Autophagy-Related Gene-Set Signals in the Genetic Architecture of Obsessive-Compulsive Disorder: A...

Obsessive-compulsive disorder (OCD) remains difficult to treat in a substantial minority of patients, despite the established first-line use of exposure and response prevention, selective serotonin reuptake inhibitors, and pharmacological augmentation. This treatment gap has encouraged renewed interest in biological axes that sit outside a narrow monoamine-centered model. In this study, we asked whether transcriptional genetic signals for OCD show enrichment in cellular maintenance pathways motivated by nicotinamide mononucleotide (NMN)-associated aging biology and whether any such pathways outperform a curated antidepressant-related gene comparator. Using OCD transcriptome-wide association study (TWAS)-derived gene-level Z statistics and an absolute Z-score ranking framework, ten NMN-motivated Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway expansions were compared against AntiDep_Genes. The strongest and most defensible finding in the OCD analysis was selective enrichment of KEGG_AUTOPHAGY_ANIMAL. Autophagy nominally outperformed the antidepressant comparator in the primary head-to-head test, with normalized enrichment score (NES) = 1.166, p = 0.041979, and difference in normalized enrichment score (dNES) = +0.2097. The signal persisted after removal of overlapping antidepressant-related genes, with cleaned NES = 1.1784 and p = 0.042479, and the autophagy set also showed a higher absolute Z-score distribution than AntiDep_Genes by Mann-Whitney U testing, p = 0.04717. Leading-edge overlap with the antidepressant comparator was minimal, limited to mechanistic target of rapamycin (MTOR). The OCD autophagy signal was driven by a mixture of core autophagy machinery and upstream signaling genes, including MAPK3, MAP1LC3A, PPP2CB, MAP2K1, GORASP2, BAD, ATG2A, ATG10, MAPK8, and CALCOCO2. This pattern suggests a signaling-execution module rather than a simple enrichment of canonical ATG genes alone. The result is biologically consistent with recent OCD genetic findings implicating cortical and hippocampal excitatory neurons and D1/D2 striatal medium spiny neurons, which are cell types with high synaptic and metabolic demands. These findings are best viewed as hypothesis-generating. They do not support NMN, nicotinamide adenine dinucleotide (NAD+) precursors, or autophagy modulators as OCD treatments, but they do identify autophagy and mitochondrial quality-control biology as plausible targets for future functional and stratification studies. The autophagy enrichment was nominally significant in the primary analysis but did not survive Benjamini-Hochberg false-discovery rate correction across the ten tested pathways (false discovery rate (FDR)-adjusted p = 0.41979); the findings should therefore be considered hypothesis-generating.