Loss of heterozygosity (LOH) studies have been used to identify sites harbouring tumour suppressor genes (TSGs) involved in tumour initiation or progression. To further elucidate the genetic mechanisms for follicular and papillary thyroid tumours development, we studied the frequency of LOH in 36 thyroid tumours (21 follicular thyroid adenomas (FAs) and 15 papillary thyroid carcinomas (PTCs)) on 10 specific genomic areas: 3p22, 3p25, 7q21, 7q31, 10q23, 10q25-26, 11q13, 11q23, 13q13 and 17p13.3-13.2 using 20 polymorphic markers. We have selected these areas for two reasons: (a) Even though LOH in thyroid neoplasms has been described in some of these areas, results are controversial, and (b) we have also studied areas described as involved in other epithelial or endocrine tumour types, but not studied up to now in thyroid neoplasms. Two areas showed a high percentage of LOH: 7q31 and 11q23. A 62% LOH was found at 7q31 in the FAs, suggesting, as other authors have proposed, that at least one TSG must be present in the vicinity of the c-met locus. The second area in frequency was at the 11q23 locus, with a 45% LOH in the FAs. This area was studied because it has been described as being involved in the development of epithelial and endocrine cancers. This locus had not been studied before in thyroid neoplasms. This result is interesting because the LOH11CR2A gene is localised at this locus. We suggest that this gene and/or an other TSG nearby may be involved in the progression to FA. In our study, a low percentage of LOH was found in the PTC samples, indicating that TSGs present in the areas we have studied are not significantly involved in their progression. Our data also suggest that TSGs located in areas where no LOH was detected (PTEN, MEN1, Cyclin D1, BRCA2 and RFC3) are not involved or do not have an important role in tumour progression.