Detection of Fast Decliner of Diabetic Kidney Disease Using Chiral Amino Acid Profiling: A Pilot Study

Biomarkers for the prediction of diabetic kidney disease are still unsatisfactory. Although D-amino acids have been shown to reflect kidney conditions, their efficacy in treating diabetic kidney disease (DKD) has not been demonstrated. This study explored the potential role of D-amino acids as progression markers for DKD, an aspect not addressed previously. We performed comprehensive D-amino acid measurements and collected the longitudinal estimated glomerular filtration rate (eGFR) data of 135 patients. We defined fast decliners (FDs) as patients exhibiting >10% decline from baseline eGFR per year and compared the D-amino acid levels of FDs and non-FDs. Then, we verified that D-amino acids could predict FDs independent of creatinine levels. In patients with diabetic kidney disease, D-serine, D-alanine, and D-proline were only detected in the blood, while 15 D-amino acids were detected in the urine. Using supervised orthogonal partial least squares analysis, blood D-serine and urine D-amino acid levels were identified as features characterizing diabetic kidney disease. Baseline blood D-serine levels and ratios did not differ between the FD and non-FD groups; however, short-term changes in blood D-serine levels differed. This study emphasized the significance of D-serine as a prognostic marker for DKD, an aspect not identified in previous research.