Metabolomic signature predicting ejection fraction recovery in recent-onset dilated cardiomyopathy

Background Recent-onset dilated cardiomyopathy (RODCM) represents a critical therapeutic window where guideline-directed medical therapy (GDMT) may induce left ventricular reverse remodeling. However, metabolic signatures distinguishing treatment responders from non-responders remain poorly characterized. Aim To identify differential plasma metabolites between responders (EF≥50%) versus non-responders (EF<50%) at 6-month follow-up using targeted metabolomics. Methods We conducted a prospective, single-center cohort study of RODCM patients managed with standardized GDMT between 2021-2024 (diagnosed ≤6 months, left ventricular ejection fraction [LVEF] <45%). Fasting plasma samples were collected at baseline from 48 RODCM patients and 48 age-, sex-, BMI-matched healthy controls, Additionally, plasma samples were also collected at 6-month follow-up from 46 patients who completed the 6-month follow-up visit. Targeted LC-MS/MS (MRM mode) was performed on a QTRAP 6500+ system, detecting 236 metabolites across >40 pathways. After QC screening (CV<20%), 119 stable metabolites underwent PCA/OPLS-DA analysis. Differential metabolites were selected by VIP>1.0 and p<0.05 (FDR-corrected), followed by KEGG pathway enrichment. Results A total of 46 RODCM patients and 46 matched controls were included in the metabolomic analysis at the baseline and follow-up. After 6 months of GDMT, 17 patients (37%) achieved EF recovery to ≥50% and were classified as responders group. The age (48.4 vs. 49.0 years), gender (male 51.9% vs. 47.1%) and baseline LVEF (32.8% vs. 30.3%, all p>0.05) were comparable between the responders and non-responders group. OPLS-DA demonstrated clear separation between responder and non-responder groups (R²Y=0.484, Q²=-0.172). We identified 10 differentially abundant metabolites, predominantly amino acids (L-leucine, L-isoleucine, L-tryptophan, L-methionine, L-lysine) and indoles (indole, indole-3-acetate). Branched-chain amino acids and aromatic amino acids were significantly elevated in the non-responder group (figure 1). Hierarchical clustering revealed distinct metabolite profiles separating the two phenotypes. Pathway analysis implicated valine/leucine/isoleucine biosynthesis, tryptophan metabolism, and protein digestion/absorption as significantly enriched pathways (p<0.05). Conclusions This metabolomic study reveals distinct amino acid and indole signatures that differentiate responders from non-responders after 6-month GDMT. The elevation of branched-chain amino acids and alterations in tryptophan metabolism in non-responders suggests potential metabolic mechanisms contributing to treatment response heterogeneity. These findings may provide novel biomarkers for early risk stratification and therapeutic targets in RODCM.