Identifying molecular mediators of the relationship between body mass index and endometrial cancer risk: a Mendelian randomization analysis

Background Endometrial cancer is the most common gynaecological cancer in high-income countries. Elevated body mass index (BMI) is an established modifiable risk factor for this condition and is estimated to confer a larger effect on endometrial cancer risk than any other cancer site. However, the molecular mechanisms underpinning this association remain unclear. We used Mendelian randomization (MR) to evaluate the causal role of 14 molecular risk factors (hormonal, metabolic and inflammatory markers) in endometrial cancer risk. We then evaluated and quantified the potential mediating role of these molecular traits in the relationship between BMI and endometrial cancer using multivariable MR. Methods Genetic instruments to proxy 14 molecular risk factors and BMI were constructed by identifying single-nucleotide polymorphisms (SNPs) reliably associated (P < 5.0 × 10−8) with each respective risk factor in previous genome-wide association studies (GWAS). Summary statistics for the association of these SNPs with overall and subtype-specific endometrial cancer risk (12,906 cases and 108,979 controls) were obtained from a GWAS meta-analysis of the Endometrial Cancer Association Consortium (ECAC), Epidemiology of Endometrial Cancer Consortium (E2C2) and UK Biobank. SNPs were combined into multi-allelic models and odds ratios (ORs) and 95% confidence intervals (95% CIs) were generated using inverse-variance weighted random-effects models. The mediating roles of the molecular risk factors in the relationship between BMI and endometrial cancer were then estimated using multivariable MR. Results In MR analyses, there was strong evidence that BMI (OR per standard deviation (SD) increase 1.88, 95% CI 1.69 to 2.09, P = 3.87 × 10−31), total testosterone (OR per inverse-normal transformed nmol/L increase 1.64, 95% CI 1.43 to 1.88, P = 1.71 × 10−12), bioavailable testosterone (OR per natural log transformed nmol/L increase: 1.46, 95% CI 1.29 to 1.65, P = 3.48 × 10−9), fasting insulin (OR per natural log transformed pmol/L increase: 3.93, 95% CI 2.29 to 6.74, P = 7.18 × 10−7) and sex hormone-binding globulin (SHBG, OR per inverse-normal transformed nmol/L increase 0.71, 95% CI 0.59 to 0.85, P = 2.07 × 10−4) had a causal effect on endometrial cancer risk. Additionally, there was suggestive evidence that total serum cholesterol (OR per mg/dL increase 0.90, 95% CI 0.81 to 1.00, P = 4.01 × 10−2) had an effect on endometrial cancer risk. In mediation analysis, we found evidence for a mediating role of fasting insulin (19% total effect mediated, 95% CI 5 to 34%, P = 9.17 × 10−3), bioavailable testosterone (15% mediated, 95% CI 10 to 20%, P = 1.43 × 10−8) and SHBG (7% mediated, 95% CI 1 to 12%, P = 1.81 × 10−2) in the relationship between BMI and endometrial cancer risk. Conclusions Our comprehensive MR analysis provides insight into potential causal mechanisms linking BMI with endometrial cancer risk and suggests targeting of insulinemic and hormonal traits as a potential strategy for the prevention of endometrial cancer. Supplementary Information The online version contains supplementary material available at 10.1186/s12916-022-02322-3.


INTRODUCTION 2
Background Explain the scientific background and rationale for the reported study. What is the exposure? Is a potential causal relationship between exposure and outcome plausible? Justify why MR is a helpful method to address the study question 3 Objectives State specific objectives clearly, including pre-specified causal hypotheses (if any). State that MR is a method that, under specific assumptions, intends to estimate causal effects

Study design and data sources
Present key elements of the study design early in the article. Consider including a table listing sources of data for all phases of the study. For each data source contributing to the analysis, describe the following: a) Setting: Describe the study design and the underlying population, if possible. Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data collection, when available. b) Participants: Give the eligibility criteria, and the sources and methods of selection of participants. Report the sample size, and whether any power or sample size calculations were carried out prior to the main analysis c) Describe measurement, quality control and selection of genetic variants d) For each exposure, outcome, and other relevant variables, describe methods of assessment and diagnostic criteria for diseases e) Provide details of ethics committee approval and participant informed consent, if relevant 5

Assumptions
Explicitly state the three core IV assumptions for the main analysis (relevance, independence and exclusion restriction) as well assumptions for any additional or sensitivity analysis 6

Statistical methods: main analysis
Describe statistical methods and statistics used 1 and 3 Title: "Identifying molecular mediators of the relationship between body mass index and endometrial cancer risk: a Mendelian randomization analysis" Abstract: "We used Mendelian randomization (MR) to evaluate the causal role of 14 molecular risk factors (hormonal, metabolic, and inflammatory markers) in endometrial cancer risk." "we used a two-sample MR approach to evaluate the causal role of 14 endogenous sex hormones, metabolic traits, and inflammatory markers in endometrial cancer risk (overall and in endometrioid and non-endometrioid subtypes). We then used multivariable MR to evaluate and quantify the mediating role of these molecular traits in the relationship between BMI and endometrial cancer risk." 6 "Observational epidemiological studies have reported associations between several hormonal, metabolic, and inflammatory factors linked to obesity and endometrial cancer, including bioavailable testosterone, sex hormone-binding globulin (SHBG), oestradiol and fasting insulin" "many previously reported molecular risk factors for endometrial cancer from conventional observational studies remain untested in an MR framework, meaning the causal relevance of these factors in disease onset is unclear" 5 and 6 7 (also table 1) The meta-GWAS "combined 17 previously reported studies from the Endometrial Cancer Association Consortium (ECAC), the Epidemiology of Endometrial Cancer Consortium (E2C2), and UK Biobank, with four studies contributing samples to more than one genotyping project. Participants were recruited from Australia, Belgium, Germany, Poland, Sweden, the UK, and the USA" 9 "we obtained single-nucleotide polymorphisms (SNPs) reliably (P < 5 x 10-8) and independently (r2 < 0.001) associated with each trait. To construct a genetic instrument for leptin, we restricted genetic variants to cis-acting SNPs." Table 1 "All studies contributing data to these analyses had the relevant institutional review board approval from each country, in accordance with the Declaration of Helsinki, and all participants provided informed consent." 27 8 and Fig. 2 "MR analysis can generate unbiased estimates of causal effects of risk factors on disease outcomes if the following assumptions are met: (i) the instrument strongly associates with the exposure ("relevance"), (ii) there is no confounding of the instrument-outcome relationship ("exchangeability"), and (iii) the instrument only affects the outcome through the exposure ("exclusion restriction")." N/A a) Describe how quantitative variables were handled in the analyses (i.e., scale, units, model) b) Describe how genetic variants were handled in the analyses and, if applicable, how their weights were selected c) Describe the MR estimator (e.g. two-stage least squares, Wald ratio) and related statistics. Detail the included covariates and, in case of two-sample MR, whether the same covariate set was used for adjustment in the two samples d) Explain how missing data were addressed e) If applicable, indicate how multiple testing was addressed 7

Assessment of assumptions
Describe any methods or prior knowledge used to assess the assumptions or justify their validity 8

Sensitivity analyses and additional analyses
Describe any sensitivity analyses or additional analyses performed (e.g. comparison of effect estimates from different approaches, independent replication, bias analytic techniques, validation of instruments, simulations) 9 Software and preregistration a) Name statistical software and package(s), including version and settings used b) State whether the study protocol and details were pre-registered (as well as when and where) i. Provide justification of the similarity of the genetic variant-exposure associations between the exposure and outcome samples "per SD (4.7 kg/m2) increase in BMI", "per increase in inverse-normal transformed (INT) nmol/L total testosterone", "per increase in natural log transformed nmol/L bioavailable testosterone", "per increase in natural log transformed pmol/L fasting insulin", "per increase in INT nmol/L SHBG" etc 14 "For traits instrumented by a single SNP, the Wald ratio was used to generate effect estimates and the delta method was used to approximate standard errors [46]. For traits instrumented by two or more SNPs, inverse-variance weighted (IVW) random-effects models were used to estimate causal effects [46]." 9 "For all traits where instruments consisted of SNPs in weak LD (i.e. leptin, IL-6 and CRP), standard errors for causal estimates were inflated to account for correlation between SNPs with reference to the 1000 Genomes Phase 3 reference panel [32,45]." 9 "A Bonferroni correction was applied as a heuristic to account for multiple testing in MR analyses for the 15 risk factors (14 molecular traits and BMI) investigated." 9 "we re-calculated causal estimates obtained from IVW models using MR-Egger regression, weighted median estimation, and weighted mode estimation" "we performed "leave-one-out" analyses for all findings showing strong or suggestive evidence of effects in IVW models" 10 9 N/A "As a sensitivity analysis we also re-performed MR analyses using sex-specific instruments where possible." "Steiger filtering was performed across all analyses to identify and subsequently remove any SNPs which explained more variance in the outcome than the exposure"

Key results
Summarize key results with reference to study objectives 15

Limitations
Discuss limitations of the study, taking into account the validity of the IV assumptions, other sources of potential bias, and imprecision. Discuss both direction and magnitude of any potential bias and any efforts to address them S6 Table   Table 1 Tables The direction of effect was inconsistent when examining the effect of BMI on total testosterone using a weighted mode model, suggesting the potential presence of horizontal pleiotropy." E.g. "in the female-specific BMI sensitivity analysis there was strong evidence for a mediating role of female-specific SHBG in the relationship between BMI and endometrioid endometrial cancer (8% mediated, 95% CI: 3 to 13%, P = 3.38 x 10-3)."

14-16
N/A 19 "Our findings supporting a causal effect of BMI on endometrial cancer risk … are larger in magnitude than those from pooled analyses of conventional observational analyses (e.g. the WCRF pooled analysis of 26 prospective studies" Supplementary figures 8, 11-17, 21-29 "Our systematic MR analysis … provided evidence for roles of elevated BMI, fasting insulin, total and bioavailable testosterone, and SHBG in risk of overall and endometrioid endometrial cancer. In mediation analyses, we found evidence that fasting insulin, bioavailable testosterone concentrations, and SHBG partially mediated the effect of BMI on overall endometrial cancer risk."

19
"There are several limitations to our analysis. First, we were unable to evaluate the role of six previously reported molecular risk factors for endometrial cancer due to the absence of reliable genetic instruments for these traits.

Data and data sharing
Provide the data used to perform all analyses or report where and how the data can be accessed, and reference these sources in the article. Provide the statistical code needed to reproduce the results in the article, or report whether the code is publicly accessible and if so, where 20

Conflicts of Interest
All authors should declare all potential conflicts of interest E.g. "important mediating roles of fasting insulin, bioavailable testosterone, and SHBG in the relationship between BMI and endometrial cancer are consistent with studies of bariatric surgery which have suggested protective effects of this procedure against endometrial cancer risk, along with reductions in insulin and bioavailable testosterone levels, and increases in SHBG levels." 20 "Potential aetiological roles of the molecular mediators identified in this analysis are consistent with the "unopposed oestrogen" hypothesis which postulates that endometrial carcinogenesis is driven by excess endogenous or exogenous oestrogen levels that are unopposed by progesterone" 20 "Our findings suggest that use of such medications may confer a favourable secondary effect of reducing endometrial cancer risk among these high-risk groups." 25 " Another possible future direction for this work is to explore the effects of excess adiposity at different life stages, for instance, comparing pre-and post-menopausal BMI, in order to evaluate any potentially independent effects of excess adiposity on endometrial cancer risk across the life-course." 21 See Funding section 27 Table 1 27 "The authors declare that they have no competing interests."