One-Year Outcomes After Belatacept Conversion in Adolescent Kidney Transplant Recipients

Abstract

Introduction: Belatacept (CTLA4-Ig) has shown efficacy in adult kidney transplantation (KT) recipients (improved graft and patient survival and reduced de novo donor-specific antibody [DSA]) compared with calcineurin inhibitors (CNIs). Its long-term benefits and monthly i.v. administration have raised interest in pediatric use, particularly in adolescents, who face an increased risk of graft loss because of nonadherence. However, data on belatacept use in pediatrics are limited. Our objective was to report the 1-year outcomes of all adolescents (N = 45) who underwent CNIs-to-belatacept conversion between 2018 and 2021, in a multicenter retrospective study conducted in the USA and France.

Methods: Indications included long-term CNI avoidance because of toxicity (e.g., histological changes, posttransplant diabetes, and tremors), suboptimal creatinine, or to improve adherence. One-year outcomes were compared with a propensity-matched cohort of adolescents remaining on CNIs.

Results: The median age was 17 years. Rejection occurred in 11 of 45 patients (24%) at a median of 10 months postconversion (7 T-cell-mediated rejections, 3 antibody-mediated rejections, 1 mixed). Belatacept was discontinued in 3 of 11 patients whereas CNIs were added in 2 of 11. Rejection and de novo DSA rates did not differ between patients on belatacept and patients on CNIs. At the individual level, estimated glomerular filtration rate (eGFR) in patients on belatacept without rejection increased significantly (median + 19%), compared with rejectors (-3%, P = 0.03) and patients on CNI (nonrejectors: -11%, P = 0.0006; or rejectors: -14%, P = 0.012). No apparent increase in infectious complications was observed.

Conclusion: Although rejection rates in pediatric patients on belatacept seemed higher than in adult cohorts, they were not significantly different from adolescent patients on CNI, underscoring the challenges related to nonadherence in this population. Tailored immunosuppression strategies, including belatacept, may offer benefits for selected adolescents, but further studies are necessary to define optimal patient selection.

Keywords: belatacept; costimulation blockade; kidney transplantation; outcomes; pediatrics.

Graphical abstract

Figure 1

Evolution of eGFR (Schwartz et al. formula). (a) One year before and after belatacept conversion (n = 45 patients, including 7 early and 38 late conversions). Data are shown as box and whisker plots with median, interquartile (box) and minimum/maximum values (whiskers). Each point represents a patient. Overall, patients’ graft function remained stable at 1 year after belatacept conversion. ∗n = 34 patients at M-12 (exclusion of patients who were converted early, n = 7, as well as patients who were transplanted < 1 year before belatacept conversion, n = 4). ∗∗n = 38 patients at M-6 (exclusion of patients who were converted early, n = 7). ∗∗∗n = 38 patients at M12 (7 patients with follow-up time < 12 months). (b) Difference between eGFR at 1 year and at baseline (ml/min per 1.73 m²) after early (n = 7) or late (n = 31) conversion to belatacept. Data are shown as median with IQR. Each point represents a patient. At the individual level, patients converted early had a median GFR improvement of 13.3 ml/min per 1.73 m² (IQR: 4.1–37.9), whereas patients converted later had a median GFR improvement of 5.7 ml/min per 1.73 m² (IQR: −1.8 to 11.8). (c) eGFR at inclusion and at 1-year follow-up in patients on belatacept (n = 39) versus matched CNI controls (n = 39). Data are shown as box and whisker plots with median, interquartile (box) and minimum/maximum values (whiskers). Each point represents a patient. Overall eGFR at 1 year did not differ between patients on belatacept and CNI controls (median: 51 [IQR: 42–61] ml/min per 1.73 m² and 57 [IQR: 38–68] ml/min per 1.73 m², respectively; P = 0.88. (d) Evolution rate of eGFR at 1 year in patients on belatacept (n = 39) versus matched CNI controls (n = 39). Data are shown as median with IQR. Each point represents a patient. When looking at the eGFR evolution rate at the individual level, patients on belatacept showed a significantly higher improvement of eGFR at 1-year (median: +13%, IQR: −3% to 32%) compared with the CNI group (median: −11%, IQR: −26% to 20%, P = 0.001). (e) eGFR at inclusion and at 1-year follow-up in rejectors and stable patients in the belatacept (n = 39, including 11 rejectors) and CNI control (n = 39, including 10 rejectors) groups. Data are shown as box and whisker plots with median, interquartile (box) and minimum/maximum values (whiskers). Each point represents a patient. At 1-year follow-up, there was no significant difference between stable and rejector patients on belatacept. In contrast, patients on CNIs who experienced rejection had significantly lower graft function than stable patients on CNI (a difference that was already present at inclusion for these patients). (f) Evolution rate of eGFR at 1 year in rejectors and stable patients in the belatacept (n = 39, including 11 rejectors) and CNI control (n = 39, including 10 rejectors) groups. Data are shown as median with IQR. Each point represents a patient. When looking at the eGFR evolution rate at the individual level, stable patients on belatacept showed significant improvement in graft function at 1 year and a significantly higher rate of eGFR progression than patients who experienced rejection on belatacept and patients on CNI (stable or rejectors). CNIs, calcineurin inhibitors; eGFR, estimated glomerular filtration rate; IQR, interquartile range.

Figure 2

Evolution of preexisting DSA in patients (n = 17) at 1-year postconversion to belatacept. Each line represents a patient. If a patient had > 1 DSA, only the immunodominant one is shown in this figure. DSA, donor-specific antibody.

Figure 3

Tacrolimus intrapatient variability over the year preceding conversion to belatacept (or the corresponding inclusion time for CNI controls). Given the challenges of capturing adherence differences directly and incorporating it into the initial matching process, we further analyzed tacrolimus levels intrapatient variability as an indirect marker of adherence. We collected all tacrolimus trough levels (T0) from both the belatacept and CNI control groups during the year preceding conversion (for patients on belatacept) or study inclusion (for patients on CNIs). Of the 39 patients in each group included in the matching, we were able to collect data from 35 patients on belatacept and 32 CNI controls. The mean number of tacrolimus trough levels (T0) collected per patient over the year prior to inclusion was 14 in patients on belatacept and 21 in CNI controls. We calculated intrapatient tacrolimus variability using the coefficient of variation (CV = SD/mean × 100) for each patient as a proxy for adherence, with higher variability in tacrolimus levels suggesting poorer adherence. Importantly, there was no statistically significant difference in the T0 CV between patients on belatacept and those on CNIs, suggesting that adherence patterns were comparable between the 2 groups. CNI, calcineurin inhibitor.

Figure 4

Rejection rate and development of de novo DSA in patients on belatacept versus matched CNI controls. Rejection rate and development of de novo DSA did not differ between groups. Cumulative incidence of rejection and development of de novo DSA were compared using the Kaplan-Meier method and log-rank test. CNI, calcineurin inhibitor; DSA, donor-specific antibody.