Smoking and pancreatic cancer: a sex-specific analysis in the Multiethnic Cohort study

Abstract

Abstract Purpose To examine whether the detrimental smoking-related association with pancreatic cancer (PC) is the same for women as for men. Methods We analyzed data from 192,035 participants aged 45–75 years, enrolled in the Multiethnic Cohort study (MEC) in 1993–1996. We identified PC cases via linkage to the Hawaii and California Surveillance, Epidemiology, and End Results Program cancer registries through December 2017. Results During a mean follow-up of 19.2 years, we identified 1,936 incident PC cases. Women smokers smoked on average less than men smokers. In multivariate Cox regression models, as compared with sex-specific never smokers, current smokers had a similar elevated risk of PC for women, hazard ratio (HR) 1.49 (95% CI 1.24, 1.79) and as for men, HR 1.48 (95% CI 1.22, 1.79) (pheterogeneity: 0.79). Former smokers showed a decrease in risk of PC for men within 5 years, HR 0.74 (95% CI 0.57, 0.97) and for women within 10 years after quitting, HR 0.70 (95% CI 0.50, 0.96), compared with their sex-specific current smokers. Both sexes showed a consistent, strong, positive dose–response association with PC for the four measures (age at initiation, duration, number of cigarettes per day, number of pack-years) of smoking exposure among current smokers and an inverse association for years of quitting and age at smoking cessation among former smokers (all ptrend’s < 0.001). Conclusion Although MEC women smoke on average less than their men counterparts, the smoking-related increase in PC risk and the benefits of cessation seem to be of similar magnitudes for women as for men.