Urinary Plasminogen as a Marker of Disease Progression in Human Glomerular Disease

Abstract

Rationale & ObjectiveGlomerular disorders have a highly variable clinical course and biomarkers that reflect molecular mechanisms underlying progression are needed. Based on our previous work identifying plasminogen as a direct cause of podocyte injury, we designed this study to test the association between urine plasmin(ogen) (i.e., plasmin and its precursor plasminogen) and end-stage kidney disease (ESKD). Study DesignMulticenter cohort study. Setting & Participants1010 patients enrolled in the CureGN Cohort with biopsy-proven glomerular disease (focal segmental glomerulosclerosis, membranous nephropathy, and immunoglobulin A nephropathy). PredictorsThe main predictor was urine plasmin(ogen) at baseline. Levels were measured by an electro-chemiluminescent immunoassay developed de novo. Traditional clinical and analytical characteristics were used for adjustment. The ratio of urine plasmin(ogen)/ expected plasmin(ogen) was evaluated as a predictor in a separate model. OutcomesProgression to ESKD. Analytical ApproachCox regression was used to examine the association between urinary plasmin(ogen) and time to ESKD. Urinary markers were Log2 transformed to approximate normal distribution and normalized to urine creatinine (Log2uPlasminogen/Cr, Log2 urinary protein/Cr [UPCR]). Expected plasmin(ogen) was calculated by multiple linear regression. ResultsAdjusted Log2uPlasminogen/Cr was significantly associated with ESKD (HR per doubling Log2uPlasminogen/Cr 1.31 (95% confidence interval [CI] 1.22-1.40), P < 0.001). Comparison of the predictive performance of the models including Log2uPlasminogen/Cr, Log2UPCR or both markers showed the plasmin(ogen) model superiority. The ratio of measured/expected urine plasmin(ogen) was independently associated with ESKD, HR 0.41 (95% CI 0.22-0.77) if ratio < 0.8 and HR 2.42 (95% CI 1.54-3.78) if ratio > 1.1 (compared to ratio between 0.8-1.1). LimitationsSingle plasmin(ogen) determination does not allow for the study of changes over time. The use of a cohort of mostly white patients and the restriction to patients with 3 glomerular disorders limits the external validity of our analysis. ConclusionsUrinary plasmin(ogen) and the ratio of measured/expected plasmin(ogen) are independently associated with ESKD in a cohort of patients with glomerular disease. Taken together with our previous experimental findings, urinary plasmin(ogen) could be a useful biomarker in prognostic decision making and a target for the development of novel therapies in patients with proteinuria and glomerular disease.