Rheumatoid Arthritis (RA) is the most common chronic inflammatory joint disease. The overall prevalence is 1% and in people older than 60 it is more than 2%. RA has auto-immune features: auto-antibodies against the Fc part of IgG, so-called Rheumatoid Factor (RF) are found more often in RA patients and more recently RA-specific auto-antibodies directed against Cyclic Citrunilated Peptides (CCP) have been discovered. Based on twin studies the contribution of genetic factors to the pathogenesis has been estimated to be about 60%. The main genetic contribution (about 40%) comes from the HLA complex. An association between HLA-DR4 and RA was already documented almost 30 years ago. This association was more prominent for severe forms of the disease. Because more HLA-DRB1 alleles appeared to be associated with RA and the products of these alleles shared a 5AA sequence in a peptide-binding pocket the so-called Shared Epitope (SE) hypothesis was formulated, the prediction being that these DRB1 molecules would bind an RA inducing peptide(s). Thus far however such (a) peptides remain elusive. Because the risk for RA associated with different but SE-identical DRB1 alleles varies considerably this SE can also not be the whole explanation for the HLA contribution to RA susceptibility/severity. A modified SE has been postulated and a role for DQ has been postulated. There is also evidence for a contribution of non-class II genes to susceptibility. About 5 years ago we have reported that certain HLA-DRB1 alleles are associated with protection from (severe) RA. The products of these alleles carry instead of the SE sequence another peptide anchor region consisting of the amino acid DERAA. In a large prospective cohort study we showed recently that these alleles indeed confer (dominant) protection both against developing RA and a severe course of the disease. This protection was observed both in the presence and the absence of SE susceptibility alleles. We are presently exploring the hypothesis that this protection is mediated by regulatory T cells reactive with the DERAA epitope. An obvious way to unravel the apparently complex association between HLA and RA is to reduce the heterogeneity of this multifactorial disease. Recently, we have discovered that SE positive DRB1 alleles are exclusively associated with CCP positive RA. The previously reported association with RF positive RA appeared to be secondary to the association with anti-CCP pos. RA. This was the case both for the association found for susceptibility and severity. Interestingly, DRB1*03 was exclusively associated with anti-CCP neg. RA. Because recent evidence puts the immune response against citrunilated proteins (CCP) as prime suspect for disease induction and progression in this subgroup of RA these observations are a big leap forwards in solving the HLA-RA puzzle.