*Purpose: The purpose of this study was to define novel clinical phenotypes based on biopsy results to enhance biomarker discovery.

*Methods: Retrospective analysis of kidney biopsy phenotypes from CTOT-08 was performed. Each subject was categorized into 1 of 6 groups based on the sum of their kidney histopathology. Group 1 had no rejection with stable kidney function. Group 2 had rejection biopsies only. Groups 3 had mixed no rejection with stable kidney function and acute kidney dysfunction (ADNR) without acute rejection. Group 4 was defined as ADNR only including BK nephropathy without any rejections. Group 5 had a combination of rejection and no rejection biopsy phenotypes. Group 6 had a clinical acute rejection (cAR) and chronic fibrosis (CFIB) pathology. Kaplan-Meier curves and Cox proportional hazard model were used to assess graft survival.

*Results: The 264 subjects were categorized into 6 groups based on kidney biopsy results as follows: Group 1: 31.4% (n=83), Group 2: 9.5% (n=25), Group 3: 12.9% (n=34), Group 4: 3.8% (n=10), and Group 5: 26.9% (n=71), and Group 6: 15.5% (n=41). There was a significant difference in graft survival among the 6 groups (p < 0.001) (Figure 1). Group 2 had significantly higher risk of graft failure (Hazard ratio (HR) 18.84 [95% CI 4.77-74.43], p-value <0.001) than Group 1. Group 6 also showed significantly higher hazard ratio for graft failure than Group 1 (HR 4.25 [95% CI 1.10-16.43], p-value 0.04) (Table 1)

*Conclusions: Novel groupings of subjects based on the combination of their pathological phenotypes can be used to predict survival rates among kidney transplant recipients. These groupings might offer the potential basis for ongoing predictive biomarker discovery and outcomes research.

« Back to 2022 American Transplant Congress