Executive Summary. Purpose and Rationale

Transcription

1 Executive Summary Purpose and Rationale The primary purpose of this Report is to update the National Council on Radiation Protection and Measurements (NCRP) Report No. 80, Induction of Thyroid Cancer by Ionizing Radiation, first published in 1985, and reprinted in NCRP Report No. 80 (NCRP, 1985a) entailed an initial analysis of the risk of thyroid cancer from: (1) external radiation from a variety of sources, including studies undertaken in Israel, Japan, and the United States; and (2) internal radiation (notably 131 I) from fallout, and diagnostic and therapeutic medical procedures. The modifying effect of ethnic background was also analyzed. The literature surveyed in NCRP Report No. 80 included 147 references, spanning the period from 1949 to That report was comprised of 11 sections and four appendices, a total of 94 pages. The general conclusions of NCRP Report No. 80 (NCRP, 1985a) were as follows: Women appear to have at least twice the risk of men for clinically apparent (thyroid) cancers at a given exposure. Data suggesting that children are more susceptible than adults warrant a 50 % reduction in risk coefficients when estimates derived for people less than or equal to 18 y at exposure are applied to a population of adults. Human experience and much animal data suggest that 131 I is less carcinogenic to the thyroid, per 0.01 Gy absorbed dose, than external radiation. Iodine-131 is considered to be no more than one-third as effective as external radiation in the induction of thyroid cancer in the general population. For the calculation of risks of fatal (thyroid) cancer, current levels of medical diagnosis and care are assumed, and a maximum of 10 % of the clinically evident radiation-induced thyroid cancers are expected to be lethal. After exposure to external irradiation, the projected overall lifetime incidence of fatal thyroid cancer would be 7.5 cases per 0.01 Gy absorbed dose to the thyroid in a general population of one million persons. This estimate is consistent 1

2 2 / EXECUTIVE SUMMARY with earlier lifetime estimates from the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR, 1977) report (5 to 15 cases), the ICRP (1977) report (five cases), and the National Academy of Sciences/ National Research Council (NAS/NRC, 1980) report (6 to 18 cases) for similar exposures. Ethnic background was found to influence the risk of radiation-induced thyroid cancer [e.g., the relative risk for Jews compared to non-jews was about 3.5 after adjusting for gender, time since exposure (TSE), and dose]. NCRP Report No. 80 acknowledged that large gaps in the existing data, the low incidence of thyroid cancer, and the small size of populations available for study make risk derivations uncertain. The report also indicated a need for further data from laboratory animals on the comparative aspects (x rays versus 131 I) of radiationinduced thyroid carcinogenesis at low doses, including other rodent strains and species exposed early and late in life and with testing for whether or not latency is dose related. Information on the carcinogenicity of 123 I and 99m Tc, both used for medical imaging of the thyroid gland because they yield superior image quality and lower doses to the thyroid, were deemed needed. Twenty-three years have passed since NCRP Report No. 80 was published. The Three-Mile Island nuclear reactor accident occurred in 1979 [but with no significant release of radioactive material (0.74 TBq of 131 I] or radiation exposures to the general surrounding population) and increased public concern about the release of fission products as a result of a nuclear reactor accident. In 1986, the Chernobyl nuclear reactor accident in Ukraine released large amounts of radioactive materials (including 1.8 EBq of 131 I) to the surrounding areas and also exposed large numbers of civilians of all age groups, including fetuses, and cleanup workers to external and/or internal radiation. Concern for the populations in and surrounding the Hanford Nuclear Reactors in Washington State arose when information about releases of radioactive materials, particularly 131 I [27 PBq (Napier, 2002)], which occurred largely between 1944 and 1947, was made public in the mid-1980s. The incidence of thyroid cancer in the United States has increased in recent years, likely due to an increased ability to detect thyroid cancer with the use of diagnostic ultrasound (Davies and Welch, 2006). Improved follow-up of patients and populations exposed (and controls) has facilitated further elucidation of short- and longterm consequences for radiation-induced thyroid cancers and increased the overall database for risk assessment. Improved risk

3 SYNOPSIS OF THIS REPORT / 3 models, surveillance procedures, and statistical approaches have been developed and employed. These collective factors provided the rationale for the present Report. This Report is intended to be comprehensive and to serve as an authoritative reference on risks to the thyroid from ionizing radiation and other relevant topics. This Executive Summary describes NCRP s key findings and conclusions and also provides a road map for the interested reader to the balance of this Report. Goals The present NCRP Report has five goals: Review all major epidemiological studies published in the English language through December 2006 that deal with thyroid and parathyroid disease related to exposure to ionizing radiations, with emphasis on the induction of thyroid cancer. Review the conclusions of earlier evaluations by NAS/NRC and UNSCEAR on the induction of thyroid disease related to exposure to ionizing radiation. Review the physics and biology associated with dose to the thyroid. Provide recommendations on the magnitude of radiation risks with doses for induction of thyroid disease, especially thyroid cancer, with emphasis on the importance of gender, age at time of exposure, TSE, exposure rate, and ethnicity. Provide recommendations on the relative biological effectiveness (RBE) of different radiations with emphasis on RBE of internal exposure of the thyroid from 131 I as compared to external exposure of the thyroid from kilovoltage xrays. These goals are addressed in seven sections and eight related appendices. The literature covered by this Report includes more than 750 references that were published from 1896 to Synopsis of this Report Section 1 provides a brief sequential outline of the contents of this Report: provision of an overview of the anatomy, physiology, and pathophysiology of the thyroid and parathyroid glands;

4 4 / EXECUTIVE SUMMARY critical review of radiation dosimetry among human cohorts exposed to medical and nonmedical radiation and subsequently evaluated for radiation-associated disease of the thyroid or parathyroid glands; derivation of absolute and relative risk factors for radiationassociated thyroid cancer; and recommendations for medical follow-up of individuals receiving significant radiation exposure to the thyroid and who have excess risk for thyroid disease, especially cancer. Of note is the fact that the reported incidence in the United States of thyroid cancer has risen from 3.6 per 100,000 in 1973 to 8.7 per 100,000 in 2002, a statistically-significant 2.4-fold increase. This increase is attributed to improved detection procedures of small papillary thyroid cancers. The mortality rate from thyroid cancers (all ages, all races, and both genders) has, however, remained low and stable, at 0.5 deaths per 100,000 persons. Section 2 presents an analysis of the anatomy and physiology, including embryological and neonatal development, of the thyroid gland. This latter aspect is particularly relevant since children (exposed in utero or during the first years of life) are shown later (Sections 4 and 5) to be particularly sensitive to radiation-induced thyroid cancer compared with exposures later in life. The thyroid gland is unique in that it concentrates iodine 500-fold and produces thyroid hormones whose molecules each can have three or four iodine atoms; this fact explains why, on the one hand, there is a daily need for iodine in the diet to maintain a healthy, functioning thyroid gland and why, on the other hand, radioactive iodines (e.g., 131 I from reactor loss of containment accidents or atomic-bomb fallout) that enter the food chain predominantly through the pasturecow-milk human pathway (cows eating 131 I-contaminated foliage; the radioiodine is concentrated in the cow s milk, which is consumed by humans, with the radioiodine concentrating in the thyroid glands) can lead to large thyroid doses and the subsequent development of thyroid cancers. The introduction of diagnostic ultrasound has greatly increased the sensitivity of medical evaluation in detecting abnormal thyroid anatomy. Thyroid cancer occurs in all age groups. Women are more prone than men to this disease. In 2006, there was a prediction of 30,180 new cases of thyroid cancer in the United States, and in this period it was expected that 1,500 people would die from this disease. There are few thyroid fatalities under the age of 40 y and there are some ethnic differences for incidence and mortality with thyroid cancer. Children are a special group of individuals whose thyroid cancers present in a

5 SYNOPSIS OF THIS REPORT / 5 manner different from that of adults; children generally have more advanced disease at the time of diagnosis (i.e., higher rates of local invasion and distant metastasis) than adults. In the past, external beam radiation therapy (EBRT) has been used in treating certain benign medical abnormalities. In addition, low doses of radioactive iodine have been used to evaluate thyroid function, and high doses of radioactive iodine have been used to treat hyper-functioning thyroid tissue and thyroid cancer. Our knowledge of genetic alterations in the thyroid in the etiology of thyroid cancers is increasing. Section 3 deals with radiation dosimetry and dose reconstruction as related to thyroid exposures. The issue is complicated for several reasons: The dose from external exposures to atomic-bomb detonations or nuclear reactor accidents has to be estimated from a number of indirect measurements and assumptions. Such considerations involve but are not limited to estimates of the radiation exposures, distance from the hypocenter of the emission site, presence or absence of shielding, and approximations of the various types of radiation emissions. The dose from internal exposures (e.g., absorbed 131 I) involves uncertainties related to the amount of radioiodine ingested or inhaled, the distribution of the internalized radioiodine in the body and biological half-lives. The fact that most of the dose from 131 I is from beta particles requires consideration of the anatomy and physiology of the thyroid gland. Most of the iodine localizes in the thyroid follicles, making estimates of the dose to the target cells more complex, especially in abnormal thyroid glands that can have follicles of varying sizes and function. In the normal thyroid gland, the distribution of radioactive iodine is reasonably homogeneous within the thyroid, thus facilitating dose estimations. The fact that 131 I concentrates in the colloid thereby reduces the dose to follicular cells at risk for cancer induction (NAS/NRC, 1996). The environmental dispersion of radioiodine is also complex, as two individuals, each equally distant but in opposite direction from the release site, may subsequently demonstrate vastly different uptakes of the radioiodine. Meteorological conditions greatly affect dispersion direction and food-chain aspects of dietary contamination from inadvertent or accidental releases of radionuclides to the environment.

6 6 / EXECUTIVE SUMMARY Potassium iodide, if orally administered just before or just after such accidental release of radioactive iodine can nearly completely block thyroidal uptake of radioiodine. The dietary sufficiency/insufficiency for iodine of each individual needs to be understood but often can only be estimated in terms of uptake of radioactive iodine. Dietary sufficiency can partly mitigate 131 I uptake, but dietary insufficiency would allow enhanced 131 I uptake. The four major cohorts exposed to internal radiation from environmental releases of radioiodine are discussed: Nevada Test Site (NTS) Marshall Islands Hanford Site Chernobyl nuclear reactor accident Each cohort presented with different exposure conditions, including releases of radionuclides over widely ranging time frames. The various approaches to these four different cohorts are thorough, from dosimetry determinations (including reconstructions) to medical follow-up and analyses. Section 3 closes with commentary and tabulated analyses of dose estimates from other epidemiological reports of thyroid cancer from external exposure during childhood, from internal ( 131 I) exposures in adults, and of thyroid nodules in relation to external or internal irradiation of thyroids in adults and children. Section 4 provides an overview of the types of studies used to determine the effects of radiation on the thyroid. This section is divided into two major parts, animal data and epidemiologic studies. Data from experiments with animals have relevance to humans because, as discussed in NCRP Report No. 150 (NCRP, 2005), thyroid carcinogenesis is essentially similar among mammalian species. This fact allows for extrapolation from animal studies (e.g., rats, mice, dogs) to humans and for design of experiments, including control of all experimental variables, which is not possible with humans. An initial observation with animal studies was that high doses yielded few thyroid cancers, but low doses yielded significant numbers of thyroid cancers. The lack of a neoplastic effect with the high doses was attributed to radiation-induced cell death, which was not observed with low doses. Thus, the lower doses were more carcinogenic than the higher doses. Irradiation of neonatal and juvenile dogs resulted in significantly more thyroid neoplasms than in mature dogs. RBE of comparable radiation exposures of

7 SYNOPSIS OF THIS REPORT / 7 mice and rats, from earlier studies of external (x-ray) radiation and internal 131 I radiation, was much higher for the external than the internal radiation, RBEs ranging from 2 to 10 (external/internal), respectively, depending on the study. However, the most recent rodent study suggests RBE of x rays and 131 I may be similar, but interpretation is not straightforward. Follicular cancers, and not papillary cancers, were the predominant cancers increased. The Long-Evans rat has a high natural rate of developing medullary carcinoma of the thyroid (>27 %), suggesting a peculiar genetic constitution. Thyroid adenomas had a different response, with 131 I being much less effective. The small size of the rat gland would result in a more uniform dose distribution. Methodologically sound epidemiological studies optimally possess: enrollment of exposed and unexposed individuals (a cohort study), or diseased and nondiseased individuals (casecontrol study); long-term (decades) follow-up; comparable study groups except for the variable of interest; precise dose estimates; range of doses; large number of participants; large range of ages of exposed individuals; and statistical control of confounding variables. The data from epidemiological studies provide the most valuable information on health effects from various radiation exposures in humans. Such studies, however, are not without uncertainties. For example, dose reconstruction involves assumptions that can include substantial uncertainties in the estimates of dose to individuals. In addition, measuring the effect is often difficult, especially when studying a disease like thyroid cancer where the incidence is very dependent on how exhaustively the population is examined or screened. Often, the most unambiguous endpoint is mortality but this endpoint is not as useful for thyroid cancer since most persons who develop thyroid cancer do not die from this disease. The most informative data for risk estimation are obtained from studies of children exposed to external radiation. These include: the Atomic-Bomb Survivors Study Rochester Thymus Study Israel Tinea Capitis Study

8 8 / EXECUTIVE SUMMARY Chicago Head and Neck Irradiation Study Boston Lymphoid Hyperplasia Study Childhood Cancer Survivor Study (United States, United Kingdom, and Canada) the Swedish Skin Hemangioma Studies (Stockholm and Gothenburg) The collective results indicate that external radiation can increase the risk of thyroid cancer; with age at the time of exposure being the most important modifying factor (i.e., children, especially under age 5 y, and adolescents are much more sensitive than adults). The effects of modifying factors (e.g., gender, ethnicity, and attained age) are less certain. Most epidemiological studies of thyroid cancer incidence following internal radiation exposure (primarily 131 I) have been less informative due to the small numbers of exposed children and adolescents. These studies are grouped within one of three types: 1. medical use of 131 I for diagnostic purposes; 2. medical use of 131 I for therapeutic purposes; and 3. environmental 131 I contamination studies. Within the first group are the: Swedish and German Diagnostic 131 I Studies; and U.S. Food and Drug Administration (FDA) Childhood Diagnostic Study. The number of thyroid cancers is small within this group despite the substantial doses received. Nearly all patients were administered 131 I during the second decade of their lives (i.e., ages 10 to 19 y), by which age the risk of radiation exposure is smaller. Within the second group are: Swedish Hyperthyroid Study; U.S. Cooperative Thyrotoxicosis Therapy Follow-Up Study; and British Hyperthyroid Study. Each of these studies has attending complications and limited utility for assessing risk. Most patients were treated as adults and for much of the collective data it appears that therapeutic application of 131 I is a safe therapy for hyperthyroidism.

9 SYNOPSIS OF THIS REPORT / 9 Within the third group are studies of: children in Nevada and Arizona who were exposed to fallout from NTS; population in the United States and Scandinavia exposed to fallout from atmospheric nuclear weapons testing; Marshall Islanders; population living downwind from the Semipalatinsk Nuclear Test Site; population living around the Hanford Site; civilian population living around Chernobyl; and cleanup workers mitigating the effects of the Chernobyl nuclear reactor accident. Unlike medical exposures, these populations were exposed to a mixture of fission products, including radioiodines with a short half-life (e.g., 133 I) as well as 131 I. For the first five groups, there were only marginal suggestions of an association between dose and thyroid cancers. For example, for the children in Nevada and Arizona who were exposed to fallout from NTS there was no statistically-significant increase in thyroid cancers, but with an analysis of combined benign and malignant thyroid tumors, a significantly increased risk was observed. For the U.S. population exposed to radioactive fallout from the atmospheric nuclear weapons testing, the only group that had a slightly increased risk was children who were 0 to 1 y at the beginning of the period of exposure (1951 to 1961). Due to small numbers and complex dosimetry, studies of the Marshall Islanders have not been very informative about the risk of thyroid cancers following exposure to 131 I. The Semipalatinsk Nuclear Test Site was used for 118 atmospheric nuclear tests between 1949 and Within the local surrounding population the prevalence of thyroid cancers was greater in women than in men, but the prevalence of thyroid cancer in the exposed group was not increased relative to that of the unexposed group. No increases in any thyroid diseases were found in studies of children exposed due to releases of 131 I at the Hanford Site. The Chernobyl nuclear reactor accident (April 1986) released a large amount (1.8 EBq) of 131 I, which resulted in the exposure of a large population (in utero fetuses to neonates, adolescents and adults) primarily through the pasture-cow-milk-human pathway. In addition, there was widespread contamination from other radionuclides, principally 137 Cs. The first reports of increases in thyroid cancer risk in children were published in 1992, only 6 y after the

10 10 / EXECUTIVE SUMMARY accident. These first reports were initially greeted with skepticism because of the short latency period and the widely held belief that 131 I was considerably less effective than external radiation exposure for causing thyroid cancer. Since these early reports, there have been comprehensive ongoing efforts to improve individual thyroid dose estimates and to follow the exposed population to determine the effects of the exposure. Twenty years after the accident, there is convincing evidence for an association between radioactive iodine exposure following childhood exposures and thyroid cancer, but risk estimates remain uncertain and the effects of modifying factors such as the amount of stable iodine in the diet need to be better understood. Birth cohort studies revealed a large increase in thyroid cancer incidence after the accident in young Ukrainian children exposed to the fallout from Chernobyl. In Belarus, 1,342 adult and seven childhood thyroid cancers were reported in the 10 y period before the Chernobyl nuclear reactor accident, whereas 4,006 adult and 508 childhood thyroid cancers were reported during the 9 y period after the accident. Long-term follow-up is needed to determine how thyroid cancer risk changes as a function of TSE. In addition to the civilian population exposures to the fallout from the Chernobyl nuclear reactor accident, analyses are under way on occupational exposures associated with its cleanup. Hundreds of thousands of civilian workers, military servicemen, scientists, and medical staff from the former Soviet Union were involved in entombing the damaged reactor and cleaning up the contaminated environment. Surveillance has included thyroid cancer incidence and mortality among this cohort of workers. In contrast to the civilian population exposures, where the major source of radiation was ingested 131 I, the cleanup workforce was mainly exposed to external radiation from gamma-ray-emitting radionuclides. There is large uncertainty with regard to individual dosimetry, but some attempt was made to control the dose limit to workers, which decreased with time (years) after the accident. The present findings, through 2006, suggest a nonsignificant trend toward increased thyroid cancers within this adult cohort of workers. It is presently unclear to what extent internal 131 I exposure contributed to the findings. Additional follow-up may clarify this complicated issue. Section 5 deals with radiation risk for thyroid neoplasms. This section begins with elaboration on the various ways risk can be measured, with emphasis on two approaches, the excess relative risk (ERR) model and the excess absolute risk (EAR) model. Both models are empirically based. The ERR model expresses excess

11 SYNOPSIS OF THIS REPORT / 11 cancer risk as being proportional to the underlying baseline rate, and is sometimes called the multiplicative model. The EAR model expresses excess cancer risk as being independent of the baseline cancer rate and that the excess cancers are simply added to the baseline cancers. The EAR model is sometimes called the additive model. Both models can have variations to account for gender, age at exposure, attained age, and TSE. The number of thyroid cancers predicted by various models is compared. Section 5 closes with a discussion of the risk of developing benign thyroid nodules following radiation exposure. Due to methodological differences, it is not possible to combine the results of different studies so tabulations of the main studies of radiation and benign thyroid nodule incidence or prevalence are presented. These 10 different studies, derived from radiation treatment of different disorders not associated with the thyroid but for which the thyroid might be expected to have had some inadvertent radiation exposure (e.g., tinea capitis, lymphoid hyperplasia), collectively show an association between radiation (dose) and risk of thyroid nodules, either as ERR or EAR. In a few instances, the 95 % confidence intervals (CIs) do not exclude one (which means the effect is not statistically significant and chance cannot be excluded as an explanation) but the overall results suggest increased risk with radiation exposure. Section 6 concerns medical follow-up of persons exposed to ionizing radiation and deals with the subsequent detection and treatment of nodular thyroid disease, both benign or malignant. These outcomes are the primary long-term sequelae of ionizing radiation of the thyroid. This section reviews briefly the significant changes that have occurred over the past 30 y, from the 1975 National Cancer Institute (NCI) workshop on Late Effects of Irradiation to the Head and Neck in Infancy and Childhood to the 1999 Institute of Medicine (NAS/IOM, 1999) report dealing with fallout and its potential consequences for thyroid disease. An evidence-based approach was used by the IOM committee. The major recommendation was that there should not be any public program or clinical policies promoting or encouraging routine screening for thyroid cancer in asymptomatic people possibly exposed to radioactive iodine from fallout of the much earlier NTS tests (1950s). The IOM committee recognized that thyroid cancer was rare in the general population, that exposure to 131 I during childhood appears to increase the risk of thyroid cancer, that it would be difficult (but not impossible) to estimate individual levels of internal 131 I body burdens, and that there was no evidence that early detection of thyroid cancer through screening programs (as opposed to routine clinical

12 12 / EXECUTIVE SUMMARY practice) improves health outcomes or has benefits that significantly outweigh risks. An informative pamphlet is available from NCI (2008). Synopsis of this Report s Conclusions and Recommendations Conclusions The conclusions of this NCRP Report differ significantly from those of the earlier NCRP (1985a) report. Major sources of new data have been published since 1985 that have resulted in a reevaluation of the risk models for thyroid cancer following radiation exposure. A pooled analysis (Ron et al., 1995) of studies of thyroid cancer following external radiation exposure was published in This analysis demonstrated a strong inverse relationship between the risk of thyroid cancer and increasing age at the time of radiation exposure and also suggested that a relative risk model was preferred over an absolute risk model. In addition, studies of the large population who were exposed when they were children and adolescents to radioiodines released as a result of the Chernobyl nuclear reactor accident have begun to provide further insight into the effectiveness of radioiodines in causing thyroid cancer. The major differences in conclusions of the NCRP (1985a) report, the current Report, and the NAS/NRC (2006) report are summarized in Table ES.1. For the population at greatest risk (ages 0 to 14 y), the current Report s preferred model predicts a lifetime risk that is up to 1.5 times greater than that in NCRP Report No. 80. For the entire population, the risk is less for the current Report than for the NCRP (1985a) report (Table 5.10). Compared to many other cancers, thyroid cancer is usually treated by surgery (thyroidectomy) and in some cases with the additional use of large doses of 131 I. The mortality from thyroid cancers is low, especially before age 40 y. Screening asymptomatic patients for thyroid cancer is not recommended for two major reasons. First, the prognosis of patients with thyroid cancer is very good with conventional medical monitoring; it is unlikely that much benefit would be derived from a screening program. Second, the prevalence of thyroid nodules is very high and the tests to distinguish thyroid cancer from benign nodules are suboptimal. Because of this, unnecessary surgery (removal of the thyroid gland) will be performed in many patients without thyroid cancer.

13 SYNOPSIS OF THIS REPORT S CONCLUSIONS / 13 Recommendations The recommendations of this NCRP Report are as follows: 1. There remains a need for better information on the relative biological effectiveness (RBE) of 131 I relative to other types of radiation (e.g., x ray, 60 Co) for induction of thyroid cancer. Animal model systems can be used for this effort since the cells of origin of thyroid cancer in humans and animals are the same, doses to the animals can be carefully controlled, as can a variety of other variables such as age, gender, diet, and genomics. There should be consideration given to the fact that high doses of ionizing radiation can kill cells and, thus, result in an underestimation of the carcinogenic effects of the exposure at lower doses. 2. Thyroid genomics is a relatively young but rapidly emerging, important field. Studies are needed of individuals with and without thyroid disease, and who had or did not have a significant thyroid radiation dose. Certain geneticallyengineered strains of mice for thyroid cancer may be useful in pursuit of Recommendation No The extensive analyses underway of the Chernobyl nuclear reactor accident should continue since there is a large cohort of individuals of all ages exposed to large internal doses of 131 I. This population provides an opportunity to study life-time risks for radiation-induced thyroid cancer from such exposures. 4. The oncogenesis of thyroid cancer needs further elucidation. The generally accepted assumption is that tissue with high cell turnover (i.e., proliferating) is more susceptible to radiation-induced effects than cells with low to no cell turnover rates. Although this assumption offers an explanation for why children are more susceptible to radiationinduced thyroid cancer than adults, the pathophysiologic mechanisms need further investigation. 5. There is a need for a better understanding of modifying factors associated with radiation-induced thyroid cancer. Age at the time of exposure, and the amount of dietary iodine have been clearly identified as important factors in the etiology of thyroid cancer. Additional information is needed about other factors that could influence the development of radiation-induced thyroid cancer, including diet, genomics, attained age, gender, and ethnicity. The effect of intensity of screening also requires further study. There is also a need to investigate the effects of varying degrees of

14 14 / EXECUTIVE SUMMARY TABLE ES.1 Tabulated comparative conclusions of NCRP Report No. 80 (NCRP, 1985a), the current NCRP Report, and the NAS/NRC (2006) report. NCRP Report No. 80 (NCRP, 1985a) This Report BEIR VII (NAS/NRC, 2006) Preferred Model Absolute risk ERR ERR Effect of Age at Time of Exposure Risk for children (<18 y) is twice the risk of adults. Strong inverse relationship between the risk of thyroid cancer and the age at the time of radiation exposure for ages <20 y. Small risk over the age of 20 y. Little if any risk over the age of 40 y. Strong inverse relationship between the risk of thyroid cancer and the age at the time of radiation exposure for ages <20 y. Small risk over the age of 20 y. Little if any risk over the age of 30 y. Effect of TSE Constant absolute risk ERR decreases considerably (40 % by 40 y) Constant ERR Gender Absolute risk for women is twice the absolute risk for men. ERR for women is equal to the relative risk for men. Relative risk for women is twice the relative risk for men.

15 SYNOPSIS OF THIS REPORT S CONCLUSIONS / 15 Internal Exposure 131 I and 125 I are no more than one-third as effective as external radiation in causing thyroid cancer. 135 I, 133 I, 132 I, 123 I, and 99m Tc are as effective as external radiation in causing thyroid cancer. All radioiodines are likely to be between 60 to 100 % as effective as external radiation in causing thyroid cancer. The absolute risk for women is about twice that for men because of the former s higher normal incidence level (i.e., EAR for women is about twice that for men, but they share a common ERR). The amount of stable iodine in the diet may be an important modifier of risk (in addition to its important effect on thyroid dose). Quantitative relationship between dose from 131 I and the development of thyroid neoplasia remains uncertain.

16 16 / EXECUTIVE SUMMARY bias in the reconstructed doses on the analysis of statistical power and the slope and confidence intervals of the dose-response relationship in an epidemiological study. This Report s specific and detailed conclusions and recommendations are presented in Section 7.