Strong excess risk of pancreatic cancer for low frequency and duration of cigarette smoking: A comprehensive review and meta-analysis

Abstract Aim Cigarette smoking is an established risk factor for pancreatic cancer but an updated quantification of the association is lacking. Our aim is to provide the most accurate and updated estimate of the dose–response relationships between cigarette smoking and pancreatic cancer risk. Methods Using an innovative approach for the identification of original publications, we conducted a systematic review and meta-analysis of epidemiological studies published on the issue up to April 2017. Random effects models were used to provide pooled estimates for the cigarette smoking status; dose–risk relationships were evaluated using one-stage random effects models with restricted cubic splines. Results Seventy-eight studies were included, providing a pooled relative risk (RR) of pancreatic cancer of 1.8 (95% confidence interval, CI: 1.7–1.9) for the current and 1.2 (95% CI: 1.1–1.2) for the former vs. never smokers. A sharp increase in pancreatic cancer risk was found already with a low number of cigarettes and up to 30 cigarettes/day (RR 2.2, 95% CI: 1.9–2.4). Similarly, the risk of pancreatic cancer steady increased after a few years of smoking up to 30 years (RR 1.8, 95% CI: 1.6–2.0). The risk rapidly decreased with increasing time since quitting and was 0.6 (95% CI: 0.5–0.7, for the former vs. current smokers) after 20 years of quitting. Conclusions The present meta-analysis indicates that pancreatic cancer risk sharply increases with a low number of cigarettes or after a few years of smoking and that it rapidly decreases a few years after cessation, although it takes almost 20 years to reach that of never smokers.