Major depressive disorder (MDD) is a complex psychiatric condition characterized by immune, oxidative, and metabolic (NIMETOX) dysregulation. Using the chronic unpredictable mild stress (CUMS) mouse model, we performed untargeted LC-MS/MS metabolomic profiling on serum samples from control, CUMS-vehicle, and five pharmacologically treated groups (fluoxetine, curcumin, simvastatin, SAMe, and PDTC). A total of 2,703 metabolites were identified. CUMS was associated with significant alterations in 166 metabolites (68 up, 98 down), with KEGG enrichment revealing perturbations in vitamin B6 metabolism, aromatic amino acid biosynthesis, the pentose phosphate pathway, and cytochrome P450-mediated metabolism. Curation and recursive feature selection yielded a high-confidence panel of 68 endogenous metabolites that showed adequate cross-validated discrimination between CUMS and control mice under repeated nested five-fold cross-validation (CV-AUC = 0.999–1.000). From this panel, eight functional NIMETOX modules (M1–M8) were constructed, capturing gut-brain axis, fatty acid oxidation, redox stress, eicosanoid inflammation, phospholipid remodeling, plasmalogens, gut-liver detoxification, and neuro-myelin integrity. All five interventions partially reversed the CUMS-induced metabolomic signature, with SAMe showing the broadest restoration across six modules. However, no treatment fully normalized the metabolome toward the control state. These findings demonstrate that CUMS induces a systemic metabolic phenotype encompassing key pathways predicted by the NIMETOX model of MDD. Module-based metabolomic profiling provides a systems-level framework for dissecting stress-related metabolic dysregulation, comparing the molecular effects of mechanistically distinct interventions.