Abstract: SA-OR044
Transcriptomic and Proteomic Profiling Identifies Patients with Elevated Inflammatory and Immune Signaling in Nephrotic Syndrome
Session Information
- Glomerular Diseases: Spotlighting Immunology and Inflammation - II
October 27, 2018 | Location: 8, San Diego Convention Center
Abstract Time: 05:42 PM - 05:54 PM
Category: Glomerular Diseases
- 1202 Glomerular Diseases: Immunology and Inflammation
Authors
- Eddy, Sean, University of Michigan, Ann Arbor, Michigan, United States
- Hamidi, Habib, University of Michigan, Ann Arbor, Michigan, United States
- Mariani, Laura H., University of Michigan, Ann Arbor, Michigan, United States
- Hartman, John R., University of Michigan, Ann Arbor, Michigan, United States
- Reich, Heather N., Toronto General Hospital, Toronto, Ontario, Canada
- Lafayette, Richard A., Stanford University, Stanford, California, United States
- Kretzler, Matthias, University of Michigan, Ann Arbor, Michigan, United States
Group or Team Name
- NEPTUNE Consortium
Background
Despite the high cost associated with renal disease, the reliability and availability of information to guide patient care using a precision medicine approach prior to reaching end-stage is relatively sparse. To address this, patient specific profiles were generated for major signaling pathways to understand the diversity of patient level signaling in a nephrotic syndrome cohort.
Methods
Transcriptomic profiles were generated from isolated glomeruli (glom) and tubulointerstitium (TI) samples from subjects with nephrotic syndrome in the NEPTUNE cohort. Patient-specific transcriptional profiles were generated from a curated set of genes for activated TNF, JAK-STAT, and major immune cell types. Patient-specific profiles were correlated with urine biomarker profiles generated from a panel of 54 markers to identify relevant surrogate markers. NMF clustering was used to identify disease-relevant molecular subtypes. Patient-specific profiles and clustering approaches were utilized in a blinded retrospective analysis to evaluate the use of high dimensional data to predict patient outcomes and help guide treatment decisions.
Results
Across the cohort, TNF and JAK-STAT pathway activation profiles in the tubulointerstitium were both positively correlated with IFTA (p<0.001). In select cases, profiles were identified with high pathway activation with little to no IFTA. MCP-1 and TIMP1 were identified as biomarkers for TNF activation with a C-statistic of 0.86. Urine IP-10 was correlated with intrarenal JAK-STAT activation (p<0.001), and together with immune cell signatures were predictive of intrarenal inflammation and immune activation. NMF clustering of glom and TI profiles from subjects each revealed four patient clusters that were associated with outcome; the poorest prognosis cluster was enriched for inflammatory and innate immunity pathways. In individual cases, profiling results were qualitatively consistent with pathology descriptors.
Conclusion
Patient-specific transcriptional and urine biomarker profiles were identified that can indicate degree of intrarenal signaling and damage, and by combing data from other domains, can be used to identify patients that may be best suited for targeted therapy approaches and/or pharmacodynamic patient monitoring in clinical settings.
Funding
- NIDDK Support