Abstract: FR-PO875
Unsupervised Learning with Network Biomarkers for Risk Stratification in IgA Nephropathy and Its Clinical Implications
Session Information
- IgA Nephropathy: Clinical, Outcomes, and Therapeutics
October 25, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Glomerular Diseases
- 1402 Glomerular Diseases: Clinical, Outcomes, and Therapeutics
Author
- Tan, Jiaxing, West China Hospital of Sichuan University, Chengdu, Sichuan, China
Background
This study integrated network biomarkers with unsupervised learning clustering (KMN) to refine risk stratification in IgA nephropathy (IgAN) and explore its clinical value, addressing the limitations of existing models that heavily rely on renal indicators and lack therapeutic guidance.
Methods
Involving a multicenter prospective cohort, we analyzed 1,460 patients and validated the approach externally with 200 additional patients. Deeper metabolic and microbiomic insights were gained from two distinct cohorts: 63 patients underwent ultra-performance liquid chromatography-mass spectrometry (UPLC-MS), while another 45 underwent fecal 16s RNA sequencing.
Results
Our study utilized hierarchical and k-means clustering methods, integrating renal, extrarenal, and network biomarkers to analyze a cohort with an average follow-up of 58.8 months. This approach identified four distinct patient clusters, with severity and prognosis worsening from Cluster 1 to Cluster 4. The k-means network biomarker (KMN) method emerged as the most effective, achieving an Area Under the Curve (AUC) of 0.77, significantly outperforming the International IgAN Prediction Tool (IIgAN-PT) and RF-RG schemes, which recorded AUCs of 0.72 and 0.69, respectively. Longitudinal analysis over three years highlighted significant differences across clusters in urinary protein and serum creatinine levels—Cluster 1 exhibited stable kidney function while Cluster 4 showed rapid progression towards renal failure. The KMN stratification facilitated personalized treatment, advocating for ACEI/ARBs for lower-risk groups and immunosuppressive therapy for higher-risk groups. It also revealed potential links between IgAN progression and alterations in serum metabolites and gut microbiota. These findings suggest a correlation, though further research is required to establish causality and deepen our understanding of IgAN's underlying mechanisms.
Conclusion
The effectiveness and applicability of the KMN scheme indicate its substantial potential for clinical application in IgAN management.