Abstract
Compared with the well-documented inverse comorbidity of common neurodegenerative diseases with some types of cancer, the association of amyotrophic lateral sclerosis (ALS) with cancer was obscure. This Mendelian randomization (MR) analysis was aimed to appraise the causal relationship of ALS with cancer. Leveraging summary statistics of genome-wide association study (GWAS) for ALS, overall-cancer and nine types of site-specific common solid cancer including colorectal cancer (CRC) in populations of European (discovery and replication) and East Asian ancestry, we investigated the causal association of ALS with cancer using inverse-variance weighted (IVW) method and complementary sensitivity analyses. We performed MR-based mediation analysis to assess possible role of serum lipid traits such as hyperlipidemia, one shared risk factor of ALS and CRC, in their causal association. We found that genetically-predicted ALS was not associated with overall-cancer and other tested types of cancer, but was causally associated with reduced risk of CRC [odds ratio (OR) 0.84; 95% confidence interval (95% CI) 0.74-0.96; p = 0.011, OR 0.32; 95% CI 0.15-0.69; p = 0.004, in datasets of European (discovery) and East Asian ancestry respectively]. We observed absences of directional pleiotropy (MR Egger, intercept = -0.02 and -0.02, p = 0.49 and 0.60), instrumental outlier (MR-PRESSO, p = 0.95 and 0.84) or heterogeneity (Cochran Q, p = 0.95 and 0.82). Null reverse causality of CRC with ALS was found in either datasets. However, we found no evidence of the inverse association of ALS with CRC in either the replication (OR 0.99; 95% CI 0.93-1.06) or the combined European datasets (OR 0.92; 95% CI 0.79-1.09). Mediation analysis in European datasets suggested that hyperlipidemia may affect the risk of CRC via ALS in an indirect manner, with the measured mediation effect of hyperlipidemia on CRC being -0.02 (95% CI -0.04 to -0.002, p = 0.03). Our two-sample MR study in trans-ethnic populations uncovered that genetically proxied ALS may be causally associated with reduced risk of CRC, providing new insight into the inverse comorbidity and etiological causality of neurodegenerative disease and cancer. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s43657-024-00159-9.