Abstract: TH-OR044
Integration of Pathomic Signatures and Spatial Transcriptomics Reveals Tubular Morphometric Trajectories and Enables Patient Stratification in Minimal Change Disease (MCD)/FSGS
Session Information
- Glomerular Precision Nephrology: Novel Mechanisms, Biomarkers, and Therapeutic Targets
November 06, 2025 | Location: Room 310A, Convention Center
Abstract Time: 05:40 PM - 05:50 PM
Category: Glomerular Diseases
- 1402 Glomerular Diseases: Clinical, Outcomes, and Therapeutics
Authors
- Fan, Fan, Department of Biomedical engineering, Emory University and Georgia Institute of Technology, Atlanta, Georgia, United States
- Smith, Cathy, University of Michigan Department of Internal Medicine, Ann Arbor, Michigan, United States
- Nair, Viji, Department of Biomedical engineering, Emory University and Georgia Institute of Technology, Atlanta, Georgia, United States
- Demeke, Dawit S., University of Michigan Department of Pathology, Ann Arbor, Michigan, United States
- Wang, Bangchen, Department of Pathology, Division of AI & Computational Pathology, Duke University, Durham, North Carolina, United States
- Ozeki, Takaya, Nagoya University Graduate School of Medicine, Department of Nephrology, Nagoya, Aichi Prefecture, Japan
- Jacobs, Jackson, Department of Biomedical engineering, Emory University and Georgia Institute of Technology, Atlanta, Georgia, United States
- Liu, Qian, Children’s Hospital of Philadelphia Research Institute, Philadelphia, Pennsylvania, United States
- Bitzer, Markus, University of Michigan Department of Internal Medicine, Ann Arbor, Michigan, United States
- Zee, Jarcy, Children’s Hospital of Philadelphia Research Institute, Philadelphia, Pennsylvania, United States
- Mariani, Laura H., University of Michigan Department of Internal Medicine, Ann Arbor, Michigan, United States
- Lafata, Kyle Jon, Duke University Department of Radiation Oncology, Durham, North Carolina, United States
- Holzman, Lawrence B., Department of Medicine, Division of Nephrology and Hypertension, University of Pennsylvania, Philadelphia, Pennsylvania, United States
- Hodgin, Jeffrey B., Department of Medicine, Division of Nephrology and Hypertension, University of Pennsylvania, Philadelphia, Pennsylvania, United States
- Madabhushi, Anant, Department of Biomedical engineering, Emory University and Georgia Institute of Technology, Atlanta, Georgia, United States
- Barisoni, Laura, Department of Pathology, Division of AI & Computational Pathology, Duke University, Durham, North Carolina, United States
- Janowczyk, Andrew, Department of Biomedical engineering, Emory University and Georgia Institute of Technology, Atlanta, United States
Background
Conventional classification of tubules into normal, injured, and atrophic imposes rigid categories on a continuous spectrum of morphological changes. Leveraging pathomic features from segmented tubular substructures, we identified tubular pathomic signatures and trajectories to capture this spectrum, aligned them with spatial transcriptomics, and assessed clinical relevance.
Methods
We applied non-negative matrix factorization to 99 pathomic features from 288,172 tubules across 254 NEPTUNE/CureGN (135 FSGS, 119 MCD/MCD-like), and 13 nephrectomies (12 reference, 1 Xenium) to identify 14 interpretable signatures representing co-occurring morphologic patterns. Monocle3 constructed pseudotime trajectories to map progressive tubular changes, reflecting potential injury and repair paths. Pathomic signatures and spatial transcriptomic markers were spatially aligned to evaluate molecular-morphologic correlations. Patient-level profiles were derived via Wasserstein distance between each patients’ tubules and reference tubules. Hierarchical clustering stratified individuals into distinct risk groups. Kaplan-Meier analysis assessed associations with two clinical outcomes (time from biopsy to 40% eGFR decline and proteinuria remission).
Results
Pan-tubular trajectories reflected kidney function decline (eGFR) (Fig. A1&3). Proximal and distal tubules-specific trajectories highlighted atrophy & hypertrophy pathways (Fig. A2). Tubules with high atrophy signature colocalized with adaptive proximal tubular markers (Fig. A4). Signature-based profiles stratified patients into clinically distinct risk groups (Fig. B).
Conclusion
Pathomic signature-derived trajectories capture continuous morphologic change linked to kidney function. These signatures enable objective tubular phenotyping and clinically relevant stratification with prognostic potential.
Funding
- NIDDK Support