Pathway-Level TWAS in ALS Reveals Targetable Axes in Bioenergetics and Autophagy Regulation

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease in which most cases appear to reflect the combined effects of many common genetic variants rather than single highly penetrant mutations. Although genome-wide association studies have identified multiple ALS risk loci, the downstream biological pathways through which these variants act remain only partly understood. To address this gap, we carried out a multi-tissue transcriptome-wide association study by integrating large ALS genome-wide association summary statistics with S-PrediXcan models derived from 14 GTEx v8 tissues, including several brain regions, spinal cord, and whole blood. We then tested pathway-level enrichment with Stouffer’s combined Z-score meta-analysis and empirical permutation using curated gene sets related to mitochondrial oxidative phosphorylation, NAD metabolism, PI3K/AKT/mTOR signaling, apoptosis, GLP-1 and insulin secretion, and senescence.The clearest signal was a protective association for oxidative phosphorylation capacity, with the HALLMARK_OXIDATIVE_PHOSPHORYLATION set showing a Stouffer’s Z of -5.43 and a permutation p value of 0.033. By contrast, higher genetically predicted activity in the HALLMARK_PI3K_AKT_MTOR_SIGNALING set was associated with increased ALS risk, with a Stouffer’s Z of +7.43 and a permutation p value of 0.0034. NAD metabolism showed more modest positive enrichment, whereas GLP-1 and insulin secretion pathways were negatively enriched, suggesting a protective direction. Senescence-associated sets were essentially null. At the gene level, TBK1, NDUFA2, NMNAT3, and PTPRN emerged as prominent drivers of these pathway signals.Taken together, the results are consistent with a genetically influenced cycle in which reduced oxidative phosphorylation capacity contributes to mitochondrial stress and altered NAD handling, while insufficient autophagic buffering linked to TBK1/mTOR signaling and reduced metabolic support from the GLP-1 axis may help sustain injury. We also outline how polygenic expression risk scores built from these driver genes could be used in the future to stratify patients for trials aimed at mitochondrial or metabolic pathways. These findings do not establish direct causality, but they do connect common-variant genetics to a set of potentially actionable biological axes in sporadic ALS.