Abstract: SA-PO0523
GLP-1 Receptor Agonists and Kidney Outcomes Across the CKD Spectrum: A Real-World Taiwanese Cohort
Session Information
- CKD: Clinical - Therapies, CKM Health, Special Populations, and More
October 24, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: CKD (Non-Dialysis)
- 2202 CKD (Non-Dialysis): Clinical, Outcomes, and Trials
Authors
- Tsai, Heng-Cheng, Division of Nephrology, Ministry of Health and Welfare Taipei Hospital, New Taipei City, Taiwan
- Ou, Shuo-ming, Department of Medicine, Division of Nephrology, Taipei Veterans General Hospital, Taipei City, Taiwan
- Liao, Ting-Ying, Division of Nephrology, Ministry of Health and Welfare Taipei Hospital, New Taipei City, Taiwan
Background
The FLOW trial established semaglutide as renoprotective in type 2 diabetes (T2DM) with chronic kidney disease (CKD), but real-world efficacy across the full CKD spectrum - particularly advanced CKD (stage 4-5) and non-diabetic CKD - is poorly defined in Asian populations. We assessed whether GLP-1 receptor agonist (GLP-1 RA) use alters eGFR and albuminuria trajectories in a real-world Taiwanese nephrology cohort.
Methods
Retrospective single-center cohort of 667 CKD patients (Jan 2023-Mar 2026; 43 GLP-1 RA users [mostly semaglutide], 624 non-users) with >=3 months between paired labs. Eligible: T2DM >=1 year or coexisting ASCVD/CKD (eGFR <60, UACR >30, or UPCR >150). Primary outcome: paired absolute eGFR change (CKD-EPI 2021); secondary: >=30% UACR or UPCR reduction. Subgroups: diabetes status by CKD stage (1-2/3/4-5). Mann-Whitney U for continuous variables; Fisher's exact for proportions; p<0.05.
Results
Overall, GLP-1 RA users had less eGFR decline than non-users (median -0.38 vs. -3.08 mL/min/1.73 m2; p=0.049). The strongest signal was in diabetic stage 4-5 CKD (baseline eGFR <30): median delta-eGFR -0.19 vs. -5.22 (difference +5.03; p=0.007), about 5 mL/min/1.73 m2 of preserved filtration. In non-diabetic CKD, >=30% UACR reduction occurred in 80.0% (4/5) of GLP-1 users vs. 25.7% (26/101) of non-users (p=0.022). A borderline non-DM relative eGFR difference of +10.1 percentage points (p=0.064) was directionally concordant with FLOW.
Conclusion
In this real-world Taiwanese CKD cohort, GLP-1 RA use was associated with attenuated eGFR decline overall and most strikingly in advanced diabetic CKD, plus higher UACR reduction rates in non-diabetic CKD. Findings extend FLOW-class evidence to Asian and non-diabetic CKD populations and support proactive GLP-1 RA use in advanced CKD.
Acknowledgment
The authors thank the nephrology, IT, and Healthcare Quality teams at MOHW Taipei Hospital for support with patient follow-up, data extraction, and Power BI dashboard work. Anthropic Claude (AI) assisted with English editing and structural drafting; the authors take full responsibility. No external funding.