Abstract: FR-PO102
Adjudication Methodology to Assess Novel Urine Biomarkers Panel Detection of Nephrotoxicant Exposure in Presence or Absence of KDIGO-Defined AKI to Support Biomarker Qualification
Session Information
- AKI: Diagnosis and Outcomes
October 25, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 102 AKI: Clinical, Outcomes, and Trials
Authors
- King, Nicholas M., Critical Path Institute, Tucson, Arizona, United States
- Coca, Steven G., Yale University School of Medicine, New Haven, Connecticut, United States
- Hoffmann, Steven C., Foundation for the National Institutes of Health, North Bethesda, Maryland, United States
- Murray, Patrick T., University College Dublin, Dublin, Ireland
- Radhakrishnan, Jai, Columbia University Irving Medical Center, New York, New York, United States
- Peron, Katrina Jolene, Critical Path Institute, Tucson, Arizona, United States
- Friedman, Gary Steven, Critical Path Institute, Tucson, Arizona, United States
Group or Team Name
- PSTC and FNIH BC Kidney Safety Project.
Background
Critical Path Institute (C-Path) convened 6 expert nephrologist adjudicators (ENAs) to interpret blinded standard & novel clinical and biomarker (SBM & NBM, respectively) data from 2 prospective studies (N=195 patients) to assess NBM sensitivity for detecting drug-induced kidney injury (DIKI) in KDIGO AKI positive & negative participants. Provided SBM and/or NBM baseline (BL) and post-baseline (PBL) patient-level data packets, each ENA determined presence/absence of KDIGO AKI and presence/absence of PBL increase in NBMs-Clusterin, Cystatin C, KIM-1, Osteopontin, NAG, NGAL. In parallel, presence/absence of AKI and DIKI were programmatically quantified using a NBM geometric mean composite measure (GMCM).
Methods
ENAs assessed patient-level electronic case report forms (eCRFs) (N=195). C-Path programmatically generated CDISC SDTM formatted tables to compare programmatic and "human-in-the-loop" adjudication decisions using SBM only [serum creatinine, serum cystatin C, eGFR, Urine Albumin & Protein-Creatinine Ratio (UACR, UPCR)]; NBM only; or combined SBM+NBM (see two 3-member adjudication panels 3-step process).
Results
ENAs unanimity or consensus was achieved in 100% of cases (N=195). Correlation between “human-in-the-loop" NBM adjudication & programmatic NBM GMCM quantitation are primary endpoints within the FDA Qualification Plan submitted to FDA.
Conclusion
NBM panel full qualification is contingent upon demonstration that NBM PBL change is more sensitive than KDIGO AKI criteria for adjudicating nephrotoxicant exposure. If qualified, academic, clinical, pharma NBM PBL and GMCM utilization will drive global clinical nephrology consensus of NBM panel implementation. Periodic re-interrogation of cumulative SBM+NBM data & NBM GMCM under pre-specified statistical analysis plans (SAP) may be a post-qualification commitment to FDA. C-Path's Biomarker Data Repository (BmDR) housing adjudicated participants’ deidentified data (n=195 patients) & patient-level SBM+NBM data from Pharma trials + SAFE-T Consortium trials (~700 participants) may address post-qualification requirements.
Case Adjudication
| Step 1 PERIOD 1: Adjudication of 50% of cases by 3 Panel A members (individual adjudication decision using SBM+eCRFs, only). PERIOD 2: Adjudication of 50% of cases by 3 Panel B members (individual adjudication decision using SBM+eCRFs, only). | Step 2 PERIOD 1: Adjudication of 50% of cases by 3 Panel A members (individual adjudication decision using NBM, only.) PERIOD 2: Adjudication of 50% of cases by 3 Panel B members (individual adjudication decision using NBM, only.) | Step 3 PERIOD 1: Adjudication of 50% of cases by 3 Panel B members (individual adjudication decision using COMBINED NBM + SBM + select eCRFs.) PERIOD 2: Adjudication of 50% of cases by 3 Panel A members (individual adjudication decision using COMBINED NBM + SBM + select eCRFs.) |
Funding
- Other U.S. Government Support – Pfizer, Lilly, Merck, AstraZeneca, Amgen, JnJ