HLA associations reveal genetic heterogeneity in psoriatic arthritis and in the psoriasis phenotype.

ABSTRACT

OBJECTIVE: Rigorously ascertained cases of psoriatic arthritis in subjects presenting to a rheumatology unit were compared with cases of psoriasis in subjects presenting to a dermatology unit, where subjects with musculoskeletal features were excluded, to address 1) the extent to which the contribution of the major histocompatibility complex (MHC) to psoriatic arthritis susceptibility resembles that in psoriasis, and 2) whether MHC genes determine quantitative traits within the psoriatic arthritis phenotype. METHODS: Separate discovery and validation subcohorts of patients recruited from a relatively homogeneous population were studied by sequence-based HLA typing, in which frequencies of the HLA-B and HLA-C alleles and haplotypes were compared. RESULTS: In patients with psoriatic arthritis, the frequency of C*0602 was lower than that in patients with psoriasis (28.7% versus 57.5%; P = 9.9 × 10(-12) ). Three haplotypes containing B*2705 or B*3901 were significantly increased in frequency in patients with psoriatic arthritis, but not in those with psoriasis. The structurally related B*3906 allele was not increased in frequency. B*27 was associated with an interval of 0.98 years between skin and musculoskeletal disease (P = 2.05 × 10(-6) ), compared with an interval of 10.14 years for C*06. Preliminary evidence suggested that B*3801 and B*08 may be associated with psoriatic arthritis susceptibility, and that allotypes encoding P2 pockets that bind side chains opposite in charge from those encoded by the B*27 and B*39 molecules may exert a protective role. CONCLUSION: These findings suggest that the psoriasis phenotype results from two patterns of MHC effect. The first involves the classic psoriasis susceptibility gene C*06, which confers more penetrant skin disease with less prevalent and more time-dependent musculoskeletal phenotype development. The second pattern appears to be mediated by HLA-B alleles, notably B*27, and includes temporally more coincident musculoskeletal involvement that is nearly equivalent in penetrance to that of the skin disease.