Skip to main content
. 2018 May 9;8:7342. doi: 10.1038/s41598-018-25634-y

Table 3.

Metaanalysis of the association of the 24 SNPs with response to MTX according to the EULAR criteria at 6 months in the four populations of RA patients.

Locus SNP MA Fixed effects Heterogeneity Random effects
OR (95% CI) P I2 P OR (95% CI) P
Candidate gene
   ABCB1 rs1045642 A 0.95 (0.8–1.2) 0.6 22 0.3 0.97 (0.7–1.2) 0.8
   ABCC1 rs35592 C 1.19 (0.9–1.5) 0.2 61 0.05 1.24 (0.8–1.9) 0.3
   AMPD1 rs17602729 A 1.08 (0.8–1.5) 0.6 42 0.2 1.14 (0.7–1.8) 0.6
   ATIC rs4673990 G 0.73 (0.6–0.9) 0.009 0 0.5 0.73 (0.6–0.9) 0.009
rs12995526 A 0.85 (0.7–1.1) 0.1 52 0.10 0.86 (0.6–1.2) 0.4
rs16853834 T 0.91 (0.7–1.3) 0.6 0 0.5 0.91 (0.7–1.3) 0.6
rs2372536 G 0.72 (0.6–0.9) 0.009 8 0.4 0.72 (0.6–0.9) 0.014
   ITPA rs2295553 C 0.81 (0.6–1.02) 0.1 0 1.0 0.81 (0.6–1.02) 0.1
rs1127354 A 0.74 (0.5–1.1) 0.2 0 0.6 0.74 (0.5–1.1) 0.2
   MTRR rs1801394 A 1.39 (1.1–1.8) 0.004 0 0.6 1.39 (1.1–1.8) 0.004
   MTHFD1 rs2236225 A 1.03 (0.8–1.3) 0.8 0 0.5 1.03 (0.8–1.3) 0.8
   SLC46A1 rs2239907 T 0.96 (0.8–1.2) 0.7 0 0.8 0.96 (0.8–1.2) 0.7
   TYMS rs699517 T 1.03 (0.8–1.3) 0.8 0 0.6 1.03 (0.8–1.3) 0.8
GWAS hit
   ARL14/PPM1L rs7624766 G 1.13 (0.9–1.4) 0.3 47 0.13 1.15 (0.8–1.6) 0.4
   BMP2 rs2650972 T 1.28 (1.02–1.6) 0.035 0 0.8 1.28 (1.02–1.6) 0.035
   DHFR rs5836788 del 1.00 (0.8–1.3) 1.0 17 0.3 1.00 (0.8–1.3) 1.0
   10p15.1 rs1901633 G 0.88 (0.7–1.1) 0.3 0 0.9 0.88 (0.7–1.1) 0.3
   14q13.1 rs4982133 A 1.19 (0.9–1.6) 0.2 0 0.7 1.19 (0.9–1.6) 0.2
   15q26.2 rs1703794 C 1.12 (0.9–1.5) 0.4 5 0.4 1.12 (0.8–1.5) 0.4
   20q13 rs6064463 C 1.17 (0.9–1.5) 0.2 0 0.5 1.17 (0.9–1.5) 0.2
   MTRR rs162040 C 1.35 (0.98–1.9) 0.1 0 1.0 1.35 (0.98–1.9) 0.1
   PTPRM rs6506569 C 0.89 (0.7–1.1) 0.3 0 0.5 0.89 (0.7–1.1) 0.3
   TYMS rs2244500 G 0.89 (0.7–1.1) 0.4 0 0.9 0.89 (0.7–1.1) 0.4
rs2847153 A 1.02 (0.8–1.4) 0.9 0 0.7 1.02 (0.8–1.4) 0.9

Summary odds ratios (OR) and their confidence intervals (95% CI) obtained with fixed effects and random effects metaanalysis are presented together with the inconsistency (I2) and p value of the Cochran Q test for heterogeneity. Multivariate logistic regression in each population included sex, age, baseline DAS28, previous use of DMARD and concomitant treatment with corticosteroids.