#AI reads Urine# Serum and urine metabolic fingerprints enable diagnosis and prognosis for IgA nephropathy
Published 30 November, 2025
This study focuses on IgA nephropathy, a globally prevalent condition for which existing diagnostic methods have limitations—renal biopsy carries risks, while conventional indicators lack sufficient specificity. It develops a dual biofluid (serum and urine) metabolic analysis method combining Nanoparticle-Enhanced Laser Desorption/Ionization Mass Spectrometry (NPELDI-MS) with machine learning. Only a tiny sample volume (1 μL) and a short time (20 seconds per sample) are required to obtain metabolic fingerprints. In a cohort including 102 IgA nephropathy patients and healthy individuals, the method achieves an AUC value of 0.81–0.95 for distinguishing patients from healthy people. It also identifies 5 key serum metabolites (such as glucose and lactic acid) and 4 key urine metabolites (such as glucose and urea). Through longitudinal tracking of 32 patients, the study finds that the change trajectories of these metabolites differ among patients with different prognoses—for instance, serum lactic acid and valine gradually decrease in patients with improved conditions, while they first decrease and then increase in those with worsening conditions. Meanwhile, it reveals disordered pathways related to the disease, such as valine metabolism and galactose metabolism. This research provides a non-invasive, efficient, and accurate tool for the early diagnosis and prognostic evaluation of IgA nephropathy, and also lays a foundation for studying the disease mechanism and developing personalized treatments.
Mater Today Bio. 2025 Oct 15:35:102428. doi: 10.1016/j.mtbio.2025.102428.
Youhe Gao
Statement: During the preparation of this work the author(s) used Doubao / AI reading for summarizing the content. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the published article.
For earlier AI Reads Urine articles：