Genetic polymorphisms in PTPN22, PADI-4, and CTLA-4 and risk for rheumatoid arthritis in two longitudinal cohort studies: evidence of gene-environment interactions with heavy cigarette smoking

Table 7 PTPN22 genotype and HLA-SE interactions according to RF status in NHS/NHSII pooled samples

PTPN22	HLA-SE	All cases^a	RF-positive cases^b,c	RF-negative cases^b,d
CC	None	1.00 (ref.)	1.00 (ref.)	1.00 (ref.)
CC	Any	1.97 (1.39–2.78)	2.37 (1.62–3.46)	1.33 (0.89–1.99)
T carrier	None	1.41 (0.87–2.27)	1.76 (0.99–3.13)	1.40 (0.78–2.51)
T carrier	Any	2.73 (1.55–4.81)	3.58 (2.02–6.35)	1.47 (0.74–2.92)
Additive interaction		AP = 0.13 (95% CI -0.40 to +0.66); P_add^e = 0.62	AP = 0.13 (95% CI -0.41 to +0.67); P_add^e = 0.64^e	AP = -0.18^f
Multiplicative interaction		P_multi^g = 0.97^g	P_multi^g = 0.72	P_multi^g = 0.60

Values are expressed as odds ratio (95% confidence interval). ^aConditional logistic regression, adjusting for parity, pack-year smoking, breast-feeding, menstrual irregularity, and age at menarche, menopausal status and postmenopausal hormone use. ^bUnconditional logistic regression adjusting for year of birth, parity, pack-year smoking, breast-feeding, menstrual irregularity, and age at menarche, menopausal status and postmenopausal hormone use. ^cRF-positive RA cases and all controls. ^dRF-negative RA cases and all controls. ^eP for AP, one of the indices of additive interactions between binary PTPN22 genotype and binary HLA-SE genotype. ^f95% confidence interval and P value for AP not applicable because of the negative estimate (antagonistic rather than synergistic). ^gP for multiplicative interaction term between binary PTPN22 genotype and binary HLA-SE genotype with one degree of freedom. AP, attributable proportion; NHS, Nurses' Health Study; RA, rheumatoid arthritis; RF, rheumatoid factor.

ISSN: 1478-6362