Table 2

Fitted ORs (95% CIs) of NAT2 and combined NAT2-mEH genotypes for different pack-years

Pack-years of smoking
NAT2 alone
 R vs. Sa,b0.66 (0.44–0.99)1.02 (0.83–1.26)1.22 (0.89–1.67)1.40 (0.92–2.14)
NAT2 and mEH
 R + VL vs. S + VLb0.64 (0.40–1.01)1.07 (0.85–1.34)1.32 (0.94–1.86)1.56 (0.98–2.48)
 S + H vs. S + VLb0.47 (0.27–0.83)1.14 (0.85–1.54)1.65 (1.07–2.56)2.19 (1.22–3.95)
 R + H vs. S + VLb0.30 (0.14–0.62)1.22 (0.84–1.78)2.19 (1.26–3.81)3.42 (1.61–7.26)
 R + H vs. S + VLc0.35 (0.16–0.78)1.27 (0.85–1.91)2.16 (1.19–3.92)3.25 (1.45–7.30)
  • a R, NAT2 rapid acetylator genotypes; S, NAT2 slow acetylator genotypes; VL, very low mEH activity genotype; H, combined low, intermediate, and high mEH activity genotypes.

  • b Logistic regression model (final model) included variables for age, gender, SR-PY, smoking status (current, ex-, and nonsmokers), years since smoking cessation, genotypes (NAT2 alone or NAT2 and mEH), interaction terms between each genotype and SR-PY, and interactions between smoking status and SR-PY.

  • c The initial model (gene-gene interaction) included variables of age, gender, SR-PY, smoking status, years since smoking cessation, mEH genotype, NAT2 genotype, interaction between smoking status and SR-PY, all two-way interactions between NAT2, mEH, and SR-PY, and the three-way interaction between NAT2, mEH, and SR-PY. This model provides algebraically equivalent ORs to the gene-gene joint-effect model.