Mutations in BRCA1 and BRCA2 genes account for the majority of familial aggregation of breast and ovarian cancers but other common genes in the population with low penetrance should be also involved in susceptibility to breast cancer. The H-ras minisatellite, located downstream of H-ras oncogene, is considered to be a likely candidate. Previous findings have estimated that as many as 1 in 11 cancers of the breast might be attributed to this region, but other studies observed inconsistent results. We propose to elucidate the potential role of H-ras locus in breast cancer, by looking at somatic alterations occurring in tumor DNAs such as the instability or the loss of heterozygosity (LOH) and by determining a potential correlation between constitutional specific H-ras alleles and clinical and/or pathological characteristics. DNA was extracted from 123 sporadic breast tumors and matched peripheral blood lymphocytes. 143 DNA samples from of peripheral blood lymphocytes from healthy donors served as a control population. The allelic diversity was determined by polymerase chain reaction analysis. Rare H-ras alleles were found to be present in about 9% of breast cancer patients while they were detected in only 1.4% of lymphocytes from healthy donors (P = 0.0044). Therefore, the risk of breast cancer is increased in patients with one or two rare alleles (odd ratio = 7.14 and 95% confidence interval = 1.94-22.27). Analyses of somatic alterations in tumor DNA have shown the lost of one allele, in general the longest, in 6.7% informative cases and an instability to H-ras locus in 6.5% tumors that appeared as a size increase of one of the two alleles. No correlation of rare H-ras alleles with clinicopathological parameters was found. Our results demonstrated an association of rare H-ras alleles with breast cancer and suggest that minisatellite H-ras may be considered as an informative marker for the breast cancer risk.