A genome-wide association study follow-up suggests a possible role for PPARG in systemic sclerosis susceptibility

INTRODUCTION: A recent genome-wide association study (GWAS) comprising a French cohort of systemic sclerosis (SSc) reported several non-HLA single-nucleotide polymorphisms (SNPs) showing a nominal association in the discovery phase. We aimed to identify previously overlooked susceptibility variants by using a follow-up strategy. METHODS: Sixty-six non-HLA SNPs showing a P value <10(-4) in the discovery phase of the French SSc GWAS were analyzed in the first step of this study, performing a meta-analysis that combined data from the two published SSc GWASs. A total of 2,921 SSc patients and 6,963 healthy controls were included in this first phase. Two SNPs, PPARG rs310746 and CHRNA9 rs6832151, were selected for genotyping in the replication cohort (1,068 SSc patients and 6,762 healthy controls) based on the results of the first step. Genotyping was performed by using TaqMan SNP genotyping assays. RESULTS: We observed nominal associations for both PPARG rs310746 (P(MH) = 1.90 × 10(-6), OR, 1.28) and CHRNA9 rs6832151 (P(MH) = 4.30 × 10(-6), OR, 1.17) genetic variants with SSc in the first step of our study. In the replication phase, we observed a trend of association for PPARG rs310746 (P value = 0.066; OR, 1.17). The combined overall Mantel-Haenszel meta-analysis of all the cohorts included in the present study revealed that PPARG rs310746 remained associated with SSc with a nominal non-genome-wide significant P value (P(MH) = 5.00 × 10(-7); OR, 1.25). No evidence of association was observed for CHRNA9 rs6832151 either in the replication phase or in the overall pooled analysis. CONCLUSION: Our results suggest a role of PPARG gene in the development of SSc.