previous.gif (1366 bytes)

Relation of Cancer To Other Thyroid Disease

The relationship of thyroid tumors to other thyroid disease is still debated. In the preceding section we discussed whether carcinomas arise from adenomas, occurring either singly or as a component of a multinodular gland. While this must happen rarely, it is not the ordinary course of events. In support of this view one may note, for example, that, whereas adenomas are rarely if ever papillary,  60-70% of all thyroid carcinomas are papillary. If carcinomas arise from adenomas, one might expect that the majority would be follicular rather than papillary, and this is not the case. Also, although carcinomas, largely of the papillary type, occur in nontoxic nodular goiters with a reported frequency of 4-17% of cases, the age of diagnosis of papillary carcinomas does not follow that for nontoxic goiter (181). Papillary carcinomas occur in children and adolescents, and reach their highest frequency during the middle decades of life. Multinodular goiter, by contrast, is infrequent in childhood, but increases with each decade. The high frequency of carcinomas detected in nodular goiter appears to reflect the efficiency of selection of patients for operation on the basis of suspicious clinical findings in the gland. Although it remains unproven, it is likely that in many or most thyroid adenomas and carcinomas, one specific mutational event leads directly to the development of the specific neoplasm.

Parathyroid adenomas occur in a small percentage of patients with thyroid cancer. The converse relationship may also exist; 2-11% of patients with parathyroid adenomas also have thyroid cancer (182-184). An important reason for this association is the induction of both tumors by X-ray exposure.

Neoplasia and Autoimmune thyroid diseases.

An increased incidence of cancer in Hashimoto's thyroiditis has been reported, but clinical experience certainly does not suggest an important relationship between this relatively common disease and thyroid cancer. Certainly the diseases can coexist in the same gland. Further, focal thyroiditis may occur as an immunologic response to thyroid cancer. In most series the coexistence of Hashimoto's thyroiditis among thyroid cancer patients is between 2-4%, but the real association, particularly with microcarcinomas, is difficult to be assessed because Hashimoto's thyroiditis is rarely operated upon. Focal thyroiditis and lymphocytic infiltration within the cancer are commonly observed, but this is another aspect, probably representing an immunologic response to thyroid cancer rather then a true autoimmune phenomenon.

Many reports on Graves' disease stress a normal or low coincidence of cancer, but several series have reported a significant association between Graves' disease and thyroid cancer, ranging from 3 to 10% (185-189). However, most of these series were surgical, and the patients were selected for surgery on the basis of suspicious nodules or large goiters. In Graves' patients treated by radioiodine, no subsequent increase in the discovery of thyroid cancer has been reported. In our review, 4 of 50 patients with thyroid cancer had coincident Graves' disease (190). Belfiore et al found the risk of thyroid cancer in Graves' disease to be increased 2-3 fold (191). Valenta and co-workers (192) reported that patients with thyroid cancer may have a LATS-like substance in the blood. A TSH-like component that cross-reacted with bovine TSH was also found in cancer patients and was not suppressed by thyroid hormone (193). This phenomenon was explained on the basis of anti-bovine TSH antibodies present in these sera, induced by prior testing or treatment with bovine TSH (194). However, TSAb can stimulate thyroid cancers when Graves' disease coexists, so the idea that TSAb might induce malignant change is tenable, but not proven. It also is possible, but unproven, that continued stimulation of a tumor may make it behave in a more aggressive manner (195, 196). Patients with Graves' disease and thyroid cancer who underwent total thyroidectomy and 131I ablation fared as well in follow-up as did patients without Graves' disease.

Hereditable Tumors (See discussion under CAUSES)

Genetic factors clearly play a role in some cancers as noted above. Individuals with familial adenomatous polyposis of the colon have an increased frequency of thyroid tumors. Also, patients with Cowden's disease and Gardner's syndrome have associated thyroid tumors, and thyroid cancer can occur in patients with Carney Syndrome. The existence of a new entity of familial non-medullary thyroid carcinoma (FNMTC) is advocated by many. Indeed, in almost all series there is the occurrence of familial differentiated, non-medullary, thyroid cancer in 2-4% of the first degree relatives. Whether this association is a casual one, perhaps due to the frequent occurrence of background nodular goiter in many families, or a true causal association is currently under investigation using molecular approaches.

Panza et al (197) found a strong association between histo- compatibility antigen HLA-DR1 and the occurrence of papillary and follicular cancer in Italy. This observation needs confirmation, and it is surprising in view of the usual sporadic rather than familial occurrence of these tumors. We found an association between HLA-DR7 and differentiated thyroid cancer in patients not exposed to x-ray. Others have found no association with HLA- DR genes. Medullary thyroid cancers are associated in inherited syndromes (MEN 2) with bilateral pheochromocytomas, parathyroid hyperplasia, neurofibromas, and mucosal neuromas. The associated ret oncogene mutations are noted above.

Occult or micro-carcinomas

The term occult carcinoma refers to tiny carcinomas (<1 cm), usually papillary and often sclerotic. Such tumors were frequently found in autopsy studies when the thyroids were sectioned completely at 1-2 mm intervals and every abnormality was studied, or during histology of glands operated for large multinodular goiters (198, 199). Nowadays, with the extensive use of thyroid ultrasound, micronodules (<1 cm) are frequently discovered in the general population , and proven malignant in a significant percentage. These tumors are probably synonymous with the "occult" tumors described at autopsy, but since they are discovered during life can no longer be considered occult. The term microcarcinomas is probably to be preferred to occult.

Tiny carcinomas, usually papillary and often sclerotic, have been found in 5.7% of thyroids of adults coming to autopsy in the United States (198). This prevalence is noted only when the thyroid is sectioned completely at 1-2 mm intervals and every abnormality is studied. The tumors have a mean diameter of 2 mm, and almost all are under 5 mm. The prevalence is best known in adults (199) and may be lower in young people. Since, in some glands, collections of psammoma bodies also exist in tiny scarred areas, it is hypothesized that such lesions may spontaneously regress. Because of their small size, they are detectable only at surgical or pathologic examination of the gland. These observations, now widely accepted, have provoked much discussion. Certainly most of these tiny tumors cannot be biologically significant, considering the low incidence of clinically recognized cancer. Most of the lesions are probably missed during routine surgical or autopsy pathologic studies. Their cause is unknown. They may be a variant of the occult sclerosing carcinomas described by Hazard, but the latter tumors are usually larger, have a predominant sclerotic component, and clearly do metastasize. It has been suggested that such lesions are in fact the cancers found after thyroid irradiation, but most likely they can explain only a small proportion of radiation-associated tumors. Most of these lesions would go undetected in a standard surgical pathologic examination, and the great majority of radiation-associated lesions are larger. In our own series, the radiation-associated lesions were on average 1.7 cm in diameter and only 14% were under 0.5 cm. Whether such "minimal"131 cancers are in fact the occasional precursors of clinically evident cancers is a moot point. It is clear that they are at present an important pathologic -- but not clinical -- entity.

continue.gif (1341 bytes)