Abstract: FR-PO0858
Mean- vs. Shape-Driven eGFR Trajectories and Relapse Burden
Session Information
- Glomerular Diseases: Practice and New Concepts Shaping Modern Care
October 23, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Glomerular Diseases
- 1402 Glomerular Diseases: Clinical, Outcomes, and Therapeutics
Authors
- Anjum, Haaris, Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States
- Gill, Nathan, Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States
- Banker, Margaret, Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States
Background
Kidney disease progression is heterogeneous, with trajectories reflecting differences in overall kidney function (mean-driven) and trajectory shape (shape-driven). However, the degree to which these respective differences contribute to relapse burden remains unclear.
Methods
We analyzed 1,587 adult participants (≥18 years) from CureGN, a prospective observational cohort study of patients with MCD, FSGS, MN, and IgA enrolled within 5 years of biopsy. Longitudinal eGFR was modeled using group-based trajectory modeling (GBTM). Trajectories were defined using non-normalized (mean-driven) and normalized (shape-driven) approaches. Individuals were assigned by highest posterior probability. These classifications were included as categorical predictors in a fractional logit model of relapse burden (proportion of follow-up time in relapse). The model included main effects for mean-driven and shape-driven trajectory groups and their interaction. No additional clinical covariates were included.
Results
The non-normalized model identified four trajectory groups: Low (n=365), Med-Low (n=376), Med-High (n=423), High (n=423). The normalized model identified two groups: Flat (n=1199) and Decreasing (n=388). Mean relapse proportion was 0.70. Lower mean eGFR groups were consistently associated with higher relapse burden, with predicted relapse proportions ranging from 0.54 in the High group to 0.84 in the Low group. Decreasing trajectories were associated with higher relapse burden than flat trajectories, though differences were smaller. No meaningful interaction was observed.
Conclusion
Mean eGFR is the primary driver of relapse burden, with trajectory shape providing modest additional information. Further study is needed to assess consistency across histologic disease categories.
Figure. eGFR trajectory groups from mean-driven and shape-driven GBTM
Funding
- NIDDK Support