External validation of the updated international IgA nephropathy prediction tool in a Chinese population

The updated post-biopsy International IgA Nephropathy (IgAN) Prediction Tool (IIgAN-PT) requires validation in diverse clinical settings. We evaluated its performance in a Chinese cohort characterized by severe pathology and aggressive immunosuppression therapy.

Patients with biopsy-confirmed IgAN from two medical centers in China were included. We calculated the predicted risk for each patient. The primary outcome was the first occurrence of end-stage kidney disease(ESKD, eGFR < 15 ml/min/1.73 m², dialysis or transplantation) or a permanent decline in estimated glomerular filtration rate to less than 50% of the baseline time value at one year after kidney biopsy. Model performance was assessed by discrimination, calibration, and reclassification.

In our cohort, 14.1% of patients experienced the primary outcome. The updated post-biopsy IIgAN-PT did not demonstrate significantly superior discriminatory ability as compared to the original biopsy IIgAN-PT. Kaplan-Meier curves for risk subgroups showed poor separation in low- and intermediate-risk groups but marked separation in the high-risk group. However, the calibration performance of the post-biopsy IIgAN-PT improved. The updated IIgAN-PT significantly improved the Net Reclassification Index(NRI) compared to the original IIgAN-PT, but showed limited improvement in Integrated Discrimination Improvement(IDI) for clinical risk reclassification. Subgroup analysis indicated that immunosuppression therapy significantly reduced the model's discriminatory ability.

In this external validation, the updated post-biopsy IIgAN-PT demonstrated improved calibration over the original IIgAN-PT for risk assessment at 1-year post-biopsy, leading to more accurate absolute risk estimates. However, due to the limitations of the updated IIgAN-PT in distinguishing between low-risk and intermediate-risk patient groups, its application value for refined risk stratification in clinical decision-making is constrained. Registry: Chinese Clinical Trial Registry, TRN: ChiCTR2400093550, Registration date: 06 December 2024.