Nephrotic Syndrome Rare Disease Clinical Research Network II

University Of Michigan At Ann Arbor, Ann Arbor MI

Investigators

Linked publications, trials & patents

Abstract

ABSTRACT Although patients with Nephrotic Syndrome (NS) present with shared clinical signs and symptoms (proteinuria, hypoalbuminemia, hyperlipidemia and edema), there is dramatic variability in prognosis and response to therapy, frustrating patients, families and their clinicians. Even within the histopathologic categories used in the current diagnostic approach (e.g. minimal change disease, focal segmental glomerulosclerosis), there is dramatic variability in disease progression and response to therapy, highlighting the underlying biological heterogeneity within the groups. Small studies with broad, clinical patient inclusion criteria have demonstrated that a subset of patients respond well to certain immunosuppressive medications, but accurate pre-treatment response of those individuals is not possible based on routine clinical parameters. This project will leverage the Nephrotic Syndrome Study Network (NEPTUNE) cohort study, a multi-center prospective study of over 600 patients with FSGS, MCD and MN with rich clinical data, kidney biopsy tissue and gene expression profiles. This study will leverage novel analysis tools to analyze the kidney tissue gene expression data to identify subgroups of patients with shared expression profiles. And, to combine this data with clinical characteristics and kidney biopsy pathology features to predict response to immunosuppressive therapy. It will utilize advanced analysis techniques which are ideal for rare disease when traditional statistical methods are not well equipped to identify relevant predictors from the large number of potential parameters across the genotype-phenotype continuum. The aims are: Aim 1: To utilize consensus non-negative matrix factorization (NMF) on glomerular and tubular genome wide mRNA expression data to identify functional subclasses of NEPTUNE participants. Aim 2: To utilize machine learning algorithms to identify predictors of response to initial immunosuppressive therapy from clinical characteristics, kidney biopsy pathology features and gene expression clusters. The objective is to improve the clinical care of patients with Nephrotic Syndrome by identifying novel, biologically-relevant predictors of treatment response. Risk stratification is useful not only to inform patients and clinicians who are making treatment decisions, but also allows stratification of patients at high risk of treatment failure into clinical trials of novel agents.