Abstract
OBJECTIVE: Genome-wide association studies carried out in rheumatoid arthritis (RA) have led to the discovery of several genetic associations with this disease. Still, the current associated genetic variations can explain only part of the genetic risk involved in RA, and it is well recognized that these genome-wide association studies are likely underpowered to detect all common disease variants. This study was undertaken to explore the genomic regions showing low-significance associations in previous genome-wide association studies of RA. METHODS: To reduce the false-positive signal fraction, we exploited pathway analysis to prioritize regions containing genes most likely to be implicated in RA. We hypothesized that true disease genes would be in a similar pathway. Therefore, genes from similar pathways but located in different regions were prioritized for replication using Prioritizer software. A total of 384 genetic variants selected from previous RA genome-wide association studies were tested in a Spanish case-control discovery cohort comprising 376 RA patients and 478 healthy controls for replication. Statistically significant associations were further validated in replication cohorts from Spain and The Netherlands. The study consisted of a total of 1,818 RA patients and 2,498 controls. RESULTS: We detected a novel genetic association between RA and the MSRA gene (rs10903323) in the Spanish combined population (P = 2.91 x 10(-5) , odds ratio [OR] 1.51). This association was further tested in our independent Dutch replication cohort. Combined analysis showed an overall association of MSRA with RA (P = 3.19 x 10(-4) , OR 1.28). CONCLUSION: Our findings indicate that a novel association in the MSRA gene is related to oxidative stress and support the notion of a major role for this process in RA.
