Table 2.
Results of the replication studya
| Replication | Meta‐analysis (anti‐CCP–negative RA)b | ImmunoChip study (anti‐CCP–positive RA)c | Ratio of OR for anti‐CCP–positive RA to OR for anti‐CCP–negative RA from the ImmunoChip studyd | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Chr. | SNP ID | Gene | Selected | OR (95% CI) | P | OR | P | OR | P | Ratio | P |
| 5 | rs71624119 | ANKRD55 | BOTH | 0.85 (0.76–0.96) | 6.0 × 10−3 e | 0.80 | 2.8 × 10−13 | 0.84 | 1.7 × 10−11 | 1.07 | 8.2 × 10−2 |
| 6 | rs10440835 | PRL | IC | 1.14 (1.04–1.26) | 8.1 × 10−3 e | 1.13 | 2.1 × 10−6 | 1.05 | 2.5 × 10−2 | 0.93 | 2.8 × 10−2 |
| 5 | rs528092 | C5orf30 | IC | 0.92 (0.83–1.01) | 8.9 × 10−2 | 0.89 | 3.3 × 10−6 | 0.92 | 5.0 × 10−4 | 1.06 | 6.4 × 10−2 |
| 8 | rs4840565 | BLK | IC | 1.12 (1.01–1.24) | 2.9 × 10−2 | 1.13 | 7.0 × 10−6 | 1.09 | 3.4 × 10−4 | 0.97 | 0.33 |
| 16 | rs28698667 | IL27 | IC | 1.06 (0.97–1.17) | 0.21 | 1.11 | 6.3 × 10−5 | 1.02 | 0.32 | 0.91 | 5.0 × 10−3 |
| 21 | rs9979383 | RUNX1 | IC | 0.96 (0.88–1.06) | 0.47 | 0.91 | 2.0 × 10−4 | 0.91 | 3.2 × 10−5 | 1.02 | 0.55 |
| 18 | rs16955629 | RAB31 | IC | 1.03 (0.92–1.15) | 0.62 | 1.11 | 2.1 × 10−4 | 1.02 | 0.34 | 0.91 | 1.0 × 10−2 |
| 2 | rs10181656 | STAT4 | IC | 1.04 (0.93–1.16) | 0.50 | 1.11 | 2.2 × 10−4 | 1.14 | 2.4 × 10−7 | 1.00 | 0.90 |
| 2 | rs888427 | CYBRD1 | IC | 0.95 (0.86–1.05) | 0.34 | 1.08 | 2.0 × 10−3 | – | – | – | – |
| 1 | rs10489912 | NFIA | UK | 0.87 (0.79–0.96) | 3.8 × 10−3 e | 0.85 | 2.5 × 10−6 | 0.97 | 0.17 | 1.06 | 4.8 × 10−2 |
| 1 | rs2249707 | SLAMF9 | UK | 0.99 (0.90–1.10) | 0.89 | 0.90 | 6.5 × 10−3 | 1.02 | 0.30 | 1.09 | 5.6 × 10−3 |
| 17 | rs2689 | HNF1B | UK | 1.04 (0.94–1.14) | 0.47 | 0.93 | 3.3 × 10−2 | – | – | – | – |
| 4 | rs6533712 | CAMK2D | UK | 0.93 (0.85–1.02) | 0.12 | 1.05 | 0.12 | – | – | – | – |
| 11 | rs613587 | FLI1 | UK | 0.89 (0.80–0.99) | 3.2 × 10−2 | 1.05 | 0.15 | – | – | – | – |
After quality control procedures, the genotypes of 14 single‐nucleotide polymorphisms (SNPs) were analyzed in the replication phase in 1,024 anti–citrullinated protein autoantibody (anti‐CCP)–negative rheumatoid arthritis (RA) cases and 6,348 controls. Chr. = chromosome; OR = odds ratio; 95% CI = 95% confidence interval.
Meta‐analysis of discovery (ImmunoChip [IC]) and replication cohorts.
Meta‐analysis across the 6 cohorts of the ImmunoChip study of the association with anti‐CCP–positive RA.
A ratio of 1.0 indicates that the effect size is the same for anti‐CCP–positive RA and anti‐CCP–negative RA. The P value for the ratio indicates whether a genetic marker is significantly differentially associated between the 2 disease subsets. The comparison with anti‐CCP–positive RA was performed only if the direction of the effect size for anti‐CCP–negative RA was consistent between the discovery and replication studies.
Significant after Bonferroni correction (P < 0.01).