Introduction
Ovarian cancer (OC) is the most lethal gynecologic cancer, which affects around 230,000 women and with 152,000 deaths worldwide each year. One of the main factors contributing to the high death-to-incidence rate of OC is the advanced stage of the disease at the time of diagnosis. [1], [2]. OC is classified into distinct histological subtypes, which differ in their origin, pathogenesis, molecular profile, risk factors, and clinical prognosis. [3]. An understanding of the epidemiology and etiology of OC based on the heterogeneity is critical for the development of prevention strategies.
Lifestyle-related risk factors are amenable to modification and may therefore be relevant targets in the prevention of OC. While many lifestyle factors have been associated with OC and its subtypes, such as education [4], coffee or tea consumption [5]–[8], dietary fat intake [9], physical activities [10]–[13], obesity [14], cigarette smoking and alcohol drinking [15]–[17], sleep duration and insomnia [18], [19], ascertaining causality and whether their modification will reduce the risk is undetermined. For example, education and obesity are closely interrelated, but their independent association with OC subtypes is uncertain. As well, smoking and coffee or tea consumption are overlapping behaviors, so they may introduce residual confounding to observational studies. Moreover, another challenge is that OC is caused by various pathologies, which have distinct pathophysiological characteristics. Most risk factors exhibited significant heterogeneity by histology, however, much of the current epidemiological data examining risk factor modification have not studied the relationships between modifiable risk factors and specific OC subtypes. A clear appraisal of the causality of these associations is of importance in updating the primary prevention strategy for OC and different histotypes.
Mendelian Randomization (MR) involves the use of genetic variants as instrumental variables in order to prove the causal effect of environmental exposure on a disease outcome. The MR estimates represent associations between genetically predicted levels of risk factors and outcomes, which makes MR estimates less likely to be affected by confounding factors than conventional observational epidemiology estimates [20], [21]. Additionally, since genetic codes are immune from environmental influences or preclinical disease, MR estimates are less prone to bias caused by reverse causation.
Herein, we conducted a comprehensive MR study to investigate the etiological role of multiple modifiable lifestyle factors on OC and its histologic subtypes.