URINARY CLUSTERIN AND MONOCYTE CHEMOATTRACTANT PROTEIN-1 (MCP-1) PREDICT LONG-TERM MAJOR ADVERSE KIDNEY EVENTS WITH NEPHROTOXICITY

Introduction:

Reliable methods to identify risks early and predict major adverse kidney events (MAKE) in drug-induced acute kidney injury (AKI) are needed. We hypothesised that elevated urine biomarkers of kidney tubular injury in patients with nephrotoxicity could detect subclinical AKI and predict long-term MAKE.

Methods:

We recruited patients admitted to a single tertiary healthcare institution in Singapore from February 2015 to February 2022 who received antimicrobials with nephrotoxic potential, including Aminoglycosides, Vancomycin, Amphotericin B, Polymyxins, Forscanet or Ganciclovir for ≥ 5 days. We measured urinary Clusterin and MCP-1 using ELISA in patients within three days of developing AKI (or final day of nephrotoxic exposure). We analysed the predictive performance of these biomarkers for MAKE, a composite of death, initiation of renal replacement therapy, or doubling of creatinine by one year.

Figure 1. Flow chart of the study cohort.

Results:

We studied a cohort of 161 patients (68% male, mean age 56 years, median eGFR 102 ml/min/1.73m²) with 48% hypertension, 35% diabetes, 39% malignancy as co-morbidities, and median nephrotoxic exposure of 14 days. Of these, 63% received vancomycin and 22% aminoglycosides. AKI developed in 26% of the patients, and 43% experienced MAKE. Biomarker levels were significantly higher (<0.001) in MAKE cases versus none: Clusterin (median 238.6ng/mL, IQR 400.2ng/mL vs. 63.5ng/mL, IQR 167.4ng/mL) and MCP-1 (median 0.63ng/mL, IQR 0.91ng/mL vs. 0.19ng/mL, IQR 0.36ng/mL). The AUROCs for predicting MAKE for Clusterin, MCP-1, and combined were 0.68, 0.69 and 0.71, respectively; the corresponding AUROC was 0.76 when both biomarkers were analysed with initial AKI, with a net reclassification index of 0.078, showing improved prediction. Patients had a stepwise increased MAKE incidence with and without elevated biomarkers and AKI.

Figure 2. Stepwise increase in the accuracy of MAKE prediction is observed when AKI status is combined with biomarker positivity (with Clusterin set at 250ng/mL and MCP-1 set at 1ng/mL). MAKE - Major Adverse Kidney Events; AKI - Acute Kidney Injury.

We derived an ideal cut-off of Clusterin (>280ng/mL) and MCP-1 (>0.4ng/mL) for predicting MAKE in 50 randomly selected patients of the cohort and tested its performance on the remaining patients (n=111), and an independent set of patients (n=28) who received platinum-based chemotherapy.

Table 1. Performance Metrics of Urinary Biomarkers for MAKE Prediction in the Validation Cohort (n=111) and Independent Cohort (n=28)

MAKE - Major Adverse Kidney Events; MCP1 - Monocyte Chemoattractant Protein -1; AKI - Acute Kidney Injury.

Conclusions:

Clusterin and MCP-1 predict MAKE in patients following nephrotoxicity with enhanced accuracy beyond the presence of initial AKI.

This abstract has been revised from the previous submission to the 14th Singapore Society of Nephrology (SSN) Annual Scientific Meeting.

I have no potential conflict of interest to disclose.

I did not use generative AI and AI-assisted technologies in the writing process.