New Score Accurately Predicts Risk of Kidney Allograft Loss

By Will Boggs MD

October 30, 2019

NEW YORK (Reuters Health) - A new eight-factor scoring system, iBox, accurately predicts the risk of allograft loss in patients receiving kidney transplants, according to an international study.

"We were very excited by the fact that the iBox proved to be robust and reliable in several situations tested, in clinical trials covering different clinical situations, and under different health systems," said Dr. Alexandre Loupy of the University of Paris and Necker Hospital, also in Paris.

"Therefore, not only is our predicting tool robust, but it is also applicable worldwide," he told Reuters Health by email.

The rates of late allograft failure have improved little over the past 15 years, and there is no robust and widely validated prognostication system for the risk of allograft failure in individual patients. Single parameters - estimated glomerular filtration rate (GFR), proteinuria, histology and HLA profiles - have failed to provide sufficient predictive accuracy.

Dr. Loupy and colleagues from France, Belgium and the U.S. sought to develop a practical risk-stratification score that could be used to identify patients at high risk of future allograft loss.

The resulting iBox score is based on eight independent predictors of allograft failure: time from transplant to evaluation, estimated GFR, proteinuria, interstitial fibrosis/tubular atrophy, microcirculation inflammation, interstitial inflammation and tubulitis, transplant glomerulopathy, and anti-HLA donor specific antibody (DSA) mean fluorescence intensity.

The model provided good discrimination ability at three years (C-index, 0.835), five years (0.819), and seven years (0.808), the researchers report in The BMJ, online September 17.

Discrimination performance was also good in the external validation cohorts in Europe (C statistic, 0.81) and in the U.S. (0.80).

When applied to three published clinical trials, iBox showed similarly accurate discrimination (C index, 0.87). The iBox prediction score also outperformed other risk scores in a systematic review.

"The iBox risk prediction system assessed the risk at a given time point, but we have shown that it can be re-evaluated at different time points after transplantation, enabling clinicians to calculate a new risk that takes into account the updated values of eGFR, proteinuria, allograft scarring, allograft inflammation, damage, and presence and concentration of anti-HLA (donor-specific antibodies)," the researchers note. "Therefore, we confirmed the iBox system's transportability for additional and updated evaluations in the patient's long-term course."

"I would like all physicians involved in transplantation to realize that they have now a reliable risk-prediction tool for long-term allograft survival that could substantially change the day-to-day care of transplanted patients," Dr. Loupy said. "Clinical decisions will be better informed, and it will also improve engagement of patients in shared decisions for medical management with objective estimation of allograft survival."

Dr. Minnie Sarwal of the University of California, San Francisco, who studies transplant rejection, but was not involved in the new work, told Reuters Health by email, "It is surprising the amount of data input required to generate the iBox score and the absolute dependence on the need to do an invasive biopsy."

"The prediction system of choice (for patients receiving kidney transplants) should be noninvasive and highly sensitive - this iBox assay would be best used for clinical trials," she said.


BMJ 2019.