Predicting kidney outcomes in autosomal dominant polycystic kidney disease a comprehensive biomarker analysis

Abstract

Background

Risk stratification tools for autosomal dominant polycystic kidney disease (ADPKD) predict kidney outcomes on a group level but lack precision in patients.

Methods We assessed the value of adding 13 prognostic biomarkers to established risk factors (sex, age, eGFR, systolic BP, Mayo Imaging Classification [MIC], and mutation type) for predicting disease progression. We included 596 patients from the Developing Intervention Strategies to Halt Progression of ADPKD cohort with >= 2 eGFR measurements and >= 1-year follow-up. Results During a mean +/- SD follow-up of 5.0 +/- 2.3 years, the mean +/- SD eGFR slope was -3.46 +/- 2.5 ml/min per 1.73 m(2) per year. Rapid disease progression (eGFR loss >= 3 ml/min per 1.73 m(2) per year) occurred in 303 patients (50.8%), and 279 patients (46.8%) reached the combined end point of kidney failure or 30% eGFR decline. Urinary albumin/creatinine, urinary monocyte chemotactic protein-1/creatinine, and serum copeptin consistently and independently predicted eGFR slope (all P < 0.001), rapid disease progression (area under the curve increasing from 0.79 [95% confidence interval (CI), 0.76 to 0.85] for a baseline model to 0.83 [95% CI, 0.81 to 0.88] when monocyte chemotactic protein-1/creatinine and copeptin were included, P = 0.006), and reaching the combined kidney end point (C-index improving from 0.806 [95% CI, 0.78 to 0.84] for a baseline model to 0.82 [95% CI, 0.79 to 0.85] for a model also containing albumin/creatinine and copeptin, P < 0.001). These results were confirmed in an independent external validation cohort (N=144) and were robust in early disease stages and in patients with moderately increased kidney volumes (MIC 1C). Conclusions Our findings suggest that incorporating these biomarkers into ADPKD risk stratification tools will improve risk prediction, even in subgroups where prognostication is most challenging and relevant.