Plasma Amino Acid Signatures Associated with Disease Progression and Hypertension in Autosomal Dominant Polycystic Kidney Disease: A Targeted Metabolomics and Machine Learning Approach.

Background: Autosomal dominant polycystic kidney disease (ADPKD) is a clinically heterogeneous disorder often leading to end-stage renal disease (ESRD). Prognostication of disease progression remains a major clinical challenge. This study aimed to identify plasma amino acid signatures associated with ADPKD progression and hypertension. Methods: We conducted targeted metabolomic analysis (LC-MS/MS) to quantify 38 plasma amino acids in 203 ADPKD patients, stratified by disease progression (rapid vs. slow) and hypertension status. Support Vector Machine (SVM) models were developed to predict outcomes using clinical data, amino acid profiles, and combined datasets. Results: Our findings revealed that specific amino acid signatures, including valine, glutamic acid, homocitrulline, and methylhistidines, were significantly elevated in both rapid progression and hypertensive groups. Isoleucine and citrulline were elevated only in rapid progressors. Phenylalanine, leucine, asparagine, and arginine were elevated in hypertensive patients. Machine learning analysis showed that integrating clinical and metabolic data modestly improved prediction for progression and hypertension. Proteinuria, glomerular filtration rate (GFR), and uric acid were the top clinical predictors; however, adding arginine, isoleucine, and 3-methylhistidine further enhanced prediction accuracy. Pathway analysis showed shared dysregulation in arginine biosynthesis and branched-chain amino acid (BCAA) metabolism. Specific amino acids were positively correlated with creatinine and uric acid and negatively correlated with GFR, and elevated levels of these metabolites were associated with increased mortality risk in survival analysis. Conclusions: Our results suggest that these plasma amino acid signatures, when combined with clinical markers, may serve as potential biomarkers for early risk stratification and precision prediction of ADPKD progression and hypertension.