Systemic sclerosis (SSc) is characterized by the presence of autoantibodies, mostly IgG, which target a limited set of nuclear proteins. These antinuclear antibodies (ANA) associate with disease subgroups and specific organ involvement. Here we show that there is mutual exclusivity of individual ANA in 130 UK SSc patients, confirm clinical associations with antibody profile and extend the analysis to include genetic data. The ANA mutual exclusivity observed leads to the possibility that SSc, in these patients, is in fact three separate diseases. An alternative explanation for exclusivity relates to the fact that optimal production of IgG antibody requires T-cell help, a process restricted by the HLA class II presentation of antigen peptide. If each autoantibody has a different and tight MHC restriction, then there is a possibility that these groups arose from a common pathway and were modified by genetics into the mutually exclusive groups observed, making the separate disease theory less tenable. In order to answer this question, we have determined MHC class II restriction precisely using high-resolution HLA genotyping (SSP) coupled with an amino acid analysis program in our 130 UK SSc patients. DRB1*11 was associated with anti-topoisomerase-I antibody (ATA)-positive patients (P = 0.007) and when combined with ATA (RR = 15.82), dcSSc (RR = 11.45), or both (RR = 21.9), represented the strongest risk factor for pulmonary fibrosis. Patients with antibodies to RNA polymerases I, II and III were associated with DQB1*0201. At the amino acid level, 20 positions in DRB1 and 20 positions in DQB1 showed some significant correlation with an ANA group. Clearly, however, the linkages to MHC class II alleles are not nearly strong enough to explain the mutually exclusive nature of the autoantibody groups and our results support, but do not prove, the separate disease theory.