Cigarette smoking and endometrial carcinoma risk: The role of effect modification and tumor heterogeneity

Purpose: The inverse relationship between cigarette smoking and endometrial carcinoma risk is well established. We examined effect modification of this relationship and associations with tumor characteristics in the National Institutes of Health-AARP Diet and Health Study. Methods: We examined the association between cigarette smoking and endometrial carcinoma risk among 110,304 women. During 1,029,041 person years of follow-up, we identified 1,476 incident endometrial carcinoma cases. Multivariable Cox proportional hazards regression models were used to estimate relative risks (RRs) and 95 % confidence intervals (CIs) for the association between smoking status, years since smoking cessation, and endometrial carcinoma risk overall and within strata of endometrial carcinoma risk factors. Effect modification was assessed using likelihood ratio test statistics. Smoking associations by histologic subtype/grade and stage at diagnosis were also evaluated. Results: Reduced endometrial carcinoma risk was evident among former (RR 0.89, 95 % CI 0.80, 1.00) and current (RR 0.65, 95 % CI 0.55, 0.78) smokers compared with never smokers. Smoking cessation 1-4 years prior to baseline was significantly associated with endometrial carcinoma risk (RR 0.65, 95 % CI 0.48, 0.89), while cessation ≥10 years before baseline was not. The association between smoking and endometrial carcinoma risk was not significantly modified by any endometrial carcinoma risk factor, nor did we observe major differences in risk associations by tumor characteristics. Conclusion: The cigarette smoking-endometrial carcinoma risk relationship was consistent within strata of important endometrial carcinoma risk factors and by clinically relevant tumor characteristics.