P10 IDENTIFICATION OF BIOMARKER CANDIDATES AND THERAPEUTIC TARGETS USING TRANSCRIPTOME-DRIVEN APPROACH

Abstract

Chronic kidney disease (CKD) affects 8-16% people worldwide, with an increasing incidence and prevalence of end-stage kidney disease (ESKD). The effective management of CKD is currently confounded by the inability to identify patients at high risk of progression while still in early stages of CKD. Identifying the mechanism responsible for progression of CKD is a critical step to contain the worldwide CKD epidemic. A molecular disease definition of CKD will allow us to develop strategies for patient risk stratification to prioritize limited health care resources, to identify targeted molecular therapies for patients at high risk and to develop biomarkers to define such patients for clinical trials. Our hypothesis is that both clinical phenotypes (e.g. GFR, proteinuria) and renal tissue alterations seen in CKD are associated with, and a consequence of, the dynamic molecular mechanism reflected in the transcriptional programs of the diseased renal tissue. Our work started from the identification and cross-validation of pathways and candidate intrarenal biomarkers for CKD progression in 261 kidney biopsies. We then used a sequential prioritization strategy to prioritize candidates for non- invasive biomarkers to allow broad clinical applicability. Next, intra-renal transcript levels of the top candidate biomarker were correlated with the urinary levels of the encoded proteins. Urinary protein levels were assessed for their correlation with CKD progression. Finally, the biomarker was tested for its ability to increase predictive power of established clinical marker panels for CKD progression prediction in three cohorts. We used Cox proportional hazards models to evaluate the predictive value of marker on CKD outcome, and used likelihood ratio tests, C-statistics, and Akaike information criterion (AIC) to assess the goodness of fit and improved prediction ability. Finally, we present our effort in identification the molecular disease mechanism, serving as targets, to develop novel therapies to prevent CKD progression. Using Epidermal Growth Factor (EGF) as a proof of principle, we could demonstrate that prediction of renal survival by eGFR and albuminuria was significantly improved by addition of uEGF to the model in diverse CKD populations with a wide spectrum of causes and stages. uEGF may contribute to the improved risk prediction as it can capture the degree of tubular differentiation and regeneration potential, mechanisms essential to retain renal function with the acute and chronic insults seen in CKD, but not be well reflected by the conventional predictors (proteinuria or baseline GFR). uEGF shows promise as an independent risk predictor of CKD progression. Inclusion of uEGF significantly improved prediction of composite end points by eGFR and proteinuria in diverse populations worldwide with a wide range of CKD. An immediate benefit of our work can be an improved stratification of CKD patients for the selections of high-risk patients into clinical trials, addressing a critical hurdle for novel molecular target validation in CKD.