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.