Transformative Research in Diabetic Nephropathy 2.0

Overview

The goal of this observational study is to learn more about kidney health in adults with diabetic kidney disease and other groups. Researchers will study kidney tissue and other samples. They want to learn how sodium-glucose cotransporter-2 (SGLT2) inhibitors, a type of diabetes medicine, may affect the kidneys. People can join only if they are already having a kidney biopsy or kidney surgery as part of their regular medical care. The main questions this study aims to answer are: * Do people who take SGLT2 inhibitors show different biological patterns in kidney tissue than similar people who do not take them? * Are these kidney tissue patterns linked with how kidney health changes over time? Researchers will compare participants who take SGLT2 inhibitors with similar participants who do not take these medicines. Participants will: Let researchers use one stored slide of kidney tissue from their regular care (no extra research biopsy) Give a blood sample and a urine sample Let researchers review medical record information over time

TRIDENT 2.0 is a multicenter observational translational study that characterizes kidney molecular and histopathologic features in relation to exposure to kidney-protective therapies, with a focus on sodium-glucose cotransporter-2 (SGLT2) inhibitors. The study leverages archived clinical kidney pathology material and harmonized clinical data to support integrated molecular-histologic analyses across participating sites. Tissue sources and central repository workflows: Formalin-fixed, paraffin-embedded (FFPE) kidney tissue sections/slides are generated from archived clinical pathology material (including clinically indicated kidney biopsies and available donor or nephrectomy specimens) and transferred under coded identifiers to a central repository for downstream molecular assays and digital pathology. Spatial transcriptomics and molecular profiling: Spatially resolved transcriptomic methods are used to generate high-resolution molecular maps while preserving tissue architecture. FFPE (or fresh-frozen, when available) sections may be processed using commercially available spatial transcriptomics platforms (e.g., 10x Genomics Visium, NanoString CosMx, Xenium) or updated technologies implemented under study governance. Standard quality control procedures evaluate tissue integrity, RNA quality, capture efficiency, and resolution of major kidney compartments and cell types. Sequencing is performed on Illumina platforms with platform-appropriate depth, followed by preprocessing using platform-specific pipelines and downstream analysis in R/Python workflows (e.g., Seurat or equivalent) with normalization and batch correction as needed. Planned analyses include identification of cell-type and sub-cell-type signatures in spatial context, mapping of injury patterns (e.g., fibrosis/inflammation/vascular remodeling), and comparative molecular profiling across disease categories and therapy exposure groups with adjustment for relevant covariates. Digital pathology and centralized histopathology review: Digitized clinical stains and available diagnostic images are used for centralized review and standardized lesion scoring by renal pathologists. When applicable, diabetic kidney disease (DKD) is classified using established renal pathology criteria. Specimen adequacy metrics are used to guide analytic inclusion and sensitivity analyses. Linked clinical data (high level): Clinical data are harmonized across sites to support clinicopathologic and molecular integration, including medication exposure history and relevant laboratory and diagnostic variables. Longitudinal clinical information is used to contextualize molecular and histologic findings for downstream modeling. Statistical and integrative analytic approach: Analytic methods include differential expression and pathway analyses with appropriate multiple-testing control and covariate adjustment. Integrative modeling may combine molecular, histologic, and clinical domains using dimension reduction and regularized approaches to derive molecular signatures associated with disease state and therapy exposure.