Risk factors for endometrial cancer: An umbrella review of the literature

Although many risk factors could have causal association with endometrial cancer, they are also prone to residual confounding or other biases which could lead to over- or underestimation. This umbrella review evaluates the strength and validity of evidence pertaining risk factors for endometrial cancer. Systematic reviews or meta-analyses of observational studies evaluating the association between non-genetic risk factors and risk of developing or dying from endometrial cancer were identified from inception to April 2018 using PubMed, the Cochrane database and manual reference screening. Evidence was graded strong, highly suggestive, suggestive or weak based on statistical significance of random-effects summary estimate, largest study included, number of cases, between-study heterogeneity, 95% prediction intervals, small study effects, excess significance bias and sensitivity analysis with credibility ceilings. We identified 171 meta-analyses investigating associations between 53 risk factors and endometrial cancer incidence and mortality. Risk factors were categorised: anthropometric indices, dietary intake, physical activity, medical conditions, hormonal therapy use, biochemical markers, gynaecological history and smoking. Of 127 meta-analyses including cohort studies, three associations were graded with strong evidence. Body mass index and waist-to-hip ratio were associated with increased cancer risk in premenopausal women (RR per 5 kg/m(2) 1.49; CI 1.39-1.61) and for total endometrial cancer (RR per 0.1unit 1.21; CI 1.13-1.29), respectively. Parity reduced risk of disease (RR 0.66, CI 0.60-0.74). Of many proposed risk factors, only three had strong association without hints of bias. Identification of genuine risk factors associated with endometrial cancer may assist in developing targeted prevention strategies for women at high risk.