Association between premorbid body mass index and amyotrophic lateral sclerosis: causal inference through genetic approaches

Abstract

Background: Inverse association between premorbid body mass index (BMI) and amyotrophic lateral sclerosis (ALS) has been discovered in observational studies; however, whether this association is causal remains largely unknown. Methods: We employed a two-sample Mendelian randomization approach to evaluate the causal relationship of genetically increased BMI with the risk of ALS. The analyses were implemented using summary statistics obtained for the independent instruments identified from large-scale genome-wide association studies of BMI (up to ~770,000 individuals) and ALS (up to ~81,000 individuals). The causal relationship between BMI and ALS was estimated using inverse-variance weighted methods and was further validated through extensive complementary and sensitivity analyses. Findings: Using 1,031 instruments strongly related to BMI, the causal effect of per one standard deviation increase of BMI was estimated to be 1.04 (95% CI 0.97~1.11, p=0.275) in the European population. The null association between BMI and ALS discovered in the European population also held in the East Asian population and was robust against various modeling assumptions and outlier biases. Additionally, the Egger-regression and MR-PRESSO ruled out the possibility of horizontal pleiotropic effects of instruments. Interpretation: Our results do not support the causal role of genetically increased or decreased BMI on the risk of ALS.