Performance of the Integrative Box (iBox) Prognostication System in African American Kidney Transplant Recipients
Authors:
- Agathe TruchotATAgathe Truchot
- Yannis LombardiYLYannis Lombardi
- Marc RaynaudMRMarc Raynaud
- Olivier AubertOAOlivier Aubert
- Gillian DivardGDGillian Divard
- Thibaut ThalamasTTThibaut Thalamas
- Brad AstorBABrad Astor
- Didier MandelbrotDMDidier Mandelbrot
- Sandesh ParajuliSPSandesh Parajuli
- Kenneth A. NewellKNKenneth A. Newell
- Christian P. LarsenCLChristian P. Larsen
- Geeta KaradkheleGKGeeta Karadkhele
- 10 September 2026
- journal article
- Published by Wolters Kluwer Health in Journal of the American Society of Nephrology
Abstract
Key Points: The integrative box showed accurate discrimination, overall fit, and clinical utility in 3866 kidney transplant recipients from North America. Predictive performance was maintained in African American and non-African American kidney transplant recipients. These results support the robustness of the integrative box as a risk prediction tool. Background: The integrative box (iBox) is a validated prognostication system that predicts long-term death-censored graft loss in kidney transplant recipients, but its performance in the specific population of African American recipients has not been fully studied. Methods: We conducted a multicenter study including 3866 kidney transplant recipients from North America, of whom 1040 (27%) were African American, to assess the impact of race on the iBox's performance in predicting graft loss. Discrimination, calibration, overall fit, and clinical utility were evaluated in the whole cohort and in African American and non-African American recipients. Results: Performance metrics for the prediction of graft loss were similar in both subgroups in discrimination (C-index 0.80 [95% confidence interval (CI), 0.77 to 0.83] in African American recipients and 0.83 [95% CI, 0.81 to 0.85] in non-African American recipients, P = 0.07) and observed/expected ratios (1.08 [95% CI, 0.95 to 1.23] and 0.99 [95% CI, 0.88 to 1.10], respectively, P = 0.28). No significant interaction between iBox score values and race was found in multivariable analysis stratified by transplant center ( P = 0.24 for the interaction term). The results were consistent between the subgroups regardless of the equation used to estimate eGFR (Kidney Recipient Specific, CKD-Epidemiology Collaboration, or Modification of Diet in Renal Disease), but with variations in calibration across equations. Conclusions: The iBox prognostication system showed accurate discrimination, overall fit, and clinical utility up to 7 years after risk evaluation in African American and non-African American kidney transplant recipients, supporting its robustness in this population. Clinical Trial registry name and registration number: ClinicalTrials.gov, NCT03474003.Keywords
Funding Information
- OrganX
- OrganX
- Poste d'accueil AP-HP/SCAI
This publication has 37 references indexed in Scilit:
- OPTN/SRTR 2022 Annual Data Report: KidneyAmerican Journal of Transplantation, 2024
- Genetic versus self-reported African ancestry of the recipient and neighborhood predictors of kidney transplantation outcomes in 2 multiethnic urban cohortsAmerican Journal of Transplantation, 2024
- Qualifying a novel clinical trial endpoint (iBOX) predictive of long-term kidney transplant outcomesAmerican Journal of Transplantation, 2023
- Augmenting the United States transplant registry with external mortality data: A moving target ripe for further improvementAmerican Journal of Transplantation, 2023
- Validation of a prediction system for risk of kidney allograft failure in pediatric kidney transplant recipients: An international observational studyAmerican Journal of Transplantation, 2023
- Machine learning does not outperform traditional statistical modelling for kidney allograft failure predictionKidney International, 2022
- Exacerbating Racial Disparities in Kidney TransplantJAMA Surgery, 2020
- Consistent Estimation of the Expected Brier Score in General Survival Models with Right‐Censored Event TimesBiometrical Journal, 2006
- A new measure of prognostic separation in survival dataStatistics in Medicine, 2004
- MULTIVARIABLE PROGNOSTIC MODELS: ISSUES IN DEVELOPING MODELS, EVALUATING ASSUMPTIONS AND ADEQUACY, AND MEASURING AND REDUCING ERRORSStatistics in Medicine, 1996