Replication of rubella vaccine population genetic studies: validation of HLA genotype and humoral response associations.

Purported genetic associations found in population studies require validation for confirmation. We previously reported rubella vaccine-induced immune responses and HLA associations in 346 adolescents, age 12-18 years (1st cohort), following two doses of a rubella-containing vaccine. We sought to replicate the associations discovered in that work by verifying these associations in a new cohort of 396 subjects, age 11-19 years (2nd cohort), all having had two doses of a rubella-containing vaccine. We found that B*2705 (median 1st cohort 20.9 IU/ml, p=0.028; 2nd cohort 20.5 IU/ml, p=0.001) and DPA1*0201 (median 1st cohort 32.5 IU/ml, p=0.048; 2nd cohort 25.8 IU/ml, p=0.025) alleles were consistently associated with lower rubella-induced antibodies. Further, DPB1*0401 (median 1st cohort 43.5 IU/ml, p=0.021; 2nd cohort 36.2 IU/ml, p=0.002) alleles were associated with higher antibody levels in both populations. The association of DRB1*04-DQB1*03-DPB1*03 (mean 1st cohort 25.2 IU/ml, p=0.011; 2nd cohort 21.4 IU/ml, p=0.032) and DRB1*15/16-DQB1*06-DPB1*03 (1st cohort 16.3 IU/ml, p=0.043; 2nd cohort 19.1 IU/ml, p=0.023) haplotypes with lower rubella-specific antibodies was observed in both studies. This study provides confirmatory evidence for an association between specific class I and II HLA markers and haplotypes with rubella vaccine-induced humoral responses and lends further weight to their influence on rubella immune responses.