Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Health Effects of }~3] Administration in Humans Small amounts of radioactive materials are often used for both observation and diagnosis of body systems and treatment of disease (see Box 2. l). The radioactive tracer iodine 13 ~ (T~3~' has been used to evaluate thyroid function in humans since the 1940s. In fact, the field of nuclear medicine developed primarily from the successful use of Ii3i as a probe to evaluate thyroid function. In the five decades since the introduction of Ii3i, thousands of radio- pharmaceuticals have been developed to evaluate function and disease in almost every tissue of the body, and millions of radioiodide tests have been performed on Americans. Today, over 10 million nuclear medicine examinations are performed annually in the United States (NCRP, 1989~. }~3] was the only radioactive tracer readily available for use in the 1950s, when the AAL thyroid study was conducted. However, serious shortcomings lunit its use in modern nuclear medicine. The major drawback of }~3} iS the relatively high radiation dose received by Me thyroid. Assuming an uptake of 25 percent of the administered radiation activity, the thyroid receives a dose of approx~nately I.3 red per microcurie~ of administered radioactivity. Another disadvantage is that principal gamma ray emissions from the radioisotope are not efficient for Unaging purposes; the 364-keV photon emitted is higher than optional for gamma scintillation cameras currently used in nuclear medicine. Today, the radioisotopes technicium 99m and }~23 are preferred in thyroid imaging procedures except when diagnosing thyroid cancer, in which case other nonradiogenic (ultrasound, palpation, or thin-needle aspiration) means may yield useful information. These radionuclides have short half-lives and decay emissions, which reduce the radiation dose to the thyroid. Their gamma energies (in the range of 150 keV) are also optimum for imaging studies. 26 body. 1See Boxes 2.1 and 2.2 for background on radioactivity and how radiation affects the human
Health Elects of i/3] Administration in Humans RADIOLOGICAL BACKGROUND OF THE AAL STUDY 27 The use of radioactive materials at the AAL for arsenal and human research studies was granted by license from the U.S. Atomic Energy Commission (AEC).2 The license allowed the use of }Eli in humans to determine uptake of iodide in human subjects before and after cold exposure. The standard procedure involved administration of I'3i in approximately 50 microcurie doses. Some subjects received multiple doses at one- to six-month intervals. The AEC license required that the investigator be properly trained in radioisotope usage and methods in order to procure, use, and dispose of radioactive materials in research studies. Dr. RodahI, the principal physician and investigator on the AAL project, had at the time of the experiments completed the required training, which involved 30 hours of participation in an instruction program and experience diagnosing and treating 49 cases using radioactive materials. Dr. Rodahl's training in the principles of radioactivity, radiological health safety, radiation measurement techniques, and biological effects was completed at the Cook County Graduate School of Medicine in Chicago. The AEC license provided for the human administration of }~3} in the form of capsules, which were to be preassayed by the supplier. The radioactive material for the AAL thyroid study was obtained from Oak Ridge National Laboratory in Tennessee, considered the most reliable supplier available. Each capsule contained approximately 50 microcuries of }~3~. Because of the relatively short half-life of li3] (eight days) and long travel the involved, the supplier had to provide capsules with higher activity than this to account for radioactive decay during transportation of the material from Oak Ridge to the AAL in Fairbanks. For instance, if the transit time was one week from the tune of manufacture at Oak Ridge to use in human subjects, the Oak Ridge facility would have had to formulate Ii3i capsules with an activity of 92 microcuries for the capsules to have 50 microcuries at the tune of administration. In some cases, the dose administered was recorded as 65 microcuries, which may have been a result of early delivery of the capsules. OVERVIEW OF EPIDEMIOLOGICAL EVIDENCE REGARDING RADIATION-INDUCED THYROID CANCER Since 1990, there have been several comprehensive reviews of the epidemiological literature concerning radiation induction of thyroid cancer in humans (NAS, 1990; Shore, 1992; Ron et al., 1 9951. Studies can be categorized into the following groups: (1) patients with various medical conditions (including benign disease and cancer) who received external beam therapy (x-rays) in which the thyroid was incidentally exposed, (2) patients given }~3} for medical and (3) individuals exposed to nuclear radiation and/or fallout diagnostic or therapeutic reasons 2AEC licenses 46-SO-1 and 33252, dated May 20, 1955. 1
28 The ALL Thyroid Function Study Box 2~..~. ~ - .: i. ~ . ~ . . ~ ~ .. ~ ,~ .... ~ ~ ~ ~ ~ ~ ~ . ~ ~ Aft. ~ .: .~ ~ ~ ~ ~ ~ ~ ~ ~ : TIC:.... - : ...: ENDS ~:D:: ''''it ~ He my - S- ~1~ ~ -:- ~d~ ~ : ..~. ~ -. ~ ~ ~ ~ 1 ~;~ --- --it ~-~Io~pes~-~m~:-~ate~-~ ails . ~ . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ :~ ~ ~ I ~ ~ ~.~ :~ ~ Pa- ~ ramots~s--I ~ === :: p=iciilar Oar HI jig ~ ~ .~..~..~ ~.~-.-Is~--~ab~r.bed -~-~t=~..~ora~ ::. ~s -God ~ ~-~ ~ if: ~ . ~ ~ . ~ iC: . ~ ~ i ~ . ~ . ~ ~ , (; ~ ~ ~ . ~ hi ... ..: .. . ~ .. ....~... ~ . .... .... , . .: . . . .. ... ~ ~ . .... . ~ ~a 1w .. ~ .~ -I ~ ., 1~ process lurougu Issue, ~ioru ~°=Mostoftbeab~ ........... ,., - ~ ..... , i ~ i , . g . .... ... .. - ~ ~ ~ . ~ ... . ~ ~ '~L~ 'a ~ ~S "me i.. - ....: . ,...- .. i- . - ~ I.. . ~ ~ ~ -I ~ ~ ~ A. ... .... : Id. :As~ . ~ -be ~..s , , it. .,- ... - ~ w~....~m . ~, ~-- A i , A. - ..~. ~ ~ .- . ~. ~ ~...ra~e .ma~ ,, ;- -- -- ~ --- .-..-' ':-- ~ . " i. .-e~...-~-.~....pam£ , , ........ ~ ... - .. -. .. ~ ~s:-.=lls:::. a. . ~- ~; ~ ~ ~ ::.~ ~ : ~patt~. . Pa I ; . ~ .............. p~ce~in Iowa - :~ ...mQle~e.~..~....~e ~ .~ ~ . . ~ ~ . ~.~. ~i; Is of ~ ~, i ~ It-- C~ei ' C""'-"'' :':''''- .- -Gil ..; ..-. .~ ~. it. ... E~'~..'..~-.'.~6 i. ~ ail - .. .. ~ .. ~ . i. . ~ ~ . . ~ ~ hi. ..... ~ .-. ~ ~ .~ ~ . ~ ~ . ~;~* ~ ~ ~. ;~ ~:aO1...Ity -~.~-.-reD:aU~-~t
Health Elects of ·~3] Administration in Humans 29 : ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~- .- I, ~ ~ ~ ~ , ~ . i: :~ ~ I. -~ .......... . - . .~ ~ ~ ~. ~ ~ : ~ ~:liv~-:dsme~i~, =~sen~ ho if- :~as~:~ ~ ~:~:~: I :~ ~ a~=tl~of: ~- ~; ~fit ~ Id: ~ radiation (including radioactive isotopes of iodine) from nuclear weapons.3 A number of generalizations about radiation-induced thyroid cancer may be drawn from these studies, and these conclusions can help provide a reasonable assessment of the radiation risk for thyroid cancer in the AAL study. 1. There are major differences in background thyroid cancer rates among the various cohorts studied, perhaps related to differences in lifestyle. The development of thyroid cancer from initiated cells (cells which have received some type of carcinogenic stimulus) is greatly dependent upon hormone status. 2. Females are two to three times more susceptible to radiogenic and nonradiogenic (i.e., background) thyroid cancer than males. The relative risks do not differ between the sexes. 3. Thyroid cancer risk from external irradiation has a significant age dependency. Risk is particularly high among children exposed within the first five years of life. There is little risk in Exposure to nuclear radiation or fallout from nuclear weapons brings substantially greater risks than incidental, diagnostic, or therapeutic exposure but the former does offer lessons of relevance.
30 The ALL Thyroic! Function Study individuals exposed after ages IS-20 years. The risk in adults (greater than age 15 years at time of exposure) may be one-eighth to one-tenth the risk in children. 4. Current epidemiological evidence on thyroid cancer induced by external irradiation is consistent with a linear dose-response curve. In children, there is convincing evidence for risk at about
Health Elects of {3] Administration in Humans 31 10 red. (The epidemiological data can be fit by a relative risk model because of the strong dependence of risk on the natural incidence of thyroid cancer in the population under study. However, an excess risk model fits the data best and includes allowances for cohort effects, latency, age at exposure, and sex.) 5. {~31 iS estimated to be 20-25 percent as effective as externally administered x-rays in producing thyroid cancer. Although the reason for this low carcinogenic potential is not well understood, it may be related to the facts that {~3} iS concentrated in the colloid, reducing the dose to the follicular cells, and that the dose rate from {~3] radiation is reduced compared with external radiation exposures. Spreading the dose over time may reduce the risk because of cellular repair of radiation injury. 6. There is a characteristic time lag (referred to as the "latent period") between external radiation exposure and clinical appearance of thyroid cancer. The minimum latent period appears to be about five years. Risk remains elevated for 30-40 years after exposure. Table 2. ~ identifies several major studies that provide risk estimates of radiation-induced thyroid cancer (Ron et al., 19951. All the studies cited involved external irradiation of subjects as a result of atomic bomb irradiation or radiation therapy for the treatment of various benign diseases or cancer (other than thyroid cancer), rather than diagnostic levels, but information from these studies can be helpful in understanding thyroid cancer risk generally. Brief descriptions of these studies follow; more detailed information is available from the comprehensive reviews by Shore (1992) and Ron et al. (19951. The largest single study of thyroid cancer risk involved Japanese survivors of the atomic bombings in 1945 (Ron et al., 19951. In this study, thyroid cancer incidence diagnosed between 1958 and 1985 was determined among 79,972 atomic bomb survivors. The average thyroid dose was 27 red (range: 1-399 red). This is the only study that includes people of both sexes who were exposed at all ages. Radiation-induced thyroid cancer has been documented in children irradiated for enlarged thymus in a Rochester, New York study begun in the early 1950s (Shore, 1992; Ron et al., 1995~. The study included 2,856 subjects treated between 1926 and 1957 and all 5,053 nonexposed available siblings. All patients were exposed before the age of ~ year. Average dose to the thyroid was 136 red (range: 3-1,100 red). Radiation therapy has been used to treat ringworm of the scalp (tinea capitis), leading to incidental irradiation of the thyroid gland. One relevant Israeli study included 10,834 persons treated in this manner between 1948 and 1960, plus 10,834 disease-free, nonuradiated matched comparison subjects and 5,392 disease-free, nonirTadiated siblings. All patients were treated before the age of 16 years. The average thyroid dose was estimated to be 9 red (range: 4-50 red). The risk of thyroid cancer in the Israeli study was determined to be substantially higher than in the other studies listed in Table 2. 1. Whether the higher coefficient is due to statistical fluctuations, to unusual susceptibility within this population, or to other factors is unclear (Shore, 1992~. Thyroid cancer has been observed in children irradiated for benign head and neck conditions (primarily enlarged tonsils and adenoids). In one study, more than 5,000 patients received head and neck radiotherapy in Chicago between 1939 and 1962. Adequate medical follow-up and dose information could be obtained for 2,634 subjects. Thyroid gland doses ranged from ~ red to 580 red, with a mean dose of 59 red. In another study from Boston, thyroid cancer incidence was compared in 1,590 irradiated children treated between 1938 and 1969 and i,499 children who were treated by surgery only for removal of enlarged tonsils or
32 The CAL Thyroid Function Study adenoids. The mean estimated thyroid dose was 24 red (range: 3-55 red). Thyroid cancer has also been observed in individuals who have been given radiation treatments for other cancers. Two studies provide limited thyroid cancer risk information. One study included 9,170 childhood cancer patients who survived two or more years. The average thyroid dose was substantial 1,250 red (range: 100-7,600 red). In a large international cohort study involving 150,000 cervical cancer patients, 43 women were identified who developed thyroid cancer at least five years after their diagnosis for cervical cancer. Eighty-one controls were matched individually to these cases. The mean thyroid doses was ~ ~ red (range: i-24 red). This is one of only a few studies which documents radiation-induced thyroid cancer in adults. Combined, the studies discussed above include about 120,000 people, nearly 700 thyroid cancers, and about 3 million person-years of follow-up (Ron et al., 19951. Except for the atomic bomb survivor and the cervical cancer patient studies, which include adult subjects, risk estimates provided in Table 2.! are based primarily on observations made in children. For childhood exposures to external radiation, the pooled absolute risk is 4.4 per million person- years per red (Ron et al , 19951 Although our understanding of thyroid cancer risk is based primarily on studies of children exposed to external radiation, there are several epidemiolocical studies that have ~r~l~A +~ ~-~-~^ By- I1~1 T^~1~ ~ ~ 1;~^ ^~-r~l ~;~ ~1~ ~1 ~ ~ _~ Y111 ~ CA I JlL )I CLI I I IC Cl I CL_I . - ~ ) I I I ~ I Jl~ A. ~ I 1.~1.~ .~=V~I ~ I ~1 l 11 1~1 1 1 11 Ill )~ 11 -A I `1 1 1( 1 1~ ~11 1 - - - ~Y~(~I fir - ~ 1 __ __. Ace_ ___ - ~ ~ ~ · ~ EVER ~ - ~ $~= ~-~ ~-^~^ ~AlAlVlV51~1 OL"~10O ~1 1 If children and adults and estimates for thyroid cancer risk. Shore (1992) provides detailed discussions of these studies in the context of an overall review and analysis of radiation-induced thyroid cancer epidemiology. Comparison of risk estunates in Tables 2. ~ and 2.2 indicate that exposure to }~3] iS not as detrimental as external radiation exposure. A few epidemiological studies of diagnostic }~3} have been conducted (the Swedish diagnostic study, the Food and Drug Administration (FDA) study, and the German diagnostic study). These studies are particularly relevant to the AAL study because they involved diagnostic levels of {~3] in which radiation doses to the thyroid are comparable (Table 2.2~. None of these studies provide clear evidence of excess thyroid cancer. The largest of these investigations was conducted by Holm and his colleagues (1988), who studied retrospectively some 35,000 patients given diagnostic doses of }~31 in Sweden between 1951 and 1969, and followed up for an average period of 20 years. The average activity administered was 52 microcuries of ]~3i, levels similar to those used in the AAL study. The mean age at the time of }~31 use was 44 years; 5 percent of the subjects were under age 20. Increased risk of thyroid cancer was observed only in patients originally examined because a thyroid tumor was suspected. Patients given li31 for reasons other than a suspected tumor were not found to be at increased risk. There was no significant excess of thyroid cancer in this study, which had good statistical power to detect effects in adults (Shore, 1992~. Shore (1992) noted that the carcinogenic potential of li3i beta particles may be as low as one-fifth that of external x-rays or gamma rays.
Health Elects of 7131 Administration in Humans TABLE 2. ~ . Risk of thyroid cancer after external radiation exposure 33 Study Population Excess Absolute Riska (Io-6 PY-rad) Excess Relative Risk (per 100 red) Exposure at ~ 15 years old Thymus A-bomb T· · ~ Plea capltls Tonsil Childhood cancer Exposure at > 15 years old Cervical cancer A-bomb 2.6 (~.7, 3.61i 2.7 (~.2, 4.6) 7.6 (2.7, ~ 3.0) ~.u I., 1/.l' ND ND 0.4 (-0. 1, 1 .4) 9.1 (3.6, 28.8) 4.7 (1.7, 10.9) 32.5 (14.0, 57.1) 2.5 (0.6, 26.0) 1.1 (0.4, 29.4) 34.9 (-2.2, 00) 0.4 (-0.1, 1.2) SOURCE: Ron et al. (1995) a9S percent confidence Innits in parentheses. ND = not determined. CALCULATIONS OF RADIATION RISK Risks of radiation-induced thyroid cancer have been estimated in Table 2.3. The risk (probability of radiation-induced thyroid cancer) is the product of the radiation dose to the thyroid and the absolute risk coefficient (excess number of cancers per million persons per red). For each population group, the average thyroid dose was determined by multiplying the average activity administered by a dose conversion factor. The average activity noted for each population group was derived from the AAL reports; the dose conversion factors (which convert the amount of radioactivity administered into a thyroid radiation dose) were based on calculations provided by Oak Ridge Institute for Science and Education based on current computer models of internal radiation dosunetry of the thyroid gland for elm. An absolute risk coefficient of 4.4 x lo-6 excess thyroid cancers per red per year at risk was used (Ron et al., 1995) as the basis for determining the risk coefficients shown in Table 2.3. This risk is based on the pooled analysis by Ron et al. (1995) of epidemiological studies. This risk estimate is based on population studies of children exposed to external beam radiotherapy and Japanese survivors of the atomic bombings. The thyroid gland is especially vulnerable to the carcinogenic effects of radiation in children. Little risk is apparent in individuals exposed after age 20 years (Ron et al., 1995~. Since the Alaska Native and white military personnel
34 o c x i_ o v ct v ._ o sit cat o cat . _ . ED The CAL Thyroid Function Study ^ cat '= ~ cat ~ cat _ ct 0 .= ~ ~ c r ~ Ct O ~ -= O Cal en X ~ O Sit ~ .= 7 ~ ~ i_ Cal A: o Cal O ~ Cal X ~ A: o . _ - C o v) ~o ~o ~ ~o ~ c~)-) ~o ~o vooo voo c~r)Q-1 ·.. ~ r ~- o ~oo voo ~o oo ~l" vvoo ooo ---- v c ~O~ 1 ~(~ ··· ~ OOO' 1 00' ~ oooo ~o') C~) 1 o ~ ~ l o ~ o ~e~ ~oo ~ ~ l h ~oo -t' ~ ~, 1 ~ ~ o o o o A ~A Cq :: ~- ~· _ o v O ~o 0 ,= ~o o = CTS ~ - ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ' 1 ~ ~ ~ ~ 3 ~ ~ ~ ,~' ~, ~ 3 ~ ;: ~ ~ u, c~ - - c 52: ._ c~ ._ c~ ~; ._ _ v ~: ~_ 0 c) v ~: v G o CSx Ct . . P~ o V)
35 ~ - u, . - p~ ~ o At pi ~ - . ~ ~ a x o ~ ~ .~ · - · - ~ G - o On 50 Ct ._ - .s - ._ 3 a: Ct U) ._ a Ct Cal Ct o ._ Cd ._ Cal U: ._ so so A: Ct Cal ._ ; - Lo m ._ O 0 ~ C) c .° Ct 8 o 8 o o o o o o o^ ~ - ^ ~ ~ 8 ° ~ 8 ° of ~Cal ~oo ~oo ~ Ct ~_ D A: o 50 U. _` .= ~ EM O~ ._._ ~ O o V ~ V C~ 0 Ct . _ ~ _~ ·U) V ~ ~ 0 Ct ~.V ;> ~ ~ o C .o C~ o oo ~ oo ~ ao . . . . . ~ ~ ~ ~ U~ 0 . . . ~ oo oo ~ oo . . . . . c~ oo oo oo . . . . . . oo ~ ~ ~ ~ V~ oo 0 0 oo ~ ~ ~ oo C~ ~o `.D o 0N C~ c ~_ cn C ~ce ~ ~ C 5 e ~ ~ ~ u 0 -~0 ._ U: - ~4 cq - Ct ~4 0 _ ~ Ct Ct C~ Ct ~: ._ C~ - ._ - C~ - ._ o ~: ~ ~: U: ~' ~ ~ S~ ~ ~ ._ ._ _ Ce Z ~ ~ _ ~ Ct Ct ·- 0 ~ ~ C~ ~3 Ct Ct C~ c ._ ') . ~ 00 0 C~ .= ~ Z; . - U~ C~ - o _ Ct I_ - 0 o cq pt . _ ^D ;> ~ O 0 a~ _ _ C) O V) U) o . - ct - c) c ~r o V) ~o - C~ D C~
36 The GAL Thyroic! Function Study subjects in the AAL study were adults and were administered }Eli, the thyroid risk coefficient of 4.4 x lo-6 was modified to account for these factors. The following factors were used to arrive at the risk coefficients listed in Table 2.3: a dose effectiveness reduction factor of one- fourth for I'3i (Shore, 1992); a sex factor of two-thirds for males and four-thirds for females (NCRP' 1985; Ron et al., 1995); an age correction factor of one-eighth for adults (Shore' 1992); and 30 years at risk following exposure (NCRP, 1985). RISK ESTIMATES FOR THE AAL STUDY The AAL conducted four types of studies to measure (1) thyroid uptake, in which the percentage of }~31 appearing in the thyroid gland was measured at various times after ]~31 administration, (2) iodine metabolism, in which urinary and salivary excretion rates and blood levels of l}3} were measured at various times after }~3} administration, (3) effects of exogenous potassium iodide treatment on thyroid uptake of }Eli, and (4) iodine metabolism in white military personnel before and after a four-week cold exposure in the field. Inspection of Table 2.3 reveals a wide range of thyroid doses and thyroid cancer risk estimates in the population groups participating in the AAL studies. Doses and risk estimates varied by more than a factor of 10. The lowest doses and risks were seen in Wainwright, Point Lay, Fort Yukon, and Point Hope Alaska Natives. Sixty-eight subjects from these communities participated in single studies and received a single administration of }~3~. Twenty-two other Alaska Native subjects participated in two studies and received two separate list administrations. All 19 white servicemen participated in both the thyroid uptake and thyroid metabolic studies. Each serviceman received two doses of llama. A group of 12 Ana~vuk Pass Eskimos and Arctic Village Indians participated in three studies and received three separate }~31 doses. Anal Pass Eskimos and Arctic Village Indians had the highest thyroid doses and calculated thyroid cancer risks. For the subjects given multiple }~3] administrations, all individual doses administered were added together and the total dose was used to estimate thyroid cancer risk. This assumes that there is no reduction in thyroid cancer risk due to protraction of the dose. The AAL Technical Report (Rodahl and Bang, 1957) identified two Wainwright subjects and one Arctic Village subject who were nursing children at the time of their participation. Since radioactive iodine can be passed on to children through the mother's milk, the childrens' thyroids may have been at risk. Thyroid activity in the nursing children was not measured directly in the AAL study. The AAL report indicated that 33 percent of the dose administered to the mother would be expected to appear in the mother's milk, and that uptake in the child's thyroid would be 33 percent of the ingested activity; thus the activity in children may be estimated to be approximately one-ninth of the activity administered to the mother. Based on these assumptions, an average thyroid activity in the nursing children of 12 red was estimated. Because the resulting thyroid dose to these children is small, the thyroid cancer risk (about I/2,000) is low. Two young Alaska Natives participated in the AAL studies. One was a 16-year-old male Point Lay Inupiat and the other a female 17-year-old Arctic Village Athabaskan Indian. For the purposes of calculating risk, these subjects were considered adults since they were older than age 15 years (Ron et al., 1995) and thus physiologically closer to adults than children.
Health Effects of i73~ Administration in Humans 37 A 30-year-old Arctic Village Athabaskan Indian may have been pregnant at the time of her participation-in the 1:~3i studies, but the Committee has no way to verify this, (the possibility was raised by recent observers calculating backwards from the age of the participant's child, a relatively unreliable technique given the variability of pregnancy lengths). The embryo/fetus is particularly radiosensitive. The nature of the biological effects to be expected from an exposure during pregnancy depends on the magnitude of the dose to the embryo/fetus and when during gestation exposure occurs. If the participant was pregnant at the time she received a single dose of 50 microcuries in October 1955, the dose to the embryo is estimated to be 0.05 red based on exposure to the embryo/fetus from accumulation of radioiodide in the mother's urinary bladder. The embryo/fetus would receive a negligible radiation dose from the mother's thyroid. Given the uncertainty, it is also unknown when this exposure might have occurred with respect to gestational age. However, because of the small dose, the risk of developmental abnormalities and other untoward pregnancy outcomes is small. SIGNIFICANCE OF CALCULATED RISKS OF RADIATION-INDUCED THYROID CANCER The last column of Table 2.3 provides lifetime estimates of the probability of thyroid cancer as a result of Il31 administration. To put these risks into perspective, it is useful to consider the natural incidence of thyroid cancer in the population and Me lifetime risk of thyroid cancer in the absence of radiation exposure. Thyroid cancer is a rare form of cancer (American Cancer Society, Inc., 19951. The American Cancer Society estimated there would be approximately 14,000 new cases of thyroid cancer in the United States in 1995. BY comparison. 1 o ~are _ _ _ ~ I_ _ ~_ 1 ~7^ Pro _ _ _ r ~ J - ----or A- ~~ ~ ~~~ 15;~UUO new cases of Breast cancer, 1/U,UUU new cases or lung cancer, and 138,000new cases of colon-rectum cancers are estimated to occur during 1995 (American Cancer Society, Inc., 1995). Assuming Cat thyroid cancer is un~fonnly distributed in a population of 260 million, the annual risk of thyroid cancer is about 5 cases per 100,000 population. Further assuming a 40- year period of risk (NCRP, 1985), the total lifetime background thyroid cancer risk would be 200 per 100,000 or ~ in 500.4 The weighted average risk among the various populations 4New information from the Alaska Native Tumor Registry for the period 1969 - 1988 (Lander et al., 1994) has demonstrated that Indian women have a higher incidence of thyroid cancer than the U.S. population as a whole, whereas the rates for Inuit men are lower, and the rates for Inuit women and Indian men are the same as for the U.S. population as a whole. Based on this report (ibid., p. 14), the annual average incidence of thyroid cancer for Inuit men is only 1 in 100,000 or a lifetime risk of 50 in 100,000 over 50 years, yielding a i-in-2,000 risk. Indian women, based on this report, are much more susceptible to the disease, with an annual average incidence rate of 9.9 per 100,000 (ibid., p. 15) or a lifetime risk of 495 per 100,000 over 50 years, yielding a lifetime risk of approx~nately 1 in 200 (double the U.S. average). However, the total population of Alaska Natives is less than 100,000 and the Committee's review of the data showed that the data differences were not statistically significant in comparison with the overall U.S. population. Although the indicated increased risk of this disease in (continued)
38 The ALL Thyroid Function Study participating in the AAL study (Table 2.3) is about ~ in 3000, a risk six times lower than the background thyroid cancer risk. Thus, participation in the AAL study added a small and insignificant amount to the background thyroid cancer risk. The greatest risks (albeit small) of thyroid cancer were seen in populations given multiple }~31 doses. In particular, He Ana~vuk Pass females and Arctic Village females who received multiple doses have calculated risks of ~ in 800 and ~ in 700, respectively (Table 2.3~. Thyroid cancer risk in these individuals is almost doubled because of the multiple }~31 administrations. However, because thyroid cancer is rare to begin with, the additional radiological risk is not statistically significant. Radiation- induced thyroid cancers caused by the AAL study would not be expected in the Alaska Native or white military personnel experunental subjects. The Committee is unaware of any reports of thyroid cancer in the irradiated population. However, there has not been a systematic medical follow-up to detennine whether any cases of thyroid cancer have appeared in the study subjects. RADIATTON GUIDELINES FOR li3] USAGE THEN AND NOW 1957 Guidelines At the time the AAL study was conducted in the mid-1950s, there were no formal guidelines concerning radiation exposure of research subjects. The AEC did approve the study, primarily based on radiological considerations; no radiation Units for diagnostic tests were in place, and l]31 was the only radioactive material available to conduct the study. Setting aside the problems of informed consent, use of special populations, and methods of subject selection addressed elsewhere in this report, the study was scientifically reasonable by the standards of the time, assuming radiation exposure as the only consideration. The prevailing view within the scientific community during the 1950s was that in order for radiation effects (particularly acute health effects such as reddening of the skin) to occur, the dose must exceed a threshold level. This threshold philosophy Implied that radiation doses below the threshold were safe and did not cause harm. In interviews with Dr. RodahI, it was clear that he accepted this philosophy then and still does now, and that he was convinced that the doses of li31 he used in the ~L thyroid uptake studies in He Inupiats, A~abascan Indians, and white military personnel were below the threshold and therefore perfectly safe. The researchers followed the threshold philosophy in administering multiple doses to some subjects. Although the cumulative effects of such multiple administrations were unknown at the time, the researchers believed that the effects of the first dose disappeared by the tune of the administration of the second dose because the second dose was given several weeks after the first, allowing for complete decay of the li31 in He fast administration. A review of medical literature of the period shows that the same general level of diagnostic dose was believed appropriate, though some researchers had begun using smaller amounts (Rail, 1956, 1957; Clark, 1956~. Athabascan women is a cause for concern, at this time the data are inadequate to warrant a revision in the current analysis.
Health Elects of i131 Administration in Humans 39 In the 1950s, scientists believed that diagnostic doses of list were not associated with cancer risk. Small doses, as used in the AAL study, were believed to be safe because no evidence had emerged that small doses, as typically used in diagnostic studies, were associated w~ `~y~u~u ~. ~ =~ study ~r~ c~ at., 1~' comlrmec the relative safety of diagnostic doses of radioiodide. It should be noted here that even today there are no existing guidelines concerning the need for medical follow-up for persons given diagnostic doses of I13l. However, the use of pregnant a possibly pregnant woman would have been considered inappropriate based on generally accepted practice in the 1950s (Clark, 1956), and lactating women would have been discouraged from breast feeding until the Ii3i cleared their system. Giving list to children was being seriously questioned at the the of the AAL research (Clark, 1956; Rail, 1957) but was not yet recommended against as a standard. The possibility that thyroid cancer could result from radiation exposure was first documented in 1950, but those cases resulted from large doses used for therapeutic purposes. In 1955, Sunpson and colleagues conducted one of the earliest studies linking thyroid cancer in children with external radiation treatment for thymic enlargement (Simpson et al., 19554. At about the same tune that studies of thyroid neoplasms subsequent to x-ray therapy for benign disease were going on, the possibility that incorporation of radiolodide in the thyroid could also be carcinogenic was also explored. Studies of the Marshall Islands inhabitants who were exposed to radioactive fallout (including radioiodide) from a thermonuclear test in 1954 and patients given high doses of radioiorlide for thyroid therapy (e.g., hyperthyroidism) suggested that radioiodide incorporation could also be carcinogenic. The Marshallese were among the most intensely studied of any group subjected to thyroid irradiation and provided the first suggestion of a link between radioiodide incorporation and thyroid cancer (Shore et al., 1986~. Appendix D contains the Air Force research protocols for the use of human volunteers in experunental research as set forth In a 1953 memo. <rail +1~;~ ~ TO TT~1~ ~_, /~1 ~ ~ _1 1 ^~ ret ~ .~ Current Guidelines This Committee was asked to address the risks to the AAL study participants under today's scientific standards. The Committee believes the AAL experiment would not be allowed by an Institutional Review Board (IRB) today because of the radiation dose to the thyroid from the ll3] to be used. The dose est~ates for even group of subjects listed in Table 2.3 exceed the current recommended radiation dosage limits approved by the FDA for research subjects (5 red for a single dose, 15 red as a cumulative dose for a number of studies conducted within one year). While there were various problems with the subject selection process, discussed elsewhere in this report, from a scientific perspective minors, lactating women, and potentially pregnant women would probably be excluded from this type of study today. The use of research subjects with obvious or modest thyroid enlargement would also have been considered a questionable practice. In addition, while the final report of the study stated that the participants were healthy, "normal" individuals, the Committee found that the physicians took the people who came to them and then performed (but not even in all cases) a rudunentary physical examination consisting of measuring height, weight, oral temperature, pulse, and blood pressure.
40 The ALL Thyroid Function Study The study design stated: "Care was taken to exclude individuals who had taken medication or x-ray contrasts- which are known to affect the iodine uptake" (Rodahl and Bang, 1957, p. 61. No Native participants could recall being asked about their physical condition or if they were taking medicines. The Committee did not speak to enough military participants to draw conclusions about their selection. The Committee believes the doctors probably had access to the military participants' medical records to determine their medical history, but such records were unlikely to exist for the Native subjects. (See Box 2.3 for additional discussion of use of Ti3i in special populations.) THE EVOLUTION OF Old llNDERSTANDING OF RADIATION HEALTH EFFECTS It was during the late 1950s that the no-threshold philosophy began to emerge and the importance of chronic health effects such as cancer was recognized. Today, radiation protection and radiological health principles are firmly grounded in the no-threshold philosophy: any dose of radiation is potentially harmful. The specific probability of harm is dependent upon the radiation dose received. Unfortunately, this no-threshold philosophy is often interpreted to mean that no radiation dose is "safe." As exemplified by this study, this interpretation of the concept of threshold is incorrect. Although a particular risk may be attributed to a given radiation dose, the probability of thyroid cancer is so small as not to be measurable in the population. The question of what is a "safe" dose is trans-scientific and requires consideration of social and economic factors in addition to dose-response data. Recently some members of the radiation science community have begun to question the assumptions giving rise to the linear no-threshold philosophy because there are experimental data that can be interpreted as supporting alternative hypotheses. This question will continue to be considered and debated in the future. In 1 950, the International Commission on Radiological Protection (ICRP) established an occupational exposure standard of 0.3 roentgen per week. This dose was based on the concept of tolerance doses related to the development of erythema, or skin reddening. ICRP exposure limits for radiation workers were based on the observation that many workers who had been in contact with radiation for a number of years at these levels had suffered no apparent radiation injury (Hendee, 19931. During the 1950s, information about the Importance of cancer as a delayed effect of radiation began to emerge from studies of the Japanese survivors of the atomic bombings at Hiroshima and Nagasaki in 1945. What emerged from the studies of the Japanese survivors and patients who had received radiation for the treatment of various medical conditions was the belief that any dose of radiation may be harmful and that the major health effect of exposure was cancer. As a consequence, occupational exposure standards were reduced to 5 roentgens per year in 1956. These limits were applicable only to individuals occupationally exposed to radiation. Human subjects exposed to radiation as part of a medical diagnostic or therapeutic procedure or who participated as volunteers in medical research studies were not subject to these limitations.
~ :::~R I~E..OF:~.3 ~ my, ~ ,,;; , ~ ~ , :~ I . , , Aft ..... ............. ~ ~ ~ -011~:111 0~ I= I~ CID~ - -I ~ aught ~ ~ ~ i:4 co~nce It 02ni~s~d-~ =t ... -~nk~.~'ne~th~ nm~.-..~men-~.~. .~.~,~ ~ preglia~ al ble time~0f ~oLlll; In I; .,. ' ; 2 , ' 2 ' 2 ~ ~ ·..2. ..2.P,,.6,,,,=~4d i'~ 'a - ~ ~ ~ of nisi :: :- Id, i10 - i: then .; . . ~.~ -0 ~ :~-' :.::'... ''' "' .~.~.~.~ ~ ~ . ~,, ,,~ ~ ~ ~ ~ u ~ ~ ~ Lw~ :~:11~. 'aft ~ ~ Act: ~ . . ~ . ~ I. ~ .. ~ r~up~:~ 1~4 ~ ~:~3; l~o~odi Beam I.. ~ ~ ~ ~ .. ~ ~ I- . -.~ ~ ~ ~ . ~ ~ ~ ~ ~ ~ I... ~ .~ ~ ~ ~ I... If. . ~ i. .~ . . .... -. . ~ ~-~ i ..~ ~ . ~ ~- ,. - ~ -~ - ..... --a .. ~.. -Gil ~ ~ I. . ~
42 The CAL Thyroid Function Study ..... i.,., it., .. ... , ~ ~ ~ , ine~---£ase:~h~ ~t .... , ..... ~ ~.~.~.~-~.~ ~ , ~ rapeutl£-~ le iii~h;~=nce=. On July 25, 1975, the FDA established for the first time limits on radiation to adult human volunteer research subjects (Department of Health, Education, and Welfare, 19751. The amount of radioactive material administered to human research subjects during the course of a research project intended to obtain basic information regarding the metabolism (e.g., kinetics, distribution, and localization) of a radioactively labeled drug should be the smallest radiation dose Mat can be administered without jeopardizing the benefits to be obtained from the study. The limit for the thyroid gland is 5 red as a single dose and 15 red as a cumulative dose from a number of studies conducted within one year. For children (research subjects under IS years of age), limits are 10 percent of adult limits. Thyroid doses below these levels are generally recognized as safe (Department of Heals and Human Services, 1990; Mossman, 1992).
