Identification of potential mediators of the relationship between body mass index and colorectal cancer: a Mendelian randomization analysis

Abstract

Background: Colorectal cancer (CRC) is the third-most-common cancer worldwide and its rates are increasing. Elevated body mass index (BMI) is an established risk factor for CRC, although the molecular mechanisms behind this association remain unclear. Using the Mendelian randomization (MR) framework, we aimed to investigate the mediating effects of putative biomarkers and other CRC risk factors in the association between BMI and CRC.

Methods: We selected as mediators biomarkers of established cancer-related mechanisms and other CRC risk factors for which a plausible association with obesity exists, such as inflammatory biomarkers, glucose homeostasis traits, lipids, adipokines, insulin-like growth factor 1 (IGF1), sex hormones, 25-hydroxy-vitamin D, smoking, physical activity (PA) and alcohol consumption. We used inverse-variance weighted MR in the main univariable analyses and performed sensitivity analyses (weighted-median, MR-Egger, Contamination Mixture). We used multivariable MR for the mediation analyses.

Results: Genetically predicted BMI was positively associated with CRC risk [odds ratio per SD (5 kg/m2) = 1.17, 95% CI: 1.08-1.24, P-value = 1.4 × 10-5] and robustly associated with nearly all potential mediators. Genetically predicted IGF1, fasting insulin, low-density lipoprotein cholesterol, smoking, PA and alcohol were associated with CRC risk. Evidence for attenuation was found for IGF1 [explained 7% (95% CI: 2-13%) of the association], smoking (31%, 4-57%) and PA (7%, 2-11%). There was little evidence for pleiotropy, although smoking was bidirectionally associated with BMI and instruments were weak for PA.

Conclusions: The effect of BMI on CRC risk is possibly partly mediated through plasma IGF1, whereas the attenuation of the BMI-CRC association by smoking and PA may reflect confounding and shared underlying mechanisms rather than mediation.

Keywords: BMI; Body mass index; CRC; Mendelian randomization; colorectal cancer; mediation analysis; obesity.

Figure 1.

To obtain valid causal estimates for mediation, the three typical Mendelian randomization (MR) assumptions should be satisfied in the context of multivariable MR. These are: (i) relevance—the single-nucleotide polymorphisms (SNPs) (G) should be strongly associated with body mass index (BMI) [given the putative mediator(s) included in the model]; (ii) exchangeability—the SNPs (G) should be independent of all confounders (U) of any of the exposures (in the model) and colorectal cancer (CRC) and (iii) exclusion restriction—there should be no other path (P) through which the SNPs affect CRC but via BMI [or the putative mediator(s) in the model]

Figure 2.

Mendelian randomization (MR) mediation analysis of genetically predicted body mass index (BMI) on colorectal cancer (CRC) risk. The forest plot (left, top and middle) presents the MR inverse-variance weighted (IVW) estimate (OR, 95% CI) of BMI on CRC risk in the MVMR model, adjusted for one putative mediator at a time (direct effects). Where there was a robust BMI-to-putative-mediator and mediator-to-CRC-risk association, the percent attenuation and 95% CI were estimated and are presented in labels. The forest plot on the right side displays the marginal associations of BMI on putative mediators (the 95% CI of the MR–IVW using a thick grey line) and mediators on CRC risk (the MR–IVW beta and 95% CI). The check marks in the first column (‘Robustness’) indicate robust marginal associations [i.e. significant associations in the IVW analysis (P < 0.05) that were qualitatively consistent in sensitivity analyses], in the second column [‘Instrument strength (MVMR)’] indicate conditional F-statistics of >10 [for both predicted body mass index (BMI) and the putative mediator in the model] and in the third column (‘Egger P-int.>0.1’) indicate that the P-value of the intercept in the MV MR–Egger models was >0.1. The forest plot at the bottom displays the total effects (OR, 95% CI) of BMI on CRC risk in univariable MR–IVW models

Figure 3.

Mendelian randomization (MR) mediation of genetically predicted body mass index (BMI) on colorectal cancer (CRC) subtypes. The forest plot shows the OR (95% CI) per SD of genetically predicted BMI (5 kg/m²), on CRC subtype in the MR inverse-variance weighted (IVW) models (total effects) and in sensitivity analyses. Where there was a robust BMI-to-putative-mediator and mediator-to-CRC-risk association, the percent attenuation and 95% CI were estimated and are presented in labels