The preventable burden of endometrial and ovarian cancers in Australia: A pooled cohort study

Objective. Evidence on the endometrial and ovarian cancer burden preventable through modifications to current causal behavioural and hormonal exposures is limited. Whether the burden differs by population subgroup is unknown. Methods. We linked pooled data from six Australian cohort studies to national cancer and death registries, and quantified exposure-cancer associations using adjusted proportional hazards models. We estimated exposure prevalence from representative health surveys. We then calculated Population Attributable Fractions (PAFs) with 95% confidence intervals (Os), accounting for competing risk of death, and compared PAFs for population subgroups. Results. During a median 4.9 years follow-up, 510 incident endometrial and 303 ovarian cancers were diagnosed. Overweight and obesity explained 41.9% (95% CI 32.3-50.1) of the endometrial cancer burden and obesity alone 34.5% (95% CI 27.5-40.9). This translates to 12,800 and 10,500 endometrial cancers in Australia in the next 10 years, respectively. The body fatness-related endometrial cancer burden was highest (49-87%) among women with diabetes, living remotely, of older age, lower socio-economic status or educational attainment and born in Australia. Never use of oral contraceptives (OCs) explained 8.1% (95% CI 1.8-14.1) or 2500 endometrial cancers. A higher BMI and current long-term MHT use increased, and long-term OC use decreased, the risk of ovarian cancer, but the burden attributable to overweight, obesity or exogenous hormonal factors was not statistically significant. Conclusions. Excess body fatness, a trait that is of high and increasing prevalence globally, is responsible for a large proportion of the endometrial cancer burden, indicating the need for effective strategies to reduce adiposity. (C) 2019 Elsevier Inc All rights reserved.