Noninherited maternal antigens do not increase the susceptibility for familial rheumatoid arthritis

Objective. It has been proposed that noninherited maternal HLA-DR antigens (NIMA) might play a role in the susceptibility for rheumatoid arthritis (RA ). This hypothesis has not been thoroughly tested in patients with familial RA, in whom genetic factors, either inherited or not, might have stronger influence than in patients with sporadic RA. We investigated the NIMA hypot hesis in a large cohort of European patients with familial RA. Methods. The distribution of NIMA, noninherited paternal antigens (NIPA), a nd inherited HLA-DR antigens was assessed in patients with familial RA from all family sets collected from 1996 onwards by the ECRAF. HLA-DRB1 oligoty ping from patients and all available nonaffected siblings and parents was c arried out. Familial RA was defined by the presence of at least 2 affected first-degree relatives in the same family. The frequencies of HLA-DR MMA an d NIPA were compared using odds ratios after stratification for HLA-DR*04, *0401, and/or *0404 and shared epitope (SE) status. NIMA/NIPA that coincide d with inherited parental HLA-DR antigens were considered redundant and wer e excluded from analysis. Results. NIMA and NIPA could be analyzed in 165 RA patients with familial R A and 84 nonaffected siblings. Patients were predominantly female, rheumato id factor positive, and had erosive disease (81, 75, and 84%, respectively) . Possession of HLA-DR*04 and *0401/*0404 alleles tended be more frequent i n patients than in nonaffected siblings but this did not reach statistical significance. SE possession was similar in patients and healthy siblings, a lthough the former had a double dose SE mon often (37.6 vs 17.8%; p = 0.002 ). Transmission of SE encoding alleles from parents to offspring was skewed only in patients [OK (95% CI) 3.56 (2.55-4.95) vs 1.16 (0.75-1.79) in nona ffected siblings]. Using the NIPA as control, the frequencies of HLA-DRB1*0 4, *0401/*0404, and SE positive NIMA were not increased in patients lacking these susceptibility alleles. The frequencies of NIMA encoding susceptibil ity alleles in DR*04 and *0401/*0404 negative patients were lower than in n onaffected siblings. Conclusion. Our results corroborate the association between RA and inherite d SE allele and do not support a role for noninherited HLA-DR maternal anti gens in the susceptibility for familial RA.