Long-Term Patterns of Tacrolimus Intrapatient Variability and Their Association with Kidney Transplant Outcomes

Le résumé fourni par la source

Background: Tacrolimus intra-patient variability (TacIPV) is a recognized marker of nonadherence and is associated with adverse kidney transplant (KT) outcomes. While previous studies have focused on short-term TacIPV, clinical implications of long-term TacIPV patterns are less understood. We aimed to classify TacIPV trajectories beyond one year post-transplant using machine learning and evaluate their association with long-term KT outcomes. Methods: We identified living donor KT recipients who were maintained on tacrolimus without regimen changes between 2001 and 2016 at a tertiary referral academic hospital in Korea. TacIPV was assessed in each of five post-transplant periods (6–12, 13–24, 25–36, 37–48, and 49–60 months) using variation independent of mean. Patients were clustered based on time-dependent TacIPV patterns using unsupervised machine learning. The primary outcome was a composite endpoint of late biopsy-proven acute rejection, doubling of serum creatinine, and death-censored graft failure. Cox regression models were used to assess the association between TacIPV patterns and outcomes. Results: Among 496 patients, four distinct TacIPV trajectory clusters were identified: Cluster 1 (57%) exhibited persistently low TacIPV, while Cluster 2 (22%) showed high TacIPV in early post-transplant period, followed by a decrease. Cluster 3 (16%) had initially low TacIPV, followed by a transient rise at one year, before returning to a low level thereafter. Cluster 4 (5%) maintained consistently high TacIPV throughout the follow-up. Patients in Clusters 2, 3, and 4 had higher incidence rates of composite outcome compared to Cluster 1, with rates of 7.18 (P=0.064), 9.61 (P=0.001), and 8.91 (P=0.059) / 100 person-years, respectively, versus 5.29 in Cluster 1. Risk of composite outcome was significantly higher in Clusters 2, 3, and 4 compared to Cluster 1, with adjusted hazard ratios of 1.45 (95%CI, 1.04–2.02), 1.77 (1.23–2.55), and 1.79 (1.03–3.12), respectively. Conclusion: Transient fluctuations in TacIPV beyond one year post-KT are associated with worse graft outcomes as persistently high TacIPV, which underscores critical importance of consistent medication adherence. Sustained adherence interventions coupled with continuous TacIPV monitoring through a multidisciplinary approach may improve long-term graft outcomes.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.