Genomic and transcriptomic insights into antipsychotic-induced changes in total cholesterol and body mass index in a multi-ancestry cohort of the US veterans

Polygenic architecture: cholesterol metabolism and lipid homeostasis loci for change in total cholesterol; BMI-related, protein homeostasis and synaptic function loci for clinically significant weight gain.

The authors report the largest and most ancestrally diverse genome-wide association study to date of longitudinal change in total cholesterol (ΔTC, n = 59,372) and body mass index (ΔBMI, n = 39,112) under antipsychotic treatment, in the Million Veteran Program across European, African American and Hispanic ancestries. Genome-wide significant and suggestive loci were identified for both traits, with stronger associations for ΔTC and for clinically significant weight gain (ΔBMI > 1.5 kg/m²) than for overall BMI change. Trans-ancestry meta-analysis highlighted shared suggestive loci, and functional annotation showed enrichment of protein homeostasis and synaptic function gene sets for marked weight gain. Summary-data-based Mendelian randomisation analyses further identified suggestive gene expression signals in whole blood and liver.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

The sample size and ancestral diversity are unprecedented for a pharmacological trait as hard to phenotype as antipsychotic-induced weight gain, and using a clinically meaningful threshold (ΔBMI > 1.5 kg/m²) visibly improves signal. This remains a preprint in which much of the findings are labelled suggestive, with no actionable variant, no validated risk score and no immediate consequence for antipsychotic selection. The value is aetiological today and potentially predictive tomorrow, provided the loci replicate in independent cohorts.

Analysis by Dr Thibaut Benquey