benign nodules or adenomatous nodules) and surgery for the specified groups (Table 3, column E; r = 0.99 and p = <0.01). This striking similarity, 4 D> o thyroid cancers was found using mean time to which is present regardless of age or the use of + thyroxine suppression, supports the notion that a i benign lesion does not evolve into a malignant one, nor do carcinomas, presumably possessing a greater degree of autonomous growth, manifest themselves clinically any earlier than benign nodules. Ron et QO w oO > oO T un Oo T Rongelap Oo Utirik & carcinomas, adenomas, and "nodules." J oS T Nm t_ al. (1989) also noted a similarity in time from radiation exposure to tumor diagnosis for 1 — Oo Total Thyroid-Absorbed Dose in Gy (internal and external) group. Only adenomas(Fig. 7b) did not develop in persons exposed beyond their teenage years. 6) Was the type of thyroid nodule induced by radiation a function of dose? It is thought that at thyroid doses above 15002000 cGy the incidence of carcinoma is decreased due to extensive cell death which leaves few cells capable of becoming neoplastic (NCRP, 1985), Years Post-Exposure Fig. & Relation of thyroid-absorbed dose to time of development of surgically conformed nodules in although there are reports of undiminished risk of thyroid cancer from external irradiation with thyroid persons who were between 1 and 6 years of age at the time of exposure on Rongelap and Utirik. However, there is a marked disparity among the four graphs for the Rongelap group. The occurrence of many adenomatous nodules in the younger Rongelap population (Fig. 7a) could have been a function of either age or high radiation dose, because the thyroid-absorbed dose was strongly age-dependent. However, none of the other three nodule types in the Rongelap exposed group (Fig. 7b-d) show the same preference for younger ages, although the number of observations is quite small. It is possible that the development of adenomas and occult papillary carcinomas, as well as overt carcinomas, was limited by thyroid cell injury resulting from the high radiation dose to the young Rongelap population. It is ironic that neoplastic nodules in the Rongelap group were numerically overshadowed by benign nonncoplastic lesions. While one might consider this to be, in a sense, a mitigation of radiation injury, the significant morbidity of benign thyroid nodules in the exposed Marshallese has been discussed (Adams et al., 1988). 5) Did benign nodules antedate carcinomas? A highly significant correlation for time of development of benign nodules (either total doses exceeding 3000 cGy (Tuckeret al., 1991). For palpable solitary nodules in the general population the usual ratio of benign to malignant lesions is about 6:1 and in someradiation-exposed groupsit can be as high as 3:1 (DeGroot et al., 1983), althoughthe ratio varies considerably depending on the definitions used. The ratios for the Marshallese are shown in Table 3, column B. In the Utirik group where the total thyroid dose was relative small the ratios for persons under 10 years of age is 3.5:1 and for those who were olderit is 3.7:1. The high benign to malignant tumorratio of 16:1 for Rongelap children who were exposed under the age of ten years and whose thyroid doses exceeded 2000 cGy is consistent with most other studies, and the likely explanation is a decrease in thyroid cancer due to extensive cell death or injury at the time of exposure. What mayalso be important, however, is the low benign to malignant tumor ratio (1.25:1) found in those Rongelap individuals with mid-range thyroid radiation doses (i.e., Rongelap individuals over the age of 10 years at exposure, who received 400 to 2000 cGy). Thus, there was a high probability in this group of a detected nodule being malignant, whereas there was a relatively low probability of malignancy in persons whose dose exceeded 2000 cGy. This may be relevant in clinical decisionmaking for nodules detected in other exposed populations. 21