Candidate composite biomarker to inform drug treatments for diabetic kidney disease.

Abstract

Current guidelines recommend renin angiotensin system inhibitors (RASi) as key components of treatment of diabetic kidney disease (DKD). Additional options include sodium-glucose cotransporter-2 inhibitors (SGLT2i), glucagon-like peptide 1 receptor agonists (GLP1a), and mineralocorticoid receptor antagonists (MCRa). The identification of the optimum drug combination for an individual is difficult because of the inter-, and longitudinal intra-individual heterogeneity of response to therapy. Using data from a large observational study (PROVALID), we identified a set of parameters that can be combined into a meaningful composite biomarker that appears to be able to identify which of the various treatment options is clinically beneficial for an individual. It uses machine-earning techniques to estimate under what conditions a treatment of RASi plus an additional treatment is different from the treatment with RASi alone. The measure of difference is the annual percent change (ΔeGFR) in the estimated glomerular filtration rate (ΔeGFR). The 1eGFR is estimated for both the RASi-alone treatment and the add-on treatment. Higher estimated increase of eGFR for add-on patients compared with RASi-alone patients indicates that prognosis may be improved with the add-on treatment. The personalized biomarker value thus identifies which patients may benefit from the additional treatment.