oEPA
United States
Environmental Protection
Agency
EPA/625/3-88/014A
May 1988
SAB Review Draft
Research and Development
Thyroid Follicular
Cell Carcinogenesis
Mechanistic and
Science Policy
Considerations
SAB
Review
Draft
(Do Not
Cite or Quote)
I
:.<*'
NOTICE
This document is a preliminary draft. It has not been formally released by EPA and should
not at this stage be construed to represent Agency policy. It is being circulated for comment
on its technical accuracy and policy implications.
-------
-------
December 15, 1987
(Edited May 1988)
THYROID FOLLICULAR CELL CARCINOGENESIS:
MECHANISTIC AND SCIENCE POLICY,CONSIDERATIONS
Prepared for the
Risk Assessment Forum
U.S. Environmental Protection Agency
Washington, DC
December 1987
TECHNICAL PANEL
Principal Authors
Richard N. Hill, M.D., Ph.D., Chair, Office of Pesticides and
Toxic Substances
Linda S. Erdreich, Ph.D., Office of Research and Development
(now with Clement Associates, Incorporated, Edison, NJ)
Orville E. Paynter, Ph.D., Office of Pesticides and Toxic Substances
Patricia A. Roberts, Office,of General Counsel
Sheila L. Rosenthal, Ph.D., Office of Research and Development
Christopher F. Wilkinson, Ph.D., Consultant, Office of Pesticides and
Toxic Substances (on leave, Cornell University)
Other Panel Members
Robert B. Jaeger, M.S., Office of Pesticides and Toxic Substances
Amal Mahfouz, Ph.D., Office of Water
Edward V. Ohanian, Ph.D., Office of Water
Dharm V. Singh, Ph.D., Office of Research and Development
RISK ASSESSMENT FORUM STAFF
Dorothy E. Patton, Ph.D., J.D., Executive Director
Judith S. Bell in, Ph.D., Science Coordinator
William P. Wood, Ph.D., Senior Environmental Scientist
Linda C. Tuxen, B.S., Technical Liaison
-------
DRAFT—DO NOT QUOTE OR CITE
This document Is a draft for review purposes only and does hot constitute
Agency policy. Mention of trade names or commerical products does not
tute endorsement or recommendation for use.
-------
TABLE OF CONTENTS
List of" Figures vii
List of Tables . .vili
Preface ix
Peer Reviewers x
I. EXECUTIVE SUMMARY 1
II. INTRODUCTION 4
III. THYROID-PITUITARY PHYSIOLOGY AND BIOCHEMISTRY 8
A. Synthesis of Thyroid Hormones 8
B. Transport of Thyroid Hormones in the Blood 9
C. Metabolism and Excretion 13
D. Physiological Actions of Thyroid Hormones 13
E. Regulation of Thyroid Hormone Synthesis/Secretion. ...... 14
IV. THYROID AND PITUITARY GLAND NEOPLASIA 17
A. Thyroid Neoplasia 17
1. Induction 17
2. Morphological Stages 19
3. Reversibility of Thyroid Effects 21
B. Pituitary Neoplasia 24
C. Molecular Considerations in Thyroid Carcinogenesis 24
1. Stimulation of Cell Division 27
a. Influence of TSH 27
b. Other Factors 28
c. Possible Controls in Thyroid Cell Division . 33,
2. Cellular Transformation 34
-------
TABLE OF CONTENTS (continued)
V. EXOGENOUS FACTORS INFLUENCING THYROID/PITUITARY
CARCINOGENESIS. ... 38
A. Physical Factors * ...... 38
B. Chemical Factors . . . 39
1. Goitrogens ; * > • 39
a. Naturally Occurring (Dietary) Substances. ....... 39
b. Synthetic Compounds 40
(i) Thionamides ,,.,... 40
(ii) Aromatic amines ......... 40
(iii) Polyhydric phenols 40
c. Modes of Action - 41
2. Enzyme Inducers ..... 41
a. Foreign Compound Metabolism and Enzymes Induction . . . 42
(i) General 42
(ii) Induction ..... 42
(iii) Different inducer types 43
b. Metabolism of Thyroid Hormones 45
c. Effect of Inducers on Thyroid Function and Morphology . 46
(i) PB-type Inducers. ... ..... 46
(ii) 3MC-type Inducers . . . . 48
(iii) Mixed type 49
3. Other Chemicals and Treatment Combinations. ........ 53
a. Other Chemicals 53
b. Combined-Treatment Studies 54
c. Summary 55
iv
-------
TABLE OF CONTENTS (continued)
C. Structure-Activity Relationships 56
i. Chemicals Producing Thyroid Neoplasms in Animals. ..... 56
2. Antithyroid Activity and Thyroid Carcinogenesis 58
a. Thionamides 60
b. Bridged Double Ring Aromatic Amines 67
c. Characteristics of Single Ring Aromatic Amines. .... 68
3. Genotoxicity and Thyroid Carcinogenesis '. . 69
a. Thionamides 70
b. Aromatic Amines 73
c. Complex Halogenated Hydrocarbons 73
d. Amitrole 73
e. Conclusions . 79
VI. HUMAN DATA ON THYROID HYPERPLASIA AND NEOPLASIA . 82
A. Thyroid-Pituitary Function 83
B. Causes of Thyroid Hyperplasia. . . 83
1. Chemical Inhibitors . 83
2. Dietary Factors 88
a. Iodine Deficiency 88
b. Other Goitrogens 89
C. Causes of Thyroid Neoplasia 91
1. Descriptive Epidemiology 92
2. Analytical Epidemiology 97
-------
TABLE OF CONTENTS (continued)
VII. DEVELOPMENT OF SCIENCE POLICY . . 101
A. Rationale • 101
B. Science Policy ........ 109
VIII. APPENDICES
i -
APPENDIX A. Combined Treatment Studies .... A-l
APPENDIX B. Single Ring Aromatic Amines. . B-l
APPENDIX C. Genotoxicity—Ethylenethiourea C-l
APPENDIX D. Genotoxicity--4,4'-Oxydianiline. D-l
APPENDIX E. Genotoxi city—Ami trole E-l
IX. REFERENCES. .
VI
-------
LIST OF FIGURES
NUMBER PAGE
1. Schematic representation of thyroid hormone biosynthesis
and secretion 9
2. lodinated compounds of the thyroid gland. . 11
3. Hypothalamic-pituitary-thyroid-peripheral organ relationships . . 12
4. Hypothetical model for thyroid carcinogenesis 26
5. Possible control points for cell division in the pre-DNA
synthetic portion of the cell cycle 35
-------
LIST OF TABLES
NUMBER
1, Effects of stimuli on thyroid cells
PAGE
. 30
2. Chemicals in the NCI/NTP bioassay program showing at least
some evidence of thyroid follicular cell neoplasia. . . , 57
3. Thionamides negative for thyroid neoplasia in NCI/NTP studies . . 59
4A. Thionoinide: relationship between antithyroid activity and
thyroid carcinogenicity - Heterocyclic compounds. ,•„.„,,,, 61
4B. Thionoinide: relationship between antithyroid activity a;nd
thyroid carcinogenicity - Thiourea derivatives ,.
.. , 62-64
5. Aromatic amines relationship between antithyroid activitiy
and thyroid carcinogenesis. . , , . i, , . . . 56-67
6. Genotoxicity data for thionarnides ...........;.,.., 71
7. Genotoxicity data for single ring aromatic amines . . ,. . , .'.. . 74
8. Genotoxicity data for bridged double ring aromatic amines .... 75
9. Genotoxicity data for miscellaneous aromatic amines ,..;.,.,,,. 7-6
10. Genotoxicity data for complex hal ogenated hydrocarbons, i. .... 77
11. Studies on humans indicating effects of chemicals on thyroid-
pituitary functions ...„.* ,,,.«... 84-86
12. Epidemiologic studies of thyroid cancer and its relationship
to goiter and thyroid nodules 98
viii
-------
PREFACE
The U.S. Environmental Protection Agency (EPA) Risk Assessment Forum was
established to promote scientific consensus on risk assessment issues and to
ensure that this consensus is incorporated into appropriate risk assessment
guidance. To accomplish this, the Risk Assessment Forum assembles experts from
throughout the EPA in a formal process to study and report on these issues from
an Agency-wide perspective.
For major risk assessment activities, the Risk Assessment Forum may estab-
lish a Technical Panel to conduct scientific review and analysis. Members are
chosen to assure that necessary technical expertise is available. Outside
experts may be invited to participate as consultants or, if appropriate, as
Technical Panel members.
The scientific analysis and policy recommendations in this report on
thyroid neoplasia are based mainly on laboratory studies in which thyroid tumors
in animals exposed to exogenous chemicals were associated with disruption in
normal thyroid-pituitary function. The Forum analysis enlarges upon a 1986
Office of Pesticide Programs report on this issue and develops science policy
recommendations for Agency-wide use.
-------
EXTERNAL PEER REVIEWERS
The following External Peer Reviewers have reviewed and commented on
an intermediate draft of this report.
Gary A. Boorman, D.V.M., Ph.D.
Chief
Michael R. Elwell, D.V.M., Ph.D.
Section Chief
Scott L. Eustis, D.V.M., Ph.D.
Section Chief
Robert R. Maronpot, D.V.M
Section Chief
Chemical Pathology Branch
Division of Toxicology Research
and Testing
National Institute of Environmental
Health Sciences
Research Triangle Park, NC
Gerard N. Burrow, M.D.
Professor and Chairman
Department of Medicine
University of Toronto
Toronto, Ontario
W. Gary Flamm, Ph.D.
Director
Ronald J. Lorentzen, Ph.D.
Assistant to the Director
Office of Toxicological Sciences
Center for Food Safety and Applied
Nutrition
Food and Drug Administration
Washington, DC
Sidney H. Ingbar, M.D., D.Sc.
Professor
Department of Medicine
Harvard Medical School/
Beth Israel Hospital
Boston, MA
R. Michael McClain, Ph.D.
Director of Toxicology
Department of Toxicology and
Pathology
Hoffmann-LaRoche Inc. Nutley, MJ
Jack H. Oppenheimer, M.D.
Professor
Departments of Medicine and of
Physio! ogy
University of Minnesota—
Health Sciences
Minneapolis, MN
David Schottenfeld, M.D.
Professor and Chairman
Department of Epidemiology
University of Michigan
School of Public Health
Ann Arbor, MI
Jerrold M. Ward.,. D.V.M.
Tumor Pathology and
Pathogenesis Section
Laboratory of Comparative
Carcinogenesis
National Cancer Institute
Frederick, MD
E. Dillwyn Williams, M.D.
Professor
Department of Pathol ogy
University of Wales College of
Medicine
Cardiff, Wales
The Technical Panel acknowledges with appreciation the special contributions
of Ms. Karlene Thomas and Ms. Pamela Bassford.
-------
I. EXECUTIVE SUMMARY
A Technical Panel of the U.S. Environmental Protection Agency's (EPA) Risk
Assessment Forum investigated potential mechanisms of action of agents that
cause thyroid follicular tumors in animals and potentially in humans in an
effort to develop a scientifically plausible approach for assessing risk due to
exposure to these agents. Based on its review of relevant scientific information,
the Technical Panel concluded that:
(1) thyroid follicular cell tumors may arise from long-term disturbances
in thyroid-pituitary hormonal feedback under conditions of reduced
circulating thyroid hormone and elevated thyroid stimulating hormone
(TSH);
(2) the steps leading to these tumors are expected to show thresholds,
such that the risks of tumor development are minimal when thyroid-
pituitary homeostasis exists; and
(3) models that assume thresholds may be used to assess the risks of
certain thyroid follicular cell tumors where there is evidence of
thyroid-pituitary imbalance.
The policy set out in this report provides guidance on determining whether
it is reasonable to presume that the observed thyroid follicular tumors are the
result of thyroid-pituitary imbalance, and on selecting appropriate procedures
to use in estimating the risks related to these tumors.
The scientific information reviewed by the Technical Panel provides sufficient
evidence to support the conclusion that a threshold mechanism is likely to apply
to the development of certain thyroid follicular tumors. In particular, several
different types of experimental treatments in laboratory animals (e.g., iodide
deficiency, subtotal thyroidectomy, chemical goitrogens) result in the formation
-------
of thyroid tumors and to some extent pituitary tumors of the cells that secrete
TSH, seemingly by the same mechanism. Tumors arise under conditions in which
there is prolonged decrease in circulating thyroid hormone and increase in TSH.
Under continued TSH stimulation, thyroid follicular cells undergo hypertrophy,
hyperplasia and, eventually, neoplasia. It appears that TSH, probably in
concert with other factors (e.g., somatomedins), acts as a stimulus for cell
division, thus increasing the pool of cells at risk for neoplastic transformation.
ioH may also play a role in the transformation process by yet undiscovered
means.
Studies in humans reveal that they respond as do animals in regard to
goitrogem'c stimuli (e.g., iodide deficiency, thionamides); there is cellular
hypertrophy and hyperplasia. Although thyroid enlargements and nodules may be
risk factors for cancer development in humans, the case for neoplastic conversion
under goitrogem'c stimulation is less well established in humans than in animals.
This suggests that humans may be less sensitive to the carcinogenic effects of
long-term TSH stimulation than animals.
In its assessment of the relevant information, the Technical Panel focused
on the following evidence: (1) a progression of events occurring under long-term
exposure to an agent, including a disruption in thyroid-pituitary homeostasis
involving reduction in thyroid hormone concentrations and increase in TSH
levels, follicular cell hypertrophy and hyperplasia, benign follicular cell
neoplasia and, possibly, malignant follicular cell neoplasia; (2) reversibility
of certain steps in the progression when thyroid-pituitary homeostasis is
reestablished; and (3) lack of consistent correlation of thyroid carcinogenicity
with the genotoxic potential of chemical classes implicated in thyroid cancer.
Based on this primary evidence, the Technical Panel developed a policy for risk
assessment of agents that cause thyroid follicular cell tumors.
-------
Briefly, the Technical Panel determined that threshold models may be
applied in dose-response assessments for those chemical substances where only
i
thyroid tumors (and relevant pituitary tumors) have been produced; the tumors
can be attributed to a disruption in thyroid-pituitary hormonal homeostasis:
and mechanisms other than thyroid-pituitary imbalance, e.g., genotoxicity, can
be ruled out. Where there are tumors at other sites and/or genotoxicity is
present, it is presumed that threshold models will not be used; however, case-
by-case determinations are possible. Threshold models will not be used where
there is no evidence of thyroid-pituitary imbalance.
Finally, the Technical Panel advised that in evaluating thyroid follicular
cell neoplasms under this policy, the risk assessment depends on full use of
the available information. In any given organism, a carcinogen may act through
more than one mechanism at one or multiple anatomical sites. Accordingly,
while use of this policy may be appropriate for assessing certain thyroid
follicular cell tumors, use of other models may be necessary to evaluate risks
at other tumor sites observed in the same study, which may result in different
risk estimates. It is incumbent upon the risk assessor to consider all relevant
risk estimates in making the final judgments on the potential human risk related
to exposure to the chemical being evaluated.
-------
II. INTRODUCTION
Responding to a request from the Office of Pesticides and Toxic Substances,
the Risk Assessment Forum established a Technical Panel to study issues raised
in an Office of Pesticide Programs (OPP) report on neoplastic ;changes in the
thyroid gland. Thyroid follicular neoplasia V, the subject of this report, is
a form that has been associated with low iodine diets, subtotal thyroidectomy,
radioactive iodine, natural goitrogens such as rape seed and cabbage,
chemotherapeutic agents such as sulfathiazole, pesticides such as amitrole,
!
industrial chemicals like polychlorinated biphenyls, and contaminants like
2,3,7,8-tetrachlorodibenzo-Mioxin. All of these agents either directly or
indirectly interfere with the normal thyroid-pituitary feedback system.
In the OPP report, "Neoplasia Induced by Inhibition of Thyroid !G"Iand
Function (Guidance for Analysis and Evaluation)," Paynter et al., (1986) postulated
that there is a causal relationship between thyroid-pituitary dysfunction and
thyroid follicular neoplasia, and further that the mechanism underlying this
relationship may be a threshold phenomenon. If this were the case, thyroid
follicular carcinogenesis would not be expected to occur below a demonstrable
threshold level of thyroid-pituitary dysfunction.
Simply stated, the OPP report described a possible mechanism for thyroid
follicular neoplasia that involves interference with the normal physiological
thyroid-pituitary hormonal feedback mechanism. It is postulated that certain
\ ;
I/ This report deals with mechanistic considerations surrounding the development
~~ of tumors of the parenchyma! cells of the thyroid. In the experimental
animal literature, such tumors are usually called follicular cell adenomas and
carcinomas. The clinical literature usually divides human follicular cell tumors
into different classes depending upon their histological features: follicular,
papillary, and anaplastic. Neoplasms of the calcitonin-secreting parafollicular
or C-cells (i.e., medullary tumors) are not considered in this report.
-------
chemicals may result in decreased levels of thyroid hormone £/ in the blood
which result in increased release of thyroid stimulating hormone (TSH) by the
anterior pituitary. This, in turn, leads to hypertrophy and hyperplasia of the
thyroid without a corresponding increase in blood thyroid hormone levels;
hyperplasia of the pituitary is also sometimes observed due to the reduced
levels of circulating thyroid hormone. After prolonged stimulation of the
thyroid-pituitary axis, thyroid (and to some extent, pituitary) hyperplasia
may progress to neoplasia. Cessation of exposure prior to the induction of
neoplasia results in a return toward the normal state. Because some degree of
thyroid-pituitary dysfunction can be accommodated within the bounds of the
normal feedback mechanism without induction of hyperplasia, a threshold for
thyroid follicular cell carcinogenesis via hyperplasia appears to be indicated.
Thus, for a chemical substance that decreases thyroid hormone levels, a dose
below which it has any effect on thyroid pituitary hormone status may be conceived
of as a threshold for the thyroid carcinogenic process.
Forum review of the issues raised in the OPP report was considered
appropriate because of the potentially significant implications for carcinogenic
risk assessment inherent in the OPP hypothesis. A risk assessment approach
based on thyroid follicular neoplasia being a threshold phenomenon would be a
significant departure from EPA's customary carcinogen risk assessment practice,
which generally uses "nonthreshold" models for extrapolation from high- to
low-dose exposures, based on the assumption that human carcinogenesis may
develop as a result of exposure to carcinogens even at the very lowest levels.
EPA's risk assessment guidelines recommend the linearized multistage model for
2/ In this report, "thyroid hormone" is often used as a collective term to
~ refer to the active thyroid hormones released from the thyroid gland into
the circulation (thyroxine and 3,5,3'-triiodothyronine).
-------
carcinogen risk assessment to place an upper bound on potential cancer risks,
in the absence of relevant biological and statistical information to the contrary
'I
(U.S. EPA, 1986). However, the guidelines also stress that all of the available
mechanistic, toxicological, metabolic, and pharmacokinetic information should
be reviewed for each chemical in making judgments about the appropriateness,
selection, and use of various extrapolation models.
The Technical Panel undertook the present analysis with three objectives:
(1) to explore the role of thyroid-pituitary relationships in thyroid carcino-
I
genesis; (2) to determine if threshold concepts might apply to the steps leading
to thyroid cancer and (3) if warranted, to develop Agency-wide guidance on how
threshold considerations may affect the estimation of risks from exposure to
chemicals that produce thyroid tumors.
The Technical Panel has studied the OPP report, as well as an extensive
number of additional studies and other information sources in order to assess
whether the hypothesis set forth in the report is consistent with available
information on human and animal thyroid neoplasia, thyroid-pituitary physiology
and function, and the mechanisms of carcinogenesis. Upon review of such
evidence, the Technical Panel agrees that under certain circumstances neoplasia
in thyroid follicular cells involves interference of thyroid-pituitary feedback
mechanisms and may involve threshold rather than nonthreshold processes. It
is recognized that when there is evidence that the thyroid follicular tumors
i
are related to an ordered linkage of steps from interference in thyroid-pituitary
status leading to depressed thyroid hormone concentrations, elevated TSH levels,
thyroid hypertrophy and hyperplasia, and neoplasia (adenoma and possibly carcinoma),
then the threshold for an earlier step becomes a threshold for the entire chain
of events. This Risk Assessment Forum report presents the findings of the
Technical Panel.
6
-------
The report has nine sections. Section I ris an Executive Summary and
this introduction constitutes Section II. Section III summarizes information on
thyroid-pituitary physiology and biochemistry, and the hormonal feedback relation-
ship between these glands. Section IV reviews the available information on the
induction of thyroid follicular neoplasia, and sets forth a hypothetical mechan-
istic model based on current information on molecular and cellular processes
relating to thyroid carcinogenesis. In Section V, exogenous factors affecting
thyroid carcinogenesis are discussed, focusing primarily on information developed
in experimental animals. Thyroid hyperplasia and neoplasia in humans are
discussed in Section VI, and Section VII develops a science policy to guide the
development of EPA risk assessments on this issue. Finally, Sections VIII and IX
are the Appendices and References, respectively.
-------
III. THYROID-PITUITARY PHYSIOLOGY AND BIOCHEMISTRY
In order to examine the possible role of pituitary,, thyroid
, and related
hormones in thyroid carcinogenesis, it is important to first understand the
physiology and biochemistry of the thyroid-pituitary hormonal system. Accordingly,
this section summarizes the nature, formation, and secretion of the thyroid
hormones and discusses the mechanisms by which circulating levels of the hormones
are regulated. References are mainly to recent reviews (see especially Paynter
et a!., 1986) rather than to the original scientific literature.
A. SYNTHESIS OF THYROID HORMONES
The thyroid hormones are synthesized in the thyroid gland and are stored
as amino acid residues of thyroglobulin, a protein constituting most of the
colloid in the thyroid follicles (Goodman and Van Middlesworth, 1980; Taurog?,
1979; Haynes and Murad, 1985). Thyroglobulin is a complex glyctiprotein made up
of two identical subunits each with a molecular weight of 330,000 daltons.
The first stage in the synthesis of the thyroid hormones is; the uptake of
iodide from the blood by the thyroid gland (Figure 1). Uptake is active in
nature (requires energy) and is effected by the so-called "iodide pump." Under
normal conditions the thyroid may concentrate iodide up to about 50-fold its
concentration in blood, and this ratio may be considerably higher when the
thyroid is active. Iodide uptake may be blocked by several anions (e.g.,
thiocyanate and perchlorate) and, since iodide uptake involves concurrent
uptake of potassium, it can be also blocked by cardiac glycosides that inhibit
potassium accumulation.
The next step in the process is a concerted reaction in which iodide- is
oxidized to an active iodine species that in turn iodinates the tyrosy! residues
of thyrogTobulin. The reaction is effected by a heme-containing peroxidase in
i
8
-------
Thyroglobulin secretion
Thyroglobulin
synthesis
Iodide pump
Oxidation
\
Incorporation
HO-/V 0
Coupling
Deiodination
ResorptionfHO
Secretion
Proteolysis
Follicle Cell
Follicular lumen
Figure 1. Schematic representation of thyroid hormone biosynthesis and secretion.
Protein portion of thyroglobulin is synthesized in rough endoplasmic reticulum.
It then travels to Golgi apparatus, where carbohydrate moieties are added, and
proceeds to the apical surface in secretory vesicles, which fuse with the apical
membrane and discharge their contents into the lumen. Iodide is pumped into
the cell of a peroxidase. At the apical surface, it is oxidized through the
action of a peroxidase. Iodine attaches to tyrosine residues in peptide linkage
in thyroglobulin. Two iodinated tyrosyl groups couple in ether linkage to form
thyroxine, which is still trapped in peptide linkage within thyroglobulin. The
secretory process requires that thyroglobulin be engulfed by pseudopods thrown
out into follicular lumen to resorb thyroglobulin into vesicles that fuse with
lysosomes. Lysosomal protease breaks thyroglobulin down to ami no acids, T4, 1%,
MIT, and DIT. T4 and 1$ are released from the cell. DIT and MIT are deiodinated
to free tyrosine and iodide, both of which are recycled back to iodinated thyro-
globulin. (DIT = Diiodotyrosine; MIT = Monoiodotyrosine).
Source: Goodman and Van Middlesworth, 1980.
-------
the presence of hydrogen peroxide. While diiodotyrosyl (DIT) residues constitute
the major products, some monoiodotyrosyl (MIT) peptides are also produced
(Figure 2). Additional reactions involving the coupling of two DIT residues or
of one DIT with one MIT residue (each with the net loss of alanine) lead to
peptides containing residues of the two major thyroid hormones, thyroxine (T/|)
and triiodothyronine (Ta), respectively (Figure 1). It is thought that these
reactions are catalyzed by the same peroxidase effecting the iodination reaction,
and it seems that both peroxidase steps are blocked by certain compounds such
as thiourea and some sulfonamides. \
The release of T4 and T3 from thyroglobulin or smaller peptides is effected
by endocytosis of colloid droplets into the follicular epithelial cells and
subsequent action of lysosomal proteases. The free hormones are subsequently
released into the circulation. It is not known whether thyroglobulin must be
hydro!yzed completely to permit release of T4 and T3.
Although T4 is by far the major thyroid hormone secreted by the thyroid
(normally about 8 to 10 times the rate of TS, although it varies as a function
of the iodine intake), it is usually considered to be a prohormone. Thus, TS is
about fourfold more potent than T4, and about 33 percent of the T4 secreted
undergoes 5'-deiodination to TS in the peripheral tissues; another 40 percent
undergoes deiodination of the inner ring to yield the inactive material reverse
TS (Figure 2). ;
B. TRANSPORT OF THYROID HORMONES IN BLOOD
On entering the circulation, both T4 and 1% are transported in strong, but
not covalent, association with plasma proteins (Figure 3). The major carrier-
protein is thyroxine-binding globulin, a glycoprotein (M.W. 63,000) that forms
a 1:1 complex with the thyroid hormones. Thyroxine-binding globulin has a very
high affinity for T4 (Ka about 1010 M) and a lower affinity for T3. Thyroxine-
10 '
-------
5
HO-/0\CH2-CH-COOH tyrosine
3 NH2
Monoiodotyrosine (MID) = 3-iodotyrosine
Diiodotyrosine (DID) = 3,5-diiodotyrosine
5' 5
HD-/0\-0-/ 0 \-CH2-CH-COOH thyronine
3' 3 NH2
Thyroxine (14) = 3,5,3' ,5'-tetraiodothyronine
Triiodothyrom'ne (13) = 3,5,3'-triiodothyronine
Reversed triiodothyrom'ne (^3) = 3,3',5'-triiodothyronine
Figure 2. lodinated compounds of the thyroid gland.
11
-------
HYPQTHALAMUS
V"
PITUITARY
THYROID
BLOOC
(TBG, TBP,
TH/ albumin) \TH
LIVER
OTHER
ORGANS
Bilary
Excretion
TRH—thyrotropin releasing hormone
TSH—thyroid stimulating hormone
TH— thyroid hormones
TBG—thyroxine binding globulin
TBP—thyroxine binding prealbumin
Figure 3. Hypothalamic-pituitary-thyroid-peripheral organ relationships.
-------
binding prealbumin and albumin also transport thyroid hormones in the blood;
the prealbumin has Ka values of about 10^ M and 10^ M for T^. and 13, respectively.
Only about 0.03 percent of the 14 in the circulation is free and available for
cell membrane penetration and thus hormone action, metabolism, or excretion.
The levels of free thyroid hormones in the circulation may be changed through
competitive binding interactions of certain drugs and other foreign compounds
(Haynes and Murad, 1985).
C. METABOLISM AND EXCRETION
As previously discussed, T4, the major hormone secreted from the thyroid,
is considered to be a prohormone and is converted to the more active T3 by
5'-monodeiodination in a variety of peripheral tissues. T4 is also metabolized
to reverse TS which is hormonally inactive and has no known function, except
perhaps as an inhibitor of the conversion of T4 to T3- Under normal conditions
the half-life of T4 is 6 to 7 days in humans.
Degradative metabolism of the thyroid hormones occurs primarily in the
liver and involves conjugation with either glucuronic acid or sulfate through
the phenolic hydroxyl group. The resulting conjugates are excreted in the bile
into the intestine. A portion of the conjugated material is hydrolyzed in the
intestine, and the free hormones thus released are reabsorbed into the blood
(enterohepatic circulation). The remaining portion of the conjugated material
(20% to 40% in humans) is excreted in the feces.
D. PHYSIOLOGIC ACTIONS OF THYROID HORMONES
While not of direct relevance to this discussion, the thyroid hormones
play numerous and profound roles in regulating metabolism, growth, and development
and in the maintenance of homeostasis. It is generally believed that these
actions result from effects of the thyroid hormones on protein synthesis.
13
-------
There is considerable evidence to suggest that many of the various biological
effects of'the thyroid hormones are initiated by the interaction of 13 with
specific nuclear receptors in target cells, presumably proteins (Oppenheimer,
1979). Recent evidence points to these receptors being the products of the
c-erb-A oncogene (Weinberger et al., 1986; Sap et al., 1986). Such interactions
can lead, directly or indirectly, to the formation of a diversity of mRNA
sequences and ultimately to the synthesis of a host of different enzyme proteins.
Qualitative and quantitative differences in the responses resulting from formation
of TVj-receptor complexes may occur in different target tissues. Such differences
may be controlled at a local cellular level and may be mediated through metabolic
or hormonal factors.
E. REGULATION OF THYROID HORMONE SYNTHESIS/SECRETION
I
Homeostatic control of thyroid hormone synthesis and secretion in the
thyroid gland is effected by a sensitive feedback mechanism that responds to
changes in circulating levels of the thyroid hormones T4 and T3- The mechanism
involves the hypothalamus and anterior pituitary of the brain, (Figure 3) (Paynter,
et al., 1986; Larsen, 1982; Houk, 1980)
Of central importance in the feedback mechanism is the tlhyroid stimulating
hormone (TSH, thyrotropin), which is secreted by the anterior pituitary gland
and causes the thyroid to initiate new thyroid hormone synthesis. Increases in
iodide uptake, the iodination of thyroglobulin, and endocytosis and proteolysis
of colloid are all observed in response to TSH stimulation. The effects of TSH
on the thyroid appear to be the consequence of binding to eel I-surface receptors
and activation of adenyl cyclase and protein kinase with subsequent phosphory-
lation of cellular proteins. Cyclic adenosine monophosphate (cAMP) can itself
i
mimic most of the actions of TSH on thyroid cells (Van Sande et al., 1983;
Roger and Dumont, 1984). Further details of the molecular biology of TSH
.14
-------
action on the thyroid are discussed elsewhere in this document (Section IV.C.).
The rate of release of TSH from the pituitary is delicately controlled by
the amount of thyrotropin-releasing hormone (TRH) secreted by the hypothalamus
and by the circulating levels of 14 and 13. If for any reason there is a
decrease in circulating levels of thyroid hormones, TSH is secreted and thyroid
function is increased; if exogenous thyroid hormone is administered, TRH secretion
is supressed and eventually the thyroid gland becomes inactive and regresses.
It appears that the plasma concentrations of both T4 and TS (and possibly
intracellular formation of T3 from T4 in the .pituitary) are important factors
in the release of TSH; they also may modulate the interaction of TRH with its
receptors in the pituitary (Goodman and Van Middlesworth, 1980; Hinkle and Goh,
1982; Larsen, 1982; Ross et al., 1986). Lastly, in the pituitary T4 undergoes
5'-mono-deiodination to T3- In the rat about 50 percent of T3 within pituitary
cells arises from this means. When serum T4 is reduced but T3 is normal,
pituitary intracellular T3 is reduced, and cells are able to respond to the
decreased serum T4 and increase TSH secretion (Larsen, 1982).
Thyroid hormone-responsive tissues contain a variable number of nuclear
receptors for thyroid hormones (mainly T3) usually in excess of several thousand
per cell (Oppenheimer, 1979). Under euthyroid conditions in the rat, usually
about 30 to 50 percent of the sites are occupied by T3, although in the pituitary
more like 80 percent of the sites are filled under physiological conditions.
The T3-receptor complex is quite labile with a half-life for dissociation of
about 15 minutes; the released T3 reenters the exchangeable cellular pool where
it can complex with another receptor or exit the cell. The half-life for J-$
clearance from the plasma in experimental animals is variable, being about 6 hr
in the rat (Oppenheimer, 1979).
Studies on the regulation of TSH output from the pituitary have indicated
15
-------
a link between 13 nuclear receptor occupancy and the mRNA levels for the T$H
subum't chains. Administration of exogenous 13 resulted in decreases in TSH
mRNA levels in the pituitaries and in transplanted pituitary tumors of thyroid-
ectomized mice within 1 day of administration (Chin et al., 1985). Subunit
messenger RNA elongation in nuclei isolated from pituitary tumors of mice
treated in vivo with 13 is decreased within 1/2 hr after hormone administration,
and mRNA levels were reduced within 1 hr (Shupnik et al., 1985). It appears
that the decrease in mRNA is either due to decreased transcription or decreased
stability of the mRNA transcripts. A straight-line relationship existed between
the proportion of nuclear 13 receptors occupied and the proportional reduction
in TSH subum't transcripts in transplanted pituitary tumors (Shupnik et al«s
1986). A 50 percent reduction in mRNA transcripts occurred when about 45
percent of the receptors were occupied; this occurred at plasma 13 levels of
about 1 ng/mL (1.5 x 10-9 M). |
Other studies have investigated the effects of withdrawal of 13 on TSH
mRNA levels in thyroidectomized mice bearing transplanted pituitary tumors
(Ross et al., 1986). Plasma TS levels dropped precipitously within 1 day after
withdrawal; plasma TSH concentrations rose fourfold between 1 and 2 days; and
tumor TSH subum't mRNA levels increased markedly between days 1 and 2.
These experiments demonstrate the rapid response of the pituitary gland to
increases and decreases in plasma TS levels. It seems that pituitary cells
modulate the levels in TSH subunit mRNAs as a function of the proportional
occupancy of the numerous nuclear receptors for 7%. ;
-------
IV. THYROID AND PITUITARY GLAND NEOPLASIA
As described in the previous section, the pituitary exerts a delicate
control over the morphological and functional status of the thyroid, and thyroid
hormones are in turn important regulators of pituitary function. It is perhaps
not surprising, therefore, that the pituitary may be affected profoundly by
factors causing thyroid gland dysfunction. Because of this close dependency,
it is appropriate to discuss thyroid and pituitary neoplasia in the same section.
A. THYROID NEOPLASIA
While, statistically, clinical thyroid cancer is not a serious human
health problem in the United States (it accounts for 0.4 percent of all cancer
and 9 in 1 million deaths annually), occult thyroid cancer discovered at autopsy
is much more common (average about 2 percent autopsies). Other thyroid lesions,
like "nodules" noted upon palpation of the thyroid, occur in about 4 to 7 percent
of adults and are of concern to physicians because they may be or develop into
thyroid malignancies (Paynter et al., 1986; De Groot, 1979; Sampson et al., 1974;
Rojeski and Gharib, 1985).
1. Induction
Thyroid neoplasia may be induced by exposure of experimental animals to a
variety of exogenous chemicals or physical agents. This is the major focus of
this paper and is discussed in some detail in Section V.
It has been recognized for some time, however, that thyroid gland follicular
cell neoplasia can also be induced in experimental animals by a number of other
factors that cause thyroid gland dysfunction, in particular those leading to
hypothyroidism. Among these factors are iodine deficiency (Bielschowsky, 1953;
Axel rod and Leblond, 1955; Schaller and Stevenson, 1966) and subtotal thyroidectomy
(Dent et al., 1956). In addition, thyroid tumors can result from the transplan-
17
-------
tation of TSH-secreting pituitary tumors (Dent et al ., 1956; Haran-Guera as
i
those noted following purposeful manipulation of TSH (e.g.., iodine deficiency).
I
In addition, treatments which raise TSH levels cooperate with irradiation in
i ' . •
increasing the frequency of thyroid tumors, while ablation of TSH 'Stimulation
(e.g., hypophysectomy) under these experimental conditions blocks tumor development
(Doniach, 1970, 1974; Nadler et al,., 1970; -MAS, 1980). Thus, part of the
irradiation-induced carcinogenicity appears to be due to or responsive to
increases in TSH levels. :
Still further support for the role of TSH in thyroid carcinogenesis comes
from experiments using chemicals which reduce circulating thyroid hormone levels
118
-------
and result in increases in TSH (see Section V.B.). Thyroid hyperplasia and
neoplasia in these cases can be blocked by doses of exogenous thyroid hormone
that reestablish thyroid-pituitary homeostasis or by hypophysectomy (for examples,
see Yamada and Lewis, 1968; Jemec, 1980).
2. Morphological Stages in Thyroid Neoplasia
The progressive morphological changes that occur in thyroid tissues in
response to prolonged elevated levels of TSH have been studied in some detail
and are qualitatively similar irrespective of the nature of the stimulus causing
TSH elevation (low iodine diet, goitrogen exposure, etc.) (Gorbman, 1947; Denef
et al., 1981; Philp et al., 1969; Santler, 1957; Wynford-Thomas et al., 1982a;
Wollman and Breitman, 1970). Following initiation of long-term TSH stimulation,
changes in the thyroid exhibit three different phases—an initial lag phase of
several days, a period of rapid growth, and a period of declining growth rate
as a plateau is attained.
During the lag or latent period, that may last for several days, thyroid
weight and DMA content remain relatively constant. Rapid changes occur in the
morphology of the gland during this period, however, characterized by resorption
of colloid from the follicular lumen and by increases in epithelial cell volume
(the cells change from a cuboidal to a more columnar form) and vascularity.
Consequently, the latent period is characterized by a redistribution of thyroid
tissue and compartment volumes and particularly by hypertrophy of the
follicular epithelial cells.
With continued TSH stimulation, the latent period is followed by a rapid
and prolonged increase in thyroid weight and size. Although all thyroid tissue
components proliferate to some extent, the major changes observed are associated
with follicular cell hyperplasia. Thus, there are dramatic increases in both
mitotic activity and in the number of follicular cells per gland (Wynford-Thomas
19
-------
et al., 1982a). There are, however, limits to the extent to which thyroid
I
hyperplasia, as well as thyroid weight and size can continue to increase.
Thus, despite a sustained TSH stimulus (e.g., administration of goitrogen) and
sustained increases in the circulating levels of TSH, mitotic activity of the
follicle cells progressively declines, and thyroid size and weight level off to
a plateau (after about 80 days of goitrogen treatment) (Wynford-Thomas et al.,
I
1982a» b). If the TSH stimulus is withdrawn for 25 days and then reintroduced,
the maximum size of the thyroid remains unchanged (Wynford-Thomas et al .r 1982b}.
Although far from definitive, the mechanism of this "desensitization" to the
stimulating effects of TSH does not appear to be due to a significant
"downregulation" (decrease) of the number of TSH receptors per cell (Witte and
McKenzie, 1981; Davies, 1985). While subsequent studies (Wynford-Thomas et
al., 1982c; Stringer et al., 1985) have failed to elucidate the desensitization
mechanism, it has been suggested that it is mediated by an intracellular change
in the follicular cell either at the receptor or postreceptor level. Clearly,
there exists an intracellular or intercellular control mechanism that limits
I
the mitotic response of thyroid follicle cells to TSH, which led Wynford-Thomas
et al. (1982c), to propose that the failure of this control mechanism might be
the first step in neoplasia. Possibly thyroid cells undergoing; repeated cell
division become irreversibly committed to a differentiated state and are no-
longer able to respond to TSH. On the other hand, cellular responsiveness to
TSH may depend upon interactions with other growth mediators.\ In support of
this, TSH-induced increases in cell number in vivo were closely correlated with
changes in receptor density for another protein growth factor (somatomedin A)
(Polychronakos et al., 1986).
Certainly, under experimental conditions of prolonged stimulation by TSH,
diffuse thyroid hyperplasia may progress to a nodular proliferation of the
20
-------
follicular cells and eventually to neoplasia (Gorbman, 1947; Money and Rawson,
1950; Griesbach et al., 1945; Doniach and Williams, 1962). While many of the
resulting tumors are benign, prolonged and excessive thyroid stimulation may
result in malignant tumors. The morphology of thyroid tumors in laboratory
rodents has been discussed in several reviews (Doniach, 1970b; Boorman 1983;
Frith and Heath, 1983). Studies with humans show a similar morphologic
progression of the thyroid up through nodular hyperplasia and "adenomatous"
lesions following prolonged stimulation by TSH (Ingbar and Woeber, 1981; see
Section VI. of this paper)
3. Reversibility of Morphological Progression to Thyroid Cancer
Several important questions arise concerning the progression of the different
morphological states towards thyroid cancer, particularly with respect to the
extent to which the progression is reversible. Thus, it is important to know
at what point (if any) and by what mechanism, the progression through hypertrophy,
hyperplasia, nodule formation, and neoplasia becomes irreversibly committed to
the formation of a malignant tumor. Undoubtedly, the final answer to these and
other questions will have to await a more thorough understanding of the molecular
biology of the complex events resulting in thyroid neoplasia (see Section
IV.C.). :
There is ample experimental evidence, however, showing that, to a significant
though unknown extent, the morphological progression towards thyroid malignancy
can be halted and at least partially reversed by removing the source of, and/or
correcting for, the excessive thyrotropic stimulation. This may be achieved
by administering adequate amounts of thyroid hormones to hypothyroid animals
(Purves, 1943: Bielschowsky, 1955; Furth, 1969; Paynter et al., 1986) or by
effecting surgical hypophysectomy (Astwood et al., 1943; Mackenzie and Mackenzie,
1943; Madler et al., 1970). Goiters in persons living in iodine-deficient areas
21
-------
tend to reverse following introduction of iodine in persons with hyperplasias
of short duration (Ingbar and Woeber, 1981; see Section VI. of this paper). In
each case, these procedures counter the effect of the source of TSH stimulation.
The extent to which morphological progression in the thyroid can be reversed,
however, clearly depends on the extent to which the process has progressed i.e.,
the severity and particularly the duration of the insult causing TSH stimulation.
On cessation of long-term goitrogen treatment or replacement of a long-term,
low-iodine diet with a high-iodine diet, the size and weight of the thyroid
typically decrease. If the pathological process has not progressed too far
(e.g., hyperplastic goiter) regression may be complete (Gorbman, 1947; Greer et
a!., 1967; Ingbar and Woeber, 1981). There is even one report that propylthi-
ouracil-induced cellular proliferation (including metastasis to the lung)
regressed, to normal when goitrogen administration to animals was stopped (Dunn,
1975). In the same study propylthiouracil-stimulated thyroid tissue transplanted
into other animals did not continue to proliferate and retain its tumorigenic
status unless the animals were treated with propylthiouracil. Others have
pointed out the need for ongoing TSH stimulation in the perpetuation of "hyper-
pi astic-neoplastic" thyroid lesions either-in the animals where the lesions
arose or in hosts receiving transplants of the material (Todd, 1986:
see Doniach, 1970b).
In contrast, little or no indication of morphological reversibility was
observed when rats that had received up to 500 ppm ethylene thiourea in their
diets for a period of 2 years were returned to a control diet (Graham et al.,
1973). In another study (Bielschowsky and Goodall, 1963) methyl thiouracil-induced
thyroid lesions in the mouse continued to progress after goitrogen administration
was stopped and replaced by thyroid hormone treatment. Most other studies
indicate varying degrees of reversibility following discontinuation of goitrogen
22
-------
administration (Arnold et a!., 1983; Wollman and Breitman, 1970; Wynford-Thomas
et al., 1982c) or return of animals from a low-iodine to a high-iodine diet
(Greer et al., 1967).
In humans it has been common practice to, use high doses of thyroid hormone
to try to suppress the growth of thyroid "nodules" and help differentiate non-
neoplastic from neoplastic growths (Rojeski and Gharib, 1985). The idea is
that preneoplastic lesions would regress upon cessation of TSH stimulation
brought about by the added hormone. Although variable success in reducing
nodule size has been noted in the past, a recent, carefully done study failed
to show any treatment-related reductions (see study and review, Gharib et al.,
1987). Thus the role of TSH in maintaining the size of human thyroid nodules
and their potential for reversal upon cessation of TSH stimulation requires
further investigation.
Typically, the reversal is marked by a reduction of thyroid gland size and
weight beginning a few days after removal of the TSH stimulus, and this is
associated with a loss of DMA indicating a decrease in the number of cells
present; some of this seems to be due to a reduction in the number of follicular
cells (Wollman and Breitman, 1970; Wynford-Thomas et al., 1982c). The mechanism
by which cells are lost from the thyroid may be cell death or migration.
Regression is associated with involution of the thyroid that involves a decrease
in vascular dilation, a marked diminution of follicular cell size and shape
(from columnar to cuboidal) and a return of follicular colloid material (Gorbman,
1947). These qualitative changes in thyroid histology almost always occur
following the removal of the TSH stimulus. However, if the goiter has been
present for several weeks, or months, the thyroid gland continues to remain at
least two to three times its normal size and weight despite a return to its
normal histological appearance (Greer et al., 1967: Wollman and Breitman, 1970;
Wynford-Thomas et al., 1982c).
23
-------
B. PITUITARY NEOPLASIA
Following chronic iodine deficiency (Axelrod and Leblond, 1955), treatment
with goitrogens (Sriesbach, 1941; 'Gries'bach et al.,, 1945) or surgical or Mil-
induced thyroi dectomy <{Doniach and Williams, 1962; Carl ton and Cries, IMS.},
the anterior pituitary frequently exhibits a loss of acidop'hiltc cells, an
increase In basophil cells, and develops swollen "thyroidectomy cells41 some ©f
which contain cytoplasmic granules. These cells contain TSH {Osamura and
Takayama, 1983) and, in the eyes of some researchers, may progress to rSB-
secreting adenomas .{-Furth et al., 1973; Bielschowsky, 19551, although ©tter
authors have failed to demonstrate tumors in such treated animalls {for instance,,
see Ohshima and Ward, 1984, 1986). Pituitary hyperplasia and neoplasia appear
to result from the same treatments causing thyroid neoplasia—conditions leading
to prolonged thyroid hormone decrease and excessive secretion of TSH by the
pituitary gland. ;
C. MOLECULAR CONSIDERATIONS IN THYROID CARC BIOGENESIS
Any hypothesis developed to explain the mechanism for carciinogenesis must
be consistent with what is known about the specific type of -cancer and the
physiological and biochemical system in which it develops. Animal experiments
have clearly shown that increased levels of TSH are associated with development
of thyroid hyperplasia and, later, with thyroid neoplasia. These end points,
hyperplasia and neoplasia, manifest two processes that are going on in the
thyroid: one is an increased commitment to cell division, which leads to
hyperplasia; the other is the transformation of normal cells into neoplast'ic
cells. Recent work at the cellular level indicates that induction of cell
division (which can lead to hyperplasia) and the transformation of normal to
altered {neoplastic) cells is the result of a complex interaction of different
24
-------
cell systems. For thyroid follicular carcinogenesis, it appears that TSH is a
major component in these interactions.
It is generally recognized that, under normal conditions, the control of
cell division requires the interaction of a number of endogenous factors which
work through a number of common pathways; exogenously added materials may also
have profound effects on this system. It seems there are at least two such
control steps centered in the pre-DNA synthetic part of the cell cycle, and TSH
is one of the factors operating there in thyroid cells. Certain protein growth
factors which operate through receptors on the cell surface are other stimuli
that influence cell division. In a similar manner, the transformation of
normal cells into an altered state with neoplastic potential also seems to be
dependent upon the interaction of different factors. TSH may also play an
active role here.
This section reviews available molecular information about the control of
cell growth in thyroid cells and their conversion to neoplastic cells, and
attempts to incorporate this information into a plausible mechanistic framework.
Figure 4 illustrates a not fully satisfactory, but hopefully instructive,
hypothetical model for the interaction of TSH and other factors in inducing
cell proliferation and transformation in the thyroid gland leading to neoplasia.
Although there are gaps in the understanding of the processes involved, what is
known about the thyroid is consistent with the existing understanding of the
components involved with the control of mammalian cell division. It is also
consistent with current thinking that carcinogenesis is a multistep process and
that multiple factors may influence its course. And finally, it accords special
weight to TSH as playing a significant role in cell proliferation and in
carcinogenesis of the thyroid gland.
25
-------
Thyroid Cell
TSH
Other factors
(EGF, phorbol esters)
Cel 1 ^__
Division
Neoplasta
'Cell
Transformation
Other influences
(mutation, oncogene
activation, growth
factors)
Figure 4. Hypothetical model for thyroid carcinogenesis.
26
-------
1 • Stimulation of Cell Division^
a. Influence of TSH--TSH interaction with its receptor on the surface of the
thyroid cell results in activation of adenyl cyclase and resultant product!an
of cAMP, the activation of the phosphatidylinositol pathway, commencement of
certain thyroid-specific differentiated functions that result in the formation
of the thyroid hormone and stimulation of cell division. Although all cultured
cells do not respond to TSH alone by increasing cell division (murine and
canine do; porcine, ovine, and human do not [see Saji et al., 1987]), the
following steps have been identified in those that do respond. Almost immediately
(within 15 to 30 minutes) after addition of TSH to, quiescent thyroid cells in
culture, there are marked increases in the levels of the mRNAs for the cellular
protooncogene, c-fos. A similar pattern is found for transcripts of the proto-
oncogene, c-myc, but the induction is delayed somewhat, with the peak occurring
at about 1 to 2 hr after TSH addition. These effects of TSH can be mimicked by
direct addition of cAMP analogs or other factors that increase cellular cAMP
(Dere et al., 1985; Tramontane et al., 1986a; Colletta et al., 1986). Interest-
ingly, human thyroid adenomas and carcinomas are characterized by c-myc expression,
which is not found in the surrounding normal thyroid tissue. In addition, like
normal cells in culture, adenoma cells respond to TSH in a dose-related manner
by increasing the levels of c-myc transcripts (Yamashita et al., 1986). This
finding in human cells is in contrast to that cited above (Saji et al., 1987).
The protein products of the c-fos and c-myc protooncogenes are thought to
play a role in the replication of cells. Both c-myc and c-fos code for proteins
that are largely restricted to the cell nucleus and appear to be functionally
linked to DNA synthesis. The latter is illustrated by experiments showing that
when monoclonal antibody to human c-myc protein is added to isolated nuclei,
there is an inhibition of DNA synthesis and replicative DNA polymerase activity;
27
-------
the inhibition can be overcome by the addition of excess c-myc protein (Studzinski
et a'l., 1986).
There is additional evidence to indicate that oncogene expression may be
an important factor in triggering cell division. For instance, Certain human
cancers have been shown to have chromosome rearrangements involving c-myc.
This relationship has been well established for cases of Burki'tt lymphoma fB-
cell cancer) (Taub et al., 1982; ar-Rushdi et al., 1983; Nishikura et al.,
1983) and to a lesser extent for certain T-cell leukemias lErikson et al,,
1986; Finger et al., 1986) It is thought that chromosomal translocations move
c-myc to the regulatory units of immune response genes in these cells and bring
about constitutive activation of the oncogene which then provides a continued
stimulus for cell proliferation (see review by Croce, 1986).
TSH also seems to affect to some extent the phosphatidylinpsitol pathway
within cells (Kasai and Field, 1982; Tanabe et al., 1984; Bone et al., 1986)
which is a major transduction system of signals across cell membranes (see
Nishizuka, 1986 and next section) as is the cAMP system. Just how this effect
of TSH may influence thyroid cell division has not yet been determined.
b. Other Factors—Experiments in a number of cell systems have identified
control points in the pre-DNA synthetic part of the cell cycle which must be
passed for cells to replicate DNA and go into cell division. For instance,
mammalian cells treated with one chemical stimulus (e.g., platelet-derived
i
growth factor which is known to stimulate c-myc) did not commence DNA synthesis
until other substances were added to the medium (Stiles et al., 1979; Smeland
et al., 1985). Current investigations on the interaction of various factors in
the control of cell division have been summarized ,by Goustin et al. (1986) and
i
Rozengurt (1986).
Work with thyroid cells also indicates that a number of growth factors and
28
-------
cell systems are operating which influence a cell's commitment to cell division.
For illustrative purposes, emphasis here will be placed on three of these:
epidermal growth factor, the protein kiriase c system (see Table 1), arid the
somatomedins.
Epidermal growth factor (EGF) is a naturally occurring polypeptide present
in a number of organs that binds to specific receptors on sensitive cells.
This binding results in activation of receptor-associated tyrosine kinase which
phosphorylates the EGF receptor and other sites and helps to bring about its
cellular action. EGF is present in adult tissues; a related growth factor,
transforming growth factor typeo(, is present in neoplasms and embryonic tissues
and may be an embryonic form of EGF. It is interesting to note that one of the
viral oncogenes, v-erbB, is a mutation of the EGF receptor gene where the
binding-site portion of the receptor has been deleted, and that this mutation
may result in constitutive activation resulting in continued cell proliferation
(Goustin et al., 1986).
There is some work that indicates that EGF plays a role in the regulation
of cellular activity and cell division in thyroid cells in culture. Its role
in vivo needs to be ascertained. Unlike TSH, EGF blocks certain differentiated
functions that typify thyroid action, such as formation of thyroglobulin by
thyroid cells in culture (Westermark et al., 1983; Bachrach et al., 1985; Roger
et al., 1986). In in vivo studies, infusion of sheep over a 24-hour period with
EGF resulted in a profound drop in serum T4 and T3 which started within 10 hours
after commencing administration. Part of this reduction in circulating thyroid
hormones appears to be due to their enhanced metabolism (Corcoran et al., 1986).
These authors cite other work which show that thyroid hormone administration
results in increased tissue levels and urinary excretion of EGF. It thus seems
that some feedback exists between levels of EGF and thyroid hormones.
29
-------
TABLE 1. EFFECTS OF STIMULI ON: THYROID CELLS
Stimulus Enzyme Induces
activity c-fos & c-myc
TSH adenyl +
cyclase
EGF tyrosine ?
kinase
TPAa protein ?
kinase c
Stimulates Effect 0n
cell differentiated Other
division functions
+ Enhances Enhances EGF
binding to
its receptor
+ Inhibits
+ Inhibits Inhibits EGF
binding to
its receptor
and tyrosine
kinase
activity
aTPA, 12-0-tetradecanoylphorbol 13-acetate, a phorbol ester.
-------
EGF also produces increases in thyroid cell division in thyroid cells.
By about one day after addition of EGF to thyroid cells in culture, there is
stimulation in DMA synthesis (Westermark et al., 1983; Roger et a!., 1986), as
was seen after administration of TSH. TSH increases the binding of EGF to its
receptor on thyroid cells and, in combination with EGF, enhances DMA synthesis
above that seen with EGF alone (Westermark et a!., 1986).
Another cell-surface related mechanism results in the activation of protein
kinase c. It is generally recognized that this system is one of the major
information-transferring mechanisms from extracellular to intracellular sites in
many cells throughout the body (see review by Nishizuka, 1986). Receptor
binding of a host of biologically active substances (e.g., hormones, neurotrans-
mitters) is followed by hydrolysis of inositol phospholipids along two paths:
one leads to calcium mobilization, the other to activation of protein kinase c.
The kinase transfers phosphate groups to various proteins which results in a
modulation of their action. Many studies have demonstrated that certain tumor
promoters in the two-stage mouse skin carcinogenesis model, including the
phorbol esters, can bind to cell receptors and activate protein kinase c (see
Nishizuka, 1986).
Phorbol esters, like EGF, inhibit differentiated thyroid cell functions
and stimulate cell division. As in other cells (Friedman et al., 1984), phorbol
esters increase protein kinase c activity and block EGF binding of its receptor
in thyroid cells (see Table 1) (Bachrach et al., 1985; Ginsberg and Murray,
1986; Roger et al., 1986). It is not known if EGF and phorbol esters stimulate
expression of the c-fos and c-myc protooncogenes in the thyroid, although there
is some evidence for this in mouse 3T3 cells (Kruijer et al., 1984; Muller et
al., 1984; Kaibuchi et al., 1986).
31
-------
A series of polypeptide substances related to Insulin and termed somatofiiedins
(insulin-like growth factors, IGFs), are known to exist which help to control
cell growth in numerous tissues (see Goustin et al.» 1986). Concentrations of
somatomedins in the blood are regulated by growth hormone* They are produced
by the liver and almost all organs of the body, seemingly the products of
mesenchymal cells (Han et al., 1987). Although they may or may not stimulate
DMA synthesis in cells when they are the only added factor, they frequently
interact significantly with other growth factors in bringing about cell division5
(Stiles et al., 1979),
In cultured rat thyroid cells very high concentrations of insulin alone
will induce cells to replicate BMA (Smith et al., 1986). It was hypothesized,
i
then demonstrated, that this effect was most likely due to cross- reactivity of
insulin with the somatomedin C CIGF-I) receptor (Tramontano et al^ 1986bv
1987: Saji et al., 1987). In rat thyroid cells TSH. and somatomedin C for
insulin) synergize in Inducing DNA synthesis, but are additive in regard to-
increasing cell growth (Tramontane etal.,, 1986b); such DWA-replication synergy
was not noted in porcine cells (Saji et al., 1987).
Although studies on thyroid cells indicate that TSH, EGF> phorbO'1 esters,
and somatomedin C (and insulin)) can each; stimulate cell division in cultured
thyroid cells, it does not mean, that these factors are the only ones. For
instance, many of the culture systems used in these studies ineTuded; serum,.
which, is known to have a number of growth factors in it. In other eases,
the culture medium was supplemented; with hormones, growth factors, and other
substances (e.g., somatostatrn, cortisol,, transferrin) which are known to
effect cell cycle traverse CBachrach et &l.y W8S;, Colletta et'al.,- 1986%
Wester-mark et al.,. 1983).
-------
c- Possible Controls in Thyroid Cell Division—As discussed earlier, It appears
that the control of cell division in mammalian cells is in the pre-DNA synthetic
portions of the cell cycle. By using combinations of substances, two control
points have been identified; both points must be passed for cells to commence
DMA replication. Although there are significant differences in response among
cell systems, factors that seem to affect the first regulatory point include
such things as platelet-derived growth factor and the c-fos and c-myc oncogenes,
whereas those operating at the second control point include somatomedin C, EGF,
and the c-ras oncogene (Stiles et al., 1979; Leof et al., 1982; see^Goustin et
al., 1986). Since TSH is also known to activate adenyl cyclase and c-fos and
c-myc expression in thyroid cells (Dere et al., 1985; Colletta et al., 1986;
Tramontano et al., 1986a), it seems possible that it may act at the first
control point. This is supported by the observation that combinations of TSH
with EGF or somatomedin C lead to enhanced DMA synthesis in thyroid cells (EGF
and somatomedin C are putative second control step agents) (Westermark et al.,
1986; Tramontano et al., 1986b, 1987).
The placement of the protein kinase c system in the control of thyroid
gland cell division is uncertain, since its effect on cell proliferation is not
enhanced by either TSH or EGF. As indicated previously, phorbol ester adminis-
tration to thyroid cells diminished EGF binding to its receptor (Bachrach et
al., 1985). It also appears that TSH itself may increase the phosphatidylinositol
pathway in addition to affecting cAMP (Bone et al., 1986). On the other hand,
the protein kinase c and adenyl cyclase systems often play complementary roles
in mammalian cells to enhance cell division and other functions (Nishizuka,
1986; Rozengurt, 1986). More information is needed in this area.
Insulin (and related substances) seem to play a facilitating role in the
thyroid. Alone in high concentrations it can induce thyroid cells in medium
33
-------
without serum to synthesize DNA, and it enables TSH to enhance this effect
t
(Wynford-Thomas et a!., 1986). Insulin is active at both control points in
certain mouse 3T3 cells as well (Rozengurt, 1986). ;
A model can be constructed for control of cell division in the thyroid
gland (Figure 5) that includes the two pre-DNA synthetic steps. The model
engenders the known effects of various factors on thyroid cells, and reflects
certain observations in other mammalian cell systems. Although the model is
not fully satisfactory, due to the inconsistencies across cell systems, it
depicts certain interactions that may exist in the thyroid gland and suggests
possible future research directions.
2. Gel1ular Transformation
As with the control of cell division, complex interactions among different
factors seem to be operating during the transformation of normal to altered
cells with neoplastic potential. Although'activation of a single oncogene is
not sufficient to produce transformation, activation of two different oncogenes
is a common means of transforming cells (see reviews by Weinberg, 1985; Barbacid,
1986). Frequently the cooperation includes an oncogene whose product is localized
to the nucleus (e.g., c-fos, c-myc) with one whose product is in the cytoplasm
(e.g., c-ras, c-src). As was mentioned previously, nuclear oncogenes can be
'* • '
activated by chromosomal translocation of the oncogene to cellular regulatory
sequences; other activation mechanisms include the insertion of viral regulatory
segments next to the nuclear oncogene, gene amplification (increase in the
number of copies of the oncogene per cell), and stabilization of the oncogene
gene product. On the other hand, cytoplasmic oncogenes tend to be activated by
point or chromosomal mutations which affect the structure of their gene products
(Weinberg, 1985).
-------
TSH
platelet-derived growth factor
insulin
Induction of adenyl cyclase
c-fos/c-myc
Pre-DNA
replication
DMA
replication
Cell
division
EGF
somatomedin C
c-ras
i nsul i n
Activation of
protein kinase c
Figure 5. Possible control points for cell division in the pre-DNA synthetic
portion of the cell cycle.
35
-------
TSH enhances c-fos and c-myc expression that may in turn interact with
other factors in bringing about cell transformation. If the stimulus for TSH
secretion from the pituitary is long-term, as in the case of continued exposure
to an antithyroid substance, it seems possible there could be continued oncogene
transcription and a continued emphasis on cell proliferation which could result
in hyperplasia. Still other stimuli (e.g., activation of a second oncogene,
certain poirit or structural mutations, interplay with growth factors) may aid
in the transformation process and bring about neoplasia.
This hypothesis is consistent with recent studies which indicate that
c-myc may be a necessary component in cellular transformation, but that it is
not sufficient in itself to bring about the condition. Studies of transgenic
mice support this conclusion (Adams et al., 1985; Langdon et a'l., 1986).
Combinations of the DMA of c-myc and the enhancer region of the Eu-immunoglobulin
locus were made and injected into fertilized mouse eggs which' were transplanted
into maternal hosts. The DMA became incorporated into the cells of the body of
the developing organism (transgenic recipients). Within a few months after
birth, almost all animals developed malignant B-cell lymphomas and died. It
seems that during development there is constitutive expression of c-myc with a
great expansion of multiple clones of B-cell precursors. However, only one
clone develops into a tumor, and this seems: to occur at variable times during
development. This has led the authors to propose that although c-myc expression
favors proliferation of B-cell precursors, some genetic event, like activation
of a second oncogene, may be required for transformation to malignancy.
Studies on the thyroid gland are consistent with the idea that c-myc
(through TSH stimulation) may interact with other stimuli in bringing about
cell transformation. For instance, an enhancement of the carcifnogenfc response
is noted when a treatment that increases TSH (e.g., iodide deficiency) follows
.'. • •• ' "•- 36 ...,'•• . , • •• . . ."".' <•<• "••'
-------
application of a genotoxic agent (e.g., irradiation, nitrosamine) (see Section
IV) which might produce a mutation that activates a second oncogene or some
other effect.
One is still faced, however, with the observation that treatments that
ensure prolonged TSH stimulation, as have been discussed previously, lead to
neoplasia. Three possibilities exist: (1) TSH simply enhances spontaneously
occurring events (e.g., mutations in regulatory sequences like oncogenes).
The finding of thyroid neoplasms in about 1 percent of some untreated laboratory
animals (Haseman et al., 1984) is in keeping with the idea that "spontaneous
mutations" might exist in control animals that might predispose animals for
development of thyroid tumors. (2) Through its effect on cell division, TSH
may expand the thyroid cell population at risk for a spontaneous event and then
promote neoplasia once a spontaneous mutation occurs. (3) TSH alone, via some
yet undisclosed mechanism, might produce cellular transformation.
37
-------
V. EXOGENOUS FACTORS INFLUENCING THYROID-PITUITARY CARC1N06ENESIS
The observations presented 1n the previous section demonstrated that
prolonged increases in TSH output are associated with thyroid cellular hypertrophy
and hyperplasia and, finally, with neoplasia in the absence of exogenously
added agents. This section summarizes known information on thyroid carcinogenesis
following application of exogenous stimuli. In the main, it, too, shows the
important role of chronic TSH stimulation in thyroid carcinogenesis. Information
on physical and chemical agents affecting thyroid-pituitary physiology and
carcinogenesis is summarized. Chemical classes associated with thyroid tumors
in the NCI/NTP animal studies are listed, and analyses are conducted on the
specific chemicals from those classes as to their antithyroid activity and
genotoxicity.
A. PHYSICAL FACTORS
External ionizing radiation is a known thyroid carcinogen in humans and
experimental animals (MAS, 1980). Internal radiation, following administration
of 131I (a{3 - and a y-radiation emitter) produces thyroid tumors in animals,
but the evidence in humans from the follow-up of treated Graves' disease patients
is less firmly established (MAS, 1980; NCRP, 1985; see Becker, 1984). A recent
paper purports the hypothesis that radioiodines may account for thyroid nodules
following the detonation of a hydrogen bonfo in the Marshall Islands in the
Pacific Ocean (Hamilton et al., 1987). Although irradiation can alter DNA and
induce mutation and, thus, influence thyroid carcinogenesis via genotoxic
mechanisms, others have speculated that the follicular cell damage induced by
irradiation may impair the gland's ability to produce thyroid hormone and,
thus, places the thyroid under conditions of long-term TSH stimulation.
38
-------
B. CHEMICAL FACTORS
1. Goitrogens
Early interest in naturally occurring chemicals causing thyroid enlargement
arose from observations that rabbits fed diets composed mainly of cabbage
leaves frequently developed goiters (Chesney et a!., 1928). Similar observations
were subsequently made with two purified synthetic chemicals (sulfaguanidine
and l-phenyl-2-thiourea) during nutritional/physiological studies with rats
(Mackenzie et al., 1941; Richter and Clisby, 1942). When it was realized that
the primary action of these and related compounds was to inhibit synthesis of
the thyroid hormones, their potential therapeutic value in hyperthyroidism
became evident.
a. Naturally-occurring (dietary) substances—These materials have been reviewed
in detail by VanEtten (1969). The early observations of goiters in rabbits
maintained on cabbage-leaf diets (Chesney et al., 1928) were followed by the
discovery that the seeds of rape and other brassica species (cabbage, brussels
sprouts, turnips, and mustard) also contained substance(s) that were goitrogenic
when incorporated into rat diets (Hercus arid Purves, 1936; Kennedy and Purves,
1941). Prolonged dietary exposure to rape seed led to the development of
adenomatous goiters (100 percent in 27 months) in rats (Griesbach et al.,
1945). L-5-Vinyl-2-thiooxazolidone {goitrin) has been identified as the active
goitrogen in turnips and the seed and green parts of other cruciferous plants.
Goitrin from these sources may be passed to humans, in the milk of cows feeding
on such plants. In humans, goitrin appears to be about as active as
propylthiouracil (Haynes and Murad, 1985). Peanuts are also reported to be
goitrogenic in rats (Srinivasan et al., 1957), the active component being the
glucoside, arachidoside.
39
-------
b. Synthetic compounds—Synthetic chemicals exhibiting goitrogenic activity may
be divided into three major structural groups: thionamides, aromatic amines*
and polyhydric phenols. The synthetic gbitrogens are discussed briefly below,
but have been extensively reviewed by Cooper (1984) and Paynter et al. (1986).
i1
(i) Thionamides. These include derivatives of thiourea and heterocyclic
compounds containing the thioureylene group. The latter includes most of the
compounds (e.g., propylthiouracil, methimazble, and carbimazole) used therapeu-
tical^ for hyper thy roidism in humans. Among the many chemicals in this group,
one nitrogen atom may be replaced by oxygen or sulfur; however, the thionamide
group is common to all. Other active compounds in this class are derivatives
of imidazole, oxazole, thiazole, thiadiazole, uracil, and barbituric acid. The
naturally occurring goitrin, present in cruciferous plants, also belongs to
this group of compounds.
(ii) Aromatic amines. Examples of compounds of this type are the sul fonamldes,
sulfathiazole, and sulfadiazine (Haynes and Murad, 1985). Optimal antithyroid
activity of this group of compounds is associated with a para-substituted
aminobenzene structure with or without aliphatic (e.g., methyl) substitution on
the ami no nitrogen. It is of interest that several methylene- and oxydianilines
(and alkyl substituted derivatives) have also been shown to possess goitrogenic
activity (Hayden et al., 1978) and like, the sulfonamides, to increase thyroid
neoplasms in rats (Weisburger et al., 1984).
(iii) Polyhydric phenols. The aritithyroid activity (hypothyroidism and goiter)
of resorcinol was first observed following the use of this material for treatment
of leg ulcers in humans (Haynes and Murad, 1985). Subsequent studies have
established that antithyroid activity is associated with compounds with meta-
polar-substituents on the benzene ring. Thus, hexyresorcinol, phloroglucinol,
2,4-dihydroxybenzoic acid, and meta-aminophenol are active, whereas catechol,
40
-------
hydroquinone, and pyrogallol are not (Paynter et al., 1986).
c. Modes of Action—Antithyroid agents belonging to structural groups i, ii, or
ill all exert their activity by direct interference with the synthesis of the
thyroid hormones in the thyroid gland. All appear to block the incorporation
of iodine into tyrosyl residues of thyroglobulin and by inhibiting the coupling
of the idotyrosyl residues into idothyronines. It was proposed by Taurog (1976)
that the antithyroid agents inhibit the enzyme peroxidase that is responsible
for the conversion of iodide to the iodinating species and the subsequent
iodination and coupling of the tyrosyl residues. This has been confirmed by
subsequent studies {Davidson et al., 1978; Engler et al., 1982) showing that
the compounds bind to and inactivate peroxidase when the heme of the enzyme is
in the oxidized state. It is likely that these compounds show some inhibitory
selectivity towards the different peroxidase-catalyzed reactions (i.e., iodination
vs. coupling) (Haynes and Murad, 1985K There is also evidence that some of
the compounds (e.g., propylthiouracil),inhibit the peripheral deiodination of
T4 to T3 (Geffner et al., 1975; Saberi et al., 1975).
Because of their ability to inhibit thyroid hormone synthesis, all of the
above compounds have the potential to reduce circulating levels of T4 and TS
and, consequently, to induce the secretion of TSH by the pituitary. As a
result, prolonged exposure to such compounds can be expected to induce thyroid
gland hypertrophy and hyperplasia and ultimately may lead to neoplasia.
2. Enzyme inducers
• • \
In addition to chemicals exerting effects directly at the thyroid, as was
summarized in the previous section, a nunfoer of others acting at peripheral
sites can cause equally profound disturbances in thyroid function and morphology.
Of particular interest are those compounds that induce hepatic and/or extrahepatic
enzymes responsible for the metabolism of many endogenous and exogenous compounds.
41 :
-------
These chemicals can increase the metabolism of thyroid hormone, result in a
reduction in circulating thyroid hormone, and stimulate an increase in TSH.
Following long-term exposure to these agents, the thyroid gland undergoes
hypertrophy and hyperplasia and finally, neoplasia.
a. Foreign compound metabolism and enzyme induction—
i. General. The enzymes responsible for the metabolism of foreign compounds
constitute a remarkably diverse group of proteins that catalyze a variety of
. '
reactions associated with either the primary {Phase I) metabolic attack on a
chemical (oxidation, reduction, hydrolysis) or with its subsequent secondary
(Phase II) metabolism (e.g., conjugation with glucuronide, sul fate, ami no
acids, and glutathione) (Testa and Jenner, 1976). The enzymes are associated
with the endoplasmic reticulum or cytosol of the liver and a nunber of extrahep-
atic tissues. The enzymes serve an important functional role in increasing the
polarity, water-solubility, and excretability of the vast majority of fat-
soluble foreign compounds and often result in a decrease in their biological
activity or toxicity. Because of the latter, they are frequently referred to
as detoxication enzymes (Wilkinson, 1984).
ii. Induction. Enzyme induction refers to the phenomenon whereby exposure of
an animal to a given foreign compound results in the enhanced activity through
de novo synthesis of a spectrum of the enzymes involved in Phase I and Phase II
metabolism (Conney, 1967). Induction typically results in am increase in the
rate at which the inducer and other compounds are metabolized and excreted.
Since the enzymes responsible for foreign-compound metabolism are thought
by many to have evolved as a biochemical defense against potentially harmful
environmental chemicals (Wilkinson, 1984), induction may be viewed as a biological
adaptation that can provide important short-term benefits for survival. On the
other hand, in the light of increasing evidence that the enzymes detoxifying
42
-------
one chemical may activate another (Cummings and Prough, 1983), there has been
concern that enzyme induction may represent a mechanism through which potentially
dangerous toxicological interactions can occur following chemical exposure.
Another cause for some concern is that several of the enzymes that participate
in foreign-compound metabolism are also known to play important roles in the
metabolism of physiologically important endogenous chemicals such as hormones.
Clearly, any changes in the levels of enzymes responsible for the synthesis or
breakdown of such compounds could lead to physiological imbalances with potentially
serious consequences (Conney, 1967).
iii. Different inducer types. Inducers of the enzymes involved in foreign-
compound metabolism have been divided into at least two different categories on
the basis of their characteristic effects on cytochrome P-450 and monooxygenase
activity (Mannering, 1971; Lu and West, 1978; Ryan et a!., 1978; Lu and West,
1980). One of these, typified by phenobarbital, led to a significant increase
in liver size and weight and caused the substantial proliferation of hepatic
endoplasmic reticulum. Induction was associated with increases in cytochrome
P-450 and a large number of monooxygenase reactions that enhanced metabolic
(oxidative) capability towards many foreign compounds. The spectrum of oxidative
reactions induced is now known to result mainly from the induction of one major
isozyme of cytochrome P-450 that, in rats, is referred to as cytochrome P-450b
(Ryan et a!., 1978). A large number of drugs and other foreign compounds
including the chlorinated hydrocarbon insecticides (DDT and its analogues and
the cyclodienes like chlordane and aldrin) exhibit induction characteristics
similar to phenobarbital and are generally referred to as "PB-type" inducers.
Early studies with the polycyclic hydrocarbon, 3-methyl cholanthrene
(3MC), clearly indicated that the induction characteristics of this compound
were quite distinct from those of PB (Mannering, 1971). In contrast to the
43
-------
latter, treatment of animals with SMC did not cause large increases in liver
size or in the proliferation of endoplasmic reticulum; neither did it result in
large increases in cytochrome P-450. Instead, SMC resulted in the formation of
a qualitatively different form of cytochrome P-450, known generally as cytochrome
P-448 and now referred to in rats as cytochrome P-450c (Mannering, 1971; Lu and
West, 1978; Ryan et al., 1978). This cytochrome is associated with a rather
limited number of oxidative reactions, the best known of which is aryl hydrocarbon
hydroxylase (AHH) (Ryan et al., 1978; Eisen et al., 1983; Conney, 1982). AHH
has received a lot of attention in recent years because of its role in the
metabolic activation of compounds like benzo[ajpyrene to potent carcinogens
(Eisen et al., 1983; Conney, 1982). Inducers of the "3MC-type" include a
number of polycyclic aromatic hydrocarbons, naphthoflavone, and several halogen-
ated dibenzo-£-dioxins; 2,3,7,8-tetrachlorodibenzo-p_~dioxin (TCOD) is the most
effective inducer of this type to be discovered (Poland and Glover, 1974). The
mechanism of action of inducers of this type involves high affinity binding to
a cytosolic receptor and subsequent migration of the inducer-receptor complex
to the nucleus where the transcript!onal effect leading to enhanced protein
synthesis is initiated (Eisen et al., 1983). Induction of this type is genetically
controlled by the so-called Ah locus in rodents and» while the true identity of
the cytosolic receptor remains unknown, it is hypothesized to be a receptor for
some hormone or other physiologically important ligand.
While the "PB-type" and "3MC-type" inducers still constitute the two major
categories of inducers, it is now recognized that a number of other types
exists, each characterized by increased levels of a distinct spectrum of isozymes
of cytochrome P-450 and other enzymes. It is also apparent that a number of
compounds share some of the characteristics of more than one group and cannot
be strictly classified. Technical mixtures of polyhalogenated biphenyls (PCBs
! •.•• -•'•'' 44 •:-'".•• •'•' • ••' ., •.' ' .
-------
and PBBs), for example, exhibit characteristics of both PB- and 3MC-type inducers
(Alvares et al., 1973), probably due to the presence in the mixtures of a
nunber of isomers representing each type.
In addition to inducing a characteristic spectrum of isozymic forms of
cytochrome P-450, many of the inducers also result in enhanced titers and
activities of other enzymes involved in foreign-compound metabolism. While
these have not been well documented, they include epoxide hydratases, glutathione
(GSH)-S-transferases and several of the transferases (UDP-trarisferases, sul fo-
transferases) associated with secondary conjugation reactions (Jacobsen et al .,
1975; Lucier et al., 1975; Ecobichon and Comeau, 1974). It has been suggested
that, like cytochrome P-450j, these enzymes may also exist in multiple isozymic
forms and that different inducers may enhance the activity of specific isozymes
with a characteristic range of substrate specificities.
b. Metabol ism of thyroid hormones—The liver not only constitutes a target
tissue for the thyroid hormones but is also an organ responsible for the metabolic
inactivation of the hormones and their elimination from the body. About half
the 14 elimination from the body of the rat occurs via the bile, whereas in
humans only about 10 to 15 percent is lost in this way (Oppenheimer, 1987).
While there appear to be quantitative differences in the relative rates of
elimination of 14 and 13, it is probable that both are excreted by a qualitatively
similar mechanism. The major pathway of elimination involves conjugation of
the phenolic hydroxyl group of T4 with glucuronic acid and biliary excretion of
the resulting glucuronide (Figure 1) (Gal ton, 1968; Bastomsky, 1973); sul fate
conjugates may also be produced and excreted. On entering the intestine a
portion of the conjugate may undergo hydrolysis by intestinal bacteria to
release free thyroid hormone that may be reabsorbed into the circulation; this
process is referred to as enterohepatic circulation. Unhydrolyzed conjugate
45
-------
cannot be reabsorbed and is excreted In the feces (Houk, 1980).
c. Effect of inducers on thyroid function and morphology--'
(i) PB-type inducers. Initial reports on the goitrogenic effects of a number
of PB-type inducers in both birds and rodents began to appear in the iitid- to late
1960s. Modest to substantial increases in thyroid weight were reported in fats
treated with phenobarbital (Japundzic, 1969; Oppenheimer et al., 1968) and
isomers of ODD (Fregly et al., 1968), in pigeons treated with J^'-DDE (Jefferies
and French, 1969), j>,jj'-DDE or dieldrin (Jefferies and French, 1972) and in
bobwhite quail exposed to jj.jj'-DDT or toxaphene (Hurst et al., 1974). Chtofdane,
another chlorinated hydrocarbon, enhanced thyroid function and caused hepatic
accumulation of 125I-T4 in rats (Oppenheimer et al., 1968). Histological
examination of the thyroids of treated animals typically showed a reduction in
/ '
follicular colloidal material and increased cellular basophilia and hyperplasia
(Fregly et al., 1968; Jefferies and French, 1972), and it was noted by several
workers that these changes were similar to those occurring in response to
increased circulating levels of TSH. Support for the effect being a response
to increased TSH, rather than a direct effect on the thyroid, is found in
studies demonstrating that the goitrogenic response of the thyroid to phenobarbital
could be prevented by hypophysectomy or the administration of 14 [Japundzic, 1969-K.
The effects of PB-type inducers on thyroid function, are now known to be
qufte complex and to involve a number of factors relating to the distribution^
tissue binding, metabolism, and excretion of thyroid hormones. Animals treated
with phenobarbital show increased hepatocellirlar binding of T^ combined with
enhanced biliary excretion of the hormone (Oppenheimer et al., 1968; 1971). In
fntact rats, these changes sinvply result from- an increaserf rate of turnover of
T4 thrat is compensated by release of TSH and enhanced thyroidal secretiort of
46
-------
new hormone. As a result, no change in serum protein-bound iodine (PBI) is
observed following treatment with phenobarbital (Oppenheimer et al., 1968). In
thyroidectomized rats, however, phenobarbital reduces serum PBI and also reduces
the hormonal effects of administered T4 (Oppenheimer et al., 1968; 1971) The
ability of phenobarbital to reduce circulating levels of exogenously supplied
T4 in a human hypothyroid patient has been reported. The major factors leading
to enhanced turnover of T4 in animals treated with PB-type indiicers seem to be
increased hepatocellular binding due mainly to proliferation of the endoplasmic
reticulum (Schwartz et al., 1969) and a modest increase in bile flow that
enhances the overall rate of biliary clearance (Oppenheimer et al., 1968).
Phenobarbital (Oppenheimer et al., 1968) and DDT (Bastomsky, 1974) cause only
minimal increase in biliary T4 excretion, and in rats treated with ODD isomers,
fecal excretion of 131I-T4 was not observed until 24hr. after hormone treatment
(Fregly et al ., 1968). While DDT slightly enhanced the proportion of biliary
I present as T4-gl ucuronide, neither PB nor DDT (Bastomsky, 1974) are reported
to have significant effects on the rate of glucuronidation of T4.
Several studies have been conducted on the effects of PB-type inducers on
thyroid hormone status in healthy human volunteers or in patients on different
drug regimens. Drugs studied include phenobarbital, carb amaze pine, rifampicin,
and phenytoin (diphenylhydantoin). Most of the studies report decreased serum
levels of T4 (both protein-bound and free) (Rootwelt et al., 1978; Faber et
al., 1985; Ohnhaus and Studer, 1983), but reports vary on the changes observed
in serum levels of T3 and rT3 depending on the type and concentration of the
inducer employed. Ohnhaus and Studer (1983) observed a relationship between
increasing levels of microsomal enzyme induction and decreasing serum levels of
T4 and rT3 in healthy volunteers treated with combinations of antipyrine and
rifampicin. An effect was only observed, however, at induction levels that
47
-------
decreased the half-life of antlpyrine by more than 60 percent. Induction of
hepatic enzymes is apparently only one of several mechanisms through which
diphenylhydantoin can reduce circulating levels of 14 (Smith and Surks, 1984).
Other possible mechanisms by which diphenylhydantoin might act include serum
protein displacement of the thyroid hormones, effects on the binding and biological
activity of 13, and even effects on hypothalamic and pituitary regulation of
TSH. Despite significantly decreased serum levels of 14, there seem few reports
of humans being placed in a hypothyroid condition as a result of treatment with
drugs that induce liver microsomal enzyme activity. An exception is the observation
that persons being maintained on exogenously supplied thyroid hormone become
hypothyroid when given diphenylhydantoin or phenobarbital unless their thyroid
hormone doses are changed (Oppenheimer, 1987). Furthermore, TSH levels never
change significantly from those observed in the controls.
(ii) 3MC-type inducers. The effects on the thyroid of 3MC-type hepatic enzyme
inducers (polycyclic aromatic hydrocarbons, TCDD, etc.) are perhaps the best
understood of the compounds under discussion. A major mechanism involved seems
to be the induction of the T4-UDP-glucuronyl transferase that constitutes the
rate-limiting step in the biliary excretion of T4 (Bastomsky, 197^). The effect
is particularly well illustrated with reference to a variety of thyroid hormone
parameters 9 days after treatment of rats with a single dose of 25 ug/kg TCDD
(Bastomsky, 1977a). Biliary excretion of 125I (during the first hour after
injection of 125I-T4 and the biliary clearance rate of plasma 125I-T4 were
increased about 10-fold. Somewhat unexpectedly, the biliary excretion of T3 was
unaffected by TCDD. As a direct consequence of these changes in metabolism and
excretion, serum T4 concentrations (but not those of 7$) were reduced to half
those in controls. Other workers have reported decreased serum T4 concentra-
tions following TCDD treatment (Potter et a!., 1983; Pazdernik and Rozman, 1985;
48
-------
Rozman et al., 1985). TCDD treatment also elevated serum concentrations of
TSH and, as a result, produced thyroid goiters (measured by elevated thyroid
weight) and enhanced 131I uptake by the thyroid. There are conflicting reports
as to whether TCDD enhances bile flow (Bastomsky, 1977a; Hwang, 1973) but this
does not seem to be a major factor in its goitrogenic action.
While TCDD is an unusually potent inducer of UDP-glucuronyl transferases,
it appears to be at least somewhat similar to compounds such as SMC (Bastomsky
and Papapetrou, 1973; Newman et al., 1971), 3,4-benzo[a]pyrene (Goldstein and
Taurog, 1968), and the polychlorinated and polybrominated biphenyls (PCBs and
PBBs) (see below) all of which have been shown to enhance the biliary excretion
of T4 at least partly by increasing the formation of T4-glucuronide. TCDD did
not uniformly increase hepatic UDP-gl ucuronyl transferase activity towards all
substrates; it enhanced activity towards j>-nitrophenol but not towards testosterone
or estrone. Its effect on the T4-transferase does not seem to have been
investigated.
Recentlys some investigators- have suggested that the explanation for the
interactions of TCDD with thyroid hormone levels is that T4 and TCDD have
common molecular reactivity properties that might allow them to react with the
same receptors (McKinney et al., 1985a, b). Indeed, McKinney and his co-workers
consider that many of the toxic effects of TCDD result directly from its action
as a thyroxine agonist. This theory contrasts with the views of Poland's group
(Poland and Knutsen, 1982) that TCDD toxicity segregates with the Ah locus and
involves TCDD binding to the cytosolic receptor. Moreover, McKinney's views
are not consistent with recent experimental results (Potter et al., 1986), and
the entire area requires more research attention.
(iii) Mixed-type. Perhaps as a result of their widespread contamination of
the environment and their wel1 documented occurrence in human foods, the
49
-------
toxicological properties of PCBs and PBBs have received considerable attention
(Kimbrough, 1974).
Daily feeding of commercial mixtures of PCB (Arochlors)or PBB (Firemaster)
to rats (5, 50, and 500 ppm) led to striking dose- and time-dependent histologicat
changes in thyroid follicular cells (Collins et al., 1977; Kasza et al., 1978).
These changes included increased vacuolization and accumulation of colloid
droplets and abnormal lysosomes with strong acid phosphatase activity in follicle
cells. Microvilli on the lumen surface became fewer in number, shortened and
irregularly branched, and Golgi bodies were smaller; at higher exposures
mitochondria were swollen with disrupted cristae. It has been suggested that
the combined presence of an abnormally Targe number of colloid droplets and
lysosomes in the follicle cells might indicate interference with the normal
synthesis and/or secretion of thyroid hormones (e.g., cleavage of active thyroxine
from thyroglobulin). PBB has been found to accumulate preferentially in the
thyroid following 20 days of treatment and was still present 5 months after
administration (Allen-Rowlands et al., 1981). Sequestrationr of PBB in the
thyroid might indicate binding to thyroidal macromolecules, and, it has been
• * '
suggested that PBB might interfere with the organisation of iodide by peroxi-
dase. More work, in this area is needed.
Instead of comprising a single layer of cuboidal or low columnar epithelium,
the follicular cells of PCB-treated animals became more columnar with multiple
layers and hyperplastic papillary extensions into the colloid. Similar foliicuHr
cell hyperplasia has been reported in other chronic (Morris et al., 1975) and
subchronic studies (Sleight et al., 1978) with PBBs. The histologieal changes,
which are similar to those observed in animals treated with TSH (SeljeTd et al,,
1971),. were accompanied by substantially decreased (>three-fold): serum thyroxine
levels in PCB-treated rats (Collins et al., 1977). Residual effects were
50 • ; •'.•• ' •
-------
observed 12 weeks after termination of exposure, probably reflecting the persis-
tent nature of the PCBs. However, it is important to note that, even in animals
exposed to the highest doses of RGBs, both the histological and functional
abnormalities were reversible and were minimal 35 weeks after cessation of
treatment.
* • ' • •
The search for a mechanistic explanation of PCB- or PBB-induced thyroid
" hyperplasia has focused on the biochemical events occurring on exposure to
these compounds. Direct effects on the thyroid cannot be discounted, and
recent evidence suggests that disturbances in thyroid hormone synthesis and
distribution may occur following long-term administration (Byrne et al., 1987).
More work is needed in this area. However, most attention has been given to
peripheral effects that modify the distribution, metabolism, and excretion of
thyroid hormones and as a consequence may cause thyroid hyperplasia indirectly
through activation of the normal feedback mechanism involving TSH. Thyroid
parameters changed following short-term oral or cutaneous administration of
PCBs to rats have been extensively studied by Bastomsky and co-workers (Bastomsky,
1974, 1977b; Bastomsky and Murthy, 1976; Bastomsky et al., 1976) and include:
(a) Increased biliary excretion (about five fold) and bile:plasma ratio
(about 12-fold) following injection of 125I-T4.
(b) Increased biliary clearance rate of plasma 125I-T4 more than 20-fold.
* (c) Modest increase in bile flow (less than two fold).
. (d) Decreased total serum and free T4 concentrations.
(e) Increased 131i uptake by thyroid.
It is apparent from these data that PCBs have effects that are similar to
both "PB-type" and "3MC-type" inducers. PCBs are reported to be potent inducers
of liver T4-UDP-glucuronyl transferase (Bastomsky and Murthy, 1976) and, as
with the "3MC-type" inducers such as TCDD, this undoubtedly accounts, at least
51
-------
partially, for the increased biliary excretion of 14. On the other hand,
also displaced the thyroid hormones from their binding proteins in the serum
(Bastomsky, 1974; Bastomsky et al., 1976), an effect usually associated more
with "PB-type" compounds. Because of its PB-like activity, it is also possible
that PCB enhances hepatic binding of 14. It may be a combination of the induction
of T4-UDP-glucuronyl transferase and the displacement from serum binding proteins
that lead to such high bile:plasma ratios of 14 following PCB treatment; much
smaller 14 bileiplasma ratios are observed with compounds like salieylate that
effect displacement but not enzyme induction (Osorio and Myant, 1963), Conversely,
the effects of changes in binding proteins on metabolism of thyroid hormone
under steady-state conditions do not seem to have been studied, and at least
some arguments can be mounted that would suggest that no change in metabolism
would occur under those conditions.
PCBs are reportedly quite specific in their ability to selectively induce
different isozymes of UDP-glucuronyl transferase. Thus, in addition to inducing
the glucuronidation of 14, the PCB-induced isozyme(s) will also enhance activity
towards £-nitrophenol (Ecobichon and Comeau, 1974) and 4-meth.ylumbelliferone
(Grote et al., 1975); PCB did not enhance the glucuronidation of bilirubin,
however (Bastomsky et al., 1975).
The effects of PCB treatment on circulating levels of Tg are clearly
different from those of T4- It has been suggested that since T3 is more active
than T4 and because it is generated peripherally by 5'-monodeiodination of T4,
T4 may be serving simply as a prohormone. It is now generally accepted, however,
that T4 does have intrinsic hormonal activity. It is of considerable interest to
note that, in contrast to the case with T4, treatment of rats with PCB does not
result in any marked change in total serum or free concentrations of ^3, While
this may result from a number of different factors (Bastomsky et al., 1976),
52
-------
no completely satisfactory explanation has yet been proposed. There is some
suggestion that the relatively constant circulating levels of 13 might be due
to enhanced thyroidal secretion and enhanced peripheral conversion of 14 or 13
in response to the PCB-induced hypothyroidism.
In summary, in addition to possible direct effects on the thyroid, mixed-type
inducers such as the PCBs and PBBs have several effects that, either alone or
in combination, reduce circulating levels of the thyroid hormones and cause the
pituitary to release TSH. These are:
(a) Induction of T4-UDP-glucuronyl transferase,
(b) Displacement of T4 from serum proteins, and
(c) Increase in bile flow.
3. Other chemicals and treatment combinations
In addition to those chemicals that act directly upon the thyroid gland to
inhibit the synthesis of thyroid hormone or act distal to that site to enhance
thyroid hormone metabolism and removal from the body (see Section VLB. for some
other agents active in humans), there is a small group of compounds that have
produced thyroid tumors in experimental animals that do not share these
characteristics. Also, several investigations have indicated that combined-
treatment regimens are associated with thyroid carcinogenic responses in excess
of that produced by either single treatment alone.
a. Other chemicals—A few compounds have been identified that produce thyroid
tumors that are not known to influence thyroid-pituitary status {see Hiasa et
al., 1982), two of which are jJ-nitroso compounds. Rats given eight injections
of J^-methyl-N-nitrosourea (NMU) over a 4-week period developed thyroid tumors by
week 36 without any development of goiter (Tsuda et al.s 1983). Likewise,
there was no evidence of diffuse follicular hyperplasia in rats given a single
dose of NMU and observed at 33 weeks, even though some animals had thyroid
53
-------
neoplasms (Ohshima and Ward, 1984). In a similar way, N-bis(2~hydroxypropyl)-
nitrosanrine (DHPN) administration for 8 weeks led to thyroid tumors by 20 weeks
without any increase in thyroid weight (Hiasa et al., 1982); this observation
was confirmed in a second laboratory (Kitahori et al., 1984). Both nitrosamines
produce tumors at sites other than the thyroid.
The nitrosamines are a notorious group of compounds as to their potential
to produce carcinogenic effects in multiple species following metabolism to
reactive intermediates. Many are genotoxic in multiple test systems for different
end effects-.
b. Combined-Treatment Studies—Although goitrogenic stimuli that increase TSH
levels (e.g., amitrole, phenobarbital, iodine deficiency) are known to induce
thyroid hyperplasia and neoplasia alone, many experiments have demonstrated an
enhancement of the neoplastic response when these treatments are combined with
other exposures. Thus, when animals are first exposed to genotoxic physical
agents (i.e., 131j or X-rays) or chemical substances (e.g., certain nitroso
compounds, 2-acetylaminofluorene) followed by a goitrogenic stimulus, carcinogenic
responses (e.g., incidence of tumor-bearing animals, multiplicity of tumors per
animal, incidence of malignancies, and tumor latency) are greater than following
single treatments alone (see Appendix A).
Some have likened this response in the thyroid to the initiation-promotion
(two-step) phenomena originally described for mouse skin. In that case, treatment
with the first agent (initiator) confers a permanent change in cells, such that
exposure (usually prolonged) to the second agent (promoter) results in neoplasms;
reversal of treatments is ineffective as to tumor production. Over time it has
become generally recognized that carcinogenesis is a multistep process that
usually includes an initiation step as well as a promotional phase (OSTP, 1985),
The thyroid combined-treatment studies are consistent with the concepts of
54
-------
initiation-promotion. The genotoxic agent might permanently alter the thyroid
cell so that its accentuated growth under a goitrogenic stimulus would result
in neoplasms. Also consistent with this notion is the finding that the effect
of the initial treatment in the thyroid is long-lived. Rats can be treated
with 4-methyl-2-thiouracil (promoter) after intervals of time at least up to
18 weeks after exposure to 2-acetyl-aminofluorene (initiator) and still go on
to show an enhanced neoplastic response (Hall, 1948). On the other hand,
protocols employing treatment with the "promoter" before the "initiator" have
not been conducted for the thyroid. Thus, the correspondence of effects in the
thyroid to those in the classical two-stage model are not established, (although
they are testable).
c. Summary--Both physical and chemical agents have been implicated in thyroid
carcinogenesis. Ionizing radiation remains the only confirmed carcinogenic
agent for the human thyroid, an observation corroborated in experimental
animals. Laboratory research has demonstrated that many substances can directly
interfere with the synthesis of thyroid hormone (e.g., certain inorganic
substances, thionamides, aromatic amines). Under conditions of reduced thyroid
hormone levels, the pituitary increases TSH stimulation of the thyroid, which
leads to a predictable set of responses including cellular hypertrophy and
hyperplasia, nodular hyperplasia, and, finally, neoplasia. Pituitary tumors
are also sometimes increased, seemingly due to the increased pituitary stimulation
resulting from lowered circulating thyroid hormone levels.
Direct thyroidal effect is not the only way chemicals produce reductions
in circulating thyroid hormone. Enzyme inducers increase the removal of thyroid
hormone from the blood which, in turn, results in stimulation of the pituitary
gland to secrete more TSH. The result, again, of long-term exposure is hypertrophy,
hyperplasia, and eventually neoplasia. Only a limited number of chemicals have
55
-------
produced thyroid follicular tumors In animals in the absence of some
antithyroid effect.
C. STRUCTURE-ACTIVITY RELATIONSHIPS
1. Chemicals producing thyroid neoplasms in animals
One means of testing hypotheses concerning the mechanism of follicular celt
thyroid carcinogenesis is to review those chemicals known to produce such
neoplasms in experimental animals. The NCI/NTP data base is a valuable source
of information because it consists of about 300 chemicals that have been subject
to a somewhat standard protocol in certain strains of rats and mice. Although
about half the chemicals tested have shown neoplastic effects at one or more
anatomical sites, only 21 chemicals have been associated with the development
of follicular cell neoplasms of the thyroid (Table 2).
These 21 compounds were not representative of the spectrum of classes of
chemicals that were tested in the bioassays. Instead there was an overabundance
of chemicals in structural classes that are known to influence thyroid hormone
status. Over half of them (13 of 21) are either thionamides (3) or aromatic
amines (10), two chemical classes that have often been linked with antithyroid
activity primarily due to peroxidase inhibition. The bulk of the remaining,
chemicals (7 of 21) are complex halogenated hydrocarbons; members of this class
are often inducers of microsomal enzymes, and at least some are known to increase
the clearance of thyroid hormone from the blood. The remaining chemical, an
organophosphorous compound, is not from a group typically linked to effects on
the thyroid. Thus, in 20 of 21 instances, there is some basis to think that
thyroid neoplasms may be related to a reduction in thyroid hormone with concomitant
increase in pituitary stimulation of the thyroid through TSH.
Although most compounds producing thyroid neoplasms are members of specific
chemical classes, not all members of those groups have been shown to produce
56=
-------
TABLE 2. CHEMICALS IN THE NCI/NTP BIOASSAY PROGRAM SHOWING AT LEAST SOME
EVIDENCE OF THYROID FOLLICULAR CELL NEOPLASIA '
1. Thionamides
N,N'-dicyclohexylthiourea
F.iP -diethyl thiourea
trTmethy1th i o u rea
2. Aromatic Amines
a. Single ring
3-amino-4-ethoxyacetam"Iide
o-anisi.dine hydrochloride
2,4-diaminoanisole sulfate
HC Blue No. 1
b. Bridged double rings
4,4'-methylenebis(N,N-dimethyl)benzenamine
4,4'-methylenedianTlTne dihydrochloride
4,4'-oxydianiline
4,4'-thiodianiline
c. Miscellaneous
C.I. Basic Red 9 rnonochloride
1,5-naphthalenediamine
3. Complex Halogenated Hydrocarbons
aldrin
chlordane
chlorinated paraffins (Ci2» 60% chlorine)
decabromodiphenyl oxide
2,3,7,8-tetrachl orodi benzo-p_-di oxi n
tetrachlorodiphenylethane (£>£'-ODD)
toxaphene
4. Organophosphous Compounds
azinphosmethyl
57
-------
such tumors. For instance, among the thionamides tested by NCI/NTP,
N,N'-dicyclothiourea, MT-diethylthiourea and trimethylthiourea yielded
positive thyroid effects whereas several others did not (see Table 3).
It, therefore, seems reasonable to postulate that while a thionamide
structure increases the chance that a chemical will produce thyroid tumors in
long-term animal tests, structure alone is not sufficient in itself to generate
such activity. The same is true for certain aromatic amines (see Section V.C.2,b.).
2. Antithyroid activity and thyroid carcinogenesis
Given that many of the chemicals producing thyroid tumors in the NCI/NTP
series come from chemical classes known to produce antithyroid effects by
inhibition of thyroid peroxidase., a review was made of specific thionamides and
aromatic amines to see if antithyroid activity was a prerequisite for thyroid
carcinogenic activity. The hypothesis was borne out for the thionamides and at
least some'of the aromatic amines.
Generally, the criteria for selecting the specific chemicals required that
they had been (1) tested for animal carcinogenicity (NCI/NTP or I.ARC review),
and (2) evaluated for antithyroid activity. However, in some cases a chemical
had been studied for carcinogenicity, but not antithyroid activity. In those
cases, structurally related compounds that had been tested for antithyroid
activity were chosen to act as surrogate indicators of a compound's antithyroid
potential.
Antithyroid activity has been measured for a number of chemicals in rats
and, to some extent, in humans. For rats, chemicals were administered orally
at different doses for 10 days. Iodine concentrations in the thyroid were
measured, and from the dose-response curve the dose that reduced the iodine
concentration to a standard level was estimated (EDc). For comparison, the
dose of thiouracil (a well-studied antithyroid agent) that reduced iodine
58
-------
TABLE 3. THIONAMIDES NEGATIVE FOR THYROID NEOPLAS1A IN NCI/NTP STUDIES
1. 2,5-dithiiobiurea
s = c
,NH2
XNH
' NH
s = c;
'NH2
5. sultanate
S = C
•S-CHe-C = CH2
Cl
2. 1ead dimethyldi thiocarbamate
/N(CH3)2
s = c
3. 1-phenyl-2-thiourea
S = C
X
NH2
6. tetraethylthiuram di sulf1de
s -
s = c:
,N(C2H5)2
N(C2H5)2
4. sodium diethyl dithiocarbamate
S = C.
•N(C2H5)2
Na"1
59
-------
concentration to the same level was also estimated (EDt). "Antithyroid activitiy
was expressed as the ratio of the estimated dose of thiouracil relative to that
for the chemical (EDt/EDc), where thiouracil (in this review) is given a value
of 100 (Astwood et al., 1945; McGinty and Bywater, 1945a, b).
For humans, antithyroid activity for a chemical was again measured against '
the effects of thiouracil (value = 100 for this review) (Stanley & Astwood,
1947). Subjects were given 1311 by mouth, and iodine in the thyroid was monitored
externally by Geiger-Muller measurement. After 1 to 2 hours, the chemical was
given orally, and the influence of the agent on the further time-course uptake
of radioactivity into the gland was evaluated. The degree to which accumulation
was affected was graded depending upon the completeness and duration of inhibition.
Usually chemicals were studied at two or more doses.
(a) Thionamides~For the heterocyclic thionamides there is strong support for
the premise that there may be a correlation between a chemical's ability to
induce thyroid tumors and its ability to inhibit significantly iodine localization
in the thyroid of rats and humans (Table 4A). For the thiourea-like thionamides
(Table 4B), namely thiourea, trimethylthiourea, and IM'-diethylthiourea,
relative antithyroid activities of about 10 or more were associated with thyroid
tumor induction. In keeping with a correlation between these effects, 2,5-di-
thiobiurea and tetraethylthiuram disulfide (with its structural analogue,
tetramethylthiuram disulfide) both lacked antithyroid activity and did not
produce thyroid neoplasia.
On the other hand, two other chemicals in the series of thiourea-like
compounds need clarification. In the case of 1-pheny 1-2-thiourea, a relative
antithyroid value of 14 was found in rats, but the long-term NCI study in rats
and mice was negative for thyroid tumors or thyroid hyperplasia. There was an
absence of any toxic manifestations in dosed rats in the Tongterm study and a
60
-------
TABLE 4A. THIONAMIDES; RELATIONSHIP BETWEEN ANTITHYROID ACTIVITY
AND THYROID CARCINOGENICTY
HETEROCYCL1C COMPOUNDS
Relative~~~
Anthithyroid Activity
{thiouracil =100)
— : —57
Neoplasms'
a/
rat
ABH
f/
human thyroid other sites
rat mouse
1. 2-thiouracil
100
100
mouse-liver
- CH
S = C
2 6-methy!thiouracil 100
S = C'
.NH - cj
- C
X
.CH3
CH
100
+ mouse-liver and
pituitary
3. 6-n-propylthi ouracil 1100
75
mouse-pituitary
S = C
4. ethylene thiourea
40
50
+ d/ mouse-liver
/NH - CH2
S = C I
XNH - CHo
61
-------
TABLE 4B. THIONAMIDES: RELATIONSHIP BETWEEN ANTITHYROID ACTIVITY
AND THYROID CARCINOGENICTY
THIOUREA DERIVATIVES
Relative
Anti thyroid Activity
(thiouracil = 100)
rat
ABH
MB
human
c/
Neoplasms
thyroid
rat mouse
other sites
1. thiourea
S s C
/NH2
N.
NH2
12
100
rat-liver, head,
face
mouse-skull
2. trimethylthiourea
10
s - c
•N-(CH3)2
'XNH-CH3
3. N.N'-diethylthiourea 40 47
/NH-C2H5
S - C
4. 2,5-dithiobi urea
S - C
S = C'
-NH2
NH
I
,NH
'NH2
62
-------
TABLE 4B. (continued)
~ Relative
Anti thyroid Activity
(thiouracil = 100)
rat
human
c/
.
ABH MB
e/
Neoplasms"
a/
f/
thyroi*
S = C'
.N(CH3)2
^S
I
,S
N(CH3)2
7. l-phenyl-2-thiourea
n 14
S = C
\
NH2
63
-------
TABLE 4B. (continued)
Relative
Antithyroid Activity
(thioracil = 100}
8. N,N'-dicyclohexyl
thiourea
9. 1,3-diethyl-
1,3-diphenyl
thiourea
C2H5
rat
human
T/
67e/
ABH MB~
Neoplasms'
thyroidi
rat mouse
other sites
KEY:
a
b
c
d
e
f
n
- from IARC reviews
- Astwood et aV., 1945
- Stanley and Astwood, 1947
- Mouse study did not examine thyroid
- McGinty and Bywater, 1945a
- from NCI studies, except thiourea (IARC review)
- not tested.
64
-------
question whether a maximum tolerated dose had been used. In addition, after 78
weeks of chemical administration, dosed animals were observed for an additional
26 weeks in rats and 13 weeks in mice before sacrifice. Since thyroid hyperplasia
is oftentimes reversible, it is possible any lesions produced by dosing may
have regressed during the observation period. Other investigators have reported
thyroid hyperplasia after 6 weeks of phenylthiourea administration to rats
(Richter and Clisby, 1942) indicating that the chemical may induce thyroid
neoplastic effects under certain conditions. Further work on this compound may
bear this out.
In the second case, !M'-dicyclohexylthiourea showed increased incidences
of thyroid follicular hyperplasia in dosed rats and mice in the NCI study, and
there were some increases in follicular cell carcinomas in male rats. Although
MT-dicylohexylthiourea has not been tested for antithyroid activity, its
structural analogue, l,3-diethyl-l,3-diphenyl thipurea failed to show significant
antithyroid effects in the rat.
(b) Bridged double"ring aromatic amines--Like the thionamides, certain aromatic
amines with double rings attached by a simple ether-like bridge, show a correlation
between antithyroid activity and thyroid carcinogenesis (Table 5). 4,4'-Methyl-
enedianiline, 4,4'-methylenebis (N^hT-dimethyl)benzenamine and 4,4'-thiodiani1ine
(chemicals no. 1 through 3, respectively) show both attributes, and although
4,4'-oxydianiline (no. 4) has not been tested for antithyroid activity, it has
close structural similarity with the other three chemicals and also produces
thyroid neoplasms. In keeping with its potential for antithyroid effects,
chemical no. 4 produced increases in the number of TSH-secreting cells in the
pituitary in rats following chronic administration (Murthy et al., 1985), and
both chemicals no. 4 and no. 1 produced thyroid enlargements in the NCI 90-day
prechronic studies. All of these observations—antithyroid activity, thyroid
65
-------
TABLE 5. AROMATIC AMINES RELATIONSHIP BETWEEN ANTITHYROID ACTIVITY AND
THYROID CARCINOGENESIS
Bridged Double Ring Compounds
1. 4,4'-methylenedianiline
dihydrochloride
Relative Antithyroid
Activity - rat
(thiouracil =100)
c/
25~
Neoplasms
thyroid
rat mouse
other sites
mouse-liver
; rat-liver
2. 4,4'-methy1enebis
(N,N-dimethyl)
benzenamine
25"
c/
mouse-liver
3. 4>4'-thiodianiline
NH2-/"o~\-S-/T\NH2
15"
d/
mouse-liver
rat-liver
4. 4,4'-oxydiani1i ne
mouse-liver,
harderian
gl and
rat-liver
5. 4,4*-sulfonyldi ani Ti ne
d/
rat-mesenchymal
NH2
-/ 0>-S-(0)-NH2
My it \__/
0
66 r
-------
TABLE 5. (continued)
Relative Antithyroid
Activity - rat
(thiouracil =100)
"57"
thyroid
rat mouse
Neoplasms
other sites
6. Michler's ke.tone
0
NH2-( 0 >-C- 0 >-NH2
\ _ /
mouse-liver
7. 4,4'-diami nodi phenylsulf oxide 12'
e/
/—\ " /—\
NH2-/ 0 VS-/0 \-NH2
8. 4,4'-methylene bis
(2-chloroani'Iine) b/
C1 Gl
NH 2-/0\-CH 2-/VV NH 2
mouse-liver,
vascular
rat-liver,
lung
9. 4,4'-methylerie bis
(2-methylaniline)
n rat-liver
CH3
-
CH3
J
KEY:
a
b
c
d
e
n
- NCI/NTP bioassay except for last two chemicals in table
- IARC review of carcinogenicity
- Astwood et al., 1945
- McGinty and Bywater, 1945b
- McGinty and Bywater, 1946a
- not tested
67
-------
enlargement in subchronic studies, and increases in the cell types of the pituitary
that>secrete TSH—are consistent with the hypothesis that bridged ring aromatic
amines induce thyroid neoplasms by reducing circulating thyroid hormone levels
and increasing TSH.
Other compounds in this series show results that are hard to interpret.
4,4'-Sulfonyldianiline (no. 5), which has an -S02~bridge between the rings, had
a low antithyroid value of 4 in rats and was negative for thyroid tumors.
Compound no. 6 with a -C(0)- bridge was also negative for thyroid tumors.
Although chemical no. 7, which has an -S{0)- bridge was negative for thyroid
neoplasms, it was associated with an antithyroid value of 12 in the rat.
Antithyroid values in the 10 to 15 range have been linked with positive thyroid
tumorigenie effects for chemical no. 3 and some of the thionamides, e.g.,
thiourea. Further studies on antithyroid activity may help to clarify this
inconsistency.
It is also interesting to note that compounds structurally identical to
4,4'-methylenedianiline (no. 1) except for substitution on the rings in the
2,2'-positions (chemicals nos. 8 and 9) are negative for thyroid tumors. It
would be interesting to measure their antithyroid activity.
In summary, for both the thionamides and bridged double ring aromatic
amines there appears to be support for concluding that there is a good relationship
between antithyroid activity and thyroid carcinogenesis, although further work
needs to be done to be able to interpret some results. It seems possible that
agents that are known to inhibit thyroid hormone output may be potential thyroid
carcinogens under certain experimental conditions.
(c) Characteristics of Single Ring Aromatic Amines—Many single ring aromatic
amines have been evaluated for carcinogenicity in experimental systems and have
shown positive effects (Clayson and Garner, 1976; Weisburger et al., 1978; see
•/• 68 . . • '. ' . '••
-------
review by Lavenhar and Maczka, 1985), but only a few of them have produced
neoplasms in the thyroid. Of the single ring compounds that have been tested ..
by the NCI/NTP (Appendix B), o-anisidine (no. 1), 2,4-diaminoanisole (no. 2),
3-amino-4-ethoxy-acetanilide (no. 3), and HC Blue No. 1 (no. 9) were the only
ones to produce thyroid neoplasms. Of these agents only 2,4-diaminoanisole
produced thyroid tumors in all four species-sex categories; the others produced
such tumors in only one group.
The single ring aromatic amines have not been examined systematically as
to their anfithyroid activity; therefore, these agents cannot be analyzed as
to the relationship between peroxidase inhibition and thyroid carcinogenesis.
However, from a preliminary review of structural analogues that have been
tested for carcinogenicity (Appendix B), there is little indication that specific
ring substitutions are influencing thyroid carcinogenic potential.
3. Genotoxicity and Thyroid Carcinogenesis
It has been generally accepted by the scientific community that mutagenesis
plays a role in carcinogenesis. In the case of thyroid follicular cell tumors,
however, it has been suggested that a hormonal feedback mechanism involving
increased output of thyroid stimulating hormone from the pituitary gland in
response to low thyroid hormone levels may be operating (Woo et al., 1985;
Paynter et al. 1986). Even though hormone imbalance may play a role in thyroid
carcinogenesis, it is important also to evaluate the mutagenic potential of
agents causing these tumors.
This section explores the relationship between the induction of thyroid
neoplasms in rodents and their outcome on several short-term tests of genotoxicity.
If the hypothesis that TSH plays a significant role in thyroid carcinogenesis
is true, one might expect that chemicals producing thyroid tumors in experimental
animals would not show genotoxic potential in any predictable way. If, instead,
69
-------
thyroid carcinogenesis were largely due to chemical reactivity and not to
hormonal derangement, then thyroid carcinogens might be genotoxic agents.
This review largely draws upon those compounds that were tested in rats and
mice for carcinogenicity by the NCI/NTP and produced thyroid neoplasms.
Structurally related compounds that did not produce thyroid tumors are included
for comparison. The genotoxicity data on these chemicals are from the NTP, >'
much of which has not been published in peer-reviewed journal;; and at least
some of which could be considered preliminary in nature.
Chemicals are divided into structural classes: thionamides, aromatic
amines, and halogenated hydrocarbons. The NTP short-term test data on many
compounds are limited and, therefore, are hard to interpret. In order to get a
better appreciation of the spectrum of genotoxic effects that may occur among
members of a chemical class, two compounds, ethylene thiourea and 4,4'-oxydiani-
line, were considered in detail (using the open literature) as; examples of
thionamides and aromatic amines, respectively. An example of the halogenated
hydrocarbon class was not included, since members of this group generally show
little indication of genotoxic potential. A third compound, amitrole, was also
included for detailed review; it does not belong to any of the above chemical
classes, but it is recognized as being an inhibitor of thyroid peroxidase as
are certain thionamides and aromatic amines.
(a) Thionamides—For the three chemicals tested by NCI/NTP that were positive
for thyroid tumors, the existing information gives 1ittle indication of significant
genotoxic potential (Table 6). Of 14 chemical-test comparisons on these agents
for both gene mutation and chromosomal effects, there are only two positive
responses. There appears to be slightly more positive genotoxicity data in the
case of thionamides that tested negative for thyroid follicular cell tumors (10
of 19 tests) than for those that tested positive. However, no firm conclusions
70
-------
TABLE 6. GENOTOXICITY DATA FOR THIONAM1DES
1. Chemicals Positive for Thyroid Tumors
MT-Dicycl ohexyl thiourea
NUN'-Di ethyl thiourea
Trimethylthiourea
6ENE
MUTATIONS
SA
B
_
- -
ML
mm
+
-
SLRL
n
•»
-
CHROMOSOMAL
EFFECTS
CA
•»
— ,
_
SCE
+
•••_
_
2. Chemicals Negative for Thyroid Tumors
1-Phenyl-2-thiourea
2,5-Dithiobiurea
Tetraethylthiuram disulfide
Sul fa! 1 ate
Lead dimethyldithiocarbamate
Sodium diethyldithiocarbamate
Symbols: SA, Salmonella reverse mutation; ML, mouse lymphoma L5178Y
cell thymiciine kinase locus; SLRL, sex-linked recessive lethal in
Drosophila; CA, chromosomal aberrations in CHO cells; SCE, sister
chromatid exchange in CHO cells; -, negative result; +, positive
result; n, not tested; w, weak positive result; ?, equivocal
result; /, results from two or more laboratories; u, under test
by NTP.
71
-------
can be made from this limited data set.
The genotoxicity of ethylene thiourea, a compound known to produce thyroid
tumors, was assessed in greater detail (see Appendix C). Although it was
concluded from the journal articles that there is evidence for genotoxicity
when ethylene thiourea is supplemented with sodium nitrite (Salmonella with
metabolic activation, in vivo cytogenetics, dominant lethal, micronucleus),
presumably via the formation of ^-nitrosoethylene thiourea, there is much less
evidence for the genotoxic potential of ethylene thiourea itself. The compound
shows little indication of gene mutation activity: negative to weakly positive
effects in bacteria, negative in Drosophila, and conflicting information in
yeast and mammalian cells in culture (negative in CHO cells and divergent
results in mouse lymphoma cells). Chromosomal effects are not demonstrated in
cells of higher eukaryotes in culture or in vivo. DMA damage tests showed
conflicting results in bacteria, yeast, and human cells in culture.
In contrast to the effects listed above, several thionamides are positive
for in vitro transformation. Thiourea, Mj_'-dicycTohexylthiourea, and ethylene
thiourea have shown positive effects in Syrian hamster cells (SHE and BHK), and
the first two also transformed rat embryo cells (Rauscher murine leukemia virus-
infected) (Heidelberger et al., 1983; Styles, 1981; Daniel and Dehnel, 1981),
However, these three chemicals and jM/-diethylthiourea were reported negative
in simian adenovirus-7 infected Syrian hamster and rat cells (Heidelberger et
al., 1983).
In sum, the lack of genotoxic effects noted with the thionamides that
produced thyroid tumors in the NCI/NTP studies is borne out by the detailed
review of ethylene thiourea. There is little indication of gene mutation or
chromosomal effects. There are conflicting results with the DMA damage tests
and in vitro transformation.
72
-------
(b) Aromatic amines—Unlike thionamides, the class of aromatic amines commonly
demonstrates genotoxlc effects for both point mutations and chromosomal effects
(Tables 7, 8, and 9). This is the case for chemicals that produced thyroid
tumors as well as for analogues that did not.
The genotoxic potential of 4,4'-oxydianiline was evaluated in more detail
using information from the published literature (Appendix 1D) to supplement that
generated by NTP (Table 3). It is concluded that it is a frame-shift and
perhaps base-pair substitution mutagen in Salmonella that requires metabolic
activation for an effect to be noted. In keeping with its mutagenic effects on
bacteria, 4,4'-oxydianiline also produced gene mutations, chromosome aberrations,
and sister chromatid exchanges (SCE) in cultured mammalian cells. However,
SCE are not increased in vivo, and two DMA damage assays in vivo gave discordant
results. In vitro transformation studies were generally positive. Thus, the
analysis of 4,4'-oxydianiline confirms the suspicion from Tables 7 through 9
that aromatic amines are genotoxic agents.
(c) Complex halogenated hydrocarbons—For the class of halogenated hydrocarbons
there are a few scattered positive genotoxicity results (3 out of 16 chemical-
test comparisons among the agents producing thyroid tumors) (Table 10), although
many compounds have not been well characterized as to gene mutations and chromo-
somal effects. Other than toxaphene, all compounds are negative in the Salmonel1 a
test. Structural analogues that have not produced thyroid tumors also show a
paucity of genetic responses (7 positives among 17 comparisons). No firm
conclusion can be drawn on these compounds because the data are limited but, in
general, it appears that complex halogenated hydrocarbons fail to demonstrate
much genotoxic potential.
(d) Amitrole--Amitro1e has not been investigated by the NTP concerning its
carcinogenicity, but from other long-term animal studies, it is known to produce
73
-------
TABLE 7. GENOTOXICITY DATA FOR SINGLE RING AROMATIC AMINES
1. Chemicals positive for Thyroid Tumors
3-Ami no-4-ethy1oxyacetani1i de
£-Anisidine hydrochloride
2,4-Diaminoanisole sulfate
HC Blue No. 1
GENE
MUTATIONS
SA
.+/+
+
+/+
+
ML
n
n
+/+
+
SLRL
_
n
n
_
CHROMOSOMAL
EFFECTS
CA
n
n
u
+
SCE
n
n
u
+
2. Chemicals Negative for Thyroid Tumors
£-Cresidine
5-Nitro-£-anisidine
£-Ani si di ne
2,4-Dimethyoxyaniline
hydrochloride
m-Pheny1enediami ne
£-Pheny1enedi ami ne hydroch1ori de
2-Nitro-£-phenylenediamine
w
•f
-h
SYMBOLS; SA, Salmonena reverse mutation; ML, mouse lymphoma L5O8Y cell
thymidine kinase locus; SLRL, sex-linked recessive lethal in
Drosophila; CA, chromosomal aberrations in CHO cells; SCE, sister
chromatid exchange in CHO cells; -, negative result;
+, positive result; n, not tested; w, weak positive result;
?, equivocal result; /, results from two or more laboratories;
u, under test by NTP.
74
-------
TABLE 8. GENOTOXICITY DATA FOR BRIDGED DOUBLE RING AROMATIC AMINES
1. Chemicals Positive for Thyroid Tumors
4,4'-Methylenedianiline
dihydrochloride
4,4'-Methylenebis (Mi-dimethyl)
benzeriamine
4,4'-Thiodianiline
4,4'-Oxydianiline
GENE
MUTATIONS
SA
+
+
+
ML
+
+/+
n
+
SLRL
n
n
n
n
CHROMOSOMAL,
EFFECTS
CA
+
n
u
+
SCE
+
n
u
+
2. Chemicals Negative for Thyroid Tumors
Michler's ketone
4,4'-Sul fonyldianiline
Sulfi soxazole
SYMBOLS:
SA» Salmonella reverse mutation; ML, mouse lymphoma L5178Y cell
thymTdine kinase locus; SLRL, sex-linked recessive lethal in
Drosophila; CA, chromosomal aberrations in CHO cells; SCE, sister
chrornatid exchange in CHO cells; -, negative result; +, positive
result; n, not tested; w, weak positive result; ?, equivocal result;
/, results from two or more laboratories; u, under test by NTP.
75
-------
TABLE 9. GENOTOXICITY DATA FOR MISCELLANEOUS AROMATIC AMINES
Chemicals Positive for Thyroid Tumors
C.I. Basic Red 9 monochloride
1,4-Naphthalenedi ami ne
GENE
MUTATIONS
SA
+/?
+
ML
+/?
n
SLRL
n
n
CHROMOSOMAL
EFFECTS
CA
M
n
SCE
-V
n
SYMBOLS; SA, Salmonella reverse mutation; ML, mouse lymphoma LS178Y cell
thymidine kinase locus; SLRL, sex-linked recessive lethal in
Drosophila; CA, chromosomal aberrations in CHO cells; SCE, sister
chromatid exchange in CHO cells; -, negative result; +, positive
result; n, not tested; w, weak positive result; ?, equivocal result;
/, results from two or more laboratories; u, under test by NTP.
76
-------
TABU- 10. GENOTOXICITY DATA FOR COMPUX,HAL06ENATED HYDROCARBONS
1. Chemicals Positive for Thyroid Tumors
Aldrin
Chlordarie
Chlorinated paraffins
60% chlorine)
Decabromodi phenyl oxide
2,3,7, 8-Tetrachl orodi benzo-£-di oxi n
£,p' -Tetrachl orodiphenyl ethane
(£,£'-'DDD)
Toxaphene
2. Chemicals Negative for Thyroid Tumors
GENE
MUTATIONS
SA
s
Ir)
_•
_
+
ML
n
It)
+
n
_
.
n
n
SLRL
n
n
n
n
—
n
n
CHROMOSOMAL
EFFECTS
CA
n
ID
n
<•»
—
u
n
SCE
n
(r)
+
n
_
_
u
n
Dieldrin
Heptachlor
Chlorinated paraffins
(C23, 43% chlorine)
PBB mixture (Fi remaster FF-1)
£,p'-Dichl orodi phenyl dichloro-
ethylene (p,p'-DDE)
_.
w
_
+
u
n
+
n
n
n
n
+/-
—
+
n
_
+
+
n
w
SYMBOLS: SA, Salmonella reverse mutation; ML, mouse lymphoma L5178Y cell
thymidine kinase locus; SLRL, sex-linked recessive lethal in Drpsophila;
CA, chromosomal aberrations in CHO cells; SCE, sister chromatTd
exchange in CHO cells; s,selected for testing by NTP; r,reagent grade;
t, technical grade; -, negative result; +, positive result; n, not
tested; w, weak positive result; ?, equivocal result; /, results from
two or more laboratories; u, under test by NTP.
77
-------
thyroid, pituitary, and liver tumors (see Paynter et al., 1986). Like the
thionamides and aromatic amines, amitrole inhibits thyroid peroxidase. Although
it lacks the thiol group of thionamides, It does show some structural similarity
(an R grouping), as illustrated with the comparison with thiourea.
-N-C-N-
s .<;*
thiourea
.N-NH
:C '
XN=CH
ami trole
Gene mutation testing of amitrole has spanned prokaryotes, yeast, insects,
and mammalian cells in culture (Appendix E). Many replications of bacterial
testing in Salmonella and E. coli have almost uniformly failed to demonstrate
mutagenic effects, which led a review group to declare amitrole negative (see
Bridges et al., 1981). Point mutation tests in Saccharomyces and Drosophila
were also negative (positive in one case; see Appendix E). Test results in
mammalian cells in culture have been conflicting, with confirmed negative
results in mouse lymphoma cells but positive effects in one laboratory for two
different loci in Syrian hamster embryo cells. Thus, submammalian testing
indicates little concern about point mutations, whereas results in mammalian
cells are positive in Syrian hamster but not mouse cells.
Testing for chromosomal effects includes evaluation of numerical aberrations,
structural aberrations, and sister chromatid exchange. Negative results have
been obtained in yeast and insect nondisjunction systems and in mammalian cells
in culture. Two in vivo mouse micronucleus assays, which can give some indication
of numerical chromosome aberrations, were also negative.
Tests for structural chromosome aberrations have been uniformly negative
and include the following: human lymphocytes in culture, mouse bone marrow
78
-------
cytogenetics, and mouse micronucleus and dominant lethal tests.
An increase was reported in the frequency of SCE in CHO cells in culture
in two studies; a negative response was recorded in a third study in the same
cells.
DNA damage tests have been performed on bacteria, fungi, and mammalian
cells in culture. Of six bacterial tests, five were reported as negative.
Thus, there is little indication in bacteria of a DNA-interactive effect. Two
of six DNA damage tests in Saccharomyces were positive. One such test in
Aspergillus gave a weak positive reaction.
Increases in unscheduled DNA systhesis have been reported in human cells.
For HeLa cells, a positive dose7response effect for amitrole was noted in the
presence of rat liver S9; no such increase was noted in the absence of exogenous
activation (Martin and MeDermid, 1981). Also, amitrole was reported in an
abstract to be positive in human EUE cells; the conditions of the study were
not gi ven.
Lastly, several positive studies have been reported for j£ vvtro transfor-
mation in Syrian hamster and rat embryo cells, which argue for some type of
genotoxic effect.
In sum, there is limited evidence for the genotoxicity of amitrole. This
effect is probably not mediated through mutagenic mechanisms: there is no
indication of the production of chromosomal mutations and, at best, the point
mutagenic evidence is inconclusive. There are indications, however, that under
some circumstances amitrole produces DNA-damaging effects. These results are
augmented by confirmed positive responses in in vitro transformation. Thus,
there is support for amitrole having a weak DNA-interactive or genotoxic effect
that probably does not involve mutation per se.
79
-------
(e) Conclusion—The review of three chemical classes demonstrating thyroid
carcinogenesis illustrates that thyroid carcinogenesis is not uniformly tied to
genotoxicity. Thionamides (and amitrole) and complex halogenated hydrocarbons
demonstrated only limited indication of a genotoxic potential, whereas aromatic
amines regularly showed positive short-term test results. Emphasis on this
point is gained from review of structural analogues from these classes that did
not produce thyroid tumors; their outcome on the tests was basically similar to
that of the thyroid carcinogens. Thus, thyroid carcinogens do not show a
consistent response on genotoxicity tests.
If we look at chemical classes as to their influence on thyroid peroxidase,
we again fail to see a consistent pattern as to their genotoxicity. Chemicals
from within the thionamides and aromatic amines (as well as amitrole) are Known
to inhibit thyroid peroxidase. However, the reviewed thionamides (and amitrole)
are generally not genotoxic, whereas the amines are active. Thus, genotoxicity
is not correlated with functional activity on peroxidase.
It is well recognized that aromatic amines are often carcinogenic in
animals and that many means are available within organisms to activate these
structures to reactive intermediates that have genotoxic potential. To the
extent that certain aromatic amines also inhibit thyroid peroxidase, it seems
possible that such agents may have two means to influence thyroid carcinogenesis:
to induce DMA damage and to increase the output of TSH from the pituitary.
Although the remarks made in the previous paragraph are representative
impressions of the data on chemical classes as a whole, they certainly do not
necessarily apply to any one chemical within a class. Many times chemicals
give a smattering of positive and negative results. In other cases, such as
with the thionamides and amitrole, the evidence indicates a general lack of
activity for some end points (e.g., gene mutations and chromosomal aberrations),
80
-------
but the potential presence for other effects (e.g., in vitro transformation).
Each of these cases makes it difficult to reach an all-inclusive position on
genotoxicity. Still, within the limits of the present review, there does not
seem to be a consistent relationship across chemical classes as to their ability
to produce genotoxic effects.
81
-------
VI. HUMAN DATA ON THYROID HYPERPLASIA AND NEOPtASIA
The goal of this section is to compare human and animal information bearing
on thyroid physiology, disruption of thyroid function and development of hyperplasta
(goiter) and neoplasia. As has been related, it has been well established
by long-term experiments in animals.that certain chemical substances and other
treatments cause thyroid nyperplasia that will progress to neoplasia. While
evaluation of laboratory experiments garners useful information on liikely
processes in humans, verification of this for human thyroid carcinogenesis
requires evaluating the weight of evidence from several different approaches
and merging data from clinical observations, studies of clinical populations,
and epidemiologic studies.
Currently, the only verified cause of thyroid cancer in humems is
x-irradiation (Ron and Modan, 1982; NCRP, 1985), and this finding is well docu-
mented in experimental animals. There are conflicting data in humans bearing
on an association of iodine deficiency and thyroid cancer, unlike the case in
animals where the association is well established. In contrast to the situation
in animal studies, no studies follow a single human population directly through
the sequence from exposure to chemical substances or initiation of some other
treatment through hyperplasia and eventually to neoplasia. Consequently, the ,
information on humans must be analyzed in separate steps, describing the role
of certain treatments on the development of hyperplasia and then describing
risk factors or antecedent conditions for thyroid neoplasia. The combination
of these two analyses allows one to make some inferences about the overall
comparability of animal models and humans regarding thyroid caircinogenesis,.
82
-------
A. THYROID-PITUITARY FUNCTION
It is widely accepted that the pituitary-thyroid axis and the nature, body
handling, and function of thyroid hormones and TSH are quite similar in experimental
animals and humans. For instance, in a review of thyroid function in humans,
Larsen (1982) presented clinical data on the feedback regulation of thyrotropin
secretion by thyroid hormones and the tissue conversion of T4 to T3 that is
basically like that in experimental animals. Recent evidence, however, helps
to point out some of the differences that may exist between animals and humans.
For instance, in the rat there is active conversion of T4 to T3 which then
regulates TSH production, whereas in humans circulating T3 may play a more
dominant role (Fish et al., 1987).
B. CAUSES OF-THYROID HYPERPLASIA
Animals and humans respond similarly to a number of treatments that disrupt
thyroid function such as 1) a lack of dietary iodide, 2) blockage of the iodide
transport mechanism (ionic inhibitors), 3) interference with the synthesis of
thyroid hormone (peroxidase inhibition), 4) suppression of thyroid activity by
high concentrations of iodide, 5) enhanced peripheral metabolism of thyroid
hormones, and 6) damage to the thyroid gland by ionizing radiation (see Sections
III and V.C. of this report; Gilman and Murad, 1975; Green, 1978; Paynter et
al., 1986; De Groot and Stanbury, 1975; Meyers et al., 1976). Each of these can
lead to goiters in humans.
1. Chemical Inhibitors
Several examples of chemical substances that influence thyroid status in
humans are summarized in Table 11 to illustrate the nature of the effects. The
agents include such things as thyroid peroxidase inhibitors (e.g., ethylene
thiourea, sulphony!ureas, resorcinol), a cation (lithium), an organiodide
(amiodarone), and inducers of mixed function oxidases (phenobarbital, PBB). In
83
-------
to
o
1— 1
u.
ce.
*— «
=5
|
r~
THYROID-P]
o
to
_!
CHEMICA
u_
o
V)
Ul
u.
• t
LLa
UJ
CJ3
•ss.
H"4
§
»— «
Q
3s
>— i
ss
ra
re
1
to
UJ
1—1
Q
rs
•
T-l
«-l
UJ
1
CO
Q
ifll
'W
«o
s_
o
i-
CO
»-
T3
«*
tn
u
CO
UJ
JQ
tn
3
to
•P
I—
CO
3C
US
I- CO
0 S-
co to
tn o
o o.
a x
UJ
11
c
ro
o
CO
o
O-.
**
: c •
S- CO
to -P CTl
. s: a: cu x»
co "O «r- C 4-> >> t-jx:^r
S- 3 S_ -•- "~ 3 C O CO
co o <0 to in -i— "t— *_
:c c a. E -p ro
-P Q >-> in cu 4— .1 co .c
CD O -C 'I— i — O i — i i — O
o
-r~
c •
O -P
co s- c
tn XJ _c cu
C CO O E
0 4-> -P
Q. c: .c co
tn "O « • «r— i
••r- *i— t/5 O O
•00 -P S- CL
•i- i- 4J >).>>.
*** O >, CO JC JC E
0 S- j= tn -P
'—I >j 4-> 3 S- S* -i-
.e o .c CO>$_>, fO >> O
i — cu .c • « in .c «s_
•an. E tn E >,
•P >•,&* tn re s* in jr
I— « JZ IO •«— SI CM -i— -P
^> * »""
•P r™~ (/5 CO
C 1 "O CO -P E I
§tn c -P -P t. 3
•i— «O t— « CO O »— 1 CU
•P "O tn c: UD
• •> v> o i— -O -O
•r— (O tn CO *O (O O CO «f—
COi-i— S-^S-+JO
O CO (O X7 +) CO S>-
s-s--oE-ai-c!cu>,
J=OS-CUC«JOS-^T
Ot-'OM- >
co tn
O
CM r-t
< A
X3 O
CO CO
«Jfr t MM
CO XJ
co s-
£- CO
•P t>
cr> co
CM CO,
CM A 1
i
O CO
!§.
< i-
-p
*CO •
•p 'ca co
o r-^
o -P O>
c c
«^f O O •
>. O O (S)
->c CH-
1 1
a. o >,-p
CO r~" *^r 3
i — co cu
CU -P -C r-
C -P O
^ co s_ .
3 Ip S_ c T-
•O 'CO CO O O
M <
as O • •
0 C CO E
•P 1. CU C i-
O « O -C O CO
c -o a. -p -P
CU CO -r- 1
CO -P C -C J3 C7>
in co O -P S- c
O -P r— -I-
•r- S-
X) CU
r— lO O
3 XJ ro r>-
•o c &- cr»
s: to u. ,-H
X!
S- CO
CU x: >>
O S- OJ T—
O i— O
'CU O. O :r-
c: cu c: Jp
O S~ 3 ll>
•
p S x>
1 O >, C
CO CU M ro CT)
XI l(_ c 1—3
ti- CU in re i- •
in cu E E x> ,r-
o. o os- -P O S- >
x: cu a. c cu o
V- >, CU E -P 1=
CO CO 3 >, »d-^_ X) cu i— ••- E
_jv C C O .C V O CU
cu
„!_, 1
1 «3 i. CO
•P E ro l—
CO CU CU
•* cu n ^^ ^- ,t
*** ^^ -^> .h~ *
(/> S- • i +J m
•p 4-> in x» si- -r- ai
C CU i — LO 3; +.'>
'*— C i — 1 •••> CU £1
-p co co s- ca i — w
CO > rO (— CO •[-'
O-r— CU E XI
CO 4-5 >»-C
S in c i 4-> x» x;i
co co g co ~.<- i— c;
• S-
CO CO
c _e
o , •>, .p
*"" CO O .
ro -^3 tn
~~-, in en
tvl CT1 3 3
CQ r— '$_
•o r-n n.x»
r-~ CM
,
J-S
^ 'i
LU 03
»,
-r- OO
ECD
CO r-H
to
CU 4->
xi in o> c:
3 ^- *t— *r— •
-I-3 'CU 'r— ;f
:in *^* to in XJ o
. S- CO *r- >r*-
CU O S- O> O 4J
, 1=
-c c: o jc ^ 3
}— -.- C 0 -P M-
! * ex
s- d
CO CU 3
•E S- 0
O O S-
rr- 1
CO O
E -OL. *
s- >,E
O .c in
C 1 ••!-
r-1 X>
•^ o
1— 3C >>
x>
x> o o
c. c s-
CO >,
^C Jd
ty) 4«> 4«>
4- T-
CU 3: <+-
34 O
S- V)
O i — >-, CU
3: 0 S- tn
s_ o .4^ •&>
r— c tn in
CO O *r— *r—
s: o j= x>
,
1
S- O>
ca !
o
H-i CM
in
o
JLS
•rP
c:
o
o
|JO O
CU CO
c: co
co s_
i — 3
>> o
iP C~
i . i 4J
-------
^^
XI
o>
3
C
•r—
P
E
O
o
, — 1
V— 1
UJ
1
CQ
d^
1—
•
M-
OJ
a:
q-
co
ro
CO
<0
P
CX
0 X
O UJ
II
c
ro
O
•r—
E
QJ
.E
O
,_
"o -
fl •
CO i —
C ro OO
n3 i^**-
S~ P CT|
| — <1) rH
,
1 *r—
CT O
CO QJ | QJ X* CO
QJ S- O Q • *> QJ CJ x?
•r- <4- CX CO E P 1 — QJ
X) >, QJ QJ s- (O S_
3 JE JE E O E QJ
p cn E o ex QJ x> •
CO-r-M— CO S QJ l|— •,— X>
JE O -i- S- to QJ
r— X) C C C CO
rO P >>••- V- 1- -r- O O
s- s- o o QJ o c:
QJ o c s_ >,p >, en
> CX QJ >>•— -r- 1-^-3: ra
QJ QJ 3 JE r— O C OO •«-
OO S- CTP >-> t/)
<+- JE -r-
•
QJ P
> E
•r- QJ XI
tO -r- C •
1 CO P ra QJ
O QJ ro i—
••- S- CX CU rO
E CXP r- E
rO QJ 3 ro OJ
S "O O E <*-
-^-»
OO
*^1"
CvJ •- 1 -•>
CO -— >, -CO >,
CO ra QJ -— ra to
•> ^H XJ i — XI CTt JE
i — co o" E • cr,— i c
CO O
'I — 1
E
t3
•l —
jn
4->
•r—
1
•t
•
r—
{LJ (t3 P""--
C P*-
(O •!«> O^
-1 OJ ^-<
QJ
• « c
P O "
s- N
O «O -Q •
CXP c:
OJ 3 O > C 5-
to e S -P
S- Q) O T-
••- J= C O
CT
C 3
O S-. •
••- XI i—
CO tO
co S_ >
12 i
-
•"^ CX O
CO(O >, S-
1- •»-> J= >>
t Ov j j^
->3^- P
"X? > 2 -*
i E o
OO QJ ro
to M— JD
X5
QJ
ro
p
co
o
•z.
, — 4
c c:
QJ O
JE M
CX rO
-*^ -p
X 3
O -Q
^-__ ,^_,
r~~ ft
OJ .
3 r-
P ro OO
o r-.
o p a>
CiJ ,
C •
O ••> ro 3:
COJ= OO
0 1- 0 H-
•," c: O c:
-^ -1- C' T-
:>i
to
CX
r~ O
•r- E -P
ex to v c
c: P E
P QJ IP
r— •!— CO rO
3 P C QJ
X* ro O S- ,
< CXi— P
X)
QJ
p
01
p
C3
CO
r*~
o
S—
p
c
o
o
o co
<-* OO
c
•r-1
0
p
>-J
d
Q)
jr.
a..
•t
•
r«
c to o
JE OO
ro P cr>
03 QJ ^H
•
co co
p p
S- 1 CO XI tO QJ
O S- S- c c cc ,i^ > r— CX
o JE , QJ
ra -P S- *
E to 3 to
•r- CO QJ
«*co O to
| — S- CX (O
<0 X QJ
-^ CX QJ O
rn •"—
OO • XI
h- • « i E ••- co
.x S- O to O QJ
*^~. QJ CX'i- S- -r-
,P P E
o o •<•. j= c c: o
_j c: «* p ft ra to
^_
o
QJ
QJ
S- X> QJ
O ro
QJ S_ QJ
!-*• >)
(O rifJ7^ »i—
S PX5
1 — 1
rO to >>
C A ,— x>
O • •> O ,
ra 3 • c: c
CX to co -r- QJ QJ
3 0 J*! E J= X>
O CX OJ O CX-^
O X QJ S- -r- X
(~> fii ^e o o o
to
*o
p
c
o
o
in cr>
CO OO
to
1 1 —
O XI >,
i~ a> c:
-O P QJ -— ~
>, ro j-; CQ
• — C CXO3
O •!- -t- Q.
Q- E CQ •—
a*
c. •
O r—
P ra LT>
c: to
3 4-> Ch
"T" QJ v— 1
1
« ••> E Pi
1— < QJ CO 3 O •
CQ J^ -i— E JO S- to
D- ro X> S- i- CXP
S- v P -r- QJ CO O
QJ-^ CX O P O QJ QJ
JZ 3 S- 1 S- 1-
P <4- >,P • O <*-
OOi— iJCS-PE QJ
.-H P 0 C
i — tOCO O JE QJ IX)
S- -r- >, P CO O
> 3 •!- P QJ -r- 3
QJP vO-i-S-EO
OO CO — X c 3S P fO C
.'
Xi
i x) c: i QJ
s_ c -i- P E
o i- -p -P •• >>
O C C S- JE
E C O QJ , •
CO O S- QJ COjC E
x. C 3 S- s M
-> -t- X> P C> ^~ -r-
JE
•p to
•i— CX
3 3 X
XI O QJ
co o i- co
o o o •
•f— r— XJ XI
P JD C QJ
QJ • rO -E
ja<- ra o
ra QJ QJ P
•f- o S- cn ro
Q E 3 ro E
1
to 1 1 O
JE S- tO QJ S-
p •> 3 • XI UO O
c P xJ o >>••- • i —
o E ra E 0 J=
E QJ QJ XI ra 1 O
E cn • ~-~~. p t— i
o -P ra E cn 3 • >>
to ra i- o o o ra
1 QJ QJ -i- CD i — XJ
CM S- > P « O S- ---.
•^ p ra ra co P o cn
o
CM
CM
J.
>>
o
JE to
ex (a
i — QJ
3 S-.
OO 3
CO
ro
ex
p
X
QJ
E
c
o
X3
QJ
3
E
r-
p
E
o
o
1— 1
r-H
QJ
Q
ra
"~
85
-------
-a
o>
3
•»r—
'•c:
;O
tJ
*
<*—*
;r-i
~tjCl
£0
[f™*
«
CD
ex
C|—
CD
to
re
CO
to
;«^J
re
:a
Of
re
s_
p
»">
CD
1—
"O
•»
o
.to
•p
o
a>
u-
CJ—
Jxl
-P
to
3
fO
-P
to
•P
"(O
CD
•3=
re
\n
O) P
t/> o.
P X
•Q 1 1 1
;n
:f*
rre
fj
:»r"
JO
re
•r-
a
^
"•r- C
cu r— re
C 3 3 .
P tO CnCO
r— C i— CM
re ••- -r-
.0 «
-P •« M
CD CO •« CU
»f i-i co -o
a .t— ? CT> •!-
to
o
•p
xz
:o
o
CCNJ
TC\J
i -a
:r— ;O)
,>> °3
"C 'C
P ••!-
_£= :tO ,-P
,D_ CO :C
rr :?> 0
^3 :S- CJ
it/) ;3 —
-o
•re s-
CD
• — to o
r— re .to
CO I_u i-H
1 1 tO
•a >> —
-P »r- r— il
s. p re -P
0 S- -r- -P
Q. >,-a c: >•> •
5-5 re r^ "c 0
P -P P .!=
r~ D_;C: re cu
.'(••- ,>V— d.'tn S_
CT* •<— p p -i— "O
S— -P C1.T3 CU
1-1 -p to •>,'•!- tn
j-j j= p :s_
co c -p :s_ CD
i-H CD CZ "TO >j .>
^~ 3 E §5 S!
.* +j
ClJ C
CD CU •
c E i- tn
•t- D-.OI E
P > P
i— i T3 i — CD -P
:ca »r- cu to o.
Q- 0- > E
.. re a> *»_ >,
-> S- -O :P to
--r- E
-• i — .to
« to re -r-
OS- T3
tn re " -r-
'CU 'O P
• >> re r— s_
,to -r- re •>>
eu o -o o,j= •
r— CO i- •!- 4-> tO
re re c: p CD
E :«\ 'cj :.»— O- tn
.CD CTl 1 . — >, re
U_ 10 rH CJ .C -CJ
si. •
p to
-P CD
C O
CD r—
E 3
-P
c cn
•i- CD
O <—
CO
i—
p
r"
;«t—
(j
•'s-
P
CO
'CD
0£.
0
1—
-•
c
CD .
•i— co
cn
c
co E
"r- 3
C *0
•CD CJ
cn- — •
P
i- i —
•P re
•f™* >
p p
cn E
CD
CD
•P S-
$$
•Q.CI-
3
to c
o
CD .r-
JC tn
.+-> to
•?—
tt c!
P CD
S-
CU
V) -.1-
p p
T3:tl_
'CD NX
i5J
•>^
i — C
c p
O -r-
jir>
0
• -C
tn 3
1 — CI_
re to
0 >,
•-•l— 'T3
,E
.CD 13
Jb -r-
o ;p
,-P
to
•cn.x:
3 -P
"lj '.^
-H tn
re
.
"^O
CD
-P
re
-+J
to
o
tn
tn
,«r—
CD
S-
3
en
p
,0.
X
CD
S-
P
•P
C
>cu
J
re
CD
S-
•p
-p
i_
p
•r~
.D,
•a
•r—
S—
j?
3
CD
tn
re
TO
tt)
to
tn
O)
tn
en
re
to
i->
O
CD
5
3
CO
_Q
lj
p
•o
•r- CD
P 0
;s- ,c
>^ CD
j; uo
,^
• «\
ci ..
1— CD
TO S
E re
re-P
.^j-
:tn P
0-
... m
'(/) >—
-,gj
4^> O
•r— 'C
CD .XI)
> i-
•i- re
•o
Ci_ _CU
• p ,C
•r-
-p ;re
"3 CD
to
CD cn
oy
to -P
-P -r-
t-
P S-
O-'CD
CD J=
S- -P
CD
C
E
r— C
0 •?
CD O
*"<§
en TO
?F— '5O TO
T3 -r- ;CU
3 :P to
•P S— e/>
:en >, CD
£: to
'.cniP -'to
'.Of—''
E « -P
re u p
••- c:
•o <=
O) ••- -P
^«- i — O
,s- o a>
re <4_
•P "CD
'-T3 ;•'£=
:CU XU : to
•C E "CD
•if ''(D HJ
E cn .re
•re .s- o
x re -i-
•c c
.tO 'CU -r-
•4J
cg_ re p
:uJ ,'Gn .CD
,,Ci>
. s
o
I—)
CD
•p »
•r- ro
O- .-1 ^
P CD
s_
O-.c:
CX J=
(O tS
.CQ
•o
ro 3
•P
X ••>
CD tn
-P CD
O
c c:
"~ s.
TO OJ
CU cv.
re •"—
tn
to *re
"• i
CD eg— ,•
en p IA
•3 O-
cn 3
4^ -P P
to en S_
'Ol CD Cn
•P -P
M — CU CD
•i— t. CD
'^> -P
r— -P CD
c tn J3
o p
E en
d) « cu
•r- • >
4— 'cn , -P
•r— re
L^U
O */) 3
•r- C
•r- CO* C
c E ••-
en
•i— cn tn
en c: CD
•i- O
.r— tn 'CD
r- 0> S.
O *+-. tn ,en
•r- « cj™ ^3 ro
,4J S- "— CU CD
tn .CD "O S- ,S-
•i— '> O O
4J CD ,+J [C '(U
•re S <" »••— ^3
P 0 10)
-.c/>;rc -P ii ii
.. .
CO
'.'p
"i
00
1
1
,
! 4^>
;a>
',:\
(Tj
'fit
i.
m
cz
tr—
'E
t/>
:a»
, ^
; ,Q
' X*
• .c
Jl
~1o
"CU
'i-
**
'"Hi—'
"P
^_
re
•c: :3S
• -t
.Ip T3
frt ir*
'> ;*>
l -***
O t/)
" ' • ,4J
'r- 'M
re cu
•f—
ir~ • Y3
r B
u *o
•-ft*
• o "O-
4J tlJ
; ' • -;s-
M- O
ni
.•U^ •
s. +j
tn ;tn .tn
t— .«- CU
cu -a s-
• .i» -i-
O) p cO
>> en
E c™ rre
3 +J
S- 'i- ,'J=
CL) '•% V
:ll JT
1 H— 'J^ O
'~a '*<—
^rO rO "CU
rf->
00 ,>, ^3
i'l — ;-C
/*
^tu.
CD
;X-
:re
CO
s'O
' f»
dj
,j=
.CJ
ip
^rci
^.
;C
-,-P
lo-
•tj
"tt
cn
•P
.•r—
p
_!_>
- ^~*
5-
o
"d.
•Q.
.«n
'O
-r-*
QJ
't/i
;'50
_Q
'_!
(0
^
r__
S^
•
f"Ti
.V)
uj
'4-
-------
each case exposures result in reduction in circulating thyroid hormone levels
and in some cases elevated TSH levels or goiters. These responses are like
those seen in animals.
Because the data base varies among the chemicals, a summary of supporting
references, including those reported in the study, is included in a separate
column entitled "data base." For example, the goitrogenic effect in humans of
sulfonylureas and of amiodarone has been reported in several clinical studies.
Differences in quantitative value of the results among studies are to be expected
because of differences in health status, age, sex, and dietary factors. In some
studies these factors are controlled (patients of similar age) or evaluated in
the analysis (sex differences).
The value of a case report in support of the hypothesis is strengthened if
cessation of treatment with the putative goitrogen or other agent is followed
by a return of thyroid function tests to normal. These temporal associations
are important in assessing the evidence for the association because subjects
are exposed to other drugs or possible confounding factors. This information,
which is important in assessing the strength of the evidence, is summarized in
the table column titled "Temporal." Prospective clinical studies provide
valuable information because subjects are euthyroid prior to exposure.
Other observations point out the comparability of response in humans as in
animals. In hypothyroid animals the cells of the pituitary enlarge and become
"thyroidectomy cells" (Baker and Yu, 1921) and, according to some authors, may •
undergo hyperplasia and finally neoplasia (see Section IV.B.). Indirect
studies in humans also demonstrate some of these findings. The bony covering
of the human pituitary, the sella turcica, normally enlarges with age up to
about 20 years and then remains essentially constant in size. Enlargement
in the sella turcica beyond normal limits is noted in cases of hypothyroidism,
87
-------
and there Is an inverse relationship between the blood levels of thyroid hormones
and sell a size and a direct one between TSH levels and size of the sell a turcica
(Yamada et al., 1976; Bigos et al., 1978). It is interesting to note that there
are also a few clinical reports linking chemical hypothyroidjsm and pituitary
adenomas, and at least some of them appear to be TSH-secreting tumors (e.g.,
Samaan et al., 1977; Katz et al.} 1980; see review by Balsam and Qppenheimer,
1975), although the case is not established with any certainty.
2. Dietary Factors :
Much of the human investigations of disruption in thyroid function following
environmental modifications have come from the study of populations where there
are dietary changes, namely deficiency of iodide and the consumption of foods
containing goitrogenic substances. ' ' • .
a. Iodine Deficiency—The most striking patterns of the geographic distribution
of populations with goiter is attributed to deficiency of iodine in the diet as
a result of low environmental iodine levels. Endemic goiter has occurred
throughout the world, particularly in mountainous areas such as the Alps,
Himalayas, and Andes, and in the United States in areas around the Great Lakes.
De Groot and Stanbury (1975) cite the report of thyroid hyperplasia in domestic
goats and in wild rodents in endemic areas of iodine deficiency in the Himalayas,
which again points out the similarity of response among mammals. Goiter incidence
has been virtually eliminated in the United States and Europe by the introduction
of iodized salt (Williams, 1977; De Groot and Stanbury, 1975; Hedinger, 1981).
Several arguments support iodine deficiency as a cause of goiter: 1)
there is an inverse correlation between iodine content of soil and water and
the appearance of goiter in the population; 2) metabolism of iodine and TH and
TSH status in patients with this disorder fits the pattern expected and is
reversed with iodine prophylaxis; and 3) there is a sharp reduction in goiter
prevalence with iodine prophylaxis (Williams, 1977; Hedinger, 1981).
88
-------
Iodine deficiency in humans can result in profound thyroid hyperplasia,
Goiters up to 5 kg (a 100-fold increase in weight) g have been observed in
iodine-deficient areas as a compensatory response to inability to synthesize
thyroid hormone. Generally, the impairment in hormone synthesis is overcome in
time, and the individual becomes clinically euthyroid, even in the presence of
some derangement in 14 and TSH levels. Often in goitrous populations repeated
cycles of hyperplasia and involution occur which can lead to multinodular
goiter. In contrast to the hyperplastic goiter, multinodular goiters do not
regress upon administration of iodine. Likewise, thyroid hormone usually has
no effect on long-standing goiters (Ingbar and Woeber, 1981). Adenomatous
hyperplasia is a less common cause of nddularity but is significant, because it
is difficult to distinguish from neoplasia, thus complicating the assessment of
the association between hyperplasta and neoplasia. As will be developed later
in this section, it does not appear that thyroid cancer is a major problem
arising from iodine-deficient goiters, in contrast to the observations in
experimental animals which indicate that tumors frequently arise under iodine-
deficient conditions.
b- Other Goitrogens—Observations of goiter distribution suggest that factors
other than iodine deficiency could be important. The incidence of goiter varies
within the population in endemic areas, and the seventy is not uniform among all
inhabitants; these suggest the presence of risk factors in addition to iodine de-
ficiency. Although it is considered unlikely that natural goitrogens in food are
a primary cause of goiter in humans, variability in response within endemic areas
has led some to conclude (De Groot and Stanbury, 1975) that "natural goitrogens
acting in concert with iodine deficiency may determine the pattern and severity
of goiter."
As discussed in Section V.B. a thionamide, goitrin, with antithyroid
89
-------
activity in animals and in humans, has been isolated from certain cruciferous
foods (e.g., turnips). It exists naturally as progoitrin, an inactive thio-
glycoside, which is hydrolyzed in vivo to goitrin.
Human data exists to illustrate the thyroid inhibiting effect of the
monovalent hydrated anion, thiocyanate (TCN), and of cyanogenic glucosides,
that are hydrolyzed in the body to thiocyanate. TCN blocks the uptake of
iodide into the thyroid. Chemicals that are metabolized to thiocyanates are
found in seeds of the plants of the genus Brassica, in Cruciferae, Compositae
and Umbelliferae. These include cabbage, kale, brussel sprouts, cauliflower,
turnips, rutabagas, mustard, and horseradish. The effect was established in man
as a result of clinical use of potassium thiocyanate (Gilman and Murad, 1975),
It has been assumed, therefore, that eating foods producing the thiocyanate
ion or goitrin contributes to endemic goiter. De Groot and Stanbury (1975) cite
studies in Australia, Finland, and England, that suggest cattle have passed
these goitrogens to humans through milk. Progoitrin has been detected in
commercial milk in goitrous regions of Finland, but not in nongoitrous regions.
Seasonal development of goiter in school children has heen related to milk from
cows fed kale (De Groot and Stanbury, 1975).
Several dietary items that are staples in some cultures contain cyanogenic
glucosides. These include cassava, sorghum, maize, and millet. In its raw
form, cassava contains toxic levels of cyanogenic glucoside, and although much
of it is removed by pounding and soaking, poorly detoxified cassava is a suspected
cause of goiter in Central Africa.
Recent studies in Africa contribute more direct evidence to support an
interactive effect of TCN (or cyanogenic glucosides) and a diet low in iodine.
In an iodine-deficient region of the Sudan where goiter prevalence may reach
55%, the frequency of large goiters is higher in rural than in urban areas
90
-------
(Eltom et al., 1985). The predominant staple food in rural Darfur is millet.
Rural subjects with goiters had statistically significantly higher levels of
TSH and 13, lower levels of 14 and free 14 index than urban subjects with
goiters. Serum TCN was significantly higher in rural subjects, but the
elevated levels of urinary TCN did not reach statistical significance. The
urinary iodine excretion, a reflection of quantity of iodine ingested, was not
significantly different between the two groups. These results are consistent
with the hypothesis that TCN overload in conjunction with iodine-deficiency
causes more severe thyroid dysfunction than iodine-deficiency alone. Evidence
of a possible effect has also been reported in North Zaire in Central Africa in
children with iodine-deficiency (Vanderpas et al., 1984).
C. CAUSES OF THYROID CANCER IN HUMANS
Epidemiologists search for clues to causes of disease and to factors that
increase an individual's risk of disease (risk factors) by examining descriptive
data or designing analytic studies. Descriptive data consist of morbidity,
mortality, or incidence rates of diseases in population groups. Incidence rates
(newly diagnosed cases in a population over a given time period) reveal patterns
of disease by age, race, sex, ethnic group, and geographic locale. These rates
and their changes over time and space identify high risk groups and provide
indirect evidence for causes of disease. Associations between host factors and
disease are hypothesized.
Analytical epidemiology consists of case-control, often termed retrospective,
and cohort or prospective studies. These studies permit greater control of
confounding factors and opportunity to link exposure and response information in
individuals. Thus, evidence for causes of disease is more direct.
As a result of descriptive and analytic epidemiologic data, radiation is a
well documented cause of thyroid cancer in humans (Schottenfeld and Gershman, 1978;
91
-------
Ron and Modan, 1982). Incidence rates for thyroid cancer rose roughly two-
fold between the 1940s and the 1970s for persons under age 55. The change in
pattern coincides with administration of x-ray for various medical treatments
and is consistent with the hypothesis that ionizing radiation is a cause of
thyroid cancer in children and young adults. Childhood irradiation was observed
more often in thyroid cancer cases than controls. Ron and Modan (1982) summarize
eight epidemiologic studies of populations exposed to x-ray therapy, atomic-bomb
explosions, and fallout from nuclear weapons testing.
The epidemiologic approach to investigating whether hyperplasia (goiter)
leads to thyroid cancer in humans is to: 1) examine descriptive data, 2) compare
the cancer rates between endemic goiter areas and goiter-free areas, 3) examine
time trends for thyroid cancer after prophylactic measures (iodine supplementation)
reduce endemic goiter frequency in a given area, and 4) evaluate whether goitrous
individuals have a greater risk of thyroid cancer or whether thyroid cancer
cases have a more frequent history of hyperplasia and nodules than controls.
These steps are summarized in the sections below.
1. Descriptive Epidemiology
Variations in cancer incidence rates by country and race may be studied to
evaluate the role of host and environmental factors on disease. Despite the
striking geographic patterns for goiter, no similar trends are detected for
incidence of thyroid carcinomas in the areas for which cancer incidence data
are available. It is one of the rarest and generally least virulent carcinomas,
and although it has increased somewhat in recent decades, purportedly because
of medical radiation exposure, it is not considered a major public health
problem (Ron and Modan, 1982).
For several countries, thyroid cancer shows rising age-adjusted incidence
rates with age and consistently higher rates for women than men, particularly
92
-------
in young adults. Rates for males range from 0.6 to 5 per 100,000 and for females
from 1.2 to 16 per 100,000. Variations by country are relatively small compared
with that for other cancer sites (about 10-fold) and are not consistently related
to geography or race. The highest age-adjusted rates in females (1967-1971)
were for Hawaiians in Hawaii (16/100,000), Iceland (16.3/100,000,), and Israeli
Jews (8.3/100,000) (Waterhouse et al., 1976).
The incidence of thyroid cancer detected clinically shows interesting
distinctions from prevalence of occult thyroid cancer detected at autopsy. At
autopsy, thyroid carcinoma is equally frequent in men and women, and high rates
have been diagnosed in populations that have unremarkable clinical rates of
thyroid cancer (Shottenfeld and Gershman., 1978). These observations have led
these authors and others to hypothesize that the host and environmental factors
that enhance the development of clinically detected thyroid cancer are different
from those that incite tumorigenesis.
Experimental evidence in several laboratory species demonstrates that
iodine deficiency, certain chemicals, and other causes of prolonged TSH stimu-
lation result in thyroid enlargements and eventually thyroid tumors. In the
absence of such information in humans other studies need to be conducted to get
some handle on human thyroid carcinogenesis.
Much of the work on the relationship between goiter and thyroid cancer has
focused on populations differing in iodine intake, since iodine-deficiency
(endemic goiter) has been and still remains a major health problem in various
parts of the world. Numerous reviews of the subject have been written which
conclude that past studies are conflicting about the role of goiter in thyroid
carcinogenesis (e.g., Alderson, 1980; Hedinger, 1981; Riccabona, 1982).
Doniach (1970a) reviews much of the information available to that time and
questions the link between endemic goiter and thyroid cancer development.
93
-------
In geographical epidemiologic studies, thyroid cancer rates are compared
in geographical areas with different goiter rates. Wegelin (1928) compared the
frequency of thyroid cancer in an autopsy series in five areas. The largest
percentage with thyroid cancer occurred in Berne, Switzerland, an area where
goiter was highly endemic. The lowest percentage of cancer appeared in Berlin
where endemic goiter was rare. Other geographic correlation studies have
followed, yet reports have been conflicting. For example, no correlations were
found in reports from Australia and Finland (Alderson, 1980; Ron and Modan,
1982), and Pendergrast (1961) found no associated increase in the cancer rates
in goiter areas in the United States compared with non-goiter areas. Hedinger
(1981) cites incidence statistics that show no decline in frequency of thyroid
malignancies despite the virtual elimination of goiter by iodine prophylaxis.
On the other hand, Wahner et al. (1966) did show a positive correlation when
they compared the incidence of thyroid cancer in Cali, Colombia, an endemic
goiter area, to similar data in New York State and Puerto Rico. Thyroid cancer
rates for both sexes were about three times higher in Colombia than in the other
two sites. ]
Several reasons may account for differing study outcomes. Some of the
correlations are based on reports of high thyroid cancer rates; generated from
pathology studies of surgery cases, and are likely to suffer from a selection
bias because thyroid disease suspected of carcinogenicity is likely to be
referred to surgery (De Groot, 1975). Different causes of cancer may result in
different histopathological types of thyroid cancer. In the United States, in
particular, radiation-induced cancer associated with therapy in childhood could
have masked a decrease associated with iodine prophylaxis. After the introduction
of iodized table salt in Switzerland and decreasing incidence of goiter, thyroid
cancer rates remained stable but an increasing proportion of thyroid cancers
94
-------
were classified as papillary (Shottenfeld and Gershman, 1977). Therefore, the
conflicting data cited above are inconclusive and difficult to interpret.
More recent geographical studies consider the histological type of thyroid
cancer. In Cali, Colombia, an endemic goiter area, at least 90% of the follicular
and anaplastic cancer specimens showed evidence of goiter, whereas about 50% of
the papillary tumors were associated.with goiter (Wanner et al., 1966). These
results suggest some relationship between goiter and the histological type of
cancer.
In Zurich, Switzerland before the advent of iodine supplementation, few
of the tumors were papillary (7.8%), whereas after that time the proportion of
papillary cancers among the total increased (33.4%) while the proportion of
follicular and anaplastic tumors decreased (Hedinger, 1981; Riccabona, 1982).
Since papillary cancers have the best prognosis and anaplastic the worst, with
follicular intermediate, these results suggest that thyroid cancer in endemic
goiter regions may be associated with more aggressive forms of cancer.
Further evidence of a relationship between iodine intake (from inadequate
to hypernormal) and the form of thyroid cancer comes from a review of thyroid
cancer cases coming to surgery in Northeast Scotland, a region with average
iodide intake, and Iceland, an island with very high iodide intake (Williams et
al., 1977). Persons from Iceland have unusually small thyroid glands, high
concentrations of iodide in plasma and the thyroid gland, and low plasma TSH
levels. Papillary cancer incidence was about five fold higher and the proportion
of papillary cancers among the total was greater in Iceland than in Scotland
(71% vs. 54%). Offsetting the difference in papillary cancers, the proportion
of follicular tumors was comparable in the two groups, but anaplastic cancers
were more common in Scotland than Iceland (19% vs. 10%).
In contrast to the above studies suggesting some relationship between
95
-------
iodide intake and the form of thyroid cancer in humans, others; fail to support
this hypothesis. For instance, Waterhouse et al. (1982) report that the relative
frequencies of the major histological types for several countries show the
highest proportion of follicular carcinoma in Sao Paulo, Brazil, Bombay, India,
and Zaragoza, Spain--all areas not noted for endemic goiter. The highest
proportion of papillary carcinoma was reported from all North America cancer
registries, and from Hawaii, Israel, and Singapore. In addition to noting the
potential for disagreement in diagnoses among experienced pathologists, the
authors conclude that the significance of these differences is unclear. There>-
fore, geograghic correlations with and without histology data are inconclusive
and do not show a consistent relationship between endemic goiter areas and
thyroid cancer rates.
Probably the most profound disruptions in thyroid functioning occur in
cases of familial goiter where there are inherited blocks in thyroid hormone
production (Stanbury et al., 1979). When left untreated, these patients develop
profound hyperplasia and nodular (benign tumor) changes, but only a very few
cases have gone on to develop thyroid carcinoma (see review by Vickery, 1981).
Like with endemic goiter, it appears that the hyperplastic thyroids in these
patients do not often undergo malignant transformation; this contrasts with the
findings in long-term animal studies where blocks in thyroid production
regularly lead to thyroid cancer.
Although not much seems to have been done concerning the follow-up of
patients with Graves' disease (hyperthyroidism) as to thyroid cancer development,
the little that has been done (a follow-up of 30,000 patients) suggests there may
not be a significant thyroid cancer problem in these cases (Dobyns et al., 1974;
see also Doniach, 1970a). [One very small study of Graves' patients suggested
a higher than expected frequency of thyroid cancer (Shapiro et al., 1970)3.
96
-------
The reason Graves' patients may be at risk is the finding that many of the
persons carry immunoglobulins in their blood which bind to the TSH receptor
on thyroid cells and, at last in vitro, act like TSH to stimulate DMA synthesis
and cell division (Valente et al., 1983; Tramontane et al., 1986b). Since
these patients frequently have enlarged thyroid glands, one can not help but
think that the immunoglobulins may stimulate thyroid cell division in vivo as
well.
The single investigation of Graves' disease patients treated with anti-
thyroid agents (i.e., thionamides) for at least one year failed to show any
thyroid cancers in over 1,000 patients (Dobyns et al., 1974). Again, this
suggests that at least circumscribed use of antithyroid drugs is not attended
with a marked, thyroid cancer risk. It should be pointed out, however, that the
goal of antithyroid treatment for Graves' disease is to bring patients into
euthyroid and not a hypothyroid status where increases in TSH may occur. Thus,
the follow-up of treated cases of Graves' disease does not provide significant
evidence to impugn or acquit antithyroid agents.
2. Analytical Epidemiology
Of all the various types of data on humans from which causal associations
can be inferred, the strongest evidence is derived from analytical epidemiology
—cohort or case-control studies—that evaluate data on individuals and suitable
controls. Analytical epidemiologic studies have helped to establish ionizing
radiation as a cause of thyroid cancer (Ron and Modan, 1982).
Three case-control studies of thyroid carcinoma in the United States have
recently been completed which evaluated risk factors for cancer including pre-
existing thyroid disease (Table 12). These studies were designed to test a
potential hypothesized role of endogenous female hormones in thyroid cancer.
Hormonal factors are suspected as a cause of thyroid cancer because of the
97
-------
TABLE 12. EPIDEMIOLOGIC STUDIES OF THYROID CANCER AND
ITS RELATIONSHIP TO GOITER AND THYROID NODULES
Odds Ratio (95% confidence limits)a Comment
Goiter Thyroid Nodules
Reference
4.5 (1.6-12.2)0 8 7 (1.6-47.5)b Women aged 18-80 McTiernan et al., 1984
10.5 (2.5-44.8JC
White women aged Preston-Martin et a}.,
15-40 1987
5.6
33 (4.5-691)d
Adjusted for age,
sex and prior
radiation exposure
Ron et al., 1987
ratio estimates risk of disease with the trait (or exposure) compared
to risk without the trait. Confidence limits that overlap 1.0 are not
significant.
bData for those unexposed to radiation. The risk for all cases was goiter
6.6 (2.8-15.6) and nodules 12.0 (2.3-63.8).
cPresence of goiter or benign nodules.
dThese data are from univariate analysis. The odds ratio of a multiple logistic
regression adjusted for age and sex were thyroid nodules (28.0) and goiter (3,8)
(not significant).
98
-------
consistently higher rates in females and the peak occurrence in females at between
ages 15 and 29 when hormonal activity is enhanced (Henderson et al., 1982; Ron
and Modan, 1982).
McTiernan et al. (1984) studied 183 women aged 18 to 80 located from a
population-based cance^ surveillance system and 394 controls. The two groups
had similar family history, weight, and smoking habits. The most common con-
founding factor in the analyses was age; therefore, relationships were adjusted
to five age groups.
History of goiter for individuals unexposed to radiation showed a statistically
significant and high odds ratio (OR) equal to 4.5. Further analysis of pre-
existing goiter by histopathological type resulted in an OR=16.-4 for foilicular
.compared with 3.3 for papillary cancer. Radiation exposure doubled the risk
for those with papillary histology, but did not change the risk for follicular.
Thyroid nodules were also a statistically significant risk factor in those
unexposed to radiation (OR=8.7) and was strongly related to papillary or mixed
papillary-foil icular thyroid cancer.
There are some potential biases in the McTiernan et al.(1984) study such
as recall bias, relatively low ascertainment rate (65%), the lack of re-evaulation
of the histopathology, and the reliance on telephone interviews rather than
medical history. However, it is doubtful that these could be the cause of
associations of the magnitude noted.
Preston-Martin et al. (1986) conducted a case-control study in which they
questioned 110 female cases aged 15 to 40 and an equal number of matched
controls. Diagnoses of cases were histologically confirmed, and thyroid disease
was recorded if a physician was consulted at least 2 years prior to the cancer
diagnosis. Statistically significant risk factors were found for thyroid
enlargement as an adolescent (OR=10) and any goiter or benign nodules (OR=10.5).
99
-------
The odds ratio of any thyroid disease was 14.5. The small number of cases of
I
follicular carcinoma prevented analysis by histological type.
Ron et al. (1987) also found increased risk with parity as well as increased
risk with goiter and nodules. This case-control study included 159 cases (109
female and 59 male) ascertained through a cancer registry and 318 controls
from the general population. A review of the pathology was included. Thyroid
nodules were evaluated separately from goiter and had a far greater4 risk (OR*33)
compared with goiter (OR=5.6); both were statistically significant. The authors
offer as caveats the fact that thyroid disease status was not medically verified
and the response rate was only 62%.
In conclusion, these three recent case-control studies in the United States
consistently showed thyroid cancer strongly related to pre-existing goiter and
to thyroid nodules (Table 12). There is insufficient evidence to identify a
quantitative difference in this relationship between follicular or papillary
tumor types. One concern is that the associations between thyroid disease and
thyroid cancer may be increased as a result of closer medical attention; after
all, there must have been some clinical indication that the patients may have
had a thyroid neoplasm prior to the time of surgery (like the presence of a
nodule in the gland). However, the consistency among studies, the strength of
the association, and the consistency with established causes (e.g., in all
studies, ORs were increased with radiation) strongly support the hypothesis
that thyroid nodules and, to a lesser degree, goiter are risk factors (potential
causes) of thyroid cancer in humans. It should be pointed out,, however, that
in the two studies that analyzed for an association between hypothyroidism and
thyroid cancer, neither showed a relationship (McTiernan et al., 1984; Ron et
al., 1987).
100
-------
VII. DEVELOPMENT OF SCIENCE POLICY
This section assembles pertinent points from the preceding review into a
rationale for a science policy. It then lays out a set of principles that will
help guide EPA in performing risk assessments on chemicals that have been shown
• or may have the potential to produce thyroid foilicular cell tumors.
A. RATIONALE FOR SCIENCE POLICY
Carcinogenesis is considered to be a multistage process in which a number
of endogenous and/or exogenous factors combine, either simultaneously or in
sequence, to disrupt normal cell growth and function. Consequently, chemical
carcinogenicity should not be viewed as a unique property of a chemical, but
rather as an outcome of the interaction of a chemical with a complex biological
system. A corollary to this is that cancer is a multifactorial disease that
may occur through a number of different mechanisms.
The development of cancer has often been divided into three major stages:
initiation, promotion, and progression. Initiation refers to the process whereby
a chemical or other agent permanently alters the DNA of the cell. Promotion
describes the subsequent processes involving the proliferation of the "transformed"
cell through several steps (e.g., hyperplasia, neoplasia) leading eventually to
a malignant tumor, while progression refers to the development of aggressive
cell behavior including local invasion and distant metastasis. It is now
recognized that initiation, promotion, and progression may each consist of
several stages involving different mechanisms. It is believed that some of
these stages are reversible and some are not; most appear to be susceptible to
modulation (enhancement or inhibition) by a variety of exogenous (e.g., diet,
stress, chemicals) or endogenous (e.g., age, sex, hormonal balance, health
status) factors.
101
-------
For most chemical substances one usually has too little relevant biological
information on mechanism of action to be able to evaluate if or how that agent may
be influencing the various stages of carcinogenesis. In some cases the results
of genotoxicity testing may give clues as to the potential to initiate carcino-
genesis, since initiation is thought to involve alterations in the DNA. However,
chemicals that can initiate carcinogenesis can very often also complete the
remaining stages in the carcinogenic process and lead to tumors.
Traditionally within EPA, chemicals that produce carcinogenic effects have
been assessed as if they are "complete" carcinogens with both initiation and
promotion components. Using this position as a basis, the Agency has generally
assumed that any exposure to the chemical substance is attended with some small
but finite risk of cancer. In modeling such dose-response relationships, an
extrapolation procedure which has a low-dose linear function has been employed
to estimate an upper bound on the additional lifetime cancer risks.
The 1986 EPA Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1986)
require the selection of a dose-response extrapolation model for each carcinogenic
agent under review. A similar directive, with guidance to aid in the selection,
is given to all federal agencies in the Office of Science and Technology Policy
Cancer Principles (OSTP, 1985). The EPA Guidelines say that the Agency "will
review each assessment as to the evidence bn carcinogenesis [sic] mechanisms
and other biological and statistical evidence that indicates the suitability of
a particular model." In the case of certain kinds of thyroid carcinogenesis,
there is considerable mechanistic information which can be used in making
judgments about model selection. The remainder of this section will be devoted
to laying out a rationale for assessing thyroid follicular cell carcinogenesis.
To fulfill their many critical functional roles, thyroid hormone levels in
the circulation are maintained under strict homeostatic control. Homeostasis
102
-------
is maintained primarily by a physiological feedback mechanism involving the
controlled synthesis and release of thyroid stimulating hormone (TSH) from the
pituitary in amounts that reflect the body's need for additional thyroid hormones,
Consequently, the thyroid and the. pituitary continually respond to both internal
(physiological) and external (environmental) stimuli that increase or decrease
the body's need for thyroid hormones. Failure to maintain homeostasis may
result in sustained increases or decreases in circulating levels of thyroid
hormones leading to hyperthyroidism and hypothyroidism, respectively.
Experimental studies in laboratory animals show that thyroid hyperplasia
and neoplasia are most often associated with prolonged exposure to excessively
high levels of TSH, irrespective of whether the latter results from endogenous
or exogenous stimuli. Thus, thyroid neoplasia may arise as a result of chronic
iodine deficiency, subtotal thyroidectomy, or the transplantation of hormonally
active pituitary tumors, all of which are associated with long-term elevated
TSH levels. Further evidence for the central role of TSH in the neoplastic
process is the finding that treatments that lower circulating levels of TSH
(e.g., hypophysectomy, thyroid hormone administration) prevent the development
of hyperplasia and neoplasia or cause the reversal of hyperplasia towards a
normal histological state.
Precise details of the mechanism through which prolonged elevated levels
of TSH may lead to thyroid neoplasia remain to be elucidated. Recent research
in molecular biology indicates that the induction of cell division (which can
lead to hyperplasia) and the change from normal to transformed (neoplastic)
cells are very complex processes. However, for certain thyroid tumors, some
of the steps seem to include the following. TSH interacts with thyroid cells
via specific plasma membrane receptors which leads to the induction of adenyl
cyclase and cellular protooncogenes (c-fos and c-myc). It appears that TSH-
103
-------
stimulated effects, in conjunction with the effects of other factors (e.g.,
i
somatomedins, epidermal growth factor, and phosphoinositol-mediated processes
involving protein kinase c), commit the thyroid cell to DMA synthesis and cell
division. Probably other interactions between TSH and other factors and influences
(e.g., mutation, oncogene activation) enhance cellular transformation. Thyroid
neoplasia, therefore, probably results from prolonged TSH stimulation in concert
with other cellular processes.
Thus, it would seem that experimental procedures (like subtotal thyroidectomy)
which stimulate increased levels of TSH may be influencing thyroid cells in at
least two different but not necessarily independent ways. First, TSH provides
a strong stimulus for cell division and the development of hyperplasia (oncogene
expression probably plays a role here). However, it seems that TSH has finite
ability to stimulate thyroid cell division both in vivo and ui vitro. Thus,
for thyroid cells to keep dividing as part of the carcinogenic process, it
would appear that they are responding to factors in addition to TSH, or the
cells themselves become changed. Second, TSH actions (like protooncogene
induction) in concert with other cellular processes lead (by some yet undis-
covered means) to neoplastic transformation.
Fitting the available information on thyroid follicular cell carcinogensis
into the "traditional" three-stage model of carcinogenesis—initiation, promotion,
and progression—is not easy (OSTP, 1985; Newell, 1986). Although the effect of
TSH (and other factors) on cell division is consistent with the concept of
promotion, a hypothesis for the way TSH might "initiate" the carcinogenic
process or enhance progression of neoplastic cells toward more malignant expression
(local invasion and distant metastasis) is less straightforward. Since little
is known about progression in thyroid carcinogenesis, remarks will be limited
to initiation. At this time it appears that oncogene expression is dependent
104
-------
upon the continued presence of TSH working via cyclic AMP. When TSH is removed,
the stimulus for oncogene expression probably ceases. Given that transformation
appears to require factors in addition to TSH, it is possible that the other
factors complete the transformation process (i.e., initiation) begun by TSH.
Another possibility would be that "spontaneous" events like mutations may occur
which complete the transformation. Since TSH, through its influence on cell
division, causes an expansion of the number of follicular cells at risk, it
seems that the total chance of a spontaneous neoplastic event would increase as
a function of the increase in cell number (assuming a constant probability of a
spontaneous mutation per cell). According to this reasoning, treatments that
increase thyroid cell number and increase mutations would be expected to enhance
the carcinogenic process; there is some support for this position. For instance,
regimens that combine a mutagenic agent (x-ray, genotoxic chemical) with an
increased output of TSH (e.g..iodide deficiency) result in an increase
carcinogenic response. The same is true for chemicals that are both mutagenic
and goitrogenic; for instance, 4,4'-methylenedianiline produces significant
increases in thyroid tumors in males and females of both rats and mice.
If the above hypothesis is valid, it would seem that TSH is not a direct
"initiator" of carcinogenesis, but rather it may allow cells to respond to
other stimuli that finally complete the initiation stage. Once transformation
occurs, TSH and other factors would be expected to promote carcinogenesis
through their influence on cell division.
Experimental observations with a number of chemicals are consistent with
the view that a major component in thyroid carcinogenesis results from prolonged
exposure of the thyroid to elevated levels of TSH. To this end, most of the
chemicals that have been shown to produce thyroid tumors in the NTP/NCI
carcinogenesis bioassay program have been compounds from structural classes
105
-------
(e.g., thionamides, aromatic amines) known to inhibit thyroid hormone synthesis.
Another, nonspecific group of compounds that have been shown capable of causing
thyroid neoplasia in laboratory animals are the inducers of hepatic mixed-function
oxidases. These materials enhance the hepatic metabolism and biliary excretion
of the thyroid hormones. The effects of both of these groups of compounds ~
inhibition of thyroid hormone synthesis or increased thyroid hormone metabolism
and elimination—result in decreased levels of circulating thyroid hormone and
a consequent increase in the level of TSH. Genotoxic activity did not correlate
with this type of thyroid carcinogenesis in any predictable way. Of the chemicals
reviewed, only the aromatic amines showed genotoxic activity for a variety of
end points.
Mechanisms other than TSH increases may influence thyroid tumor response.
For those substances where gene mutations and structural chromosome aberrations
may be induced, there is the possibility that a single or limited number of
chemical-cell interactions may influence carcinogenesis.
In evaluating the nature of the dose-response relationship for chemicals
that appear to have produced thyroid tumors via their influences on thyroid-
pituitary status and an increase in TSH, several points should be kept in mind.
Together these factors provide support for levels of TSH that are not associated
with carcinogenic risk (i.e.» subthresholds).
1. The proper maintenance of homeostatic control of circulating levels of
the thyroid hormones requires some optimal, non-zero level of TSH.
2. TSH and the thyroid hormones must be continually replaced, since their
residence in the body is finite (T4: rat plasma tx/2 = 12-24 hours; human plasma
t1^ = 5-9 days) (see Thomas and Bell, 1982).
3. The feedback mechanism through which thyroid homeostasis is maintained
depends ultimately on TSH-stimulated thyroid hormone synthesis by the thyroid
106
-------
gland. The de novo synthesis of thyroid hormones is dependent on two separate
receptor-mediated, dose-dependent steps. One of these occurs in the pituitary
gland where receptors monitor circulating levels of the thyroid hormones and
respond by releasing appropriate amounts of TSH; the second occurs in the
thyroid gland itself where receptors respond to TSH. Hormone-receptor complexes
are short lived and a supply of hormones must be present on an ongoing basis to
interact with their receptors (T3-receptor dissociation tl/2 = 15 min). In both
the pituitary and the thyroid there exists a large number of receptors (several
thousand per cell) for thyroid hormones and TSH, respectively, and the response
of each gland is likely to be graded in nature and dependent on the number of
receptors occupied at any one time. By analogy with other receptor-mediated
reactions and from the information accumulated on the binding of T3 by pituitary
cell receptors, a large number of receptors must be occupied to elicit a response.
4. The effects of excessive TSH on thyroid cell histology/pathology (e.g.,
hypertrophy, hyperplasia) are reversible if the TSH stimulus is removed early
in the process.
5. Thyroid cell proliferation and transformation involve several different
steps and require a number of factors in addition to TSH. Some of the factors
that may be operative work through receptors themselves and most likely require
multiple site occupancy for effect.
6. Thyroid carcinogenesis seems to require long-term disruption in thyroid-
pituitary status leading to elevated levels of TSH (and reduced levels of the
thyroid hormones).
Humans appear to be quite similar to laboratory animals in their responses
to goitrogenic stimuli. Thus, iodine deficiency, partial thyroidectomy (surgical
or 131i), and administration of antithyroid agents (e.g., thionamides) result in
reduced thyroid hormones levels and increased levels of TSH, and can lead to
107
-------
thyroid hypertrophy and hyperplasia. As In experimental animals thyroid enlarge-
ment and nodular lesions have been implicated as possible antecedents tp thyroid
cancer in humans.
In spite of these qualitative similarities, however, there is some evidence
that humans may not be as sensitive quantitatively to thyroid cancer as exper-
imental animal species. For instance, experimental animals readily respond to
reduced iodide intake with thyroid cancer development. The case with humans j.$
much less certain. Although there is profound hyperplasia with "adenoroatpus1'
changes, the case for malignant transformation is only suggestive and has not
been demonstrated with any certainty. Even with congenital goiters where there
are inherited blocks in thyroid hormone synthesis, only a few thyroid cancers
have been reported in the literature. Humans also may .be less sensitive to the
effects of 131i. Although the data are very soft, there does not seem to be any
profound indication of a cancer problem in persons with Graves' disease where £
significant proportion of patients have .aut.oa.Qtibodies that stimulate the
thyroid like TSH. In a like manner, these same patients treated with antithyrpid
compounds do not seem to show increases in thyroid cancer.
In contrast to the observations mentioned above, the finding of thyroid
cancer in human autopsy studies in the United States is not unlike that seen in
animal studies. For instance, about 1 percent of control Fischer 344 rats
develop thyroid cancer over a lifetime, while autopsy prevalences of tumors in
humans that were not noted during life range from 0.9 to 5,7 percent (about 2
percent average) in different studies. Few of the human tumors are manifest,,
since clinically significant thyroid tumors occur in only about 3 of 10.0.,,POP
persons and constitute only about 0,5 percent of all cancer deaths.
108
-------
B. SCIENCE POLICY
It is generally accepted that carcinogenesis ,s a long-term, complicated
and multistep process with numerous causes. Although it is very difficult to
prove that carcinogenesis proceeds via specific, discrete steps, 1n certain
cases accunulated evidence becomes persuasive enough to presume that certain
processes are operative, and this infection can be used as the basis for an
approach to estimate human cancer risk. This is the case for the induction of
certain folliailar cell neoplasms of the thyroid gland.
Studies over the last several decades in multiple laboratories and using a
number of different treatment regimens (e.g., iodine deficiency) have demonstrated
the significance of long-term thyroid-pituitary hormonal imbalance in thyroid
carcinogenesis. A consistent progression of events is noted: reduction in
thyroid hormone concentrations, elevation in TSH levels, cellular hypertrophy
and hyperplasia, nodular hyperplasia, and neoplasia. Hyperplasia and sometimes
neoplasia of the pituitary may also be seen. A block in any of the early steps
acts as a block for subsequent steps including tumor development, and cessation
of treatment at an early stage in the progression results in regression toward
normal thyroid structure and function. Based on these observations and the
rationale set out above in Section VILA., the Agency concludes that:
1. thyroid follicular cell tumors may arise from long-term disturbances
in thyroid-pituitary feedback under conditions of reduced circulating
thyroid hormone and elevated TSH levels;
2. the steps leading to these tumors are expected to show thresholds,
such that the risks of tumor development are minimal when thyroid-
pituitary homeostasis exists; and
109
-------
3. models that assume thresholds may be used to assess the risks of
thyroid follicular cell tumors where there is evidence of thyroid-
pituitary hormonal imbalance.
There are special considerations that must be addressed before applying
this policy to any chemical substance that has produced thyroid tumors and is
subject to review by the Agency. It is recognized that some thyroid tumors
seem to arise from mechanisms other than thyroid-pituitary imbalance. It is
also known that chemical substances may impact living cells in a number of
different ways and, therefore, may be producing toxic effects by different
mechanisms. Thus, two basic questions must be satisfactorily addressed in the
risk assessment of chemicals under review in determining whether and how to
apply the policy. The first is a qualitative issue which addresses whether it
is reasonable to presume that the neoplasms are due to thyroid-pituitary imbalance.
A corollary issue is the extent to which other carcinogenic mechanisms can be
discounted. The second question concerns the procedures to be employed in
estimating the risks of these agents. Criteria for addressing these issues are
developed below.
The answers to the first question allow one to assign chemicals producing
thyroid tumors to one of three categories. The assignation is based upon
knowledge as to whether the chemical disrupts thyroid-pituitary feedback,
whether tumors other than thyroid follicular cell (and relevant pituitary)
tumors are found, and whether mechanisms other than thyroid-pituitary imbalance
may apply to the observed tumor response. The guidance on how to proceed with
the quantisation of risk varies with the category, as follows.
1. Threshold considerations should be applied in dose-response assessments
for those chemical substances where (a) only thyroid tumors (and
relevant pituitary tumors) have been produced; (b) the tumors can be
110
-------
attributed to a disruption in thyroid-pituitary hormonal homeostasis;
and (c) potential mechanisms other than thyroid-pituitary imbalance
(e.g., genotoxicity) can be disregarded.
2. Special attention should be given to chemicals (a) that have induced
thyroid tumors (and relevant pituitary tumors) that may be due to
thyroid-pituitary imbalance, and (b) where there is also evidence of
either a genotoxic potential or the induction of neoplasms at sites
other than the thyroid (or pituitary). Generally, those cases will
be approached using various principles laid out in the EPA Guidelines
for Carcinogen Risk Assessment. A strong rationale must be articulated
for handling these agents otherwise.
3. For those chemicals producing thyroid tumors that do not seem to be
acting via thyroid-pituitary hormonal inhibition, dose-response
assessments will be performed in accordance with the EPA Guidelines
for Carcinogen Risk Assessment.
The application of this guidance is contingent upon the careful assessment
of all information hearing on the carcinogenicity of each chemical subject to
review. It calls for an evaluation of the types of thyroid (and pituitary)
tumors and any other tumor types as well as preneoplastic and other toxicological
lesions that are produced.. It also requires a careful analysis of relevant
mechanistic information bearing on the assessment of carcinogenicity. In
certain cases data gaps may necessitate further testing and research before an
assessment based on this policy can be completed. The remainder of this section
will be devoted to a discussion of some of the factors that should be considered
in the assessment of chemicals producing thyroid tumors.
One essential factor is whether the thyroid tumors can be attributed to
disruption of thyroid-pituitary hormonal balance. In addressing whether this
111
-------
is the case, the presence of several indicators should be considered.
1. Goitrogenic activity in vivo (i.e., thyroid follicular cell hypertrophy
and hyperplasia)
2. Clinical chemistry indication of changes in thyroid and pituitary
functional parameters (e.g., reduced thyroid hormone arid increased TSH
serum concentrations).
3. Specific evidence that the agent either reduces thyroid hormone synthesis
(e.g., inhibits iodine uptake) or increases thyroid hormone clearance
(e.g., enhances biliary excretion).
4. A progression of lesions under long-term exposure to an agent, showing
cellular hypertrophy and hyperplasia, nodular hyperplasia, and neoplasia
(benign and possibly malignant tumors).
5. Other studies bearing on the hypothesis that thyroid-pituitary imbalance
may be operative, like reversibility of lesions following cessation of
the treatment.
6. Structure-activity analysis of the agent under review to see if it
belongs to a class of compounds that shows a correlation with the
induction of thyroid tumors.
For each chemical that shows thyroid follicular cell carcinogenic effects,
the above points are reviewed as a whole, and an overall judgment is made as to
the likelihood the tumors may be due to a disruption in thyroid-pituitary status.
Since the data base on chemicals will vary considerably, precise criteria as to
what constitutes adequate evidence cannot be given, but at a minimum information
from items 1, 2, and 3, with some indication of dose-response, are essential
in making these judgments. In addition, several other of the above lines of
evidence in support of the hypothesis are valuable, and while it is unlikely
that one will ever amass direct proof of the hypothesis, enough supportive
112
-------
information should be available so that the position is scientifically
reasonable.
Another important point is the extent to which genotoxicity may account
for the observed tumor effects. Short-term jn vitro and in vivo tests for
various end points, including gene and chromosomal mutations and DNA-damaging
capability, should be reviewed to get an idea of the spectrum of effects that
may be produced in somatic cells. V It is recognized that a chemical seldom
produces all positive or all negative responses in such tests; therefore, a
case-by-case judgment must be made of the likelihood the chemical's carcinogenic
effects may be due to its genotoxic activity. It should be pointed out that
for the purposes of this policy, it is necessary to evaluate potential carcinogenic
mechanisms and not just the correlation between short-term test results and
carcinogenicity.
Certain short-term test end points have more intuitive relevance to carcino-
genicity than others. End points such as gene and chromosomal mutations readily
fit into what is known about carcinogenic mechanisms, whereas less can be said
about the applicability of other end points like sister chromatid exchange or
mitotic gene conversion. Even with mutations, there are different ways, at
least theoretically, that chemicals might induce them and that may be relevant
to dose-response considerations. It is conceivable that gene mutations arise
from single (or a limited number of) chemical-cellular interactions, whereas at
least two (and probably more) would be required for stable structural aberrations,
and most likely many interactions would be needed to induce numerical chromosome
aberrations.
- The Agency s Guidelines for Mutagenicity Risk Assessment should be consulted
but with the understanding that making a judgment of mutagenic risk to future
generations germ cell risk) involves an important aspect not relevant to carcino-
genicity evaluation Valuable reviews of short-term tests and testing results
can be found in publications of the EPA Gene Tox program (U.S. EPA, 1988).
113
-------
Neoplasms that occur in addition to thyroid follicular tumors (and relevant
pituitary tumors) must be carefully evaluated as to mechanistic considerations.
There is no a priori evidence that elevation in TSH serum concentration is
associated with tumors at sites other than the thyroid. However, it is recognized
that most organs of the body are responsive to thyroid hormone, and thus tteo-
plastic development may in some way be modified under conditions that result in
reduced circulating thyroid hormone concentration. This eventuality should be
considered and evaluated. In addition, the role of target-organ toxicity,
immunologic suppression and any other relevant biological properties of the
chemical under study should be reviewed in assessing the significance of these
tumors. Metabolic and pharmacokinetic considerations are also relevant.
The last major point in the evaluation of thyroid carcinogens is the way
to quantitate carcinogenic risk when it is judged that the tumors are associated
with thyroid-pituitary imbalance and threshold concepts apply. The traditional
way the Agency has dealt with thresholds is to use a no-observed-adverse effect
level (NOAEL) for the critical effect as a measure of potency and then to use
uncertainty factors to estimate exposure levels (dose rates) where it is antic-
ipated there will be little chance of risk in humans. An alternate means of
expressing a degree of concern is to calculate a margin of exposure, the ratio
of the NOAEL to anticipated human exposure. The larger the ratio, the less
likely the exposure will be cause for concern, while the smaller the ratio, the
greater the concern. Historically, the Agency has used threshold concepts to
evaluate the risks related to target-organ and systemic toxicity and developmental
114
-------
and reproductive toxicity. .£/ ' •
Based on current Agency practices, the procedures used for evaluating risks
from systemic toxicants and other threshold-relevant end points may be employed.
For those thyroid follicular cell tumors that are conceived as arising from an
imbalance in the thyroid-pituitary axis, one needs to use toxicological parameters
that give some indication that thyroid-pituitary homeostasis has been disrupted.
End points that should be considered as bases of NOAELs include such things as
increases in thyroid weight, decreases in circulating thyroid hormone, and
increases in TSH concentration as a function of chemical dose. It is expected
that these end points will show deviations from normal at doses lower than or
equal to those showing increases in thyroid tumors. A NOAEL is determined for
each meaningful toxicological end point, and the one from among those reviewed
that demonstrates the lowest NOAEL is called the critical effect, that is, the
most sensitive indicator of a pertubation in thyroid-pituitary balance. Either
uncertainty factors or anticipated human exposure are used with this NOAEL to
calculate measures associated with human risks. When chemically exposed groups
of humans are available, clinical chemistry measurements (e.g., serum TSH
concentrations) and other measures are useful in evaluating risks by comparing
the distribution of values in this group as compared to a control group.
In evaluating thyroid follicular cell neoplasms under this policy, the risk
assessment depends on full use of the available information. As indicated above,
£/ Because the Agency recognizes that the traditional techniques are not
necessarily the most sophisticated means of extrapolating risks it is
important to investigate alternative means of extrapolatinq risks in situations
involving thresholds. Some of these alternatives are being considered for
investigation or are already under development within the Agency. These include
the use of a combination of high-to-low dose modeling and uncertainty factors
and considerations of initiation-promotion phenomena in biologically based
models. The Agency should actively pursue the application of some of these
alternatives to the evaluation of human risks for thyroid "threshold" carcinogens
in place of the traditional way the Agency has dealt with threshold considerations
115
-------
in any given organism, a carcinogen may act through more than one mechanism at
one or multiple anatomical sites. Accordingly, while use of this policy may be
appropriate for assessing certain thyroid follicular cell tumors, use of other
models may be necessary to evaluate risks at other tumor sites observed in the
same study, which may result in different risk estimates. It is incumbent upon
the risk assessor to consider all relevant risk estimates in making the final
judgments on the potential human risk related to exposure to the chemical being
evaluated.
116
-------
APPENDIX A
COMBINED TREATMENT STUDIES
Test Animal Treatment A
Treatment B
Results
Reference
Wistar rat
(female)
Lister rat
(male &
female)
Lister rat
(male &
female)
Wistar rat
(male)
Wistar rat
(male)
AAF (2.5 mg
gavage, 4-6x
for one week)
AAF (100 mg/L
in drinking
water for
13 mo.
131I (30 uCi,
X-rays
(300 rad
to neck)
DHPN (70 mg/
100 g bw gi ven
sc once/wk for
4 or 8 wk)
MTU (0.1 g/L
in drinking
water up to
21 wk)
MTU (1 g/L in
drinking water
for 13 mo. con-
current with AFF).
MTU (1 g/L in
in drinking
water for
15 mo)
MTU (1 g/L in
drinking water
for 15-18 mo)
Amitrole (2000 ppm
in diet for 12 wk)
Combined treatment
showed multiple
adenomas/gland. MTU
alone caused hyper-
plasia or single tumors.
AAF stated as having
no tumor effect
Combined treatment
showed multiple
adenomas when interval
between treatments
extended for 4-18 wk.
Combined treatment
showed more adenomas/
gland than single
treatment groups.
Combined treatment
produced more adenomas/
gland and malignancies
not seen in single
treatment groups.
Combined treatment
increased incidence of
tumor-bearing animals
and malignancies that
were not seen with
single treatments.
Amitrole after 4 wk
of DHPN induced thyroid
adenomas at 91% and
carcinomas at 9%.
No tumors with DHPN
or amitrole alone.
Amitrole accelerated
development of adenomas
and increased carcinomas
after 8 wk of DHPN (no
amitrole - 58% adenomas,
18% carcinomas; with
amitrole - 100% adenomas,
42% carcinomas). No tumors
with amitrole alone.
Hall, 1948
Doniach, 1950
Doniach, 1953
Christov, 1975
Hiasa et al.,
1982a
(continued on the following page
A-l
-------
APPENDIX A. (continued)
Test Animal Treatment A
Treatment B
Results
Reference
Histar rat
(male)
DHPN (70 mg/
100 g bw given
sc once/wk for
for 4 or 6 wk)
PB (500 ppm in
.diet for 12 wk)
BB (500 ppm in
diet for 12 wk)
Wistar rat
(male)
Wistar rat
(male)
DHPN (single
sc dose of
280 mg/100
g bw)
DHPN (single
sc dose of
280 mg/100
g bw)
PB (500 ppm
in diet for
6, 12 or 19
wks)
PTU (1500 ppm
in diet for 19
wk)
PB after 4 wk of DHPN Hiasa et a!.,
induced thyroid adenomas 1982b
at 66% and carcinomas at
10%. No tumors with DHPN
or PB alone.
PB after 6 wk of DHPN
acclerated development
of adenomas and induced
carcinomas (no PB-23%)
adenomas, no carcinomas;
with PB-100% adenomas,
251 carcinomas; no tumors
with PB alone).
PB after 4 wk of DPHN
induced thyroid adenomas
(23%) but no carcinomas.
No tumors with BB alone.
BB after 6 wk of DHPN '
accelerated development
of adenomas and induced
a small number of
carcinomas (no BB
- 23% adenomas, no
carcinomas; with BB -
45% adenomas, 10%
carcinomas; no tumors
with BB alone).
PB for 12 or 19 wk Hiasa et al,,
after DHPN enhanced 1983
development of thyroid
adenomas. PB for 19 wk
after DHPN induced
thyroid carcinomas at 12%.
Mot seen with DHPN alone. '
PB alone produced no tumors.
PTU after DHPN enhanced Kitahori
development of thyroid et at.,
follicular cell adenomas 1984.
and induced carcinomas
(no PTU - 19% adenomas,
0% carcinomas; with PTU
- 100% adenomas, 52%
carcinomas). PTU alone
produced no tumors.
(continued on the following page
A-Z
-------
APPENDIX A. (continued)
Test Animal
Wistar rat
(male)
F344/NO
rat
(male)
F344/NO
rat
(male)
Treatment A
DHPN (single
ip dose of 280
mg/lOOg bw)
NMU (single iv
dose of 41.2
ing/kg bw)
NMU (single
iv dose of
41.2 mg/kg
Hi»i \
UW )
Treatment B
MDA (1000 ppm
in diet for 19
wk)
Iodine deficient
diet after 2 wk
until 20 or 33
• \
wk)
Iodine deficiency
after 2 wk until
52 and 77 wk
Results
MDA after DHPN enhanced
development of thyroid
tumors and induced
carcinomas (no MDA -
28% tumors, 0% carci-
nomas; with MDA - 90$
tumors, 9.5% carcin-
omas). MDA alone
produced no tumors.
Iodine deficiency after
NMU enhanced development
of thyroid follicular cell
adenomas and carcinomas
(NMU alone - 10% adenomas
at 20 wk and 70S adenomas
at 33 wk, 10% carcinomas
at 33 wk; NMU with iodine
deficiency - 100$ adenomas
at 20 wk and 100% carcinomas
at 33 wk; no tumors following
iodine deficiency alone).
Iodine deficiency after
NMU enhanced development
of the thyroid follicular
cell carcinomas (NMU alone;
32$ carcinomas at 52 wk;
NMU with iodine deficiency
90% at 52 wk).
Reference
Hiasa
et al . ,
1984
Ohshima and
Ward, 1986
Ohshima and
Ward, 1984
Wistar
rat
(female)
NMU (40 mg/kg
bw by gavage
for 3 days)
MTU [1 g/L in
drinking water
from 4 wk after
NMU until death
(60 wk)
Iodine deficiency alone
induced mostly thyroid
adenomas and a few carcin-
omas (40$ adenomas at 52 wk,
60$ adenomas at 77 wk, and
10$ carcinomas at 77 wk).
Combined treatment resulted
in appearance of thyroid
follicular cell adenomas
(within 13 wk) and carcinomas
(after 16 wk) that metasta-
ized to the lung (after
30 wk). No single treatment
groups were included, and the
fate of untreated controls was
not described.
Schaffer
and Muller,
1980
(continued on
lowing page)
A-3
-------
APPENDIX A. (continued)
Test Animal Treatment A Treatment B
Results
Reference
F344
rat
(female)
NMU fsingle
iv dose of 50
rag/kg bw)
PTU (3, 10, and
30 mg/L in
drinking water)
uCI)
(1 and 10
F344 rat
(male)
NMU (20 mg/kg
1p 2x/wk for 4
wk)
PB (0.05% in
diet for 32 wk)
PTU after NMU induced
development of thyroid
adenomas and carcinomas
{NMU alone - no tumors;
with 3 mg/L PTU - 17%
adenomas, 23% carcinomas;
with 10 and 30 mg/L PTU
- 100% carcinomas). No
PTU alone group was
included.
No thyroid tumors.
PB after NMU-induced
thyroid papillary
carcinomas. NMU alone
did not induce tumors.
PB was not tested alone.
Milmore
et al., 1962
Tsuda et al.,
1983
KEY: AAF, 2-acetylaminofluorene; MTU, 4-methyl-2-thiouracil •
DHPN, N-bls(2-hydroxypropy1)nitrosamine; amitrole, 3-ami no-1,2,4-tri azole;
EM, P5 ™*hrbi«a1-\BB> barblta1; PTU> Propylthiouracil; MDA, methylenedianiline;
NMU, ^-methyl-N-mtrosourea.
-------
APPENDIX B
SINGLE RING AROMATIC AMINES
Several structurally related, single-ring aromatic amines have been tested
for carcinogenicity and are illustrated in the accompanying table. Of the 11
structural analogues, only o-anisidine, (no. 1), 2,4-diaminoanisole (no. 2),
3-amino-4-ethoxyacetanilide (no. 3), and HC Blue No. 1 (no. 9) were positive for
thyroid tumors.
Although the first three chemicals share amino and methoxy substituents in
the ortho position on the ring, other tested chemicals with this conformation
(no. 4, no. 5) did not produce thyroid tumors. Both chemicals, no. 2 and no.
3, have amino groups in the meta position on the ring; however, compound no. 8,
which also has this configuration, lacked thyroid tumor activity. Chemicals
no. 2 and no. 3 also shared amino and methoxy groups in the para positions;
compounds no. 6 and 7 with these constituents were negative for thyroid tumors.
Likewise, for HC Blue No. 1 (no. 9), which showed a thyroid tumor response in
the NTP bioassay, structural analogues no. 10 and 11 failed to show this response
Thus, it is not readily apparent which, if any, substitutions on the ring may
impact thyroid tumor activity.
B-l
-------
APPENDIX B. (continued)
SINGLE RING AROMATIC AMINES:
STRUCTURE-ACTIVITY RELATIONSHIPS AMONG CHEMICALS TESTED BY THE NCI/NTP
1.
2.
o-anisidine
NH2
/0\-OCH3
2,4-diaminoanisole
Thyroid Tumors Other
Rat Mouse Rat
M F M F M F
+ bladder bladder
kidney
+ + + + skin skin
liver liver
Tumors
Mouse
M F
bladder bladder
i
liver
NH2
NH2-/0\-XH3
3. 3-amino-4-ethoxy-
acetanilide
NH2
/ "p\
CH3-C~NH-/ 0 VOCH3
4. p-cresidine
NH2
CH3-/0\-XH3
5. 5-nitro-o-anisidine
NH2
N02-/0\-OCH3
bladder bladder bladder bladder
nasal nasal
liver - - liver
ski n ski n
zynto al zymb al
clitoral liver
gland
(continued o.n the foilowi:ng;pagej
B-2
-------
APPENDIX B. (continued)
6. p-anisidine
NH2-/~0\OCH3
7. 3,4-dlmetnoxyanil ine
,OCH3
NHW 0 VOCH3
8. m-di phenyleneaxine
NH2
9. HC Blue No. 1
N02
/*' '"•' "\
Thyroid Tumors
Rat Mouse
M F M F
Rat
Other Tumors
M
(HOCH2-CH2)2N-( 0 )-NH-CH3
\ _ /
10. p-phenyl enedlamlne
NH2-/T\-NH2
11. 2-nitro-p_- phenyl ene-
dlamlne
N02
. _ /
Y 0\-NH2
\ . .. ./
Mouse
M
liver lung liver liver
B-3
-------
-------
APPENDIX C
GENOTOXICITY: ETHYLENE THIOUREA
1. GENE MUTATIONS
A. BACTERIA
Salmonella (Ames)
Reported
Effect
N-nitrosoethylenethiourea (+)
"
G46
G46
multiple strains
mouse/rat host mediated
G46
multiple strains
mouse host mediated
646, TA 1530
multiple strains
TA 1950
mouse host mediated
(TA 1950)
multiple strains
mouse host mediated
(TA 1950)
mill ti pi e strai ns/repl i cati ons
in different labs
multiple strains/replications
in different labs
E. coli
(-N02")
(+N0p-)
(-NO?)
2
(w)
-?
(+ N0-)
(-N0?v)
WP2
WP2
(-NO?-)
(-)
<+>
(+) TA 1530
only
(+) TA 1530
only
(+) in all
(w)
(^)
(w)
'
(w) TA 1535
only
(w) TA 1535
(-) all
others
Reference
Seller, 1974
Seller, 1977
Shirasu et al.,
1977
Schupbach and
Hummler, 1977
Anderson and
Styles, 1978
Autio et al ^
1982
Moriya et al.,
1983
Braun et al.,
1977
Mortelmans
et al., 1986
Bridges et al.,
1981
Shirasu et al .,
1977
KEY: (+) positive
(w) weak positive
(?) equivocal
(-) negative
(continued on following page)
C-l
-------
APPENDIX C. (continued)
B. EUKARYOTIC MICROORGANISMS
Saccharomyces (XV 185-14C)
Schlzosaccharomyces
C. HIGHER EUKARYOTES
Mouse lymphoma cells (TK)
Mouse lymphoma cells
Chinese hamster ovary
(several loci)
Drosophila XLRL
Drosophila XLRL
Drosophila XLRL
2. CHROMOSOME EFFECTS
A. NUMERICAL ABERRATIONS
Saccharomyces mitotic aneuploidy
Mouse micronucleus (see B, below)
B. STRUCTURAL ABERRATIONS
Chinese hamster ovary cells
Chinese hamster ovary cells
Chinese hamster ovary cells
Mouse micronucleus (B6C3F1)
Mouse micronucl eus (ICR)
Mouse micronucleus (CD-I)
(+) requires Mehta and
S9 vonBorstel, 1981
(-) Loprieno, 1981
Jotz and Miitchel, 1981
NTP, 1986
Carver et al., 1981
(-) Valencia and
Houtchens, 1981
(-) injection Woodruff et al,, 1985
(?) feeding
NTP, 1986
Parry and Sharp, 1981
Shirasu et al., 1977
Natarajan and
vanKesteren-van
Leeuwen, 1981
NTPS 1986
Salamone et al., 1981
Kirkhart, 1981
Tsuchimoto and
Matter, 1981
(continued on following page)
KEY: (+) positive
(w) weak positive
(?) equivocal
(-) negative
C-2
-------
APPENDIX C. (continued)
Mouse micronucleus (-NaN02)
Mouse micronucleus
Mouse dominant lethal
Mouse dominant lethal
Mouse dominant lethal
Mouse dominant lethal (+ NaN02) {+)
preimplantation loss
postimplantation loss
3.
Chinese hamster bone
marrow
Rat bone marrow
Drosophila reciprocal
traiislocation
C. SISTER CHROMATID EXCHANGES
Chinese hamster ovary cells
Chinese hamster ovary cells
Chinese hamster ovary cells
Chinese hamster ovary cells
Mouse in vivo (CBA/J)
DNA DAMAGE
1- S"btil is (rec)
jL- £211 (pol A)
KEY: (+) positive
(w) weak positive
(?) equivocal
(-) negative
(w) without
S9
{-) with S9
Seller, 1975
Schupbach and Hummler, 1977
Shirasu et al., 1977
Schupbach and Hummler, 1977
Schupbach and Hummler, 1977
Teramoto et al., 1978
Seiler, 1977
Shirasu et al., 1977
NTP, 1986
Evans and Mitchel, 1981
Natarajan and
vanKesteren - van
Leeuwen, 1981
Perry and Thomson, 1981
NTP, 1986
Paika et al., 1981
Kada, 1981
Green, 1981
(continued on following page)
C-3
-------
APPENDIX C. (continued)
£. col 1 (rec)
E.* col 1 (rec, pol A)
E. coli (pol A)
JL* coli (1 anb da induction)
Saccharomyces mitotic cross-over
Saccharomyces mitotic gene
conversion
Saccharomyces mitotic gene
conversion
(+) with S9
(w) without
S9
(-) with S9
Ichinotsubo et al., 1981
Tweats, 1981
Rosenkranz et al., 1981
Thomson,, 1981
Kassinova et al.» 1981
Jagannath et al., 1981
Zimmemann and
Scheel, 1981
Saccharomyces (JDI) mitotic gene (+) without Sharp and Perry, 1981a
conversion S9
Saccharomyes (RAD) differential growth
Unscheduled DNA synthesis
WI-38 cells (-)
Human fibroblasts (-)
Mouse sperm morphology (-)
Mouse sperm morphology (-)
Sharp and Perry, 1981b
Robinson and
Mitchell, 1981
Agrelo and Amos, 1981
Wyrobek et al., 1981
Tophan, 1980
4. IN VITRO TRANSFORMATION
Baby hamster kidney (BHK 21)
Baby hamster kidney (BHK 21)
Syrian hamster embryo, adenovirus (-)
infected (SHE-SA7)
Daniel ctnd Dehne, 1981
Styles, 1981
Hatch et al., 1986
Key: {+) postive
(w) weak positive
(?) equivocal
(-) negative
C-4
-------
APPENDIX D
GENOTOXICITY: 4,4'-OXYDIANlLINE
1. GEME MUTATION
A. BACTERIA
Salmonella (Ames)
TA 98
TA 100
TA 98
TA 100
TA 98
TA 100
TA 97
TA 98
TA 100
TA 1535
TA 1537
Reported Effect
(+) requires S9
(+) assayed only in
presence of S9
(w) requires S9
( + ) requires S9
(+) requires S9
(+) requires S9
(+) requires S9
(+) requires S9
( + ) with or without S9
(+) requires hamster S9
(+) assayed only with S9;
requires hamster S9
Reference
Lavoie et al.,
1979
Parodi et al.,
1981
Tanaka et al.',
1985
NTP, 1987
(personal
communication
Dr. Errol
Zeiger)
B. EUKARYOTES
Mammalian cell s in culture
Mouse lymphoma
2. CHROMOSOME EFFECTS
Chinese hamster ovary cells
structural chromosome (+)
aberrations
sister chromatid
exchanges
Rat bone marrow
sister chromatid
exchanges
NTP, 1986
NTP, 1986
Parodi et al.,
1983
KEY:
(+) positive
(w) weak positive
(?) equivocal
(-) negative
(continued on following page)
D-l
-------
APPENDIX D. (continued)
3. DMA DAMAGE
Unscheduled DNA synthesis
(rat hepatocytes)
_™ vivo (-) Mirsalis
in vitro (-) et al,, 1983
4. IN VITRO TRANSFORMATION
Syrian hamster embryo cells (?) Tu et a]., 1986
Enhancement of vi rus Hatch et al,,
infected transformation of
Syrian hamster embryo cells
Key: (+) - positive
(w) - weak positive
(?) - equivocal
(-) - negative
D-2
-------
APPENDIX E
GENOTOXICITY: AMITROLE
1. GENE MUTATIONS
A. BACTERIA
Reported Effect
Salmonella (Ames)
TA 1950, mouse host
mediated (~N02~)
(+N02-) (w)
E." col i
(WPZuvrA (P))
(WP2uvrA)
(WP2uvrA/pKM101)
Streptomyces
B. EUKARYOTIC MICROORGANISMS
Saccharomyces (RV)
(w)
KEY: (+) positive
(w) weak positive
(?) equivocal
(-) negative
Reference
See multiple bacterial
tests summarized in
Bridges et al., 1981
McCann and Ames, 1976
Braun et al., 1977
Dunkel, 1979
Rosenkranz and Poirier, 1979
Moriya et al., 1983
NTP, 1986
Venitt and Crofton-
Sleigh, 1981
Matsushima et al., 1981
Matsushima et al., 1981
Carere et al., 1978
Mehta and vonBorstel, 1981
(continued on following page)
E-l
-------
APPENDIX E. (continued)
C. HIGHER EUKARYOTES
Drosophila XLRL
- -
(feeding, ?; injection,
Mouse lymphoma L5178Y cells (TK) (-/-/-)
Syrian hamster enbryo cells
(ouabain)
(6-thioguanine)
2. CHROMOSOME EFFECTS
A. NUMERICAL ABERRATIONS
Saccharomyces (D6)
Aspergillus mitotic
nondisjunction
Drosophila sex chromosome
nondisjunction
B. STRUCTURAL ABERRATIONS
Human lymphocytes jjn vitro
Mouse micronucleus (B6C3F1)
(CD-I)
Mouse dominant lethal
(Ha, 1 CR)
(w)
Laamanen et al., 1976
Vogel et al., 1980
Vogel et al., 1981
NTP, 1986
Woodruff et al,, 1985
NTP, 1986
Tsutsui et al., 1984
Tsutsui, et al., 1984
Parry and Sharp, 1981
Bignami et al., 1977
Laamanen et al., 1976
Meretoja et al;, 1976
Sal omone, et al., 1981
Tsuchimoto and
Matter, 1981
Food and Drug Res., 1978
KEY: (+) positive
(w) weak positive
(?) equivocal
(-) negative
(continued on the following page)
E-2
-------
APPENDIX E. (continued)
C. OTHER EFFECTS
sister chromatid exchange
(CHO)
(CHO)
3. DMA DAMAGE
Bacillus sub til is
E. coli
Rec
Rec
Rec
Rec
Rec
Pol A
Lambda prophage
induction
Saccharomyces cerivisiae
(D3) mitotic
cross over
(race XI1)
mitotic cross over
(D4) mitotic gene
conversion
(D7) mitotic gene
conversion
(JD1) mitotic
gene conversion
Perry and Thomson, 1981
NTP, 1986
Kada, 1981
Green, 1981
Ichinotsubo et al., 1981
Mamber et al., 1983
Tweats, 1981
Rosenkranz et al., 1981
Thomson, 1981
Simmon, 1979
Kasinova et al., 1981
Jagannath et al.,
1981
Zimmerman and
Scheel 1981
Sharp and Perry, 1981
1981a
KEY: ( + ) positive
(w) weak positive
(?) equivocal
(-) negative
(continued on the following page)
E-3
-------
APPENDIX E. (continued)
(RAD) cell growth
Aspergillus mitotic cross
Unscheduled DMA synthesis (HeLa)
MLV integration enhancement
(C3H2K)
Mouse sperm head abnormality
4. IN VITRO TRANSFORMATION
Syrian hamster embryo cells
Baby hamster kidney cells (BHK)
Rat Embryo cell's
(Rauscher murine leukemia
virus infected)
(w)
Sharp and Perry, 1981b
Bignami et al,, 1977
Martin and McDermid, 1981
Yoshikur and Matsiishima, J,98J
Tophan, 1980
Dunkel et al,, 1981
Tsutsui, et al., 1980
Styles, 1980
Styles, 1981
Daniel and Dehnel, 1981
Dunkel et al., 1981
NTP, 1983
KEY:
(+)
(w)
(?)
(-)
- positive
- weak positive
- equivocal
- negative
E-4
-------
REFERENCES
Adams, J.M.; Harris, A.M.; Pinkert, C.A.; Corcoran, L.M.; Alexander, U.S.
Cory, S.; Palmiter, R.D.; Brinster, R.L. (1985) The c-myc oncogene
driven by immunoglobulin enhancers induces lymphoid malignancy in transqenic
mice. Nature 318:533-8.
Agrello C.; Amos H. (1981) DNA repair in human fibroblasts. In: de Serres, J.;
•Ashby, J.,1981. Evaluation of Short-Term Tests for Carcinogens". Report of the
International Collaborative Program. Vol. 1, Progress in mutation research.
Elsevier, N.Y. p. 528-32.
Alderson M.R. (1980) Thyroid cancer epidemiology. Recent Results Cancer Res.
/ O • J. "~££ •
Allen-Rowlands, C.F.; Castracane, V.D.; Hamilton, M.G; Seifter, J. (1981)
Effect of polybrominated biphenyls (PBB) on the pituitary -thyroid axis of
the rat. Proc. Soc. Exp. Biol. Med. 160:506.
Alvares, A. P.; Bickers, D.R.; Kappas, A. (1973) Poly chlorinated biphenyls-
? "6W n^e7n!i!o?UCer °f cytoch™me P-448 in the liver. Proc. Natl. Acad.
oCl . U.SA /U: 1321.
J ^1978) Appendix II. The bacterial mutation test.
Arnold D.E.; Krewski, D.R., Junkins, D.B.; McGuire, P.P.; Moodie, C.A.;
Munro, i.e. (1983) Reversibility of ethyl enethiourea-induced thyroid
lesions,, Toxicol Appl Pharmacol . 67:264-73.
ar-Rushdi, A.; Nishikura, K.; Erikson, J.; Watt, R.; Rovera, G.; Croce, C.M.
(1983) Differential expression of the translocated and the untrans-
located c-myc oncogene in Burkitt lymphoma. Science 222:390-3.
Astwood, E.B.; Bissell, A.; Hughes, A.M. (1945) Further studies on the
chemical nature of compounds which inhibit the function of the thyroid
gland. Endocrinology 37:456-81.
Astwood, E.B.; Sullivan, B.; Bissell, A.; Tyslowitz, R. (1943) Action of
certain sulfonamides and of thiourea upon the function of the thyroid
gland of the rat. Endocrinology 32:210-25,
Autio, K.; von Wright, A.; Pyysalo, H. (1982) The effect of oxidation of the
sulfur atom on the mutagenicity of ethylene thiourea. Mut. Res. 106:27-31.
Axelrod, A.A.; Leblond, C.P. (1955) Induction of thyroid tumors in rats by
a low iodine diet. Cancer 8:339-367. '
Bachrach, L.K.; Eggo, M.C.; Mak, W.W.; Burrow, G.N. (1985) Phorbol esters
stimulate growth and inhibit differentiation in cultured thyroid cells
Endocrinology 116:1603-9.
Bahn A.K. J.L.; Mills P.J ; Snyder, Gann; P.H.; Houten, L.; Bialik, 0.; Hollmann,
in "Orkers 'exposed t0 P<>lybro.1n.ted
R-l
-------
Baker, B.L; Yu, Y-Y. (1971) HypophyseaT changes induced by thyroid deficiency
and thyroxine administration as revealed by immunQCheraical staining.
Endocrinology 89:996-1004.
Balsam, A.; Oppenheimer, J.H. (1975). Pituitary tumor with primary hypothy-
roidism. N.Y. State J. Med. 75:1737-41.
Barbacid, Mv (1986) Oncogenes and human cancer: cause or consequence:
Carcinogenesis 7:1037-42.
Bastomsky, C.H. (1973) The biliary excretion of thyroxine and its glueuronic
acid conjugate in normal and Gunn rats. Endocrinology 92:35-40.
Bastomsky, C.H.; Papapetrou, P.O. (1973) Effect of methylcholanthrene on biliary
thyroxine excretion in normal and Gunn rats. J. Endocrinol. 56:267-73,
Bastomsky, C.H. (1974) Effects of polychlorinated biphenyl mixture (Arochlor 1254)
and DDT on biliary thyroxine excretion in rats. Endocrinology 95:1150-5.
Bastomsky,, C.H.; Solymoss, B.; Zsigmond, G; Wyse, J.M. (1975) On the. mechanism
of polychlorinated biphenyl-induced hypobilirubinaemia. Clin. Chim. Acta
61:171-4.
Bastomsky, C.H.; Murthy, P.V.N. (1976) Enhanced in vitro hepatic glucuroni-
dation of thyroxine in rats following cutaneous application or ingestion
of polychlorinated biphenyls. Can. J. Physiol. Pharmacol. 54:23-6.
Bastomsky, C.H.; Murthy, P.V.N.; Banovac, K. (1976) Alterations of thyroxine
metabolism produced by cutaneous application of microscope immersion oil:
effects due to polychlorinated biphenyls. Endocrinology 98:1309-14.
Bastomsky, C.H. (1977a) Enhanced thyroxine metabolism and high uptake goiters
in rats after a single dose of 2,3,7,8-tetrachlorodibenzo-p_-dioxin.
Endocrinology 101:292-6.
Bastomsky, C.H. (1977b) Goitres in rats fed polychlorinated biphenyls.
Can. J. Physiol. Pharmacol. 55:288-92.
Becker, D.V. (1984) Choice of therapy for Graves' hyperthyroidism: N. Eng.
J. Med. 311:464-6.
Bielschowsky, F. (1953) Chronic iodine deficiency as a cause of neoplasia in
thyroid and pituitary of aged rats. Brit. J. Cancer 7:203-13.
Bielschowsky, F. (1955) Neoplasia and internal environment. Brit. J. Cancer
9:80-116.
Bielschowsky, F.; Goodall, C.M. (1963) A reassessment of the thyroid tumors
induced by goitrogens in mice. Proc. Univ. Otago Med. School (Dunedin,
New Zealand) 41:3-4.
Bignami, M.; Aulicino, F.; Velcich, A.; Carere, A.; Morpurgo, G. (1977)
Mutagenic and recombinogenie action of pesticides in Aspergillys nidulans.
Mut. Res. 46:395-402. " •'•" " " •'- '' '
R-2
-------
Bigos, S.T.; Ridgway, E.G.; Koun'des, I.A.; Maloof, F. (1978) Spectrum of
pituitary alterations with mild and severe thyroid impairment. J. Clin.
Endocrine!. Metab. 46:317-25.
Bone, E.A.; Ailing, D.W.; Grollman, E.F. (1986) Norepinephrine and thyroid-
stimulating hormone induce inositol phosphate accumulation in FRTL-5
cells. Endocrinology 119:2193-200
Boorman, G.A. (1983) Follicular cell hyperplasia, thyroid, rat. In: "Endocrine
System" (Jones, T.C., Mohr, V.; Hunt, R.D., Eds.). Springer-Verlag,
Berlin, pp. 176-84.
Braun, R.; Schoneich, J.; Ziebarth, D. (1977) In vivo formation of N-nitroso
compounds and detection of their mutagenic activity in the host-mediated
assay. Cancer Res. 37:4572-9.
Bridges, B.A.; MacGregor, D.; Zeiger, E. (1981) Summary report on the performance
of bacterial mutation assays. In: de Serres, J; Ashby.J. (1981) "Evaluation
of Short-Term Tests for Carcinogens". Report of the International Collabora-
tive Program. Vol. 1, Progress in mutation research. Elsevier, N.Y. p. 49-67.
Bull, 6.M.; R. Fraser. (1950) Myxoedema from resorcinol ointment applied to
leg ulcers. Lancet i:851-5.
Byrne, J.J.; Carbone, J.P.; Hanson, E.A. (1987) Hypothyroidison and abnormalities
in the kinetics of thyroid hormone metabolism in rats treated chronically
with polychlorinated biphenyl and polybrominated biphenyl. Endocrinology
121:520-7.
Carere, A.; Ortali, V.A.; Cardamore, G.; Torracca, A.M.; Raschetti, R. (1978)
Microbiological mutagenicity studies of pesticides in vitro. Mutat. Res.
57:277-286.
Carl ton, W.W.; Gries, C.L. (1983) Adenoma and carcinoma, pars distal is, rat.
In: "Endocrine System" (Jones, T.C.; Mohr, V.; Hunt, R.D., Eds.). Springer-
Verlag, Berlin, pp. 134-44.
Carver, J.H.; Salazar, E.P.; Knize, M.G.; Wandres, D.L. (1981) Mutation
induction and multiple gene loci in Chinese hamster ovary cells: the
genetic activity of 15 coded carcinogens and noncarcinogens. In: de
Serres; Ashby, 1981. p. 594-601.
Chesney, A.M.; Clawson, T.A.; Webster, B. (1928) Endemic goitre in rabbits,
Incidence and characteristics. Bull. Johns Hopkins Hosp. 43:261-77.
Chin, W.W.; Shupnik, M.A.; Ross, D.S.; Habener, J.F.; Ridgway, E.G. (1985)
Regulation of the c<- and thyrotropin /? -subunit messenger ribonucleic
acids by thyroid hormones. Endocrinology 116:873-8.
Christov, K. (1975) Thyroid cell proliferation in rats and induction of tumors
by X-rays. Cancer Res. 35:1256-62.
Clayson, D.B.; Garner, R.C. (1976) Carcinogenic aromatic amines and related
compounds. In: Searle, C.E., Ed. "Chemical Carcinogens" ACS Monograph 173.
American Chemical Society. Washington, DC, p. 366-461.
R-3
-------
Colletta, 6.; Cirafici, A.M.; Vecchio, 6. (1986) Induction of the c-fos oncpgene
by thyrotropic hormone in rat thyroid cells in culture. Science 223:458^60.
Collins, W.T.; Jr., Capen, C.C.; Kasza, L.; Carter, C.; Dailey, R.E. (1977)
Effect of polychlorinated biphenyl (PCB) on the thyroid gland of rats.
Am. J. Path. 89:119-36.
Conney, A.H. (1967) Pharmacological implications of microsornal enzyme induction.
Pharmacol. Rev. 19:317-366.
Conney, A.H. (1982) Induction of microsomal enzymes for chemicals and carcino-*
genesis by polyaromatic hydrocarbons: G.H.A. Clowes Memorial lecture
Cancer Res. 42:4875-917.
Cooper, D.S. (1984) Antithyroid drugs. N. Eng. J. Med. 311:1353-62,
Corcoran, J.M.; Waters, M.J.; Eastman, C.J.; Jorgensen, G. (1986) Epidermal
growth factor: effect on circulating thyroid hormone levels in sheep.
Endocrinology 119:214-7.
Croce, C.M. (1986) Chromosome trans!ocations and cancer. Cancer Res, 46:6019^23,
Cummings, S.W.; Prough, R.A. (1983) Metabolic formation of toxic metabolites.
In: "Biological Basis of Detoxication" (Caldwell, J.; Jacoby, W.B., Eds,).
Academic, London.
Daniel, M.R.; Dehnel, J.M (1981). Cell transformation with baby hamster
kidney cells. In: de Serres, J.: Ashby, J., 1981. "Evaluation of Short-Term
Tests for Carcinogens". Report of the International Collaborative Program.
Vol. 1, Progress in mutation research. Elsevier, N.Y. p. 626-37.
Davidson, B.; Soodak, M.; Neary, J.T.; Strout, H.V.; Kieffer, J.D.; Mover, H,;
Maloof, F. (1978) The irreversible inactivation of thyroid peroxidase by
methylmercaptoimidazole, thiouracil and propylthiouracil in vitro and its
relationship to in vivo findings. Endocrinology 103:871^52.
Davies, T.F. (1985) Positive regulation of the guinea pig thyrotropin receptor.
Endocrinology 117:201-7.
De Groot, L.J. (1979) Thyroid Neoplasia. In: DeGroot, L.J. et al., Eds.
"Endocrinology". Grune and Stratton, New York, 1:509-21.
De Groot, L.J.; Stanbury, J.B. (1975) "The Thyroid and its Diseases." John
Wiley and Sons, N.Y. Chapters 4, 11, and 13.
Denef, J.F.; Haumont, S.; Cornette, C.; Beckers, C. (1981) Correlated
functional and morphometric study of thyroid hyperplasia induced by iodine
deficiency. Endocrinology 108:2352-8.
Dent, J.N.; Godsden, E.L.; Furth, J. (1956) Further studies on induction and
growth of thyrotropic pituitary tumors in mice. Cancer Res. 16:171-4.
Dere, W.H.; Hirayu, H.; Rapoport, B. (1985) TSH and cAMP enhance expression of
the myc proto-oncogene in cultured thyroid cells. Endocrinology 117:2249-51.
R-4
-------
de Serres, J.; Ashby, J., Eds. (1981) "Evaluation of Short-Term Tests for
Carcinogens". Report of the International Collaborative Program. Vol. 1,
Progress in mutation research. Elsevier, N.Y.
Dobyns, B.M.; Sheline, 6.E.; Workman, J.B.; Thompkins, E.A.; McConahey, W.M.;
Becker, D.V. (1974) Malignant and benign neoplasms of the thyroid in
patients treated for hyperthyroidism. A report of the cooperative
thyrotoxicosis therapy follow-up study. J. Clin. Endocrinol. Metab.
38:976-98.
Doniach, I. (1950) The effect of radioactive iodine alone and in combination
with methylthiouracil and acetylaminofluorene upon tumour production in
the rat's thyroid gland, Brit. J. Cancer 4:223-34. i
Doniach, I. (1953) The effect of radioactive iodine alone and in combination
with methylthiouracil upon tumour production in the rat's thyroid gland.
Brit. J. Cancer 7:181-202.
Doniach, I.; Williams, E.D. (1962) The development of thyroid and pituitary
tumors in the rat two years after partial thyroidectomy. Brit. J. Cancer
16:222-31.
Doniach, I. (1970a) Aetiological considerations of thyroid carcinoma. In:
"Tumours of the Thyroid Gland." (Smithers, D., Ed.). Edinburgh, E.;
Livingstone,S. 6:55-72.
Doniach, I. (19705) Experimental thyroid tumors. In: "Tumors of the Thyroid
Gland." (Smithers, D., Ed.). Edinburgh, E.; Livingstone,S. 6:73-199.
Doniach, I. (1974) Carcinogenic effect of 100, 250, and 500 rad X-rays on
the rat thyroid gland. Br. J. Cancer 30:487-95.
Dunkel, V.C. (1979) Collaborative studies on the Salmonella/microsome muta-
genicity assay. J. Assoc. Off. Anal. Chem. 62:874-82.
Dunkel, V.C.; Pienta, R.J.; Sivak, A.; Traulik, A. (1981) Comparative neoplastic
transformation responses of BALB/3T3 cells, Syrian hamster embryo cells,
and Rauscher murine leukemia virus-infected Fischer 344 rat embryo cells
to chimney carcinogens. J. Nat!. Cancer Inst. 67:1303-15.
Dunn, T.B. (1975) "The Unseen Fight Against Cancer: Experimental Cancer Research:
Its Importance to Human Cancer". Charlotte, NC: Bates Publishing, p. 111.
Ecobichon, D.J.; Comeau, A.M. (1974) Comparative effects of commercial
Arochlors on rat liver enzyme activities. Chem. Biol. Interact. 9:341-7.
Eisen, H.G.; Hannah, R.R.; Legraverend, C.; Okey, A.B.; Nebert, D.W. (1983)
In: "Biochemical Actions of Hormones" (Litwack, G., Ed.), Vol. 10. Academic,
N.Y., pp., 227-57.
Eltom, M;; Salih, M.A.H.; Bostrom, H.; Dahlberg, P.A. (1985) Differences in
aetiology and thyroid function in endemic goiter between rural and urban
areas of the Darfur region of the Sudan. Acta Endocrinol. 108:356-60.
R-5
-------
Engler, H.; Taurog, A.; Nakashima, T. (1982) Mechanism of inactivation of
thyroid peroxidase by thioureylene drugs. Biochem. Pharmacol. 31:3801-6.
Erikson, J.; Finger, L.; Sun, L.; ar-Rushdi, A.; Nishikura, K.; Minowada, J,;
Finan, J.; Emanuel, B.S.; Nowell, P.C.; Croce, C.M, (1986) Deregulation
of c-myc by trans!ocation of the -locus of the T-cell receptor in T-cell
leukemia. Science 232:884-6.
Evans, E.L.; Mitchell, A.D. (1981) Effects of 20 coded chemicals on sister
chromatid exchange frequency in cultured Chinese hamster ovary cells. In:
de Serres; Ashby, 1981. p. 539-50.
Faber, J.; Lumholtz, I.B.; Kirkegaard, C.; Poulsen, S.; Holme Jorgensen, P.;
Siersbaek-Nielsen, K.; Frus, T. (1985) The effects of phenytoin
(diphenylhydantoin) on the extrathyroidal turnover of thyroxine, 3,5,3'-
triiodothyrom'ne, 3,3',5'-triiodothyronine and 3',5'-diodothyronine in
man. J. Clin. Endocrinol. and Metab. 61:1093-9.
Field, J.B., Dekker, A.; Titus, G.; Kerins, M.F.; Worden, W.; Frumess R. (1979)
In vitro and in vivo refractoriness to thyrotropin stimulation of iodine
organification and thyroid hormone secretion. J. Clin. Invest. 64:265-71,
Finger, L.R.; Harvey, R.C.; Moore, R.C.A.; Showe, L.C.; Croce, C.M. (1986) A
common mechanism of chromosomal translocation in T- and B-cell neoplasia.
Science 234:982-5.
Fish, L.H.; Schwartz, H.L.; Cavanaugh, J.; Steffes, M.W.; Bantle, J.P.; Oppenheimer;
J.H. (1987) Replacement dose, metabolism, and bioavailability of levq-
thyroxine in the treatment of hypothyroidism. Role of triiodothyromne
in pituitary feedback in humans. N. Engl. J. Med. 316:764-70.
Food and Drug Research (1978) A study to determine the potential of amitrole
to induce dominant lethal mutations in Ha(ICR) mice. FDRL Report No.
5502; Feb. 28, 1978.
Fregly, M.J.; Waters, I.VI.; Straw, J.A. (1968) Effect of isomers of ODD on
thyroid and adrenal function in rats. Can. J. Physiol. Pharmacol. 45:59-66.
Friedman, B.A.; Frackelton, Jr., A.R.; Ross, A.H.; Connors, J.M.; Fujiki, H.;
Sugimura, T.; Rosner, M.R. (1984) Tumor promoters block tyrosine-specific
phosphorylation of the epidermal growth factor receptor. Proc. Nat!.
Acad. Sci. USA 81:3034-8.
Frith, C.H.; Heath, J.E. (1983) Adenoma, thyroid, mouse. In: "Endocrine System"
(Jones, T.C., Mohr, V. and Hunt, R.D., Eds.). Springer-Verlag* Berlin,
p. 184-191.
Furth, J.; Moy, P.; Hershman, J.M.; Ueda, G. (1973) Thyrotropic tumor syndrone,
A multiglandular disease induced by sustained deficiency of thyroid hormones.
Arch. Pathol. 96:217-26.
Galton, V.A. (1968) The physiological role of thyroid hormone metabolism. In:
"Recent Advances in Endocrinology," 8th ed. (James, V.H.T., E
-------
Geffner, D.L.; Azukizawa, M.; Hershman, J.M. (1975) Propylthiouracil blocks
extrathyroidal conversion of thyroxine to triiodothyronine and augments
thyrotropin secretion in man. J. Clin. Invest. 55:224-9.
Gharib, H.; James, E.M.; Charbonean, J.W.; Naessens, J.M.; Offord, K.P.; Gorman,-
C.A. (1987) Suppress!ve therapy with levothyroxine for solitary thyroid
nodules: a double-blind controlled clinical study. N. Engl. 0. Med.
317:70-5.
Gilman, A.; Murad, F. (1975) Thyroid and antithyroid drugs. In: "The Pharma-
cological Basis of Therapeutics., 5th ed., (L.S. Goodman and Gilman, A.,
Eds.) MacMillan Publishing Co., Inc., New York.
Ginsberg, J.; Murray, P.G. (1986) Protein kinase C activators modulate
differentiated thyroid function in vitro. FEES Letts. 206:309-12.
Goldstein, J.A.; Taurog, A. (1968) Enhanced biliary excretion of thyroxine
glucuronide in rats pretreated with benzpyrene. Biochem. Pharmacol.
17:1049-56.
Goodman, H.M.; van Middlesworth, L. (1980) The thyroid gland. In: "Medical
Physiology" (Mountcastle, V.B., Ed.). C.V. Mosby Co., St. Louis, 2:1495-518.
Gorbman, A. (1947) Thyroidal and vascular changes in mice following chronic
treatment with goitrogens and carcinogens. Cancer Res. 7:746-58.
Goustin, A.S.; Leof, E.B.; Shipley, G.D.; Moses, H.L. (1986) Growth factors
and cancer. Cancer Res. 46:1015-29.
Graham, S.L.; Hansen, W.H.; Davis, K.J.; Perry, C.J (1973) Effects of one-year
administration of ethylene thiourea upon the thyroid of the rat. J. Agr.
Food Chem. 21:324-9.
Green, M.H.L. (1981) A differential killing test using an improved repair-
deficient strain of Escherichia coli. In: de Serres,J.; Ashby, J., 1981 .
"Evaluation of Short-Term Tests for Carcinogens". Report of the International
Collaborative Program. Vol. 1, Progress in mutation research. Elsevier,
N.Y. p. 183-94.
Green, W.L. (1978) Mechanism of action of antithyroid compounds In: "The Thyroid"
(Werner, .C. and Ingbar, S.H.,Eds.) Harper & Row, New York. p. 77-87.
Greer, M.A.; Studer, H.; Kendall, J.W. (1967) Studies on the pathogenesis of
colloid goiter. Endocrinology 81:623-32.
Griesbach, W.E. (1941) Studies on experimental goitre II. Changes in the
anterior pituitary of the rat, produced by Brassica seed diet. Brit. J.
Expt. Pathol. 22:245-9.
Griesbach, W.E.; Kennedy, T.H.; Purves, H D. (1945) Studies on experimental s
goiter. VI. Thyroid adenomata in rats on Brassica seed diets. Brit. J.
Exp. Pathol. 26:18-24.
Grote, W.; Schmoldt, A.; Dammann, H.G. (1975) The metabolism of foreign compounds
in rats after treatment with polychlorinated biphenyls (PCBs). Biochem.
Pharmacol. 24:1121.
R-7
-------
Hall, W.H. (1948) The role of initiating'and promoting factors in the patho-
genesis of tumours of the thyroid. Brit. J. Cancer 2:273-80.
Hamilton, I.E.; vanBelle, G.; LoGerfo, J.P. (1987) Thyroid neoplasia in Marshall
Islanders exposed to nuclear fallout. J. Amer. Med. Assn. 258:629-36.
Han, V.K.; D'Ercole, A.J.; Lund, P.K. (1987) Cellular localization of somatomedin
(insulin-like growth factor) messenger RNA in the human fetus. Science
236:193-7.
Haran-Guera, N.; Pullar, P.; Furth, J. (1960) Induction of thyrotropin-
dependent thyroid tumors by thyrotropes. J. Endocrinol. 66:694-701.
Haseman, J.K.; Huff, J.; Boorman, G.A. (1984) Use of historical control data
in carcinogenicity studies in rodents. Toxicol. Path. 12:126-35
Hatch, G.G.; Anderson, T.M.; Lubet, R.A.; Kouri, R.E.; Putnam, D.L,; Cameron,
J.W.; Nims, R.W.; Most, B.; Spalding, J.W.; Tennant, R.W.; Schechtman,
L.M. (1986) Chemical enhancement of SA7 virus transformation of hamster
embryo cells: Evaluation by inter!aboratory testing of diverse chemicals.
Environ. Mut. 8:515-31.
Hayden, D.W.; Wade, G.G.; Handler, A.H. (1978) Goitrogenic effect of 4,4r-
oxydianiline in rats and mice. Vet. Pathol. 15:649-62.
Haynes, R.C.; Jr.; Murad, F. (1985) Thyroid and antithyroid drugs. In: "The
Pharmacological Basis of Therapeutics" (Gilman, A.G., Goodman, L.S., Rail,
T.W. and Murad, F., Eds.), 7th edition, Macmillan, N.Y., p. 1389-1411.
Hedinger, C. (1981) Geographic pathology of thyroid diseases. Pathol. Res.
Pract. 171 285-92.
Heidelberger, C.; Freeman, A.E.; Pienta, R.J.; Sivak, A.; Bertram, J.S.;
Casto, B.C.; DunkeT, V.C.; Francis, M.W.; Kakunaga, T,; Little, J.B.;
Schechtman, L.M. (1983) Cell transformation by chemical agents - a review
and analysis of the literature. A report of the U.S. Evironmental
Protection Agency's Gene-Tox Program. Mut. Res. 114:283-385.
Henderson, B.E.; Ross, R.K.; Pike, M.C.; Casagrande, J.T. (1982) Endogenous
hormones as a major factor in human cancer. Cancer Res. 42:3232-9.
Hercus, C.E.; Purves, H.D. (1936) Studies on experimental and endemic goiter.
J. Hyg. 36:182-203.
Hiasa, Y.; Ohshima, M.; Kitahori, Y.; Yuasha, T.; Fujita, T.; Iwata» C. (1982)
Promoting effects of 3-amino-l,2,4-triazole on the development of thyroid
tumors in rats with N-bis(2-hydroxypropyl)nitrosamine. Carcinogenesis
3:381-4.
Hiasa, Y.j Kitahori, Y.; Ohshima, M.; Fujita, T.; Yuasa, T.; Konishi, N.;
Miyashiro, A. (1982) Promating effects of phenobarbital and barbital on
development of thyroid tumors in rats treated with N~bis(2-hydroxypropyl)-
nitrosamine. Carcinogenesis 3:1187-90.
R-8
-------
Hiasa, Y.; Kitahon, Y.; Konishi, N.; Enoki, N.; Fujita, T. (1983) Effect of
varying the duration of exposure to phenobarbetal on its enhancement of
N.-bis(2-hydroxypropyl)nitrosamine-induced thyroid tumori genes Is fn male
Wistar rats. Carcinogenesis 4:935-7.
Hiasa, Y.; Kitahori, Y.; Enoki, N.; Konishi, N.; Shimoyama, T. (1984)
4,4 -Diami nodiphenylmethane: promoting effect on the development of
thyroid tumors in rats treated with N-bis(2-hydroxypropyl)nitrosamine. 0.
Nat). Cancer Inst. 72:471-6. ~
Hinkle, P.M.; Goh, K.B.C. (1982) Regulation of thyrotropin-releasing hormone
receptors and responses by L-triiodothyronine in dispersed rat pituitary i
cell cultures. Endocrinology 110:1725-3.
Houk, J.C. (1980) Homeostasis and control principles. In ."Medical Physioloav"
(Mountcastle, V.B , Ed.). C.V. Mosby, Co., St. Louis, 1:246-67.
Hunton, R.B.;; Wells, M.V.; Skipper, E.W. (1965) Hypothyroidism in diabetics
treated with sulphony1 urea. Lancet p. 449-51.
Hurst,.J.6.; Newcomer, W.S.; Morrison, J.A. (1974) Some effects of DDT
toxapherie and polychlorinated biphenyl on thyroid function in bobwhite
quail. Poult. Sci. 53:125-33.
Hwang, S.W. (1973) Effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin on the biliary
excretion of indocyanine green in rat. Environ. HlthT Perspect. 5:227-31.
IARC (International Agency for Research on Cancer). (1974) IARC Monographs
on the evaluation of carcinogenic risk of chemicals to man. Some antithyroid
and related substances nitrofurans and industrial chemicals. WHO IARC
Lyon, France. Vol. 7.
Ichinotsubo, D.; Mower, H.; Mandel, M. (1981) Testing of a series of paired
compounds (carcinogenic and noncarcinogenic structural analogue) by DMA
repair-deficient E_. coli strains. In: de Serres, J.; Ashby, J., 1981.
Evaluation of Short-Term Tests for Carcinogens". Report of the International
Collaborative Program. Vol. 1, Progress in mutation research. Elsevier
N.Y. p. 195-8. '
Ingbar S.H.; Woeber K.A. (1981) The thyroid gland. In: "Textbook of Endocrin-
ology, 6th ed. (Williams, R.H., Ed.), W.B. Saunders, Philadelphia, p. 117-247.
Jacobsen, M.M.; Levin, W.; Conney, A.H. (1975) Studies on bilirubin and steroid
glucuromdation by rat liver microsomes. Biochem. Pharmacol. 24:655-65.
Jagannath, D.R.; Vultaggio, D.M.; Brusick, D.J (1981) Genetic activity of 42
coded compounds in the mitotic gene conversion assay using Saccharomyces
cerevisiae strain D4. In: de Serres, J.; Ashby, J. (1981) "Evaluation of
Short-Term Tests for Carcinogens". Report of the International Collaborative
Program. Vol. 1, Progress in mutation research. Elsevier, N.Y. p. 456-67.
Japundzic M.M. (1969) The goitrogenic effect of phenobarbital-Na on the rat
thyroid. Acta Anat. 74:88-96
R-9
-------
Jefferies, D.J.; French, M.C. (1969) Avian thyroid: effect of p.p'-DDT on
size and activity. Science 166:1278-80. ""~
Oefferies, D.J.; French, M.C. (1972) Changes induced in pigeon thyroid by
£,£'-DDE and dieldrin. J. Wildlife Mgt. 36:24-30.
Jemec, B. (1980) Studies of the goitrogenic and tumorigenie effect of two
goitrogens in combination with hypophysectomy or thyroid hormone treatment
Cancer 45:2138-48.
Jotz, M.M.; Mitchell, A.D. (1981) Effects of 20 coded chemicals on the forward
mutation frequency at the thymedine kinase locus in L5178Y mouse lymphoma
cells. In: de Serres; Ashby, 1981. p. 580-93.
Kada, T. (1981) The DMA damaging activity of 42 coded compounds in the Rec-
Assay. In: de Serres; Ashby, 1981. p. 175-82.
Kaibuchi, K.; Tsuda, T.; Kikuchi, A.; Tanimoto, T.; Yamashita, T.; Takai, Y.
(1986) Possible involvement of protein kinase C and calcium ion in growth
factor-induced expression of c-myc oncogene in Swiss 3T3 fibroblasts.
J. Biol. Chem. 261:1187-92.
Kasai, K.; Field, J.B. (1982) Properties of enzyme activities involved in
protein phosphorylation-dephosphorylation of thyroid plasma membranes.
Biochim. Biophys. Acta 718:125-34.
Kassinova, 6.V.; Kovaltsova, S.V.; Marfin, S.V.; Zakharov, I A. (1981)
Activity of 40 coded compounds in differential inhibition and mitotic
crossing-over assays in yeast. In: de Serres; Ashby, 1981. p. 434-55.
Kasza, L.; Collins, W.T.; Capen, C.C.; Garthoff, L.H.; Friedman, L. (1978)
Comparative toxicity of polychlorinated biphenyl and polybrominated biphenyl
in the rat thyroid gland: light and electron microscopic alterations
after subacute dietary exposure. J. Environ. Path. Toxicol. 1:587-99.
Katz, M.S.; Gregerman, R.I.; Horvath, E.; Kovacs, K.; Ezrin, C. (1980) Thyro-
troph cell adenoma of the human pituitary gland associated with primary hypo-
thyroidism: clinical and morphological features. Acta Endocrinol. 85:41-8.
Kennedy, T.H.; Purves, H.D. (1941) Studies on experimental goitre. I. The
effects of Brassica seed diets on rats. Brit. J. Exp, Pathol. 22:241-7.
Kimbrough, R.D. (1974) The toxicity of polychlorinated polycyclic compounds
and related chemicals. Crit. Rev. Toxicol. 2:445-98.
Kirkhart, B. (1981) Micronucleus test on 21 compounds. In: de Serres; Ashby,
1981. p. 698-704.
Kitahori, Y.; Hiasa, Y.; Konishi, N.; Enoki, N>; Shimoyama, T.; Miyashiro, A.
(1984) Effect of propylthipuracil on the thyroid tumorigenesis induced
by N-bis(2-hydroxypropyl) nitrosamine in rats. Carcinogenesis 5:657-60.
R-10
-------
Kruijer, W.; Cooper, J.A.; Hunter, T.; Verma, I.M. (1984) Platelet-derived
growth factor induces rapid but transient expression of the c-fos qene and
protein. Nature 312:711-6,
Laamanen, I.; Sorsa, M.; Bamford, D.; Gripenberg, U.; Meretoja, T. (1976)
Mutagenicity and toxicity of amitrole. I. Drosophila tests. Mut. Res.
40:185-90
Lane, R.J.M.; Clark, F.; McCollum, J.K. (1977) Oxyphenbutazone-induced goitre.
4 Postgrad. Med. J. 53:93-5.
Langdon, W.Y.; Harris, A.M.; Cory, S.; Adams, J.M. (1986) The c-myc oncogene
v perturbs B lymphocyte development in E -myc transgenic mice. Cell 47:11-8.
Larsen, P.R. (1982) Thyroid-pituitary interaction. N. Eng. J. Med. 306:23-32.
Lavenhar, S.R.; Maczka, C.A. (1985) Structure-activity consideration in risk
assessment: a simulation study. Toxicol. Indust. Health 1:249-59.
Lavoie, E.; Tulley, L.; Foro, E.; Hoffman, D. (1979) Mutagenicity of amino-
phenyl and mtrophenyl ethers, sulfides, and disulfides. Mut. Res. 67:123-31.
Leof, E.B.; Wharton, W.; van Wyk, J.J.; Pledger, W.J (1982) Epidermal growth
factor (EGF) and somatomedin C regulate 61 progression in competent
BALB/C-3T3 cells. Exp. Cell Res. 141:107-15.
Loprieno, N. (1981) Screening of coded carcinogenic/noncarcinogenic chemicals
by a forward mutation system with the yeast Schizosaccharoymyces pombe. In-
de Serres; Ashby, 1981. p. 424-33. ~~ ~
Lu, A.Y.H.; West, S.B. (1978) Reconstituted mammalian mixed-function oxidases-
requirements, specifications, and other properties. Pharmacol. Therap.
(A) 2:337-358.
Lu, A.Y.H.; West, S.B. (1980) Multiplicity of mammalian microsomal cytochromes
P-450. Pharmacol. Rev. 31:277-95.
Lucier, G.W.; McDaniel, O.S.; Hook, G.E.R, (1975) Nature of the enhancement
of hepatic uridine diphosphate glucuronyl transferase by 2,3,7,8-tetra-
chlorodibenzo-jD-dioxin in rats. Biochem. Pharmacol. 24-325-32
€ -
MacKenzie, C.G.; MacKenzie, J.B. (1943) Effect of sulfonamides and thioureas
on the thyroid gland and basal metabolism. Endocrinology 32:185-209.
Mackenzie, J.B.; Mackenzie, C.G.; McCollum, E.V. (1941) Effect of sulfanilyl-
guanidine on thyroid of the rat. Science 94:518-9.
Mamber, S.W.; Bryson, V.; Katz, S.E. (1983) The Escherichia coli WP2/WP10Q rec
assay for detection of potential chemical carcinogens. MutTRes. 119:135-44.
Mannering, 6.J. (1971) In: "Fundamentals of Drug Metabolism and Drug Disposition"
(LaDu, B.N., Mandel, H.G. and Way, E.L., Eds ). Williams and Wilkens,
Baltimore.
R-ll
-------
Martin, C.N.; McDermid, A.C. (1981) Testing of 42 coded chemicals for their
ability to induce unscheduled DMA repair synthesis in HeLa cells In:
de Serres; Ashby, 1981. p. 532-7.
Martino, E.; Safran, M.; Aghini-Lombardi, F.; et al. (1984) Environmental
iodine intake and thyroid dysfunction during chronic anriodarone therapy
Ann, Intern. Med. 101:28-34.
Matsushima, T.; Takamoto, Y.; Shirai, A.; Sawamura, M.; Sugimura, T. (1981)
Reverse mutation test on 42 coded compounds with the E. coli WP2 system.
In: de Serres; Ashby, 1981. p. 387-95. »
McCann, J.; Ames, B.A. (1976) Detection of carcinogens as mutagens in the
Salmonella/microsome test: Assay of 300 chemicals: Discussion. Proc. »
Natl. Acad. Sci. USA 72:5135-9.
McGinty, D.A.; Bywater, W.G. (1945a) Antithyroid studies I. The goitrogenic
activity of some thioureas, pyrimidines and miscellaneous compounds. J.
Pharmacol. Exp. Therap. 84:342-57.
McGinty, D.A.; Bywater, W G. (1945b) Antithyroid studies. Ill, The goitrogenic
activity of certain chemotherapeutically active sulfones and related
compounds. J. Pharmacol. Exp. Therap. 85:129-39.
McKinney, J.D.; Chae, K.; Jordan, S.; Luster, M.; Tucker, A. (1985a) 2,3,7,8-
Tetrachlorodibenzo-p_-doxin (TCDD) binds the nuclear receptor for thyroxine.
Toxicologist 5:201.
McKinney, J.D.; Fawkes, J.; Jordan, S.; Chae, K.; Oatley, S.; Coleman, R.E.;
Briner, W. (1985b) 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) as a potent
and persistent thyroxine aqonist: a mechanistic model for toxicity based on
molecular reactivity. Env. Hlth. Perspect. 61:41-53.
McTiernan, A.M.; Weiss, N.S.; Dalling, J.R. (1984) Incidence of thyroid cancer
in women in relation to previous exposure to radiation therapy and history
of thyroid disease. J. flat!. Cancer Inst. 73:575-81.
Mehta, R.D.; R.C. von Borstel (1981) Mutagenic activity of 42 encoded compounds
in the haploid yeast reversion assay, strain XV 185-14. In: de Serresj
Ashby, 1981. p. 414-23.
Meretoja, T. et al. (1976) Mutagenicity and toxicity of amitrole. II. Human ?
lymphocyte culture tests. Mut. Res. 40:191-6. "
Meyers, F.H.; Jawetz, E.; Goldfien, A. (1976) Thyroid and antithyroid drugs. *
In: "Review of Medical Pharmacology," 5th ed. Lange Medical Publishers,
Los Altos, CA.
Milmore, J.E.; Chandraskaran, V.;. Weisburger, J.H. (1982) Effects of hypothyroidistff
on development of nitrosomethylurea-induced tumors of the mammary gland,
thyroid gland and other tissues. Proc. Soc. Exp. Biol. Med. 169:487-93.
Mirsalis, J»; Tyson, K.; Beck, J.; Loh, E.; Steinmetz, K.j Contreras, C.j
Austere, L.; Martin, S.; Spalding, J. (1983) Induction of unscheduled
DMA syntheses (UDS) in hepatocytes following in vitro and in vivo treatment.
Environ. Mut. (abst.). 5:482.
R-1E
-------
Money, W.L.; Rawson, R.W. (1950) The experimental production of thyroid
tumors in the rat exposed to prolonged treatment with thiouracil. Cancer
3:321-3b.
Moriya, M.; Ohta, T.; Watana.be, K..;. Miyazawa, T.; Kato, K.; Shirasu, Y. (1983)
Further mutagenicity studies on pesticides in bacterial reversion assay
systems,. Mut. Res. 116:185-216.
Mortelmans, K.; Haworth, s.; Lawlor, T.; Speck, W.; Tainer, B.; Zeiger, E.
(1986) Salmonella mutagenicity tests: II. Results from the testing of
270 chemicals.Environ. Mut. 8, suppl. 7:1-119.
Mortensen, J.D.; Woolner, L.B.; Bennett, W.A. (1955) Gross and microscopic find-
ings in clinically normal thyroid glands. J. Clin. Endocrinol. 15:1270-80.
Moulding, T.; Fraser, R. (1970) Hypothyroidism related to ethionamide. Am.
Rev. Respir. Dis. 101:90-4.
Muller, R.; Bravo, R.; Burckhardt, J. (1984) Induction of c-fos gene and
protein by growth factors precedes activation of c-myc. Nature 312:716-20.
Murthy, A.S.; Russfield, A.B.; Snow, G.J. (1985) Effect of 4,4'-oxydianiline
on the thyroid and pituitary glands of F344 rats: a morphologic study
with the use of the intnunoperoxidase method. J. Nat!. Cancer Inst. 74:203-8.
Nadler, N.J.; Mandavia, M.; Goldberq, M. (1970) The effect of hypophysectomy
?Sno ? exPenmenta] Production of rat thyroid neoplasms. Cancer Res. 30:
1909-11,
MAS (1980) "The Effects on Populations of Exposure to Low Levels of Ionizing
Radiation. National Research Council , National Academy of Sciences
Washington, DC. '
NCRP (1985) Induction of thyroid cancer by ionizing radiation. NCRP Report No.
«0. National Council on Radiation Protection and Measurement, Bethesda, MD.
Newman, W.C.; Fernandez, R.C.; Slayden, R.M.; Moon, R.C. (1971) Accelerated
biliary thyroxine excretion in rats treated with 3-methylcholanthrene.
Proc. Soc. Exp. Biol. Med. 138:899-904.
Mon KAL*r~Rus*di' A'; Enkson, J.; Watt, R.; Rovera, G.; Croce, C.M.
(1983) Differential expression of the normal and the translocated human
c-myc oncogenes in B cells. Proc. Natl. Acad. Sci. USA 80:4822-6.
Nishizuka Y. (1986) Studies and perspectives of protein kinase C. Science
: 30b-12.
Morris , J.M.; Kociba, R.J.; Schwetz, B.A.; Rose, J.Q.; Humiston, C.6.; Jewett,
G.L.; Gehnng P.J.; Mailhes, J.B. (1975) Toxicology of octabromobiphenyl
and decabromodiphenyl oxide. Environ. Hlth. Perspect. 11:153-61.
Nastaranjan, A.T.; van Kesteren-van Leeuwen, A.C. (1981) Mutagenic activity of
20 coded compounds in chromosome aberration/sister chromatid exchange assav
using Chinese hamster ovary (CHO) cells. In: de Serres; Ashby, 1981. p. 551-9.
R-13
-------
Nowell, P.C. (1986) Mechanisms of tumor progression. Cancer Res. 46:2203-7.
NTP (1983) NTP Technical Bulletin, No. 9, p.7. National Toxicology Program,
Public Health Service, US Department of Health and Human Services, Research
Triangle Park, NC.
NTP (1984) National Toxicology Program. Fiscal year 1984 annual plan* US
Department of Health and Human Services: Public Health Service, p. 8.
NTP (1986) CHEMTRACK. [Data base]. Research Triangle Park, NC: National
Toxicology Program—National Institute of Environmental Health Sciences *
US Department of Health and Human Services. Results Report, October, 1986.
Ohnhaus, E.E.; Studer, H. (1983) A link between liver microsomal enzyme activity ^
and thyroid hormone metabolism in man. Br. J. Clin. Pharmacol. 15:71-6*
Ohshima, M.; Ward, J.M. (1984) Promotion of N-methyl-N-nitrosourea-induced
thyroid tumors by iodine deficiency in F34~4/NCr rats. J. Natl. Cancer
Inst. 73:289-96.
Ohshima, M.; Ward, J.M. (1986) Dietary iodine deficiency as a promoter and
carcinogen in male F344/NCr rats. Cancer Res. 46:877-83.
Oppenheimer, J.H.; Bernstein, 6.; Surks, M.I. (1968) Increased thyroxine
turnover and thyroidal function after stimulation of hepatocellular binding
of thyroxine by phenobarbital. J. Clin. Invest. 47:1399-406*
Oppenheimer, J.H.; Shapiro, H.C.; Schwartz, H.L.; Surks, M.I. (1971) Dissociatiott
between thyroxine metabolism and hormonal action in phenobarbital-treated
rats. Endocrinology 88:115-9.
Oppenheimer, J.H. (1979) Thyroid hormone action at the cellular level. Science
203:971-9.
Oppenheimer, J.H. (1987) Personal communication.
Osamura, R.Y.; Takayama, S. (1983) Histochemical identification of hormones
in pituitary tumors, rat. In "Endocrine System" (Jones, T.C*, Mohr, U.;
Hunt, R.D., Eds.). New York: Springer-Verlag p. 130-4.
Osorio, C.; Myant, N.N. (1963) Effect of salicylate on the biliary excretion
of thyroxine in the rat. Endocrinology 72:253-7. • 1
OSTP, Office of Science and Technology Policy (1985) Chemical carcinogens: a
review of the science and its associated principles. Fed. Reg 50:10371-442. *
Paika, I.J.; Beauchesne, M.T.; Randall, M.; Schreck, R.R.; Latt, S.A. (19011 irt.
vivo SCE analysis of 20 coded compounds. In: de Serres; Ashby, 1981. p. 673-&U-
Parodi, S.; Taningher, M.; Russo, P.; Part a, M.; Tamaro, M.-j Monti-Bragadfr, C.
(198.1) DNA-damaging activity in vivo and bacterial mutargenicity of sixteen
aromatic amines and azo-derivatives, as related quantitatively to their
carcfnogenicity. Carcfnogenesis 2:1317-26.
R-14
-------
Parodi, S.; Zunino, A.; Ottaggio, L.; DeFerrari, M.; Santi, L. (1983) Lack of
correlation between the capability of inducing sister-chromatid exchanges
in vivo and carcinogenic potency for 16 aromatic amines and azo derivatives.
Mut. Res. 108:225-38.
Parry, J.M.; Sharp. D.C. (1981) Induction of mitotic aneuploidy in the yeast
. strain 06 by 42 coded compounds. In: de Serres; Ashby, 1981. p. 468-80.
Paynter, O.E.,; Burin, G.J.; Jaeger, R.B.; Gregario, C.A. (1986) Neoplasia
induced by inhibition of thyroid gland function (guidance for analysis and
evaluation). Hazard Evaluation Division, U.S. Environmental Protection
Agency, Washington, D.C.
Pazdernik, T.L.; Rozman, K.K. (1985) Effect of thyroidectomy and thyroxine on
2,3,7,8«tetrachlorodibenzo-p-dioxin immunotoxicity. LifeSci. 36:695-703.
Pendergrast, W.J.; Milmore, B.K.; Marcus, S.C. (1961) Thyroid cancer and
thyrotoxicosis in the United States their relation to endemic goiter. J.
Chron. Dis. 13:22-38.
Perry, P.E.; Thomson, E.J. (1981) Evaluation of the sister chromatid exchange
method in mammalian cells as a screening system for carcinogens. In:
de Serres; Ashby, 1981. p. 560-9.
Philp, J.R.; Crooks, J.; MacGregor, A.G.; Mclntosh, J.A.R. (1969)! The growth
curve of the rat thyroid under a goitrogenic stimulus Brit! J. Cancer
23:515-23.
Poland, A.; Glover, E. (1974) Comparison of 2,3,7,8-tetrachlorodibenzo-£-dioxin,
a potent inducer of aryl hydrocarbon hydroxylase, with 3-methylcholanthrene.
Mol. Pharmacol. 10:349-59.
i
Poland, A.; Knutsen, J.C. (1982) 2,3,7,8-Tetrachlorodibenzo-p-dioxin and
related halogenated aromatic hydrocarbons examination of the'mechanism of
toxicity. Ann. Rev. Pharmacol. Toxicol. 22:517-54.
Polychronakos, C.; Guyda, H.J.; Patel, B.; Pooner, B'.I. (1986) Increase in the
number of type II insulin-like growth factor receptors during propylthioura-
cil-induced hyperplasia in the rat thyroid. Endocrinology 119:1204-9.
Potter, C.L.; Moore, R.W.; Inhorn, S.L.; Hagen, T.C.; Peterson, R^E. (1986)
Thyroid status and thermogenesis in rats treated with 2,3,7,8-tetrachloro-
dibenzo-p_-dioxin. Toxicol. Appl. Pharmacol. 84:45-55.
Potter, C.L.; Sipes, I.G.; Russel, D.H. (1983) Hypothyroxinemia and hypothermia
in rats in response to 2,3,7,8-tetrach1orodibenzo-£-dioxin administration.
Toxicol. Appl. Pharmacol. 69:89-95.
Preston-Martin, S.; Bernstein, L.; Pike, M.C.; Maldonado, A.A.; Henderson, B.E.
(1987) Thyroid cancer in young women related to prior thyroid disease and
pregnancy history. Br. J. Cancer. 55:191-5.
Purves, H.O. (1943) Studies on experimental goitre. IV. The effect of
diiodotyrosine and thyroxine on the goitrogenic action of brassica seeds.
Brit. J. Exptl. Pathol. 24:171-3.
R-15
-------
Riccabona, G. (1982) Thyroid cancer and endemic goiter. In: "Endemic Goiter
and Endemic Cretinism," (J.B Stanbury and .B.S. Hetzel, Eds») Wiley, N.Y.
p. 333-350.
Richter, C.P.; Clisby, K.H. (1942} Toxic effects of the bitter-tasting phenyl-
thiocarbamide. Arch. Path. 33:46-57.
Robinson, D.E.; Mitchell, A.D. (1981) Unscheduled DMA synthesis response in
human fibroblasts, WI-38 cells, to 20 coded chemicals. In: de Serres;
Ashby, 1981. p. 517-27. *
Roger, P.P.; Dumont, J.E. (1984) Factors controlling proliferation and
differentiation of canine thyroid cells cultured in reduced serum conditions. *
Effects of 1-thyrotropin, cyclic AMP, and growth factors. Mol. Cell.
Endocri no!. 36:79-93.
Roger, P.P.; Reuse, S.; Servais, P.; Van Heuversivyn, B.; Dumont, J.E. (1986)
Stimulation of cell proliferation and inhibition of differentiation expression
by tumor-promoting phorbol esters in dog thyroid cells in primary culture
Cancer Res. 46:898-906.
Rojeski, M.T.; Gharib, H. (1985) Nodular thyroid disease: evaluation and
management. N. Eng. J. Med. 313:428-36.
Ron, E.; Kleinerman, R.A.; Boice, J.D.; Livolsi, V.A.; Flannery, J.T.; Fraumeni,
J.F. (1987) A population-based case-control study of thyroid cancer.
J. Nat!. Cancer Inst. 79:1-12.
Ron, E.; Modan, B. (1982) Thyroid Cancer. In: "Cancer Epidemiology and Prevention."
(Schottenfield, D. and Fraumeni, J.F.-, Eds.). Saunders, Philadelphia.
p. 837-54.
Rootwelt, K.; Ganes, T.; Johannessen, S.I. (1978) Effect of carbamazepine,
phenytoin and phenobarbitone on serum levels of thyroid hormones and
thyrotropin in humans. Scand. J Clin. Lab. Invest. 38:731-6.
Rosenkranz, H.S.; Hyman, J.; Leifer, Z. (1981) DMA polymerase deficient assay.
In: de Serres; Ashby, 1981. p. 210-8.
Rosenkranz, H.S.; Poirfer, L.A. (1979) Evaluation of the mutagenicity and
DMA-modifying activity of carcinogenicity and noncarcinogen in microbial 1
systems. J. Nat!. Cancer Inst. 62:873-92.
Ross, D.S.; Ellis, M.F.; Ridgway, E.C. (1986) Acute thyroid hormone withdrawal *
rapidly increases the thyrotropin oc- and (3 -subunit messenger ribonucleic
acids in mouse thyrotropic tumors. Endocrinology 118:1006-10.
Rozengurt, E. (1986) Early signals in the mitogenic response. Science 234:161-6.
Ryan, D.; Lu, A.Y.H.; Levin, W. (1978) Purification of cytochrome P-450 and
P-448 from rat liver microsomes. Meth. Enzymol. 52:117-123.
Safaeri, M.; Sterling, F.H.; Utiger, R.D. ,(1975) Reduction in extrathyroidal
triiodothyronine production by propylthiouracil in man. J. Clinv Invest.
55:218-23.
R-16
-------
Saji, M.; Tsushima, T.; Isozaki, 0.; Murakami, H.; Ohba, Y.; Sato, K.; Arqi,
M.; Mariko, A.; Shizume, K (1987) Interaction of insulin-like growth
factor I with porcine thyroid cells cultured in monolayer. Endocrinology
121:749-56.
Salamone, M.F.; Heddle, J.A.; Katz, M. (1981) Mutagenic activity of 41 compounds
- in the in vivo micronucleus assay. In: de Serres; Ashby, 1981. p. 686-97.
Samaan, N.A.; Osborne, B.M.; MacKay, B.; Leavens, M.E.; Duello, T.M.; Halmi,
N.S. (1977) Endocrine and morphologic studies of pituitary adenomas
secondary to primary hypothyroidism. J. Clin. Endocrinol. Metab. 45:903-11.
Sampson, R.J,,; Woolner, L.B.; Bahn, R.C.; Kurland, L.T. (1974) Occult thyroid
carcinoma in Olmsted County, Minnesota: prevalence at autopsy compared
with that in Hiroshima and Nagasaki, Japan. Cancer 34:2072-6.
Santler, J.E. (1957) Growth in the cell populations of the thyroid gland of
rats treated with thiouracil. J. Endocrinol. 15:151-61.
Sap, J.; Munoz, A.; Damm, K.; Goldberg, Y.; Ghysdael, J.; Leutz, A.; Beug, H.;
Vennstromi B. (1986) The c-erb-A protein is a high-affinity receptor for
thyroid hormone. Nature 324:635-40.
Schaffer, R.; Muller, H.A. (1980) On the development of metastasizing tumors
of the thyroid gland after combined administration of m'trosomethyl urea
and methylthiouracil. J. Cancer Res. Clin. Oncol. 96 281-5.
Schaller, R.T.; Stevenson, J.K. (1966) Development of carcinoma of the thyroid
in iodine deficient mice. Cancer 19:1063-80.
Schottenfeld, D.; Gershman, S.T. (1978) Epidemiology of thyroid cancer. CA-A
Cancer «J. Clin. 28:66-87.
Schwartz, H.L.; Bernstein, G.; Oppenheimer, J.H. (1969) Effect of phenobar-
bital administration on the subcellular binding of 1251-thyroxine in rat
liver: importance of microsomal binding. Endocrinology 84:270-87.
Schupbach, M.; Hummler, H. (1977) A comparative study on the mutagenicity of
ethylenethiourea in bacterial and mammalian test systems. Mut. Res.
56:111-20.
Seiler, J.P. (1974) Ethylenethiourea (ETU), a carcinogenic and mutagenic
metabolite of ethylenebisdithiocarbamate. Mut. Res. 26:189-91.
Seiler, J.P. (1975) In vivo mutagenic interaction of nitrite and ethylene-
thiourea. Experientia 31:214-5.
Seiler, J.P. (1977) Nitrosation in vitro and in vivo by sodium nitrite, and
mutagenicity of nitrogenous pesticide's. Mut. Res. 48:225-36.
Seljelid, R.; Helminen, H.J.; Thies, G. (1971) Effect of long-term suppression
and stimulation of rat thyroid with special reference to lysosomes.
Exp. Cell Res. 69:249-55.
R-17
-------
31?
™ in
CD
CO
O
O
m
-o
\n
co
CO
WOE
— ?• Q.
rr o
o o
S.5 3
*
a*f s?i
0 0
r* < o
ro o
01 X
Q. m
o. o
re 7*
: m
-
-t CD
> m c
CD 3 3
<. »•
D -i CD
O O Q.
"I*
CD Q>
3 fl*-
i'3 fc
3
O)
00
CD
3D
CD
(O
CD
0)
m
30
TJ
O
cn
------- |