Up to 10% of adults will develop a thyroid nodule in their lifetime. Although the majority of these are benign, more than 12,000 new cases of thyroid cancer develop each year in the United States. The diagnostic tools available to discriminate benign from malignant neoplasms are not reliable, which results in a large number of unnecessary surgeries. Furthermore, prognostic markers are not particularly useful. Although surgical resection and radioiodine ablation is curative for most thyroid cancers, there is still no effective treatment for undifferentiated thyroid cancers which no longer trap 1311. A better understanding of the biology of these frequently occurring endocrine neoplasms is needed to address these clinical shortcomings. Thyroid tumors are believed to arise as a result of an accumulation of somatic mutations affecting genes involved in growth control. There is increasing evidence that inactivation of tumor suppressor genes may be important in the microevolution of thyroid neoplasms. Loss of function of tumor suppressor genes removes a growth-inhibitory control, often through mutational inactivation of both alleles, as these genes evoke the tumorigenic phenotype in a recessive manner. One of the alleles is often lost as part of a chromosomal deletion, which commonly extends beyond the region coding for the gene of interest. Consistent regions of allelic loss can therefore provide clues to the localization of tumor suppressor genes. Allelic loss is relatively infrequent in human thyroid tumors. Some sporadic follicular, but not papillary, neoplasms have loss of heterozygosity for markers in the long arm of chromosome 11 (11q13)(1). This region is known to contain several genes associated with tumorigenesis, including the gene conferring predisposition to multiple endocrine neoplasia type I (MEN1). Although it has not yet been cloned, the MEN1