Association of allergic diseases and asthma with colorectal cancer: a Mendelian randomization study

Abstract

Background As observational studies showed inconsistent findings, the association of the different phenotypes of allergic diseases and asthma with colorectal cancer (CRC) remains controversial. This might be attributed to confounding and using Mendelian randomization (MR) design can resolve the issue of confounding. Methods a MR study was performed using genetic variants as instrumental variables to estimate the causal effect of genetically proxied liability to the different phenotypes of allergic diseases and asthma on CRC risk. We obtained summary statistics from genome-wide association studies of allergic diseases (including 360,838 individuals [n =136 variants]), different phenotypes of asthma (recruiting 457,822 subjects [n = 595 variants]), and CRC using 98,715 participants. Findings The inverse-variance weighted (IVW) model suggested an inverse association of genetically proxied liability to asthma, hay fever, or eczema with CRC (OR, 0.88; 95% CI, 0.82–0.94; P = 1.39E-13). This inverse association was supported by the other sensitivity analyses (MR–Egger regression: OR, 0.82; 95% CI, 0.69 –0.99; P = 4.36E-02; weighted median: OR, 0.92; 95% CI, 0.85–0.99; P = 2.15E-02). The IVW model showed weak evidence of association between genetically proxied liability to the different phenotypes of asthma (adulthood-onset, childhood-onset, non-allergic, and allergic asthma) and CRC. Interpretation Genetically proxied liability to allergic diseases (asthma, hay fever, or eczema) were inversely associated with CRC. However, assessing genetically proxied liability to allergic asthma alone with CRC revealed weak evidence of association. Future MR studies assessing hay fever or eczema with CRC can help clarify what is driving the association.