Do SNPs in folate pharmacokinetic pathway alter levels of intracellular methotrexate polyglutamates and affect response? A prospective study in Indian patients.

This study investigated the impact of seven polymorphisms in genes of folate transport and (de)glutamation pathway on methotrexate polyglutamate levels and response in patients with rheumatoid arthritis. This prospective study included patients with rheumatoid arthritis. They were treated with methotrexate (up to 25 mg per week) for 24 weeks and categorized by EULAR response criteria into responders (good and moderate) and non-responders. Using real-time Taqman discrimination assay, SNPs were genotyped-rs1045642 (ABCB1 3435C>T), rs1128503 (ABCB1 1236C>T), rs10106 (FPGS 1994A>G), rs1544105 (FPGS G>A), rs11545078 (GGH 452C>T), rs3758149 (GGH -401C>T), and rs1051266 (RFC1 80G>A). RBC methotrexate polyglutamate(MTX-glu) levels were determined at 4, 8, 16, and 24 weeks using by reverse phase HPLC using C-18 column followed by post column photo-oxidation. This study included 117 patients with rheumatoid arthritis (M:F = 14:103). The mean dose of methotrexate at 24 weeks was 22.0 ± 4.0 mg, with data on DAS28(3) at 24 weeks available in 96 patients-61 responders and 35 non-responders. Minor allele of GGH 452C>T had an association with non-response (odds ratio 2.9, 95% CI 1.4-5.6) and assuming the dominance of C, the recessive genetic model found GGH 452C>T CC genotype (odds ratio 9.5, 95% CI 1.2 to 76.0) was significantly associated with response. However, there was no difference in MTX-glu levels among the various genotypes of this SNP (p = 0.9). Other SNPs were neither associated with response nor with alteration in methotrexate polyglutamate levels. On logistic regression, GGH 452C>T CC genotype and DAS28(3) at baseline were independent predictors of response. GGH 452C>T CC genotype was associated with response to methotrexate. None of the SNPs affected MTX-glulevels.