Opposing Ageing-Related Transcriptomic Profiles Distinguish Opioid Exposure Liability From Dependence Liability

Opioid use disorder (OUD) involves separable stages of exposure and dependence, but many genetic studies have treated these liabilities as a single phenotype. In this study, transcriptome-wide association study (TWAS) methods were applied to the Psychiatric Genomics Consortium Substance Use Disorders 2020 opioid meta-analysis summary statistics to examine ageing-related gene pathways across brain tissues. Genetic liability for opioid exposure was strongly associated with reduced predicted expression of complement components and synaptic-pruning genes, together with negative signals in astrocyte and senescence-related gene sets. By contrast, liability for dependence among opioid-exposed individuals showed positive enrichment in AMP-activated protein kinase-mechanistic target of rapamycin nutrient-sensing pathways, mitochondrial oxidative phosphorylation, especially DLD, and selected nicotinamide adenine dinucleotide/sirtuin pathways, accompanied by presynaptic and DNA repair-associated signals. Many ageing-related gene sets showed opposing directional profiles between exposure and dependence contrasts, indicating that these stages are associated with partially antagonistic transcriptomic profiles. Gene-level analyses identified influential drivers and directional sign-flips, including CASP7, TERF2, STX1A, and ADCY6, which further supported the stage-specific pattern. A dedicated post-hoc robustness inventory was performed, including threshold variation, family-wise false discovery rate pooling, tissue-specific decomposition, gene-set overlap analysis, major histocompatibility complex/complement flagging, and leave-one-out influence analysis. In this inventory, 131 of 4913 pooled statistical tests survived global Benjamini-Hochberg false discovery rate correction, and 100 of 102 evaluable pathway-profile entries satisfied an opposing exposure-versus-dependence criterion. These findings suggest that opioid exposure and opioid dependence are biologically distinguishable at the level of ageing-related brain gene expression. The results nominate stage-specific risk modules and biomarker hypotheses that may inform future precision approaches to OUD prevention and treatment while remaining hypothesis-generating until validated in functional and longitudinal studies.