Abstract: TH-PO0598
Artificial Intelligence: Driven Gene Therapy for Alport Syndrome
Session Information
- Monogenic Kidney Diseases: Glomerular
November 06, 2025 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Genetic Diseases of the Kidneys
- 1201 Genetic Diseases of the Kidneys: Monogenic Kidney Diseases
Authors
- Wu, Yu, AlloDx (Shanghai) Biotech Co, Ltd, Shanghai, China
- Cuello Garcia, Haider, AlloDx (Shanghai) Biotech Co, Ltd, Shanghai, China
- Shen, Yanjun, AlloDx (Shanghai) Biotech Co, Ltd, Shanghai, China
- Li, Lu Peil, AlloDx (Shanghai) Biotech Co, Ltd, Shanghai, China
- Cao, Changling, AlloDx (Shanghai) Biotech Co, Ltd, Shanghai, China
- Shi, Haifeng, AlloDx (Shanghai) Biotech Co, Ltd, Shanghai, China
- Jiang, Tingya, AlloDx (Shanghai) Biotech Co, Ltd, Shanghai, China
Background
Alport syndrome is prevalent hereditary kidney disease, affects approximately 1 in 5000 individuals in the Unites States and characterized by a mutation in the gene encoding collagen IV (COL4), despite recent therapeutic advances, there remains not truly effective treatments. We present an artificial intelligence (AI)-driven gene therapy for severe X-linked Alport syndrome (XLAS) in males, focusing on the optimization of mini collagen α5 sequences to restore functional α3, α4, α5 heterotrimer assembly.
Methods
Our AI-powered gene therapy for male XLAS using deep-learning algorithms, The construct was evaluated in vitro transfecting HEK293 cells with AI predicted, truncated mini collagen α5 sequences and imaging with Split NanoLuc technology for luciferase activity. This bioluminescence-based approach enabled highly sensitive and quantitative detection of α3 -α4 - miniα5 trimer formation. Moreover, we treated five Col4a5 R471X mice using package adeno-associated virus, and 1 month post-treatment in tail vein injection, levels of Blood urea nitrogen (BUN), Creatinine (Scr), and Urinary Albumin (UA) were tested.
Results
We obtained 6 sequences of collagen α5 sequences and 2 significantly enhanced the assembly of native-like heterotrimers in our luciferase NanoLuc test (Figure 1A). In Col4a5 R471X mice, treatment led to substantial improvements in renal function. UA (922.80 VS 194.55, p=0.0015), BUN (55.14 VS 22.67, p=<0.0001), Scr (3119.19 VS 835.03, p=<0.0001), and UACR (3050.45 VS 270.41, p=<0.0001) pre- and post-treatment respectively (Figure 1B). Immunofluorescence microscopy further confirmed the restoration of α5 chain expression in the glomerular and tubular basement membranes, indicating the normalization of type IV collagen architecture.
Conclusion
Our study validates the feasibility of an AI-informed approach in rectifying the molecular defect of XLAS, offering a novel treatment strategy. Future research will focus on refining the AI algorithm to treat diverse genetic mutations, optimizing systems for enhanced therapeutic efficacy, and advancing toward clinical trials, moreover, Provides a new treatment avenue for personalized therapies