Some studies suggest that genetic determinants of longevity might reside in the polymorphisms for genes that regulate immune responses as HLA genes. Some HLA alleles or haplotypes that confer resistance to infectious disease respectively via peptide presentation or via antigen non specific stimulation, have been selected all through evolution. Some of the highly conserved haplotypes are composed by the HLA-DRDQ alleles and in some instances HLA-DQ alleles play a central role in the association to disease. The relatively homogeneous Sardinian population, an ancient genetic isolate having a gene pool characterized by a relative little genetic flow from the various populations invading the island during the last 3000 years, may represent a suitable sample for association studies addressed to dissect the complex traits as longevity. So we have evaluated the HLA-DQA1 and HLA-DQB1 allele frequencies in 123 centenarians from Sardinia and 92 controls randomly selected from the inhabitants of the same country. HLA-DQ alleles typing was carried out by PCR-SSP. HLA-DQA1 and HLA-DQB1 allele frequencies were evaluated by gene count and haplotype frequencies calculated by a maximum likelihood algorithm (Arlequin Software, University of Geneva). 2 x 2 tables were constructed to determine statistical significance (2 test with Yate's correction) of differences in HLA haplotype frequencies between centenarians and controls. Obtained P values were multiplied for the number of alleles under study (Bonferroni's correction). Distribution frequency analysis suggested that within each group, the haplotype distributions were consistent with those predicted by the Hardy-Weinberg equilibrium. The frequencies of HLA- DQA1-DQB1 were not significantly different in centenarians (both in women and in men) respect to controls. Moreover, we have investigated the association between heterozygous HLA genotypes and longevity, considering that heterozygote subjects are thought more fit for survival than homozygote ones. On the basis of haplotype distribution, we have been able to show an equal percentage of heterozygote subjects in centenarians and controls (11.4 vs. 15.2). Therefore these data suggest that II class HLA DQ genes allele could not influencing life span expectation according to the other reports. On the other hand with the exception of IDDM and Celiac Disease most of studies on association among HLA class II loci and disease have individuated the HLA-DR and not -DQ loci as the major susceptibility component for ageing associated diseases.