Werner & Ingbar's The Thyroid: A Fundamental & Clinical Text, 9th Edition

70A.Epidemiology and Pathogenesis

Arthur B. Schneider

Elaine Ron

The thyroid gland is an uncommon site of cancer, accounting for only 0.85% and 2.5% new cases of cancers among men and women, respectively, in the United States. Because of its favorable prognosis, thyroid carcinoma causes an even lower percentage of cancer deaths, 0.21% and 0.30% for men and women, respectively (1). Since 1980, the incidence of thyroid carcinoma in the United States has been increasing rapidly (2) (Fig. 70A.1). As shown in Figure 70A.1, among women the increase has been particularly steep since 1993 (4.3% per year, p < .05), whereas among men the rate has been increasing 1.9% per year (p < .05) since 1980. Thyroid carcinoma was the most rapidly increasing malignancy in women and the second most rapidly increasing in the general population (2) (Fig. 70A.2).

FIGURE 70A.1. Time trend in the United States for thyroid carcinoma incidence and mortality for men (left) and women (right) by year. The vertical scale for women is reduced by twofold compared with men. The data are fit with a program set to allow up to three connected lines. (Data from Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer Statistics Review, 1975–2000. Bethesda, MD: National Cancer Institute, 2003.)

FIGURE 70A.2. Change in the incidence of various malignancies in the United States over the time interval 1975 to 2000. (Data from Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer Statistics Review, 1975–2000. Bethesda, MD: National Cancer Institute, 2003.)

In contrast to incidence, the changes in mortality rates were much smaller. From the 1980s to 2000, the mortality rate for thyroid carcinoma increased by 1.0% per year (p < .05) among men, whereas there was no change (0.0% per year) in women. The increasing incidence is due in part to more sensitive diagnostic methods that find thyroid carcinomas with little or no propensity to progress. Improved treatment, a decline in anaplastic thyroid carcinoma (3), and the fact that the increase is primarily in early stage papillary carcinomas explain why, despite the increasing incidence, there has been little or no change in mortality.

The increasing incidence of thyroid carcinoma is not confined to the United States, as seen in Table 70A.1. To understand this apparently uniform increase in thyroid carcinoma incidence, birth cohort analyses have been performed. In Connecticut, where the incidence of papillary thyroid carcinoma in both women and men has been increasing since 1935, while the rates for other histologic types generally have been stable for the past two to three decades, the results of a birth cohort analysis fit the hypothesis that the increase was due to radiation treatment to the head and neck area of children, which is the only proven carcinogen (8). Similar results were observed in Canada and Norway (4,9). However, other factors, such as improved diagnosis, may be important as well. In some countries the incidence of thyroid carcinoma is no longer increasing (10,11).

TABLE 70A.1. TIME TRENDS FOR THYROID CARCINOMA IN VARIOUS PARTS OF THE WORLD


  

Rate of Increase (%/yr)


Geographical Area

Time Interval

Women

Men


United States (2)

1993–2000;

4.3

1.9

 1980–2000

 Canada (4)

1970–1996

2.7

2.9

Australia (5)

1982–1997

10.7

8.3

France (6)a

1978–1997

8.1

6.2

Sweden (3)

1958–1981

2.0

1.7

Vaud, Switzerland (7)a

1974–1998

2.8

2.9


aPapillary only.

Papillary thyroid carcinoma is the predominant histologic form in most parts of the world (12,13,14). The distribution has changed over time due to an increase in papillary carcinoma and a decrease in anaplastic carcinoma. From 1993 to 1997 the U.S. Surveillance, Epidemiology, and End Results (SEER) cancer registries reported 85.3% papillary (compared with 73.0% during 1973–1987), 10.9% follicular, 1.7% medullary, 0.78% anaplastic, and 1.3% other thyroid carcinomas in whites and 72.3% papillary, 20.5% follicular, 3.0% medullary, 0.5% anaplastic, and 4.2% other thyroid carcinomas in blacks (12,14). The worldwide distribution is similar, although in some areas the papillary/follicular ratio is smaller. In Sweden, from 1958 to 1987 there were 7,906 thyroid carcinomas reported to the national cancer registry. After excluding 947 (12%) that were actually follicular adenomas, 52% were classified as papillary thyroid carcinoma, 28% as follicular thyroid carcinoma, and 20% as anaplastic or medullary thyroid carcinoma (15). During this interval, the overall incidence of thyroid carcinoma increased, the incidence of papillary carcinoma increased, and the incidence of anaplastic carcinoma decreased (3). Similarly, in Canada and Switzerland, the increases over similar periods of time were confined to papillary carcinomas (Table 70A.1). A limitation in interpreting these trends is that a new histologic classification was published in 1988 (16). This caused some follicular carcinomas to be reclassified as a follicular variant of papillary carcinoma.

In the United States, thyroid carcinoma occurs about three times more frequently in women than in men, and this ratio is relatively constant over differ racial and ethnic groups. However, thyroid carcinoma incidence varies among racial and ethnic groups (2). Age-adjusted rates are about twofold higher in white men and women compared with black men and women, whereas white Hispanics have nearly the same rates as non-Hispanic whites. The highest rates among U.S. residents are found in Filipino men and women, about twice those of whites.

External radiation is the clearest pathogenetic factor associated with thyroid carcinoma. Because thyroid carcinoma is two to three times more common in women than in men, especially during the reproductive ages (17), hormonal factors probably are involved (Fig. 70A.3). However, after years of study, epidemiologic data are still inconclusive. Other environmental factors (e.g., diet) have been implicated in the etiology of thyroid carcinoma. Many of these factors are thought to operate through the action of thyrotropin (TSH). There is considerable evidence from animal experiments that prolonged TSH stimulation can cause thyroid carcinoma, but in humans the evidence is not as clear (18). Inherited genetic factors are related to thyroid carcinoma in familial adenomatous polyposis syndrome, Cowden's disease, and probably in other familial occurrences of thyroid carcinoma.

FIGURE 70A.3. Thyroid carcinoma incidence in the United States by age and sex. (Data from Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer Statistics Review, 1975–2000. Bethesda, MD: National Cancer Institute, 2003.)

RADIATION

Although radiation can damage cells in several ways, it is generally accepted that it primarily causes carcinoma by its effects on DNA. The site of this initial DNA damage and the path leading to the eventual carcinoma involve multiple, as yet incompletely understood, steps (19). While radiation can initiate and possibly promote thyroid carcinogenesis, additional factors probably are required before clinically evident carcinoma occurs. Factors that increase TSH secretion may not be sufficient to cause thyroid carcinoma, but may stimulate tumorigenic growth. Thus, even after radiation damage has occurred, giving thyroid hormone treatment may prevent the progression to clinically important thyroid tumors.

External Radiation and Thyroid Carcinoma

The relationship between radiation and thyroid carcinoma was first recognized by Duffy and Fitzgerald in 1950 (20). They found that an unusually large fraction of their childhood thyroid carcinoma patients had a history of radiation therapy for benign conditions of the head and neck. This relationship was subsequently confirmed by many epidemiologic studies (21,22).

Several difficulties arise in studying the relationship between radiation exposure and thyroid carcinoma. Because thyroid carcinoma is a rare disease, few studies have sufficient statistical power to adequately quantify risk; the very good survival rate of thyroid carcinoma patients requires that incidence rather than mortality be assessed; because radiation exposure frequently occurs at a young age, people often are unaware of, or uncertain about, their exposure; and finally, the diagnosis of thyroid tumors is highly dependent on the extent of the procedures used to look for them (diagnostic bias). In the case control studies, cases of thyroid carcinoma were identified by their entry into a tumor registry or by their admission to a hospital. The control subjects were comparable subjects without thyroid carcinoma. Information on risk factors, such as radiation exposure, was obtained retrospectively and the distribution in the two groups compared. In case control studies, cases may report exposure to risk factors more completely than controls (recall bias). In the cohort studies, exposure to radiation generally was documented, and the characteristics and amount of exposure was known. The frequency of thyroid carcinoma in the radiation-exposed group was compared with a group of similar subjects who had little or no exposure. Therefore, in cohort studies, recall bias was minimized, but diagnostic bias could have been important. The evidence in Table 70A.2 is especially strong because multiple studies conducted in various locations using different methodologies report similar findings.

TABLE 70A.2. SELECTED EPIDEMIOLOGIC STUDIES OF THE RELATIONSHIP BETWEEN EXTERNAL RADIATION AND THYROID CARCINOMA


 Study Population

  


 Location (reference)

Age at Exposure

Exposed (n)

Nonexposed (n)

Comments


Cohort studiesa

   Boston, tonsils (24)

Children

1,192

1,063

Elevated risk for nodules

   Chicago, tonsils (25)

Children

2,643

0

ERR/Gy = 2.5

   China, background radiation (26)

All ages

1,001

1,005

No effect

   China, radiology workers (27)

Adults

27,000

26,000

Doses unknown, relative risk = 2.1

   Israel, tinea capitis (28)

Children

10,834

16,226

ERR/Gy = 32.5

   Japan, atomic bomb (29)

All ages

41,234

38,738

ERR/Gy = 4.7 (children), 0.4 (adults)

   Marshall Islands, fallout (30)

All ages

250

600

ERR/Gy = 0.3 (children), 0.5 (adults)

   New York City, tinea capitis (21)

Children

2,200

1,400

ERR/Gy = 7.7 (not significant)

   Rochester NY, thymus (31)

Children

2,475

4,991

ERR/Gy = 9.1

   Utah-Nevada-Arizona, fallout (32)

Children

1,055

1,418

ERR/Gy = 7 for benign and malignant neoplasms combined, carcinoma not significant

 

Cases (n)

Controls (n)

 


Nested case control studiesa

   International, cervical cancer (33)

Adults

43

81

ERR/Gy = 34.9

   International, childhood cancer (34)

Children

22

82

ERR/Gy = 1.1


aIn the cohort studies, individual doses were estimated, and dose-response relationships were evaluated. In the nested case control studies, cases had thyroid carcinoma, and controls did not. In the two case control studies, the cases were derived from 150, 000 patients treated for carcinoma of the uterine cervix and 9, 170 children treated for carcinoma, respectively. Controls in the Japan atomic bomb study were exposed to < 0.01 Sv and in the Utah-Nevada-Arizona fallout study to < 0.05 Gy.

ERR, excess relative risk. All of the estimates of ERR are taken from reference 15, except for the Marshall Island study taken from reference 30 and the Utah-Nevada-Arizona study taken from reference 32.

Ron et al (23) conducted a comprehensive analysis of radiation exposure and thyroid carcinoma, combining the data from seven studies that had individual thyroid dose estimates. Their analysis of childhood exposure, which included nearly 500 patients with thyroid carcinomas, demonstrated a strong positive association between radiation dose and thyroid carcinoma. Based on an excess relative risk model (i.e., risk increases multiplicatively with dose), a linear dose-response relationship fit the data well. A consistent and strong relationship between radiation exposure, possibly at doses as low as 0.1 Gy (10 rad) and thyroid carcinoma was found (21,22,23). At doses below 0.1 Gy, results have been equivocal, but a linear nonthreshold dose-response relationship fits the data well.

Several additional observations, with important clinical implications, follow from the analysis of the pooled data. First, there is a strong inverse relationship between age at exposure and risk. In fact, there was little evidence for a radiation effect among persons exposed after age 15 years (Fig. 70A.4). Second, women tended to be more sensitive to the effects of radiation overall, although the difference between men and women is not statistically significant and is not consistent across studies. Third, the risk remains elevated several decades after the initial exposure. Between 5 and 30 years after exposure the risk is essentially constant. After 30 years, it appears to decline, but still remains elevated. However, depending on the model used to describe the data, the magnitude and duration of the risk can vary considerably, and more work is needed to determine which model is most appropriate (35).

FIGURE 70A.4. Relative risk for thyroid carcinoma by age at radiation exposure. The dose-response relationship for age at exposure < 15 years is compared with the dose-response for age at exposure ≥15 years. (Data from Ron E, Lubin JH, Shore RE, et al. Thyroid cancer after exposure to external radiation: a pooled analysis of seven studies. Radiat Res 1995;141:259.)

Internal Radiation and Thyroid Carcinoma

It is now clear from the Chernobyl accident experience that exposure to radioactive iodine (RAI) during childhood is associated with an increased risk for thyroid carcinoma (36,37). Nevertheless, in the past few years, additional evidence has been obtained that adds support to the safety of using iodine 131 in the clinical setting. The following two sections summarize the evidence for both of these observations and discuss the likely explanations for this apparent paradox.

Iodine 131 Releases into the Environment

Even prior to the Chernobyl accident, there was concern about the carcinogenic potential of 131I. In part, this came from experiments in animals, but it was unclear if these findings could be extrapolated to humans (38). Some people living on certain atolls of the Marshall Islands who were exposed to fallout from a nuclear test explosion in 1954 subsequently developed thyroid tumors, including carcinomas (30,39,40). However, their radiation exposure came from a combination of 131I, other more rapidly decaying isotopes of iodine, and external gamma radiation.

Shortly after the accident at the Chernobyl power plant, which released an estimated 23 to 46 MCi of 131I, reports began to appear of thyroid carcinoma occurring in exposed children. Because of the unusually short latency and the intense thyroid screening performed in the area, it was not immediately clear whether these carcinomas were attributable to exposure to 131I from the accident. The data collected during the years since the accident now indicate that the sharp increase in childhood thyroid carcinoma is associated with exposure to 131I from Chernobyl (41,42,43,44,45). This conclusion is drawn from several lines of evidence. An international panel of pathologists has reviewed the thyroid carcinoma diagnoses and has confirmed about 95% them. The incidence of childhood thyroid neoplasms is far higher than the incidence prior to the accident, and the cases have been confined almost exclusively to children who were alive, or in utero, at the time of the accident. As has been demonstrated in studies of external radiation, the risk for developing thyroid carcinoma in the Chernobyl area has been found to increase with decreasing age at exposure. Furthermore, studies have linked thyroid dose with thyroid carcinoma incidence (44,46,47). For example, a case control study conducted in Belarus has demonstrated a statistically significant relationship between childhood thyroid carcinoma and individually estimated radiation doses (46). In addition, based on detailed dose reconstruction data from Ukraine, Belarus, and Russia, Jacob et al (47) correlated mean regional thyroid doses with thyroid carcinoma incidence and found a linear dose-response relationship. The risk estimates were similar to those reported for external radiation (23,48). However, only carefully conducted long-term studies will provide an accurate and complete account of the dose-response relationships.

There are several distinctive features of the Chernobyl-related cases of childhood thyroid carcinoma. The latency was shorter than previously reported for external radiation–related cases. This may be a result of intense thyroid screening, the promotion of thyroid tumorigenesis by iodine deficiency (43,49,50,51,52,53,54,55), or the extremely large number of exposed individuals, which increases the chance to observe an uncommon disease. Furthermore, many of the cases had a unique histologic pattern of papillary carcinoma with a large solid component (56,57,58,59,60), and cases appeared to be rapidly growing and aggressive, with infiltration and lymph node involvement. Some of these findings are similar to those of thyroid carcinoma in children in general (61). Finally, the pattern of somatic mutations in the carcinomas, particularly in the ret proto-oncogene, appeared to be distinctive (see Chapter 70B).

In addition to the Marshall Islands and Chernobyl, 131I was widely dispersed into the environment in the United States and elsewhere by above-ground nuclear tests (62) and as a by-product of plants preparing isotopes for use in nuclear weapons (63). An epidemiologic study of children exposed to nuclear fallout from weapons testing at the Nevada Test Site found a significant association between all thyroid nodules and dose, but not for thyroid carcinoma separately (32). In a detailed evaluation of thyroid doses received by Americans during the testing period, the estimated average collective dose was 2.0 cGy. For people under age 20 at the time of exposure, the mean dose was about 10 cGy (62). An ecologic study of U.S. thyroid carcinoma incidence and mortality rates and 131I doses from the bomb tests was recently conducted (64). No association with cumulative dose was found, but associations were observed for children exposed under 1 year of age and those in the 1950 to 1959 birth cohort. Although it is impossible to determine the number of potential excess thyroid carcinomas caused by exposure to the tests, a National Academy of Sciences (USA) committee calculated that, if there is an excess, it is probably no more than 11,000 cases and that about 45% of them have already been diagnosed (65).

The Hanford nuclear site released 131I into the atmosphere, particularly in its early years, as part of the process of preparing fissionable materials for the U.S. atomic program. Prior to studying the potential consequences of the resulting exposure, methods were devised to estimate (“reconstruct”) individual thyroid doses. Subsequently, study individuals were selected to represent the spectrum of doses, to determine if there were any dose-response relationships for several thyroid diseases, particularly thyroid carcinoma. In the Hanford Thyroid Disease Study, the thyroid glands of 3,441 people who were born in the vicinity of the Hanford nuclear plant between 1940 and 1946 were thoroughly examined, including by ultrasound imaging. Among the 3,193 participants who lived in the Hanford site region during the years of atmospheric emissions, the mean and median thyroid doses were 18.6 and 10.0 cGy, respectively. No evidence of an increased risk for developing benign or malignant thyroid neoplasms associated with childhood exposure to the atmospheric releases of RAI from the Hanford Nuclear site was found (66).

The reasons for the apparent differences in the findings at Chernobyl and at Hanford are not clear. The roles of short-lived isotopes, iodine deficiency (43,49,50,51,52,53,54,55), and genetic predisposition need to be elucidated before the Chernobyl or the Hanford findings can confidently be extrapolated to other settings.

Medical Uses of Iodine 131

A great deal of effort has gone into assessing the relationship between 131I used for therapeutic or diagnostic medical purposes and the development of thyroid carcinoma (67,68,69,70,71,72,73,74,75,76,77,78). These studies generally have been reassuring about the use of 131I, but small increased risks for anaplastic thyroid carcinoma have been reported following 131I therapy for hyperthyroidism (75,76). The U.S. Thyrotoxicosis Therapy Follow-up Study Group reported that among 35,593 hyperthyroid patients treated during 1946 to 1964 and followed through 1990, 63% of whom were treated with 131I, total carcinoma mortality was not associated with 131I use (76). However, the standardized mortality rate for thyroid carcinoma was slightly, but significantly, elevated, resulting in a very small number of excess thyroid carcinomas (76). Most of the deaths occurred soon after 131I treatment, suggesting that the underlying thyroid disease may have been involved in the development of thyroid carcinoma or may have influenced the surveillance or reporting on death certificates (74). In a study conducted in Birmingham, England, 634 cancers were diagnosed among 7,417 hyperthyroid patients treated with 131I (75,79). Although the overall cancer incidence and mortality rates were lower than expected compared with age, sex, and period-specific national rates, both the thyroid cancer incidence and mortality rate were significantly higher than expected.

In a study of 10,552 hyperthyroid patients treated with 131I in Sweden, there was a nonsignificant 30% increased risk for thyroid carcinoma that was more pronounced among the toxic nodular goiter patients (69). No dose-response relationship was demonstrated. Although the number of excess cases of thyroid cancer was small in these investigations, the increased relative risk suggests that these patients need continued follow-up.

The health effects following diagnostic 131I have been studied most thoroughly in Sweden (72,73,78). Among 36,792 predominantly adult patients administered diagnostic 131I and followed through 1998, 24,010 had scans for reasons other than suspicion of a thyroid tumor and did not have a history of exposure to external radiation. The average absorbed thyroid dose was estimated to be 0.94 Gy. The number of observed thyroid carcinomas was very close to the number expected [respectively, 36 vs. 39.5 SIR (standardized incidence ratio) = 0.91; confidence interval (CI) 0.64–1.26]. In contrast, there was an increased risk among patients whose scan was performed due to suspicion of a tumor. Among patients with a history of external radiation exposure, the risk was about 10-fold. However, there was no evidence that the 131I exposure contributed to the risk, because there was no effect of age or dose of 131I exposure (78). When a sample of approximately 1,005 female patients examined with 131I and 248 nonexposed female patient controls were examined for palpable thyroid nodularity, no excess was found among exposed women compared with nonexposed women. However, among the exposed women there was a significant relationship between 131I dose and the prevalence of nodules (72).

It is of special interest to know if diagnostic 131I exposure in children is associated with thyroid carcinoma; however, the available data are too sparse for any conclusions. After a mean of 20 years, examinations were performed on 789 people who had a history of diagnostic 131I procedures (mean thyroid dose of 100 cGy) before age 18 and 1,118 controls who had other thyroid-related tests. Although no radiation effect was observed based on two exposed and three nonexposed cases (relative risk 0.86; CI 0.14–5.13), this study provides little relevant information (77). In this study, only 147 people were no more than 10 years of age at the time of 131I administration, and the study was severely limited by the low and differential participation rates for exposed and nonexposed subjects (35% and 41%, respectively), which could result in substantial selection bias.

In general, studies of adult 131I exposure for therapeutic and diagnostic purposes continue to be reassuring, but some aspects of the studies suggest a small effect, apparently related to 131I exposure, on thyroid nodularity, thyroid carcinoma incidence, and thyroid carcinoma mortality. Although a causal relationship is possible, it may be that the observations represent the nature of the underlying thyroid condition or an increase in surveillance and diagnostic misclassification.

Comparison of External and Internal Radiation

The reasons for the apparent difference between external and internal radiation are not known with certainty. One factor is undoubtedly age at exposure, which has a very strong modifying effect for external radiation (Fig. 70A.4). Because most of the patients included in studies of the medical uses of 131I have been adults, the results cannot be extrapolated to children. Another factor is that compared with the instantaneous dose received from external radiation, the lower dose rate of 131I may allow repair of radiation damage. Finally, when 131I is administered, it results in a wide range of doses to different areas of the thyroid, whereas external radiation exposes the entire thyroid to the same dose, making it difficult to compare the two types of radiation. Exposures to 131I, as occurs in the medical setting, may not be applicable to the more prolonged exposures people generally receive from fallout or living near nuclear production facilities.

Preventing Radiation-Related Thyroid Carcinoma

Functioning nuclear plants continue to generate iodine isotopes in large quantities that could, in the case of an accident or terrorist attack, pose a threat to the surrounding population. Potassium iodide (KI), if taken promptly, effectively reduces the dose to the thyroid (80). Following the Chernobyl accident, KI was widely administered to residents of Poland (81). The success of the program and the rarity of observed side effects support the distribution of KI as an effective means of reducing 131I exposure in the event of a nuclear accident (82). The World Health Organization and the U.S. Food and Drug Administration have both adopted age-based guidelines for the use of KI in case of an emergency, taking into account the varying sensitivity of the thyroid shown in Figure 70A.4 (83). In some countries the distribution of KI has been achieved, whereas in the United States evaluation is continuing, and implementation varies from state to state (84).

Evaluation of Irradiated Patients

An essential part of evaluating a person with a history of irradiation is determining the type of radiation, the site or sites treated, the age at treatment, and the dose (21,23, 85,86). Among these factors, the dose received by the thyroid is the most important, because risk increases linearly with increasing dose. The dose timing may modify risk, with fractionation possibly reducing it. Age at the time of radiation exposure is an independent risk factor, with younger age at exposure associated with greater risk. There is a greater spontaneous thyroid carcinoma risk for women than men, but it is unclear whether radiation increases the risk for thyroid carcinoma equally for women and men.

External irradiation formerly was used to treat a wide range of benign conditions during childhood. These included enlargement of the thymus, tonsils, adenoids, and cervical lymph nodes, as well as pertussis, asthma, bronchitis, tinea capitis, and acne. In most cases the radiation treatment resulted in an increased risk for thyroid carcinoma. Other commonly used forms of radiotherapy involved local application of radioactive plaques to treat hemangiomas, other localized lesions, and enlarged tonsils, as well as radium rods inserted through the nose into the nasopharynx as treatment for impaired eustachian tube function and hearing loss. Following treatment for hemangioma, the mean dose to the thyroid was about 0.2 Gy, and the excess relative risks were significantly increased (87,88) and consistent with those from the pooled analysis of external radiation (23). In contrast, treatment with nasopharyngeal radium rods resulted in very low doses to the thyroid (mean dose < 0.02 Gy) and no excess of thyroid carcinomas (89,90,91,92).

Currently, external radiation of the neck for malignant conditions such as childhood cancer, Hodgkin's disease, and carcinoma of the larynx is used widely. Such treatment often results in subclinical or overt hypothyroidism (93,94,95), and sometimes in nodular thyroid disease and thyroid carcinoma (96,97). Although the dose to the thyroid following radiation therapy for breast carcinoma frequently is high, no dose-related excess of thyroid carcinoma was found in two recent studies (98,99). Although diagnostic radiologic examinations may confer some risk for thyroid carcinoma, a large case control study in Sweden (484 cases of thyroid carcinoma and 484 controls) found no association with prior diagnostic radiography (100). A smaller case control study from Sweden, using medical records and dentist cards to estimate radiation exposure, reached the same conclusion (101). Thus, it is important to obtain the best radiation history possible, because it is needed to decide whether further evaluation, particularly thyroid imaging, should be performed.

How to examine a patient with a history of radiation exposure is a deceptively simple question (102). Thyroid palpation has limited sensitivity, but whether imaging should be used remains controversial. The limitations of palpation are shown by the following illustrative studies involving radiation-exposed individuals. In one small series, about half of the nodules found by ultrasonography that were larger than 1.5 cm were not palpable (103). In a larger series, about two thirds of the palpable nodules were not confirmed by ultrasonography (104). The use of imaging is supported by the observation that many people with nodular thyroid disease, including carcinoma, were discovered solely by imaging (103,105). On the other hand, ultrasonography is associated with considerable observer variation and may be too sensitive because approximately one third of adult women have ultrasound-detectable thyroid nodules (106,107). Some clinicians believe that even if imaging identifies otherwise undetectable nodules, they are too small to be of clinical consequence and can be safely ignored until they become palpable.

To determine the clinical importance of a nonpalpable nodule larger than 10 mm in diameter, ultrasound-guided fine-needle aspiration (FNA) is the method of choice (108). However, it is important to know if FNA is as reliable in radiation-exposed patients. Two reports provide some reassurance, although they also point out the limitations of FNA (109,110). The sensitivity and specificity rates of FNA were similar for a given size of nodule to those reported for nodules not associated with radiation, but because radiation-exposed thyroids frequently have multiple nodules, small carcinomas are more likely to be missed.

However, these small carcinomas may not be clinically important. A reasonable approach is to reserve ultrasound imaging for irradiated patients who are at especially high risk (102,111).

There is no evidence that subclinical or overt hypothyroidism or thyrotoxicosis result from the level of radiation dose used to treat benign childhood conditions (112). However, there is some, as yet unconfirmed, data suggesting a relationship between autoimmune thyroid disease and radiation (113,114). The evaluation and treatment of patients with nodular disease is discussed further in Chapters 68 and 69 and in the next section of this chapter.

Patients with a currently normal thyroid and a history of irradiation should be examined periodically. Radiation-related nodular disease continues to occur for as long as it has been possible to study patients irradiated between 1939 and 1962 (25,115,116). Patients at especially high risk should have thyroid ultrasound imaging as part of their follow-up examination. Ultrasound results should be interpreted with caution, given the high prevalence of abnormalities found in the general adult population. Examples of high-risk patients are those with one or more of the factors listed in TABLE 70A.3. An examination interval of 1 to 2 years and an imaging interval of 3 to 5 years, continued indefinitely, seem prudent.

TABLE 70A.3. RISK FACTORS ASSOCIATED WITH RADIATION-INDUCED THYROID TUMORS


Amount of radiation exposure

Young age at exposure to radiation

High serum thyroglobulin concentration

Other radiation-related tumor (s)


Prophylactic therapy to prevent the occurrence of nodular thyroid disease should be considered in patients who received radiation treatment (117,118). One report demonstrated the effectiveness of thyroid hormone therapy in preventing the appearance of nodules in irradiated patients, but this was not confirmed by another (119,120). In patients at high risk, thyroxine (T4) therapy has potential benefits that probably outweigh its risks. In this instance a reasonable definition of high risk is more than one of the factors listed in TABLE 70A.3. Also, an abnormal or equivocal thyroid imaging finding, such as a nodule seen by ultrasonography that is too small to aspirate, would contribute to a classification of high risk. Even if only benign nodules occur less frequently in T4-treated patients, the reduction of both anxiety and the likelihood of surgery is important.

Other head and neck tumors may arise in patients with a history of childhood radiation exposure, some of which can have equal or greater clinical consequence for the patient than thyroid tumors. Parathyroid adenomas and hyperparathyroidism have been reported in people who received radiation therapy (121,122,123,124). Whether the parathyroid glands show the marked age-related sensitivity charac-teristic of the thyroid gland is not clear. After radiation exposure, salivary gland tumors most commonly occur in the parotid glands and are usually readily evident to the patient. About two thirds are benign, and most of these are of the mixed cell variety. Mucoepidermoid carcinomas were the most common malignant form. A dose-response relationship was observed among atomic bomb survivors and the Michael Reese patients (125,126). Salivary gland neoplasms can occur many years after radiation exposure (125,126,127,128). Benign and malignant neural tumors also can be radiation related (129,130,131). Benign neural tumors, particularly neurilemomas and acoustic neuromas, occasionally occur as multiple tumors in an individual following childhood radiation therapy (130,131). Finally, an elevated risk of nonmelanoma skin carcinoma has been observed on the face and scalp among several of the irradiated groups studied (29,132,134).

Clinical Features of Radiation-Related Thyroid Neoplasms

A history of radiation exposure has two major clinical implications: the increased risk for developing thyroid nodules and the increased risk for a thyroid nodule being malignant. In the Michael Reese tonsil group (mean dose to the thyroid ~60 cGy), more than one third of radiation-exposed patients who had thyroidectomies had a thyroid carcinoma (116).

Follow-up studies so far indicate that external radiation-related thyroid carcinomas behave the same as other thyroid carcinomas in both children and adults (135,136,137,138). A nested case control study of 364 thyroid carcinomas (91 in radiation-exposed patients vs. 273 matched unexposed controls) was performed in France. Cases had a higher frequency of multifocal carcinomas and more frequent local residual carcinoma. However, the rates of recurrence and thyroid carcinoma-related mortality did not differ. Therefore, therapy and follow-up should be similar to that provided to other patients with thyroid carcinoma. As mentioned above, the behavior of the Chernobyl-related carcinomas may be especially aggressive. Whether this behavior is due to the young age at diagnosis and iodine deficiency at the time of exposure remains to be determined.

The thyroid carcinomas that arise in relation to radiation treatment are almost all well-differentiated papillary or papillary-follicular carcinomas. Case reports of anaplastic thyroid carcinoma occurring after radiation treatment are rare (139). There is no evidence that the well-differentiated radiation-related thyroid carcinomas are more likely to undergo transition to more aggressive or less differentiated malignancies. It is possible, however, that as the population ages, more aggressive carcinomas will be seen, as occurs in the general population (140).

All irradiated patients who have had nodular thyroid disease treated by thyroidectomy should receive thyroid hormone treatment, even if enough of the gland remains to maintain normal thyroid hormone secretion. This recommendation is based on the observation that nodules continue to occur in these patients with equal or greater frequency compared with patients who did not have surgery. When thyroid hormone therapy is given after thyroidectomy, the frequency of recurrence is reduced (141).

The carcinomas in the children from the Chernobyl region have presented with a high frequency of advanced features. Of the approximately 1,500 cases from Belarus, Ukraine, and Russia, 45% were classified as pT4 (142). Treatment of more advanced cases has included RAI. In 249 patients treated with one or more courses of RAI, only 129 (51.8%) were completely ablated, suggesting increased resistance to RAI therapy (143). In another group of 209 Belarus children treated in Germany with 755 RAI treatment courses, 84% of the carcinomas were ablated and none progressed following the treatment (142).

FAMILIAL THYROID CARCINOMA

The existence of two familial syndromes that include carcinoma of the thyroid of follicular cell origin supports the existence of genetic factors in the pathogenesis of thyroid carcinoma. Thyroid carcinomas occur in association with familial adenomatous polyposis (FAP) and its subtype, Gardner's syndrome, which is FAP associated with osteomas, epidermoid cysts, and desmoid tumors. Both are dominantly inherited conditions with mutations in the APC gene. In fact, thyroid carcinoma is considered to be the most common noncolonic malignancy in these syndromes. As reviewed by Hizawa et al (144), the risk for thyroid carcinoma is increased by more than 100-fold over the general population. In the Leeds Castle Polyposis database, 45 patients (1.2%) had thyroid carcinoma (145). Because there was only one death from thyroid carcinoma and the frequency was relatively low, Bulow et al concluded that screening for thyroid carcinoma is not indicated. However, it seems prudent to perform careful palpation of the thyroid in any patient with these syndromes. The thyroid carcinomas in these syndromes have two distinguishing features. They tend to occur at an early age, often before 35 years (146), and they have a distinct histologic pattern (147,148). The carcinomas are papillary carcinomas with a cribriform pattern and solid, spindle cell–containing areas. One clue to the pathogenesis of these carcinomas comes from the observation that the germline mutations in APC of cases with thyroid carcinoma tend to cluster in a region of the gene associated with extracolonic features of the disease (149). Another clue is that there is no loss of heterozygosity or biallelic mutations in the APC gene (in six cases) and that ret/ptc1 activation is often (in four of the six cases) found (150). Whether reduced APC activity facilitates ret translocaction remains to be determined.

The second familial syndrome associated with thyroid carcinoma is Cowden's disease, a rare autosomal-dominant disorder involving the development of multiple hamartomatous polyps, mucocutaneous pigmentation, and extraintestinal manifestations. Thyroid abnormalities, including thyroid carcinomas, are common in these cases (151,152). Mutations in the PTEN gene, a tumor suppressor gene that functions as a protein phosphatase, cause the syndrome.

Studies of family patterns of nonmedullary thyroid carcinoma have provided increasing evidence supporting the existence of additional hereditary factors. In a systematic study of 576 people with thyroid carcinoma in Utah, using extensive family history records and the state tumor registry, 28 first-degree relatives were found to have thyroid carcinoma (153). The expected number of cases was only 3.3, and compared with 27 other sites of malignancy, this was the largest increase found. In Norway, the number of thyroid carcinomas in 5,457 first-degree relatives of patients with thyroid carcinoma was about fivefold greater than expected (154). Using the Swedish Family Cancer Registry covering 9.6 million people, Hemminki and colleagues have examined the familial risk for thyroid carcinoma (155). When a parent had papillary or follicular thyroid carcinoma, the risk for an offspring developing thyroid carcinoma was 7.8 (95% CI 3.9–13.2) for a son and 2.8 (95% CI 1.4–4.5) for a daughter. These familial risks were higher than for any other cancer site.

There have been multiple reports of family aggregates of papillary thyroid carcinoma. For example, in a clinical study, 6% of 226 patients with papillary carcinoma reported having at least one relative who also had a papillary thyroid carcinoma (156). In two case control studies, relative risks of 5 and 10 were found in close relatives of patients with nonmedullary thyroid carcinoma (157,158).

There is some debate about the behavior of the thyroid carcinomas found in family clusters. In one series, the high frequency of lymph node and local invasion was cited as reason for aggressive treatment (159). In a review of 15 case series, containing 87 families with 178 cases, evidence for aggressive presentation was present in only 6 (160). Although the investigator did not advocate therapy more aggressive than usual, it should be noted that the overall frequency, where it was reported, of distant metastases (10%) and recurrences (29%) was high. Two other studies showed high rates of recurrence (16% and 44%) and advocated aggressive therapy (161,162). So far, other than APC and PTEN, mentioned above, no gene, including the two potential candidate genes, FAP and ret, has been identified as a cause of familial thyroid carcinoma (163,164,165).

It has been hypothesized that there are genetic factors related to the probability of developing radiation-induced thyroid tumors. If so, then patients who have one radiation-associated tumor (thyroid, salivary, or benign neural) might be more likely to develop a second tumor than other patients without a tumor but with comparable radiation exposure and other risk factors. Although earlier observations supported this possibility (130,166), more recent data do not (167). Another way to evaluate a genetic sensitivity to radiation tumorigenesis is to see whether siblings who were both irradiated had a higher rate of concordance than expected. Earlier analysis of thyroid nodules, but not thyroid carcinoma, showed that concordance occurred more often than would be expected by chance (168). However, when all radiation-related neoplasms are considered together and other known risk factors are taken into account, more recent analyses show that there is no excess concordance.

PREEXISTING THYROID DISEASE

Thyroid carcinoma is often preceded by other thyroid abnormalities, including endemic and sporadic goiter, benign thyroid nodules, lymphocytic thyroiditis, and Graves' disease, all of which are common. Whether patients with these abnormalities should be considered at increased risk for developing thyroid carcinoma is uncertain. Despite considerable efforts to resolve this question, results remain inconclusive. Many case control studies of thyroid carcinoma have revealed more preexisting benign thyroid nodules and goiter in the carcinoma patients than in the control subjects (157,169,170,171,172,173,174,175,176). The risks have been high, especially for nodules. A pooled analysis of 14 case control studies (177,178,179,180) with 2,725 cases and 4,776 controls was conducted to allow a systematic approach to analyzing and interpreting major hypotheses for thyroid carcinoma etiology. Large risks were associated with a self-reported history of goiter (odds ratio 5.9) or benign nodules (odds ratio 30) among women (177). Case control studies, however, are especially subject to recall bias, in which cases may remember all previous thyroid diseases while controls may forget them. This is less of a problem in prospective studies. In one prospective study of women with benign thyroid conditions in Boston, there was a significant excess of thyroid carcinoma mortality among patients with thyroid adenomas (181). In a cohort study of 204,964 members of a large health plan, 196 thyroid carcinomas occurred, and a history of goiter (as well as Asian race, educational attainment, family history, and radiation exposure) was a risk factor (relative risk 3.36; CI 1.82–6.20) (182). In a multiethnic case control study designed to explain the higher incidence of thyroid carcinoma among Asian women living in the San Francisco area, prior goiter and nodules were significant explanatory factors (183). Perhaps the most convincing study is one from Denmark (184,185). In the period 1977 to 1991 there were 57,326 patients discharged from Danish hospitals with a diagnosis of myxedema, thyrotoxicosis, or goiter. The subsequent incidence of thyroid carcinoma was ascertained through the Danish Cancer Registry and compared with thyroid carcinoma incidence in the general population. The incidence of thyroid carcinoma was increased among all three patient groups but was substantially higher among goiter patients. However, prospective studies remain subject to potential ascertainment bias (i.e., one thyroid disorder could draw attention to another). In the Danish study, the thyroid carcinoma risk decreased with time following hospital discharge, but the risk remained elevated even 10 years later. The cumulated data now suggest that a history of goiter or benign thyroid nodules is a strong thyroid carcinoma risk factor.

Recent genetic data suggest that thyroid neoplasms may progress from benign tumors to well-differentiated carcinomas to anaplastic carcinomas, as somatic mutations accumulate (see next section). The epidemiologic data relating previous nodular thyroid disease to thyroid carcinoma are consistent with this model.

HORMONAL AND REPRODUCTIVE FACTORS

Thyroid carcinoma, like most other thyroid diseases, occurs more frequently in women than men, suggesting that hormonal factors are involved in its pathogenesis. In England and Wales, the female to male ratio was highest at the time of puberty. From puberty to menopause, the difference between females and males declined consistently (17). In Switzerland, the female to male ratio at puberty for papillary carcinoma was high, but for follicular carcinoma the ratio was highest between the ages of 25 and 44 years (186). These findings suggest that hormonal events occurring at puberty or during the early reproductive years might be important in influencing the development of papillary thyroid carcinoma, but few significant relationships between hormonal and reproductive factors and thyroid carcinoma have been found, and often the findings have not been consistent across studies (157,170,171,173,174,187,188,190,190). In a pooled analysis of thyroid carcinoma etiology, there was a small but significantly positive trend for risk to increase with increasing age at menarche (179).

Results from some, but not all, epidemiologic studies indicate that parity may increase the risk for thyroid carcinoma (157,170,171,173,191,192,193). The pooled analysis found a slightly elevated risk associated with having had at least one child but no relationship with number of children (179). The most convincing data come from a prospective study of 1.1 million Norwegian women of reproductive age. Based on almost 1,000 thyroid carcinomas, a significant trend for increasing risk with increasing parity was demonstrated (191). In an attempt to distinguish the role of the biologic effects of pregnancy from lifestyle or environmental exposures, the number of children of male thyroid carcinoma patients was studied (194). No trend was found, which led the researchers to conclude that the parity effect was due to biologic changes during pregnancy.

Results from the pooled analysis also suggest that thyroid carcinoma occurs more often among women who are older when they first give birth (179). Data from some studies suggest that women with a history of spontaneous or induced abortion, particularly during the first pregnancy, have an enhanced risk for thyroid carcinoma (31,157,170,189,190), that the risk of thyroid carcinoma may be elevated among women seeking medical care for fertility problems (195,196), and that women undergoing hysterectomy may be at increased risk for developing thyroid carcinoma (192,197), but these findings were not confirmed in the pooled analysis (179).

Three studies, from Washington State, the San Francisco area, and Kuwait, each suggest that recent pregnancy, within about 5 years, is a thyroid carcinoma risk factor and that the risk is even higher with more than one recent pregnancy (198,199,200). Whether these observations are confirmed and, if so, whether they are related to hormonal effects or are incidental to increased pregnancy-related medical attention, remains to be seen.

Other suggested risk factors for thyroid carcinoma in women are exogenous estrogens, including oral contraceptives (170,173,201), lactation suppressant drugs (201), postmenopausal estrogen therapy (201), and fertility drugs (189). These associations, however, were usually weak and not dose dependent. In the recent pooled analysis, current oral contraceptive users had a moderately increased risk, which disappeared 10 or more years after discontinuing use (178). No significant risks were reported for use of hormone replacement therapy or fertility drugs, but the odds ratio for drugs used to suppress lactation was elevated. Although positive associations between hormonal and reproductive factors and the incidence of thyroid carcinoma have been found in some studies, they are generally weak and not always consistent across studies.

Two studies, both conducted among irradiated individuals, have looked at reproductive factors and benign thyroid tumors (202,203). One study showed a protective effect of pregnancy, whereas, in apparent contradiction, both showed a nonsignificant increase in risk with number of pregnancies.

DIETARY FACTORS

Iodine

A relationship between iodine-deficient endemic goiter and thyroid carcinoma has been suspected since Wegelin (204) reported more thyroid carcinoma at autopsy in Bern, an endemic goiter area, compared with Berlin, a nonendemic goiter area. Since then, several studies of the effects of iodine supplementation on the incidence of thyroid carcinoma have, with at least one exception, failed to show a decline. In Switzerland, thyroid carcinoma mortality decreased after the introduction of iodized salt (205), but it did not decrease in Italy or the United States (206,207). In Austria, the incidence continued to increase after the introduction of iodine prophylaxis (208). In Sweden, the incidence of thyroid carcinoma has continued to increase in both iodine-deficient and iodine-sufficient areas (3).

The effects of iodine intake on specific histologic types of thyroid carcinomas are clearer. In endemic goiter areas, follicular and, perhaps, anaplastic thyroid carcinoma predominate. When iodine supplementation is introduced, the proportion of papillary carcinomas increases and that of follicular carcinomas decreases (18,209,210). In Hawaii, where iodine intake is extremely high, the proportion of papillary carcinoma is also high (189). In Sweden, the risk for follicular carcinoma was higher in iodine-deficient areas than in iodine-sufficient areas, whereas the pattern was opposite for papillary carcinoma (211). Similar findings were reported in Sicily (212). These data suggest that follicular carcinomas are related to iodine deficiency and prolonged TSH stimulation.

There are several studies in which iodine intake was evaluated, and a pooled analysis of case control studies of fish and shellfish intake has recently appeared (213). In a case control study in Hawaii in which iodine intake from food sources and supplements was quantitated, iodine intake was higher in the patients with thyroid carcinoma than in the control subjects (189). There have been seven case control studies in which fish and shellfish consumption was examined as a surrogate measure of iodine intake. In all but the Hawaiian study the dietary data were extremely limited and, therefore, the findings should be interpreted with caution. Because shellfish generally have a higher iodine content than fish, they are a better indication of high iodine intake. Although the results were not always statistically significant, fish or shellfish consumption was reported more frequently among cases than controls in four studies (157,173,189,214). In a Norwegian population-based study a similar conclusion was reached based on the observation that among women, being married to a fisherman was a risk factor for thyroid carcinoma (215). In contrast, results from northern Italy and Vaud, Switzerland, indicated a protective effect of fish (216). In a study conducted in northern Sweden, the results were equivocal: adult consumption of fish appeared to be associated with a slightly decreased risk for thyroid carcinoma, and shellfish intake with a higher risk (188). The pooled analysis of these studies suggests that high fish consumption is not associated with thyroid carcinoma, except in areas of iodine deficiency, where it may have a beneficial effect (213).

Other Dietary Factors

Some dietary factors have been related to thyroid carcinoma, but many findings have been reported in single studies or have been inconsistent. In most studies, vegetables, particularly cruciferous ones, appear to be associated with a reduced risk for thyroid carcinoma (157,189,216). This would be a somewhat surprising finding, since these vegetables contain natural goitrogens. However, in a pooled analysis of 10 case control studies, the risk reduction associated with cruciferous vegetables is of borderline significance for moderate consumption (odds ratio 0.87; CI 0.75–1.01) and not significant for high consumption (odds ratio 0.94; CI 0.80–1.10) (217). In another pooled analysis, of Italian and Swiss studies, pasta, bread, pastry, and potatoes were associated with an increased risk (216), and in studies from Norway and Sweden, consumption of butter and cheese was associated with an increased risk (218). In a study from Hawaii, the cases reported eating more fat, protein, and carbohydrates than the control subjects. The higher total calorie intake among cases is consistent with an association of thyroid carcinoma with obesity seen in the Hawaiian study (189), as well as two other epidemiologic studies (157,201). Finally, in studies conducted in Greece and Japan, coffee consumption appeared to protect against thyroid carcinoma, but not in Los Angeles County (219,220,221).

SMOKING, PHARMACEUTICAL AGENTS, AND TOXINS

Cigarette smoking has been associated with a decreased risk for thyroid carcinoma in three recently reported case control studies (192,222,223). In Washington state 410 women with papillary carcinoma were compared with 574 controls. Ever having smoked 100 cigarettes was associated with a decreased odds ratio of 0.7 (CI 0.5–0.9) (223). In Northern Norway and Central Sweden 191 papillary and follicular cases were compared with 341 controls (192). Smoking was associated with an odds ratio of 0.69 (CI 0.47–1.01), and in premenopausal women it was 0.60 (CI 0.38–0.96). In Canada, 1,224 cases were compared with 2,659 controls (222). The risk ratios were reduced in male (0.77; CI 0.58–1.02) and female (0.71; CI 0.60–0.84) smokers. Notably, there were dose-response relationships for the duration and amount smoked. The relative uniformity of the estimates of risk reduction and the dose-response relationships support the physiologic relevance of the observation. However, two other recent reports, from the San Francisco area and Los Angeles County, did not find an effect of smoking (182,221). It is not clear why cigarette smoking should reduce the risk for thyroid carcinoma, but current smoking has been shown to reduce TSH levels in some studies (224).

Although no pharmaceutical agent or toxin has been implicated in the etiology of thyroid carcinoma in humans, there is reason to remain open to this possibility. Drugs such as lithium and phenobarbital are known to cause goiter and increased serum TSH concentrations, making it reasonable to suspect that certain drugs could cause or promote the growth of thyroid carcinoma. In rodents, a variety of drugs and other chemicals cause thyroid growth and thyroid neoplasms (225). None of them have been shown to be thyroid mutagens. Rather, they appear to act by interfering with the synthesis of thyroid hormones or by altering the peripheral metabolism of thyroid hormones (226). In either case, the increased TSH secretion, possibly combined with subsequent random events or events promoted by the TSH stimulation, may lead to thyroid neoplasms (227,228). Patients with congenital goiter have thyroid glands that are subjected to intense TSH stimulation until appropriate treatment is given. Rarely, such patients develop thyroid carcinoma, supporting the possibility that intense TSH stimulation contributes to thyroid carcinogenesis in humans (229,230).

SUMMARY

Radiation is the clearest risk factor for thyroid carcinoma in humans, but a history of goiter or benign nodules also appears to significantly increase the risk for developing thyroid carcinoma. One of the major problems in understanding the pathogenesis of thyroid carcinoma is the need to take histology into account. The four major histologic types of thyroid carcinoma (papillary, follicular, anaplastic, and medullary) may have different risk factors, and the rarity of the disease makes it extremely difficult to study the histologic types separately. Although pooling data from several studies of radiation-induced thyroid carcinoma has helped resolve questions regarding the shape of the dose-response curve and effect modification, similar pooling of data from etiologic studies of thyroid carcinoma so far has been less informative. The roles of iodine, diet, reproductive and hormonal factors, and genetics still remain unclear.

REFERENCES

1. Jemal A, Murray T, Samuels A, et al. Cancer statistics, 2003. CA Cancer J Clin 2003;53:5–26.

2. Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer Statistics Review, 1975–2000. Bethesda, MD: National Cancer Institute, 2003. seer.cancer.gov/csr/1975_2000.

3. Pettersson B, Coleman MP, Ron E, et al. Iodine supplementation in Sweden and regional trends in thyroid cancer incidence by histopathologic type. Int J Cancer 1996;65:13–19.

4. Liu S, Semenciw R, Ugnat AM, et al. Increasing thyroid cancer incidence in Canada, 1970–1996: time trends and age-period-cohort effects. Br J Cancer 2001;85:1335–1339.

5. Burgess JR. Temporal trends for thyroid carcinoma in Australia: an increasing incidence of papillary thyroid carcinoma (1982–1997). Thyroid 2002;12:141–149.

6. Colonna M, Grosclaude P, Remontet L, et al. Incidence of thyroid cancer in adults recorded by French cancer registries (1978–1997). Eur J Cancer 2002;38:1762–1768.

7. Levi F, Randimbison L, Te VC, et al. Thyroid cancer in Vaud, Switzerland: an update. Thyroid 2002;12:163–168.

8. Zheng TZ, Holford TR, Chen YT, et al. Time trend and age-period-cohort effect on incidence of thyroid cancer in Connecticut, 1935–1992. Int J Cancer 1996;67:504–509.

9. Akslen LA, Haldorsen T, Thoresen SØ, et al. Incidence pattern of thyroid cancer in Norway—influence of birth cohort and time period. Int J Cancer 1993;53:183–187.

10. Hrafnkelsson J, Jonasson JG, Sigurdsson G, et al. Thyroid cancer in Iceland 1955–1984. Acta Endocrinol (Copenh) 1988;118: 566–572.

11. Akslen LA, Haldorsen T, Thoresen SO, et al. Incidence of thyroid cancer in Norway 1970–1985: population review on time trend, sex, age, histological type and tumour stage in 2625 cases. APMIS 1990;98:549–558.

12. Correa P, Chen VW. Endocrine gland cancer. Cancer 1995; 75:338–352.

13. Ron E. Thyroid cancer. In: Schottenfeld D, Fraumeni Jr JF, eds. Cancer epidemiology and prevention, 2nd ed. New York: Oxford University Press, 1996:1000–1021.

14. Parkin DM, Whelan SL, Ferlay J, et al. Cancer incidence in five continents. Vol. VIII. Lyon: IARC Scientific, 2002.

15. Lundgren CI, Hall P, Ekbom A, et al. Incidence and survival of Swedish patients with differentiated thyroid cancer. Int J Cancer 2003;106:569–573.

16. Hedinger C, Williams ED, Sobin LH. International histological typing of thyroid tumours, 2nd ed. Berlin: Springer-Verlag, 1988.

17. dos Santos Silva I, Swerdlow AJ. Sex differences in the risks of hormone-dependent cancers. Am J Epidemiol 1993;138:10–28.

18. Williams ED. TSH and thyroid cancer. Horm Metab Res Suppl 1990;23:72–75.

19. Schneider AB, Robbins J. Ionizing radiation and thyroid cancer. In: Fagin J, ed. Thyroid cancer. Boston: Kluwer Academic, 1998:27–57.

20. Duffy BJ, Fitzgerald P. Thyroid cancer in childhood and adolescence: a report on twenty-eight cases. Cancer 1950;10:1018–1032.

21. Shore RE. Issues and epidemiological evidence regarding radiation-induced thyroid cancer. Radiat Res 1992;131:98–111.

22. UNSCEAR (United Nations Scientific Committee on the Effects of Radiation). Sources and effects of ionizing radiation. New York: United Nations, 2000.

23. Ron E, Lubin JH, Shore RE, et al. Thyroid cancer after exposure to external radiation: a pooled analysis of seven studies. Radiat Res 1995;141:259–277.

24. Pottern LM, Kaplan MM, Larsen PR, et al. Thyroid nodularity after childhood irradiation for lymphoid hyperplasia—a comparison of questionnaire and clinical findings. J Clin Epidemiol. 1990;43:449–460.

25. Schneider AB, Ron E, Lubin J, et al. Dose-response relationships for radiation-induced thyroid cancer and thyroid nodules: evidence for the prolonged effects of radiation on the thyroid. J Clin Endocrinol Metab 1993;77:362–369.

26. Wang Z, Boice JD Jr, Wei L, et al. Thyroid nodularity and chromosome aberrations among women in areas of high background radiation in China. J Natl Cancer Inst 1990;82:4784–185.

27. Wang JX, Boice JD Jr, Li BX, et al. Cancer among medical diagnostic X-ray workers in China. J Natl Cancer Inst 1988; 80:344–350.

28. Ron E, Modan B, Preston D, et al. Thyroid neoplasia following low-dose radiation in childhood. Radiat Res 1989;120:516–531.

29. Thompson DE, Mabuchi K, Ron E, et al. Cancer incidence in atomic bomb survivors. II. Solid tumors, 1958–1987. Radiat Res 1994;137(2 suppl):17–67

30. Robbins J, Adams WH. Radiation effects in the Marshall Islands. In: Nagataki S, ed. Radiation and the thyroid. Amsterdam: Excerpta Medica, 1989:11–24.

31. Shore RE, Hildreth N, Dvoretsky PM, et al. Thyroid cancer among persons given x-ray treatment in infancy for an enlarged thymus gland. Am J Epidemiol 1993;137:1068–1080.

32. Kerber RA, Till JE, Simon SL, et al. A cohort study of thyroid disease in relation to fallout from nuclear weapons testing. JAMA 1993;270:2076–2082.

33. Boice JD, Jr., Engholm G, Kleinerman RA, et al. Radiation dose and second cancer risk in patients treated for cancer of the cervix. Radiat Res 1988;116:3–55.

34. Tucker MA, Jones PH, Boice JD Jr, et al. Therapeutic radiation at a young age is linked to secondary thyroid cancer: the Late Effects Study Group. Cancer Res 1991;51:2885–2888.

35. Shore RE, Xue X. Comparative thyroid cancer risk of childhood and adult radiation exposure and estimation of lifetime risk. In: Thomas G, Karaoglou A, Williams ED, eds. Radiation and thyroid cancer. Singapore: World Scientific, 1999:491–498.

36. Robbins J, Schneider AB. Thyroid cancer following exposure to radioactive iodine. Rev Endocrin Metab Dis 2000;1:197–203.

37. Williams ED. Chernobyl, 15 years later, correlation of clinical, epidemiological and molecular outcomes. Ann Endocrinol 2003;64:72

38. National Council on Radiation Protection and Measurement. Induction of thyroid cancer by ionizing radiation. NCRP Report No. 80. Bethesda, MD: National Council on Radiation Protection and Measurement, 1985.

39. Conard RA. Late radiation effects in Marshall Islanders exposed to fallout 28 years ago. In: Boice JD Jr, Fraumeni JR, eds. Radiation carcinogenesis: epidemiology and biological significance. New York: Raven, 1984:57–71.

40. Takahashi T, Schoemaker MJ, Trott KR, et al. The relationship of thyroid cancer with radiation exposure from nuclear weapon testing in the Marshall Islands. J Epidemiol 2003;13:99–107.

41. Becker DV, Robbins J, Beebe GW, et al. Childhood thyroid cancer following the Chernobyl accident: a status report. Endocrinol Metab Clin North Am 1996;25:197–211.

42. Robbins J. Lessons from Chernobyl: the event, the aftermath fallout: radioactive, political, social. Thyroid 1997;7:189–192.

43. Robbins J, Schneider AB. Radioiodine-induced thyroid cancer: studies in the aftermath of the accident at Chernobyl. Trends Endocrinol Metab 1998;9:87–94.

44. Sobolev B, Heidenreich WF, Kairo I, et al. Thyroid cancer incidence in the Ukraine after the Chernobyl accident: comparison with spontaneous incidences. Radiat Environ Biophys 1997;36: 195–199.

45. Buglova EE, Kenigsberg JE, Sergeeva NV. Cancer risk estimation in Belarussian children due to thyroid irradiation as a consequence of the Chernobyl nuclear accident. Health Phys 1996; 71:45–49.

46. Astakhova LN, Anspaugh LR, Beebe GW, et al. Chernobyl-related thyroid cancer in children of Belarus: a case-control study. Radiat Res 1998;150:349–356.

47. Jacob P, Goulko G, Heidenreich WF, et al. Thyroid cancer risk to children calculated. Nature 1998;392:31–32.

48. Jacob P, Kenigsberg Y, Goulko G, et al. Thyroid cancer risk in Belarus after the Chernobyl accident: comparison with external exposures. Radiat Environ Biophys. 2000;39:25–31.

49. Ashizawa K, Shibata Y, Yamashita S, et al. Prevalence of goiter and urinary iodine excretion levels in children around Chernobyl. J Clin Endocrinol Metab 1997;82:3430–3433.

50. Kasatkina EP, Shilin DE, Rosenbloom AL, et al. Effects of low level radiation from the Chernobyl accident in a population with iodine deficiency. Eur J Pediatr 1997;156:916–920.

51. Bleuer JP, Averkin YI, Abelin T. Chernobyl-related thyroid cancer: What evidence for role of short-lived iodines? Environ Health Perspect 1997;105:1483–1486.

52. Gembicki M, Stozharov AN, Arinchin AN, et al. Iodine deficiency in Belarussian children as a possible factor stimulating the irradiation of the thyroid gland during the Chernobyl catastrophe. Environ Health Perspect 1997;105:1487–1490.

53. Karaoglou A, Chadwick KH. Health consequences of Chernobyl and other radiation accidents—Report on the European Union Cluster Contractors' workshop (San Miniato, Italy, 17–22 June 1997). Radiat Environ Biophys 1998;37: 1–9.

54. Ishigaki K, Namba H, Takamura N, et al. Urinary iodine levels and thyroid diseases in children; comparison between Nagasaki and Chernobyl. Endocr J 2001;48:591–595.

55. Robbins J, Dunn JT, Bouville A, et al. Iodine nutrition and the risk from radioactive iodine: a workshop report in the Chernobyl long-term follow-up study. Thyroid 2001;11:487–491.

56. Nikiforov YE, Rowland JM, Bove KE, et al. Distinct pattern of ret oncogene rearrangements in morphological variants of radiation-induced and sporadic thyroid papillary carcinomas in children. Cancer Res 1997;57:1690–1694.

57. Williams ED, Tronko ND. Molecular, cellular, biological characterization of childhood thyroid cancer. International Scientific Collaboration on Consequences of Chernobyl Accident. Luxembourg: European Commission, 1996.

58. Nikiforov Y, Gnepp DR. Pediatric thyroid cancer after the Chernobyl disaster—pathomorphologic study of 84 cases (1991–1992) from the Republic of Belarus. Cancer 1994;74: 748–766.

59. Lushnikov EF, Abrossimov AY. Histological characterization of papillary thyroid carcinoma in children, adolescents and young adults in Russia after the Chernobyl accident. In: Yamushita S. Chernobyl: message for the 21st century. New York: Elsevier Amsterdam, 2002:239–244.

60. Williams D. Cancer after nuclear fallout: lessons from the Chernobyl accident. Nat Rev Cancer 2002;2:543–549.

61. Harach HR, Williams ED. Childhood thyroid cancer in England and Wales. Br J Cancer 1995;72:777–783.

62. National Cancer Institute. Estimated exposures and thyroid doses received by the American people from I-131 in fallout following Nevada atmospheric nuclear bomb tests. Bethesda, MD: National Institutes of Health, 1997.

63. Shipler DB, Napier BA, Farris WT, et al. Hanford environmental dose reconstruction project: an overview. Health Phys 1996; 71:532–544.

64. Gilbert ES, Tarone R, Bouville A, et al. Thyroid cancer rates and I-131 doses from Nevada atmospheric nuclear bomb tests. J Natl Cancer Inst 1998;90:1654–1660.

65. National Research Council and Institute of Medicine. Exposure of the American people to iodine-131 from Nevada nuclear-bomb tests: review of the National Cancer Institute report and public health implications. Washington, DC: National Academy Press, 1999.

66. Davis S, Kopecky KJ, Hamilton TE. Hanford Thyroid Disease Study Final Report. Seattle: Fred Hutchinson Cander Research Center, 2002.

67. Dobyns BM, Sheline GE, Workman JB, et al. Malignant and benign neoplasms of the thyroid in patients treated for hyperthyroidism: a report of the Cooperative Thyrotoxicosis Therapy Study. J Clin Endocrinol Metab 1974;38:976–998.

68. Holm L-E, Wilkund KE, Lundell GE, et al. Thyroid cancer after diagnostic doses of iodine-131: a retrospective cohort study. J Natl Cancer Inst 1988;80:1132–1138.

69. Holm L-E, Hall P, Wiklund K, et al. Cancer risk after iodine-131 therapy for hyperthyroidism. J Natl Cancer Inst 1991;83: 1072–1077.

70. Hall P, Berg G, Bjelkengren G, et al. Cancer mortality after iodine-131 therapy for hyperthyroidism. Int J Cancer 1992;50: 886–890.

71. Hall P, Lundell G, Holm L-E. Mortality in patients treated for hyperthyroidism with iodine-131. Acta Endocrinol 1993;128: 230–234.

72. Hall P, Furst CJ, Mattsson A, et al. Thyroid nodularity after diagnostic administration of iodine-131. Radiat Res 1996;146: 673–682.

73. Hall P, Mattsson A, Boice JD Jr. Thyroid cancer after diagnostic administration of iodine-131. Radiat Res 1996;145:86–92.

74. Cooper DS. Radioiodine for hyperthyroidism—where do we stand after 50 years? JAMA 1998;280:375–376.

75. Franklyn JA, Maisonneuve P, Sheppard MC, et al. Mortality after the treatment of hyperthyroidism with radioactive iodine. N Engl J Med 1998;338:712–718.

76. Ron E, Doody MM, Becker DV, et al. Cancer mortality following treatment for adult hyperthyroidism. JAMA 1998;280:347–355.

77. Hahn K, Schnell-Inderst P, Grosche B, et al. Thyroid cancer after diagnostic administration of iodine-131 in childhood. Radiat Res 2001;156:61–70.

78. Dickman PW, Holm LE, Lundell G, et al. Thyroid cancer risk after thyroid examination with I-131: a population-based cohort study in Sweden. Int J Cancer 2003;106:580–587.

79. Franklyn JA, Maisonneuve P, Sheppard M, et al. Cancer incidence and mortality after radioiodine treatment for hyperthyroidism: a population-based cohort study. Lancet 1999;353: 2111–2115.

80. Becker DV, Braverman LE, Dunn JT, et al. The use of iodine as a thyroid blocking agent in the event of a reactor accident. Report of the Environmental Hazards Committee of the American Thyroid Association. JAMA 1984:659–661.

81. Nauman J, Wolff J. Iodide prophylaxis in Poland after the Chernobyl reactor accident: benefits and risks. Am J Med 1993; 94:524–532.

82. Becker DV, Zanzonico P. Potassium iodide for thyroid blockade in a reactor accident: administrative policies that govern its use. Thyroid 1997;7:193–197.

83. US Food and Drug Administration Center for Drug Evaluation Branch. Guidance: potassium iodide as a thyroid blocking agent in radiation emergencies. Rockville, MD: 2001.

84. Schneider AB, Becker DV, Robbins J. Protecting the thyroid from accidental or terrorist-instigated 131I releases. Thyroid 2002; 12:271–272.

85. UNSCEAR (United Nations Scientific Committee on the Effects of Radiation). Sources and effects of ionizing radiation. New York: United Nations, 1994.

86. Hall P, Holm L-E. Radiation-associated thyroid cancer—facts and fiction. Acta Oncol 1998;37:325–330.

87. Lundell M, Hakulinen T, Holm L-E. Thyroid cancer after radiotherapy for skin hemangioma in infancy. Radiat Res 1994; 140:334–339.

88. Lindberg S, Karlsson P, Arvidsson B, et al. Cancer incidence after radiotherapy for skin haemangioma during infancy. Acta Oncol 1995;34:735–740.

89. Ronckers CM, van Leeuwen FE, Hayes RB, et al. Cancer incidence after nasopharyngeal radium irradiation. Epidemiology 2002;13:552–560.

90. Ronckers CM, Land CE, Hayes RB, et al. Late health effects of childhood nasopharyngeal radium irradiation: nonmelanoma skin cancers, benign tumors, and hormonal disorders. Pediatr Res 2002;52:850–858.

91. Sandler DP, Comstock GW, Matanoski GM. Neoplasms following childhood radium irradiation of the nasopharynx. J Natl Cancer Inst 1982;68:3.

92. Yeh H, Matanoski GM, Wang N, et al. Cancer incidence after childhood nasopharyngeal radium irradiation: a follow-up study in Washington County, Maryland. Am J Epidemiol 2001; 153:749–756.

93. Kaplan MM, Garnick MB, Gelber R, et al. Risk factors for thyroid abnormalities after neck irradiation for childhood cancer. Am J Med 1983;74:272–280.

94. Fleming ID, Black TL, Thompson EI, et al. Thyroid dysfunction and neoplasia in children receiving neck irradiation for cancer. Cancer 1985;55:1190–1194.

95. Pasqualini T, Iorcansky S, Gruneiro L, et al. Thyroid dysfunction in Hodgkin's disease. Cancer 1989;63:335–339.

96. Schneider AB. Cancer therapy and endocrine disease: radiation-induced thyroid tumours. In: Sheaves R, Jenkins PJ, Wass JA, eds. Clinical endocrine oncology. Oxford: Blackwell Science, 1997:514–517.

97. Hancock SL, Cox RS, Mcdougall IR. Thyroid diseases after treatment of Hodgkin's disease. N Engl J Med 1991;325:599–605.

98. Huang J, Walker R, Groome PG, et al. Risk of thyroid carcinoma in a female population after radiotherapy for breast carcinoma. Cancer 2001;92:1411–1418.

99. Adjadj E, Rubino C, Shamsaldim A, et al. The risk of multiple primary breast and thyroid carcinomas—role of the radiation dose. Cancer 2003;98:1309–1317.

100. Inskip PD, Ekbom A, Galanti MR, et al. Medical diagnostic x rays and thyroid cancer. J Natl Cancer Inst 1995;87:1613–1621.

101. Hallquist A, Nasman A. Medical diagnostic x-ray radiation—an evaluation from medical records and dentist cards in a case-control study of thyroid cancer in the northern medical region of Sweden. Eur J Cancer Prev 2001;10:147–152.

102. Sarne DH, Schneider AB. External radiation and thyroid neoplasia. Endocrinol Metab Clin North Am 1996;25:181–195.

103. Schneider AB, Bekerman C, Leland J, et al. Thyroid nodules in the follow-up of irradiated individuals: comparison of thyroid ultrasound with scanning and palpation. J Clin Endocrinol Metab 1997;82:4020–4027.

104. Inskip PD, Hartshorne MF, Tekkel M, et al. Thyroid nodularity and cancer among Chernobyl cleanup workers from Estonia. Radiat Res 1997;147:225–235.

105. Favus M, Schneider A, Stachura M, et al. Thyroid cancer occurring as a late consequence of head-and-neck irradiation. N Engl J Med 1976;294:1019–1025.

106. Brander A, Viikinkoski P, Nickels J, et al. Thyroid gland—US screening in middle-aged women with no previous thyroid disease. Radiology 1989;173:507–510.

107. Jarl¿v AE, Nygârd B, Hegedus L, et al. Observer variation in ultrasound assessment of the thyroid gland. Br J Radiol 1993; 66:625–627.

108. Fischer AH, Bond JA, Taysavang P, et al. Papillary thyroid carcinoma oncogene (RET/PTC) alters the nuclear envelope and chromatin structure. Am J Pathol 1998;153:1443–1450.

109. Hatipoglu BA, Gierlowski T, Shore-Freedman E, et al. Fine-needle aspiration of thyroid nodules in radiation-exposed patients. Thyroid 2000;10:63–69.

110. Kikuchi S, Perrier ND, Ituarte PHG, et al. Accuracy of fine-needle aspiration cytology in patients with radiation-induced thyroid neoplasms. Br J Surg 2003;90:755–758.

111. Sarne DH, Schneider AB. Evaluation and management of patients exposed to childhood head and neck irradiation. Endocrinologist 1995;5:304–307.

112. Schneider AB, Favus MJ, Stachura ME, et al. Plasma thyroglobulin in detecting thyroid carcinoma after childhood head and neck irradiation. Ann Intern Med 1977;86:29–34.

113. Nagataki S, Shibata Y, Inoue S, et al. Thyroid diseases among atomic bomb survivors in Nagasaki. JAMA 1994;272:364–370.

114. Pacini F, Vorontsova T, Molinaro E, et al. Prevalence of thyroid autoantibodies in children and adolescents from Belarus exposed to the Chernobyl radioactive fallout. Lancet 1998;352: 763–766.

115. Schneider A, Bekerman C, Favus M, et al. Continuing occurrence of thyroid-nodules after head and neck irradiation: relationship to plasma thyroglobulin concentration. Ann Intern Med 1981;94:176–180.

116. Schneider AB, Shore-Freedman E, Ryo UY, et al. Radiation-induced tumors of the head and neck following childhood irradiation: prospective studies. Medicine 1985;64:1–15.

117. Kaplan MM. Thyroid hormone prophylaxis after radiation exposure. In: Robbins J, ed. Treatment of thyroid cancer in childhood. Springfield, VA: National Technical Information Service, 1994:143.

118. Robbins J. Thyroid suppression therapy for prevention of thyroid tumors after radiation exposure. In: Degroot LJ, Frohman LA, Kaplan EL, et al, eds. Radiation-associated thyroid carcinoma. New York: Grune & Stratton, 1976:419.

119. Murphy ED, Scanlon EF, Garces RM, et al. Thyroid hormone administration in irradiated patients. J Surg Oncol 1986;31: 214–217.

120. Degroot LJ. Diagnostic approach and management of patients exposed to irradiation to the thyroid. J Clin Endocrinol Metab 1989;69:925–928.

121. Tisell LE, Carlsson S, Fjalling M, et al. Hyperparathyroidism subsequent to neck irradiation: risk factors. Cancer 1985;56: 1529–1533.

122. Fujiwara S, Sposto REH, Akiba S, et al. Hyperparathyroidism among atomic bomb survivors in Hiroshima. Radiat Res 1992; 130:372–378.

123. Schneider AB, Gierlowski TC, Shore-Freedman E, et al. Dose-response relationships for radiation-induced hyperparathyroidism. J Clin Endocrinol Metab 1995;80:254–257.

124. Rasmuson T, Damber L, Johansson L, et al. Increased incidence of parathyroid adenomas following x-ray treatment of benign diseases in the cervical spine in adult patients. Clin Endocrinol (Oxf) 2002;57:731–734.

125. Land CE, Saku T, Hayashi Y, et al. Incidence of salivary gland tumors among atomic bomb survivors, 1950–87: evaluation of radiation-related risk. Radiat Res 1996;146:28–36.

126. Schneider AB, Lubin J, Ron E, et al. Salivary gland tumors after childhood radiation treatment for benign conditions of the head and neck: dose-response relationships. Radiat Res 1998;149:625–630.

127. Saku T, Hayashi Y, Takahara O, et al. Salivary gland tumors among atomic bomb survivors, 1950–1987. Cancer 1997;79: 1465–1475.

128. Modan B, Chetrit A, Alfandary E, et al. Increased risk of salivary gland tumors after low-dose irradiation. Laryngoscope 1998;108:1095–1097.

129. Ron E, Modan B, Boice JD Jr, et al. Tumors of the brain and nervous system after radiotherapy in childhood. N Engl J Med 1988;319:1033–1039.

130. Schneider AB, Shore-Freedman E, Weinstein R. Radiation-induced thyroid and other head and neck tumors: occurrence of multiple tumors and analysis of risk factors. J Clin Endocrinol Metab 1986;63:107–112.

131. Sznajder L, Abrahams C, Parry DM, et al. Multiple schwannomas and meningiomas associated with irradiation in childhood. Arch Intern Med 1996;156:1873–1878.

132. Shore RE, Albert RE, Reed M, et al. Skin cancer incidence among children irradiated for ringworm of the scalp. Radiat Res 1984;100:192–204.

133. Ron E, Modan B, Preston D, et al. Radiation-induced skin carcinomas of the head and neck. Radiat Res 1991;125:318–325.

134. Ron E, Preston DL, Kishikawa M, et al. Skin tumor risk among atomic-bomb survivors in Japan. Cancer Causes Control 1998; 9:393–401.

135. Schneider AB, Recant W, Pinsky S, et al. Radiation-induced thyroid carcinoma: clinical course and results of therapy in 296 patients. Ann Intern Med 1986;105:405–412.

136. Rubino C, Cailleux AF, Abbas M, et al. Characteristics of follicular cell–derived thyroid carcinomas occurring after external radiation exposure: results of a case control study nested in a cohort. Thyroid 2002;12:299–304.

137. Samaan N, Schultz P, Ordonez N, et al. A comparison of thyroid carcinoma in those who have and have not had head and neck irradiation in childhood.J Clin Endocrinol Metab 1987; 64:219–223.

138. Viswanathan K, Gierlowski TC, Schneider AB. Childhood thyroid cancer—characteristics and long-term outcome in children irradiated for benign conditions of the head and neck. Arch Pediatr Adolesc Med 1994;148:260–265.

139. Shimaoka K, Getaz EP, Rao U. Anaplastic carcinoma of thyroid: radiation-associated. NY State J Med 1979;79:874–877.

140. Mazzaferri EL, Young RL. Papillary thyroid carcinoma: a 10 year follow-up report of the impact of therapy in 576 patients. Am J Med 1981;70:511–518.

141. Fogelfeld L, Wiviott MBT, Shore-Freedman E, et al. Recurrence of thyroid nodules after surgical removal in patients irradiated in childhood for benign conditions. N Engl J Med 1989; 320:835–840.

142. Reiners C, Biko J, Demidchik EP, et al. Results of radioactive iodine treatment in children from Belarus with advanced stages of thyroid cancer after the Chernobyl accident. In: Yamushita S. Chernobyl: message for the 21st century. New York: Elsevier Amsterdam, 2002:205–214.

143. Oliynyk V, Epshtein O, Sovenko T, et al. Post-surgical ablation of thyroid residues with radioiodine in Ukrainian children and adolescents affected by post-Chernobyl differentiated thyroid cancer. J Endocrinol Invest 2001;24:445–447.

144. Hizawa K, Iida M, Aoyagi K, et al. Thyroid neoplasia and familial adenomatous polyposis/Gardner's syndrome. J Gastroenterol 1997;32:196–199.

145. Bulow C, Bulow S, Berk T, et al. Is screening for thyroid carcinoma indicated in familial adenomatous polyposis? Int J Colorectal Dis 1997;12:240–242.

146. Perrier ND, van Heerden JA, Goellner JR, et al. Thyroid cancer in patients with familial adenomatous polyposis. World J Surg 1998;22:738–743.

147. Harach HR, Williams GT, Williams ED. Familial adenomatous polyposis associated thyroid carcinoma: a distinct type of follicular cell neoplasm. Histopathology 1994;25:549–561.

148. Fenton PA, Clarke SEM, Owen W, et al. Cribriform variant papillary thyroid cancer: a characteristic of familial adenomatous polyposis. Thyroid 2001;11:193–197.

149. Cetta F, Montalto G, Gori M, et al. Germline mutations of the APC gene in patients with familial adenomatous polyposis-associated thyroid carcinoma: results from a European cooperative study. J Clin Endocrinol Metab 2000;85:286–292.

150. Cetta F, Curia MC, Montalto G, et al. Thyroid carcinoma usually occurs in patients with familial adenomatous polyposis in the absence of biallelic inactivation of the adenomatous polyposis coli gene. J Clin Endocrinol Metab 2001;86:427–432.

151. Thyresson HN, Doyle JA. Cowden's disease (multiple hamartoma syndrome). Mayo Clin Proc 1981;56:179–184.

152. McIver B, Eberhardt NL. Cowden disease and the PTEN/ MMAC1 gene. In: Akamizu T, Kasuga M, Davies TF (eds), Genetics of Complex Thyroid Diseases. New York, Tokyo: Springer. 2002: 151–175.

153. Goldgar DE, Easton DF, Cannon-Albright LA, et al. Systematic population-based assessment of cancer risk in first-degree relatives of cancer probands.J Natl Cancer Inst 1994;86: 1600–1608.

154. Frich L, Glattre E, Akslen LA. Familial occurrence of nonmedullary thyroid cancer: a population-based study of 5673 first-degree relatives of thyroid cancer patients from Norway. Cancer Epidemiol Biomarkers Prev 2001;10:113–117.

155. Hemminki K, Dong CH. Familial relationships in thyroid cancer by histo-pathological type. Int J Cancer 2000;85:201–205.

156. Stoffer SS, Van Dyke DL, Bach JV, et al. Familial papillary carcinoma of the thyroid. Am J Med Genet 1986;25:775–782.

157. Ron E, Kleinerman R, Boice JD Jr, et al. A population-based case-control study of thyroid cancer. J Natl Cancer Inst 1987;79:1–12.

158. Pal T, Vogl FD, Chappuis PO, et al. Increased risk for nonmedullary thyroid cancer in the first degree relatives of prevalent cases of nonmedullary thyroid cancer: a hospital-based study. J Clin Endocrinol Metab 2001;86:5307–5312.

159. Grossman RF, Tu SH, Duh QY, et al. Familial nonmedullary thyroid cancer: an emerging entity that warrants aggressive treatment. Arch Surg 1995;130:892–899.

160. Loh KC. Familial nonmedullary thyroid carcinoma: a meta-review of case series. Thyroid 1997;7:107–113.

161. Alsanea O. Familial nonmedullary thyroid cancer. Curr Treat Options Oncol 2000;1:345–351.

162. Uchino S, Noguchi S, Kawamoto H, et al. Familial nonmedullary thyroid carcinoma characterized by multifocality and a high recurrence rate in a large study population. World J Surg 2002;26:897–902.

163. Brunaud L, Zarnegar R, Wada N, et al. Chromosomal aberrations by comparative genomic hybridization in thyroid tumors in patients with familial nonmedullary thyroid cancer. Thyroid 2003;13:621–629.

164. Kraimps JL, Bouin-Pineau MH, Amati P, et al. Familial papillary carcinoma of the thyroid. Surgery 1997;121:715–718.

165. Malchoff CD, Sarfarazi M, Tendler B, et al. Familial papillary thyroid carcinoma is genetically distinct from familial adenomatous polyposis coli. Thyroid 1999;9:247–252.

166. Cohen J, Gierlowski TC, Schneider AB. A prospective study of hyperparathyroidism in individuals exposed to radiation in childhood. JAMA 1990;264:581–584.

167. Mihailescu D, Shore-Freedman E, Mukani S, et al. Multiple neoplasms in an irradiated cohort: pattern of occurrence and relationship to thyroid cancer outcome. J Clin Endocrinol Metab 2002;87:3236–3241.

168. Perkel V, Gail MH, Lubin J, et al. Radiation-induced thyroid neoplasm: evidence for familial susceptibility factors. J Clin Endocrinol Metab 1988;66:1316–1322.

169. McTiernan AM, Weiss NS, Daling JR. Incidence of thyroid cancer in women in relation to previous exposure to radiation therapy and history of thyroid disease. J Natl Cancer Inst 1984; 73:575–581.

170. Preston-Martin S, Bernstein L, Pike MC, et al. Thyroid cancer among young women related to prior thyroid disease and pregnancy history. Br J Cancer 1987;55:191–195.

171. Franceschi S, Fassina A, Talamini R, et al. Risk factors for thyroid cancer in Northern-Italy. Int J Epidemiol 1989;18:578–584.

172. Levi F, Franceschi S, Lavecchia C, et al. Previous thyroid disease and risk of thyroid cancer in Switzerland. Eur J Cancer 1991; 27:85–88.

173. Preston-Martin S, Jin F, Duda MJ, et al. A case-control study of thyroid cancer in women under age 55 in Shanghai (People's Republic of China). Cancer Causes Control 1993;4:431–440.

174. Wingren G, Hatschek T, Axelson O. Determinants of papillary cancer of the thyroid. Am J Epidemiol 1993;138:482–491.

175. Akslen LA, Sothern RB. Seasonal variations in the presentation and growth of thyroid cancer. Br J Cancer 1998;77:1174–1179.

176. Memon A, Varghese A, Suresh A. Benign thyroid disease and dietary factors in thyroid cancer: a case-control study in Kuwait. Br J Cancer 2002;86:1745–1750.

177. Franceschi S, Preston-Martin S, Dal Maso L, et al. A pooled analysis of case-control studies of thyroid cancer. IV. Benign thyroid diseases. Cancer Causes Control 1999;10:583–595.

178. La Vecchia C, Ron E, Franceschi S, et al. A pooled analysis of case-control studies of thyroid cancer. III. Oral contraceptives, menopausal replacement therapy and other female hormones. Cancer Causes Control 1999;10:157–166.

179. Negri E, Dal Maso L, Ron E, et al. A pooled analysis of case-control studies of thyroid cancer. II. Menstrual and reproductive factors. Cancer Causes Control 1999;10:143–155.

180. Negri E, Ron E, Franceschi S, et al. A pooled analysis of case-control studies of thyroid cancer. I. Methods. Cancer Causes Control 1999;10:131–142.

181. Goldman MB, Monson RR, Maloof F. Cancer mortality in women with thyroid disease. Cancer Res 1990;50:2283–2289.

182. Iribarren C, Haselkorn T, Tekawa IS, et al. Cohort study of thyroid cancer in a San Francisco Bay area population. Int J Cancer 2001;93:745–750.

183. Haselkorn T, Stewart SL, Horn-Ross PL. Why are thyroid cancer rates so high in Southeast Asian women living in the United States? The bay area thyroid cancer study. Cancer Epidemiol Biomarkers Prev 2003;12:144–150.

184. Mellemgaard A, From G, Jorgensen T, et al. Cancer risk in individuals with benign thyroid disorders. Thyroid 1998;8:751–754.

185. From G, Mellemgaard A, Knudsen N, et al. Review of thyroid cancer cases among patients with previous benign thyroid disorders. Thyroid 2000;10:697–700.

186. Levi F, Franceschi S, Te VC, et al. Descriptive epidemiology of thyroid cancer in the Swiss canton of Vaud. J Cancer Res Clin Oncol 1990;116:639–647.

187. Akslen LA, Nilssen S, Kvâle G. Reproductive factors and risk of thyroid cancer: a prospective study of 63,090 women from Norway. Br J Cancer 1992;65:772–774.

188. Hallquist A, Hardell L, Degerman A, et al. Thyroid cancer: reproductive factors, previous diseases, drug intake, family history and diet: a case-control study. Eur J Cancer Prev 1994;3: 481–488.

189. Kolonel LN, Hankin JH, Wilkens LR, et al. An epidemiologic study of thyroid cancer in Hawaii. Cancer Causes Control 1990; 1:223–234.

190. Levi F, Franceschi S, Gulie C, et al. Female thyroid cancer: the role of reproductive and hormonal factors in Switzerland. Oncology 1993;50:309–315.

191. Kravdal Ø, Glattre E, Haldorsen T. Positive correlation between parity and incidence of thyroid cancer: new evidence based on complete Norwegian birth cohorts. Int J Cancer 1991; 49:831–836.

192. Galanti MR, Hansson L, Lund E, et al. Reproductive history and cigarette smoking as risk factors for thyroid cancer in women: a population-based case-control study. Cancer Epidemiol Biomarkers Prev 1996;5:425–431.

193. Modan B, Ron E, Lerner-Geva L, et al. Cancer incidence in a cohort of infertile women. Am J Epidemiol 1998;147:1038–1042.

194. Glattre E, Kravdal Ø. Male and female parity and risk of thyroid cancer. Int J Cancer 1994;58:616–617.

195. Brinton LA, Melton LJ III, Malkasian GD Jr, et al. Cancer risk after evaluation for infertility. Am J Epidemiol 1989;129: 712–722.

196. Ron E, Lunenfeld B, Menczer J, et al. Cancer incidence in a cohort of infertile women. Am J Epidemiol 1987;125:780–790.

197. Luoto R, Grenman S, Salonen S, et al. Increased risk of thyroid cancer among women with hysterectomies. Amer J Obstet Gynecol 2003;188:45–48.

198. Memon A, Darif M, Al Saleh K, et al. Epidemiology of reproductive and hormonal factors in thyroid cancer: evidence from a case-control study in the Middle East. Int J Cancer 2002;97:82–89.

199. Rossing MA, Voigt LF, Wicklund KG, et al. Reproductive factors and risk of papillary thyroid cancer in women. Am J Epidemiol 2000;151:765–772.

200. Sakoda LC, Horn-Ross PL. Reproductive and menstrual history and papillary thyroid cancer risk: the San Francisco Bay Area thyroid cancer study. Cancer Epidemiol Biomarker Prev 2002;11:51–57.

201. McTiernan AM, Weiss NS, Daling JR. Incidence of thyroid cancer in women in relation to reproductive and hormonal factors. Am J Epidemiol 1984;120:423–435.

202. Shore RE, Hildreth N, Dvoretsky P, et al. Benign thyroid adenomas among persons X-irradiated in infancy for enlarged thymus glands. Radiat Res 1993;134:217–223.

203. Wong FL, Ron E, Gierlowski T, et al. Benign thyroid tumors: general risk factors and their effects on radiation risk estimation. Am J Epidemiol 1996;144:728–733.

204. Wegelin C. Malignant disease of the thyroid and its relation to goiter in men and animals. Cancer Rev 1928;3:297.

205. Wynder EL. Some practical aspects of cancer prevention. N Engl J Med. 1952;246:573

206. Franceschi S, Talamini R, Fassina A, et al. Diet and epithelial cancer of the thyroid gland. Tumori 1990;76:331–338.

207. Pendergrast WJ, Milmore BK, Marcus SC. Thyroid cancer and thyrotoxicosis in the United States: their relation to endemic goiter. J Chronic Dis 1961;13:22

208. Bacher-Stier C, Riccabona G, Totsch M, et al. Incidence and clinical characteristics of thyroid carcinoma after iodine prophylaxis in an endemic goiter country. Thyroid 1997;7:733–741.

209. Bakiri F, Djemli FK, Mokrane LA, et al. The relative roles of endemic goiter and socioeconomic development status in the prognosis of thyroid carcinoma. Cancer 1998;82:1146–1153.

210. Feldt-Rasmussen U. Iodine and Cancer. Thyroid 2001;11: 483–486.

211. Pettersson B, Adami HO, Wilander E, et al. Trends in thyroid cancer incidence in Sweden, 1958–1981, by histopathologic type. Int J Cancer 1991;48:28–33.

212. Belfiore A, La Rosa GL, Padova G, et al. The frequency of cold thyroid nodules and thyroid malignancies in patients from an iodine-deficient area. Cancer 1987;60:3096–3102.

213. Bosetti C, Kolonel L, Negri E, et al. A pooled analysis of case-control studies of thyroid cancer. VI. Fish and shellfish consumption. Cancer Cause Control 2001;12:375–382.

214. Glattre E, Haldorsen T, Berg JP, et al. Norwegian case-control study testing the hypothesis that seafood increases the risk of thyroid cancer. Cancer Causes Control 1993;4:11–16.

215. Frich L, Akslen LA, Glattre E. Increased risk of thyroid cancer among Norwegian women married to fishery workers—a retrospective cohort study. Br J Cancer 1997;76:385–389.

216. Franceschi S, Levi F, Negri E, et al. Diet and thyroid cancer: a pooled analysis of four European case-control studies. Int J Cancer 1991;48:395–398.

217. Bosetti C, Negri E, Kolonel L, et al. A pooled analysis of case-control studies of thyroid cancer. VII. Cruciferous and other vegetables (International). Cancer Cause Control 2002;13:765–775.

218. Galanti MR, Hansson L, Bergstrom R, et al. Diet and the risk of papillary and follicular thyroid carcinoma: a population-based case-control study in Sweden and Norway. Cancer Cause Control 1997;8:205–214.

219. Linos A, Linos DA, Vgotza N, et al. Does coffee consumption protect against thyroid disease? Acta Chir Scand 1989;155: 317–320.

220. Takezaki T, Hirose K, Inoue M, et al. Risk factors of thyroid cancer among women in Tokai, Japan. J Epidemiol 1996;6:140–147.

221. Mack WJ, Preston-Martin S, Bernstein L, et al. Lifestyle and other risk factors for thyroid cancer in Los Angeles county females. Ann Epidemiol 2002;12:395–401.

222. Kreiger N, Parkes R. Cigarette smoking and the risk of thyroid cancer. Eur J Cancer 2000;36:1969–1973.

223. Rossing MA, Cushing KL, Voigt LF, et al. Risk of papillary thyroid cancer in women in relation to smoking and alcohol consumption. Epidemiology 2000;11:49–54.

224. Fisher CL, Mannino DM, Herman WH, et al. Cigarette smoking and thyroid hormone levels in males. Int J Epidemiol 1997; 26:972–977.

225. Capen CC. Mechanisms of chemical injury of thyroid gland. Prog Clin Biol Res 1994;387:173–191.

226. Hurley PM. Mode of carcinogenic action of pesticides inducing thyroid follicular cell tumors in rodents. Environ Health Perspect 1998;106:437–445.

227. Capen CC. Mechanistic data and risk assessment of selected toxic end points of the thyroid gland. Toxicol Pathol 1997;25: 39–48.

228. Williams ED. Mechanisms and pathogenesis of thyroid cancer in animals and man. Mutat Res 1995;333:123–129.

229. Cooper DS, Axelrod L, Degroot LJ, et al. Congenital goiter and the development of metastatic follicular carcinoma with evidence for a leak of nonhormonal iodide: clinical, pathological, kinetic, and biochemical studies and a review of the literature. J Clin Endocrinol Metab 1981;52:294–306.

230. Medeiros-Neto G, Gil-Da-Costa MJ, Santos CLS, et al. Metastatic thyroid carcinoma arising from congenital goiter due to mutation in the thyroperoxidase gene. J Clin Endocrinol Metab 1998;83:4162–4166.