Post-Chernobyl Thyroid Carcinoma in Belarus Children and Adolescents

0021-972X/97/$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1997 by The Endocrine Society

Vol. 82, No. 11 Printed in U.S.A.

Post-Chernobyl Thyroid Carcinoma in Belarus Children and Adolescents: Comparison with Naturally Occurring Thyroid Carcinoma in Italy and France* FURIO PACINI, TATIANA VORONTSOVA, EUGENI P. DEMIDCHIK, ELEONORA MOLINARO, LAURA AGATE, CRISTINA ROMEI, ELENA SHAVROVA, EUGENY D. CHERSTVOY, YURIY IVASHKEVITCH, ELVIRA KUCHINSKAYA, MARTIN SCHLUMBERGER, GIUSEPPE RONGA, MAURO FILESI, AND ALDO PINCHERA Institute of Endocrinology (F.P., E.M., L.A., C.R., A.P.), University of Pisa, Pisa 56124, Italy; Institute of Radiation Medicine (T.V., E.S., E.K.) and Oncology-Pathology Department (E.D., E.C., Y.I.), State Medical University, Minsk 220600, Belarus; Medicina Nucleare (G.R., M.F.), Clinica Medica II, Universita` La Sapienza, Roma 00140, Italy; and Institute Gustave Roussy (M.S.), Villejuif 94805, France ABSTRACT After the Chernobyl nuclear accident (April 26, 1986), childhood thyroid carcinoma had a great increase in Belarus and Ukraine, as a consequence of the exposure to iodine radioactive fallout. The epidemiological and clinical features of the disease were studied in 472 patients less than 21 yr old at diagnosis, with differentiated thyroid carcinoma, representing 97.7% of all thyroid carcinomas diagnosed in Belarus between May, 1986, and December, 1995. The results were compared with those of 369 subjects of the same age group, with naturally occurring thyroid carcinoma, observed in Italy and France. Between 1986 and 1989, the number of thyroid cancer cases per year ranged from 3– 8 and increased to 31 in 1990, to 66 in 1991, to 72 in 1992, to 93 in 1993, to 96 in 1994, and to 90 in 1995. The age at diagnosis was 14 yr or less in 78.8% (children group) and more than 14, but less than 21, yr in the remaining subjects (adolescents group). Mean (6SD) age at the time of the accident was 4.4 6 3.4 yr (3.2 6 2.3 in children and 8.9 6 2.7 in adolescents), the majority of the patients (62.9%) being 5 yr old or less. The time interval between the accident and the diagnosis (latency period) decreased progressively from 7.5 6 1.6 yr in children 0 –2 yr old at the time of the accident to 6.0 6 1.6 yr in those 9 –11 yr old. Since 1993, the yearly distribution of new cases showed a decrease in the subjects 9 yr old or more at the time of the accident but not in those 5 yr old or less. This could not be accounted for by a shift of exposed subjects to an age group at diagnosis not included in this study, because only subjects less than 12 yr of age at the time of the accident were considered in this analysis. Mean age at diagnosis in Belarus patients was 11.3 6 3.1 yr (10.1 6 2.3 in children and 15.7 6 1.4 in adolescents), whereas, among pa-

tients with naturally-occurring thyroid carcinomas from Italy and France, the majority of cases were diagnosed after 14 yr of age (mean age at diagnosis: 14.6 6 4.2 yr). The female-to-male ratio was significantly higher in Italy and France (2.5/1), compared with the ratio of patients from Belarus (1.6/1). Most of the tumors were papillary in both series, but a relatively high proportion of follicular carcinomas (P 5 0.0001) was found in Italy/France (15.2%), as opposed to 5.3% in Belarus. Extrathyroidal extension and lymph node metastases were more frequent in Belarus (49.1%, P 5 0.0001; and 64.6%, P 5 0.002, respectively) with respect to Italy/France (24.9% and 53.9%, respectively). Thyroid lymphocytic infiltration and circulating antithyroperoxidase antibody were more frequent in Belarus patients. Our analysis of Belarus thyroid cancer patients less than 21 yr old showed that the post-Chernobyl increase in thyroid carcinomas involved both children and, to a much lesser extent, adolescents. Subjects 5 yr old or less at the time of the accident accounted for the majority of the patients. No evidence of a decrease in the number of new cases was observed in this age group, as opposed to older subjects. These data support the concept that subjects who were younger at the time of radiation exposure had, and continue to have, a greater risk of developing thyroid carcinoma and strongly suggest that this age group should be carefully monitored in the future. When compared with naturally occurring thyroid carcinoma of the same age group observed in Italy and France, the post-Chernobyl Belarus thyroid carcinomas affected younger subjects, were less influenced by gender, were virtually always papillary, had a greater aggressiveness at presentation, and were more frequently associated with thyroid autoimmunity. (J Clin Endocrinol Metab 82: 3563–3569, 1997)

A

and continuing up to now (1–3). According to the Belarus medical authorities, before 1989, thyroid carcinoma in Belarus was uncommon and had the same incidence as found in European countries and in the United States. An increase of childhood thyroid carcinoma after the Chernobyl accident also has been reported in the Ukraine (4, 5) and, to a lesser extent, in the Russian Federation (6). The size of the increase in the number of childhood thyroid cancers is too high to be explained by increased awareness. Based on the geographical and temporal distribution of the cases, there is strong evidence that the increased incidence of childhood thyroid

FTER the Chernobyl accident on April 26, 1986, a great increase in the prevalence of childhood thyroid carcinoma was observed in Belarus, only 4 yr after the accident Received January 27, 1997. Revision received June 24, 1997. Accepted July 18, 1997. Address all correspondence and requests for reprints to: Furio Pacini, M.D., Istituto di Endocrinologia, Via Paradisa, 2, 56124 Pisa, Italy. * This work was supported, in part, by grants from: European Comunities; project JSP4 (COSU CT 93– 00-52 and COSU CT 94 – 00-90); project NuFiSa (FI4CCT960002); Associazione Italiana Ricerca sul Cancro; Consiglio Nazionale delle Ricerche, Rome (ACRO project 95.00417.PF39).

3563

3564

PACINI ET AL.

cancer is caused by radiation exposure, as a result of the Chernobyl nuclear reactor accident (most likely, to the huge amounts of iodine radioisotopes released by the damaged Chernobyl reactor) (7–9). This outbreak of radiation-induced childhood thyroid carcinoma raises the question of whether it differs in its biological and clinical behavior, with respect to childhood thyroid cancer with no previous history of radiation exposure. Moreover, 10 yr after the accident, it remains to ascertain whether the increase in thyroid carcinoma will be limited to children or will extend to adolescents and young adults. To answer these questions, we studied the clinical and epidemiological features of thyroid carcinomas diagnosed in the Republic of Belarus, after the Chernobyl accident, from May, 1986, to December, 1995, both in children (age at diagnosis: ,14 yr; n 5 372) and in adolescents and young adults (age at diagnosis: .14 to ,21 yr; n 5 100), here referred to as adolescents. The clinical and epidemiological features of these patients were compared with those of 369 children and adolescents that, in the past 20 yr, received treatment for thyroid carcinoma in 2 centers in Italy (Pisa and Roma; n 5 219) and one center in France (Villejuif; n 5 150). Subjects and Methods Patients Belarus patients. The Belarus study group included 472 patients (288 females, 184 males) with thyroid carcinoma diagnosed when they were less than 21 yr old. According to the official data of the Belarus medical authorities, these represent 97.7% of all cases of thyroid carcinoma registered in Belarus in that specific age group, from May 1986 to December 1995. Patients were classified as children (age ,14 at diagnosis; n 5 372) or adolescents (age at diagnosis .14 and ,21 yr; n 5 100). With the exception of 2 children born in 1988, all patients included in this study were already born, or were in uterus (n 5 9), at the time of the Chernobyl accident and were living in regions of Belarus, to a variable extent, contaminated by radioactive fallout. As indicated in previous reports (10), the number of cases of thyroid carcinoma was grossly related to the degree of radioactive contamination, more than half of the cases (51.9%) being observed in the most exposed region of Gomel and less than 3% in the least contaminated region of Vitebsk (Fig. 1). In particular, 245 (205 children and 40 adolescents) were living in Gomel,

FIG. 1. Degree of radioactive contamination and number of post-Chernobyl thyroid cancer cases in children and adolescents in various regions of Belarus. The number of cases and percentage of all cases are indicated for each region.

JCE & M • 1997 Vol 82 • No 11

106 (82 children and 24 adolescents) in Brest, 58 (40 children and 18 adolescents) in Minsk, 28 (19 children and 9 adolescents) in Grodno, 24 (14 children and 10 adolescents) in Mogilev, and 11 (5 children and 6 adolescents) in Vitebsk. Initial treatment of all patients was carried out in the Thyroid Center of the Belarus Ministry of Health in Minsk and consisted of total (or near-total) thyroidectomy in 173 patients, subtotal thyroidectomy in 75, lobectomy in 217, and undetermined in 7. Western European patients. The control group of patients, with thyroid carcinoma and no previous history of radiation exposure, consisted of 369 subjects (266 females, 103 males) of the same age groups, consecutively observed in the past 20 yr at the Institute of Endocrinology of the University of Pisa, Italy (n 5 116), at the Nuclear Medicine Department, University of Roma, Italy (n 5 103), or at the Institute Gustave-Roussy, Villejuif, France (n 5 150). Initial treatment was near-total thyroidectomy in 307 patients, lobectomy plus isthmectomy in 56, less than lobectomy in 4, and undetermined in 2. It is worth noting that the initial treatment was clearly different in the Belarus and Italy/France series, because of local conditions and therapeutic strategies. This difference may well have important implications, with regard to the outcome of the disease, an issue which has not been taken into consideration in the present study.

Methods A centralized data base, including all cases of thyroid cancer registered in Belarus, is located in Minsk. The data included in the present study refer to all registered cases of the concerned age groups, with no selection, except for 13 cases with insufficient information because of initial treatment outside Belarus. Cancer classification used by the Belarus pathologists was that recommended by WHO. Many cases have been revised, on several occasions, by Western European expert pathologists, who have confirmed the diagnosis in almost all cases (11). The clinical records of all patients were examined. The features, the age, and the sex distribution of the thyroid carcinomas observed in Belarus and in the centers of Italy and France were compared. A moderate increase in thyroid cancer in adults, commonly recognized as the consequence of ascertainment bias, has been reported in Belarus. We have not considered the adults in this study, because of the difficulty in collecting reliable and verifiable data in this age group. Statistical analysis was performed using the Student’s t test, x2 test, and linear regression analysis, as appropriate. The x2 test was always performed to compare two qualitative variables in the two populations (Belarus vs. Italy/France). Antithyroperoxidase autoantibodies (anti-TPO) were measured by RIA (Sorin Biomedica Diagnostics, Saluggia, Italy; normal range: ,10 U/mL); antithyroglobulin autoantibodies (anti-Tg) were measured by

POST-CHERNOBYL THYROID CARCINOMA IN BELARUS

3565

immunoradiometric assay (Biocode, Solessin, Belgium; normal range: 5–50 U/mL).

Results Belarus patients: age at radiation exposure and age at diagnosis

As reported in Fig. 2, the number of thyroid cancers per year in Belarus ranged from 3 to 8 between 1986 and 1989. Then, a progressive increase was observed, continuing up to 1995: 31 cases in 1990, 66 in 1991, 72 in 1992, 93 in 1993, 96 in 1994, and 90 in 1995. The increase was observed mainly in children but also occurred in adolescents, with a similar trend. Figure 3 shows the distribution of cases of thyroid cancer according to the age of the patients at the time of the accident. At that time, most of the patients who subsequently developed thyroid carcinoma were in the childhood group, 62.9% of all patients being less than 5 yr old. At diagnosis, the distribution of Belarus cases had a peak at 10 yr (Fig. 4). Considering all patients, the mean age at the time of the Chernobyl accident and at the time of diagnosis was 4.4 6 3.4 yr (range 0 –17.6) and 11.3 6 3.1 yr (range 3.9 –20.1), respectively. Considering separately the two age groups (children and adolescents), the mean age at the time of the accident and at the time of diagnosis was 3.2 6 2.3 and 10.1 6 2.2 in children, and 8.9 6 2.7 and 15.7 6 1.4 in adolescents, respectively. Only 11 cases of thyroid carcinoma were observed in subjects who were born after the Chernobyl accident; 9 of them were generated before, but were still in uterus at the time of the accident. The distribution of cases according to the latency period (the time elapsed between exposure and diagnosis) showed a similar trend in children and adolescents. The mean latency period was 6.9 6 1.9 yr (range 1–10.2), with no significant difference between children (6.9 6 1.8) and adolescents (6.8 6 2.0). As shown in Fig. 5, the age at diagnosis progressively increased, in relation to the age at the accident. The correlation coefficient was highly significant, both when all patients were considered together (r 5 0.84; P 5 0.0001) and when the two age groups were considered separately: r 5 0.67, P 5 0.0001 in children and r 5 0.68, P 5 0.0001 in adolescents. As shown in Fig. 6A, a weak (but significant)

FIG. 3. Age distribution of Belarus thyroid cancer patients at the time of the accident (1986). The black column (f) and the white column (M) represent patients diagnosed during childhood and during adolescence, respectively. The first column (,1 yr) includes nine patients who were in uterus at the time of the accident. The square bracket indicates the total number and percentage of thyroid cancer patients who were 5 yr old or less at the time of the accident.

negative correlation was found between the latency period and the age at the time of the accident, longer latency periods being observed in younger subjects. To overcome the bias possibly deriving from the shift of the subjects who were more than 11 yr old at the time of the accident to an age group at the time of diagnosis not included in this study (i.e. $ 21 yr), the data were further analyzed by excluding subjects older than 11 yr at the time of the accident. As shown in Fig. 6B, a weak negative correlation between latency period and age at accident was found also after exclusion of these subjects. Mean latency periods were 7.5 6 1.6 yr, 7.2 6 1.7 yr, 6.6 6 1.8 yr, and 6.0 6 1.6 yr for the age groups 0 –2 yr, 3–5 yr, 6 – 8 yr, and 9 –11 yr at the time of the accident, respectively. The slight decrease observed was statistically significant (P 5 0.0001). The yearly distribution of new cases of thyroid carcinoma, in relation to the age at the time of the accident, is shown in Fig. 7. To avoid the bias mentioned above, subjects 11 yr old or more at the time of the accident were excluded from this analysis. It is apparent that, starting from 1990, the patients of the age group of 5 yr or less accounted for the majority of the cases in each year of observation. The number of cases among the subjects 9 yr old or more decreased since 1993, and no cases of this age group were registered in 1995. Belarus vs. Western European controls

FIG. 2. New cases of childhood (f) and adolescent (M) thyroid carcinoma in Belarus, registered yearly from 1986 to 1995.

In Belarus, 288 patients were females (227 children, 61 adolescents) and 184 were males (145 children, 39 adolescents), with a female to male (F/M) ratio of 1.6/1, with no difference between children and adolescents. In Italy and France, 266 patients were females (113 children, 153 adolescents) and 103 were males (45 children, 58 adolescents), with an F/M ratio of 2.5/1 (P , 0.0001), significantly higher with respect to Belarus patients; there was no difference in F/M ratio between children and adolescents. The distribution of the cases in the control group from Italy and France, according to the age at diagnosis, increased progressively with the age, the majority (57.4%) of the pa-

3566

PACINI ET AL.

JCE & M • 1997 Vol 82 • No 11

FIG. 4. Age distribution at the time of diagnosis of thyroid cancer patients from Belarus and from Italy and France.

FIG. 5. Correlation between age at the time of the accident and age at diagnosis in Belarus children and adolescents. Points before the 0 line are patients in uterus.

tients being diagnosed after the age of 14 yr. The mean age at diagnosis was 14.6 6 4.2. In contrast, in Belarus, most of the cases (87.9%) were diagnosed before the age of 15 yr and the mean age was 11.3 6 3.1 (Fig. 4). As shown in Table 1, most Belarus tumors (443 5 93.8%) were papillary (with or without a follicular component), 25 were pure follicular (5.3%), 2 were medullary (0.4%), 1 was Hurthle cell carcinoma (0.2%), and 1 was undifferentiated carcinoma (0.2%). In the Western European series, 303 tumors (82.1%) were papillary (with or without a follicular component), 56 (15.2%) follicular, and 10 (2.7%) unclassified. Two cases of hereditary medullary thyroid carcinoma were excluded from the Italian series because they were diagnosed by screening of pedigrees with multiple endocrine neoplasia type 2. The differ-

ence between papillary and follicular carcinomas of Belarus and Western Europe was statistically significant (P 5 0.0001, by x2). Almost half (49.1%) of the Belarus tumors extended outside the thyroid gland (188 children and 44 adolescents). Lymph node involvement was present in 64.6% of the cases (247 children and 58 adolescents), and distant metastases were found in 7.8%. In the control group, tumors extending outside the thyroid gland were much less (24.9%), lymph node involvement was present in a slightly lower percentage (53.9%), both when considering all cases and when considering only papillary thyroid carcinomas, and distant metastases were found in 17.3%. It is worth noting that, with a few exceptions, the Belarus patients were not submitted to 131-I whole-body scan after thyroidectomy, and only lung metastases visible by chest x-ray could be detected. In the control group, also distant metastases (recognized shortly after surgery by 131-I whole-body scan) were counted. Thus, the incidence of distant metastases was not comparable in the 2 groups and were not statistically analyzed. When considering only lung metastases detectable by chest x-ray, their prevalence was similar in the groups (7.7% in Italy/France; 7.8% in Belarus). Autoimmune phenomena (Table 2)

Lymphocytic infiltration was studied in 52 Belarus tumors and was present in 27 (51.9%) cases. The lymphocytic infiltration was diffuse and typical of chronic autoimmune thyroiditis in 5 cases and focal in the remaining 22 cases. This aspect was analyzed in 96 patients of the Italian series and 22.1% of cases were found to have lymphocytic infiltration (mostly focal). In a subgroup of 171 Belarus children, serum samples were available for thyroid autoantibodies determination. The fre-


3567

FIG. 7. Distribution of new cases of thyroid carcinoma observed in each year, in relation to the age at the time of the accident. The following age groups were considered: z, 0 –2 yr; f, 3–5 yr; s, 6 – 8 yr; M, and 9 –11 yr.

FIG. 6. Correlation between age at the time of the accident and latency period of thyroid cancer in all Belarus patients (panel A) and in children less than 11 yr of age at the time of the accident (panel B).

quency of anti-TPO (with or without anti-Tg) was significantly higher (P , 0.01) in Belarus patients than in Italian patients (46% vs. 23.1%). No antibody data are available in the French series. Discussion

The occurrence of a great increase of childhood thyroid carcinoma in Belarus, Ukraine and (to a lesser extent) in the Russian Federation after the Chernobyl accident has been documented in previous reports (4 – 6). Evidence based on geographical and temporal distribution of the cancer cases and the occurrence of very few cases in children born after the accident supports the concept that such increase resulted from exposure to radiation, radioiodine isotopes released from the damaged Chernobyl nuclear reactor being the most likely causative agents (7–10). In the present study, the clinical and epidemiological features of the thyroid carcinoma occurring in Belarus was studied and compared with the naturally occurring thyroid carcinoma observed in Italy and France. Previous reports (1– 4, 9, 10) on post-Chernobyl thyroid carcinoma focused on patients who were children at the time of diagnosis. The present study was extended to the thyroid carcinomas occurring in subjects more than 14, but less than 21, yr old when diagnosed, here referred to as adolescents.

Our data indicate that adolescents showed an increase of thyroid carcinoma similar to, although much less pronounced than, that observed in children. The distribution of the cases according to the year of diagnosis showed that the increase in children reached its peak in 1993, with a trend to a plateau in the following years. The question of whether this finding indicates that the epidemic of radiation-induced thyroid cancer is approaching its peak, rather than resulting from epidemiological bias caused by the shift of exposed subjects from childhood to the adult cohort, may be raised. As indicated above, we addressed this issue by including in our analysis subjects more than 14, but less than 21 yr old at the time of diagnosis, 6 –18 yr old at the time of the accident. If the lack of a further increase of the number of thyroid carcinomas in children observed in the most recent years was caused by a shift of the exposed subjects from childhood to adolescence, one would expect a compensatory increase in the adolescents. This was not observed, supporting the concept that the epidemic might be in the process of approaching its peak. It is worth noting that the trend to plateau in the outbreak of post-Chernobyl thyroid carcinoma suggested by the present study does not exclude the possibility of a second peak occurring in the next few decades and having a longer latency period, similar to that observed after external radiation to the head and neck (12–15). The distribution of the patients, with regard to the age at the time of the accident showed that the younger subjects (less than 5 yr old) accounted for the majority of the cases: 62.9% of all patients and 79.8% of the children. This is consistent with the notion that the thyroid gland of young children is more sensitive to the carcinogenic effects of radiation (16). However, the possibility that, to some extent, this could be caused by the variable time interval between exposure to radiation and observation among the different age groups should also be taken into account. To overcome this problem, we analyzed the yearly distribution of new cases by considering only subjects with an equal period of follow-up. The relative preponderance of children less than 5 yr old at the time of the accident was confirmed in each year of observation since 1990, further supporting the concept that the younger age groups were at greater risk. Interestingly, a

3568

JCE & M • 1997 Vol 82 • No 11

PACINI ET AL.

TABLE 1. Histotype and tumor extension of thyroid cancer in children and adolescents (Belarus vs. Italy and France) Belarus (n 5 472)

Histology Papillary Follicular Medullary Hu¨rthle Anaplastic Unknown Extension Extrathyroid Lymph nodes Distant met.c

Italy/France (n 5 369)

n

(%)

n

(%)

443 25 2 1 1 0

(93.9) (5.3) (0.4) (0.2) (0.2)

303 56 (2)b 0 0 10

(82.1) (15.2)

232 305 37

(49.1) (64.6) (7.8)

92 199 64

(24.9) (53.9) (17.3)

P 5 0.0001a

(2.7) P 5 0.0001 P 5 0.002

By x2 limited to papillary and follicular. Excluded from analysis because detected by genetic screening of medullary thyroid carcinoma. c In Belarus patients the presence of distant metastases was assessed by x-rays; in Italy/France, by x-rays and a b

131

I-WBS.

TABLE 2. Circulating antithyroid autoantibodies and tumoral lymphocytic infiltration in Belarus and Italian children

Belarus Italy a

Anti-TPO

Anti-TG

Both

Limphocytic infiltration

79/171 (46.0 %)a 24/103 (23.3 %)

30/171 (17.5 %) 10/103 (9.7 %)

17/171 (9.9 %) 6/103 (5.8 %)

27/52 (51.9 %)a 22/96 (22.9 %)

P , 0.01 by x2.

trend to a decrease in the number of thyroid cancer cases was observed in the subjects who were 9 yr old or more at the time of the accident, with no new cases being observed in 1995. These data suggest that the risk of developing thyroid carcinoma in the next few years persists in the subjects exposed to radiation at a very young age, whereas it may decrease in the older age groups. The most striking difference between the thyroid cancers observed in Belarus children and adolescents after the Chernobyl accident and the naturally-occurring cancers observed in the same age groups in Italy and France, was in the age distribution. The Belarus patients showed a distribution pattern resembling a Gaussian curve, with a peak at 10 yr, the large majority of the patients being diagnosed before or at the age of 14 yr. In contrast, the Western European series showed a progressive increase up to the age of 18 yr, the majority of the cases being diagnosed after the age of 14 yr. The difference is in keeping with the concept that the Belarus thyroid carcinoma outbreak was related to a single event, i.e. exposure to radiation action within a limited and defined period of time and having a greater carcinogenic effect in the younger age group. In Belarus, as well as in Italy and France, females were more frequently affected than males, but the preponderance of females was much more pronounced in the Western European series. Similar findings have been found in the series from England and Wales (17). This suggests that the role of predisposing factors commonly associated to the female sex is less relevant in cases of radiation-induced thyroid tumors than in cases of naturally occurring thyroid carcinomas. Papillary carcinomas accounted for most of the thyroid tumors found in both series, but the proportion of this histotype was clearly greater in Belarus than in the Western European centers, whereas the reverse was true for follicular carcinomas. The high prevalence of papillary carcinomas in

the Belarus patients should not be regarded as surprising, because this histotype is found in virtually all patients developing thyroid carcinoma after external radiation to the head and neck regions (12–15). When compared with the Western European series, at histology the Belarus patients showed more frequent extrathyroidal extension of the tumor to the surrounding tissues. This and other histological evidence of a relatively greater aggressiveness of post-Chernobyl thyroid tumors, with regard to the naturally occurring papillary carcinomas, has been reported previously in other studies (18, 19). In both series, cervical lymph node metastases were documented at surgery in many cases, in keeping with the notion that these metastases are frequently present at the time of diagnosis in children and adolescents with thyroid cancer, and not infrequently are the clue leading to the recognition of the primary tumor. Admittedly, although significant, the difference in the prevalence of lymph node metastases in the Belarus and the Western European series was small and its significance, with regard to biological behavior of the tumor, remains to be established. The relatively higher prevalence of lymph node metastases found in the Belarus series is not caused by the lower frequency of follicular carcinomas, because it was still present when the comparison was limited to patients with papillary thyroid cancer. It is worth noting that the finding of lymph node metastases and of extrathyroidal extension of the tumor in a high proportion of cases provides strong support to the view that increased awareness did not play a major role in the detection of the postChernobyl thyroid cancer in Belarus. Lymphocytic infiltration of the thyroid with the pattern of focal thyroiditis (in most cases) and of diffuse chronic thyroiditis (in some) was found in about half of the 52 Belarus patients in whom these changes could specifically be assessed. In a similar percentage of larger numbers of patients


(n 5 171), serum antibodies to thyroid peroxidase and (less frequently) to thyroglobulin were detected. These figures should be compared with the nearly 20% frequency of lymphocytic infiltration and humoral thyroid autoimmunity found in this and other Italian series (20). Unfortunately, it is impossible to have, as control, an unexposed group of Belarus thyroid cancer patients with measured antibodies. We have studied a population of children from an unexposed Belarus village (Braslav), apparently with no thyroid disease, in which the prevalence of serum antithyroid antibodies was 3% (unpublished observations), not different from that found in normal children of Italy (21). All together, these data suggest that thyroid autoimmune reactions may be related to radiation exposure. This interpretation is in keeping with the observation of thyroid autoimmune phenomena in several survivors of the atomic bomb explosions, as well as in subjects exposed to radiation fallout in the Marshall Islands (22, 23). Although preliminary (requiring confirmation in a larger population, including controls not affected by thyroid cancer), these data suggest that thyroid autoimmunity may be an additional and important consequence of the Chernobyl accident. In conclusion, our analysis of Belarus thyroid cancer patients less than 21 yr old showed that the post-Chernobyl increase in thyroid carcinomas involved both children and, to a much lesser extent, adolescents. Subjects 5 yr old or less at the time of the accident accounted for the majority of the patients. Ten years after the accident, no evidence of a decrease in the number of new cases is observed in this age group, as opposed to older subjects. These data support the concept that subjects of younger age at radiation exposure had, and continues to have, a greater risk of developing thyroid carcinoma (24) and strongly suggest that this age group should be carefully monitored in the future. When compared with naturally occurring thyroid carcinoma of the same age group observed in Italy and France, the post-Chernobyl Belarus thyroid carcinomas affected younger subjects, were less influenced by gender, were virtually always papillary, had a greater aggressiveness at presentation, and were more frequently associated with thyroid autoimmunity.

5. 6.

7.

8. 9.

10.

11.

12. 13. 14. 15. 16. 17. 18.

19.

20.

References 1. Kazakov US, Demidchik EP, Astakhova LN. 1992 Thyroid cancer after Chernobyl. Nature. 359:21. 2. Baverstock K, Egloff B, Pinchera A, Ruchti C, Williams D. 1992 Thyroid cancer after Chernobyl. Nature. 359:21–22. 3. Demidchik E, Kazakov VS, Astakhova LN, Okeanov AE, Demidchik YE. 1994 Thyroid cancer in children after the Chernobyl accident: clinical and epidemiological evaluation of 251 cases in the Republic of Belarus. In: Nagataki S, ed. Nagasaki Symp., Chernobyl: Update and Future. Amsterdam: Excerpta Medica, Elsevier Press; 21–30. 4. Tronko N, Bogdanova T, Kommisarenko I, et al. 1996 Thyroid cancer in children and adolescents in Ukraine after the Chernobyl accident (1986 –1995).

21. 22. 23. 24.

3569

In: Karaoglou A, Desmet G, Kelly GN, Menzel HG, eds. The Radiological Consequences of the Chernobyl Accident, ERU 16544 EN. Luxembourg: European Commission; 683– 690. Likhtarev IA, Sobolev BG, Kairo IA, et al. 1995 Thyroid cancer in the Ukraine. Nature. 375:365. Tsyb AF, Parshkov EM, Shakhtarin VV, Stepanenko VF, Skvortsov VF, Chebotareva IV. 1996 Thyroid cancer in children and adolescents of Bryansk and Kaluga regions. In: Karaoglou A, Desmet G, Kelly GN, Menzel HG, eds. The Radiological Consequences of the Chernobyl Accident, ERU 16544 EN. Luxembourg: European Commission; 691– 697. Commission of The European Communities. 1993 Thyroid cancer in children living near Chernobyl. Williams D, Pinchera A, Karaoglou A., Chadwick KH (eds). Report EUR 15248 EN. Luxembourg: Office for Official Publications of the European Communities. 1995 Results of the IPHECA pilot projects and related national programmes. In: Health Consequences of the Chernobyl Accident. Summary Report. Geneva: World Health Organization. Sobolev B, Likhtarev I, Kairo I, Tronko N, Oleynik V, Bogdanova T. 1996 Radiation risk assessment of the thyroid cancer in Ukrainian children exposed due to Chernobyl. In: Karaoglou A, Desmet G, Kelly GN, Menzel HG, eds. The Radiological Consequences of the Chernobyl Accident, ERU 16544 EN. Luxembourg: European Commission; 741–748. Demidchik EP, Drobyshevskaya IM, Cherstvoy ED, et al. 1996 Thyroid cancer in children in Belarus. In: Karaoglou A, Desmet G, Kelly GN, Menzel HG, eds. The Radiological Consequences of the Chernobyl Accident, ERU 16544 EN. Luxembourg: European Commission; 677– 682. William ED, Cherstvoy E, Egloff B, et al. 1996 Interaction of pathology and molecular characterization of thyroid cancers. In: Karaoglou A, Desmet G, Kelly GN, Menzel HG, eds. The Radiological Consequences of the Chernobyl Accident, ERU 16544 EN. Luxembourg: European Commission; 699 –714. Shore RE, Woodlard ED, Hildreth N, et al. 1985 Thyroid tumors following thymus radiation. J Natl Cancer Inst. 74:1177–1184. Favus MJ, Schneider AB, Stachura ME, et al. 1976 Thyroid cancer occurring as a late consequence of head-and-neck irradiation. Evaluation of 1056 patients. N Engl J Med. 294:1019 –1025. Refetoff S, Harrison J, Karanfilski BT, Kaplan E, DeGroot L, Bekerman C. 1975 Continuing occurrence of thyroid carcinoma after irradiation to the neck in infancy and childhood. N Engl J Med. 292:171–175. DeGroot LJ, Paloyan E. 1973 Thyroid carcinoma and radiation: a Chicago endemic. JAMA. 255:487–501. Shore RE. 1992 Issues and epidemiological evidence regarding radiationinduced thyroid cancer. Radiat Res. 131:98 –111. Harach HR, Williams ED. 1995 Childhood thyroid cancer in England and Wales. Br J Cancer. 72:777–783. Cherstvoy E, Pozcharskaya V, Harach HR, Thomas GA, Williams ED. 1996 The pathology of childhood thyroid carcinoma in Belarus. In: Karaoglou A, Desmet G, Kelly GN, Menzel HG, eds. The Radiological Consequences of the Chernobyl Accident, ERU 16544 EN. Luxembourg: European Commission; 779 –784. Bogdanova T, Bragarnik M, Tronko ND, Harach HR, Thomas GA, Williams ED. 1996 The pathology of thyroid cancer in Ukraine post Chernobyl. In: Karaoglou A, Desmet G, Kelly GN, Menzel HG, eds. The Radiological Consequences of the Chernobyl Accident, ERU 16544 EN. Luxembourg: European Commission; 785–789. Pacini F, Mariotti S, Formica N, Elisei R, Capotorti E, Pinchera A. 1988 Thyroid autoantibodies in thyroid cancer: incidence and relation with tumour outcome. Acta Endocrinol (Copenh). 119:373–379. Mariotti S, Sansoni P, Barbesino G, et al. 1992 Thyroid and other organspecific autoantibodies in healthy centenarians. Lancet. 339:1506 –1508. Nagataki S, Shibata Y, Inoue S, Yokoyama N, Izumi M, Shimaoka K. 1994 Thyroid disease among atomic bomb survivors in Nagasaki. JAMA. 272:364 –370. Conrad RA, Peglia DE, Larson PR, et al. 1980 Review of medical findings in a Marshallese population twenty-six years after accidental exposure to radioactive fallout, BNL 51261, NTIS, January. Nikiforov Y, Gnepp DR, Fagin JA. 1996 Thyroid lesions in children and adolescents after the Chernobyl disaster: implications for the study of radiation tumorigenesis. J Clin Endocrinol Metab. 81:9 –14.