Purpose: Although smoking has not been associated with overall ovarian cancer risk, a different impact on various histotypes has been reported. Our aim is to provide an accurate, up-to-date estimate of the dose-risk relationships between cigarette smoking and epithelial ovarian cancer, overall and by histotypes.
Methods: Using an innovative approach for the identification of original study publications, we conducted a systematic review and meta-analysis of epidemiological studies published on the topic until September 2018. Summary relative risks (RR) for cigarette smoking were estimated using random-effects models; dose-risk relationships were evaluated using one-stage random-effects models with restricted cubic splines.
Results: Thirty-seven studies were considered in the meta-analysis. The summary RRs for current versus never smokers were 1.05 (95% confidence interval CI 0.95-1.16) for overall ovarian cancer, 1.78 (95% CI 1.52-2.07) for mucinous, 0.77 (95% CI 0.65-0.93) for clear cell, 0.81 (95% CI 0.73-0.91) for endometrioid, and 1.05 (95% CI 0.94; 1.17) for serous cancer. The risk of borderline mucinous (RR 2.09) and serous (RR 1.16) tumors was higher than for invasive cancers (RR 1.44 and 0.95, respectively). For mucinous cancer, risk was noticeably higher with smoking intensity and duration (RR 2.35 for 20 cigarettes/day, and 2.11 for 20 years of smoking). A non-significant linear relation was found with smoking intensity, duration, and time since quitting for overall ovarian cancer and other histotypes.
Conclusions: This uniquely large and comprehensive meta-analysis confirms that although cigarette smoking does not appear to be a risk factor for ovarian cancer, and it is even slightly protective for some rare histotypes, there is a strong dose-risk relationship with mucinous ovarian cancer.
Keywords: Dose–response relationship; Histotypes; Invasiveness; Meta-analysis; Ovarian cancer; Tobacco smoking.