcinogenicity - Human
Exposure (NTP, 1985b) 4-110
4.5.4.2. Evidence for Carcinogenicity - Animal
Studies (NTP, 1985b) 4-110
4.5.4.3. Evidence for Activity in Short-term
Tests (NTP, 1985b) 4-110
4.5.5. Heptachlor 4-110
4.5.5.1. Evidence for Carcinogenicity - Human
Exposure (IARC, 1979b) 4-110
4.5.5.2. Evidence for Carcinogenicity - Animal
Studies (IARC, 1979b) . . . . 4-111
4.5.5.3. Evidence for Activity in Short-term
Tests (IARC, 1979b) 4-111
4.5.6. Hexachlorocyclopentadiene 4-111
4.5.6.1. Evidence for Carcinogenicity - Human
Exposure (U.S. EPA, 1984) 4-111
-------
CONTENTS (continued)
4.5.6.2. Evidence for Carcinogenicity - Animal
Studies (U.S. EPA, 1984) 4-112
4.5.6.3. Evidence for Activity in Short-term
Tests (U.S. EPA, 1984) 4-112
4.5.7. Isodrin 4-112
4.5.7.1. Evidence for Carcinogenicity - Human
Exposure (HSDB, 1985) 4-112
4.5.7.2. Evidence for Carcinogenicity - Animal
Studies (HSDB, 1985). ...,,,..... 4-112
4.5.7.3. Evidence for Activity in Short-term
Tests (HSDB, 1975) 4-112
5. RISK ESTIMATION FROM ANIMAL DATA 5-1
5.1. SELECTION OF DATA 5-1
5.1.1. Aldrin 5-1
5.1.2. Dieldrin 5-2
5.2. CHOICE OF EXTRAPOLATION MODELS 5-2
5.3. INTERSPECIES DOSE CONVERSION 5-3
5.4. ESTIMATION OF CARCINOGENIC POTENCY 5-4
5.4.1. Aldrin 5-4
5.4.2. Dieldrin 5-24
6. REFERENCES 6-1
APPENDIX A: DATA EVALUATION RECORD: rERMITICIDES—ALDRIN
MUTAGENICITY--MULTIPLE GENETIC TOXICOLOGY STUDIES A-l
APPENDIX B: DATA EVALUATION RECORD: TERMITICIDES—DIELDRIN
MUTAGENICITY--MULTIPLE GENETIC TOXICOLOGY STUDIES B-l
-------
TABLES
3-1 Chemical and physical properties of aldrin and dieldrin ... 3-2
4-1 Distribution of radioactivity in male rats 1 and 82 days
after dosing with [l^C]-aldrin for 3 months 4-7
4-2 Relative concentrations of excretion products in male rats
fed aldrin for 12 weeks and observed for an additional 12
weeks 4-23
4-3 Toxicity of aldrin following long-term feeding to mice,
rats, and dogs 4-33
4-4 Toxicity of dieldrin following long-term feeding to
mice, rats, and dogs 4-36
4-5 Summary of dietary carcinogenesis bioassays of aldrin in mice 4-48
4-6 Carcinogenesis bioassay of aldrin and dieldrin in C3HeB/Fe
mice exposed by dietary mixture 4-51
4-7 Survival data and liver lesions in C3H mice following two-
year dietary exposure to aldrin and dieldrin 4-52
4-8 Reevaluation of liver lesions of the FDA carcinogenesis
bioassays of aldrin and dieldrin. 4-54
4-9 Incidences of hepatocellular carcinoma in B6C3F1 mice
exposed to aldrin in dietary mixtures ............ 4-56
4-10 Summary of dietary carcinogenesis bioassays of aldrin in rats 4-57
4-11 Incidence of liver lesions and tumors as related to percent
survival in Osborne-Mendel rats administered aldrin in the
diet 4-60
4-12 Tumor incidences in Osborne-Mendel rats fed aldrin in the
diet 4-63
4-13 Summary of dietary carcinogenesis bioassays of dieldrin in
mice. . 4-67
4-14 Incidences of tumors in CF-1 mice administered dieldrin in
the diet for 132 weeks 4-77
4-15 Incidences of tumors in CF-1 mice administered dieldrin in
the diet for 128 weeks 4-79
4-16 Incidences of liver tumors in CF-1 mice administered 10 ppm
dieldrin in the diet for various periods of time 4-81
vii
-------
TABLES (continued)
4-17
4-18
4-19
4-20
4-21
4-22
5-1
5-2
5-3
5-4
5-5
5-6
5-7
5-8
5-9
5-10
5-11
Incidences of liver tumors in CF-1 mice administered 10 ppm
dieldrin in the diet for 110 weeks
Incidences of hepatocellular carcinomas in B6C3F1 mice
exposed to dieldrin in dietary mixture
Incidences of liver tumors in CF-1 mice administered 10 ppm
dieldrin in the diet for 110 weeks
Induction of liver tumors in CF-1 mice administered dieldrin
in the diet for lifetime exposure
Summary of tumors observed in C57BL/6J, C3H/He, and B6C3F1
mice treated with dieldrin in the diet at 10 ppm .
Summary of dietary carcinogenesis bioassays of dieldrin in
Cancer data sheet for derivation of potency of aldrin from
tumor incidence in mice (Davis, 1965)
Cancer data sheet for derivation of potency of aldrin from
tumor incidence in mice (Davis and Fitzhugh, 1962). .....
Cancer data sheet for derivation of potency of aldrin from
tumor incidence in mice (NCI- 1978a)« < , , , , ,,,,,.
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Davis, 1965)
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Davis, 1965)
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Walker et al., 1972)
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Walker et al., 1972)
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Walker et al., 1972)
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Walker et al . , 1972)
Cancer data sheet for derivation of potency of dieldrin from
4-83
4-85
4-87
4-89
4-91
4-92
5-5
5-7
5-8
5-9
5-10
5-11
5-12
5-13
5-14
5-15
5-16
vm
-------
TABLES (continued)
5-12
5-13
5-14
5-15
5-16
5-17
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Thorpe and Walker, 1973)
Cancer data sheet for derivation of potency of dieldrin from
Cancer data sheet for derivation of potency of dieldrin from
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Meierhenry et al . , 1983)
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Meierhenry et al ., 1983)
Cancer data sheet for derivation of potency of dieldrin from
tumor incidence in mice (Meierhenry et al., 1983')
5-17
5-18
5-19
5-20
5-21
5-22
-------
FIGURES
4-1 Chemical structure and numbering system of aldrin 4-2
4-2 Chemical structures of dieldrin metabolites ......... 4-10
4-3 Metabolic pathways of aldrin 4-14
4-4 Metabolic pathways of dieldrin. 4-16
4-5 Chemicals selected for structure-activity analysis 4-107
-------
PREFACE
The Carcinogen Assessment Group within the Office of Health and Environ-
mental Assessment has prepared this carcinogenicity assessment at the request
of the Hazard Evaluation Division within the Office of Pesticides and Toxic
Substances. The assessment contains a qualitative and quantitative evaluation
of the carcinogenicity data available as of November 1985.
xi
-------
1.1.1.2.2. Dieldrin. Twelve carcinogenicity bioassays of dieldrin have been
conducted with mice; all but one were judged adequate in design. Seven strains
of mice were used: CsHeB/Fe, CsH, CBA/J, Swiss-Webster, CFi, B6C3Fi, and
C57BL/60. In all studies, either benign or malignant liver tumors were ob-
served. The authors of six of these studies indicated that the tumors were
malignant (hepatocellular carcinomas), whereas in five others the tumors were
diagnosed as hepatomas. In three of these five, a pathology reevaluation was
performed by other pathologists who classified the lesions as being malignant.
In three studies, all with CF^ mice, many of the hepatocellular carcinomas had
metastasized to the lung. In addition, in one study, a significant (p < 0.05)
reduction in latency period was observed. In one of the studies with CF^ mice,
a slight but significant (p < 0.05) increase in the incidences of pulmonary
adenomas and carcinomas, lymphoid tumors,.and other tumors was seen in female
mice at 1 ppm. The doses used in the bioassays ranged from 0.1 to 20 ppm
with dietary administration for 18 to 24 months.
Seven long-term carcinogenicity studies were conducted with rats, three
of which were considered adequate carcinogenicity assays. The others suffered
mainly from too few animals, too high mortality, too short a duration, and/or
inadequate pathology examination or reporting. Four strains were used: Car-
worth, Osborne-Mendel, Holtzman, and Fischer 344. Doses ranged from 0.1 to 285
ppm in the diet generally for 2 years or more. Although liver pathology was
generally associated with exposure to the chlorinated insecticides, there was
no firm evidence of a carcinogenic response in any of the studies.
1.1.1.3. Mutagenicity—
1.1.1.3.1. Aldrin. An evaluation of mutagenicity is hampered by a paucity of
data on point mutations and cytogenetics. However, the available data indicate
that aldrin is probably not mutagenic in procaryotes and does not induce DNA
1-4
-------
damage or clastogenic effects. A report of induced unsceduled DNA. synthesis
(UDS) was discounted due to technical deficiencies.
1.1.1.3.2. Di'eldrin. The data are insufficient to establish a complete genetic
toxicology profile. However, the results of 13 studies support the conclusion
that dieldrin is not mutagenic in procaryotes. However, two studies reported
positive results, one with S. typhimurium and the other with Chinese hamster
V79 cells. Both, however, were seriously flawed so that the data are of ques-
tionable value. However, based on these data, dieldrin is classified as a pre-
sumptive mutagen for S. typhimurium and for Chinese hamster V79 cells. Dieldrin
was reported to be weakly clastogenic in mammalian somatic cells and nonclasto-
genic in male germinal cells. There is no evidence for genotoxicity in yeast
or primary rat and mouse hepatocytes or for primary DNA damage in bacteria. A
positive result of UDS in SV-40 transformed cells is questionable based upon
numerous technical deficiencies in the study.
1.1.1.4. Structural Relationship—Three compounds, structurally related to
aldrin and dieldrin, have induced malignant liver tumors in animals. Chlordane,
heptachlor, and chlorendic acid have produced liver tumors in mice, and chlo-
rendic acid has also produced liver tumors in rats.
1.1.2. Quantitative Analysis
1.1.2.1. Aldrin—Three data sets are suitable for quantitative risk estimation:
male and female C^tt mice, and male B6C3Fi mice. The most sensitive sex and
strain tested is female C^ti mice. From these the potency can be estimated at
23 per mg/kg/day.
The most sensitive species tested is mice. There are three potency esti-
mates, ranging from 23 down to 12 per mg/kg/day, with a geometric mean of 17 per
mg/kg/day. Because humans may be as sensitive as the most sensitive animal spe-
cies, the potency for the general population is estimated at 17 per mg/kg/day.
1-5
-------
These estimates are plausible upper bounds for the increased cancer risk
from aldrin, meaning that the true risk is not likely to exceed these estimates
and may be lower. These estimates have been calculated using the Agency's
preferred methodology in the absence of specific physiologic, metabolic, or
kinetic information: the linearized multistage model with doses scaled accord-
ing to relative body surface area. Other assumptions about the dose-response
model or the interspecies dose adjustments may result in lower estimates.
The molecular potency index, which is the potency expressed in terms of
molecular weight, has been used to rank suspect carcinogens according to poten-
cy. The index is computed by multiplying the general-population potency by the
molecular weight. The molecular potency index for aldrin is 6.2 x 10^ per mmol/
kg/day. This places aldrin in the most potent quartile of suspect carcinogens
ranked by the Carcinogen Assessment Group (CA6).
1.1.2.2. Dieldrin—Thirteen data sets are suitable for quantitative risk esti-
mation. The most sensitive sex and strain tested is male CF^ mice. From these
the potency can be estimated at 55 per mg/kg/day. This estimate may, however,
be misleading because 100% of the treated animals developed cancer. The second
most sensitive estimate is 28 per mg/kg/day.
The most sensitive species tested is mice. There are 13 potency estimates,
ranging from 55 down to 7 per mg/kg/day, with a geometric mean of 16 per mg/kg/
day. Because humans may be as sensitive as the most sensitive animal species,
the potency for the general population is estimated at 16 per mg/kg/day.
These estimates are plausible upper bounds for the increased cancer risk
from dieldrin, meaning that the true risk is not likely to exceed these esti-
mates and may be lower. These estimates have been calculated using the Agency's
preferred methodology in the absence of specific physiologic, metabolic, or
kinetic information: the linearized multistage model with doses scaled accord-
1-6
-------
ing to relative body surface area. Other assumptions about the dose-response
model or the interspecies dose adjustments may result in lower estimates.
The molecular potency index for dieldrin is 6.1 x 103 per mmol/kg/day.
This places dieldrin in the most potent quartile of suspect carcinogens ranked
by the CA6.
1.2. CONCLUSIONS
1.2.1. Aldrin
For aldrin, the finding of hepatocellular carcinomas in male B6C3Fj mice
and the finding of hepatomas in C$\ and CsHeB/Fe mice, which were later diag-
nosed as hepatocellular carcinomas, in both males and females, constitute as
a first approximation, "sufficient" evidence of carcinogenicity in animals,
according to criteria in the EPA's Guidelines for Carcinogen Risk Assessment.
However, the Guidelines call for a careful analysis of the nature of the mouse
liver tumor only response to ascertain whether there is a sufficiently strong
reason to downgrade the evidence to "limited." This downgrading was found not
to be appropriate since the tumor increases were not marginal in male B6C3Fi
mice (which have a high spontaneous incidence of liver tumors), there was a
dose-related increase in the proportion of tumors that were malignant, and the
response occurred in both males and females. In addition, the response occurs
in CsH and C3HeB/Fe mouse strains with a low spontaneous incidence of liver
tumors, which is a response not subject to the downgrading factors.
One factor which would argue for a "limited" classification is that the
two CsH strains are genetically closely related and the B6C3Fi strain derives
half of its gene pool from these strains. However, the fact that the epoxide
metabolite, dieldrin, produces liver tumors in seven different mouse strains
indicates strongly that the carcinogenic effect of aldrin is not limited to a
restricted genetic range of mice. Although in rats, one adequate study with
1-7
-------
aldrin and three adequate studies with dieldrin were negative or equivocal,
the multiplicity of mouse strains responding is considered to be enough for
classification of "sufficient" evidence in animals.
In the absence of adequate evidence in humans, the sufficient animal evi-
dence amounts to an overall classification of B2, or "probable" carcinogen for
aldrin.
For aldrin, the carcinogenic potency, obtained by averaging estimates from
the most sensitive species tested (mice), is 17 per mg/kg/day. The potency
using the most sensitive sex and strain is 23 per mg/kg/day. These are plaus-
ible upper bounds for the increased cancer risk from aldrin, meaning that the
true risk is not likely to exceed these estimates and may be lower. These
estimates have been calculated using the Agency's preferred methodology in
the absence of specific physiologic, metabolic, or kinetic information: the
linearized multistage model with doses scaled according to relative body sur-
face area. Other assumptions about the dose-response model or the interspecies
dose adjustment may result in lower estimates. The molecular potency index for
aldrin is 6.2 x 10^ per mmol/kg/day. This places aldrin in the first (most
potent) quartile of suspect carcinogens ranked by the CAG.
1.2.2. Dieldrin
For dieldrin, the mouse liver tumor response in seven mouse strains
Fe, CsH, CBA/J, CF^, Swiss-Webster, B6C3Fi, and C57BL/6J) along with the appear-
ance of pulmonary adenomas and carcinomas and lymphoid tumors in one strain
(CFi) would justify the preliminary classification of dieldrin as having "suf-
ficient" evidence in animals. Based on the nature of the response there is no
reason for downgrading the classification to "limited," since the carcinoma
response is unmistakably strong, it occurs in both males and females, and at
both high and low doses. The occurrence in several strains implies that it is
1-8
-------
not a genetically isolated finding. The carcinogenicity studies in rats for
dieldrin were considered negative. However, the ability of most studies to
detect carcinogenicity was compromised due to too few animals, too high mor-
tality, and too short duration. Although liver pathology was associated with
exposure to the chlorinated insecticides, there was no firm evidence of a
carcinogenic response. In the absence of adequate evidence, the sufficient
animal evidence amounts to an overall classification of "B2" or probable car-
cinogen for dieldrin.
For dieldrin, the carcinogenic potency, obtained by averaging estimates
from the most sensitive species tested, is 16 per mg/kg/day. The potency using
the most sensitive sex and strain, is 55 per mg/kg/day. This estimate may,
however, be misleading because 100% of the exposed animals developed cancer.
The second most sensitive estimate is 28 per mg/kg/day. These are plausible
upper bounds for the increased cancer risk from dieldrin, meaning that the true
risk is not likely to exceed these estimates and may be lower. These estimates
have been calculated using the Agency's preferred methodology in the absence of
specific physiologic, metabolic, or kinetic information: the linearized multi-
stage model with doses scaled according to relative body surface area. Other
assumptions about the dose-response model or the interspecies dose adjustment
may result in lower estimates. The molecular potency index for dieldrin is
6.1 x 10-3 per mmol/kg/day. This places dieldrin in the first (most potent)
quartile of suspect carcinogens ranked by the CAG.
Although there is always uncertainty in extrapolating potency from animals
to humans, our confidence in these animal potency estimates is relatively
high. This confidence is based not only on the tight clustering of aldrin and
dieldrin potency estimates, but on other factors as well. The potency estimates
for aldrin are consistent with the potency estimates for dieldrin, one of its
,1-9
-------
metabolites. In addition, aldrin's potency (and dieldrin's as we'll) is based
in part on studies of mice with low background tumor rates. A study of female
C3HeB/Fe mice, for example, saw the tumor incidence increase from 4% in controls
to 85% in the treated group. Potencies derived from high-background strains are
consistent with potencies from low-background strains. Furthermore, potencies
derived from mouse liver tumors are consistent with potencies derived from rat
tumors for two other related substances, chlordane and heptachlor epoxide.
Thus, mouse liver tumors can provide a basis for estimating cancer potency in
humans, recognizing the current issues associated with use of this type of data
and attendant uncertainties in a risk estimate based upon it.
1-10
-------
2. INTRODUCTION
Aldrin and dieldrin were first synthesized in the laboratory in 1948 with
commercial production in the United States first reported in 1950 (IARC, 1974a,
b). Aldrin and dieldrin are organochlorine pesticides extensively used in the
1960s and early 1970s for a large variety of pesticidal uses. Most uses were
banned in the United States in 1974 under the Federal Insecticide, Fungicide,
and Rodenticide Act (39 FR 37246, October 18, 1974). Uses were also restric-
ted in many other countries at about the same time. The approved uses in the
United States now are mainly for termite control by direct soil injection and
for nonfood seed and plant treatment. Production of aldrin and dieldrin in
the United States was also discontinued along with the restrictions of 1974,
so that virtually all aldrin and dieldrin now used in the United States is
imported. Exposure of humans has been primarily from dermal and inhalation
exposure related to the application as pesticides; however, due to the per-
sistence and bioaccumulation of dieldrin, considerable exposure has occurred
through ingestion of contaminated water and food products. In contrast, little
human exposure from environmental sources of aldrin occurs, as it is readily
converted to dieldrin by-direct epoxidation in the environment. During the
early 1970s, nearly all humans sampled had measurable tissue levels of dieldrin
and, due to its stability and persistence, dieldrin is still detected in humans.
However, the levels are considerably lower. Aldrin and dieldrin are usually
considered together, as they are structurally-related cyclodiene insecticides,
have similar uses, and because aldrin is readily epoxidized to dieldrin in the
body or the environment.
It is the purpose of this report to review the currently available litera-
ture on the toxicity and carcinogenicity of both aldrin and dieldrin and conduct
2-1
-------
an assessment as to the potential careinogenicity risk to humans. The Carcino-
gen Assessment Group (CA6) conducts a "weight-of-evidence" evaluation that
weighs data concerning innate biological activity including metabolism, toxic
effects as related to duration and levels of exposure, results of mutagenicity
and other short-term tests, long-term animal bioassays, and, where available,
data derived from epidemiologic studies. Since mutagenicity is considered as
one major mechanism for cancer induction, those studies are considered suppor-
tive to the in vivo animal or human data.
Following a review of the data, those data most relevant and amenable to
mathematical extrapolation are selected for the risk assessment. Several
mathematical models, as described in EPA's Guidelines for Carcinogen Risk
Assessment (U.S. EPA, 1986) are used. On the basis of the '"weight of evidence"
and risk extrapolation, the substance is classified as a potential human car-
cinogen according to the EPA guidelines.
2-2
-------
3. GENERAL BACKGROUND INFORMATION
3.1. CHEMICAL AND PHYSICAL PROPERTIES
Those properties most relevant to carcinogenesis and human exposure are
listed in Table 3-1.
3.1.1. Identification
DieSdrin
Molecular Formula:
Molecular Weight:
C12H8C16
364.93
C12H8C160
380.93
60-57-1
CAS Number: 309-00-2
3.1.2. Synonyms (IARC, 1974a, b)
3.1.2.1. Aldrin—
I,2,3,4,10.10-hexach1oro-l,4>4a,8,8a-hexahydro-endo>exo-l,4:5,8-
dimethanonapthalene;
hexachlorohexahydro-endo-exo-dimethanonaphthalene;
1,2,3,4,10 JO-hexachloro-l,4s4a,5,8,8a-hexahydro-endp_-l,4,-exo-5,8"
dimethanonaphthalene;
3-1
-------
TABLE 3-1. CHEMICAL AND PHYSICAL PROPERTIES
OF ALDRIN AND DIELDRIN
Physical properties
Aldrin
Dieldrin
Boiling point
Melting point
Specific gravity
Vapor pressure
Solubility
Evaporation
Half-life (t1/2)
from water
Log octanol/water
partition
coefficient
Odor threshold
Conversion factor
NF - not found.
Decomposes
104°-105.5°C
(technical grade 49°-60°C)
NF
7.5 x lO'5 mmHg at 20°C
1.4 x 10-4 mmHg at 25°C
27 yg/L at 27°C in water
Soluble in most organic
solvents
185 hr at 25°C and 1 m
depth
3.01
0.017 mg/kg water
1 ppm = 14.96 mg/m3.
at 25°C and 760 mmHg
Decomposes
176°-177°C
1.75
3.1 x 10'6 mmHg at 20°C
7.78 x 10-7 mmHg at 25°C
186 yg/L at 25°C in water
Soluble in acetone and
benzene
12,940 hr at 25°C and 1 m
depth
NF
0.041 mg/kg water
1 ppm = 15.61 mg/m3
at 25°C and 760 mmHg
3-2
-------
l,2,3,4,10,10-hexachloro-l,4,4a,5,8,8a-hexahydro-l,4-ieriido-^xo-5,8-
dimethanonaphthalene;
l,2,3,4,10,10-hexachloro-l,4,4a,5,8,8a-hexahydro-^xo-l,4-^ndo-5,8-
dimethanonaphthalene
Compound 118 ENT 15,949
HHDN
3.1.2.2. Dieldrin—
endo-exo isomer of l,2,3,4,10,10-hexachloro-6,7-epoxy-r,4,4a,5,6,7,8,
8a-octahydro-l,4,5,8-dimethanonaphthalene
l,2,3>4,10>10-hexach1oro-6>7-epoxy-l,4,>4a>5>6,7>8,8a-octahydro-endo-exo-
1,4:5,8-dimethanonaphthalene
l,2,3,4,10,10-hexachloro-6,7-epoxy-l,4,4a,5,6,7,8,8a-octahydro-en
-------
Most uses in the United States were banned by the EPA in 1974. While aldrin
and dieldrin are approved for use on non-food seed and plants, the Chemical
Economics Handbook (1985) reported that current use is for termite control
exclusively, and that all quantities used are imported. The use of aldrin and
dieldrin has also been banned or highly restricted in many other countries
(IARC, 1974a, b).
3.3. ROUTES AND PATTERNS OF EXPOSURE
Exposure to aldrin and dieldrin can occur by virtually all major routes,
including inhalation, skin absorption, or by ingestion of contaminated food or
water.
The potential for inhalation exposure is greatest for pesticide applicators
or residents in buildings where termite treatment occurs. However, during the
period when there was extensive use of these pesticides, atmospheric pollution
was fairly common. In 1976, more than 85% of atmospheric air samples tested by
the EPA contained aldrin or dieldrin with levels as high as 2.8 ng/m^ resulting
in an intake of up to 0.098 yg/day (U.S. EPA, 1985). In the IARC (1974a, b)
review, the intake by occupationally exposed workers was estimated to range
from 0.72 to 1.10 mg/person/day compared to 0.025 mg/person/day for the general
population during 1967 and 1968.
Estimates of potential dermal exposure during orchard spraying ranged from
14.2 to 15.5 mg/hour, and the potential respiratory exposure was estimated to
be 0.03 to 0.25 mg/hour from 1963 to 1967.
Aldrin and dieldrin are widely applied to vast areas of agricultural land
and aquatic areas in the United States and in most parts of the world until
government restrictions were imposed. As a result, both fresh and marine waters
were contaminated, with exposure via ingestion of water being an important
route of exposure.
3-4
-------
Aldrin itself is readily converted to dieldrin. Dieldrin is stable,
highly persistent in the environment, lipophilic in nature, and accumulates in
the food chain. Due to these factors, the levels of aldrin in water and foods
have been low, with those for dieldrin considerably higher and more persistent.
Weaver et al. (1965) found dieldrin in all major U.S. river basins more often
than any other pesticide and at a mean concentration of 7.5 ng/L. In 1976,
dieldrin was present in many fresh waters in the United States at even higher
concentrations, ranging from 5 to 396 ng/L in surface waters and from 1 to 7
ng/L in drinking water. A 1975 survey in the United States found dieldrin in
117 of 715 water samples. Aldrin was also detected at concentrations of 15
to 18 ng/L (U.S. EPA, 1976). Harris et al. (1977) reported concentrations of
1 ng/L in Seattle, Washington, and Cincinnati, Ohio; 2 ng/L in Miami, Florida,
and Ottumwa, Iowa; and as high as 50 ng/L in New Orleans, Louisiana.
It has been estimated by MacKay and Wolkoff (1973) that dieldrin has by
far the longest half-life of the chlorinated hydrocarbons in water 1 meter (m)
in depth, calculating a half-life for aldrin and dieldrin of 10.1 days and 723
days, respectively, compared to 3.5 days for DDT and 289 days for lindane.
This long half-life in water combined with the potential for bipconcentration
by aquatic organisms further enhances the hazard of these two pesticides
(Wurster, 1971).
The EPA (1980) estimated that 99.5% of all human beings in the United
States had dieldrin residues in their tissues. Wurster (197l) claimed that
these residue levels were mainly due to contamination of foods of animal ori-
gin. The overall concentration of dieldrin in the diet in the United States
was calculated by Epstein (1976) to be approximately 43 ng/g of food consumed.
3-5
-------
-------
4. HAZARD IDENTIFICATION
4.1. METABOLISM AND PHARMACOKINETICS
Aldrin and dieldrin are related pesticides, and in biological systems,
aldrin is rapidly converted to its epoxide, dieldrin. For all practical pur-
poses, therefore, the kinetic functions and metabolism of these chlorinated
cyclodienes are identical. Part of the confusion regarding the metabolism of
these chemicals resides in the nomenclature. There are two systems that number
the ring carbons differently. The Chemical Abstracts Services (CAS) and Inter-
national Union of Pure and Applied Chemistry (IUPAC) systems define the chem-
icals as derivatives of octahydromethanonaphthalene, while the von Baeyer-IUPAC
system defines the chemicals as derivatives of tetracyclo[6.2.1.13,6,92,7]-
dodec-9-ene. Figure 4-1 shows the structures of aldrin for the two systems.
For this discussion, the von Baeyer-IUPAC system is used.
4.1.1. Absorption
4.1.1.1. A1drin—Quantitative data on the absorption of aldrin from the gas-
trointestinal tract of animals are not available. However, metabolic studies
indicate that aldrin is absorbed slowly via the portal venous system following
oral administration. A time-course study of the distribution and metabolism of
aldrin in the liver, kidneys, stomach, and small intestine of neonate (24-hour-
old) Sprague-Dawley rats following a single oral dose was studied by Farb et
al. (1973). The concentration of aldrin in the liver increased to about 13% of
the dose during the first 72 hours after dosing, whereas in 'the stomach and
small intestine, aldrin was detected through the sixth day after dosing.
Dermal absorption has not been specifically investigated in laboratory
animals, but acute and subchronic dermal toxicity studies indicate that percu-
taneous absorption does occur. Moreover, the rate of absorption is apparently
4-1
-------
C1 , C1
C1
C1
C1
Figure 4-1. Chemical structure and numbering system of aldrin:
(1) the von Baeyer-IUPAC system, and (2) CAS and IUPAC systems,
4-2
-------
affected by the vehicle used, such that the application of aldrin in Ultrasene
or oil is several-fold more toxic to rabbits than the application of aldrin as
a dry powder (Treon and Cleveland, 1955).
In humans, dermal absorption of aldrin was reported to be 7.8% +_ 2.9% of
the applied dose in acetone over a period of 5 days (Feldmann and Mai bach,
1974). However, the accuracy of this value is questionable because the dose
used in both the topical and intravenous (control study used to correct the
data of the topical application study) studies was only 1 yCi/subject, and
excretion was only determined in the urine and accounted for'"only 2% to 5% of
the dose following injection. In addition there was a large variability among
the subjects. Moreover, the rate of percutaneous absorptipn may be affected by
the amount of adipose fat in and/or the metabolizing capability of the skin.
The absorption of aldrin following inhalation exposure has been demonstra-
ted in humans (Beyermann and Eckrich, 1973; Bragt et al., 1984); about 20% to
50% of the inhaled aldrin vapor was retained. However, the data presented by
Beyermann and Eckrich (1973) were inadequate, and an unpublished study by Bragt
et al. (1984) was not available for review. Animal studies were not available,
but the uptake, metabolism, and disposition of aldrin by isolated, perfused
rabbit lungs was reported by Mehendale and El-Bassiouni (1975"). Recirculating
and single-pass experiments were conducted with an artificial medium perfused
through the pulmonary artery of ventilated lungs. Aldrin was taken up by
simple diffusion from the vascular space, and then quickly diffused into the
intercellular and intracellular spaces. Aldrin was then metabolized at a slow-
er rate to dieldrin in the endoplasmic reticulum of the lung; metabolism was
dose dependent, and up to 70% of the aldrin was epoxidized in 1 hour at low
doses (< 4 x 10"6 M). Dieldrin was then distributed into two pools, a saturable
and a nonsaturable pool. An equilibrium was established quickly with respect
4-3
-------
to the net uptake of dieldrin or aldrin. After steady-state conditions were
established, the disappearance of aldrin from the perfusate was attributed to
its metabolic turnover rather than to further uptake by the lung. There was no
evidence for irreversible binding or potential for long-term storage of aldrin
or dieldrin in the lung; they apparently can be completely effluxed from the
lung by simple diffusion into the blood.
4.1.1.2. Pi eldri n—Quanti tati ve data on the absorption of dieldrin from the
gastrointestinal tract of animals has not been reported. Absorption through
the gastrointestinal tract depends upon the vehicle used (Heath and Vandekar,
1964). Metabolic studies indicate that absorption is slow and initially occurs
via the portal venous system. Immediately following a single oral dose of
[36Qi]_ or [14c]_dieldrin to rats, little if any radioactivity was found in the
lymph system (Heath and Vandekar, 1964; latropoulos et a!., 1975). Large con-
centrations of radioactivity were still present in the stomach and small intes-
tine 24 to 48 hours after dosing, whereas concentrations in the liver decreased,
The subsequent redistribution of dieldrin among body tissues (from about 12
hours after dosing) apparently involves the lymphatic system (latropoulos et
al., 1975).
The in situ absorption of [14C]-dieldrin across the isolated small
intestine of rats also occurred at a slow rate. About 8% and 15% of the [-^C]
was absorbed after 1- and 5-hour incubations, respectively (Tanaka et al.,
j/
1981). /
The absorption of dieldrin through the skin occurs at levels great enough
to cause toxic effects, but quantitative data are not available. Most of the
dieldrin absorbed through the skin of guinea pigs, dogs, and monkeys is accumu-
lated in the subcutaneous fat (Sundaram et al., 1978a, b).
4-4
-------
Studies on the absorption of dieldrin following inhalation exposure were
not available. The absorption of dieldrin by isolated perfused rabbit lungs
was studied by Mehendale and El-Bassiouni (1975). The uptake of dieldrin was
rapid in the lung and there was no evidence for long-term storage. Apparently,
dieldrin can be completely effluxed from the lung by simple diffusion into the
blood.
The absorption of dieldrin in humans is expected to resemble that observed
in animals. In one survey conducted by Stacey and Tatum (1985) in Perth,
Australia, the effect of treating homes with organochlorine pesticides was
correlated with pesticide residue levels in human breast milk. Fourteen donors
provided milk on a monthly basis for a 1-year period. There was a strong
correlation between the dieldrin levels in the milk and the treatment of the
houses with aldrin.
4.1.2. Tissue Distribution
4.1.2.1. Aldrin—There is no direct information on the distribution or elimi-
nation of aldrin by humans following its transfer from the gastrointestinal
tract, or skin, into the circulating blood. In humans, aldrin is rarely if
ever found in human blood or other tissues, except in cases of acute poisoning
by accidental or intentional ingestion of massive doses.
A number of animal studies have been conducted which yield information on
tissue distribution. In one study by Farb et al. (1973), the concentrations of
aldrin in the liver decreased from 13% at 6 hours to 3% and 0.1% of the dose
after 24 and 72 hours, respectively. The only metabolite detected in aldrin-
treated animals was dieldrin, which was detected in the liver as early as 2
hours after dosing. The concentration of dieldrin increased, reaching a maxi-
mum of 31% of the administered dose at 18 hours, then rapidly decreased to less
than 10% at 48 hours. Dieldrin concentrations in the liver decreased slowly
4-5
-------
between 48 and 96 hours, leveling off at about 5% of the dose at days 4 to 7.
In the kidneys, unknown metabolite A and dieldrin were detected 2 and 6 hours
after dosing and through day 10; the distribution of metabolite A and dieldrin
in the stomach and small intestine was similar to that observed in the kidneys
(Farb et al ., 1973).
The distribution of [^C] in the carcass, fat, and other tissues of male
rats given daily oral doses of [^4C]-aldrin (4.3 yg/rat) for 3 months accouted
for 7.2% of the total administered dose. The [14C] stored in rats, calculated
as the difference between administered and excreted C^C], reached maximum
levels (steady-state) at 53 days. The estimated amount of [*4C] stored at this
time was 0.13 to 0.15 ppm of total rat weight (Ludwig et al., 1964). Residues
dropped to about 0.37% 82 days after the final dose, but the rates of elimina-
tion were not determined (Table 4-1). Most of the administered [^C] was
excreted in the feces and urine.
4.1.2.2. Dieldrin—The tissue distribution of dieldrin following ingestion
was investigated in two studies by Deichmann et al. (1969, 1975). In the
first, Deichmann et al. (1969) determined the concentrations of dieldrin in the
fat, liver, and blood of male beagle dogs given oral doses of aldrin in corn
oil at 0.6 mg/kg/day, 5 days/week for 10 months. The mean dieldrin concentra-
tion in blood during months 4 to 10 of dosing did not show any specific pattern
and ranged between 0.1 and 0.22 ppm. During the following 12 months, after
dosing was terminated, dieldrin concentrations gradually dropped to about 0.075
ppm. Fat and liver biopsies showed increasing concentrations of dieldrin in
these tissues during months 1 to 10 of dosing, and at 10 months, the levels
reached about 70 and 20 ppm, respectively. When aldrin administration was
discontinued, dieldrin concentrations gradually decreased, and by month I2f had
dropped to 25 and 6 ppm in fat and liver, respectively.
4-6
-------
TABLE 4-1. DISTRIBUTION OF RADIOACTIVITY IN MALE RATS
1 AND 82 DAYS AFTER DOSING WITH [14C]-ALDRIN FOR 3 MONTHS3
Total excretion:
Feces
Urine
Carcass
Fat
Other tissues
Percent
Day 1
80.80
7.90
3.60
1.77
1.83
of [14C] administered dose
after final dose
Day 82
(12 weeks)
90.25
9.29
0.21
0.11
0.05
aMean value from two rats.
SOURCE: Ludwig et al., 1964.
4-7
-------
In the second study, Deichmann et al. (1975) measured the concentration of
dieldrin in the abdominal fat and total extractable lipids from mice fed diets
containing 5 or 10 ppm aldrin for seven generations, except for the F4 genera-
tion. For each generation, residues were determined in the parents after 260 +_
15 days on the test diets. All groups of animals in the F]_, F£, and F3 genera-
tions showed significant increases in dieldrin retention. These concentrations
in mice were generally related directly to increases in total body lipids, and
were higher in females than in males. Dieldrin concentrations in total lipids
from animals of the P^, Fj_, ?2> anc* ^3 generations receiving 10 ppm of aldrin
ranged between 115 and 121 ppm in males and 149 and 159 ppm in females.
Tissue distribution of [^C] following oral distribution of [l4C]-dieldrin
to rats in a single dose indicated that the initial rapid uptake of C^C] by
the liver during the first 12 hours after dosing is followed by a biphasic
decrease and redistribution of dieldrin and/or its metabolites among body
tissues. Redistribution of radioactivity suggests the accumulation and storage
of unchanged dieldrin in body fat (Hayes, 1974; latropoulos et al., 1975).
Hayes (1974) reported that the concentrations of dieldrin in fat increased to
about 30% of the dose (10 mg/kg) by 8 hours after dosing and continued to in-
crease, reaching about 50% of the dose by 24 to 48 hours. Although residues in
the kidneys, brain, and muscle attained maximum levels about 2 to 4 hours after
dosing (0.08% to 4% of the dose), these concentrations remained at the same
levels 48 hours after dosing and then gradually decreased at an unspecified rate
(Hayes, 1974). There is some evidence that C^C] residue levels in the brain,
blood, liver, and subcutaneous fat and the profile of metabolites in urine and
feces are similar and independent of the dose following oral or intraperitoneal
administration of [14C]-dieldrin to rats (Lay et al., 1982).
Sex differences have been noted in the tissue distribution of [^C] resi-
4-8
-------
dues in rats 9 days after administration of a single oral dose at about 0.108
and 0.153 mg/kg in males and females, respectively. Greater [l^C] residue
levels were found in the muscle, fat, and liver of females when compared to
males, whereas residues in the kidneys and stomach were greater in males than
in females. In addition, all of the radioactivity in female tissues except the
stomach and small intestine contained unchanged dieldrin. In contrast, male
tissues contained various amounts of metabolites (< 1.0% to 85.6% of extract-
able radioactivity), namely trans-6,7-dihydroxydihydroaldrin, 9-hydroxydiel-
drin, dicarboxylic acid, and pentachloroketone (PCX) (Figure 4-2) (Matthews et
al., 1971).
Differences among species in tissue distribution of [^C] and/or metabo-
lites have also been reported for CFE rats and CF^ and LACG mice (Hutson,
1976). Although residues in only three tissues were determined, in general,
[l^C] residue concentrations in liver (about 1.5 ppm) and fat (12 to 14 ppm)
were greater in mice than in rats, whereas [^c] residue concentrations in
kidneys (3.5 ppm) were much greater in rats. Unchanged dieldrin was the major
constituent in the livers, with concentrations in mice (0.8 to 1 ppm) being 8-
to 10-fold those in rats. Relatively higher concentrations of 9-hydroxydiel-
drin were found in the livers of mice (0.27 to 0.51 ppm) and PCK in the kidneys
of rats (2.48 ppm). Fat tissues in rats and mice contained only dieldrin. In
addition, the data indicated that a relatively small amount of dieldrin may be
available at any given time for hepatic metabolism and enterohepatic circula-
tion in rats. Pretreatment of animals with dieldrin for 28 days prior to
dosing with [^4C]-dieldrin did not greatly alter the distribution and levels of
radioactivity in the tissues, but the concentrations of total (labeled and
unlabeled) dieldrin (as noted by GLC) and its metabolites increased, especially
in mice tissues (Hutson, 1976).
4-9
-------
H02-C-H
Figure 4-2. Chemical structures of dieldrin metabolites:
trans-6,7-dihydroxy-dihydroaldrin (I), 9-hydroxydieldn'n (II),
dicarboxylic acid (III), pentachloroketone (IV).
4-10
-------
Following repeated oral dosing (2 to 104 weeks), an equilibrium or steady
state was reached between the intake, storage, and excretion of dieldrin in
various strains of rats and beagle dogs (Deichmann et al..,. 1968; Walker et al.,
1969; Barron and Walton, 1971; Davison, 1973). Steady-state kinetics were
determined by measuring, at various time intervals, either total radioactivity
excreted or that found in fat, blood, and liver. In rats, a steady state was
reached 4 to 39 weeks after receiving diets containing various sublethal con-
centrations of dieldrin (0.04 to 50 ppm). Equilibrium was reached earlier in
rats receiving relatively higher doses of dieldrin, which the authors specula-
ted as possibly associated with the induction of microsomal metabolism. In
dogs, an approximate plateau in blood residue levels was noted after 12 to 18
and 18 to 30 weeks in dogs receiving 0.005 and 0.05 mg/kg, respectively (Walker
et al., 1969). Statistically significant relationships were found between the
concentrations of dieldrin in the blood, fat, liver, and brain of rats and dogs
receiving dieldrin orally at various concentrations (rats, 0, 0.1, 1, or 10 ppm
in the diet; dogs, 0, 0.005, or 0.05 mg/kg for 2 years [Walker et al.s 1969]).
The steady-state concentration in adipose tissue of male Osborne-Mendel
rats placed on diets containing 25 ppm of dieldrin for 8 weeks was reported to
be 50 mg/kg (50 ppm) dieldrin. The elimination of dieldrin residues from fat
was rapid when the rats were subsequently placed on untreated diets, the esti-
mated half-life being about 4 to 5 days (Barren and Walton, 1971). The pharma-
cokinetics of dieldrin elimination from the liver, brain, blood, and adipose
tissue of Carworth Farm E male rats placed on diets containing 10 ppm dieldrin
for 8 weeks was studied by Robinson et al. (1969). Dieldrin storage in the
four tissues was as follows: adipose tissue » liver > brain > blood. The
half-lives of dieldrin in the liver were estimated to be 1.3 and 10.2 days for
the initially rapid and subsequently slower phases of elimination, respectively.
4-11
-------
Similar values were estimated for blood. The estimated half-lives of dieldrin
in adipose tissue and brain were 10.3 and 3 days, respectively. Based on
similar mathematical equations, Moriarty (1975) estimated a half-life of 10.3
days for dieldrin in plasma. However, the data may be confounded by the fact
that the estimated half-life for plasma residues also included some measurement
for residues present in the plasma lipid fraction.
Sex differences were noted in steady-state tissue residues between male
and female Sprague-Dawley and Carworth Farm E rats (Walker et al., 1969;
Davison, 1973) but not in beagle dogs (Walker et al., 1969). The [14C] residue
levels in the whole carcass of female Sprague-Dawley rats were 3 to 4 times
those found in males (Davison, 1973), whereas dieldrin residues in blood, fat,
and liver of female Carworth Farm E rats were 2- to 10-fold those of males
(Walker et al., 1969).
In rhesus monkeys fed daily diets containing 2 ppm of dieldrin for 260
days, daily excretion never equaled the daily dose administered. By extra-
polation, it was estimated that a steady state would have been reached after
350 to 400 days of dosing. In one male and one female monkey sacrificed after
260 days of dosing, fat contained the highest concentrations of C^^C] residues
(54.4 and 6.1 ppm dieldrin, respectively). Lesser levels were found in other
tissues. Animals subsequently placed on untreated diets for 130 days showed
a marked decrease in tissue residue levels. However, appreciable amounts re-
mained in the fat (21.4 and 13 ppm in males and females, respectively). Fat
extracts contained only unchanged dieldrin (Muller et al., 1979).
The concentrations of dieldrin found in the liver of male rhesus monkeys
(0.22 to 23.25 ppm) fed diets containing various concentrations of dieldrin (0
to 5 ppm) for 69 to 74 months were dose dependent (Wright et al., 1978). Most
of the radioactivity was located in the liver microsomes. The authors suggested
4-12
-------
that rhesus monkeys and humans exhibited similar pharmacokinetics and had con-
siderably greater concentrations of dieldrin in the fat, liver, and blood when
compared to rodents receiving similar or higher doses of dieldrin, thus indi-
cating a slower metabolic clearance in primates than in rodents.
While there is little information on long-term retention of dieldrin, the
U.S. EPA National Human Adipose Tissue Survey (U.S. EPA, 1985) reported geo-
metric-mean levels of dieldrin in human adipose tissue from 1970 to 1983 ex-
cluding 1980 and 1982. These mean values ranged from a maximum of 0.22 ppm in
1971 to a minimum value of 0.05 ppm in 1981. There was a trend toward lower
levels over time. ,
4.1.3. Metabolism
4.1.3.1. Aldrin—Following absorption from the gastrointestinal tract, aldrin
is readily epoxidized in mammals to dieldrin (Figure 4-3). Epoxidation of
aldrin to dieldrin has also been demonstrated in plants, insects, and micro-
organisms (Soto and Deichmann, 1967; Brooks and Harrison, 1966). The in vivo
epoxidation of aldrin to dieldrin in mammals and detection of dieldrin body fat
was first reported by Winteringham and Barnes (1955) in experiments with mice.
Similar findings were reported for other species receiving unlabeled aldrin,
including beef and dairy cattle, pigs, sheep, rats, and poultry (Bann et al.,
1956), or [36Cl]-aldrin, in rats and pigs (Hunter et al., 1960).
The epoxidation of aldrin to dieldrin is catalyzed by an enzyme (or en-
zymes) found in the hepatic microsomes; the enzyme has been called aldrin
epoxidase. Further metabolism of dieldrin is relatively slow and, consequent-
ly, aldrin has been used as a substrate to measure the epoxidase activity in
many in vitro studies. Aldrin epoxidation by hepatic microsomes from pigs was
reported by Brooks and Harrison (1966). Dieldrin was readily formed when
aldrin was incubated with the microsomes in a phosphate buffer for 10 minutes
4-13
-------
CI
Cl CI
CI
Aldrin
Dieldrin
/ \
Pentachloroketone
trans-Diol
ci
H02-C-
9-Hydroxydieldrin
co2 - H
Dicarboxylic acid
Figure 4-3. Metabolic pathways of aldrin.
4-14
-------
at 30°C in the presence of a NADPH-generating system.
Wolff et al. (1980) studied the epoxidation of aldrin in a reconstituted
enzyme system using two purified isozymes of cytochrome P-450. Dieldrin forma-
tion was faster (apparent Km of 7 _+ 2 mM) when cytochrome P-450, isolated from
hepatic microsomes of phenobarbital-induced rats, was the enzyme used than when
the 3-methylcholanthrene-inducible form, P-448, was used (apparent Km of 27 +_
7 mM).
Aldrin epoxidation occurs readily with five purified cytochrome P-450
isozymes from (uninduced) male ICR mouse liver (Levi and Hodgson, 1985), The
rate of dieldrin formation ranged between 5.1 and 11 nmol dieldrin formed/nmol
cytochrome P-450/30 minutes, with cytochrome P-450A1 being the most active.
Davies and Keysell (1983) demonstrated that the rate of aldrin epoxidation by
hepatic microsomal preparations from Wistar rats was sex dependent, being fast-
er in males than in females. The apparent Km and Vmax for hepatic cytochrome
P-450 from females were 78.8 mM and 0.39 nmol/nmol cytochrome P-450/minute,
respectively. These values were 28% higher and 94% lower, respectively, than
hepatic cytochrome P-450 from males.
In humans, aldrin epoxidation occurs readily in vitro upon incubation with
liver microsomes. The rate of aldrin epoxidation by liver microsomes from 28
subjects was directly related to the cytochrome P-450 content (McMannus et al.,
1984).
4.1.3.2. Dieldrin—The metabolic pathways of dieldrin in laboratory animals are
presented in Figure 4-4. The only monohydroxy derivative detected, 9-hydroxy-
dieldrin, may be conjugated with glucuronic acid. The 6,7-transdihydroxydi-
hydroaldrin is a biotransformation product of the epoxy group and may be con-
jugated with glucuronic acid or oxidized to the tricyclic dicarboxylic acid.
Pentachloroketone (PCK) is one of the few examples known of a metabolite of a
4-15
-------
Cl Cl
Cl
Cl
Dieldrin
/ \
Pentachloroketone
trans-Diol
ci
H02.C-\H ^
CO2-H
Dicarboxylic acid
0 9-Hydroxydieldrin
Figure 4-4* Metabolic pathways of dieldrin.
4-16
-------
foreign compound that is formed by a chemical skeletal rearrangement (Figure
4-4). . ,
Dieldrin is hydroxylated to 9-hydroxydieldrin by liver microsomal mono-
oxygenase from rats, and the reaction is inhibited by the addition of the mono-
oxygenase inhibitor sesamex (Matthews and Matsumura, 1969). Its isomer anti-
9-hydroxydieldrin is neither an in vitro nor an in vivo metabolite, which may
be due to the inaccessibility of the anti C-12 hydrogen of dieldrin to oxida-
tive attack by the mono-oxygenase. Further, oxidation of 9-hydroxydieldrin to
9-ketodieldrin apparently does not occur, since oxidation would also involve
attack of the inaccessible anti C-9 hydrogen atom (Bedford and Hutson, 1975). '
The 9-hydroxydieldrin glucuronide is formed both in vitro and in vivo.
It has been identified in the bile of dosed rats (Heath and Vandekar, 1964;
Chipman and Walker, 1979), but is extensively deconjugated in the intestine
and excreted in the feces in the free form (Bedford and Hutson, 1975; Hutson,
1976). However, 9-hydroxydieldrin has also been found in the urine of monkeys
(Muller et al., 1975). The 9-hydroxydieldrin glucuronide is formed rapidly in
vitro from dieldrin (Hutson, 1976) or 9-hydroxydieldrin (Matthews et al., 1971)
upon incubation with liver microsomes from rat and uridine diphosphoglucuronic
acid. The rate of conversion from dieldrin was 0.0028 nmol/min/mg protein
(Hutson, 1976), whereas about 67% of the 9-hydroxydieldrin was converted to the
conjugate (Matthews et al., 1971). The hydroxyl conjugate has also been isola-
ted and identified in mouse urine (Hutson, 1976).
Dieldrin can also be metabolized by epoxide hydratase to form 6,7-trans-
dihydroxydihydro-aldrin, which was the first metabolite to be isolated and
identified from rabbits (Korte and Arent, 1965). However, Matthews and
McKinney (1974) reported that the cis-diol is also formed in vitro from diel-
drin by rat liver microsomes. Apparently, the cis-diol can be metabolized
4-17
-------
further to other metabolites or to the trans-diol by means of a microsomal
epimerase, whereas the trans-diol formed in vitro is not metabolized further.
It was suggested that the formation of the cis- and the trans-diol may occur
via a hydroxyketodihydro intermediate (Matthews and McKinney, 1974; Bedford
and Hutson, 1975). The compound 6,7-trans-dihydroxydihydroaldrin is further
oxidized to aldrin dicarboxylic acid (ADA) or conjugated to glucuronic acid
(Figure 4-4). Elimination of the ADA occurs mainly in the urine of mice and
rats (Baldwin et a!., 1972; Hutson, 1976) and in the feces of rats (Hutson,
1976), whereas the diol conjugate has been identified in urine from rabbits and
monkeys (Muller et al., 1975, 1979). The diol glucuronide is also formed j_n_
vitro upon incubation of 6,7-trans-dihydroxydihydroaldrin and hepatic micro-
somal preparations from rabbits or rats in the presence of uridine diphosphp-
glucuronic acid and NADPB. About 40% to 50% of the substrate is converted to
the conjugate (Matthews and Matsumura, 1969).
PCK, also known as Klein's metabolite, is a major urinary metabolite in
male rats, but it is only found in trace amounts in the urine of male mice
(Matthews et al., 1971; Baldwin et al., 1972; Hutson, 1976). PCK is also found
in trace amounts in the urine of female rats (Matthews et al., 1971). It has
also been identified in the urine of rats fed diets containing aldrin (Klein et
al., 1968). PCK is also formed in vitro by hepatic microsomes and its produc-
tion is inhibited by sesamex (Matthews and Matsumura, 1969). The mechanism of
formation of this skeletal rearrangement product has been the subject of much
speculation. Bedford and Hutson (1975) suggested that PCK is the product of
rearrangement of the same intermediate that leads to 9-hydroxy-dieldrin (Figure
4-4).
4.1.3.2.1. Pi eldrin Metabolism in Humans. Very little is known about the
metabolism of dieldrin in humans. Routes of exposure for dieldrin handlers are
4-18
-------
by inhalation, ingestion, and dermal. For the general population, the major
source of dieldrin exposure is believed to be the ingestion of contaminated
food. It has been shown that exposures due to inhalation of cigarette''smoke
(Domanski et al., 1977) and through drinking water (Banquet et al., 1981) are
relatively insignificant. The absorption of dieldrin following exposure by any
of these routes' has not been well characterized. Feldmann and Maibach (1974)
investigated the percutaneous absorption of 4 mg/cm^ [l^Cl-dieldrin applied to
the ventral forearms of male volunteers; this concentration represented the
amount that would be deposited by a thin film of 0.25% pesticide solution. It
was reported that only about 8% of the applied dose was absorbed based on
urinary [14C] excretion when compared to the excretion of a known dose adminis-
tered intravenously. This conclusion, however, is questionable because of the
large variability of the data, the small amounts of dieldrin administered, the
lower recovery of [-^C] in the urine (2% to 5% of the dose) following intrave-
nous injection, and the possible effect on absorption by cutaneous metabolism
or adipose tissue content. No human studies were found that examined the
absorption of dieldrin via inhalation and only limited information was found on
absorption via ingestion.
Most of what is known about the pharmacodynamics of dieldrin in humans
following absorption comes from a two-part study (Hunter and Robinson, 1967;
Hunter et al., 1969) in which normal male volunteers ingested either 10, 50,
or 211 mg of dieldrin per day for 18 to 24 months. The authors reported that
dieldrin concentrations in blood and fat increased with increasing dosages
of the pesticide. For higher doses of dieldrin, a significant correlation
existed between the concentration of dieldrin in the blood and in fat. The
concentration of dieldrin in adipose tissue was approximately 156-fold the
concentration of dieldrin in blood. The time-dependent increase of dieldrin
4-19
-------
concentration in blood and adipose tissue approached an asymptote in an expo-
nential manner, indicating that there is a finite upper limit to the storage
of dieldn'n by each person that is dependent on that person's daily intake of
dieldrin. This upper limit for the storage of dieldrin corresponds to a
balance between the amount of dieldrin ingested and the amount eliminated
daily, and was observed at about 15 months. The equilibrium value is charac-
teristic of a person and his particular daily intake. At about month 15,
dieldrin levels in the blood of the group receiving 211 pg HEOD/day was
0.0143 ppm and the concentration in adipose tissue was 2.12 ppm, giving a fat-
to-blood ratio of 148 (Hunter and Robinson, 1967). Measurements of blood
dieldrin concentrations for 8 months following the end of dosing on these same
volunteers indicated that dieldrin is eliminated at an exponential rate that
follows first-order kinetics. The reported half-life for dieldrin elimination
was 369 days (Hunter et al., 1969). However, this value is only an estimate
based on a limited number of samples, and may be confounded with adventitious
ingestion of dieldrin in foodstuffs (estimated at about 14 mg/day).
Moriarty (1975) estimated a half-life of about 2150 days for dieldrin in
plasma. It therefore appears that humans, like primates, exhibit a slower
metabolic clearance than rodents (Wright et al., 1978).
Additional information on body burdens of dieldrin and its mobilization
during times of stress was reported by Hunter and Robinson (1967). In measur-
ing the adipose tissue concentrations of dieldrin, it was found that the body
burden of dieldrin is related to the lipid mass of an individual and that for
a given chronic dose of dieldrin, a lean person will attain higher relative
adipose tissue concentrations of dieldrin but will retain less of the admini-
stered dose than a person having more body fat. The authors also concluded
that the catabolism of fat resulting from surgery or from sudden weight loss
4-20
-------
due to dieting did not lead to significant increases in blood dieldrin levels.
Consequently, the body burden of dieldrin in the general population does not
constitute a risk for intoxication as a result of tissue mobilization at times
of illness or weight loss.
Eckenhausen et al. (1981) measured dieldrin levels in the blood of mothers
and their babies, as well as in the mothers' breast milk and placental tissue
and the babies' meconium. Measurable levels in all samples were detected. The
reported ranges were mothers' blood, < 0.1 to 4.1 ng/mL; umbilical cord blood,
< 0.1 to 2,6 ng/mL; babies' blood, < 0.1 to 4.6 ng/mL; breast milk, 1.4 to 3.3
ng/mL; placental tissue, 1.9 ng/g (mean); and meconium, 3.1 ng/g (mean). It
was concluded that breast-fed babies did not have higher blood levels of diel-
drin than bottle-fed babies, and that the placenta offered only a partial
barrier against the transfer of dieldrin from the mother to her baby. The
levels of dieldrin present were too low to allow for metabolite analysis.
Dieldrin and/or its metabolites have been detected in human urine (Cueto
and Hayes, 1962; Cueto and Biros, 1967) but were not characterized further.
Studies on dieldrin and/or its metabolites in human feces other than in meco-
nium have not been found.
4.1.4. Excretion
4.1.4.1. Aldrin—The metabolism and excretion of [l^C]-aldrin following daily
rf'
administration by gavage to male rats at a dose of 4.3 mg/rat for 3 months was
studied by Ludwig et al. (1964). The elimination of [-^C], expressed as per-
cent of dose administered weekly per rat, increased from 30.5% on the second
day of dosing to about 103% during weeks 10 to 12. Most of the [14C] was
excreted in the feces; about 48% of the dose given in week 1 was found in the
feces and 2% in the urine. For weeks 10 to 12, fecal excretion accounted for
93% to 94% of the weekly dose, whereas the urinary excretion accounted for 8.8%
4-21
-------
to 9.1%. It was found that after about 53 days of dosing, the amount of the
radioactivity that was administered each day was equal to the amount excreted,
so that subsequent doses of aldrin did not result in further accumulation in
the body. It should be noted that Quaife et al. (1967) concluded that the
above study failed to provide unequivocal evidence for the equilibrium concept
(i.e., between intake and storage of aldrin) following chronic feeding of al-
drin because only one dose level was used and the equilibrium was established
and maintained for only a month before dosing was terminated. Nonetheless, the
results provided by Ludwig et al. (1964) indicate that such an equilibrium may
indeed be attained, especially since aldrin can induce liver enzymes leading to
a higher rate of aldrin metabolism following repeated dosing.
After termination of dosing, the excretion of C^C] decreased rapidly, and
about 98% and 99.5% of the total [l^C] administered had been eliminated by 6 and
12 weeks after the final dose, respectively. Chromatographic analysis of
diethyl ether extracts of urine and methanol extracts of feces, collected at
various intervals of the study, revealed the predominant excretion products
were hydrophilic metabolites as shown in Table 4-2. Aldrin concentration in
urine and feces decreased during the dosing period, whereas dieldrin concentra-
tions were not greatly changed. Although the decrease in aldrin concentrations
in urine and feces during the dosing period suggests increased rates of metab-
olism due to enzyme induction, some of the aldrin and dieldrin present in urine
may also have been due to contamination by feces in the metabolism cages (Ludwig
et al., 1964).
One of the polar metabolites found in the urine of rats fed diets
containing 25 ppm aldrin for an unspecified period of time was identified as
l,l,2,3a,7a-pentachloro-5,6~epoxydecahydro-2,4,7-metheno-3Hcyclopenta[a]-
pentalen-3-one (pentachloroketone) (Klein et al., 1968). Pentachloroketone was
4-22
-------
TABLE 4-2. RELATIVE CONCENTRATIONS OF EXCRETION PRODUCTS IN MALE RATS
FED ALDRIN FOR 12 WEEKS AND OBSERVED FOR AN ADDITIONAL 12 WEEKS
Week
0.29
(Day 2)
3
12
20b
Specimen
Urine
Feces
Urine
Feces
Urine
Feces
Urine
Feces
AldMn
25
45
2
15
1
10-15
5
Compound/Hetabol
Dleldrln
10
15
5-9
15
5-9
15
3-8
20
1tea
HydrophlUc
metabolites
65
40
89-93
70
90-94
70-75
92-97
75
Expressed as percent of radioactivity found In extracts.
^Eight-week recovery period after final dose.
4-23
-------
also identified in the urine of rats fed diets containing 25 ppm dieldrin
(Figure 4-4) (Klein et al., 1968). Other known metabolites of dieldrin, namely
9-hydroxydieldrin, the trans-diol, and dicarboxylic acids, have not been identi-
fied in excreta of animals dosed with aldrin. These metabolites should be
formed readily. However, neither qualitative nor quantitative data are avail-
able.
4.1.4.2. Dieldrin—The excretion of radioactivity following administration of
single oral doses of [^^C]-dieldrin to rats, mice, monkeys, and chimpanzees is
mainly in the feces, accounting for 5% to 62% of the dose (Matthews et al.,
1971; Muller et al., 1975; Hutson, 1976), whereas in rabbits, elimination is
greater in urine, accounting for 1.5% to 2.2% of the dose (Muller et al.,
1975). The ratio of radioactivity excreted in the feces and urine is about 10
to 18 in rats, 18 to 38 in mice, 0.2 in rabbits, 3.6 in monkeys, and 3.8 in
chimpanzees. Similar findings were reported following repeated dosing of
dieldrin to rabbits twice weekly for 22 weeks for a total dose of about 56 to
58 mg/kg (Korte and Arent, 1965) and to rhesus monkeys receiving 2 ppm daily in
the diet for 260 days (Muller et al., 1979). However, the ratio of radioactiv-
ity excreted in feces versus urine in monkeys decreased with time of dosing, so
that about 40% of the radioactivity was excreted in urine and 60% in the feces
after 100 days of dosing.
Although Muller et al. (1975) did not find a sex-related difference in the
amounts of radioactivity excreted in feces and urine of male and female rats
and mice, Matthews et al. (1971) reported that metabolism and excretion of
dieldrin were 3 to 4 times more rapid in male (about 31% of the dose) than in
female (about 10% of the dose) rats. The difference was attributed to the
greater ability of males to metabolize dieldrin to its more polar metabolites.
The total concentration of dieldrin excreted in the feces of male and female
4-24
-------
rats 9 days after dosing was 0.65% and 2.45% of the dose, respectively, whereas
9-hydroxydieldrin accounted for about 21.5% and 5.76& of the dose in males and
females, respectively. Most of the radioactivity extracted from urine (50% to
65%) in males was PCK and traces of 6,7-trans-dihydroxydihydroaldrin, whereas
in females the diol accounted for most of the extractable radioactivity (5% to
25%).
Species differences have been reported for the excretion of dieldrin and/
or its metabolites between male CFE rats and male CFi and LAC6 mice (Baldwin et
al., 1972; Hutson, 1976). Eight days after a single oral dose (3 mg/kg), fecal
elimination of [-^c] was slightly greater and faster in rats (62.4% of the dose)
than in mice (27.2% and 51.4% of the dose) (Hutson, 1976). Rats excreted 9-
hydroxydieldrin as the major component in the feces, together with the diol and
aldrin-derived dicarboxylic acid (ADA), whereas unchanged dieldrin was the major
component of [^C] excreted by mice with lesser amounts of 9-hydroxydieldrin,
the diol, and ADA (Baldwin et al., 1972; Hutson, 1976). Rats and mice excreted
unchanged dieldrin, ADA, and PCK in urine; however, PCK was a major urinary
metabolite in rats, whereas it was only present in minute trace amounts in
mouse urine. Another difference between rats and mice was the excretion of a
polar complex of five metabolites by mice (about 50% of the urinary [14C]).
One of these metabolites was 9-hydroxydieldrin glucuronide, present at about
0.1% of the dose in both strains of mice (Hutson, 1976).
In a comparative study conducted by Muller et al. (1975), unchanged diel-
drin was found in the feces, but not in the urine, of male and female mice
(5.5% and 3.2%, respectively), rats (0.8% and 2.8%, respectively), and rabbits
(0.3% and 0.5%, respectively), male rhesus monkeys (9.0%), and a female chim-
panzee (3.2%). The animals were dosed with [-^Cl-dieldrin at 0.5 mg/kg and
excreta was collected for 10 days. The major metabolite found in rats, monkeys,
4-25
-------
and the chimpanzee was 9-hydroxydieldrin, whereas the diol was the major
metabolite found in mice and rabbits. However, the concentrations of each
metabolite in the feces and urine were not specified.
The major metabolite excreted in the urine of rabbits (sex not specified)
receiving repeated oral doses of [l^C]-dieldrin was 6,7-trans-dihydroxydihydro-
aldrin and accounted for about 86% of the total urinary radioactivity (Korte and
Arent, 1965). In monkeys fed diets containing 2 ppm of [l^C]-dieldrin for 260
days, 9-hydroxydieldrin and unchanged dieldrin were the major compounds excreted
in the feces. As the concentration of [l^C] excreted in the urine increased
with the duration of dosing, the diol and, subsequently, 9-hydroxydieldrin and
the diol conjugate were found. At the end of the 360-day experiment, the pro-
portions of dieldrin and its metabolites, expressed as percent of total [^c]
excreted, were: dieldrin, 24%; 9-hydroxydieldrin, 52%; 6,7-dihydroxyhydro-
aldrin, 8%; and the diol conjugate, 14% (Muller et al., 1979).
Total elimination of C^C] following intraperitoneal or intravenous injec-
tion of [l^Cl-dieldrin to male rats was roughly equal to or less than that
observed following oral dosing, and occurred primarily in the feces (about 90%)
(Cole et al., 1970; Lay et al., 1982). Elimination in the feces occurs via.the
bile, since most of the [l^C]-dieldrin administered intraperitoneally or intra-
venously is excreted via the bile within 7 days (Heath and Vandekar, 1964; Cole
et al., 1970). Excretion in bile of cannulated male rats reached maximal rates
•
about 0.5 to 5 hours (Chipman and Walker, 1979; Tanaka et al., 1980). The rate
of [l^c] elimination in the bile increased threefold following pretreatment of
rats with phenobarbital (Chipman and Walker, 1979) and was three times as rapid
in perfused livers from males when compared to females (Klevay, 1970).
4-26
-------
4.2. TOXIC EFFECTS
4.2.1. Human Studies
4.2.1.1. Aldrin—The acute lethal dose of aldrin in humans by the oral route
presumably is in the range of other mammals, 20 to 95 mg/kg (Hodge et alo,
1967). This is consistent with the lethal dose of 70 mg/kg as reported by
Hayes (1971) although Jager (1970) reported the acute lethal dose to be about
5 g/kg. It has also been reported that humans recovered following exposure to
aldrin. The signs of acute poisoning are primarily related to the central ner-
vous system, with hyperexcitability, convulsions, depression, and death. Based
on data from studies oh monkeys, Hodge et al. (1967) estimated that a prolonged
oral intake of aldrin at 17.5 mg/day would endanger the life to a 70-kg human,
whereas 3.5 mg/day could be tolerated.
There have only been a few cases of severe aldrin poisoning reported. The
first was reported by Spiotti (1951), who described a case in which a 23-year-
old man intentionally ingested an amount of aldrin equivalent to 25.6 mg/kg
body weight. Convulsions began in about 20 minutes, and abnormal electroen-
cephalographic (EEG) tracings revealing generalized cerebral dysrhythmia were
seen shortly thereafter. By 5 months the rhythm was essentially normal and the
man recovered with no apparent lasting effects.
The World Health Organization (1969), as reported by Hayes (1982), pub-
lished data on 58 cases of poisoning in Kenya resulting from consumption of
seed grain treated with aldrin, sometimes in combination with other pesticides.
None of the cases involved convulsions; however, in five of the severe cases,
effects on the pupils were observed.
A few reports have documented toxicity following prolonged or chronic expo-
sure. Again the predominant effects were related to the central nervous system.
In India, 12 cases of neurotoxicity resulted from repeated consumption of wheat
4-27
-------
over a 6- to 12-month period when aldrin powder and gammexane power had been
accidentally mixed with the wheat. Convulsions were included among the repor-
ted clinical symptoms. The EEG tracings were consistent with a diagnosis of
organochlorine insecticide poisoning (Gupta, 1975).
Jager (1970) reported the effects of aldrin and dieldrin following chronic
exposure from monitoring workers at the Shell Chemical Company. He reported
that 33.2 mg/kg/day was tolerated by workers for up to 15 years. Above this
level, some individuals showed signs of intoxication including hyperexcitabil-
ity and convulsions.
Hoogendam et al. (1962) reported the results of a study designed to devel-
op a useful test for measuring possible subclinical intoxication of workers who
had been exposed to aldrin, dieldrin, and endrin at a pesticide manufacturing
and formulating plant in the Netherlands. Prior to this study, convulsions
followed by abnormal EEGs were observed at the plant following acute or pro-
longed exposures. In this study, EEGs were made of 122 exposed workers who
were compared to 122 male office workers used as controls. Twenty-five of the
exposed workers and eleven of the control group exhibited EEG abnormalities.
The incidence in the exposed workers was significantly different than controls
by chi-square test (p < 0.05).
4.2.1.2. Dieldrin—Dieldrin was also reported to be highly toxic by ingestion,
inhalation, and skin absorption, with a tolerance of 0.25 mg/m^ of air (Hawley,
1981). The majority of the data on the toxic effects of dieldrin in humans are
the result of accidental intoxication or by adults with suicidal intent. The
onset of clinical intoxication is almost always rapid. As in the case of
aldrin, the acute lethal dose is estimated to be in the range of 20 to 95 mg/kg
(Hodge et al., 1967). Princi (1954) reported survival of one man who ingested
a dose of about 44 mg/kg.
4-28
-------
More recently, however, Hayes (1982) reported that ingested dosages of
about 10 mg/kg of dieldrin resulted in fatalities. Acute intoxication was
usually followed by a brief period of excitation or drowsiness, muscle twitch-
ing, convulsions, and coma. Most subjects regained consciousness and recovered
completely. Death, when it did occur, was usually prognosticated clinically by
evidence of hypothermia.
* •
Garrettson and Curley (1968) reported an accidental poisoning in which a
2-year-old girl and her 4-year-old brother ingested a 5% solution of dieldrin.
Generalized convulsions began within 15 minutes after ingestion, and the young
girl died before medical treatment arrived. The older child sustained convul-
sive seizures for 7-1/2 hours although he eventually recovered.
Hayes (1975) previously had discussed a series of sublethal cases of occu-
pational poisoning by dieldrin that occurred in various countries. In 90 cases,
the individuals showed full epileptoid convulsions, and 11 suffered recurrence
of illness up to several months after the last exposure. The occurrence of
illness among the sprayers was proportional to the intensity and duration of
occupational exposure.
Van Raalte (1977) reported 32 clinical cases of intoxication following di-
eldrin exposure during a 13-year period (1954-1962). Convulsions were observed
in 19 of the cases.
Other than the neurological manifestations, effects on other organ systems
have rarely been reported. However, Hamilton et al. (1978) reported a case of
dieldrin-induced immunohemolytic anemia observed in Iowa. The person had
hemolytic anemia with a positive direct antiglobin (Coombs) test and a positive
Ham test in the blood serum. The serum contained antibodies selectively active
against erythrocytes coated with dieldrin.
Versteeg and Jager (1973) discussed studies at a pesticide plant in the
4-29
-------
Netherlands. Of 233 long-term workers, no permanent adverse effects were ob-
served.
4.2.2. Laboratory Animal Studies
4.2.2.1. Acute Studies—The lethal oral dose (1059) of aldrin/dieldrin for
most species ranges from 3 to 100 mg/kg. This includes mice, rats, hamsters,
guinea pigs, dogs, rabbits, monkeys, and humans (Hodge et al., 1967; RTECS,
1985). With both aldrin and dieldrin, the signs of acute toxicity are primari-
ly related to the central nervous system (CNS). The acute effects upon the CNS
include intoxication, hyperactivity, hypersensitivity to auditory and tactile
stimuli, hyperexcitability, loss of appetite (anorexia) and body weight, hyper-
nea, salivation, tremors, depression, convulsions, coma, and ultimate death
(Borgmann et al., 1952; Hodge et al., 1967).
4.2.2.2. Subchrom'c Studies—In short-term (subchronic) feeding studies with
mice and rats, the major toxic effects exerted by aldrin/dieldrin are mortal-
ity, depressed body weight, increase in liver-to-body weight ratio, and histo-
pathologic changes in the liver known as "chlorinated hydrocarbon insecticide
rodent liver (CHIRL)." CHIRL lesions consist of enlarged centrilobular hepatic
cells with increased cytoplasmic oxypillia and peripheral migration of the
basophilic granules. These lesions are only seen in the liver of rodents
exposed to chlorinated pesticides (Borgmann et al., 1952; Treon and Cleveland,
1955; Hodge et al., 1967; Fitzhugh et al., 1964; NCI, 1978b).
Treon and Cleveland (1955) reported that dogs were more susceptible than
rats to the toxic effects of aldrin and dieldrin. Diets containing 20 to 50
ppm aldrin were given to the dogs 5 or 6 days per week. The dogs died after
a period of 2 days to 3.2 months. When dieldrin was administered at the same
dose level, fatalities occurred in a period of 5 days to 1.3 months. Both
aldrin and dieldrin induced diffuse degenerative changes in the brain, liver,
4-30
-------
and kidneys of these dogs.
Keane and Zavon (1969) demonstrated a direct relationship between dieldrin
levels in the blood and the occurrence of intoxication. Repeated daily.oral
administration of 0.2, 1, or 2 mg/kg dieldrin to mongrel dogs was continued
until intoxication (between days 18 and 85) occurred. A direct relationship
between the dieldrin concentration in the blood and severity of clinical signs
of intoxication was observed; on the first day of muscle spasms, the average
concentration of dieldrin in the blood was 0.38 to 0.50 mg/mL.
Burchfield et al. (1976) examined the effects of dieldrin exposure upon
the primate brain by measuring the EEG changes in rhesus monkeys. Dieldrin was
administered to groups of three monkeys either as a single "large dose" of 4,
mg/kg intravenously or a series of 10 "small doses" of 1 mg/kg intramuscularly
at weekly intervals. The results indicated that both single exposure and a
series of subclinical exposures to dieldrin altered the frequency spectrum of
the spontaneous EEG for up to 1 year.
Singh and Jha (1982) drenched four female goats each day with aldrin at
the rate of 2.5 mg/kg body weight for 22 days. The control group consisted of
three goats that received no treatment. On study days 16, 20, and 22, an in-
crease in the serum alkaline phosphatase level of treated goats was observed;
however, the increase was not statistically significant when compared to the
controls. The raised values of inorganic phosphorus on days 20 and 22 were
statistically different from control values. No significant change in the
serum calcium level was observed during the study. Microscopic examination
of the costochondral junction revealed a drastic reduction in the width of
proliferating, maturing, and degenerating cartilage cells during the aidrin
intoxication.
4-31
-------
4.2.2.3. Chronic Studies—In long-term noncarcinogenic feeding studies, both
aldrin and dieldrin have been tested at different dose levels in various strains
of mice and rats. The overall no-effect level in long-term feeding studies in
rats is 0.5 ppm of either aldrin or dieldrin. At feeding levels of 1 ppm and
above, an increasing, dose-related hepatomegaly and histologic changes in the
liver characterized as CHIRL occurred. At levels of 10 ppm and above, typical
signs of organochlorine toxicity occurred such as irritability, tremors, and
convulsions.
The long-term feeding studies conducted with aldrin and dieldrin in mice,
rats, and dogs are summarized in Table 4-3 (aldrin) and Table 4-4 (dieldrin).
To date, there has been only one long-term study reported on either aldrin
or dieldrin in Syrian golden hamsters. Cabral et al. (1979) conducted a life-
time feeding study in hamsters with dieldrin. Male and female hamsters were
fed a diet containing 25, 60, or 180 mg/kg for up to 120 weeks. A decreased
survival for females at week 70 was observed. Both males and females at the
low and high doses exhibited a marked retardation of growth. A dose-related
increase in the incidence of hepatic cell hypertrophy in treated hamsters was
observed.
There has been one report on the effects of dieldrin on rhesus monkeys.
Groups of five male rhesus monkeys were given diets containing 0, 0.1, 0.5, 1,
1.75, or 5 ppm (0.0002 to 0.07 mg/kg body weight) dieldrin for approximately 6
years. After two monkeys in the 5 ppm group had died, the level of exposure to
the remaining three animals was reduced to 2.5 and later to 1.75 ppm. Subse-
quently, one of these animals had its dieldrin intake progressively increased
until, at the end of the second year, it was receiving dieldrin at the initial
dietary concentration of 5 ppm. Clinical and hematologic examinations, liver
and kidney function studies, urinalysis, and pathology revealed no abnormalities,
4-32
-------
TABLE 4-3. TOXIC1TY OF ALDR1N FOLLOWING LONG-TERM FEEDING TO HICE, RATS, AND DOGS
.fa
CO
Species Strain No. /Sex
House C3HeB/Fe 100/H.F
House C3HeB/Fe 100/H.F
House C3HeB/Fe 100/H.F
House B6C3F1 50/H
50/F
Rat Sprague-Dawley 10/H.F
Rat CF-1 40/H.F
Rat Osborne-Hendel 12/M.F
Rat Holtzman 15/-b
Rat Osborne-Hendel 30/H.F
Dose
10 ppm
10 ppm
10 ppm
4, 8 ppm
3, 6 ppm
5, 10, 50.
TOO, ISO ppm
2.5, 12.5,
25 ppm
0.5. 2. 10
50. 100,
150 ppm
250 ppm
5 ppm
Exposure Effects
2 yr Reduced llfespan; enlarged
mesenterlc lymph node
2 yr Reduced llfespan; Increase
In benign hyperplasla and
hepatomas
2 yr Hepatic vein thrombosis
80 wk Hyper excitability
.80 wk Reduced llfespan
2 yr Increased mortality; Increased
liver/body wt. ratio; CHIRLa
changes In the liver at the
100 and 150 ppm levels.
2 yr Increased liver/body wt. ratios
In males at 2.5 level and In
females at 12.5 level; CHIRL at
all dose levels.
2 yr A dose-related Increase,
In mortality at the 50, 100,
or 150 ppm levels; Increased
liver /body wt. ratio (not dose
related); CHIRL was observed at
all levels; hemorrhaglc and/or
distended urinary bladder associ-
ated with nephritis was observed
In male rats at the 100 and 150
ppm levels; gross enlargement
of the liver In rats 50 ppm and
above.
375 days Liver cell necrosis; nuclear
hypertrophy; abnormal mitosis.
2 yr None
Reference
Davis and
FHzhugh
(1962)
Davis (1965)
Rueber (1977)
NCI (1978a)
Borgmann (1952)
Treon and
Cleveland
(1955)
FHzhugh et al.
(1964)
Song and
Harvllle (1964)
Delchmann et al.
(1967)
(continued on the following page)
-------
TABLE 4-3. (continued)
Species Strain
No./Sex
Dose
Exposure
Effects
Reference
Rat Osborne-Hendel 50/H.F
-p.
CO
Rat Sprague-Dawley 50/F
Osborne-Hendel -50/F
Rat Osborne-Hendel 50/F
Rat Osborne-Hendel 12/H.F
Dog Beagle
2/H.F
20, 30. 50 ppra 31 no
20, 50 ppra
20. 50 ppra
30, 60 ppra
0.5. 2. 10
50. 100.
150 ppra
1, 3 ppm
2 yr
2 yr
74-80
wk
2 yr
16 mo
Dose-related tremors and clonlc
convulsions particularly In
females; reduced llfespan for
females at 50 ppm; Increased
liver/body wt. ratio In males
at 30 and 50 ppra; a moderate
Increase (not dose-related) In
the incidence of hepatic centri-
lobular cloudy swelling and
necrosis was observed in both
sexes.
Reduced survival at 50 ppm
Reduced survival at 50 ppm
Organchlorlne intoxication
(tremors, convulsions, hyper-
excitability) and reduced body
weights were observed during
the second year.
Decreased survival in rats fed
50 ppm and higher, with 33X at
150 ppm dead at the end of 1
year; female rats at 150 ppm
died from renal and hepatic
necrosis; male rats showed a
higher incidence of chronic
nephritis at 50 ppm and
above.
Hepatomegaly; local hyalin
(droplet) degeneration of
hepatocytes; vacuolization
of kidney tubules and liver
changes in the 3 ppm females;
Increase in liver/body wt.
ratios at the 3 ppm level.
Delchmann et al.
(1970)
Delchmann (1974)
NCI (1978a)
Rueber (1980)
Treon and
Cleveland
(1955)
(continued on the following page)
-------
TABLE 4-3. (continued)
Species Strain
No./Sex
Dose
Exposure
Effects
Reference
.£»
O1
Dog Mongrel 2/H.F 0.2,0.5. 25 mo Mortality occurred In dogs at Fltzhugh et al.
1.2 the 2 and 5 mg/kg level during (1964)
weeks 2-5; marked weight loss;
common lesions observed were fatty
degeneration of the liver, fat In
the renal tubules, and reduced .
erythrold cells In the bone marrow.
Dog Beagle 6/M 0.6 mg/kg 10 mo Hyperexcliability, tremors Delchmann et al.
and weight loss; cloudy (1969)
swelling and fatty degeneration
In the liver and hypertrophy of
hepatocytes.
aCHIRL = chlorinated hydrocarbon Insecticide rodent liver.
bData not given.
-------
TABLE 4-4. TOX1CITY OF DIELDR1H FOLLOWING LOHG-TERH FEEDING TO MICE, RATS, AND DOGS
I
w
CT>
Species Strain
House C3HeB/Fe
House C3HeB/Fe
House CF-1
House CF-1
House CD-I
House C3HeB/Fe
House B6C3F1
Rat CF-1
Rat Osborne-Hendel
No. /Sex
100/H.F
100/H.F
125/H.F
200/H.F
30/H.F
100/H.F
100/H.F
50/H.F
20/H.F
12/H.F
Dose
10 ppra
10 pp»
0.1, 1
10 pp»
10 ppn
0.15-15 ppm
10 ppm
2.5. 5 ppm
2.5. 12.5.
25 ppn
0.5. 2. 10.
50. 100. and
150 ppm
Exposure Effects
2 yr Reduced Itfespan
2 yr None
2 yr None
2 yr Decreased survival rate; palpable
intra-abdewiinal masses after 9
months; half the males and
females died or were sacrificed
by the 15th month.
2 yr Increased mortality and enlarged
liver In both sexes
25 mo Time- and dose-related hepatic
. nodules
2 yr Hepatic vein thrombosis
80 wk Tremors, Irritability, abdominal
distension
2 yr Increase In liver /body wt. ratio
in both sexes at 2.5 ppm; CHIRLa
at all levels
2 yr Dose-related Increase 1n mortality
at 50 ppm and higher; liver /body
wt. ratio was Increased In males
at 10 ppm and above; CHIRL was
observed in all treatment groups;
hemorrhaglc and/or distended
urinary bladder associated with
nephritis was observed at 100 and
150 ppm; gross enlargement of the
liver in all rats fed 50 ppm and
above.
Reference
Davis and
Fltzhugh (1962)
Davis (1965)
Walker et al.
(1972)
Thorpe and
Walker (1973)
Benitz et al.
(1977)
Reuber (1977)
NCI (1978b)
Treon and
Cleveland (1955)
Fltzhugh et al.
(1964)
(continued on the following page)
-------
TABLE 4-4. (continued)
Species Strain No./Sex
Dose
Exposure
Effects
Reference
Rat CF
25/H.F
Rat Osborne-Hendel 50/N.F
i
co
Rat Flscher-344 . 24/H.F
Rat Osborne-Hendel 50/H.F
Rat Osborne-Hendel 12/H.F
0.1. 1, 2 yr Irritability, tremors, and con- Walker et al.
10 ppm vulslons; increase In liver to (1969)
body wt. ratio In the 1 and
10 ppm females; CHIRL
observed at the 10 ppm; chronic
nephritis and myocardial fibrosls
in all groups.
20/30, 29 mo Dose-related tremors and clonlc Deichmann et al.
50 ppm convulsions especially In fe- (1970)
males; lower survival for females
at 30 and 50 ppm; significant
increase In liver/body wt. ratio
at 30 ppm; moderate (not dose
related) Increase in the Incidence
of hepatic centrilobular cloudy
swelling and necrosis in both
sexes at all levels.
2. 10. 2 yr Central nervous system disorders NCI (1978b)
50 ppm (convulsions, tremors, nervous
behavior) In high-dose males and
females at 80 weeks.
29 ppm 80 wk Organochlorine intoxication NCI (1978b)
65 ppm 59 wk (tumors, convulsions, hyper-
excitability) during the second
year.
0.5. 1,2, 2 yr Decreased survival In rats fed Reuber (1980)
10, 50. 100, at 50 ppm and above; 42X at
150 ppm 150 ppm dead by the end of one
year; female rats died of renal
hepatic necrosis; males had
higher incidence of chronic
nephritis at the 100 and 150 ppm
levels.
(continued on the following page)
-------
TABLE 4-4. {continued)
Species Strain
No./Sex
Dose
Exposure
Effects
Reference
Rat Albino
40/H
Dog Beagle
CO
oo
Dog
Mongrel
2/H.F
Dog
Beagle
5/N.F
2 mg/kg/day 6 mo
2/H.F 1. 3 ppm 16 mo
0.2. 0.5. 1. 25 mo
2. 5. 10 rag/
kg/day
0.005. 0.05 2 yr
rag/kg/day
The RNA content and SCOT and Shakoori et al.
alkaline phosphatase (AP) (1982)
activities increased 4.38, 2.55,
and 5.93 times, respectively, while
the bilirubin, cholesterol, and total
protein contents decreased BOX, 67X.
and SOX, respectively. Histologi-
cally, hepatic cells increased 40X
in size, while the nuclear and
nucleolar size increased 12X and
754, respectively.
1 female at the 3 ppm had vacuol- Treon and
ation of the distal renal tubules; Cleveland
no toxic signs at 1 ppm. (1955)
All dogs at the 1, 2. 5. and FHzhugh et al.
10 ing/kg levels died by the (1964)
43rd week; weight loss and
convulsions were observed at the
higher levels; common lesions
observed were fatty degeneration
of the liver; fat in the renal
tubules; reduced erythrold cells.
In females at 0.05 rag/kg, liver/ Walker et al.
body wt. ratio and in both sexes (1969)
serum alkaline phosphate activity
were increased.
aCHIRL = chlorinated hydrocarbon insecticide rodent liver.
-------
No differences were noted in liver/body-weight ratios and liver DNA and RNA of
the test animals when compared to those of control animals. No subcellular
changes were seen in the hepatocytes. Dose-related increases in microsomal
cytochrome P-450 and in the activity of the liver mono-oxygenase enzyme system
were observed at the two top dose levels. The concentrations of dieldrin in
the subcutaneous fat of the monkeys fed 0.1 ppm were very similar to those
measured in humans receiving a similar daily intake of dieldrin per kilogram of
body weight. The concentrations in the monkey livers, however, were approxi-;
mately 200 times higher than in male rats receiving a two-times greater daily
intake of dieldrin and were very similar to the concentration in the livers of
male mice daily ingesting approximately 50 times more dieldrin per kilogram of
body weight (Zavon and Stemmer, 1975; Wright et al., 1978).
4.2.2.4. Teratology and Reproduction Studies—Ball et al. (1953) observed
alterations of the estrous cycle of female rats fed a diet containing 20 ppm of
aldrin for 10 to 25 weeks.
Treon and Cleveland (1955) studied the effects of aldrin/dieldrin on rat
reproduction. Aldrin and dieldrin, at concentrations of 2.5, 12.5, and 25 ppm,
were incorporated into the diets of Carworth rats for three generations. Aldrin
at the 12.5 ppm level and dieldrin at the 2.5 ppm level reduced the number of
pregnancies. This effect was not observed at the other levels for both
compounds. Feeding of parent rats during the period of suckling had an effect
on the incidence of mortality among the pups. Pup mortality was severe at the
12.5 and 25 ppm levels, and was slight to moderate at the 2.5 ppm level for
both aldrin and dieldrin. No differences were observed in the pup weight at
weaning between the treated and untreated rats.
Good and Ware (1969) fed male and female CFW Swiss mice with 5 ppm of
dieldrin for 120 days, beginning 30 days before mating. The dieldrin-fed mice
4-39
-------
produced litters significantly smaller in weight and in number of pups than
controls.
Harr et al. (1970) investigated the effect of dieldrin on reproduction in
Wistar rats. Dieldrin was fed to male and female rats at concentrations rang-
ing from 0.08 to 40 ppm and monitored for dam survival, conception rate, pup
survival, and weaned litter size. These values were normal in rats fed at con-
centrations of 0.08 and 0.16 ppm. In rats fed 0.31 to 1.25 ppm, a reduction in
dam survival and conception rate was noted. In rats fed at concentrations
between 2.5 and 10 ppm, the pups died in convulsions (43%) or starved (57%).
Deaths of pups due to starvation occurred because both the dams and the pups
were too hyperesthetic to permit adequate nursing. Milk curd in the stomach of
pups dying when 1 to 3 days old contained dieldrin at concentrations of 2 to 20
ppm. Dam survival, conception rate, pup survival, and weaned litter size were
affected in female rats fed dieldrin at 20 to 40 ppm. Most of the pups died in
convulsions. The authors reported that the highest dietary level consistent
with normal reproduction was 0.24 ppm.
Boucard et al. (1970, as cited in Dix et al., 1977) observed teratogenic
effects of dieldrin in rats and mice given a low dose of 2.5 mg/kg and a higher
dose of 3.4 mg/kg. These doses were given as a single or repeated dose during
gestation. No dose response was demonstrated; however, no statistical analyses
were performed, and the incidences of spontaneous abnormalities in these ani-
mals were not presented.
Deichmann et al. (1971) reported subnormal reproduction in dogs exposed to
aldrin. Eleven (eight females and three males) beagle dogs were dosed with
aldrin at 0.15 mg/kg (4 females) and at 0.30 mg/kg (4 females and 3 males) by
capsule 5 days per week for a total of 14 months. Following discontinuation of
the feeding period, attempts were made to breed all experimental female dogs
4-40
-------
with experimental male dogs. The estrous cycle was delayed by 7 to 12 months
in the female dogs. Mammary development and milk production were severely
depressed. Stillbirth occurred in one female at the 0.15 mg/kg level and in
two females at the 0.30 mg/kg level. Two females fed aldrin at the 0.15 mg/kg
level did not come into heat during the 8 months following discontinuation of
the feeding. No malformations were observed.
Deichmann (1972) presented data on the effects of aldrin and dieldrin when
fed to Swiss mice for six generations. Swiss mice were exposed to aldrin at 3,
5, 10, and 25 ppm and to dieldrin at 3, 10, and 25 ppm in their diet for six
generations. The indices measured were fertility, gestation, viability, lac-
tation, and survival of the pups. Marked effects on fertility, gestation,
viability, lactation, or survival were noted in the parent (first) and second
generation, and in their offspring (first and second litters) fed 25 ppm aldrin
and dieldrin. Less marked but still significant effects were found in the
first and second generations and their offspring that were fed aldrin at 3, 5,
and 10 ppm'and dieldrin at 3 and 10 ppm.
In a study by Ozoukwa and Sleight (1972), pregnant guinea pigs were given
divided doses of dieldrin (up to 120 mg/kg) during gestation. Abortions oc-
curred in guinea pigs after administration of 100 ppm (3 mg/kg) dieldrin in
the diet, a dose that was also toxic to the mothers. Dyspnea and convulsions
were observed commonly in the dieldrin-treated guinea pigs. Several animals
became excitable 2 or more hours after administration of dieldrin, but had CNS
depression before death. Mild hepatic fatty metamorphosis and slight vacuolar
degeneration in the kidneys were seen in dieldrin-treated guinea pigs. Swell-
ing of the mitochondria occurred in cells of the cerebral cortex and cerebellum
after acute dieldrin toxicosis.
Ottolenghi et al . (1974) reported the teratogenic effects of aldrin and
. 4-41
-------
dieldrin in mice and hamsters. Pregnant Syrian golden hamsters and CD-I mice
were given single oral doses of aldrin or dieldrin at one-half the LD5Q (ham-
sters at 5, 30, and 50 mg/kg, and mice at 2.5, 15, and 25 mg/kg, respectively).
The hamsters were treated orally on days 7, 8, or 9 of gestation, and mice on
day 9 of gestation. Both aldrin and dieldrin caused a significant increase in
fetal death in hamsters treated on days 7 and 8. Hamsters treated on day 8
also had the highest number of anomalies (i.e., open eyes, webbed feet, cleft
palate, and others). A statistically significant (p < 0.01 to 0.03) reduction
in the fetal weight in hamsters treated on three different days was also
observed. In contrast to the hamsters, no significant effects were observed
in the weight or survival of fetuses of treated and control mice. However,
treated mice showed a significant (p < 0.01) increase in the total number of
anomalies. Open eye and webbed foot were more frequent after aldrin; cleft
palate and webbed foot were more frequent in the dieldrin group. The authors
reasoned that both species are susceptible to the teratogenic action of the
pesticides; however, the reduced teratogenic effect observed in mice may be
due to the lower doses used.
Chernoff et al. (1975) tested dieldrin and the photoproduct, photodiel-
drin, in CD-I mice and CD rats. Pregnant mice and rats were given daily oral
doses of 1.5, 3, and 6 mg/kg/day of dieldrin and 0.15, 0.30, and 0.60 mg/kg/day
of photodieldrin in peanut oil from days 7 to 16 of gestation. In mice at 6
mg/kg, reduced body weight gain and increases in liver-to-body weight ratio
were observed. An increase in supernumerary ribs and decreased numbers of
caudal ossification centers were noted in the fetuses of mice given 6 mg/kg
dieldrin. These findings at the highest dose are a reflection of a fetotoxic
response secondary to maternal toxicity. No evidence of teratogenicity without
concomitant maternal effects was observed at the given dose levels. The high-
4-42
-------
est dose of dieldrin (6 mg/kg) and photodieldrin (0.60 mg/kg) resulted in mor-
talities (41% and 15%, respectively). Dieldrin was not found to be teratogenic
in either the CD rat or the CD-I mouse at these dose levels. Fetal toxicity at
the highest dose was observed. Photodieldrin was not teratogenic or fetotoxic
in either the CD rat or CD-I mouse at these dose levels.
Dix et al. (1977) compared the use of two vehicles (corn oil and DMSO)
with various doses of dieldrin in CF-1 mice. The corn oil groups received 1.5
and 4.5 mg/kg/day of dieldrin orally; young (7 weeks) virgin mice were used and
the pregnancy rate was very low. With the few mice that survived to termina-
tion, the only significant effect was delayed ossification in fetuses of the
mice administered the 4 mg/kg level. The DMSO experiments were conducted with
older (10 weeks) mice of proven fertility. Fetuses of these mice exhibited a
significant (p < 0.05) increase in the incidence of delayed ossification and
extra ribs. However, the DMSO controls also had a high incidence of delayed
ossification (48) and extra ribs (71) compared to the untreated controls, where
the number of mice with delayed ossification and extra ribs were 7 and 6,
respectively. The authors attributed this to the toxic effects of DMSO. DMSO
also produced a reduction in maternal and fetal body weights, whereas the corn
oil did not. No differences were observed in the mean litter size, number of
resorptions, or fetal death with either vehicle. No dieldrin-related terato-
genic effects were observed.
Dix et al. (unpublished results) also reported that no teratogenic effects
were observed in the offspring of groups of pregnant banded Dutch rabbits dosed
with up to 6 mg/kg/day of dieldrin in carboxyl methyl cellulose, from days 6 to
18 of gestation.
4.3. MUTAGENICITY
Evaluation of the published mutagenicity studies on aldrin and dieldrin
4-43
-------
have recently (June 1986) been completed by the Office of Pesticides Program
(OPP) and presented in reports prepared by Dynamac Corporation. The detailed
descriptions and evaluations are appended to this document. Presented below
are the overall interpretations of the data taken directly from the OPP aldrin
and dieldrin reports, respectively.
4.3.1. Aldrin
The paucity of acceptable studies and the existence of major gaps in the
data (point mutation and cytogenetics) severely hinders the development of a
definitive genetic toxicology profile for aldrin. However, since aldrin is
readily converted both in vivo and in vitro to dieldrin, the genetic toxicology
assessment prepared for dieldrin can be used to infer a genetic activity pro-
file for aldrin.
Although no published bacterial gene mutation assays with sufficient
primary data were reviewed, the survey studies with ^. typhimurium and £. coli
indicate that aldrin is probably not mutagenic in procaryotes. This finding
is supported by the overwhelming conclusion of negative mutagenic effects for
dieldrin.
No mammalian gene mutation assays with aldrin were located; the single
mammalian cell point mutation assay of dieldrin showed an inconclusive, uncon-
firmed mutagenic response. A major data gap, therefore, exists for both com-
pounds.
No conclusions could be drawn from the in vitro or in vivo somatic cyto-
genetic assays with aldrin; therefore a major data gap remains. However,
aldrin was not genotoxic in bacteria, yeast, or rat hepatocytes.
Nonactivated and S9-activated aldrin induced statistically significant in-
creases in UDS in SV-40 transformed (VA-4) human fibroblasts. Dieldrin, also
investigated in this study, caused a similar response. However, the validity
4-44
-------
of this assay was seriously compromised by numerous technical deficiencies, the
observation of strong positive responses by compounds generally accepted as
weakly genotoxic or nongenotoxic substances, and the lack of positive controls.
4.3.2. Dieldrin
The published articles reviewed are not considered sufficient to establish
a complete genetic toxicology profile for dieldrin. However, certain consis-
tencies emerge which allow us to make reasonable interpretations and to recom-
mend further studies which might generate useful data.
Dieldrin was adequately tested in three acceptable Ames assays with
sufficient primary data to conclude that nonactivated and rat or hamster S9-
activated dieldrin is not mutagenic in S_. typhimurium. The results of the
seven survey Ames tests with dieldrin support this conclusion. However, based
on the inclusive results of a single Ames test, dieldrin, in the presence of
mouse S9 activation, is considered presumptively mutagenic in _$_. typhimurium.
We caution, however, that this finding has not been confirmed and technical
deficiencies preclude acceptance of the results as valid.
The three _E. coli reverse mutation survey studies with dieldrin further
support the conclusion that the chemical is not mutagenic in procaryotes.
In the single published mammalian gene mutation assay, nonactivated diel-
drin induced an inconclusive, mutagenic response in Chinese hamster V79 cells.
The validity of this assay was seriously compromised because of technical
deficiencies, reporting of "positive" responses with compounds that have been
previously reported as either weakly mutagenic or not mutagenic, and the lack
of a positive control. Although we classified dieldrin as an unconfirmed
presumptive positive mutagen for Chinese hamster V79 cells, we have serious
reservations about the validity of the study and recommend that these results
be verified in a completed assay, i.e., three or more noncytotoxic doses, with
4-45
-------
and without metabolic activation, and with known mutagens as posit.ive controls.
Although the majority of doses assayed were cytotoxic, dieldrin elicited
dose-related clastogenic responses in both cultured human lung and mouse bone
marrow cells. Dieldrin, therefore, may be considered weakly clastogenic in
mammalian somatic cells.
By contrast, dieldrin administered by gavage or by intraperitoneal injec-
tion proved to be nonclastogenic in male germinal cells; HEOD, a major compo-
nent of dieldrin, was also negative in male germinal cells.
Survey repair studies indicated that dieldrin does not induce primary DNA
damage in bacteria. Data from studies with sufficient primary data to draw
meaningful conclusions also indicated that dieldrin was not genotoxic in yeast
or primary rat and mouse hepatocytes.
However, both S9-activated and nonactivated dieldrin induced a statisti-
cally significant increase in unscheduled DNA synthesis in SV-40 transformed
human fibroblasts. The validity of this assay is questionable because of
numerous technical deficiencies, the reporting of strong "positive" responses
by weakly genotoxic or nongenotoxic substances, and the lack of acknowledged
positive controls.
It is noteworthy that all assays reporting that dieldrin adversely affects
genetic material were either flawed by inadequate study designs (Article Nos.
16, 17, and 26) or showed the greatest activity at cytotoxic doses (Article No.
18).
4.4. CARCINOGEN1CITY
This section presents the evidence relating to the carcinogenic potential
of aldrin or dieldrin based on long-term animal bioassays or results of epi-
demiologic studies.
4-46
-------
4.4.1. Animal Studies
4.4.1.1. Aldrin—
4.4.1.1.1. Studies with Mice. Table 4-5 summarizes the carcinogenicity bio-
assays of aldrin conducted with mice that are described in this section.
4.4.1.1.1.1. Food and Drug Administration Study (Davis and Fitzhugh,
1962). The authors alluded to a previous 2-year aldrin mouse feeding study
that raised the suspicion of tumorigenicity of aldrin although the results were
considered inconclusive because the majority of the animals were not available
for pathologic examination. No reference was provided and no further informa-
tion could be found on that study.
In this study, a group of 215 C3HeB/Fe mice were fed a dietary mixture
containing 10 ppm aldrin (purity not specified) for a period of up to 2 years.
The control group consisted of 217 mice. The number of mice/sex was not given,
but they were approximately equally divided by sex. Mice that died during the
experiment were necropsied, as were all 2-year survivors. The extent of path-
ology examination was not clear. Tissues from animals with gross lesions and
from some grossly normal mice were preserved. After fixation, the tissues
were reexamined and were selected for microscopic study. Slides were prepared
from all gross lesions in which neoplasia was suspected, from lungs of mice
that had hepatic masses, and also from some animals that had gross pneumonia,
intussusceptions, or certain other incidental gross abnormalities. No break-
down as to actual tissues examined was provided.
The average survival of the aldrin-treated groups was approximately 2
months less than controls. Intercurrent diseases, pneumonia, and intestinal
parasitism were present and may have influenced the long-term survival rate.
Results, reported for both sexes combined, indicated a statistically
significant (p < 0.001) increase in the incidence of hepatomas (hepatic cell
4-47
-------
adenomas) in the treated animals as compared to controls (Table 4-6). Only one
other tumor (lung), was reported in the aldrin group. The incidence of hepato-
mas, based on necropsied mice, was 23% in the aldrin group and 7% in controls.
The hepatic cell adenomas were described as expanding nodules of hepatic paren-
chyma! tissue, usually with altered lobular architecture, and morphologically
ranging from very benign lesions to borderline carcinomas. The authors con-
cluded that aldrin had significantly increased the incidence of histologically
benign liver tumors. Dr. M. Reuben conducted an independent reevaluation of
the liver lesions and considered most of the hepatomas to be liver carcinomas.
Drs. Popper, Farber, and Firminger concurred with Reuber's evaluation (Epstein,
1975). The value of this study was compromised by the poor survival rate, lack
of detailed pathology, loss of a large percentage of the animals to the study,
and failure to treat the results in males and females separately. Despite
these inadequacies, the study revealed evidence for hepatocarcinogenicity of
aldrin in CsH mice.
4.4.1.1.1.2. Food and Drug Administration (Davis, 1965). As a follow-up
to the previous FDA study, Davis (1965) administered aldrin (purity not speci-
fied) at 0 or 10 ppm in the diet, to groups of 100 male and 100 female C^ti mice
for 2 years (Table 4-7). The pathology examination was similar to that conduc-
ted in the other FDA study. Again, survival in the treated group was reduced
compared to the control group, although no breakdown of data by sex and by time
of death was given. There was no indication as to the time of tumor detection
or deaths in treated versus control groups. The incidence of hepatic hyper-
pi asi a and benign hepatomas in the aldrin group was approximately double that
of controls, whereas the number of hepatic carcinomas was about the same
(Table 4-7). Dr. Reuber also reevaluated the liver lesions from this study and
concluded that most of the hepatomas were actually carcinomas. Drs. Popper,
4-50
-------
TABLE 4-6. CARCINOGENESIS BIOASSAY OF ALDRIN AND DIELDRIN IN C3HeB/Fe
MICE EXPOSED BY DIETARY MIXTURE
Observation
Hice Initially on experiment
Average weeks on experiment
Average weeks on experiment
for mice with hepatomas
Survivors at 18 months
Survivors at 24 months
Mice discarded at autopsy
Mice potentially available
for pathology examination
Number of mice with hepatic
cell adenomas
Aldrln
(10 ppm)
215
51.8
80
32 (15%)
2 (1%)
64 (30%)
151
35 (23%)a
Dleldrln
(10 ppm)
218
51.4
77
33 (15%)
8 (4%)
70 (32%)
148
36 (24%)a
Control
217
59.8
89
47 (22%)
11 (5%)
83 (38%)
134
9 (7%)
Data not broken down by sex. •
3S1gn1f1cantly different from controls (p < 0.001),
SOURCE: Davis and FHzhugh, 1962.
4-51
-------
TABLE 4-7. SURVIVAL DATA AND LIVER LESIONS IN C3H MICE FOLLOWING
TWO-YEAR DIETARY EXPOSURE TO ALDRIN AND DIELDRIN
Observation
Mice Initially on experiment
Survivors at 18 months
Survivors at 24 months
Hepatic hyperplasla
Benign hepatoma
Hepatic carcinoma
Combined hepatoma and hepatic
hyperplasla
Aldrln
(10 ppm)
200
121 (6054)
31 (15%)
72
65 (32.5%)a
3 (1.5%)
68 (34°/.)a
Dleldrln
(10 ppm)
200
117 (58%)
39 (19%)
71
69 (34.5%)a
5 (2.5%)
74 (37%)a
Control
200
150 (75%)
62 (31%)
38
27 (13.5%)
4 (2%)
31 (15.5%)
aS1gn1f1cantly different from controls by the Fisher Exact Test (p < 0.05).
SOURCE: Davis, 1965.
4-52
-------
Farber, and Firminger concurred in Reuben's diagnosis (Epstein, 1975). A sum-
mary of Reuben's diagnosis is pnesented in Table 4-8.
Some of the same deficiencies as seen in the 1962 FDA study wene also
evident in this one; namely, a lack of detailed pathology examination and
failune to pnesent data acconding to sex. Howeven, the sunvival at 18 on 24
months was acceptable in the contnols. The evidence fon an oncogenic nesponse
(whethen benign on malignant) is substantial in male and female C3H mice.
4.4.1.1.1.3. Song and Hanville (1964). A total of 55 mice of two stnains
(C3H and CBA) wene administened aldnin in the diet (punity not specified) at 15
ppm fon an unspecified time. Ten mice senved as contnols. A companion study
with die!dm'n was also conducted. Seven aldnin on dieldnin tneated mice had
liven tuitions by 330 to 375 days. No othen details of this study wene neponted.
Little useful information can be obtained fnom this study due to the lack of
details pnovided.
4.4.1.1.1.4. National Cancen Institute (1978a). Abioassay of technical-
gnade aldnin (95% pune) was conducted using gnoups of 50 animals of each sex of
B6C3F1 mice. The aldnin was administened as a dietany mixtune fon 80 weeks,
with an additional obsenvation fon 10 to 13 weeks. Two dose levels wene used.
Time-weighted avenage (TWA) doses wene 4 on 8 ppm fon males and 3 on 6 ppm fon
females. Matched contnols consisted of untneated gnoups of 20 male mice and 10
female mice.
Pooled contnols, used fon statistical evaluation, consisted of the matched-
contnol gnoups combined with 92 untneated male and 79 untneated female mice
fnom si mi Ian bioassays of othen chemicals. All sunviving mice wene sacnificed
at 90 to 93 weeks.
Mean body weights of tneated mice wene similan to those of contnols duning
the finst yean of the study, but lowen than those of the contnols duning the
4-53
-------
TABLE 4-8. REEVALUATION OF LIVER LESIONS OF THE FDA CARCINOGENESIS
BIOASSAYS OF ALDRIN AND DIELDRIN
Observation
Hale mice
Number of mice examined
Average survival (weeks)
Hyperplasla
Hyperplastlc nodules
Hepatocellular carcinomas
Female mice
Number of mice examined .
Average survival (weeks)
Hyperplasla
Hyperplastlc nodules
Hepatocellular carcinomas
Aldrln
(10 ppm)
91
86
3%
1354
82%a
85
80
6%
8%
85%a
Dleldrln
(10 ppm)
71
91
3%
10%
87%a
71
81
4%
8%
87%a
Control
73
89
12%
18%
30%
53
93
11%
13%
4%
3S1gn1f1cantly different from controls by the Fisher Exact Test
(p < 0.05).
SOURCE: Evaluation by Reuber; as cited 1n Epstein, 1975.
4-54
-------
second year. Hyperexcitability was observed in all treated groups with in-
creasing frequency and severity during the second year. Aldrin produced a
dose-related positive trend in the mortality of female mice, but not of male
mice.
In male mice, there was a significant dose-related increase in the hepato-
cellular carcinomas when compared with either matched controls (p =0.001) or
pooled controls (p < 0.01). The incidence in the low-dose male group was
significantly increased when compared to pooled controls (p < 0.05), and the
incidence in the high-dose male group was significantly increased when compared
with matched controls (p =0.002) or pooled controls (p < 0.001) (Table 4-9).
There was no significant increase (p < 0.05) in the female test groups. The
authors concluded that aldrin was carcinogenic for the livers of male B6C3Fi
mice producing hepatocellular carcinomas.
4.4.1.1.2. Studies with Rats. Table 4-10 summarizes the carcinogenicity bio-
assays of aldrin conducted with rats that are described in this section.
4.4.1.1.2.1. Kettering Laboratory (Treon and Cleveland, 1955). Aldrin
(95% pure) was administered in the diet to groups of 40 Carworth rats of each
sex commencing at age 27 to 28 days at concentrations of 2.5, 12.5, or 25 ppm
for a period of 2 years. Groups of 40 animals of each sex, which were fed only
the powdered diet, served as controls. The mortality of the treated rats was
greater than controls with only about one-half surviving at the end of the
2-year period in the 2.5 and 12.5 ppm groups and only about 40% surviving in
the high-dose (25 ppm) groups. Weight gain was comparable for all groups.
Pathology examination revealed increases in liver weights and the induction of
nonspecific, degenerative changes in the hepatic cells typical of lesions pro-
duced by chlorinated hydrocarbon compounds. No tumor information was reported
by Treon and Cleveland (1955).
4-55
-------
TABLE 4-9. INCIDENCES OF HEPATOCELLULAR CARCINOMA IN B6C3F1 MICE
EXPOSED TO ALDRIN IN DIETARY MIXTURES
Observation Male mice Female mice
Pooled controls
Matched controls
Low dose3
High doseb
17/92 (18X)C
3/20 (15X)d
16/49 (33X)e
25/45 (56X)f
3/78 (4X)
0/10 (OX)
5/48 (10%)
2/43 (5X)
aLow Dose * 4 ppm - male mice; 3 ppm - female mice.
^Hlgh Dose * 8 ppm - male mice; 6 ppm - female mice.
Cstat1st1cally significant dose-related trend (p < 0.001) by Cochran-
Armltage test.
Statistically significant dose-related trend (p = 0.001) by Cochran-
Armltage test.
Statistically significant (p = 0.048) when compared to pooled controls by the
Fisher Exact Test.
^Statistically significant when compared to matched controls (p = 0.002)
or pooled controls (p < 0.001) by the Fisher Exact Test.
SOURCE: NCI, 1978a.
4-56
-------
TABLE 4-10. SUMMARY OF DIETARY CARCINOGENESIS BIOASSAYS OF ALDRIN IN RATS
Substance
tested Strain
Number/Sex
Dose(s)
(ppm)
Duration
Effects
Comments
Reference
>95X pure Carworth
40 H&F/dose
99X pure Osborne-Hendel 12 H&F/dose
i
01
NS
Holtzman
15 (sex not
specified)
95X pure Osborne-Hendel 30 H&F/dose
0
2.5
12.5
25
0
0.5
2
10
50
100
150
250
24 mo
24 mo
12 mo
25 mo
"Increased liver
weights
""Chlorinated hydrocarbon"
type lesions
"No Increase In tumors
related to aldrin
"Significant
Increase (p < 0.05)
liver to body weight
ratio in males
"Higher total tumors
in aldrin low-dose groups
"No increase in liver tumors
""Chlorinated Insecticide"
type lesions
"Liver cell necrosis
"No tumors reported
"High mortality Treon and
in dose groups Cleveland
"Limitations of (1955)
study:
-small number of
animals surviving
-Inadequate report-
ing of data
"Inadequate test
"High mortality at Fitzhugh et
50 ppm and above al. (1964)
"Inadequate test
"No treatment effects
on: survival, liver to
body-weight ratio, gross
pathology, tumor Inci-
dences
"Limitations of
study:
-too few animals
-too short
duration
-Inadequate report-
Ing of data
"Inadequate test
"Limitations of
study:
-dose too low
-Inadequate
reporting of
data
"Inadequate test
Song and
Harville
(1964)
Deichmann et
al. (1967)
(continued on the following page)
-------
TABLE 4-10. (continued)
Substance
tested Strain
95X pure Osborne-Hendel
Oose(s)
Hunber/Sex (ppra) Duration
50 H&F/dose 0 31 no
100 MM/controls 20
30
SO
Effects Conments Reference
•Liver pathology 'High mortality at Delchraann et
•Ho treatment-related 50 pp» al. (1970)
tunor Incidences 'Comparable
survival at other
doses
•Reevaluatlon by
Reuber on sample
concluded Delchmann
underestimated
Incidences of
malignant tuaors.
•Only 84% of the
animals were micro-
scopically examined
•Inadequate test
•£>
I
01
CO
Osborne-Hendel 50 F/dose
95X pure Osborne-Hendel
50 H&F/dose
10 H&F/matched
controls
68 H/pooled
controls
70 F/pooled
controls
0
20
50
0
30
60
24* mo
Hales: 74
vik treat-
ment plus
37-38 wk
observation;
Females: 80
wk treat-
ment plus
32-33 wk
observation;
Total = 111-
113 wk
•No increase in mammary
tumors
"No treatment-
related tumors were
observed
•Final pathology
not reported
•Inadequate test
•Body weight gain
of aldrln-treated
and rats was less
than controls
•No significant
difference in
mortality between
treated and control
groups.
•Adequate test
•Suggestive evidence
for thyroid folli-
cular cell
and adrenal cortex
tumor.
Delchmann
(1974)
NCI (1978a)
(Chu
et al.,
1981)
(Griesemer
and Cueto,
1980}
NS=Not specified.
-------
Cleveland (1966) reported the tumor incidences observed in the 1955
study. In both male and female rats, the incidences of tumors were less in
all treated groups than in the controls. A variety of tumors was seen in
several organs of both treated and control groups with no apparent treat-
ment relationship. .
This study has limited value as a carcinogenicity test because of the
following deficiencies: small number of animals at*risk at 2 years and the
inadequate reporting of specific animal data as related to survival and associ-
ated lesions. However, the dose levels were adequate, as some toxicity and
increased mortality were observed in the treated groups.
4.4.1.1.2.2. Food and Drug Administration (Fitzhugh et al., 1964). Aldrin
(> 99% purity) was administered to 12 male and 12 female Osborne-Mendel rats in
the diet at dose levels of 0, 0.5, 2, 10, 50, 100, or 150 ppm for a period of 2
years. The rats were 3 weeks of age when the aldrin diet was started.
There was no significant effect on growth rate. However, survival was
markedly decreased in aldrin groups at 50 ppm or more. Increases in liver to
body weight ratio were reported at all dose levels and were statistically
increased from controls (p < 0.05) at 10 ppm and greater in males and at all
dose levels for females. Chronic nephritis, which occurred more commonly in
males than females, was reported for high-dose levels.
Only 68% of all animals were examined histologically. Total tumor inci-
dence (all doses combined) in aldrin-treated animals was 36% compared to an in-
cidence in controls of 18%. However, the increase was particularly large among
rats at the lower dosage levels. The lower incidence at higher doses may have
been related to the decreased survival at the high doses (Table 4-11). There
was no predominant tumor type with tumors in various organs including the lungs,
breast, and lymphoreticular system. Although no hepatomas or hepatocellular
4-59
-------
TABLE 4-11. INCIDENCE OF LIVER LESIONS AND TUHORS AS RELATED TO PERCENT
SURVIVAL IN OSBORNE-HENDEL RATS ADMINISTERED ALDRIN IN THE DIET
Survival percent
Dose level
(ppm)
0
0.5
2
10
50
100
150
12 mo
92*
100
100
96
79
63
33
18 mo
75X
75
83
67
63
42
17
24 mo
50%
50
50
42
25
17
4
Liver
lesions
0/1 7a
0/19
0/19
0/22
12/18
11/11
9/9
Rats with
tumors
3/1 7b
10/19
7/19
8/22
5/18
5/11
1/9
alnc1dence of rats with liver lesions of "chlorinated Insecticide" type
per number microscopically examined.
^Incidence of rats with tumors/number microscopically examined.
SOURCE: Fltzhugh et al., 1964.
4-60
-------
carcinomas were diagnosed, a high incidence of "chlorinated insecticide" lesions
were observed at 50 ppm and above.
4.4.1.1.2.3. Song and Harvilie (1964). Groups of 15 Holtzman rats of un-
specified sex were fed a control diet supplemented with 250 ppm aldrin (purity
not specified) for approximately 1 year. Liver cell necrosis and other changes
were seen; however, no tumors were found. Two animals from each group were
sacrificed "periodically." The test is inadequate as a carcinogenicity bio-
assay because of its short duration, lack of substantive data, and inadequate
number of animals.
4.4.1.1.2.4. Deichmann et a!. (1967). Groups of 30 male and 30 female
Osborne-Mendel rats were fed a control diet or a diet supplemented with 5 ppm
aldrin (95% pure) for a period of 25 months. This level of aldrin had little
observable effect as: a) survival at 24 months was 55% in the controls and 65%
in the 5 ppm aldrin group, b) there was no increase in liver-to-body weight
ratios, c) no difference in gross pathology, and d) no difference in overall or
specific tumor incidences. The value of this study is questionable as a test
for possible carcinogenicity for the following reasons: a maximum tolerated
dose was not achieved and there were no lifetable data or data on mortality
from organ-specific tumors.
4.4.1.1.2.5. Deichmann et.al. (1970). Groups of 50 male and 50 female
Osborne-Mendel rats were administered aldrin (95% pure) in the diet at final
concentrations of 20, 30, or 50 ppm, commencing at the age of weaning; dosage
for the first 10 weeks was half the final levels. Groups of 100 rats of each
sex served as controls. Experiments were terminated at 31 months. Companion
dieldrin studies were also conducted at the same time and by the same proto-
col. There was no difference in body weight between the aldrin dose groups and
controls. Survival was comparable in all male groups and in the two low-dose
4-61
-------
groups. Mean survival was reduced in the 50 ppm group (13.0 months) as compared
to controls (19.5 months).
A moderate but not dose-related increase in the incidence of centrilobular
cloudy swelling and liver necrosis was noted in all aldrin-treated male and
female rats, but not in controls. There was no evidence for a carcinogenic
response in male or female rats fed aldrin at 20, 30, and 50 ppm. The inci-
dence of all tumors is presented in Table 4-12.
While there was no reported increase in tumor incidence, the value of the
study was compromised as the result of the lack of pathology data for many ani-
mals. The authors reported that all animals were grossly examined. However,
only 86% of the aldrin-exposed and 80% of the controls were microscopically
examined. In the high-dose female group, only 62% were actually examined
microscopically. There is also some concern as to the accuracy of the tumor
diagnoses and reporting. Reuber (as cited in Epstein, 1975) conducted an
independent reevaluation of most tissues of the 30 ppm dieldrin-exposed rats.
Based on this, Reuber claimed that the authors may have underestimated or
under-reported the incidence of malignant tumors by approximately threefold.
With the reduced survival at the high doses and the insufficient pathology
examination this study is not considered an adequate carcinogenicity test.
4.4.1.1.2.6. Deichmann (1974). In a summary that included little ex-
perimental detail, Deichmann reported the results of feeding aldrin (purity
not specified) at 0, 20, or 50 ppm to groups of 50 female Osborne-Mendel rats
and 50 female Sprague-Dawley rats, commencing at weaning. The feeding of
aldrin was associated with a dose-related reduced period of survival with the
primary cause of death listed as pneumonitis.
This feeding experiment was initiated for the purpose of studying the
effect of aldrin on mammary tumors. The authors concluded that there was no
4-62
-------
TABLE 4-12. TUMOR INCIDENCES IN OSBORNE-MENDEL RATS
FED ALDRIN IN THE DIET
Dose level
(ppm)
0
20
30
50
Hales
19/753 (2554)
5/45 (11%)
7/46 (15%)
4/45 (9%)
Females
60/80 (75%)
20/47 (43%)
24/44 (55%)
11/31 (35%)
aNumber of rats with tumors/number rats examined h1sto!og1cally.
SOURCE: Delchmann et al., 1970.
4-63
-------
indication of an increase in mammary tumor incidence related to the feeding of
20 or 50 ppm aldrin to the two strains of rats. The final pathology diagnoses
were not available at the time of the report, so that the distribution of
malignant and benign tumors could not be stated.
4.4.1.1.2.7. National Cancer Institute (1978a). Groups of 50 Osborne-
Mendel rats of each sex were administered aldrin (95% pure) in the diet at 30
or 60 ppm. The duration of the studies was 111 to 113 weeks with male rats
treated for 74 weeks, followed by 37 to 38 weeks of observation; female rats
were treated for 80 weeks, followed by 32 to 33 weeks of observation. Matched
controls consisted of groups of 10 untreated rats of both sexes. The pooled
controls, which were used for statistical evaluation, consisted of the matched-
control groups combined with 58 untreated males and 60 untreated females from
similar bioassays of other chemicals. A comprehensive pathology examination
was performed with only a minor loss of animals to the study.
Mean body weights of treated rats were lower than those of the controls
during the second year of the study, with hyperexcitability observed in all
treated groups with increasing frequency and severity during the second year.
Aldrin produced no significant effect on the mortality of rats of either sex.
There was an increased incidence of follicular-cell adenoma and carcinoma
of the thyroid gland in both male (matched controls 3/7, pooled controls 4/48,
low dose 14/38, high dose 8/38) and in female rats (matched controls 1/9, pooled
controls 3/52, low dose 10/39, high dose 7/46). These incidences were signifi-
cant in the low-dose but not in the high-dose groups of both males (p = 0.001)
and females (p = 0.009) when compared with pooled controls. However, they were
not significant with matched controls. The total number of rats in matched con-
trols was very small. Furthermore, cortical adenoma of the adrenal gland was
significant (p = 0.001) in aldrin-treated female rats in low-dose (8/45) but
4-64
-------
not in high-dose (1/48) groups when compared with pooled controls (0/55).
Because these increased incidences were not consistantly significant when com-
pared with matched controls rather than pooled control groups, NCI concluded
that the incidences of these tumors were not associated with the treatment.
But by present standards these results will be considered equivocal.
4.4.1.1.3. Studies with Other Species.
4.4.1.1.3.1. Treon and Cleveland (1955). Groups of two beagle dogs of
both sexes were administered control diets or diets supplemented with 1 or 3
ppm of aldrin (95% pure) and were sacrificed between 15 and 16 months.
In the 3 ppm aldrin group, liver weights of treated animals were increased
compared to the controls, and both of the females and one of the males were
found to have minor changes in the liver, characterized by local hyaline (drop-
let) degeneration of the hepatic cells. In the same treatment group, both of
the females had vacuolation of the epithelial cells lining the distal renal
tubules, and one of the males had slight renal tubular degeneration. In the 1
ppm dieldrin group, one female had vacuolation of the distal renal tubules.
The study is unacceptable as a test of possible carcinogenicity because of
the small number of animals tested and the short-term nature of the test rela-
tive to the normal lifespan of dogs.
4.4.1.1.3.2. Fitzhugh et al. (1964). A total of 12 mongrel dogs (5
males, 6 females, 1 of unspecified sex), ranging in age from 11 months to 6
years, were fed aldrin at dosages of 0.2 to 10 mg/kg/day, 6 days/week for
periods up to 25 months. However, survival at 1, 2, or 5 ppm was very poor
(five dogs died by 25 weeks and another at 49 weeks). Except for one dog that
died at 4 days, the dogs fed at 0.2 and 0.5 mg/kg/day survived until 24 to 25
months at which time they were sacrificed. Animals fed aldrin at 0.5 mg/kg/day
or more showed gross toxic effects, including loss of weight and convulsions,
4-65
-------
and they died progressively earlier with increasing dose level. Histologic
effects were limited to fatty degenerative changes in the liver and kidneys and
bone marrow changes in those dogs fed aldrin at 1 mg/kg/day or more. No gross
or microscopic effects were seen in dogs fed aldrin at 0.2 mg/kg/day.
The study is unacceptable as a test of possible carcinogenicity because
of the small number of animals tested and the short-term nature of the test
relative to the normal lifespan of dogs.
4.4.1.2. Dieldrin—
4.4.1.2.1. Studies with Mice. Table 4-13 summarizes the carcinogenesis bio-
assays conducted with dieldrin and discussed in this section.
4.4.1.2.1.1. Food and Drug Administration (Davis and Fitzhugh,1962).
The authors alluded to a previous 2-year dieldrin mouse feeding study that
raised the suspicion of tumorigenicity of dieldrin in which the results were
considered inconclusive because the majority of the animals were not available
for pathologic examination. However, no reference was provided and no further
information could be found on that study.
In this study, a group of 218 CsHeB/Fe mice were fed a dietary mixture
containing 10 ppm dieldrin (purity not specified) for 2 years. The control
group consisted of 217 mice. The number of mice/sex was not given, but they
were approximately equally divided by sex. Mice that died during the experi-
ment were necropsied, as were all 2-year survivors. The extent of pathology
examination was not clear. Tissues from animals with gross lesions and from
some grossly normal mice were preserved. After fixation, the tissues were
reexamined and selected for microscopic study. Slides were prepared from all
gross lesions in which neoplasia was suspected, from lungs of mice with hepatic
lumps and from some animals that had gross pneumonia, intussusceptions, or cer-
tain other incidental gross abnormalities. No breakdown as to actual tissues
examined was provided.
4-66
-------
TABLE 4-13. SUMMARY OF DIETARY CARCINOGENESIS BIOASSAYS OF DIELDRIN IN NICE
Substance
tested Strain
Dose(s)
Number/Sex (ppm) Duration Effects
Comments Reference
NS
C3HeB/Fe
215 H/F (treated)
217 H/F (controls)
0 (controls) 24 mo
10
Hepatoroas:
controls- 9/134 (7X)
10 ppm- 36/148 (24X)a
NS
C3H
100 N&F/dose
0
10
24 mo
Hepatoroas:
controls-27/200 (13.5X)
10 ppm- 69/200 (34.5X)6
NS
C3HeB/FeJ
CBA/J
55 treated (sex
not specified);
10 controls (sex
not specified)
375 days
15
Liver tumors:
7 treated
mice had tumors
"Reevaluatlon by Davis and
Reuber and other Fitzhugh
pathologlsts con- (1962)
eluded that most Reuber (as
liver tumors were cited in
hepatocellular Epstein.
carcinomas. 1975)
"Limitations of
study:
-poor survival rate
-many lost animals
-data not tabulated
by sex, time of
death/tumors
"Adequate test
°Reevaluat1on by Davis
Reuber and other (1965)
pathologlsts con- Reuber (as
eluded that most cited In
liver tumors were Epstein,
hepatocellular 1975)
carcinomas.
"Limitations of
study:
-lack of detailed
pathology
-data not tabulated
by sex, time of
death/tumors
"Adequate test
"Lack of detail as Song and
to sex, tine of Harvllle
death, tumors, etc. (1964)
"Limited value
"Inadequate test
(continued on the following page)
-------
TABLE 4-13. (continued)
01
oo
Substance
tested Strain
Technical Swiss-Webster
Grade
>99X CF-1
pure
>99* CF-1
pure
Oose(s)
Number/Sex (ppn) Duration
100 H/F/dose 0 Not clear
3
10
300 H/F (ctr) 0 132 wk
125 H/F (0.1 ppn) 0.1
125 H/F (1.0 ppra) 1
200 H/F (10 ppm) 10
30 H/F/dose 0 128 wk
78 M/F /controls 1.25
2.5
5
10
20
Effects
Hepatic Nodules;
Dose-related
0 ppm-OX
3 ppm-2.5X
10 pprn-48%
Liver Tumors:
Dose-related
Increased In type A,
type B. and total
liver tumors.
Statistically
significant (p < 0.05)
at 10 ppm for both
total liver tumors
and type B tumors
(carcinomas); lung.
lymphold and other
tumors In female mice
Liver Tumors:
Dose-related, except
at 10-20 ppm where
early mortality
reduced number at
risk
Comments Reference
"Reevaluatlon by HacDonald
Reuber and original et al.
pathologist. Many (1972)
nodules reclassl- Reuber (as
fled as hepato- cited In
cellular carclno- Epstein,
mas. (1975)
•Limitations of
study:
-only 71X histology
examination
-Inadequate report-
Ing of data
"Adequate study
"Carcinomas mainly Walker et
In 10 ppm groups al. (1972)
"Limitations of
study
-Inadequate report -
ing of individual
deaths and asso-
ciated tumors
"Adequate study
"Liver enlargements Walker et
detected by 36 al. (1972)
weeks .
"20 ppm-high
toxlcity early
deaths
(continued on the following page)
-------
TABLE 4-13. (continued)
^
<£>
Substance
tested Strain
>99X CF-1
pure
>99X CF-1
pure
>99X CF-1
pure
Dose(s)
Number/Sex (ppra) Duration Effects
(continued)
5 ppm
Hales-
87X (controls 12%)
Feroales-
60% (controls 10X)
Statistically sig-
nificant (p < 0.05) at
2.5 ppm and above
60 H/F/dose 0 128 wk Liver Tumors:
78 M/F/controls 10 Dose-related.
Increases In total
tumors, type A and
type B tumors
29 N/F/dose 0 Varied from Liver Tumors:
10 2-64 wk Significant
Increase (p < 0.05)
with 64 wk of
exposure; not sig-
nificant for 32 or
less wk of exposure
30/H/F/dose 0 110 wk Liver Tumors:
45 H/F /controls 10 Significant increase
(p < 0.01) In type B
tumors of which many
metastaslzed to the
lungs
Shortened latency period
Comments
(continued)
°<10 ppm- minimal
toxldty
"No individual data
"Adequate study
"Liver enlargements
by 40-50 wk In
all dose groups
"Adequate study
"Demonstrates need
for long-term
exposure to Induce
effect
"Adequate study
"Liver enlargements
by 50 wk;
"Adequate study
Reference
Walker et
al. (1972)
Walker et
al. (1972)
Thorpe and
Walker
(1973)
(continued on the following page)
-------
TABLE 4-13. (continued)
Substance
tested Strain
95% pure B6C3F1
>99X pure CF-1
Huraber/Sex
SO M&F/dose
20 H (controls)
10 F (controls)
19-82 H/group
Dose(s)
(ppm)
0
2.5
5
0
Duration
80 wk
treatment
plus 10-13
wk obser-
vation
110 wk
Effects Comments
Hepatocellular 'Dose-related
Carcinomas: mortality (n
Males: HD-16/45 (36X)a females.
LO-12/49 (24X)a "Adequate test
Ctrl (H)-3/18(17X)
Ctrl (P)-17/92(18X)
Females: HD-3/46 (7X)
LD-3/50 (6X)
Ctrl (H)-1/18(6X)
Ctrl (P)-7/91(8X)
Liver Tumors; "Limitations of
Reference
NCI (1978a)
Tennekes
•p»
o
10
10 ppra-48/59 (SIX) study et al.
0 ppm-6/60 (10X); -Only males tested (1981)
39X type B (carcinoma) "Adequate study
at 10 ppn and only IX
In controls. Lung
raetastasta in 14X of
10 ppm animals; OX for
controls. Significant
(p < 0.01)
>99.9X CF-1
pure
1800 mice used
1n study. Number
varied from 17
to 297 per groups
equally divided
by sex
0
1
2
5
10
20
0 Lifetime
.1
.5
Liver Tumors:
Dose-related increase
In Incidences at 10 ppm
and greater. Also
dose-related reduction
in induction time
at 0.1 ppm and greater
"No linear relation-
ship between
median total dose
or median time-to-
tumor formation
and daily exposure
levels.
Tennekes
et al.
(1982)
(continued on the following page)
-------
TABLE 4-13. (continued)
Substance
tested Strain
Number/Sex
Dose(s)
(ppra)
Duration
Effects
Comments
Reference
>99% pure
CeH/He
B6C3F1
C57/BL/6J
SO/50 M
62/76 H
71/69 H
(Dose/control
groups)
0
10
85 wk exposure
plus 47 wk
observation
(total=132 wk)
(continued)
In females and 1 In
males.
Hepatocellular
Carcinomas:
Significant Increase
(p < 0.05) In all 3
"Also benign liver
tumors In B6C3F1
and C57BL/6J mice
(continued)
"Authors suggest
this Indicates en-
hancement or promot-
ing effect rather
than Initiation.
"Adequate study
"Limitations of test Helerhenry
-Only males tested et al.
"Adequate study (1983)
strains.
^significantly different from controls (p < 0.001).
Significantly different from controls (p < 0.05).
NS = Not specified; Ctrl (N) = matched controls; Ctrl (P) = pooled controls.
-------
Average survival in the dieldrin-treated group was approximately 2 months
less than in controls. Intercurrent diseases, pneumonia, and intestinal para-
sitism were present and may have influenced the long-term survival rate.
Results, reported for both sexes combined, indicated a statistically sig-
nificant (p < 0.001) increase in the incidence of hepatomas (hepatic cell
adenomas) in the treated animals as compared to controls (Table 4-6). The
incidence of hepatomas, based on necropsied mice, was 24% in the dieldrin group
and 7% in the controls. The hepatic cell adenomas were described as expanding
nodules of hepatic parenchymal tissue, usually with altered lobular architec-
ture, and morphologically ranging from very benign appearance to borderline
carcinomas. The authors concluded that dieldrin had significantly increased
the incidence of histologically benign liver tumors. Dr. M. Reuber conducted
an independent reevaluation of the liver lesions, and considered most of the
hepatomas to be liver carcinomas. Drs. Popper, Farber, and Firminger concurred
with Dr. Reuber1s evaluation (Epstein, 1975).
The value of this study was compromised by the poor survival rate, lack
of detailed pathology, loss of a large percentage of the animals to the study,
and failure to treat the data for males and females separately. Despite these
inadequacies, the study revealed evidence for the hepatocarcinogenicity of
dieldrin in C$\ mice.
4.4.1.2.1.2. Food and Drug Administration (Davis, 1965). As a follow-up
to the previous FDA study, Davis (1965) administered dieldrin (purity not
specified) at 0 or 10 ppm in the diet to groups of 100 male and 100 female C^ti
mice for 2 years. The pathology examination was similar to that conducted in
the other FDA study. Again, survival in the treated group was reduced compared
to the control group, although no breakdown of data by sex and by time of death
was given. There was no indication as to the time of tumor detection or death
4-72
-------
in treated versus control groups. The incidence of benign hepatomas in the
dieldrin group was significantly increased (p < 0.05) with incidence of 32.5%
versus 13.5% in controls. The number of hepatic carcinomas was about the same
(Table 4-7).
Dr. Reuber also reevaluated the liver lesions from this study and con-
cluded that most of the hepatomas were actually carcinomas and that the in-
creased incidences were statistically significant (p < 0.05) for both males and
females. Drs. Popper, Farber, and Firminger checked representative histology
and concurred with Reuben's diagnosis (Epstein, 1975). A summary of Reuber's
diagnosis is presented in Table 4-8.
Some of the same deficiencies as seen in the 1962 FDA study were also
evident in this one; namely, a lack of detailed pathology examination and
failure to present data according to sex. However, the survival at 18 or 24
months was acceptable in the controls. The evidence for an oncogenic response
(whether benign or malignant) is substantial in male and female C3H mice.
4.4.1.2.1.3. Song and Harville (1964). A total of 55 mice of two strains
(C3H and CBA) were administered dieldrin (purity not specified) in the diet at
15 ppm for an unspecified time. Ten mice served as controls. A companion
study with aldrin was also conducted. Seven dieldrin- or aldrin-treated mice
had liver tumors by 330 to 375 days. No other details of this study were re-
ported. Little useful information can be obtained from this study due to the
lack of details provided.
4.4.1.2.1.4. MacDonald et al. (1972). This unpublished report was re-
viewed and evaluated by Epstein (1975). Technical grade dieldrin, dissolved
in corn oil, was administered in the diet at 0, 3, or 10 ppm to groups of 100
weanling Swiss-Webster mice of both sexes. The duration of the experiment was
unclear. No data were given on the time of death of any animal. Only 71% of
4-73
-------
the mice were examined histologically. Decreases in the incidence of spontane-
ous mammary carcinomas were noted in treated groups. No significant weight
loss or changes in liver-to-body weight ratios occurred in mice in the test
groups.
The authors claimed that dieldrin is not carcinogenic, but it induced a
dose-related increase in nonneoplastic lesions of the liver, such as regenera-
tive hypertrophy and nodulation or regenerative hyperplasia. The incidence of
nodulation, for both sexes combined, was 0%, 2.5%, and 48% in the control, 3
ppm, and 10 ppm groups, respectively.
The histopathology was reevaluated by Reuber (cited in Epstein, 1975) and
others Deichmann (1973, cited in Epstein, 1975). Anderson, one of the original
authors of the study, reversed his original diagnosis of liver lesions from
nonneoplastic to neoplastic in approximately half of the only 14 sections
originally examined. Of the livers from 10 mice fed 10 ppm dieldrin in which
nodulation and nonneoplastic lesions were originally reported, Reuber diagnosed
hepatocellular carcinoma in seven instances; these diagnoses were confirmed
by Firminger. The results of the histopathologic reevaluation confirm the
hepatocarcinogenicity of dieldrin in mice, although revised incidence figures
cannot be given, since the reevaluation was based on a relatively small number
of sections.
4.4.1.2.1.5. Walker et al. (1972). Walker et al. (1972) conducted two
studies on dieldrin carcinogenicity for Shell Research Limited (Tunstall
Laboratory). The second study comprised four experiments. Specific-pathogen-
free CFi mice of both sexes were fed dieldrin (< 99% HEOD) in the diet com-
mencing at 4 weeks of age. The duration varied somewhat with the specific
experiment, but all were for at least 2 years. Other mice were fed 4-amino-
2,3-dimethylazobenzene and served as positive controls. In all studies, except
4-74
-------
one, study No. 2 (Experiment 2.2), mice were fed diets sterilized by ethylene
oxide and were maintained under specifiopathogen-free conditions. Initially,
the mice were housed five per cage, but after the 6th week, they were placed in
individual cages.
Several factors complicate the interpretation of the results of these
studies. Statistical analyses of tumor incidence data were not presented. In
addition, individual animal data were not reported, thereby restricting the
possibility of independent analysis. Some animals were sacrificed when palpable
masses became large, even if they were otherwise healthy. This lowered the
incidence of tumors, tumor size, and probability of metastases. It is not
clear how the incidences of tumors were derived, as these were based on an
undefined number of animals.
Liver tumors were classified as Type A or Type B. The Type A tumor
appeared as solid cords of closely packed parenchymal cells with a morphology
and staining affinity little different from the rest of the normal parenchyma
with no fibrous capsule. They were classified as benign tumors. The Type B
tumor appeared as areas of papilliform and adenoid growth with some areas of
cells proliferating in confluent sheets and often with foci of necrosis.
Nuclear abnormalities were often present. The vast majority of mice with Type
B tumors also had Type A tumors. In many cases, the two types of lesions were
contiguous. Type B tumors were rare in controls.
Study No. 1
In this study, diets containing nominal concentrations of 0, 0.1, 1, and
10 ppm dieldrin (> 99% pure) were fed to groups of 600, 250, 250, and 400 CF-1
mice, respectively, divided equally by sex. The animals were 4 weeks of age,
at initiation of treatment. The duration of treatment was 132 weeks. The con-
trols and 10 ppm groups were abdominally palpated weekly, commencing at 16
4-75
-------
weeks. Once masses were detected, animals were palpated twice weekly and those
animals with very big masses were sacrificed. Sections of the brain, heart,
lungs, liver, kidneys and testes or ovaries, and of any macroscopic lesions,
were prepared for microscopic examination. Palpable liver masses were iden-
tified in the 10 ppm groups by 9 months, with one-half having died or been
sacrified by 15 months, as compared with 20 to 24 months for the controls.
Palpable masses were never detected in the 0.1 or 1 ppm groups, whose life-
spans were similar to those of untreated controls. Surviving animals were
sacrified from 132 to 140 weeks. The absence of data on the time of death or
sacrifice of animals with tumors made accurate determination of survival rates
impossible.
An increased incidence of liver tumors was present at all dietary concen-
trations of dieldrin and this increase was dose related (Table 4-14). Type B
tumors were rare in controls, with none seen in females and only 4% in males.
While there was an increase in type A tumors in all treated groups, the in-
crease in type B tumors was mainly in the 10 ppm groups.
Dieldrin also induced a significant increase (p < 0.05) in the incidence
of pulmonary adenomas and carinomas in female mice, and also lymphoid and
"other" tumors in female mice. The study was an acceptable test for carcino-
genicity with highly positive results in the induction of hepatocellular
carcinomas and a slight increase in other tumors, especially in the treated
female groups.
Study No. 2
Experiment 2.1—In this experiment, designed to examine the existence of a dose-
response relationship between intake of dieldrin and formation of liver tumors,
groups of 30 CF-1 mice of each sex received ethylene oxide-sterilized diets con-
taining 1.25, 2.5, 5, 10, or 20 ppm dieldrin (> 99% pure) for 128 weeks. The
4-76
-------
TABLE 4-14. INCIDENCES OF TUMORS IN CF-1 NICE ADMINISTERED DIELDRIN IN THE DIET FOR 132 WEEKS
Dose
(Ppm)
0
0.1
1
10
0
0.1
1
10
No. of
mice
288
124
111
176
297
90
87
148
Percent
Total
20
26
31
94a
13
27
37
92a
with ^^ver
Type A
16
22
23
37a
13
23
31a
37a
tumors
Type B
4
4
8
57a
0
4
6
553
Percent with
Adenoma
HALES
33
38
38
18
FEMALES
16
26
34a
10
lung tumors
Carcinoma
8
11
12
1
6
13b
14b
0
Percent with
lymphold tumors
35
21
20
24
40
50b
54b
5
Percent with
other tumors
6
3
5
2
7
9
17b
1
Statistically significant (p > 0.01).
Statistically significant (p < O.OS).
SOURCE: Walker et al.. 1972.
-------
control group consisted of 78 animals of both sexes. The number of animals
available for examination in the highest dose groups (20 ppm) was much reduced
because of early losses from acute toxicity. In these 20 ppm groups, about
25% of the males and nearly 50% of the females died during the first 3 months
of the experiment, with the remainder sacrificed by 12 months due to liver
enlargements. The liver enlargements were detected after 36 weeks. No signs
of dieldrin toxicity were seen in the 10 ppm or lower groups. However, intra-
abdominal masses were detected after week 40 in the 10 ppm groups, and most of
the mice were sent to necropsy by 16 to 17 months. This was similar to the
earlier results at 10 ppm. Palpable masses were detected in the 5 ppm group
after week 75 and in the 2.5 ppm animals after week 100, with morbidity of
these groups only slightly increased. No liver enlargements could be palpated
at 1.25 ppm.
The incidences of liver tumors are presented in Table 4-15. It is note-
worthy that a clear, statistically significant (p < 0.05) increase in liver
tumors of both type A and type B and in total liver tumors occurs from 1.25
to 5 ppm, reaching 87% in males and 60% in females. The decrease at 10 and
20 ppm may be attributable to losses due to toxicity so that fewer mice were
actually at risk after 1 year. As the individual data were not presented, a
life-table analysis could not be performed. At dose levels of 2.5 ppm and
above, a statistically significant (p < 0.05) increase in liver tumors was
observed. While increased at 1.25 ppm, the increases were not statistically
significant (p < 0.05).
Experiment 2.2—This study was designed to determine the effect of the method
of diet sterilization and dieldrin-induced liver changes. Diets were steril-
ized with gamma-irradiation or were not sterilized. Groups of 30 male and 30
female CF-1 mice were fed diets of 0 or 10 ppm dieldrin sterilized by gamma-
4-78
-------
TABLE 4-15. INCIDENCES OF LIVER TUMORS IN CF-1 MICE ADMINISTERED
DIELDRIN IN THE DIET FOR 128 WEEKS
Dose
(ppm)
0
1.25
2.5
5
10
20
0
1.25
2.5
5
10
20
No. of
mice
78
30
30
30
11
17
78
30
28
30
17
21
Percentage
Total
HALES
123
20
433
87a
453
713
FEMALES
10b
17
433
60a
53«
38a
of mice with
Type A
12b
13
40a
77a
36a
18a
10b
17
39a
433
413
243
liver tumors
Type B
Ob
7
3
10a
9
53a
Ob
0
4
173
123
143
Statistically different from controls (p < 0.05).
bstat1st1cally significant dose-related trend by Cochran-Armltage trend
test (p < 0.05).
SOURCE: Walker et al., 1972.
4-79
-------
Company, the mice were not sacrificed when abdominal masses became large, but
only when the animals were moribund (anorexic and clinically affected).
Diets containing 10 ppm dieldrin (> 99% pure) were fed to groups of 30
CF-1 mice of both sexes for 110 weeks. The control group consisted of 45 mice
of both sexes. The mice were bred and maintained in individual cages under
specific-pathogen-free conditions. Treatment commenced at 4 weeks of age; all
survivors were sacrificed after the 110-week exposure. Moribund animals were
sacrificed during the study; all animals were grossly examined. The sections
of brain, heart, lungs, liver, kidneys, and testes or ovaries, and all macro-
scopic lesions were processed-anetmicroscopically examined. In the mice fed
dieldrin, liver enlargement was detected after 50 weeks in both sexes with
mortality increased after 22 months. Liver lesions were arbitrarily classified
as Type A or Type B as previously described by.Walker et al. (1972).
Dieldrin induced a statistically significant\increase (p < 0.01) in the
incidence of liver tumors in both sexes as compared with controls (Table 4-17).
Metastases were found only in the lungs of animals with Type B tumors. As in
the previous study by Walker et al. (1972), Type B tumors were not detected in
female controls and were present in only a small proportion (4%) of male con-
trols.
In contrast to the earlier study of Walker et al. (1972), tumor incidences
versus time were presented. Liver tumors appeared much earlier in the treated
animals than in controls in that 100% and 50% of all dieldrin-treated males and
females, respectively, that died by 17 months had liver tumors; no liver tumors,
however, were found in controls at that time.
4.4.1.2.1.7. National Cancer Institute (1978a). A bioassay of technical-
grade 'dieldrin (> 96% purity) for possible carcinogenicity was conducted by
-------
TABLE 4-17. INCIDENCES OF LIVER TUHORS IN CF-1 HICE ADMINISTERED
10 PPM DIELDRIN IN THE DIET FOR 110 WEEKS
Dose
(ppm)
0
10
0
10
No. of
mice
45
30
44
30
Percentaqe of mice
Total
24
100a
23
87a
Type A
HALES
20
47a
FEMALES
23
40b
with liver tumors
Type B
4
53a
0
470
Lung metastases
0
3
0
17a
aS1gn1f1cantly different from controls (p < 0.01).
bS1gn1f1cantly different from controls (p < 0.05).
SOURCE: Thorpe and Halker, 1973.
4-83
-------
administered dieldrin at one of two doses for 80 weeks followed by observation
for 10 to 13 weeks. Time-weighted average doses were 2.5 or 5 ppm. Matched
control groups consisted of 20 untreated male mice and 10 untreated female
mice. Pooled controls, used for statistical evaluations, consisted of the
matched-control groups and untreated animals from similar bioassays for other
chemicals (92 male and 79 female mice). All surviving mice were killed at
90 to 93 weeks. The protocol called fo.r a comprehensive gross and microscopic
examination of all animals.
During the second 6-month period, the treated animals showed treatment-
related toxicity, mainly neurologic in origin, e.g., hyperexcitability, hyper-
activity, and tremors. The toxic effects appeared first in the males and later
on in the females. Mean body weights were similar in the dieldrin-exposed and
control groups during the first year, but were slightly lower during the second
year in the treated groups although not statistically significant. Survival in
all groups was excellent with at least 80% of all groups alive at 90 weeks. No
significant difference in survival was noted. Based upon achieving a maximum
tolerated dose without excess toxicity and increased mortality, the test was
considered an acceptable assay for carcinogenicity.
In male mice, there was a significant dose-related increase in hepato-
cellular carcinomas when compared to pooled controls (p = 0.02). The incidence
in high-dose males was also significantly increased (p = 0.025) when compared
with the pooled controls (Table 4-18). No other tumor type was increased in
the treated groups. The authors concluded that dieldrin induced a significant
increase in the incidence of hepatocellular carcinomas in high-dose male B6C3F1
mice.
4.4.1.2.1.8. Tennekes et a!. (1979, 1981). In these studies the incidence
of liver tumors was determined in groups of 19 to 82 male CF-1 mice fed control
4-84
-------
TABLE 4-18. INCIDENCES OF HEPATOCELLULAR CARCINOMAS IN B6C3F1 MICE
EXPOSED TO DIELDRIN IN DIETARY MIXTURE
Study group
Pooled Controls
Matched Controls
Low Dose (2.5 ppm)
High Dose (5 ppm)
Male mice
17/92 (18X)a
3/18 (1%)
12/50 (24%)
16/45 (36%)b
Female mice
7/91 (8%)
1/18 (6%)
3/50 (6%)
3/46 (2%)
Statistically significant dose-related trend (p < 0.02) by Cochran-
Armltage test.
Statistically significant when compared to pooled controls (p = 0.025)
by the Fisher Exact Test.
SOURCE: NCI, 1978a.
4-85
-------
or die!drip-supplemented (10 ppm) diets for 110 weeks and maintained on either
of two types of bedding (shredded filter paper or softwood sawdust) and two
types of diet (commercial or purified semi synthetic). Animals were bred and
reared on the specified diet or bedding. The males, upon weaning, were then
housed together at a maximum of five/cage. All animals, during the study,
found dead or moribund were necropsied. While all organs were grossly inspec-
ted, only the liver and lungs were examined histologically. The mice were
weanlings at time of initiation of dieldrin administration. A total of 139
mice were treated with dieldrin with 252 mice fed the control diet. Liver
lesions were classified according to the arbitrary method of Walker et al.
(1972) as Type A or B. Mice were sacrificed at 15, 52, 65, 92, and 110 weeks.
Dieldrin administration resulted in the relatively early appearance of
nodular hepatic lesions, the first being observed in a mouse approximately
43 weeks of age.
Of 59 dieldrin-treated mice killed at 65 weeks of age, 48 (81%) had liver
tumors as compared to only 3/60 (5%) in the control groups. Type B tumors were
not present in control animals, although they were found in 25% of the animals
in the dieldrin groups. Only one animal was found with a hepatocellular car-
cinoma metastasis to the lung.
The incidence of liver tumors for the total 110 weeks was similar to that
seen at 65 weeks in that 81% of treated mice had liver tumors compared to only
10% of the controls. The major difference, however, was in the number of
hepatocellular carcinomas (39% vs. 1%) and the number of lung metastases (14%
vs. 0%) (Table 4-19). These results were highly significant (p < 0.01) and
represented a substantial carcinogenic response.
4.4.1.2.1.9. Tennekes et al. (1982). This study was undertaken to inves-
tigate the dose-response characteristics of dieldrin-mediated enhancement of
4-86
-------
TABLE 4-19. INCIDENCES OF LIVER TUMORS IN CF-1 MICE ADMINISTERED
10 PPM DIELDRIN IN THE DIET FOR 110 WEEKS
Dose
(ppm)
0
10
No. of
mice
252
139
Percentaqe of mice
Total
10
81a
Type A
9
42a
with liver
Type B
1
393
tumors
Lung metatases
0
14
aStat1st1cally significant (p < 0.01).
SOURCE: Tennekes et al., 1981.
4-87
-------
liver tumor formation in CF^ mice. Mice were bred and maintained under spe-
cific pathogen-free conditions. A total of 1800 CF^ mice were administered
dieldrin (> 99.9% purity) in the diet beginning at 3 weeks of age and continued
for lifetime. Treatment groups are indicated in Table 4-20. The animals were
multiply housed, but separated into individual cages after 6 weeks of treat-
ment. Health and behavior were observed daily. Weekly abdominal palpation
was started after 16 weeks of treatment; when intra-abdominal masses were
detected, the animal was palpated twice weekly and it was sacrificed when the
enlargement was considered to be detrimental to health. Liver tumors were
classified arbitrarily as Type A or B by the method of Walker et al. (1972).
Liver tumor incidence was calculated on the basis of censored survival
data, censorship being imposed on deaths due to other causes, natural or arti-
ficial.
Tumor incidence was higher in treated than in control animals and appears
to be dose-related through the 10 ppm groups for both sexes (Table 4-20). The
reduced response in the 20 ppm groups may be due to compound toxicity, since
the authors stress that a significant proportion of both male and female mice
exposed to 20 ppm dieldrin died from acute intoxication within the first 13
weeks of treatment.
Dieldrin exposure levels were also associated with a significant reduction
of the time-to-tumor development, as compared to controls. The lowest dieldrin
exposure level associated with a significant reduction (p < 0.05) of median
liver tumor induction period was 0.1 ppm in females and 1 ppm in males.
While there was no linear relationship between the median total dose or
the median time-to-tumor formation and the daily dieldrin exposure level, the
authors suggested that this was not inconsistent with the concept that this com-
pound is devoid of initiating potential, and operates by enhancing the effect
4-88
-------
TABLE 4-20. INDUCTION OF LIVER TUMORS IN CF-1 MICE ADMINISTERED
DIELDRIN IN THE DIET FOR LIFETIME EXPOSURE
Sex
Female
Male
Expt.
no.
1
2
1
1
2
2
1
2
2
1
2
1
1
2
2
1
2
Dleldrln
(ppm)
Q
0
0.1
1
2.5
5
10
10
20C
0
0
0.1
1
2.5
5
10
20C
NO.
of mice
297
78
120
117
28
30
148
17
21
289
78
124
m
30
30
176
17
Percent animals
with liver tumors
Total
12
10
21
28
46
60
96
53
38
20
12
27
32
47
87
95
71
Type B
0
0
3
5
7
17
60
12
14
4
0
5
9
7
10
60
53
Median liver
tumor Induction
period (wk)
129 (>123)a
127 (>123)
119 (>117)b
116 (113-120)
106 (89-108)
92 (85-99)
64 (64-66)
50 (>47)
48 (>41)
133 (>130)
>128
130 (121-134)
118 (116-122)b
111 (>109)
99 (91-101)
67 (65-68)
41 (39-46)
aNumbers 1n parenthesis Indicate 95% confidence Intervals.
^Lowest dleldrln exposure level associated with a significant reduction of
median liver tumor Induction period, p < 0.05.
cln the 20 ppm dleldrln treatment group, 11/21 females and 5/17 males died
from acute Intoxication within the first 13 weeks of treatment.
SOURCE: Tennekes et.al,, 1982.
4-89
-------
of genetically linked oncogenic factor in CF^ mouse liver.
4.4.1.2.1.10. Meierhenry et al. (1983). Dieldrin (> 99% pure) was con-
tinuously fed in the diet for 85 weeks to 50 CsH/He, 62 B6C3Fls and 71 C57BL/6J
male mice starting at the time of weaning. Fifty C3H/He, 76 B6C3Fls and 69
C57BL/6J male mice served as controls. The chow had been analyzed and found
to be free of a panel of insecticides by Gulf South Research Institute. Eight
weeks after the initiation of the experiment and at intervals up to 132 weeks
of age, at least three mice in each group were sacrificed and examined. No
additional information on animals at risk versus time was provided. Dieldrin
(10 ppm) induced a significant increase compared with controls, in the inci-
dence of hepatocellular carcinomas in all three strains of mice (Table 4-21).
Hepatocellular carcinomas were defined as liver nodules that contained tra-
becular areas histologically. The incidence of benign hepatic tumors was also
significantly increased in B6C3Fi and C57BL/6J mice.
4.4.1.2.2. Studies with Rats. Table 4-22 summarizes the carcinogenesis bio-
assays of dieldrin conducted with rats and discussed in this section.
4.4.1.2.2.1. Treon and Cleveland (1955). Dieldrin (> 85% purity) was
administered in the diet of 40 Carworth rats of both sexes at concentrations
of 2.5, 12.5, or 25 ppm, commencing at age 27 to 28 days and continued for 2
years. Groups of 40 animals of both sexes, which were fed powdered or pelleted
diets, served as controls. Dieldrin treatment did not result in an increased
mortality or decreased growth rate as compared to controls. No gross patho-
logical or histological data, or any data on tumor incidences were reported.
However, Cleveland (1966) reported that many tumors had been seen in the 1955
study. This study is unacceptable as a carcinogenicity test because of the
following deficiencies: small number of animals at risk beyond 1 year; no
gross pathology or histology data; and no data on tumor incidence even though a
4-90
-------
TABLE 4-21. SUMMARY OF TUMORS OBSERVED IN C57BL/6J, C3H/He, AND B6C3F1
MICE TREATED WITH DIELDRIN IN THE DIET AT 10 PPM
Strain
C57BL/6J
C3H/He
B6C3F1
Dietary
aldrln
(ppm)
0
10
0
10
0
10
Time of
observation
of first
tumor (wk)
56
46
37
12
57
49
No. of
mice 1n
treatment
group
69
71
50
50
76
62
Percent with
hepatocellular
carcinoma
0
30a
•12
38a
4
423, b
Percent with
benign
hepatic
tumors
14
283
18
20
4
29a
aS1gn1f1cantly different from controls (p < 0.01).
bTwo mice had pulmonary metastases.
SOURCE: Melerhenry et al., 1983.
4-91
-------
TABLE 4-22. SBNKARY OF DIETARY CARCIHOGINESIS BIOASSAYS OF DIEIDRIN IN RATS
ID
ro
Substance
tested Strain Number/Sex
>85X pure Carworth 40/sex/dose
100X pure Osborne-Hendel 12/sex/group
Dose(s)
(ppm) Duration
0 2 yr
2.5
12.5
25
0 2 yr
0.5
2
10
50
100
150
Effects
No treatment-related
Increase In tumors;
No Increased Mortality
or decreased growth In
dleldrln-treated animals
Dose-related decrease
In mortality > 2 ppn.
Dose-related Increase
In liver to body ratios;
Chronic nephritis at
high doses, esp. males;
No liver tumors, al-
though liver lesions
> 50 ppM. Total tumors
Increased (24X In
treated vs. 18X In
controls)
Comments
•Limitations of
study
-Small numbers
alive at 1
year;
-No pathology data
reported
"Inadequate study
"Reuber diagnosed
liver cancers at
100 ppm and con-
firmed multiple-
site tumors
"Limitations of
study
-too few animals
-high mortality
> 2 ppm
"Inadequate study
Reference
Treon and
Cleveland
(1955)
Cleveland
(1966)
Fitzhugh et
al. (1964)
Reuber (as
cited in
Epstein
(1975)
Unspec-
ified
purity
Holtzman
15/N or F/group
0
250
285
No tumor data reported
"Limitations of
study
-too few animals
-too short duration
-lack of data
reported
"Inadequate study
Song and
Harvllle
(1964)
(continued on the following page)
-------
TABLE 4-22. (continued)
ID
CO
Substance
tested Strain
>99X pure Carworth
Technical Osborne-Hendal
Grade
(purity
not
specified)
Dose(s)
Number/Sex (ppm) Duration
25/sex/treated 0 2 yr
group; 45/sex/ 0.1
treated/control 1
10
50/sex/dose 0 31 mo
100/sex/controls 20
30
50
Effects
No effect on body weight
High mortality in all
groups
Dose-related Increase
In liver to body-weight
raUo
No treatment-related
Increase in tumors
Liver pathology at
10 ppm
No effect on body
weight
Survival decreased
at 50 ppm
Liver pathology; no
increase In tumors
with dieldrin exposure
Comments Reference
"Reevaluated by Walker et
Stevenson-confirmed al. (1969)
pathology results Stevenson
•Limitations of et al.
study (1976)
-too few animals
at risk > 1 year;
-only 58X animals
examined
-individual animal
data not presented
"Inadequate study
"Reevaluatlon by Delchmann
Reuben on sample et al.
concluded Delchmann (1970)
underestimated Reuber
Incidences of (as cited
malignant tumors in Epstein,
"Limitations of 1975)
study
-only 84X of animals
microscopically
examined
-control histology
not available
"Hlnlmally accept-
able study
(continued on the following page)
-------
TABLE 4-22. (continued)
10
Substance
tested Strain Number/Sex
955C pure Osborne-Hendel 50/sex/dose
10/sex matched
controls
50 males and 60
females -colony
controls
Dose(s)
(ppn)
0
29
65
Duration
80 uk of
exposure plus
30-31 wk of
observation
Effects
Body weights lower In
second year;
Increase In mortality
of treated rats;
No treatment-related
Increase In tumors
Comments
"While Increased
mortality observed
In treated groups,
It was not signifi-
cantly different at
18 months
"No limitations of
study
"Adequate study
Reference
NCI (1978a)
Technical
Grade
(purified)
Fischer 344
24/sex
0
2
10
50
104-105 wk
No treatment-related
change In body weight
or mortality;
No increase In tumors
related to dieldrln
exposure
"Limitations of
study
-Number of animals
small
'"Adequate study
NCI (1978b)
-------
later report indicated that tumors had been detected.
However, Cleveland (1966) reported the tumor incidence data for these
dieldrin feeding studies conducted by Treon and Cleveland (1955). Cleveland
reported that the incidences of tumors in male and female rats fed dieldrin for
up to 2 years was of the same magnitude as controls. However, the Treon and
Cleveland (1955) study is considered unacceptable as a carcinogen!city test for
the following reasons: number of animals at risk from late-developing tumors,
while not clear, appears to be low; no survival data were presented; data were
presented only for all types of tumors combined; and specific histopathology
data were not presented.
4.4.1.2.2.2. Food and Drug Administration (Fitzhugh et al., 1964).
Dieldrin (100% purity) was administered for 2 years at 0, 0.5, 2, 10, 50, 100,
or 150 ppm to groups of 12 male and 12 female Osborne-Mendel rats, starting
at 3 weeks of age and continuing for 2 years. By 6 months, there was a dose-
related decrease in survival at dieldrin levels > 2 ppm, and the effect was
progressively more marked from 6 to 24 months. A significant and dose-related
increase in liver-to-body weight ratio was reported, although this did not
always occur in both sexes. Chronic nephritis occurred (more commonly in males
than in females) at the high-dose levels.
Only 68% of all animals were examined histologically. Sections of liver,
kidney, testis, as well as tumors and gross abnormalities were made. Total
tumor incidence (all doses combined) in dieldrin-treated animals was 24% com-
pared to an incidence in controls of 18%. According to independent statistical
analyses of these data (Fears, as cited in Epstein, 1975), the multiple-site
carcinogenic effect is statistically significant (significance level not given)
for females and for both sexes combined in low-dose groups. Fitzhugh et al.
(1964) reported no hepatomas or hepatocellular carcinomas, although character-
4-95
-------
istic "chlorinated insecticide" lesions were said to be generally dose-related,
being particularly marked at > 50 ppm.
Tumor histopathology from this experiment plus additional unpublished
material from a separate subsequent FDA study was independently reevaluated by
Reuber (as cited in Epstein, 1975) and others. The higher incidence of multi-
ple-site tumors in dieldrin-treated animals was confirmed, with statistical
significance (significance level not given) at the combined dieldrin 0.5, 2,
and 10 ppm levels. In contrast to Fitzhugh et al. (1964), who reported no
liver tumors, Reuber found liver cancers in 4/7 females and 1/11 males in
the 100 ppm dieldrin group; the incidence is statistically significant (p =
0.0264), both sexes combined in relation to controls (no hyperplastic nodules
or carcinomas were found in the livers of control animals). Selected histo-
logical material was also examined by Drs. Firminger and Farber (Reuber, as
cited in Epstein, 1975) who confirmed the above histological findings.
4.4.1.2.2.3. Song and Harville (1964). Groups of 15 Holtzman rats of
unspecified sex were fed for approximately 1 year with control diets, diets
supplemented with 250 ppm dieldrin (unspecified purity), or diets supplemented
with 15 ppm dieldrin, increased by 10 ppm every 2 weeks until 285 ppm was
reached. Two animals from each group were sacrificed periodically. This study
is unacceptable as a carcinogenicity test because of its short duration, lack
of substantive data, and inadequate number of animals.
4.4.1.2.2.4. Walker et al. (1969). Groups of 25 male and 25 female Car-
worth Farm strain E rats were administered nominal concentrations of 0.1, 1, or
10 ppm dieldrin (> 99% pure) in the diet for 2 years, commencing at 5 weeks of
age. Controls consisted of groups or 45 rats of each sex. The pathology
protocol included gross examination of all animals dying or sacrificed during
the study with microscopic examination of most major organs.
4-96
-------
No effect on body weight occurred at any dose level. Mortality was high
in all groups, so that groups of only 10 to 16 treated and 10 control rats sur-
vived to the end of the test. A treatment-related increase in liver-to-body
weight ratio occurred only in females.
Only 58% of all the animals were examined histologically. Organochlorine
liver changes were reported only in the 10 ppm group at incidences of 4% (1/23)
in males and 26% (6/23) in females. The authors concluded in this study and
in their subsequent reevaluation (Stevenson et al., 1976) that no treatment-
related increase in tumor incidence occurred in rats. A review of the data as
presented supports that conclusion.
However, this study has limitations as a valid test for carcinogenicity
for the following reasons: small number of animals at risk for late-develop-
ing tumors; high premature mortality in all groups; small percentage of animals
examined histologically; and failure to present individual animal data in suf-
ficient detail.
4.4.1.2.2.5. Deichmann et al. (1970). Groups of 50 male and 50 female
Osborne-Mendel rats were administered dieldrin (technical grade) in the diet at
final concentrations of 20, 30, or 50 ppm, commencing at the age of weaning;
dosage for the first 10 weeks was half the above final levels. Groups of 100
rats of both sexes served as controls. The survivors were sacrificed at 31
months.
Feeding dieldrin at the above concentrations had no adverse effect on body
weight. Mean survival, however, was reduced in the 50 ppm group (16.6 months)
compared to controls (19.5 months). Grossly-observed tumors, lesions and sec-
tions of all livers, kidneys, and lungs were to be stained and examined micro-
scopically. However, the authors' tables indicate that sections were examined
for only 87% of dieldrin-exposed animals and 81.5% of the controls.
4-97
-------
A moderate but not dose-related increase in the incidence of centri-
lobular cloudy swelling and liver necrosis was noted in all dieldrin-treated
male and female rats but not in controls. The authors reported a decrease in
tumors in treated animals as compared to controls. Based upon an independent
reevaluation of the histopathology for a limited sample of 36 males and 40
females in the 30 ppm dieldrin group, Reuber (as cited by Epstein, 1975) con-
cluded that the original authors underestimated or under-reported the inci-
dence of malignant tumors by approximately threefold. However, the limited
scope of this reevaluation, along with the fact that control histology was not
available, precluded conclusions on carcinogenicity (Epstein, 1975).
4.4.1.2.2.6. National Cancer Institute (1978a). Groups of 50 Osborne-
Mendel rats of both sexes were administered dieldrin (95% pure) at one of two
doses in the diet. Low-dose rats were treated for 80 weeks, followed by obser-
vation for 30 to 31 weeks. Treatment of high-dose rats was terminated after 59
weeks and followed by 51 to 52 weeks of observation. Time-weighted average
doses were 29 or 65 ppm. Matched controls consisted of 10 untreated rats of
both sexes. Pooled controls, used for statistical evaluation, consisted of
the matched control groups combined with untreated animals (58 males and 60
females) from similar bioassays. All surviving animals were sacrificed at 110
to 111 weeks. A comprehensive pathology examination was performed with only a
minor loss of animals to the study.
Mean body weights of treated rats were comparable to controls during the
first year but lower than the controls during the second year of the study.
Hyperexcitability was observed in all treated groups with increasing frequency
during the second year, especially in high-dose animals. There was a marked
increase in mortality in treated rats during the first 90 weeks; however, due
to a high rate of mortality in controls in the last 20 weeks, survival was not
4-98
-------
statistically dose responsive.
In general, there was no predominance of tumors in any particular organ
associated with dieldrin-treatment. The only possible exception was an in-
crease in the combined incidence of adrenal cortical adenomas or carcinomas in
the low-dose females (6/45) versus the pooled controls (0/55). While this was
statistically significant (p = 0.007), the authors discounted the finding due
to inconsistency in dose-response and variability in other NCI studies.
The authors concluded that none of the tumors occurring in Osborne-
Mendel rats treated with dieldrin could clearly be associated with treatment.
The study is considered an adequate test for carcinogenicity in rats.
4.4.1.2.2.7. National Cancer Institute (1978b). Groups of 24 Fischer 344
rats of both sexes were administered dieldrin (purified technical grade; purity
not indicated) in feed at 2, 10, or 50 ppm for 104 to 105 weeks. Matched
controls consisted of groups of 24 untreated rats of both sexes. All surviving
rats were killed at 104 to 105 weeks. A comprehensive pathology examination
was performed.
Body weights and survival were essentially unaffected by treatment.
Adequate numbers of animals were available for meaningful statistical analyses
of the incidences of tumors. A variety of neoplasms occurred in control and
treated rats, but the incidences were not related to treatment.
The authors concluded that under the conditions of this bioassay, dieldrin
was not carcinogenic in Fischer 344 rats. While the numbers of animals was
small, the dose levels were sufficiently high without inducing unacceptable
mortality or toxicity. It therefore serves as an acceptable bioassay of the
carcinogenicity of dieldrin in rats.
4.4.1.2.3. Studies with Other Species.
4.4.1.2.3.1. Cabral et al. (1979). Groups of Syrian golden hamsters were
4-99
-------
fed a diet containing 0 (40 animals of both sexes), 20 (34 males, 33 females),
60 (32 males, 34 females), or 180 (41 males, 38 females) ppm of dieldrin (99%)
for their lifespan. The age of the animals at initiation of treatment was not
clear. The experiment was terminated at 120 weeks of age, when the last sur-
vivor was sacrificed. A necropsy was performed on all animals, and the thy-
roid, lungs, liver, kidneys, spleen, and gonads together with all organs show-
ing macroscopic abnormalities were examined histologically. Mortality was
high, especially in the control group for which survival at 70 weeks of age
was only 30% in males and 15% in females. Hamsters treated with 180 ppm diel-
drin lived longer than those in all other groups (percent survival at 70 weeks
of age was 49% for males and 24% for females). Both males and females fed 20
and 180 ppm dieldrin showed a marked retardation of growth. The livers in
treated groups showed hepatic cell hypertrophy, and the incidence increased
with increasing doses of dieldrin.
No hepatomas were observed in controls, but hepatomas were found in one
female and one male in the 180 ppm dieldrin group at 85 and 94 weeks of age,
respectively. Although the total tumor incidence in treated animals was not
significantly different from that in controls, more animals in the treated
groups had more than one tumor than animals in the control groups.
This experiment is unacceptable as a test for possible carcinogenicity for
the following reasons: high premature mortality; small numbers of animals at
risk from late-developing tumors; and no lifetable data presentation.
4.4.1.2.3.2. Treon and Cleveland (1955). Groups of two beagle dogs of
each sex were administered control diets or diets supplemented with 1 or 3 ppm
dieldrin for 15 and 16 months and then sacrificed. Liver weights of treated
animals were increased, even at 1 ppm. Vacuolization of the distal renal
tubules was reported in one female of the 3 ppm dieldrin group. This study is
4-100
-------
unacceptable as a test of possible carcinogenicity because of the small number
of animals tested and the short-term nature of the study relative to the normal
lifespan of dogs.
4.4.1.2.3.3. Fitzhugh et al. (1964). A total of 14 mongrel dogs (seven
of each sex), ranging in age from 5 months to 4 years, were fed dieldrin (100%
pure) in dosages of 0.2 to 10 mg/kg/day, 6 days/week for periods up to 25
months. Animals fed 0.5 mg/kg/day or more showed gross toxic effects, includ-
ing loss of weight and convulsions, and they died progressively earlier with
increasing dose level. Histological effects were limited to fatty degenerative
changes in the liver and kidneys and bone marrow changes in those dogs fed 0.5
mg/kg/day dieldrin or more. No gross or microscopic effects were seen in dogs
on dieldrin at 0.2 mg/kg/day. The study is unacceptable as a test of possible
carcinogenicity because of the small number of animals tested and the short-
term nature of the test relative to the normal lifespan of dogs.
4.4.1.2.3.4. Walker et al. (1969). Groups of five male and five female
beagle hounds received, by capsule, daily oral doses of 0.005 or 0.05 mg/kg
(equivalent to approximately 0.1 or 1 ppm) dieldrin (> 99% HEOD), commencing at
4 to 7 months of age. The animals were sacrificed after 2 years. At sacrifice,
the health and body weights of the treated dogs, as compared to controls, were
normal, and there was no observed evidence of gross or microscopic changes in
any organs. The study is unacceptable as a test of possible carcinogenicity
because of the small number of animals tested and the short-term nature of the
test relative to the normal lifespan of dogs.
4.4.1.2.3.5. Wright et al. (1978). Groups of five male rhesus monkeys,
all approximately 4 years old, were fed control diets or diets supplemented with
technical-grade dieldrin (88.4% purity) at initial concentrations of 0.01, 0.1,
0.5, 1, or 5 ppm. One of the monkeys in the 5 ppm group died after 4 months of
4-101
-------
exposure. Dieldrin intoxication was the likely cause of death and, consequent-
ly, the dieldrin content of the diet of this group was reduced to 2.5 ppm fol-
lowing a brief period of being fed the control diet. One of the four remaining
monkeys in this group died of endocrine failure 6 months from the outset of the
experiment. The exposure to dieldrin was not considered to be a contributory
factor in its death. However, as a precautionary measure, the concentration of
dieldrin in the diet of the three survivors was further reduced to 1.75 ppm at
9 months. Subsequently, one of these animals had its dieldrin intake progres-
sively increased until, at the 23rd month from the outset, it was receiving
dieldrin at the initial dietary concentration of 5 ppm. The experiment was
terminated after approximately 6 years.
Apart from small changes associated with the hepatocellular endoplasmic
reticulum, the overall study revealed no effects that could be attributed to
dieldrin. Increases in the activity of the liver microsomal mono-oxygenase
system, determined in^ vitro with a wide range of substrates, provided the most
sensitive criterion for an effect; significant increases, to a maximum of
threefold, were observed in the 1.75 and 5 ppm treatment groups and these
changes were paralleled by increases in cytochrome P-450.
The study is unacceptable as a test of possible carcinogenicity because of
the small number of animals tested and the short-term nature of the test rela-
tive to the average lifespan of the rhesus monkey of 20 years.
4.4.1.2.3.6. Zavon and Stemmer (1975). Thirty-one rhesus monkeys were
fed dieldrin (purity not indicated) for 3 years at levels of 0.01, 0.1, 0.5,
1, and 5 ppm. Weight and body functions remained normal throughout the 3 years
for most animals. No pathology data were reported. This study is not an
acceptable bioassay for carcinogenicity due to the small numbers of animals
and short duration of the treatment as compared to the average lifespan of the
4-102
-------
rhesus monkey.
4.4.2. Epidemiologic Studies
The epidemiologic data for the carcinogenicity of aldrin and dieldrin
consists of two studies. The first is a study of a group of pesticide produc-
tion workers in Holland (Van Raalte, 1977). The cancer cases are presented as
case reports. The second is a retrospective cohort mortality study of workers
in a pesticide manufacturing plant (Ditraglia et al., 1981). Limitations of
both these studies preclude the association of exposure with cancer.
4.4.2.1. Van Raalte (1977)—The author reported his observations on a group of
1000 industrial workers employed at pesticide manufacturing and formulation
plants in Pernis, Holland. The number of plants was not reported. Exposure to
aldrin and dieldrin began in 1954, with exposure to endrin and telodrin begin-
ning later (no specific information was provided). The results focused on
clinical parameters, and the incidence of cancer was discussed only in quali-
tative terms. Of the 1000 workers, a subset of 166 men with dieldrin exposure
was studied in detail. Fifty-one were over 50 years of age. Two cases of
cancer were reported. Among men with "comparatively mild exposure," there was
a cancer death due to gastric cancer. This case had 5 years of "comparatively
mild" exposure. The other case was a lymphosarcoma in a man with 7 years of
"very mild exposure." Both of these cases occurred before 1964. The authors
stated that no new cases had occurred in the last 11 years, but the exact time
period could not be determined. Tumors were not observed in the more heavily
exposed workers. No other causes of death were reported that might mask a
higher cancer incidence. The authors concluded that there was no excess tumor
incidence related to dieldrin exposure. Many study limitations are noted.
Exposure was uncertain as to chemicals and concentration levels. Statistical
methods were not described, no control group was utilized, and the period of
4-103-
-------
follow-up after exposure was probably inadequate to detect a carcinogenic
effect. Thus, this study is inadequate to determine the carcinogenicity of
aldrin and dieldrin.
4.4.2.2. Ditraglia et al. (1981)—A retrospective cohort mortality study of
employees at four organochlorine pesticide manufacturing plants was under-
taken. All workers (race and sex not specified) who had at least 6 months
of employment prior to December 31, 1964, were included in the study popula-
tion. Vital status was determined for each worker as of December 31, 1976,
through the Social Security Administration (SSA), state motor vehicle offices,
U.S. Postal Mail Corrective Services, and "other" sources.
Four separate cohorts representing four pesticide plants composed the
study populations. One plant had manufactured aldrin, dieldrin, and endrin and
had done so since 1946. Other pesticides manufactured at this plant included
organobromines and organophosphates beginning in 1955 and 1956, respectively.
Other chemical products included numerous precursors to the chemicals. The
cohort at this plant consisted of 1,155 individuals, representing 24,939 per-
son-years of observation. Ten percent were lost to follow-up. The three
remaining plants were not involved in the production of aldrin or dieldrin and,
thus, results are not discussed here for these three plants.
Death certificates for all known decedents were obtained and coded by a
nosologlst to the ICD-A in effect at the time of death. Those with an unknown
vital status were assumed alive as of December 31, 1976. The U.S. white male
age-, calender time-, and cause-specific mortality rates were used to calculate
the standardized mortality ratios. Statistical significance between the ob-
served and expected values were tested with the Poisson distribution. Confi-
dence intervals were presented for the SMR estimates, although the method of
calculation was not reported by the authors.
4-104
-------
Statistically significant (p < 0.05) SMR deficits were observed for all
causes of death (SMR = 84, observed = 173, 95% CI 72 to 98) and for diseases
of the circulatory system (SMR = 77, observed = 69, 95% CI 60 to 97). Statis-
tically significant (p < 0.01) SMR increases were observed for nonmalignant
respiratory disease (SMR = 212, observed = 22, 95% CI 133 to 320). The SMR
for all malignant neoplasms was 82. This was not statistically significant.
However, increased SMRs were observed for other specific neoplasms including
cancer of the esophagus (SMR = 235), cancer of the rectum (SMR = 242), cancer
of the liver (SMR = 225), and cancer of the lymphatic and hematopoietic
systems (SMR = 147). None of these was statistically significant.
An analysis by latency was also carried out. Standardized mortality
ratios were calculated for deaths due to all malignant neoplasms according to
latency, which was defined as the number of years from date of first employ-
ment. Latency was categorized into three time periods: (1) less than 10 years
since first employed, (2) 10 to 19 years since first employed, and (3) 20 or
more years since first employed. SMRs of 53, 111, and 64 were reported for the
three latency time periods, respectively. The individual point estimates were
not statistically significant.
This study revealed excess risk of cancer at various tissue sites for all
workers at this plant, though none was statistically significant. However,
there are limitations to this study. No information on quantitative exposures
were available. It was not possible to assess the effects of aldrin or diel-
drin independently of the other toxic chemicals at the plant. No attempt was
made to exclude or adjust for the effects of sex or race, and no effort was
made to control for other confounding variables such as other chemicals pro-
duced in the plant, smoking, or alcohol consumption. However, the size of the
study populations would preclude any real analysis of these variables. Workers
4-105
-------
with little or no occupational exposure to these chemicals (i.e., office work-
ers, etc.) were included in the cohort. Thus, risk may be underestimated for
workers involved in the day-to-day manufacture of aldrin and dieldrin. Also,
since vital statistics were unknown for 10% of the workers, the risks may be
understated. The size of the study population was small in spite of a 25-year
or more period of follow-up. Thus, the power of this study to detect a sta-
tistically significant result is limited. This study provides insufficient
evidence to link aldrin or dieldrin exposure to cancer.
4.5. STRUCTURE-ACTIVITY RELATIONSHIPS
Seven chemicals were identified as important for evaluating the structure-
activity of aldrin and dieldrin for carcinogenicity. These chemicals are
chlordane, chlorendic acid, endrin, endosulfan, heptachlor, hexachlorocyclopen-
tadiene, and isodrin; their structures are presented in Figure 4-5.
4.5.1. Chlordane
4.5.1.1. Evidence for carcinogenicity - human exposure (IARC, 1979a)—Case
reports suggest a relationship between exposure to chlordane or heptachlor
(either alone or in combination with other compounds) and blood dyscrasias.
Another publication has also suggested an association with acute leukemia; an
association between both pre- and postnatal exposure to technical-grade chlor-
dane and the development of neuroblastomas in children was also suggested.
These data are inadequate to assess the carcinogenicity of chlordane.
4.5.1.2. Evidence for carcinogenicity - animal studies (IARC, 1979a)—Chlor-
dane (analytical grade) was tested in four experiments in mice and in four
experiments in rats by oral administration. It produced hepatocellular car-
cinomas in mice of both sexes and in four different strains; for rats, one
study revealed an increase in hepatocellular adenomas. These data provide suf-
ficient evidence for the carcinogenicity of chlordane.
4-106
-------
ChSordans
CI
ci—t
ci-
• ci
Chlor©nds0 acid
C!
Cl
Ci-
C1-,
OH
Heptachlor
Ci
ci -,
Cl-i
Endosuifan
Ssodrin
cr
ci—
HexachSorooycl@p@ntadi@n@
Ci2
Ci-,
ci-
Figure 4-5. Chemicals selected for structure-activity analysis.
4-107
-------
4.5.1.3. Evidence for activity in short-term tests (IARC. 1979a)—Chlordane
induced gene conversions in yeast but was not shown to be mutagenic in bacteria.
It induced mutations in mammalian cells in culture, but was negative in dominant
lethal tests in mice. These data are inadequate to assess the carcinogenicity
of chlordane.
4.5.2. Chlorendic Acid
4.5.2.1. Evidence for carcinogenicity - human exposure (NTP, 1985b)—Chlorendic
acid is manufactured in an essentially closed system. Although this procedure
would seem to minimize human exposure, there are no published data on the level
of occupational exposure to Chlorendic anhydride or chlorendic acid. These
data are inadequate to assess the carcinogenicity of chlorendic acid.
4.5.2.2. Evidence for carcinogenicity - animal studies (NTP, 1985b)—In an
NTP bioassay, there was clear evidence of carcinogenicity for chlorendic acid
in male F344/N rats as shown by increased incidences of neoplastic nodules of
the liver and acinar cell adenomas of the pancreas. Increased incidences of
alveolar/bronchiolar adenomas and preputial gland carcinomas may also have been
related to the administration of chlorendic acid. There was also evidence of
carcinogenicity in female F344/N rats as shown by increased incidences of neo-
plastic nodules and of carcinomas of the liver. There was evidence of carcino-
genicity for chlorendic acid in male B6C3Fi mice as shown by increased inci-
dences of hepatocellular adenomas and of hepatocellular carcinomas. There was
no evidence of carcinogenicity in female B6C3Fi mice given chlorendic acid in
the diet at concentrations of 620 or 1250 ppm for 103 weeks. These data pro-
vide sufficient evidence for the carcinogenicity of chlorendic acid.
4.5.2.3. Evidence of activity in short-term tests (NTP. 1985b)—Chlorendic
acid was not mutagenic in strains TA100, TA98, TA1535, or TA1537 of _S.
typhimurium in the presence or absence of Aroclor 1254-induced male Sprague-
4-108
-------
Dawley rats or male Syrian hamster liver S9 when tested according to the pre-
incubation protocol. Chlorendic acid was mutagenic in the L5178Y/TK+/-mouse
lymphoma cell forward mutation assay in the absence of S9 and was not tested
in the presence of S9. There was no mutagenic response in the absence of
severe toxicity. The toxicity curve was sharp, going from relative total
growth of 74% to 5% at doses of 1600 to 1700 mg/mL. With such a sharp toxic
response, the increase in mutant count and mutant frequency was observed only
at the higher dose; this response was replicated in another experiment. A
response in which only one toxic response is observed is considered as an in-
direct mechanism for which the mutation is not due to the direct interaction of
the chemical with DNA. This assay, as performed, does not answer this question.
These data are inadequate to assess the carcinogenicity of chlorendic acid.
4.5.3. Endrin
4.5.3.1. Evidence for carcinogenicity - human exposure (IARC, 1974c)--A study
was undertaken in 1968 on 233 workers employed in a factory that had been
manufacturing aldrin and dieldrin since 1954 or 1955, endrin since 1957, and
dieldrin during 1958 to 1965. Lengths of exposure ranged from 4 to 13.2 years
,(average, 7.6 years). One hundred and eighty-one workers were still employed
by the same firm at the time of the study, and their average age was 41 years
(range, 22 to 64). Only two deaths had occurred, and one was due to stomach
cancer. The 52 workers who left the company were the subject of a subsequent
report. The average age at the time of this survey was 47.4 years (range,
29 to 72); the average occupational exposure was 6.6 years (4.0 to 12.3); and
the average time since the end of exposure was 7.4 years (4.5 to 16). Only one
death was recorded, and this had not been caused by cancer. These data are
inadequate to assess the carcinogenicity of endrin.
4-109
-------
4.5.3.2. Evidence for cardnogenicity - animal studies (NTP, 1985b)_—There was
no evidence of carcinogenicity in B6C3F1 mice or Osborne-Mendel rats fed endrin,
although there was early mortality in both species. Another rat study was
inadequate. These data are inadequate to assess the carcinogenicity of endrin.
4.5.3.3. Evidence for activity in short-term tests (NTP, 1985b)—Endrin did
not cause mutations in NTP Salmonella mutagenicity tests. These data are
inadequate to assess the carcinogenicity of endrin.
4.5.4. Endosulfan
4.5.4.1. Evidence for carcinogenicity - human exposure (NTP, 1985b)— Specific
mention of endosulfan in analytic epidemiologic studies was not found. Medical
reports mention fainting, convulsions, and unconsciousness following exposure
to endosulfan; however, recovery was rapid. These data are inadequate to
assess the carcinogenicity of endosulfan.
4.5.4.2. Evidence for carcinogenicity - animal studies (NTP, 1985b)_—In an NTP
bioassay, no conclusion was drawn on the carcinogenicity of endosulfan for male
rats or male mice because of early deaths in these groups. Under the conditions
of the bioassay, endosulfan was not carcinogenic in female Osborne-Mendel rats
or in female B6C3F1 mice. These data are inadequate to assess the carcinogen-
icity of endosulfan.
4.5.4.3. Evidence for activity in short-term tests (NTP, 1985b)—No evidence
of the mutagenicity of endosulfan was noted in the standard Ames test and in
the mouse micronucleus test. These data are inadequate to assess the carcino-
genicity of endosulfan.
4.5.5. Heptachlor
4.5.5.1. Evidence for carcinogenicity - human exposure (IARC, 1979b)—Case
reports suggest a relationship between exposure to heptachlor or chlordane
(either alone or in combination with other compounds) and blood dyscrasias.
4-110
-------
Another publication has also suggested an association with acute leukemia;
an association between both pre- and postnatal exposure to technical grade
chlordane containing heptachlor and the development of neuroblastomas in chil-
dren was also suggested. These data are inadequate to assess the carcinogen-
icity of heptachlor.
4.5.5.2. Evidence for carcinogenicity - animal studies (IARC, 1979b)—Hepta-
chlor containing about 20% chlordane was tested in one experiment in mice and
in one in rats by oral administration. Liver carcinomas were observed in mice
of both sexes. In rats, the results suggest a carcinogenic effect on the
thyroid in females. Heptachlor (97% pure) was also inadequately tested in one
experiment in rats by oral administration. A reevaluation of unpublished
studies involving the oral administration of heptachlor of unspecified purity
to mice and rats of other strains confirms the hepatocarcinogenicity of hepta-
chlor for mice and suggests a carcinogenic effect in female rats. These data
provide sufficient evidence for the carcinogenicity of heptachlor.
4.5.5.3. Evidence for activity in short-term tests (IARC. 1979b)—Heptachlor
was not mutagenic in _S. typhimurium or Drosophila melanogaster and was negative
in dominant lethal tests in mice. These data are inadequate to assess the car-
cinogenicity of heptachlor.
4.5.6. Hexachlorocyclopentadi ene
4.5.6.1. Evidence of carcinogenicity - human exposure (U.S. EPA, 1984)—
Although there is human experience with respect to mortality, there is only
limited information on the morbidity results in those exposed to hexachlorocy-
clopentadiene (HEX). Acute inhalation produces a high prevalence of headaches
and severe irritation of the eyes, nose, throat, -and lungs. Dermal contact can
cause severe burns. Epidemiologic studies have generally shown no significant
differences in mortality between workers exposed to HEX in the workplace and
4-111
-------
the general population. Although a significant excess of deaths from cerebro-
vascular disease was reported in one study, the deaths showed no consistent
pattern with duration of employment or follow-up. Current human exposure is
limited to improper handling and disposal and proximity to either manufactur-
ing sites utilizing HEX or disposal sites. These data are inadequate to assess
the carcinogenicity of HEX.
4.5.6.2. Evidence of carcinogenicity - animal studies (U.S. EPA, 1984) — In
vivo bioassays have not been conducted; however, an inhalation bioassay has
been scheduled by the NTP.
4.5.6.3. Evidence for activity in short-term tests (U.S. EPA, 1984)—The
available evidence suggests that HEX is not a mutagen. Negative mutagenicity
results were obtained in bacteria, liver epithelial cells, Drosophila, mouse
lymphoma cells, and in the mouse dominant lethal test. Furthermore, HEX did
not induce unscheduled DNA synthesis in rat hepatocytes. HEX did not induce
in vitro cell transformation in assays using BALB/3T3 cells. These data are
inadequate to assess the carcinogenicity of HEX.
4.5.7. Isodrin
4.5.7.1. Evidence of carcinogenicity - human exposure (HSDB, 1985)—No data
on occupational exposure to this isomer of aldrin were found.
4.5.7.2. Evidence of carcinogenicity - animal studies (HSDB, 1985)—The tox-
icity of isodrin is stated to be twice that of aldrin (species tested and
parameters examined were not specified). No data on the carcinogenicity of
isodrin were found.
4.5.7.3. Evidence for activity in short-term tests (HSDB, 1985)—Isodrin is
converted to endrin by liver microsomes, and endrin does not cause mutations
in NTP Salmonella mutagenicity tests. These data are inadequate to assess the
carcinogenicity of isodrin.
4-112
-------
5. RISK ESTIMATION FROM ANIMAL DATA
5.1. SELECTION OF DATA
For some chemicals, several studies in different animal species, strains,
and sexes at several doses and different routes of exposure may be available.
A choice must be made as to which data sets should be used to quantify human
risk by low-dose extrapolation. The following procedure is used to make this
choice. The animal studies are evaluated qualitatively to assure that only
properly conducted studies are used. The tumor incidence data are then sepa-
rated according to organ sites and tumor types. The data sets used in the
model are the ones-in which the incidence of tumors is statistically signifi-
cantly higher in at least one test dose level compared to controls or where
the tumor incidence rate shows a significant trend with respect to dose level.
Both biological and statistical considerations are used to select appropriate
data sets.
Because humans may be as sensitive as the most sensitive animal species,
potency estimates from the most sensitive species tested can be averaged to
estimate potency for the general population. Because some subpopulations may
be more sensitive than the general population, the potency estimates from the
most sensitive sex and strain tested are also reported. This approach is con-
sistent with EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1986).
5.1.1. Aldrin
Three data sets are suitable for quantitative risk estimation. These are
both male and female C^ti mice in the Davis (1965) study as reevaluated by
Reuber and cited in Epstein (1975), and male B6C3F! mice in the NCI (1978a)
bioassay.
5-1
-------
5.1.2. Dieldrin
Thirteen data sets are suitable for quantitative risk estimation. These
include both male and female C3H mice in the Davis (1965) study as reevaluated
by Reuben, male and female CF-1 mice in the Walker et al. (1972) study, male and
female CF-1 mice in the Thorpe and Walker (1973) study, male B6C3F1 mice in the
NCI (1978a, b) bioassay, male CF-1 mice in the Tennekes et al. (1981) study, and
male C57BL/6J, C3H, and B6C3F1 mice in the Meierhenry et al. (1983) study.
5.2. CHOICE OF EXTRAPOLATION MODELS
Mathematical models are used to estimate human excess cancer risks associ-
ated with exposure to a given chemical. The models are fit to tumor incidence
data from animal bioassays, and an extrapolation is made to low doses to esti-
mate human risk. The animal bioassay data used to predict human cancer risk
from exposure to aldrin and dieldrin are described in the previous section.
Although no single mathematical model is recognized as the most appro-
priate, some mechanisms of the process have become known. If one accepts the
linearity of tumor formation and if the carcinogenic agent of interest acts in
the same manner as carcinogenic agents acting in the background, then the added
effect of the carcinogen on tumor formation at low doses is linear. Thus, a
model that is linear at low doses is appropriate. The linearized multistage
,-./>•
model, which has these properties, is used in this assessment to calculate
human risk. This model takes the following form:
P(d) = 1 - exp [-(q0 + q:d + q£d2 + ... + qkdk)]
where q-j >_Q, i = 0, 1, ..., k dose levels, and d = dose.
5-2
-------
Cancer risks can also be estimated using models such as the probit, logit,
Vlelbu.ll, one-hit, and gamma multi-hit. The linearized multistage model leads
to an upper limit on risk.
5.3. INTERSPECIES DOSE CONVERSION
The low-dose extrapolation process involves the use of animal tumor inci-
dence data to predict human cancer risk. Differences exist between humans and
experimental and pharmacokinetic mechanisms. Hence, animal dosages need to be
converted to human equivalent dosages. The conversion accounts for noncontinu-
ous feeding, differences in lifetime, and size differences. In the absence of
information to the contrary, this assessment uses the surface area correction
recommended by EPA's Guidelines for Carcinogen Risk Assessment (U.S. EPA,
1986). Therefore, a factor of the cube root of the ratio of the animal to
human body weights is applied to animal data to calculate a human equivalent
dose. This equation is:
DH = DA (WA/70)1/3
where
DH = human equivalent dosage (mg/kg/day)
DA = animal dosage (mg/kg/day)
WA = weight of the animal (kg) assumed (0.03 kg for mice,
0.35 kg for rats)
70 = weight of a man (kg).
The surface area correction factor used in this assessment is (0.03/70)!/3 =
0.08. In the absence of information on human absorption, tissue distribution,
metabolism, and excretion, this assessment makes no adjustment for potential
differences between animals and humans.
5-3
-------
5.4. ESTIMATION OF CARCINOGENIC POTENCY
Estimates of carcinogenic potency can be obtained by fitting the linear-
ized multistage model, described in Section 5.2 to each data set described in
Section 5.1. Table 5-1 summarizes 16 potency estimates obtained in this way.
Tables 5-2 to 5-17 describe the data sets used for these estimates.
5.4.1. Aldrin
Three data sets are suitable for quantitative risk estimation: male and
female C$\ mice, and male B6C3Fi mice. The most sensitive sex and strain tested
is female CsH mice. From these the potency can be estimated at 23 per mg/kg/day,
The most sensitive species tested is mice. There are three potency esti-
mates, ranging from 23 down to 12 per mg/kg/day, with a geometric mean of 17
per mg/kg/day. Because humans may be as sensitive as the most sensitive animal
species, the potency for the general population is estimated at 17 per mg/kg/
day.
These estimates are plausible upper bounds for the increased cancer risk
from aldrin, meaning that the true risk is not likely to exceed these estimates
and may be lower.
The molecular potency index, which is the potency expressed in terms of
molecular weight, has been used to rank suspect carcinogens according to poten-
cy. The index is computed by multiplying the general-population potency by the
molecular weight. The molecular potency index for aldrin is 6.2 x 10^ per
mmol/kg/day. This places aldrin in the most potent quartile of suspect car-
cinogens ranked by the CAG.
The unit risk in air, which is the potency expressed in terms of yg/m^
ambient air concentration, is used to estimate risk when exposures are ex-
pressed as air concentrations. The unit risk in air is the increased cancer
risk to a person who throughout life breathes air contaminated with 1 yg/m^ of
5-4
-------
TABLE 5-1. HUMAN POTENCY ESTIMATES BY CHEMICAL
Sex, species, Tumor site Potency
Chemical strain and type (mg/kg/day)"1 Reference
Aldrln Male, mice, Liver, carcinoma 18
C3H
Aldrln Female, mice, Liver, carcinoma 23
C3H
Aldrln Male, mice, Liver, carcinoma 12
B6C3F1
Davis (1965),
reevaluated by
Reuber (cited
1n Epstein, 1975)
Davis (1965),
reevaluated by
Reuber (cited
1n Epstein, 1975)
NCI (1978a)
Dleldrln Male, mice,
CF-1
Dleldrln Female, mice,
CF-l
Dleldrln
Dleldrln
Dleldrln
Dleldrln
Dleldrln
Dleldrln
Dleldrln
Male, mice,
CF-1
Female, mice,
CF-1
Male, mice,
CF-1
Female, mice
CF-1
Male, mice
CF-1
Female, mice
CF-1
Male, mice,
B6C3F1
Liver, carcinoma 22
Liver, carcinoma 25
Liver, carcinoma 25
Liver, carcinoma 28
Liver, carcinoma 15
Liver, carcinoma 7.1
Liver, carcinoma 55
Liver, carcinoma 26
Liver, carcinoma 9.8
Davis (1965),
reevaluated by
Reuber (cited
In Epstein, 1975)
Davis (1965),
reevaluated by
Reuber (cited
In Epstein, 1975)
Walker et al. (1972)
Walker et al. (1972)
Walker et al. (1972)
Walker et al. (1972)
Thorpe and Walker (1973)
Thorpe and Walker (1973)
NCI (1978a)
(continued on the following page)
5-5
-------
TABLE 5-1. (continued)
Sex, species. Tumor site
Chemical strain and type
Potency
(mg/kg/day)-'
Reference
Dleldrln Hale, mice
CF-1
Dleldrln Hale, n»1ce
C57BL/60
Dleldrln Hale, mice
C3H/He
Dleldrln Hale, mice
B6C3F1
Liver, carcinoma 18
Liver, carcinoma 7.4
Liver, carcinoma 8.5
Liver, carcinoma 11
Tennekes et al. (1981)
Helerhenry et al. (1983)
Helerhenry et al. (1983)
Helerhenry et al. (1983)
5-6
-------
TABLE 5-2. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF ALDRIN
FROM TUMOR INCIDENCE IN NICE
Compound: aldrln
Species, strain, sex: mice, C3H, female
Body weight: 0.030 kg (assumed)
Length of experiment: 2 years
Length of exposure: 2 years
Tumor site and type: liver, carcinoma
Route, vehicle: oral, diet
Human potency: 23 per mg/kg/day
Experimental
dose (ppm)
0
10
Transformed
dose (mg/kg/day)
0
1.3
Human equivalent
dose (mg/kg/day)
0
0.104
Incidence
No. responding/
No. examined
2/53
72/85
SOURCE: Davis, 1965 (evaluated by Reuber; as cited 1n Epstein, 1975).
5-7
-------
TABLE 5-3. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF ALDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: aldrln
Species, strain, sex: mice, C3H, male
Body weight: 0.030 kg (assumed)
Length of experiment: 2 years
Length of exposure: 2 years
Tumor site and type: liver, carcinoma
Route, vehicle: oral, diet
Human potency: 18 per mg/kg/day
Incidence
Experimental Transformed Human equivalent No. responding/
dose (ppm) dose (mg/kg/day) dose (mg/kg/day) No. examined
0
10
0
1.3
0
0.104
22/73
75/91
SOURCE: Davis and FHzhugh, 1962 (evaluated by Reuber; as cited In Epstein,
1975).
5-8
-------
TABLE 5-4. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF ALDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: technical grade aldrln
Species, strain, sex: mice, B6C3F1, male
Body weight: 0.030 kg (assumed)
Length of experiment: 93 weeks
Length of exposure: 80 weeks
Tumor site and type: liver, carcinoma
Route, vehicle: oral, diet
Human potency: 12 per mg/kg/day
Experimental
dose (ppm)
0
4
8
Transformed
dose {mg/kg/day)
0
0.52
1.04
Human equivalent
dose (mg/kg/day)a
0
0.04
0.08
Incidence
No. responding/
No. examined
17/92
16/49
25/45
aHuman equivalent doses were multiplied by 80/93 to reflect exposure less
than the duration of the experiment.
SOURCE: NCI, 1978a.
5-9
-------
TABLE 5-5. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mice, C3H, male
Body weight: 0.030 kg (assumed)
Length of experiment: 2 years
Length of exposure: 2 years
Tumor site and type: liver, carcinoma
Route, vehicle: oral, diet
Human potency: 22 per mg/kg/day
Incidence
Experimental
dose (ppm)
0
10
Transformed
dose (mg/kg/day)
0
1.3
Human equivalent
dose (mg/kg/day)
0
0.104
No. responding/
No. examined
22/73
62/71
SOURCE: Davis, 1965 (evaluated by Reuber; as cited 1n Epstein, 1975).
5-10
-------
TABLE 5-6. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUHOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mice, C3H, female
Body weight: 0.030 kg (assumed)
Length of experiment: 2 years
Length of exposure: 2 years
Tumor site and type: liver, carcinoma
Route, vehicle: oral, diet
Human potency: 25 per mg/kg/day
Incidence
Experimental
dose (ppm)
0
10
Transformed
dose (mg/kg/day)
0
1.3
Human equivalent
dose (mg/kg/day)
0
0.104
No. responding/
No. examined
2/53
62/71
SOURCE: Davis, 1965 (evaluated by Reuber; as cited 1n Epstein, 1975).
5-11
-------
TABLE 5-7. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mice, CF-1, male
Body weight: 0.030 kg (assumed)
Length of experiment: 33 months
Length of exposure: 24 months
Tumor site and type: liver, carcinoma
Route, vehicle: oral, diet
Human potency: 25 per mg/kg/day
Experimental
dose (ppm)
0
0.1
1
10
Transformed
dose (mg/kg/day)
0
0.013
0.13
1.3
Human equivalent
dose (mg/kg/day)
0
0.001
0.01
0.104
Incidence
No. responding/
No. examined
58/288
32/124
34/111
165/176
SOURCE: Walker et al., 1972.
5-12
-------
TABLE 5-8. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mice, CF-1, female
Body weight: 0.030 kg (assumed)
Length of experiment: 33 months
Length of exposure: 24 months
Tumor site and type: "liver, carcinoma
Route, vehicle: oral, diet
Human potency: 28 per mg/kg/day
Experimental
dose (ppm)
0
0.1
1
10
Transformed
dose {mg/kg/day)
0.0013
0.013
0.13
1.3
Human equivalent
dose {mg/kg/day)
0.0001
0.001
0.01
0.104
Incidence
No. responding/
No. examined
39/297
24/90
32/87
136/148
SOURCE: Walker et a!., 1972.
5-13
-------
TABLE 5-9. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mouse, CF-1, male
Body weight: 0.030 kg (assumed)
Length of experiment: 128 weeks
Length of exposure: 128 weeks
Tumor site and type: Hver
Route, vehicle: oral, diet
Human potency: 15 per mg/kg/day
Experimental Average dally Human equivalent
dose (ppm) dose (mg/kg/day) dose (mg/kg/day)
00 0
1.25 0.16 0.012
2.5 0.33 0.025
5 0.65 0.049
10 1.30 0.098
20 2.60 0.196
Tumor
Incidence
9/78
6/30
13/30
26/30
5/11
12/17
SOURCE: Walker et al., 1972.
5-14
-------
TABLE 5-10. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mouse, CF-1, female
Body weight: 0.030 kg (assumed)
Length of experiment: 128 weeks
Length of exposure: 128 weeks
Tumor site and type: liver
Route, vehicle: oral, diet
Human potency: 7.1 per mg/kg/day
Experimental
dose (ppm)
0
1.25
2.5
5
10
20
Average dally
dose {mg/kg/day)
0
0.16
0.33
0.65
1.30
2.60
Human equivalent
dose {mg/kg/day)
0
0.012
0.025
0.049
0.098
0.196
Tumor
Incidence
8/78
5/30
1 2/28
18/30
9/17
8/21
SOURCE: Walker et a!., 1972.
5-15
-------
TABLE 5-11. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mouse, CF-1, male
Body weight: 0.030 kg (assumed)
Length of experiment: 110 weeks
Length of exposure: 110 weeks
Tumor site and type: liver
Route, vehicle: oral, diet
Human potency: 55 per mg/kg/day
Experimental Average dally Human equivalent Tumor
dose (ppm) dose (mg/kg/day} dose (mg/kg/day)a Incidence3
000 11/45
10 1.30 0.098 30/30
aTo calculate human potency, a tumor Incidence of 29/30 was used and the
dose was adjusted to 29/30 of the above value.
SOURCE: Thorpe and Walker, 1973.
5-16
-------
TABLE 5-12. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mouse, CF-1, female
Body weight: 0.030 kg (assumed)
Length of experiment: 110 weeks
Length of exposure: 110 weeks
Tumor site and type: liver
Route, vehicle: oral, diet
Human potency: 26 per mg/kg/day
Experimental Average dally Human equivalent Tumor
dose (ppm) dose (mg/kg/day) dose (mg/kg/day) Incidence
0 00 10/44
10 1.30 0.104 26/30
SOURCE: Thorpe and Walker, 1973.
5-17
-------
TABLE 5-13. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: technical-grade dleldrln
Species, strain, sex: mice. B6C3F1, male
Body weight: 0.030 kg (assumed)
Length of experiment: 93 weeks •
Length of exposure: 80 weeks
Tumor site and type: liver, carcinoma
Route, vehicle: oral, diet
Human potency: 9.8 per mg/kg/day
Incidence
Experimental
dose (ppm)
0
2.5
5
Transformed
dose (mg/kg/day)
0
0.325
0.65
Human equivalent
dose (mg/kg/day )a
0
0.026
0.052
No. responding/
No. examined
17/92
12/50
16/45
aHuman equivalent doses were multiplied by 80/93 to reflect exposure less
than the duration of the experiment.
SOURCE: NCI. 1978a.
5-18
-------
TABLE 5-14. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROH TUMOR INCIDENCE IN MICE
Compound: Dleldrln
Species, strain, sex: mouse, CF-1
Body weight: 0.030 kg (assumed)
Length of experiment: 110 weeks
Length of exposure: 110 weeks
Tumor site and type: liver
Route, vehicle: oral, diet
Human potency: 18 per mg/kg/day
male
Experimental
dose (ppm)
0
10
Average dally
dose (mg/kg/day)
0
1.30
Human equivalent
dose (mg/kg/day)
0
0.104
Tumor
Incidence
25/252
113/139
SOURCE: Tennekes et al., 1981.
5-19
-------
TABLE 5-15. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mouse, C57BL/63, male
Body weight: 0.030 kg (assumed)
Length of experiment: 132 weeks
Length of exposure: 85 weeks
Tumor site and type: liver, hepatocellular carcinoma
Route, vehicle: oral, diet
Human potency: 7.4 per mg/kg/day
Experimental
dose (ppm)
0
10
Average dally
dose (mg/kg/day)
0
1.30
Human equivalent
dose (mg/kg/day)a
0
0.104
Tumor
Incidence
0/69
21/71
aHuman equivalent doses were multiplied by 85/132 to reflect exposure less
than the duration of the experiment.
SOURCE: Helerhenry et a!., 1983.
5-20
-------
TABLE 5-16. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mouse, C3H/He, male
Body weight: 0.030 kg (assumed)
Length of experiment: 132 weeks
Length of exposure: 85 weeks
Tumor site and type: liver, hepatocellular carcinoma
Route, vehicle: oral, diet
Human potency: 8.5 per mg/kg/day
Experimental
dose (ppm)
0
10
Average dally
dose (mg/kg/day)
0
1.30
Human equivalent
dose (mg/kg/day)a
0
0.104
Tumor
Incidence
6/50
19/50
aHuman equivalent doses were multiplied by 85/132 to reflect exposure less
than the duration of the experiment.
SOURCE: Melerhenry et al., 1983.
5-21
-------
TABLE 5-17. CANCER DATA SHEET FOR DERIVATION OF POTENCY OF DIELDRIN
FROM TUMOR INCIDENCE IN MICE
Compound: dleldrln
Species, strain, sex: mouse, B6C3F1, male
Body weight: 0.030 kg (assumed)
Length of experiment: 132 weeks
Length of exposure: 85 weeks
Tumor site and type: liver, hepatocellular carcinoma
Route, vehicle: oral, diet
Human potency: 11 per mg/kg/day
Experimental
dose (ppm)
0
10
Average dally
dose (mg/kg/day)
0
1.30
Human equivalent
dose (mg/kg/day )a
0
0.104
Tumor
Incidence
3/76
26/62
aHuman equivalent doses were multiplied by 85/132 to reflect exposure less
than the duration of the experiment.
SOURCE: Melerhenry et a!., 1983.
5-22
-------
a pollutant. With a linear dose-response curve, risks at any concentration can
be computed by multiplying the unit risk by the concentration. The unit risk
is computed by converting a 1 yg/m3 concentration to a mg/kg/day dose and then
multiplying by the potency. For a 70-kg person who breathes 20 m3/day, an air
concentration of 1 yg/m3 is equivalent to a dose of:
(1 yg/m3) (10-3 mg/yg) (20 m3/day) / (70 kg) = 2.89 x 10-4 mg/kg/day.
Multiplying this dose by the potencies calculated above gives unit risks
of 4.9 x 10"3 per yg/m3 and 6.6 x 10"3 per yg/m3 for potencies of 17 and 23,
respectively.
The unit risk in water, which is the potency expressed in terms of yg/L
drinking water concentration, is used to estimate risk when exposures are
expressed as water concentrations. The unit risk in water is the increased
cancer risk to a person who throughout life drinks water contaminated with 1
yg/L of a pollutant. With a linear dose-response curve, risks at any concen-
tration can be computed by multiplying the unit risk by the concentration. The
unit risk is computed by converting a 1 yg/L concentration to a mg/kg/day dose
and then multiplying by the potency. For a 70-kg person who drinks 2 L/day, a
water concentration of 1 ug/L is equivalent to a dose of:
(1 yg/L) (10-3 mg/yg) (2 L/day) / (70 kg) = 2.9 x 10-5 mg/kg/day.
Multiplying this dose by the potencies calculated above gives unit risks
of 4.9 x 1Q-4 per yg/L and 6.6 x 10'^ per yg/L for potencies of 17 and 23,
respectively.
5-23
-------
5.4.2. Dleldrin
Thirteen data sets are suitable for quantitative risk estimation. The most
sensitive sex and strain tested is male CF^ mice in the Thorpe and Walker (1973)
study. From these data the potency can be estimated at 55 per mg/kg/day.
The most sensitive species tested is mice. There are 13 potency
estimates, ranging from 55 down to 7 per mg/kg/day, with a geometric mean of
16 per mg/kg/day. Therefore the potency for the general population is
estimated at 16 per mg/kg/day.
These estimates are plausible upper bounds for the increased cancer
risk from dieldrin, meaning that the true risk is not likely to exceed these
estimates and may be lower.
The molecular potency index for dieldrin is 6.1 x 103 per mmol/kg/day.
This places dieldrin in the most potent quartile of suspect carcinogens ranked
by the CAG.
The potency estimates yield unit risks in air of 4.6 x 10~3 per
ug/rn3 and 1.6 x 10~2 per yg/m3. The unit risks in water are 4.6 x 10~4 per
pg/L and 1.6 x 10~3 per yg/L.
5-24
-------
6. REFERENCES
Baldwin, M.K.; Robinson, J.; Park, D.V. (1972) A comparison of the metabolism
of HEOD (dieldrin) in the CF-1 mouse with that in the CFE rat. Food
Cosmet. Toxicol. 10:333.
Ball, W.U; Kay, K.; Sinclair, J.W. (1953) Observation on toxicity of aldrin.
I. Growth and estrus in rats. Arch, Ind. Hyg. Occup. Med. 7:292-300.
Bann, J.M.; DeCino, T.J.; Earle, N.W.; Sun, Y.P. (1956) The rate of aldrin
and dieldrin in the animal body. J. Agric. Food Chem. 4:937-941.
Barquet, A.; Morgade, C.; Pfaffenberger, C.D, (1981) Determination of organo-
chlorine pesticides and metabolites in drinking water, human blood serum,
and adipose tissue. J. Toxicol. Environ. Health 7:469-479.
Barren, R.L.; Walton, M.S, (1971) Dynamics of HEOD (dieldrin) in adipose
tissue of the rat. Toxicol. Appl. Pharmacol. 18:958.
Bedford, C.T.; Hutson, D.H. (1975) The comparative metabolism in rodents of
the isomeric insecticides dieldrin and endrin. Chem. Ind. 10:440.
Benitz, K.F.; Roth, R.N.; Coulston, F. (1977) Morphologic characteristics of
hepatic nodules induced by mi rex and dieldrin. Toxicol. Appl. Pharmocol„
41:154-155.
Beyermann, K,; Eckrich, W. (1973) Gas-chromatograph,ische Bestimmung von
Insecticid-Spuren in Luft. Z. Anal. Chem. 265:4-7.
Borgmann, A.R.; Kitselman, C.H.; Dahm, P.A.; Pankaskie, J.E.; Dutra, F.R.
(1952) Toxicological studies of aldrin on small laboratory animals.
Unpublished report of Kansas State College, July, 34 pp.
Boucard, M.; Beaulaton, I.S.; Mestre, S.R.; Allieu, M. (1970) Etude experi-
mental e de la teratogenese: influence de la peri ode et de la duree du
traitment. Therapie 25:907-913.
Bragt et al. (1984) As cited in U.S. Environmental Protection Agency Special
Data Call-in on Aldrin Termiticide, Data Packet on Metabolism, July 27,
1984.
Brooks, G.T.; Harrison, A. (1966) Metabolism of aldrin and dihydroaldrin by
houseflies (M. domestica L.) in vivo and by housefly and pig liver micro-
somes. Life"~Sc"i. 5:2315-2320.
Burchfield, J.L.; Duffy, F.H.; Sim, V.M. (1976) Persistent effects of sarin
and dieldrin upon the primate electroencephalogram. Toxicol. Appl.
Pharmacol. 35:365-379.
Cabral, J.R.P.; Hall, R.K.; Bronczyk, S.A.; Shubik, P. (1979) A carcinogen-
icity study of the pesticide dieldrin in hamsters. Cancer Lett. 6:241-246.
6-1
-------
Chemical economics handbook. (1985) Men!o Park, CA: SRI International,
Section 638.5030K.
Chernoff, N.; Kavlock, R.J.; Katherin, J.R.; Dunn, J.M.; Haseman, J.K. (1975)
Prenatal effects of dieldrin and photodieldrin in mice and rats. Toxicol.
Appl. Pharmacol. 31:302-308.
Chipman, J.K.; Walker, C.H. (1979) The metabolism of dieldrin and two of its
analogues: the relationship between rates of microsomal metabolism and
rates of excretion of metabolites in the male rat. Biochem. Pharmacol.
28:1337-1345.
Chu, K.C.; Cueto, C. Jr.; Ward, J.M. (1981) Factors in the evaluation of 200
National Cancer Institute carcinogen bioassays. J. Toxicol. Environ.
Health 8:251-280.
Cleveland, P.P. (1966) A summary of work on aldrin and dieldrin toxicity at
the lettering Laboratory. Arch. Environ. Health 13:195.
Cole, J.F.; Klevay, L.M.; Zavon, M.R. (1970) Endrin and dieldrin: a compari-
son of hepatic excretion in the rats. Toxicol. Appl. Pharmacol. 16:547.
Cueto, C., Jr.; Hayes, W.J., Jr. (1962) The detection of dieldrin metabolites
in human urine. J. Agric. Food Chem. 10:366-369.
Cueto, C., Jr.; Biros, F.J. (1967) Chlorinated insecticides and related
material in human urine. Toxicol. Appl. Pharmacol. 10:261.
Davies, M.W.; Keysell, S.R. (1983) Sex differences in the metabolism of aldrin
by rat liver preparations. J. Pharm. Pharmacol. 35(Suppl.):82.
Davis, K.J. (1965) Pathology report on mice fed aldrin, dieldrin, heptachlor
or heptachlor expoxide for two years. Internal FDA memorandum to Dr. A.J.
Lehrman, July 19, 1965.
Davis, K.J.; Fitzhugh, 0.6. (1962) Tumorigenic potential of aldrin and diel-
drin for mice. Toxicol. Appl. Pharmacol. 4:187-189.
Davison, K.L. (1973) Dieldrin-^C balance in rats, sheep, and chickens. Bull,
Environ. Contam. Toxicol. 10:16-24.
Deichmann, W.B. (1972) Toxicology of DDT and related chlorinated hydrocarbon
pesticides. J. Occup. Med. 14:285-292.
Deichmann, W.B. (1973) Progress report on the lifetime feeding studies with
Swiss-Webster mice for diets supplemented with 0, 3, and 10 ppm dieldrin
(prepared for the U.S. EPA and Shell International Research, The Hague,
September 21, 1973).
Deichmann, W.B. (1974) Statement of testimony for aldrin/dieldrin hearings
suspension phase, before the U.S. Environmental Protection Agency, August
23, 1974.
6-2
-------
Deichmann, W.B.; Keplinger, M.; Sala, F.; Glass, E. (1967) Synergism among
oral carcinogens in the simultaneous feeding of four tumorigens to rats.
Toxicol. Appl. Pharmacol. 11:88-103.
Deichmann, W.B.; Dressier, I.; Keplinger, M.; MacDonald, W.E. (1968) Reten-
tion of dieldrin in blood, liver, and fate of rats fed dieldrin for six
months. Ind. Med. Surg. 37:837-840.
Diechmann, W.B.; Kiplinger, M.; Dressier, I.; Sala, F. (1969) Retention of
dieldrin and DDT in the tissue of dogs fed aldrin and DDT individually
and as mixture. Toxicol. Appl. Pharmacol. 14:205-213.
Deichmann, W.B.; MacDonald, W.E.; Blum, E.; Bevilacqua, M.; Radomski, J.;
Keplinger, M.; Balkus, M. (1970) Tumorigenicity of aldrin, dieldrin and
endrin in the albino rat. Ind. Med. Surg. 39:426-434.
Deichmann, W.B.; MacDonald, W.E.; Beasley, A.G.; Cubit, D. (1971) Subnormal
reproduction in beagle dogs induced by DDT and aldrin. Ind. Med. Surg.
40:10-20.
Deichmann, W.B.; MacDonald, W.E.; Cubit, D.A. (1975) Dieldrin and DDT in the
tissues of mice fed aldrin and DDT for seven generations. Arch. Toxicol.
34:173-182.
Ditraglia, D.; Brown, D.P.; Namekata, T.; Iverson, N. (1981) Mortality study
of workers employed at organochlorine pesticide manufacturing plants.
Scand. J. Work Environ. Health 7:140-146 (Suppl. 4).
Dix, K.M.; Vanderpauw, C.L.; McCarthy, W.V. (1977) Toxicity studies with
dieldrin: teratological studies in mice dosed orally with HEOD. Tera-
tology 16:57-62.
Domanski, J.J., et al. (1977) Relation between smoking and levels of DDT and
dieldrin in human fat. Arch. Environ. Health 32:196.
Eckenhausen, F.W.; Bennet, D.; Beynon, K.I.; Elgar, K.E. (1981) Organochlorine
pesticide concentrations in perinatal samples from mothers and babies.
Arch. Environ. Health 36:81-92.
Epstein, S.S. (1975) The carcinogenicity of dieldrin. Part I. Sci. Total
Environ. 4:1-52.
Epstein, S.S. (1976) Case-study 5: aldrin and dieldrin suspension based on
experimental evidence and evaluation and societal need. Ann. N.Y. Acad.
Sci. 271:187.
Farb, R.M.; Sanderson, T.; Moore, B.G.; Hayes, A.W. (1973) Interaction: the
effect of selected mycotoxins on the tissue distribution and retention of
aldrin and dieldrin in the neonatal rat. Presented at the 8th Inter-
America Conference on Toxicology and Occupational Medicine.
Fears, T. (1974) Exhibit 52, Testimony at hearing on aldrin/dieldrin. (Cited
in Epstein, 1975.)
6-3
-------
Feldmann, R.J.; Maibach, H.I. (1974) Percutaneous penetration of some
pesticides and herbicides in man. Toxicol. Appl. Pharmacol. 28:126-132.
Fitzhugh, O.6.; Nelson, A.A.; Quaife, M.L. (1964) Chronic oral toxicity of
aldrin and dieldrin in rats and dogs. Food Cosmet. Toxicol. 2:551-562.
Garrettson, L.K.; Curley, A. (1968) Dieldrin studies in a poisoned child.
Arch. Environ. Health 19:814-272.
Good, E.E.; Ware, G.W. (1969) Effects of insecticides on reproduction in the
laboratory mouse. IV. Endrin and dieldrin. Toxicol. Appl. Pharmacol.
14:201-203.
Griesemer, R.A.; Cueto, C. Jr. (1980) Toward a classification scheme for
degrees of experimental evidence for the cardnogenicity of chemicals for
animals. Molecular and cellular aspects of carcinogen screening tests.
IARC Scientific Publications No. 27. International Agency for Research
on Cancer, Lyon, France.
Gupta, P.C. (1975) Neurotoxicity of chronic chlorinated hydrocarbon insecti-
cide poisoning—a clinical and electro-encephalographic study in man.
Indian J. Med. Res. 63:601-605.
Hamilton, H.E.; Morgan, D.P.; Simmons, A. (1978) A pesticide (dieldrin)-
induced immunohemolytic anemia. Environ. Res. 17:155-164.
Harr, J.R.; Claeys, R.; Bone, J.F.; McCorcle, T.W. (1970) Dieldrin toxicosis:
rat reproduction. Am. J. Vet. Res. 31:181-189.
Harris, R.H., et al. (1977) Carcinogenic hazards of organic chemicals in
drinking water. In: Hiatt et al., eds. Origins of human cancer. New
York: Cold Spring Harbor Laboratory, p. 309.
Hawley, G.G. (1981) The condensed chemical dictionary, 10th ed. New York:
Van Nostrand Reinhold Co., p. 433.
Hayes, W.O., Jr. (1971) Toxicology of pesticides. Baltimore, MD: Williams &
Wilkins Co.
Hayes, W.J., Jr. (1974) Distribution of dieldrin following a single oral
dose. Toxicol. Appl. Pharmacol. 28:485-492.
Hayes, W.J., Jr. (1975) Toxicology of pesticides. Baltimore, MD: Williams &
Wilkins Co., pp. 327-370.
Hayes, W.J., Jr. (1982) Pesticides studied in man. Baltimore, MD: Williams &
Wilkins Co., pp. 234-247.
Hazardous Substances Technical Data Bank (HSDB). Data base. (1985) Bethesda,
MD: National Library of Medicine, U.S. Department of Health and Human
Services.
6-4
-------
Heath, D.F.; Vandekar, M. (1964) Toxicity and metabolism of dieldrin in rats.
Br. J. Ind. Med. 21:269r279.,7 ,.
Hodge, H.C.; Boyce, A.M.; Deichmann. W.B.; Kraybill, H.F. (1967) Toxicology
and no-effect levels of aldrin and dieldrin. Toxicol. Appl. Pharmacol.
10:613-675.
Hoogendam, I.; Versteeg, J.P.J.; DeVlieger, M. (1962) Electroencephalograms
in insecticide toxicity. Arch. Environ. Health 4:92-100.
Hunter, C.6.; Robinson, J. (1967) Pharmacodynamics of dieldrin (HEOD). I.
Ingestion by human subjects for 18 months. Arch. Environ. Health 15:614-
626.
Hunter, C.G.; Rosen, A.; Williams, R.T.; Reynolds, J.G.; Borden, A.N. (1960)
The fate of aldrin, dieldrin, and endrin in the mammal. Meded. Landbouw-
hogesch. Opzoekingsstn. Staat Gent 25:1296-1307 (as cited in Soto and
, Deichmann, 1967).
Hunter, C.G.; Robinson, J.; Roberts, M. (1969) Pharmacodynamics of dieldrin
(HEOD) ingestion by human subjects for 18 to 24 months, and postexposure
for 8 months. Arch. Environ. Health 18:12-21.
Hutson, D.H. (1976) Comparative metabolism of dieldrin in the rat (CFE) and
in two strains of mouse (CF-1 and LACG). Food Cosmet. Toxicol. 14:577-591.
latropoulos, M.J.; Milling, A.; Muller, W.F.; Nohynek, G.; Rozman, K.; Coulston,
F.; Korte F. (1975) Absorption, transport and organotropism of dichloro-
biphenyl (DCB), dieldrin, and hexachlorobenzene (HCB) in rats. Environ.
Res. 10:384-389.
International Agency for Research on Cancer (IARC). (1974a) IARC monographs
on the evaluation of the carcinogenic risk of chemicals to humans: aldrin.
Vol. 5. Lyon, France: IARC, pp. 25-38.
International Agency for Research on Cancer (IARC). (1974b) IARC monographs
on the evaluation of the carcinogenic risk of chemicals to humans:
dieldrin. Vol. 5. Lyon, France: IARC, pp. 125-156.
International Agency for Research on Cancer (IARC). (1974c) IARC monographs
on the evaluation of the carcinogenic risk of chemicals to humans: endrin.
Vol. 5. Lyon, France: IARC, pp. 157-171.
International Agency for Research on Cancer (IARC). (1979a) IARC monographs
on the evaluation of the carcinogenic risk of chemicals to humans:
chlordane. Vol. 20. Lyon, France: IARC, pp. 45-65.
International Agency for Research on Cancer (IARC). (1979b) IARC monographs
on the evaluation of the carcinogenic risk of chemicals to humans:
heptachlor and heptachlor epoxide. Vol. 20. Lyon, France: IARC,
pp. 129-154.
6-5
-------
Jager, K.W. (1970) Aldrin, dieldrin, endrin and telodrin: an epidemiological
and toxicological study of long-term occupational exposure. New York:
Elsevier Pub. Co., pp. 121-131.
Keane, W.T.; Zavon, M.R. (1969) The validity of a critical blood level for
the prevention of dieldrin intoxication. Arch. Environ. Health 19(1)36-44.
Klein, A.K.; Link, J.D. and Ives. N.F. (1968) Isolation and purification of
metabolites found in the urine of male rats fed aldrin and dieldrin. J.
Assoc. Anal. Chem. 51:895-898.
Klevay, L.M. (1970) Dieldrin excretion by the isolated perfused rat liver: a
sexual difference. Toxicol. Appl . Pharmacol . 17:813-815.
Korte, F.; Arent, H. (1965) Metabolism of insecticides. IX(I). Isolation
and identification of dieldrin metabolites from urine of rabbits after oral
administiMUion of .iieldrin-l^c. Life Sci . 4:2017-2026.
Lay, O.R.; Malik, O.K.; Klein, W.; Korte, F. (1982) Effects of dosing and
routes of administration on excretion and metabolism of carbon-14-labeled
dieldrin in rats. Chemosphere 11:1231-1242.
Levi, P.E.; Hodgson, E. (1985) Oxidation of pesticides by purified cytochrome
P-450 isozymes from mouse liver. Toxicol. Lett. 24:221-228.
Ludwig, W.; Weis, J.; Korte, F. (1964) Excretion and distribution of
and its metabolites after oral administration for a long period of time.
Life Sci. 3:123-130.
MacDonald, W.E.; Anderson, W.A.D.; Bevilacqua, M.; Blum, E.; Deichmann, W.B.
(1972) The tumorigenici ty of dieldrin in the Swiss-Webster mouse (as
cited in Epstein, 1975).
MacKay, D.; Wolkoff, A.M. (1973) Rate of evaporation of low-solubility con-
taminants from water bodies to atmosphere. Environ. Sci. Technol . 7:611.
Matthews, H.B.; Matsumura, F. (1969) Metabolic fate of dieldrin in the rat.
J. Agric. Food Chem. 17:845-852.
Matthews, H.B.; McKinney, J.D. (1974) Dieldrin metabolism to cis-dihydro-
aldrindiol and epimerization of cis- to trans-dihydroaldrindiol by rat
liver microsomes. Drug Metab. Dispos. 2:333-340.
Matthews, H.B.; McKinney, J.D.; Lucier, G.W. (1971) Dieldrin metabolism,
excretion and storage in male and female rats. J. Agric. Food Chem.
19:1244-1248.
McMannus, M.E.; Boobis, A.R.; Minchin, R.F.; Schwartz, D.M.; Murray, S.;
Davies, D.S.; Thorgeirsson, S.S. (1984) Relationship between oxidative
metabolism of 2-acetylaminofluorine, dibrisoquine, and aldrin in human
liver microsomes. Cancer Res. 44:5692-5697.
6-6
-------
Mehendale, H.M.; El-Bassiouni i-E'vAs (1975) ^Uptake and disposition of aldrin
and dieldrin by isolated perfused rabbit lung. Drug Metab. Dispos. 3:543-
556.
Meierhenry, E.F.; Reuber, B.H.; Gershwin, M.E.; Hsieh, L.S.; French, S.W.
(1983) Dieldrin-induced mallory bodies in hepatic tumors of mice of dif-
ferent strains. Hepatology 3:90-95.
Moriarty, F. (1975) Organochlorine insecticides: persistent organic pollu-
tants. New York: Academic Press, pp. 29-72.
Muller, W.; Nohynek, G.; Woods, G.: Korte, F.; Coulston, F. (1975) Compara-
tive metabolism of dieldrin-l^C in mouse, rat, rabbit, rhesus monkey and
chimpanzee. Chemosphere 2:89-92.
Muller, W.; Nohynek. G.; Korte, F.; Coulston, F. (1979) [Absorption, body
distribution, metabolism and excretion of dieldrin in nonhuman primates
and other laboratory animals.] Z. Naturforsch May-June 34C(5-6):340-345.
National Cancer Institute (NCI). (1978a) Bioassay of aldrin and dieldrin for
possible carcinogenicity. DHEW Publication No. (NIH) 78-821. Carcino-
genesis Technical Report Series No. 21.
National Cancer Institute (NCI). (1978b) Bioassay of aldrin and dieldrin for
possible carcinogenicity. DHEW Publication No. (NIH) 78-822. Carcino-
genesis Technical Report Series No. 22.
National Toxicology Program (NTP). (1985a) Report of the Ad Hoc Panel on
Chemical Carcinogenesis Testing and Evaluation. Research Triangle Park,
NC, p. 280.
National Toxicology Program (NTP). (1985b) Bioassay of chlorendic acid for
possible carcinogenicity. DHEW Publication No. (NIH) 85-2560. Carcino-
genesis Technical Report Series No. 304.
Ottolenghi, A.D.; Haseman, J.K.; Suggs, F. (1974) Teratogenic effects of
aldrin, dieldrin and endrin in hamsters and mice. Teratology 9:11-16.
Ozoukwa, M.; Sleight, S.D. (1972) Dieldrin toxicosis: fetotoxicosis, tissue
concentrations and microscopic and ultrascopic changes in guinea pigs.
Am. J. Vet. Res. 33:579-583.
Princi, F. (1954) Toxicity of the chlorinated hydrocarbon insecticides. In:
Atti dell'xi Cong. Int. Med. Labor, pp. 253-272.
Quaife, M.L.; Winbush, J.S.; Fitzhugh, O.G. (1967) Survey of quantitative
relationships between ingestion and storage of aldrin and dieldrin in
animals and man. Food Cosmet. Toxicol. 5:39-50.
Registry of toxic effects of chemical substances (RTECS). (1985) Data base.
Cincinnati, OH: National Institute for Occupational Safety and Health.
6-7
-------
Reuben, M.D. (1974) Exhibit 42, Testimony at hearings on aldrin/dieldrin.
(as cited in Epstein, 1975)
Reuber, M.D. (1977) Hepatic vein thrombosis in mice ingesting chlorinated
hydrocarbons. Arch. Toxicol. 38:163-168.
Reuber, M.D. (1980) Significance of acute and chronic renal disease in
Osborne-Mendel rats ingesting dieldrin or aldrin. Clin. Toxicol. 17:159-
170.
Robinson, J.; Roberts, M.; Baldwin, M.; Walker, A.I.T. (1969) The pharmaco-
kinetics of HEOD (dieldrin) in the rat. Food Cosmet. Toxicol. 7:317-332.
Shakoori, A.R.; Rasul, Y.6.; Syed, S.A. (1982) Effect of dieldrin feeding for
six months on albino rats. Biochemical and histological changes in liver.
Pakistan J. Zoo!. 14:191-204.
Singh, K.K.; Oha, 6.0. (1982) Bone growth during chronic aldrin intoxication
in goats. Res. Vet. Sci. 32:282-288.
Song, 0.; Harville, W.E. (1964) Carcinogenic!ty of aldrin and dieldrin in
mouse and rat liver. Fed. Proc. Fed. Am. Soc. Exp. Biol. 23:336.
Soto, A.R.; Deichmann, W.B. (1967) Major metabolism and acute toxicity of
aldrin, dieldrin, and endrin. Environ. Res. 1:307-322.
Spiotti, E.O. (1951) Aldrin poisoning in man. Arch. Ind. Hyg. Occup. Med.
4:560-566.
Stacey, C.I.; Tatum, T. (1985) House treatment with organochlorine pesti-
cides and their levels in human milk - Perth, Western Australia. Contam.
Toxicol. 35(2):202-208. Bull. Environ. Health 40(2):102-108.
Stevenson, D.E.; Thorpe, E.; Hunt, P.F.; Walker, A.I.T. (1976) The toxic
effects of dieldrin in rats: re-evaluation of data obtained in a two-year
feeding study. Toxicol. Appl. Pharmacol. 36:247-254.
Sundaram, K.S.; Damodaran, V.N.; Venkitasubramanian, T.A. (1978a) Absorption
of dieldrin through monkey and dog skin. Indian J. Exp. Biol. 16:101-103.
Sundaram, K.S.; Damodaran, V.N.; Venkitasubramanian, T.A. (1978b) Absorption
of dieldrin through skin. Indian J. Exp. Biol. 16:1004-1007.
Tanaka, R.; Fujisawa, S.; Nakai, K. (1981) Study on the absorption and pro-
tein binding of carbaryl, dieldrin and paraquat in rats on a protein diet.
J. Toxicol. Sci. 6:1-11.
Tanaka, R.; Fujisawa, S.; Nakai, K.; Minagawa. K. (1980) Distribution and
biliary excretion of carbaryl, dieldrin and paraquat in rats: effects on
diets. J. Toxicol. Sci. 5:151-162.
6-8
-------
Tennekes, H.A.; Wright, A.S.; Dix, K.M. (1979) The effects of dieldrin, diet,
and other environmental components on enzyme function and tumor incidence
in livers of CF-1 mice. Arch. Toxicot.^2:197-212.
Tennekes, H.A.; Wright, A.S.; Dix, K.M,; Koeman, J.H. (1981) Effects of diel-
drin, diet, and bedding on enzyme function and tumor incidence in livers
of male CF-1 mice. Cancer Res. 41:3615-3620.
Tennekes, H.A.; Edler, L.; Kunz, H.W. (1982) Dose-response analysis of the
enhancement of liver tumor formation in CF-1 mice by dieldrin. Carcino-
genesis 3:941-945.
Thorpe, E.; Walker, A.I.T. (1973) The toxicology of dieldrin (HEOD). Part II.
Comparative long-term oral toxicology studies in mice with dieldrin, DDT,
phenobarbitone, beta-BHC and gamma-BHC. Food Cosmet. Toxicol. 11:433-441.
Treon, J.F. and Cleveland, F.P. (1955) Toxicity of certain chlorinated hydro-
carbon insecticides for laboratory animals, with special reference to
aldrin and dieldrin. Agric. Food Chem. 3:402-408.
U.S. Environmental Protection Agency (U.S. EPA). (1976) National interim
primary drinking water regulations. EPA-570/9-76-003. ' t
U.S. Environmental Protection Agency (U.S. EPA). (1980) Ambient water qual-
ity criteria document for aldrin/dieldrin. EPA-440/5-80-019. NTIS PB81-
117301. Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH.
U.S. Environmental Protection Agency (U.S. EPA). (1984) Health assessment
document for hexachlorocyclopentadiene. EPA-600/8-84-001F. NTIS PB85-
124915. Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH.
U.S. Environmental Protection Agency (U.S. EPA). (1985) National Human
Adipose Tissue Survey. Data base.
U.S. Environmental Protection Agency (U.S. EPA). (1986) Guidelines for
carcinogen risk assessment. Federal Register 51(185):33992-34003.
Van Raalte, H.G.S. (1977) Human experience with dieldrin in perspective.
Ecotox. Environ. Safety 1:203.
Versteeg, J.P.J.; Jager, K.W. (1973) Long-term occupational exposure to the
insecticides aldrin, dieldrin, endrin, and telodrin. Br. J. Ind. Med.
30:201.
Walker, A.I.T.; Stevenson, D.E.; Robinson, J.; Thorpe, E.; Roberts, M, (1969)
The toxicology and pharmacodynamics of dieldrin (HEOD): Two-year oral
exposures of rats and dogs. Toxicol. Appl. Pharmacol. 15:345-373.
Walker, A.I.T.; Thorpe, E.; Stevenson, D.E. (1972) The toxicology of dieldrin
(HEOD). Long-term oral toxicity studies in mice. Food Cosmet. Toxicol.
11:415-432.
6-9
-------
Weaver, L., et al. (1965) Chlorinated hydrocarbon pesticides in major U.S.
river basins. Public Health Rep. 80:481.
Winteringham, F.P.W.; Barnes, J.M. (1955) Comparative response of insects and
mammals to certain halogenated hydrocarbons used as insecticides. Physiol.
Rev. 35:701-739.
Wolff, T.; Grain, H.; Huang, M-T.; Miwa, G.T.; Lu, Y.H. (1980) Aldrin epox-
idation catalyzed by purified rat-liver cytochromes P-450 and P-448. Eur.
0. Biochem. 111:545-551.
Wright, A.S.; Donninger, C.; Greenland, R.D.; Stemmer, K.L.; Zavon, M.R. (1978)
The effects of prolonged ingestion of dieldrin on the livers of male
rhesus monkeys. Ecotoxicol. Environ. Safety 1:477-502.
Wurster, C.F. (1971) Aldrin and dieldrin. Environment 13:33.
Zavon, M.R.; Stemmer, K.L. (1975) The effect of dieldrin ingestion in rhesus
monkeys: a six-year study. University of Cincinnati, Kettering Labora-
tory, Cincinnati, OH. Unpublished.
6-10
-------
APPENDIX A
EPA: 68-02-4225
OYNAMAC No. 1-77-A4
June 25, 1986
DATA EVALUATION RECORD
TERMITICIDES—ALORIN
Mutagenicity— Multiple Genetic Toxicology Studies
STUDY IDENTIFICATION: Mutagenicity—Multiple genetic toxicology studies.
APPROVED BY:
I. Cecil Felkner, Ph.D. Signature: .L, (^JL J]JJ*
Department Manager
Dynamac Corporation Date: _ (r. - 15
A-l
-------
1. CHEMICAL: Aldrin.
2. TEST MATERIAL; N/A.
3. STUDY/ACTION TYPE: Mutagenicity-multiple genetic toxicology studies
4. STUDY IDENTIFICATION; N/A.
5. REVIEWED BY;
Brenda Worthy, M.T.
Principal Reviewer
Dynamac Corporation
Signature:
Date:
Nancy E. McCarroll, B.S.
Independent Reviewer
Dynamac Corporation
Signature:
Date:
A
6. APPROVED BY;
I. Cecil Felkner, Ph.D.
Genetic Toxicology
Technical Quality Control
Dynamac Corporation
Irving Mauer, Ph.D.
EPA Reviewer
EPA Section Head
Jane E. Harris, Ph.D.
Signature:
Date :
Signature:
Date: _
c, -
Signature:
Date : _
7" 3-
A-2
-------
I. INTRODUCTION
The purpose of this document Is 'to summarize from published articles
the review findings of in vitro and in vivo genetic toxicology
assays conducted with aldrin. Findings which were evaluated as part
of a screening assay and reported limited primary data are presented
in Table 1. Although survey findings had insufficient data to
evaluate assay acceptability, they offer useful information in the
overall genetic toxicology review of aldrin.
Table 2 contains the review findings of published primary data with
sufficient information to make an assessment of their acceptability.
The results of published articles reviewed on aldrin are discussed
according to genetic effect categories; the data included in the
survey studies consisted of 7 gene mutation assays (Category 1), 1
chromosomal aberration assay (Category 2), and 2 assays that
assessed other mutagenic mechanisms (Category 3); the data from.
primary studies consisted of 3 chromosomal aberration assays
(Category 2) and 5 assays in Category 3.
II. SURVEY STUDIES;
A. Summary:
1. Gene Mutation (Category 1):
Aldrin was reported negative (Article No. 1) using the plate
incorporation method with five strains of Salmonella
tvphimurium (TA1535, TA1537, TA1538, TA98, and TA100) in the
presence or absence of rat S9 activation, at doses up to
5000 v»g/ml. Negative results were reported for the
concentration agar gradient assay (Article No. 2) using S.
tvphimurium 646, C3076, 03052, TA1535, TA1537, TA1538, TA98,
and TA100 with or without rat S9 activation, over a 1-10,000
fold concentration range. Nonactivated aldrin was also
negative for reverse mutation (omitted from list of positive
compounds) at 20 yg/disc in S. tvphimurium strains TA1535,
TA1536, TA1537, and TA1538 (Article No. 3).
Aldrin was investigated in four survey Escherichia coli
tryptophan reversions assays. Nonactivated and 59-activated
aldrin, assayed up to 5000 yg/plate, did not increase
tryptophan reversion in E. coll WP2 her (Article No. 1).
Qualitative negative results were reported over a 1-10,000-
fold nonactivated and S9-activated concentration range of
aldrin in the concentration agar gradient test with E.. coli
WP2 and WP2 uvrA (Article No. 2). In two nonactivated spot
tests (Article Nos 3 and 4), 20 wg and 1 mg/disc aldrin
were negative in £. coli WP2 and WP2 her or E.. coli WP2,
respectively.
A-3
-------
TABLE 1. SUMMARY Of HUTAGEHICITY TESTING OF A10RIH
Genetic
Article/Authors/Study Identification Endpolnt
1 Further inutagemlcity studies on Gene
pesticides in bacterial reversion Nutation
assay systeras/Boriya, H., Ohta,
T., Watanabe, K., Hlyazawa, T..
Kato. K., Shlrasu, Y./Hutat. Res.
116: 185-216, 1983.
2 Chemically Induced unscheduled DMA Gene
synthesis in primary rat hepatocyte nutation
cultures: A comparison with bac-
terial rnutagenkity using 218 com-
pounds/Probst, G. S., HcHahon, R.
E., Hill. L. E.» Thompson, C. Z.,
Epp, J. K., Seal, S. B. /Environ.
Hutagen. 3: 11-22, 1981.
3 Hutagenidty screening of pesti- Gene
cides in the microbial system/ Mutation
Shlrasu, Y., Hoiriya, H.. Kato, K.,
Furuhashi, A., Kada. T./Hutat.
Res. 40: 19-3Qi, 1976.
DNA damage
Indicator
Organism Purity
S. typhi- Not
Buriu* reported
TA100, TA98.
TA1535,
TA1537. and
TA1538
E. coli
MP2 her
S. typhi- Not
rauriwa reported
646, C3076,
D3052.
TA1535,
TA1537,
TA1538,
TA98, and
TA100
E. coli
MP2 and
HP2 uvrA~
S. typhi- Not
raurium reported
TA1535,
TA1536,
TA1537,
TA1538
E. coli
WP2 and
WP2 her
B. subtil is
H17(Rec+)
and H45
(fier)
Application
Salmonella Hicrosome
Plate Assay (Anes
Test) and E. coli
HP2 her tryptophan
reversion assay
Concentration agar
gradient plate; +/-
S9 activation
Disc Test Cells were
exposed for 2 days
at 37 C -S9 activa-
tion
Disc Test Cells were
exposed for 24 hr at
37 C -S9 activaton
Dose
Range Response Conclusion
Tested Negative Negative
up to
5000»g/
plate
unless
cytotoxlc;
+/- S9
activation
1- Negative Negative
10,000 not In-
fold eluded in
cone. tables
range showing
positive
responses
20 |ig/ Negative Negative
disc by omis-
sion from
list of
positive
results
20 |ig/ Negative Negative
disc • by omis-
sion from
list of
positive
results
Cowroent
Preliminary screening
assay (228 pesticides)
Preliminary screening
assay for 218 com-
pounds of various
chemical classes
Preliminary screen-
ing assay of 166
pesticides
Preliminary screening
assay of 166 pesti-
cides
-------
TABLE 1. SUMMARY OF HUTAGENICIIY TESTING OF ALORIN (Continued)
Art 1 c 1 e/Authors/Study
Genetic
Identification Endpolnt
4 Hutagenlcity of Dichlorvos/ Gene
As hwood -Smith, H. J.
Ring, R. /Nature 240:
5 Hutagenlcity testing
, Trevino, J.. Mutation
418-420. 1972.
of herbicides, Chromo-
fungicides and Insecticides. I. somal
Chromosome aberrations In V1c1a aberra-
faba/Njagl, G. D. E.
N. B./Cytologla 46:
>
i
in
6 Is. vitro breakage of
DNA by mutagens and
Griffin. III. D. E.,
W. E./Hutat. Res. 52
1978.
, Gopalan, H. tlons
169-172, 1981.
plasraid DNA damage
pesticides/
H111.
: 161-169,
Indicator
Organism
E coli
WP2
Root tip
cells of
Vlcla faba
E. cp_H .
collclno-
genlc
plasmld
El
(CoLEl)
Purity Application
Not Disc Test Cells were
reported exposed for 48 hr at
37 C -S9 activation
40* The roots of Intact
seedlings were treated
for ~ 5 hours.
Cells were collected
every 30 minutes
during treatment
Not [He3H] thyraine
reported labelled cells were
exposed for an un-
specified time, lay-
ered on a alkaline
sucrose gradient on
which fraction analy-
sis was made to assess
DNA strand breakage.
-S9 activation
Dose
Range
10% of
active
Ingre-
dient
~ Img
10. 50.
100,
1000,
5000.
and
10,000
ppm
1 mg/mL
Response
Negative
Positive
for:
mitotic
Inhibi-
tion,
pynosls,
premature
chromo-
some con-
densation,
plasmoly-
sls.
Reported
to be dose
related
Negative
Conclusion
Negative
Results
suggest
severe
cytotox-
1city; but
no assessment
for chromo-
somal aberra-
tions
Negative
Comment
Preliminary screening
assay of 15 pesticides
Results are from
anaphase cells. This
study was intended
as a screening assay
of 12 pesticides
Preliminary screening
assay for development
of a new procedure
-------
TABLE 2. SUMMARY OF HUTAGEHIC1TY TESTING OF ALOR1H
Genetic Indicator Dose
Article/Authors/Study Identification Endpolnt Organism Purity Application Range
7 The comparative cytogenetic ef- Chromosomal Human perl- Not Cells cultured for 19.125,
fects of aldrln and phosphamldon/ aberration pheral blood reported 50 hrs were exposed 38.25,
Georgian, l./Hutat. Res. 31: ' for 22 hrs. 76.5
103-108. 1975. tig/ml
Chromosomal Rat and mouse 5 rats and 5 nice 9.56,
aberration bone marrow received one 1p dose; 19.125,
sacrificed 24 hr 38.25,
after dosing 76.50
K3/9
Za
*
Response Conclusion
Positive Inconclusive
at lowest evidence for
dose; a positive
however, response
dose was
cytotoxic
(19.125)
iig/mt
All cells
died at
higher
doses
100 and Inconclusive
50X mor- evidence for
tality a positive
at 76.5 response
and 38.25
vg/g;
positive
at 38.25
and 19.125
iig/roL
Negative
at 9.56
vg/mL
Comment
Aberrations observed
near limit of cell
survival; gaps
included in
X aberrant cells;
no. of aberrations/
cell slightly lower
than Mltomydn C;
slides not coded
Aberrations observed,
near limit of animal
survival.
Gaps Included in %
aberrants.
No positive control;
slides not coded.
Comments on both
studies: 1, limited
information on
methodology (i.e.,
staining procedures)
2, presumed reactive
dose was comparable
for in vivo and in
vitro studies
3, positive effects
noted near limit of
cell and animal
survival .
-------
TABU 2. SUMMARY OF MUTAGENIC1TY TESTING OF ALDRIN (Continued)
Article/Authors/Study Identification
8 Hutagenicity studies Involving
aldrin, endosulfan, dlmethoate,
phosphamidon, carbaryl, and
ceresan/Usha Rani. H. V., Reddl, 0.
S.. Reddy, P. P. /Bull. Environ.
Contain. Toxicol. 25:277-282, 1980.
9 Detection of chemical mutagens by
the dominant lethal assay In the
mouse/Epstein, S. S., Arnold, E.,
Andrea, J.. Bass, W., Bishop, Y./
Toxicol. Appl. Pharmacol. 23: 288-
325, 1972.
Genetic
Endpoint
Chromosomal
aberrations
(mlcronu-
c lea ted
poly-
chromatic
erythrocytes)
Chromosomal
breaks in
male ger-
minal
Dominant
lethal assay
Indicator
Organism Purity
Swiss male Not
mice bone reported
marrow
Mouse sper- Not
matogenesls reported
cycle (8 week
mating sequence)
Application
Six mice received
two equal oral doses
separated by a 24 hr
Interval
A single 1p dose
(7 males low dose
9 males high dose,
1 male mated to
3 females/week)
Dose
Range Response Conclusion
13 mg/ Negative Unacceptable
kg
8 and Negative Negative
40 mg/ 5/9 or 56* (Acceptable)
kg of males
died at
40 mg/kg;
At week 5:
1 .2 Increase
in early
deaths/
female in
high dose
Comment
Only one dose assayed;
no evidence of a cyto-
toxic or toxic ef-
fects; only one sam-
pling Interval; no
positive control
assayed.
Increase was not
significant, and no
other increases were
noted throughout
study; assayed up to
a toxic dose
Five consecutive
dally administrations
by gavage (10 males/
group; 1 male
mated to 3 females/
week)
0.5 and Negative: Results are
1 mg/kg/ 20% males uninter-
day 1n 1.0 mg/ pretable
kg group
and 10% in
0.5 mg/kg
group
died;
At wk 2
In high '
dose 1.0
early deaths/
pregnancy
At wk 8:
decrease In
% pregnancy
at both doses;
1.5 increase
In early
deaths per
pregnancy and
100% pregnant
females with
early death
observed 1n 2
females at
the low dose
only
As reported by
authors, increases
were not significant
and these variances
occurred late In the
study (wk B) and only
in the low dose;
however, they did
not rule out dose
reversal
-------
TABLE 2. SUMMARY OF HUIAGEH1CI1Y 1ESTIHG OF ALORIH (Continued)
Article/Authors/Study Identification
10 Chemically Induced unscheduled DMA
synthesis in primary rat hepatocyte
cultures: A comparison with bac-
terial mutageniclty using 218 com-
pounds/Probst, G. S., HcHahon, R.
E.. Hill, L. E., Thompson, C. Z.,
Epp, J. K., Neal, S. B. /Environ.
Hutagen. 3: 11-22. 1981.
11 Effect of pesticides on scheduled
and unscheduled DNA synthesis of
rat thymocytes and human lympho-
cytes/Rocchi, P.. Perocco, P..
Alberghini, W. , F1n1. A., Prodi,
G./Arch. Toxicol. 45: 101-108,
1980.
»
0
Genetic
Endpoint
DNA
repair
UDS
Scheduled
DNA
Synthesis
(SOS)
Unsched-
uled
ONA
Synthesis
(UDS)
Indicator
Organism Purity
Primary Not
rat hepato- reported
cytes;
Unscheduled
DNA Synthesis
(UDS)
Rat thymo- 95X
cytes
Human lympho-
cytes
Human lympho-
cytes
Application
Cells were exposed
for 5 and 18-20 hr
with tritiated thy-
midlne
Dose
Range
8 doses
ranging
from
0.5 to
1 ,000
Response Conclusion
Negative Negative
(Acceptable)
Comment
The cone, scored was
100 moles/ml, the
highest noncytotoxic
dose.
nmoles/ml
Preliminary Study;
Cells were exposed
for 4 hrs with tri-
triated thymidlne
UDS Assay Cells were
exposed for 4 hr with
hydroxyurea and tri-
tiated thyroidlne
Cells were irradiated
with UV light, exposed
for 4 hr with hydro-
xyurea and tritiated
thymidlne to measure
the effects of the
chemical on DNA repair
after UV
damage
10.
100.
and
1000
iig/mL
lOOpg/
ml
100Mg/
ml
Positive N/A
Inhibition
of SDS
10, 35,
and
96X at 10.
100. and
1000 ug/mL,
respectively
Weak posi- Inconclusive
tive; evidence for
slight a genotoxic
Increase response
in UDS
Weak inter- N/A
ference
with DNA
repair
Range-finding study to
determine that dose
which Inhibited syn-
thesis by SOX;
selected dose for UDS
assay with human
lymphocytes was
100 pg/ml
Multiple doses should
have been assayed
Correlation between
interference with DNA
repair and genotox-
icity is not clear.
-------
TABLE 2. SUMMARY OF HUTAGEN1CIIY ItSI IMG OF ALUKIN (Continued)
Article/Authors/Study Identification
12 Pesticide Induced DNA damage and
its repair in cultured human cells/
Ahmed, F; E., Hart, R. W., Lewis,
N. J./Mutat. Res. 42: 161-174,
1977.
13 Evaluation of the alkaline elution/
rat hepatocyte assay as a predictor
of carcinogenlc/mutagenlc potential
Sina, J. F., Bean, C. L.,
Dysart, G. R., Taylor. V. I.,
Bradley, M. O./Mutat. Res. 113:
357-391, 1983.
to
14 Cytogenic effect of some chloror-
ganic Insecticides on mouse bone-
marrow cell nuclel/Markaryan, 0.
S./Genetlka 2(1): 132-137. 1966.
Genetic
Endpoint
DNA repair
(UDS)
DNA
damage
(DNA strand
breakage)
In vivo
spindle
apparatus
inhibition
Indicator
Organism Purity
SV-40 trans- Not
formed human reported
cells (VA-4)
Rat Not
hepatocytes reported
Anaphase Not
and early reported
telophase in
male mouse
bone marrow
cells
Application
UPS Assay Cells were
treated with trltiated
thyraldine +/- S9
activation and exposed
for 1, 3, 5, 8 and 12
hours
Cells were
exposed for
3 hours.
Single 1p
administration
to 3 male mice;
bone marrow
harvested 21 hr.
posttreatment
Dose
Range
1. 10.
100
and
1000 |iH
0.03,
0.3
and
3mH
0.002
mg/g
Response
Positive
at all
doses
+/- S9
activation
(quali-
tative
results)
Reported
positive
at cyto-
toxlc
doses
Positive;
Increase
in
nuclear
distur-
bances;
marked
increase
in chromo-
some and
chromatid
bridges and
fragments
Conclusion
Inconclusive
presumptive
positive
Inconclusive
evidence for
a positive
response
Unacceptable
technically
deficient
Comment
; The use of a trans-
formed cell line is
not recommended
The reporting of
qualitative results
prohibits an accurate
assessment of the
data; assay is also
technically flawed.
Noncarcinogens that
gave results similar
to aldrin (-1- at
cytotoxlc doses) were
classified as "false
positives" by the
authors
; Dataware unlnter-
pretable due to
. staining protocol
(probably high
incidence of
artifacts); unable
to determine if
Increases in nuclear
disturbances Include
severe cytotoxlc signs
observed at 4X of
LD50 (50 mg/kg)
-------
2. Chromosomal Aberrations (Category 2):
Meristematic root cells of Vicia faba were treated with 8
doses of aldrin ranging from 10 to 10,000 ppm (Article
No. 5). Root samples were collected at varying times up to
310 minutes, stained, and scored for chromosomal aberrations
and cytological effects. The results from anaphase cells
collected 120 minutes following exposure to 100 ppm showed
increased mitotic inhibition, formation of pynotic nuclei,
premature chromosome condensation, and plasmolysis. The
authors stated that the cytotoxic effects were dose and time
related; however, there was no apparent increase in
chromosomal aberrations. We assess that aldrin was severely
cytotoxic but not clastogenic in this study.
Other Mutagenic Mechanisms (Category 3):
Nonactivated aldrin, at 20 yg/disc, did not cause
preferential inhibition of Bacillus subtilis repair-
deficient H45, compared to the repair-competent HI7 strain
(Article No. 3).
One mg/mL aldrin did not increase the number of plasmid
breaks in E.. coli CoL El when compared to the control
(Article No. 6).
B. Conclusion - Survey Data;
There was insufficient data presented in the survey studies to
draw unequivocal conclusions about the genotoxic responses to
aldrin. However, the results suggest that aldrin did not cause
gene mutations or DNA damage in bacteria, or induce chromosomal
aberrations in plants.
III. STUDIES WITH PRIMARY DATA:
1. Gene Mutation (Category 1);
No studies were submitted for this category; therefore, a sig-
nificant data gap exists.
2. Chromosomal Aberrations (Category 2);
a. Mammalian Cells; Georgian (Article No. 7) exposed synchro-
nized human peripheral blood cells to 19.125, 38.25 and 76.5
vg/mL aldrin. No cells survived the 22-hour treatment
with 38.25 and 76.5 yg/mL. At the lowest dose assayed,
which was reported as cytotoxic, there was a 20% increase in
chromosomal aberrations per cell as compared to the negative
control. No cells survived the two higher concentrations,
and a mammalian metabolic activation system was not employed
in this study.
A-10
-------
Our reviewers noted, however, that gaps were included in the
scored aberrations, and that only 84 cells were counted for
the 19.125 vg/ml_ dose group, whereas 273 cells each were
scored for the negative and positive (Mitomycin C, 1 yg/mL)
controls, respectively. The authors stated that the increase
in aberrations occurred near the limit of cell survival, and
that the observed chromosomal lesions are probably not
perpetuated in other abnormal cells. We assess that the
inclusion of gaps in the total aberration counts and the
increase in aberrant cells near the limit of cell survival
are insufficient evidence for a clastogenic response. The
study is, therefore, inconclusive for a positive response.
Georgian (Article No. 7), also administered single ip
injections of 9.56, 19.125, 38.25, and 76.50 ug/g aldrin to
five rats and mice (sex not specified) per dose; bone marrow
cells were collected 24 hours later. All mice and rats
receiving 76.50 vg/g and 50% of the 38.25 vg/g dose
groups died. Although increased chromosomal aberrations
were reported in the 38.25 pg/g group, only 26 cells were
scored for surviving rats and 30 for surviving mice.
Approximately 7- and 3-fold increases in chromosomal
aberrations per cell were reported for rats and mice,
respectively, exposed to 19.125 ug/g aldrin. The results
for the lowest dose (9.56 vg/g) were not reported; however,
the authors indicated that no chromosomal aberrations were
found. Hence, there were no quantitative data available to
be evaluated at the lowest dose. As noted for the in vitro
cytogenetic assay performed by Georgian, gaps were included
in the. total number of aberrations. The author formed the
same conclusions about chromosomal aberrations near the
limit of survival. We assess that these data are deficient
in some details, and are insufficient for concluding that
aldrin is clastogenic in rodent bone marrow cells. The
study is, therefore, inconclusive for a positive response.
Usha et al. (Article No. 8) performed a micronucleus assay
in mouse bone marrow using aldrin. Six male mice received
two equal oral administrations of aldrin (13 mg/kg total)
separated by 24-hour intervals. Six hours after the second
administration, animals were sacrificed by cervical
dislocation and bone marrow cells were stained. Two thousand
polychromatic erythrocytes (PCE) were scored for micronuclei
and the ratio of PCE to normochromatic erythrocytes (NCE)
was calculated. No toxic signs were reported and the ratio
of PCErNCE was comparable to the control. No significant
increase in the frequency of micronuclei was observed. We
assess that the assay was technically flawed because only one
dose was evaluated, no evidence of a cytotoxic or clinically
toxic effect was observed, at the single sampling interval,
and a positive control was not evaluated. We concluded that
the test material was not adequately investigated; therefore,
the study is unacceptable.
A-ll
-------
b. Germinal Cells: Epstein et al. (Article No. 9) investigated
the clastogenic effects of 172 chemicals in mouse dominant
lethal assays. Aldrin was administered to 7 or 9 male mice
using a single ip dose of 8 or 40 mg/kg, respectively; in a
second experiment, 10 male mice/group received an oral daily
dose of 0.5 or 1 mg/kg/day for 5 consecutive days. The
concurrent negative control group consisted of 10 males.
For both experiments, male mice were mated 1 male to 3
females for 8 weeks. In the acute assay 56% (5/9) of the
males died at the highest dose (40 mg/kg). A nonsignificant
increase (p >0.05) in early deaths/female was recorded at
week 5; no other increases in measured parameters were
noted; therefore, the study is acceptable.
In the subacute assay, 10% (1/10) and 20% (2/10) of the males
died in the 0.5 and 1.0 mg/kg dose groups, respectively.
The following nonsignificant changes were observed: at week
2, an increase in early deaths/pregnancy at the high dose;
at week 8, a decrease in percent pregnancy (both doses) and
increases in early fetal deaths and percent pregnant females
with early deaths. The authors stated that since the
changes in dominant lethal parameters were not significant
and were generally noted at the low dose, the results did
not suggest a positive effect; they did not, however, rule
out the possibility of dose reversal. We concluded this
assay cannot be properly interpreted.
c. Conclusion for Chromosomal Aberrations; No conclusion can be
drawn from these in vitro and in. vivo somatic cytogenetic
assays with aldrin; therefore, a major data gap exists.
Based on the dominant lethal studies (Article No. 9) aldrin
was not clastogenic in the acute assay; however, the results
of the subacute assay cannot be properly interpreted.
3. Other Hutagenic Mechanisms (Category 3);
a. DNA Repair/Damage; Probst et al. (Article No. 10) tested 8
aldrin doses ranging from 0.5 to 1000 nM/mL in an in vitro
unscheduled DNA-synthesis (UDS) assay using primary rat
hepatocytes. Aldrin did not cause an increase in UDS at the
highest noncytotoxic dose (100 nM/mL). Although 100 nM/mL
was the only dose scored in this article, the study was
properly controlled, and is acceptable.
Rocchi et al. (Article No. 11) tested aldrin's action on
semi conservative DNA synthesis (SDS) to determine the 50%
inhibition dose with rat thymocytes. They assayed aldrin
for its potential to induce unscheduled DNA synthesis (UDS)
in human lymphocytes in the presence of hydroxyurea and
A-12
-------
tritiated thymidine without metabolic activation. In
addition, they tested aldrin's ability to interfere with
human lymphocyte repair' capability following damage exerted
by ultraviolet (UV) rays.
Based on the results of the inhibition of thymidine uptake
assay, aldrin caused a 50% inhibition in rat thymocytes at a
dose of 100 yg/mL; this dose was used in the UOS assay
with human cells. Induced DOS repair effects were assessed
after calculations were made to eliminate the effect of SDS
inhibition. Aldrin induced an increase in DOS over the
control. Similarly, with human cells damaged by UV irradia-
tion and dosed with aldrin (100 yg/ml), the chemical
slightly interfered with UDS repair. We assess that since
the observed effects were noted at only one dose, the
evidence for a genotoxic response is inconclusive. There-
fore, the assay should be repeated with multiple doses in
the presence and absence of metabolic activation.
Ahmed et al. (Article No. 12) evaluated aldrin at doses of
1, 10, 100 and 1000 yM with or without S9 activation in a
UDS assay with SV-40 human (simian virus) transformed cells
(VA-4). The authors reported only qualitative positive
responses at all nonactivated and S9-activated doses fol-
lowing the 8-hour treatment. We assess that the reporting
of qualitative results prohibits an accurate assessment of
any dose-related response. In addition, the cells used in
this assay were not grown under reduced serum or arginine
deprivation conditions in conjunction with hydroxyurea to
insure blockage of S-phase synthesis. Several agents which
are generally considered nongenotoxic or weakly genotoxic
(e.g., Oiquat, 2,4-0 fluid, and carbaryl) induced strong
responses in this study. We discussed the findings of this
assay with Dr. Gregory Probst,1 Lilly Research Labs. He
concurred with our assessment and noted that cells which may
have escaped hydroxyurea treatment could easily be misinter-
preted as undergoing UDS synthesis, thereby increasing the
chances of a false positive.
In addition, the use of a transformed cell line to assess
primary DMA damage does not conform with recommended3
procedures, which use non-transformed diploid human cells.
Probst, G. S., Lilly Research Labs., Greenfield, IN, personal communica-
tions 4/86.
t
Mitchell, A. D., Casciano, D. A., Meltz, M. L., Robinson. D. £., San, R.,
H. C., Williams, G. M., and Von Halle, E. S. Unscheduled DNA synthesis
tests, a report of the U.S. Environmental Protection Agency Gene-Tox
Program. Mutat. Res. 123(1983): 363-410.
A-13
-------
We classify aldrin as an inconclusive presumptive genotoxih
for SV-40 transformed human cells. We recommend confirma-
tion of these data in a rigorously controlled experiment
that conforms to the currently recommended protocols for UDS
assays with human cell lines.
Sina et al. (Article No. 13) evaluated the -DNA-damaging
activity of 91 chemicals, which included aldrin, in a
ONA-strand breakage assay using primarily rat hepatocytes.
Cells were exposed to three aldrin doses of 0.03, 0.3, and
3 mM for 3 hours. Cytotoxicity was measured either by
glutamate-oxaloacetate transaminase release or by trypan
blue exclusion. DMA strand breaks were analyzed by alkaline
elution, and the ONA content was determined fluorimetrically.
In contrast to the authors' criteria for a positive response,
(3-fold increase in elution rates at <30% toxicity), aldrin
was listed as positive. The authors reported a dose-related
increase in single strand breaks (SSB) at the two highest
doses; however, no activity was seen at the low dose. Based
on the authors' statement that trypan blue exclusion proved
to be a more reliable indicator of cell viability and a >30%
reduction in cell viability was indicative of a biologically
significant cytotoxic effect, we concluded from the data
that the two active doses were cytotoxic (4 and 62% viability
at 3 and 0.3 mM, respectively). Interestingly, several
nonmutagens/noncarcinogens (cupric sulfate, phenanthrene,
toluene, and 2-nitrobiphenyl) elicited responses similar to
aldrin but were classified as "false positives." The
authors stated that "Each of these compounds, however,
caused SSB only when cytotoxicity was greater than 30%. DNA
damage produced by these compounds may, therefore, be
secondary to cell lysis."
We assess that the results for aldrin did not provide
convincing evidence of a genotoxic effect.
b. Inhibition of Spindle Apparatus: Harkaryan (Article No. 14)
studied the effects of aldrin on the bone marrow of male
mice. He reported that anaphase and early telophase
harvested cells in mice injected ip with 0.002 mg/g showed
increased frequencies of "nuclear disturbances," total
chromosome rearrangements (chromosome and chromatid bridges
and fragments) and "adhesions." We discussed the findings
of this study with Dr. James Ivett,3 and concur with his
recommendation that the study is unacceptable for the
following reasons:
1. The method used to prepare anaphase and telophase plates
(slices of bone marrow stained with acetocarmine) tends
3
Ivett, 0. L., Hazleton Biotechnologies, Vienna, VA (personal communica-
tions, 4/86.
A-14
-------
to increase the number of artifacts, and lends itself to
a planer effect (causes chromosomes to be excluded from
a field being viewed); this makes the distinction between
true chromosomal rearrangements and artifact difficult.
2. The number of animals investigated (three male mice) is
probably not an inadequate sample population.-
3. The inclusion of cytotoxic effects (vacuolation,
karyopycnosis, and karyolysis) in the total percent
nuclear disturbances suggests that the assayed dose,
which was approximately 4% of the LD5Q, was severely
cytotoxic.
4. The slides were not reported as being coded.
We concluded, therefore, that the results could not be
interpreted, and that these findings should be excluded from
the genetic assessment of aldrin,
c. Conclusions for Other Genetic Mechanisms:
Only one published article (No. 10) in the category provided
convincing evidence that aldrin is not genotoxic; this was
the UDS assay using primary rat hepatocytes.
Based on the inconclusive and unconfirmed findings of a UDS
assay with transformed human cells, aldrin is listed, with
reservations, as presumptively genotoxic in human cells. We
caution, however, that these findings are suspect and should
be confirmed. The additional UDS assay with human lympho-
cytes or the DNA-strand break assay with primary rat
hepatocytes were equivocal and provided insufficient
evidence of a genotoxic response.
IV. OVERALL INTERPRETATION OF STUDY RESULTS AND RECOMMENDATIONS
1. Interpretation
The paucity of acceptable studies and the existence of major
gaps in the data (point mutation and cytogenetics) severely
hinders the development of a definitive genetic toxicology
profile for aldrin. However, since aldrin in readily converted
both in. vivo and in. vitro to dieldrin, the genetic toxicology
assessment prepared for dieldrin can be used to infer a genetic
activity profile for aldrin.
Although no published bacterial gene mutation assays with suffi-
cient primary data were reviewed, the survey studies with S.
tvphimurium and E_. coli indicate that aldrin is probably not
mutagenic in procaryotes. This finding is supported by the over-
whelming conclusion of negative mutagenic effects for dieldrin.
A-15
-------
No mammalian gene mutation assays with aldrin were located; the
single mammalian cell point mutation assay of dieldrin showed an
inconclusive, unconfirmed mutagenic response. A major data gap,
therefore, exists for both compounds.
No conclusions could be drawn from the in vitro or jm vivo
somatic cytogenetic assays with aldrin; therefore a major data
gap remains. However, aldrin was not clastogenic in an acute
dominant lethal assay in mouse germinal cells.
In agreement with both survey and acceptable DNA damage/repair
assays with dieldrin, aldrin was not genotoxic in bacteria,
yeast, or rat hepatocytes.
Nonactivated and S9-activated aldrin induced statistically
significant increases in UDS in SV-40 transformed (VA-4) human
fibroblasts. Dieldrin, also investigated in this study, caused
a similar response. However, the validity of this assay was
seriously compromised by numerous technical deficiencies, the
observation of strong positive responses by compounds generally
accepted as weakly genotoxic or nongenotoxic substances, and the
lack of positive controls.
2. Recommendations
Although no persuasive evidence of adverse genetic effects was
uncovered in the review of the available genetic toxicology
assay with aldrin, adequate tests are needed to assess the
following:
1. In. vitro and in vivo gene mutation assays.
2. In. vitro and In. vivo chromosomal aberration assays.
3. IJQ. vitro UDS assays with SV-40 transformed (VA-4) cells and
non-transformed human cells.
In conclusion, we assess that the most useful information will
come from experiments on promotion activity, since interference
with cellular communication has been shown for dieldrin and
other related chlorinated hydrocarbons such as heptachlor and
chlordane. We therefore recommend these additional tests be
performed with aldrin:
1. In vitro metabolic cooperation and/or transformation with
promotion assay.
2. In. vivo promoter activity assays.
A-16
-------
APPENDIX B .
EPA: 68-02-4225
TASK: 077-A3
June 23, 1986
DATA EVALUATION RECORD
TERMITICIQES--DIELDR1N
Mutagenicity—Multiple Genetic Toxicology Studies
STUDY IDENTIFICATION; Mutagenicity—Multiple genetic toxicology studies,
APPROVED BY:
I. Cecil Felkner, Ph.D.
Department Manager
Oynamac Corporation
Signature:
..
Date: (a ~
B-l
-------
1. CHEMICAL: Oieldrin.
2. TEST MATERIAL: N/A.
3. STUDY/ACTION TYPE; Mutagenlcity—multiple genetic toxicology studies.
4. STUDY IDENTIFICATION: N/A.
REVIEWED BY:
Brenda Worthy, M.T.
Principal Reviewer
Dynamac Corporation
Nancy E. McCarroll, B.S.
Independent Reviewer
Dynamac Corporation
Signature:
Date:
Signature:
Date:
6 -
APPROVED BY;
I. Cecil Felkner, Ph.D.
Genetic Toxicology
Technical Quality Control
Dynamac Corporation
Irving Mauer, Ph.D.
EPA Reviewer
Gane E. Harris, Ph.D.
EPA Section Head
Signature:
Date:
Signature:
Date:
Signature:
Date:
(./-2-3-U
C
B-2
-------
I. INTRODUCTION:
The purpose of this document is to summarize the review findings from
published articles of in. vitro and in vivo genetic toxicology assays
conducted with dieldrin. Articles that evaluated dieldrin as part of
a screening assay and reported limited primary data are presented in
Table 1. Although survey articles had insufficient data to evaluate
assay acceptability, they offer useful information in the overall
genetic toxicology review of dieldrin.
Table 2 contains the review findings of published articles with
sufficient data to make an assessment of assay acceptability. The
results of all published articles reviewed on dieldrin are discussed
based on categories of genetic effect according to EPA Guidelines:
survey data from 8 gene mutation assays (Category 1) and 6 assays
that assessed other mutagenic mechanisms (Category 3); primary data
from 5 gene mutation assays (Category 1), 5 chromosomal aberration
assays (Category 2), and 8 assays classified as other mechanisms
(Category 3).
II. SURVEY STUDIES (Table 1);
A. Summary;
1. Gene Mutation (Category 1):
Dieldrin was reported negative in four survey articles
(Article Nos. 1, 2, 3, and 5) which reported plate incorpora-
tion assays with strains of Salmonella typhimurium (TA1535,
TA1537, TA1538, TA98, or TA100) in the presence or absence of
rat S9 activation; doses ranged from 4 to 5000 vg/mL,
depending on solubility or toxicity. Similarly, dieldrin was
negative in two spot tests (Article No. 4 and 7) with S.
tvphimurium strains TA98 and TA100 or TA1535, TA1536, TA1537,
and TA1538 without S9 activation, at doses of 0.05 and
1: .
B-3
-------
1 mg/d1sc (Article No. 4) and at 20 yg/d1sc (Article
No. 7). The chemical was also reported negative over a
1-10,000-fold concentration range using a concentration agar
gradient method (Article No. 6) with S. typhimurium strains
646, 03052, C3076, TA1535, TA1537, TA1538, TA98, and TA100
with and without rat S9 activation.
Oieldrin was negative when assayed with Escherichia coli
strains WP2, WP2uyrA and/or WP2hcr (Article Nos. 3, 6, 7, and
8). The test chemical was assayed at levels up to 5000 yg/mL
+/- S9 activation in a plate incorporation assay (Article
No. 3); over a 1-10,000 fold dose range in the concentration
agar gradient test +/-S9 activation (Article No. 6), and
using the nonactivated spot test at 20 pg/disc and
1 mg/disc (Article Nos. 7 and 8).
2. Other Hutagenic Mechanisms (Category 3);
Oieldrin, assayed at levels up to 1000 jig/mL with or
without S9 activation, did not cause preferential inhibition
of £. coli DNA repair-deficient WP67 or CM871 in a micro-
suspensiojn assay (Article No. 5). Similarly, nonactivated
20 vg/disc dieldrin was not genotoxic in Bacillus subtilis
DNA repair-deficient M45 (Article No. 7). Oieldrin
(0.1 mg/mL) was assayed with E_. coli cells which carry
plasmid El (Article No. 9) to assess breakage in Col El
plasmid DNA. The chemical did not increase the number of
plasmid breaks relative to the negative control.
Dieldrin's ability to cause DNA damage in Chinese hamster V79
cells was assessed using the alkaline elution method (Article
No. 10). It was assayed at nine doses ranging from 0.001 to
10 mH with S9 activation, and reported as negative.
B-4
-------
The potential for dieldrin to affect DNA synthesis in
testicular cells of male mice was examined in Article
No. 11. Mice were orally administered a single dose of
50 mg/kg dieldrin. The authors reported that the chemical
inhibited DNA synthesis as assessed by a decrease in uptake
over the control in tritiated thymidine by testicular cells.
In another article inhibition of DNA synthesis by dieldrin
was also assessed in human HeLa cells but only in the
presence of S9 activation (Article No. 12). The authors
reported that dieldrin inhibited HeLa cell DNA-synthesis at
-4
4x10 M, and that after the chemical was removed, the
rate of DNA synthesis continued to decline. The authors
suggested that the continued depression of DNA synthesis
after removal of the chemical: can be used to distinguish
between DMA-damaging chemical's and those that inhibit DNA
synthesis but do not cause DNA damage. Although we agree
that the assay may distinguish DNA inhibition from DNA repair
synthesis, the lack of quantitative data for example on UDS
does not permit us to assess the validity of this assumption.
B. Conclusion:
The data presented in these survey articles were insufficient to
form unequivocal conclusions. However* the data collectively
suggest that dieldrin is not mutagenic nor does it cause DNA
damage in bacteria. Although dieldrin was shown to inhibit DNA
synthesis in mammalian cells, the distinction between cellular
toxicity and direct damage to DNA was not clearly established.
III. STUDIES WITH PRIMARY DATA (Table 2);
1. Gene Mutation (Category 1);
a. Bacteria: In an acceptable study, Haworth et al. (Article
No. 13) assayed dieldrin using a 20-minute preincubation
B-5
-------
plate Incorporation method with S. typhimurium strains
TA1535, TA1537, TA98, and TA100 at dose levels of 33, 100,
333, 1000, and 3333 pg/plate with and without rat or
hamster S9 activation. The chemical precipitated at the
three highest doses. Dieldrin did not cause an increase in
the number of histidine revertants in any of the tester
strains, with or without rat or hamster S9 activation. The
positive controls demonstrated the sensitivity of the assay
to detect a mutagenic response.
In Article No. 14, Glatt et al. assayed dieldrin with S.
typhimurium strains TA98, TA100 (+/- rat S9 activation),
TA1535, and TA1537 (-S9 activation) at six doses ranging from
10 to 3000 yg/plate. In addition, strain TA98 was dosed
with rat S9-activated dieldrin in the presence or absence of
the epoxide hydrolase inhibitor and glutathione depletor
1,1,1-trichloropropene 2,3-oxide (TCPO). Exposure was for 3
days at 37°C using the plate incorporation method. Precipita-
tion was observed at the three highest doses (300, 1000, and
3000 vg/plate). Dieldrin was negative in this study with
or without S9 activation whether TCPO was present or absent.
The positive controls demonstrated the sensitivity of the
test to detect a mutagenic response. We assess that the
study was well designed, properly controlled, and that
dieldrin was assayed up to the limit of solubility; the study
is acceptable.
Marshall et al. (Article No. 15) assayed the chemical using
S. typhimurium strains TA1535, TA1536, TA1537, and TA1538 at
doses from 50 to 1000 yg/plate, with or without rat S9
activation; cytotoxic levels were reported as >2500 vg/plate
-S9 and >1000 vg/plate +S9, but these data were not included
in the article. The authors reported that dieldrin did not
induce a mutagenic response. Although the more sensitive
strains., TA98 and TA100 were not evaluated, we assess that
the study was properly controlled and is acceptable.
B-6
-------
Majumdar et al. (Article No. 16) tested dieldrin with S>
typhimurium strains TA1535, TA98, and TA100 using the plate
incorporation method. Cells were exposed to 1, 25, and
50 yg/mL with or without S9 activation and scored at 72
hours. Cytotoxicity (decrease in the number of revertants)
was reported at the highest dose when compared to the two
lower doses. Dieldrin caused an increase in the number of
revertants in strains TA98 and TA100 either in the presence
or absence of mouse S9 activation, but only with S9 activation
in strain TA1535. The highest number of revertants was
induced at 25 yg/mL; however, 1 yg/ml also induced a
mutagenic response that was twofold or greater than the
solvent controls. We assess, however, that a repeat assay
should be performed because positive controls were not
included; the background frequency of the TA98 solvent
controls was lower and the nonactivated solvent control for
i
TA1535 was higher than those published by Ames et al.
Additionally the authors stated that the reduction in
revertants at 50 wg/mL may have been due to cytotoxicity.
It is noteworthy that the results of 10 other Ames assays
with dieldrin, reviewed here, showed that the test material
was noncytotoxic at doses up to 1000 yg/plate. The study
is, therefore, considered inconclusive, and dieldrin is
classified as a presumptive mutagen under mouse S9-activated
conditions.
b. Mammalian Cells: Ahmed et al. (Article No. 17) assayed
dieldrin for mutation at the ouabain locus in Chinese hamster
lung (V79) cells. Based on a preliminary cytotoxicity test
with dieldrin at doses >100 yM, a dose of 10 yM was
chosen for the mutation assay. Nonactivated dieldrin caused
Ames, B. N., McCann, J., Yamasaki, E. "Methods for detecting carcinogens
and mutagens with the Sa1mone11a/mamma1i an microsome mutagenicity test.
Mutat. Res. 31(1975): 347-364.
B-7
-------
an Increase 1n the number of revertants with 77.8% survival.
However, the study was Incomplete because exogenous metabolic
activation was not used. These data are needed to establish
whether metabolism has an effect on the test material's
mutagenic potential.
Additionally, the study results are considered inconclusive
because:
1. Only a single dose of dleldrin was evaluated for
mutagenlcity.
2. The presumptive mutagenic response was not confirmed in a
repeat assay at the concentration range expected to give
dose-response data.
3. Carbaryl and 2,4-D fluid, which have been repeatedly
shown by others to be either weak mutagens or
nonmutagens, were reported to give strong positive
responses.
4. No positive control was included to ensure an appropriate
level of sensitivity.
Although we assess that the study provides inconclusive
evidence for a positive response, nonactivated dieldrin is
classified as an unconfirmed presumptive mutagen in Chinese
hamster lung cells.
c. Conclusion for 6ene Mutation (Category 1): Dieldrin was
tested in 4 different studies using the S. typhimurium plate
Incorporation assay. It was reported that dieldrin did not
Induce a mutagenic response with or without rat S9 activation
1n three acceptable studies (Article Nos. 13, 14, and 15).
However, 1n Article No. 16, it was reported that the number
of revertants increased with or without mouse S9 activation.
B-8
-------
The weight of evidence from both survey (Table 1) and primary
data (Table 2) indicates that dieldrin is not mutagenic in
bacteria in the presence or absence of rat S9. However, an
inconclusive presumptive positive response with or without
mouse S9 activation was observed in a single as-say. Hence,
we assess that it is unlikely that dieldrin causes gene
mutations in bacteria, unless a validated repeat assay using
mouse S9 can be submitted for review.
The presumptive mutagenlcity of a single nonactivated dose of
dieldrin in Chinese hamster lung cells was not confirmed.
Hence the evidence that dieldrin causes gene mutations in
mammalian cells is inconclusive; the results of this study
should be confirmed 1n valid mammalian gene mutation assays.
2. Chromosomal Aberrations (Category 2):
a. Somatic Cells: Majumdar et al. (Article No. 18) exposed
cultured human lung cells (WI-38) to nonactivated dieldrin at
1, 10, and 30 yg/ml. Dieldrin induced significant (p = <0.01)
dose-related increases in the number of chromosomal aberra-
tions (chromatid breaks, fragments, chromosome interchanges,
and rings.), at all doses. The authors also reported a decrease
in the mitotic index from 8.0 to 1.5% with increasing doses
of dieldrin compared to control (11.0) and the highest dose
tested resulted in a cell degeneration of 50% at 24 hrs. We
assess that the dose-related response indicates a clastogenic
effect. However, the assay should also be performed with 59
activation; therefore the study is inconclusive.
The same investigators administered single ip doses of 1, 30,
or 50 mg/kg of dieldrin to STS mice (sex not specified). Two
hours prior to the 24-hour sacrifice and bone marrow harvest,
the mice received an ip injection of colcemld. The chemical
caused an increase in chromosomal aberrations (chromatid
B-9
-------
breaks, fragments, chromosome Interchanges, and rings). They
reported decreases in the mitotic index from 6.0 to 2.8% with
increasing doses of dieldrin compared to controls (10.5).
Although this assay was performed with only one cell harvest
interval and without a positive control we assess that the
dose-related response indicates a clastogenic effect;
therefore the study is acceptable.
Dean et al. (Article No. 19) administered single oral doses
of 30 or 60 mg/kg HEOD to male and female Chinese hamsters
(oral LO _, 120 mg/kg). At 1.5 hours prior to the 8- and
24-hour sacrifices, the mice received an ip dose of colcemid.
The authors reported that the test chemical did not cause an
increase in the number of chromosomal aberrations at the
8- or 24-hour intervals. However, since no clinical toxicity
or cytotoxic effects were reported, and no positive control
was used, this study is unacceptable.
The same investigators monitored the lymphocytes of workers
in a dieldrin manufacturing plant. They screened 17
presumably non-exposed controls, 9 former and 12 current
plant workers, and reported no differences in chromosomal
aberrations. Levels of dieldrin and/or other manufacturing
components were not reported.
Germinal Cells: To assess the chromosomal effects of HEOD on
male germinal cells, Dean et al. (Article No. 19) performed
two separate dominant lethal tests. In the first experiment,
8 male mice were administered single oral doses of HEOO at
12.5 or 25 mg/kg; the concurrent negative control group
consisted of 16 animals. Each male was mated to 3 females
per week for 8 weeks. The authors reported that there were
no differences in the number of early fetal deaths between
the dosed and the control groups; however, the pregnancy rate
was low for all groups, i.e., 65.1% in the controls and 65.9
B-10
-------
and 64.3% in the dosed groups. Fetal Implantations were
statistically decreased at week 1 in the high-dose group and
at week 3 in the low-dose group. In addition, the authors
indicated that housing conditions may have been "substandard"
(although these conditions were not detailed); -therefore, a
second experiment was performed. Male mice were dosed once
orally with 12.5, 25, or 50 mg/kg of HEOD and each male was
mated to 3 females per week for 5 weeks. The authors
reported that clinical toxicity was observed in the mid and
high doses. Under the presumably "improved" housing condi-
tions, the pregnancy rate for the dosed groups was >80.2%
compared to the controls (76.6%). There was no reduction in
the number of fetal implantations or increases in the number
of early fetal deaths compared to controls over the 5-week
mating period.
We assess that HEOO at the doses tested (12.5, 25, and
50 mg/kg) did not cause a clastogenic response in male
germinal cells over a 5-week period. Although it is recom-
mended that the dominant lethal assay be performed over an
8-week period, we assess that the study 1s acceptable for all
stages of the spermatogenic cycle, except differentiating
spermatogonia and stem cells.
Epstein et al. (Article No. 20) evaluated the chromosomal
effects of dieldrin among 172 other chemicals tested on mouse
germinal cells in dominant lethal assays. Dieldrin was
administered to male mice as a single ip dose at 5.2 or
26 mg/kg in one experiment, and in a second experiment male
mice received oral daily doses of 2 or 3 mg/kg/day for 5
consecutive days. In both experiments male mice were mated,
1 male to 3 females per week for 8 weeks. The authors
reported, in both experiments, that dieldrin did not cause
significant differences in the number of fetal implantations
or total fetal deaths when compared to the control groups.
B-ll
-------
We assessed that at the doses tested, dieldrin did not cause
any clastogenic effects in mouse germinal cells. Although no
cytotoxic responses were reported, the test chemical was
assayed at approximately the LD which was 30 mg/kg; 1/9
deaths occurred at 26 mg/kg, the highest dose "tested. The
study is acceptable.
c. Cone 1 usion for Chrompsoma1 Aberratlions: In single uncon-
firmed tests dieldrin was reported to induce chromosomal
aberrations in human lung cell cultures (WI-38) and mouse
bone marrow cells. HEOD (99%) did not cause chromosomal
aberrations in hamster bone marrow; however, there was no
evidence of cytotoxicity; therefore, the study was not
acceptable. Neither dieldrin nor HEOD induced chromosomal
breaks in mouse germinal cells in two dominant lethal studies.
Based on this evidence dieldrin may be considered presump-
tively clastogenic in somatic mammalian cells, requiring
confirmation in adequately controlled repeat experiments.
3. Other Mutagenic Mechanism (Category 3):
a. DNA Damage/Repair: Dean et al. (Article No. 19) examined
HEOD in six independent host-mediated assays in which
Saccharomvces cerevisiae strain D4 was scored for mitotic
gene conversion. In assays 1-3 male mice were dosed once
orally at 25 and 50 mg/kg and inoculated ip with yeast
cells. In assays 4-6, male mice were dosed orally at 5 or
10 mg/kg/day for 5 consecutive days, and on day 5 mice were
inoculated ip with yeast cells. All animals were sacrificed
5 hours after inoculation. No toxic or cytotoxic effects
were observed in the acute phase of the host mediated assays;
however, in a concurrent dominant lethal assay, evidence of
chemical intoxication was observed following the acute
administration of 50 mg/kg HEOD. In the subacute assays
animals receiving the chemical at 10 mg/kg/day x 5
B-12
-------
were lethargic throughout the dosing intervals. The chemical
did not cause an increase in the numbers of S. cerevisiae
adenine or tryptophan convertants relative to the controls at
any dose level in either experiment. We assess, therefore,
that HEOD was evaluated at the appropriate doses, in properly
controlled host mediated assays; the study is acceptable.
Probst et al. (Article No. 21) tested dieldrin in an
unscheduled ONA-synthesis (DOS) assay with primary rat
hepatocytes at 8 doses ranging from 0.5 to 1000 nM/mL.
Dieldrin did not cause an increase in UOS at the highest
noncytotoxic dose (500 nM/ml) scored. The study was properly
controlled, and is acceptable.
In another UOS assay, Klauning et al. (Article No. 22)
assessed the DNA damaging potential of dieldrin in mouse
-4 -5 -6
hepatocytes at concentrations of 10 , 10 , and 10 M.
No increase in net grains/nucleus was reported. Phenobarbital
pretreated mouse hepatocytes exposed to comparable doses of
dieldrin also showed no increases in net grains/nucleus. The
study was properly controlled and is acceptable.
Rocchi et al. (Article No. 23) tested dieldrin's action on
semiconservative DNA synthesis (SOS) to determine the 50%
inhibition dose with rat thymocytes. They assayed aldrin for
its potential to induce unscheduled ONA synthesis (LIDS) in
human lymphocytes in the presence of hydroxyurea and
tritiated thymidine without metabolic activation. In
addition, they tested aldrin's ability to interfere with
human lymphocyte repair capability following damage exerted
by ultraviolet (UV) rays.
Based on the results of the inhibition of thymidine uptake
assay, dieldrin caused a 50% inhibition in rat thymocytes at
a dose of 100 yg/mL; the dose was used in the UDS assay
B-13
-------
with human cells. Induced UOS repair effects were assessed
after calculations were made to eliminate the effect of SOS
inhibition. Dieldrin induced an increase in UDS over the
control. Similarly, with human cells damaged by UV irradia-
tion and dosed with dieldrin (100 yg/mL), the chemical
slightly interfered with UOS repair. We assess that since
the observed effects were noted at only one dose, the
evidence for a genotoxic response is inconclusive; there-
fore, the assay should be repeated with multiple doses in the
presence and absence of metabolic activation.
Ahmed et al. (Article No. 24) evaluated S9 activated and
nonactivated dieldrin at doses of 1, 10, and 100 yM in a
UDS assay using SV-40 (virus-transformed) human cells (VA-4).
The authors reported a significant (p=<0.05) net increase in
UOS at 1 yM, (12.5±1.8 grains/nucleus) versus the control
(3.5+0.9 grains/nucleus). However, the results for the
mid- and high-doses were reported only as qualitative (+)
positive responses with or without S9 activation after 8
hours of exposure. The lack of primary data for the mid- and
high-dose prohibits an assessment of any dose-related
response.
The cells used in this assay were not grown under reduced
serum or arginine deprivation conditions in conjunction with
hydroxyurea to insure blockage of S-phase synthesis. Several
agents which are generally considered nongenotoxic or weakly
genotoxic (e.g., Diquat 2,4-0 fluid, and carbaryl) induced
strong responses in this study. We discussed the findings of
i
this assay with Dr. Gregory Probst, Lilly Research Labs.
Probst, G. S., Lilly Research Labs., Greenfield, IN, personal communica-
tions 4/86.
B-14
-------
He concurred with our assessment and noted that cells which
may have escaped hydroxyurea treatment could easily be mis-
interpreted as undergoing UDS synthesis, thereby increasing
the chances of a false positive.
In addition the use of a transformed cell line to assess
2
primary DNA damage does not conform with recommended
procedures, which use non-transformed dlplold human cells.
We classify dieldrin as an inconclusive presumptive genotoxin
in SV-40 transformed human cells. We recommend confirmation
of these data in a rigorously controlled experiment that
conforms to the currently recommended protocols for UDS
assays with human cell lines.
Ahmed et al. (Article No. 24) also assessed dieldrin's
potential to cause ONA strand breakage in VA-4 cells using
photolysis of BrdU at a wavelength of 313 nm light. The
authors reported that the test material increased the number
of strand breaks/10 Oaltons at the 100 MM dose. Since
the control data were not reported, and the use of a
transformed cell line was employed, we assess the results to
be inconclusive evidence for a positive response.
b. Inhibition of Spindle Apparatus:
Markaryan (Article No. 25) studied the effects of dieldrin on
bone marrow cells of male mice. He reported that anaphase
and early telophase harvested cells in mice injected ip with
2
Mitchell, A. 0., Casciano, 0. A., Meltz, M. L., Robinson, 0. E., San, R.
H. C., Williams, G. M., and Von Halle, E. S. Unscheduled DNA synthesis
tests, a report of the U.S. Environmental Projection Agency Gene-Tox
Program. Hutat. Res. 123(1983): 363-410.
B-15
-------
0.0012 mg/g dieldrin had Increased frequencies of "nuclear
disturbance," total chromosome rearrangements (chromosome and
chromatid bridges, and fragments), and "adhesions." We dis-
3
cussed the findings of this study with Dr. James Ivett,
and concur with his recommendation that the study is
unacceptable for the following reasons:
1. The method used to prepare anaphase and telophase plates
(slices of bone marrow stained with acetocarmine) may
cause an increase in the number of artifacts, and lends
itself to a planer effect (complete chromosomal displays
are not visible in all viewing fields). This makes the
distinction between true chromosomal rearrangements and
artifact difficult.
2. The number of animals investigated (three male mice) do
not represent a sufficient sample population.
3. The inclusion of cytotoxic effects (vacuolation, karyo-
pycnpsis, and karyolysis) in the total percent nuclear
disturbances suggests that the assay dose (approximately
4X of the LDen) was severely cytotoxic.
4. The slides were not reported as being coded.
We concluded, therefore, that the results are unacceptable
because they cannot be interpreted, and that these findings
should not be included in the genetic assessment of dieldrin.
3
Ivett, 0. L., Hazleton Biotechnologies, Vienna, VA (personal communica-
tions), 4/86.
B-16
-------
c. Conclusion for Other Hutagenic Mechanisms (Category 3):
Published articles that evaluated the potential of dieldrin
to interact with and damage DNA provided evidence that "the
chemical appears not to be genotoxic in yeast, or in rat or
mouse primary hepatocyte cultures. Based on the inconclusive
and unconfirmed findings of a UDS assay with transformed
human cells, dieldrin is listed, with reservations, as
presumptively genotoxic in human cells. We caution, however,
that these findings are suspect and should be confirmed. The
results of the spindle mechanisms/inhibition assay performed
with dieldrin could not be interpreted.
4. Epigenetic Effects:
Wade et al. (Article No. 26) reported a new technique to measure
cell-to-cell communication, which they termed, "gap fluorescence
redistribution after photobleaching" (FRAP). To support the
ability of this method to measure interference with gap junction
communication, the investigators included a known tumor promoter,
12-0-tetradecanoylphorbol-13-acetate (TPA), a nontumorigenic
phorbol ester, 4-phorbol-12,13-didecanoate (4PO), and the
pesticide, dieldrin.
Cultured human teratocarcinoma cells were labeled with 6-carboxy-
fluorescein diacetate and exposed to 7 yg/mL of dieldrin (a
reported noncytotoxic dose). Pairs of contacting cells were
selected and one of each pair was photobleached (dye was photo-
chemically destroyed) by an argon ion laser beam (ACAS 470
laser/imaging). Transfer or redistribution of the dye from the
unbleached cell to the bleached cell was measured over time using
a series of image scans.
B-17
-------
-Oieldrin blocked the fluorescence recovery of photobleached
cells. Fifteen minutes after bleaching the fluorescence
intensity of dieldrin-treated photobleached cells was =1000
units, compared to the control (=14,000) units. It was
reported; however, data were not included in article that the
tumor promoter, TPA, also blocked fluorescence recovery in
exposed cell cultures, and the authors stated that inhibition of
transport by TPA may be related to its tumorigenicity. In
contrast, the nontumorigenic phorbol ester, 4PD, did not block
transport; again however, the 4PO data were not included.
Although this is a new technique that has not been fully
validated, we assess these preliminary results indicate that
dieldrin can inhibit junctional communication in cultured
mammalian cells.
IV. OVERALL INTERPRETATION OF RESULTS AND RECOMMENDATIONS:
1. Interpretation:
The published articles reviewed are not considered sufficient to
establish a complete genetic toxicology profile for dieldrin.
However, certain consistencies emerge which allow us to make
reasonable interpretations and to recommend further studies which
might generate useful data.
Dieldrin was adequately tested in three acceptable Ames assays
with sufficient primary data to conclude that nonactivated and
rat or hamster S9-activated dieldrin is not mutagenic in S.
typhimurium. The results of the seven survey Ames tests with
dieldrin support this conclusion. However, based on the
inconclusive results of a single Ames test, dieldrin, in the
presence of mouse 59 activation, is considered presumptively
mutagenic in Salmonella. We caution, however, that this finding
B-18
-------
has not been confirmed and technical deficiencies preclude
acceptance of the results as valid.
The three £. coli reverse mutation survey studies with dieldrin
further support the conclusion that the chemical is hot mutagenic
in procaryotes.
In the single published mammalian gene mutation assay, non-
activated dieldrin induced an inconclusive, mutagenic response in
Chinese hamster V79 cells. The validity of this assay was
seriously compromised because of technical deficiencies, reporting
of "positive" responses with compounds that have been previously
reported as either weakly mutagenic or not mutagenic, and the
lack of a positive control. Although we classified dieldrin as
an unconfirmed presumptive positive mutagen for Chinese hamster
V79 cells, we have serious reservations about the validity of the
study and recommend that these results be verified in a complete
assay, i.e., three or more noncytotoxic doses, with and without
metabolic activation, and with known mutagens as positive
controls.
Although the majority of doses assayed were cytotoxic, dieldrin
elicited dose-related clastogenic responses in both cultured
human lung and mouse bone marrow cells. Dieldrin, therefore, may
be considered weakly clastogenic in mammalian somatic cells.
By contrast, dieldrin administered by gavage or ip proved to be
nonclastogenic in male germinal cells; HEOD, a major component of
dieldrin, was also negative in male germinal cells.
Survey repair studies indicated that dieldrin does not induce
primary DNA damage in bacteria. Data from studies with
sufficient primary data to draw meaningful conclusions also
indicated that dieldrin was not genotoxic in yeast or primary rat
and mouse hepatocytes.
B-19
-------
However, both S9-act1vated and nonactivated dieldrin induced a
statistically significant increase in UDS in SV-40 transformed
human fibroblasts. The validity of this assay is questionable
because of numerous technical deficiencies, the reporting of
strong "positive" responses by weakly genotoxic or nongenotoxic
substances, and the lack of acknowledged positive controls.
It is noteworthy that all assays reporting that dieldrin adversely
affects genetic material were either flawed by inadequate study
designs (Article Nos. 16, 17, and 26) or showed the greatest
activity at cytotoxic doses (Article No. 18).
In a promising new technique, Wade et al. tested dieldrin against
the known tumor promoter TPA to demonstrate the effectiveness of
the FRAP technology to detect interference with cellular
communication. We assess that their results suggest that
dieldrin may act by an epigenetic mechanism possessing
promotional activity.
2. Recommendations:
Although the evidence from the majority of available published
articles with procaryotic gene mutation assays shows that
dieldrin does not induce gene mutations, its mutagenic potential
for S. typhimurium with or without mouse S9-activation has not
been fully investigated. Likewise, mutations of Chinese hamster
V79 cells to ouabain resistance with or without mouse S9 should
be validated. A reassessment of UDS using human diploid cell
lines in a well controlled experiment would also be useful in
completing information which, at present, constitutes a data gap.
We assess that the most useful information, however, will come
from experiments on promotion activity. Wade et al. demonstrated
that dieldrin may interfere with intercellular communication.
B-20
-------
The following tests are recommended to confirm the potential
promotional activity of the test material:
1. In. vitro transformation with promotional assay
2. in vivo promoting activity assay
3. In vitro UOS assays with SV-40 (Simian virus) transformed
cells (VA-4) and non-transformed human cells.
*U.S. GOVERNMENT PRINTING OFFICE: 1987.7i»8J21/67066 REGION NO. 4
B-21
-------
S1™
Jl
=•
O1
O
O
Oi
CO
o
o
CD
II
O ' 01 Dl
C IT O «
Q) — ~
j " a
to m
2s
||
C -i
3 ®
o 9
c? o
Q. O
0.°
3 o,
If
> me
O O Q.
------- |