Urinary Kidney Injury Biomarker Profiles in Healthy Individuals and After Nephrotoxic and Ischemic Injury

Two observational studies were conducted to support an initiative to qualify translational kidney safety biomarkers as clinical drug development tools that identify tubular injury prior to changes in estimated glomerular filtration rate (eGFR). Normal healthy volunteers provided three morning spot urine collections over 4 weeks. Patients undergoing surgical resection and intrathoracic cisplatin for malignant pleural mesothelioma provided urine samples pre- and postoperatively at 4, 8, and 12 hours and daily for 6 days. Using receiver-operating characteristics curves, "statistically significant thresholds" established peak longitudinal changes for 8 biomarkers to differentiate mesothelioma patients who developed acute kidney injury (AKI) from normal healthy volunteers. We also assessed "medically significant thresholds" to differentiate mesothelioma patients who did vs. did not develop AKI. Statistically and medically significant thresholds for a fold-change from baseline of urine creatinine (UCr)-normalized values were established for 6 biomarkers: clusterin (2.2, 5.1); osteopontin (3.1, 7.1); N-acetyl-ß-D-glucosaminidase (2.7, 8.1); kidney injury molecule-1 (4.3, 7.5); cystatin C (1.8, 4.5); neutrophil gelatinase-associated lipocalin (2.9, 7.8). For urine albumin and total protein, thresholds were established based on UCr-normalized absolute values: (> upper limit normal, > 10× upper limit normal). Statistically significant thresholds for all biomarkers outperformed eGFR at discriminating mesothelioma subjects exposed to cisplatin from healthy volunteers, demonstrating their utility for enhancing safe drug development. Medically significant thresholds provide perspective on when patients begin to exhibit AKI. These studies have established guideposts for confirmatory studies with additional cohorts and nephrotoxicants to formally qualify the selected biomarkers with worldwide regulatory authorities.