Supplementary Material for: Association Between Smoking Cessation and the Risk of Cholangiocarcinoma and Ampulla of Vater Cancer: A Nationwide Cohort Study

Introduction: The association between smoking cessation and intrahepatic and extrahepatic cholangiocarcinoma (iCCA and eCCA) risk is unclear. Furthermore, the association in individuals with preexisting risk factors is unknown. We aimed to investigate the association between smoking status (especially smoking cessation) and CCA risk according to individuals’ glycemic status. Methods: In this nationwide cohort study, 9,520,629 adults without cancer who underwent national health screening by the Korean National Health Insurance Service in 2009 were followed up through 2018. The hazard ratios (HRs) and 95% confidence intervals (CIs) for CCA were estimated after adjusting for potential confounders. Results: During the 78.3 person-years of follow-up, 16,236 individuals were newly diagnosed with CCA. Quitters had a significantly lower risk of iCCA and eCCA compared to current smokers in all glycemic status groups (all P<0.01). The HRs (95% CIs) for iCCA in current smokers and quitters were 1.33 (1.24–1.43) vs. 0.98 (0.90–1.06) in individuals with normoglycemia, 1.49 (1.37–1.63) vs. 1.17 (1.06–1.28) in individuals with prediabetes, and 2.15 (1.96–2.37) vs. 1.58 (1.42–1.75) in individuals with diabetes, compared to never-smokers with normoglycemia. Current smokers with diabetes or prediabetes had a synergistically increased risk of iCCA (all P<0.01). However, quitters with diabetes and prediabetes had an iCCA risk comparable to that of never-smokers. Analysis of eCCA yielded similar results. Smoking was not independently associated with the risk of the ampulla of Vater cancer. However, smoking combined with diabetes or prediabetes was associated with an increased risk of the ampulla of Vater cancer (all P<0.05). Discussion/Conclusion: Smoking cessation was associated with a reduced risk of CCA, despite the synergistically increased risk in current smokers with diabetes and prediabetes. Our findings suggest a crucial opportunity to reduce the risk of CCA. More individualized and intensive cancer prevention education is needed against CCA.