Early Determination of Tacrolimus Concentration-Dose Ratio Identifies Risk of Allograft Loss in Kidney Transplantation

Abstract

Introduction: Fast tacrolimus-metabolizing kidney transplant recipients (KTRs) (i.e., tacrolimus trough-level/total daily dose [C0/D < 1.05]) have poorer allograft function; however, their identification in a real-life setting is challenging. We investigated the reproducibility of tacrolimus metabolic status during the first months after transplantation and its association with long-term allograft outcomes.

Methods: All KTRs between 2000 and 2019 with a functional allograft at 1 month and receiving tacrolimus in our center were included. Fast or slow tacrolimus metabolizers were classified according to the time spent with a C0/D < 1.05 (> 75% = High, < 25% = Low) at various time points posttransplantation. We first determined the earliest accurate time for patient categorization by investigating C0/D variability during the first months. Second, a multivariate cause-specific Cox model studying allograft outcomes was performed in groups identified by their status determined from the earliest accurate timepoint after transplantation.

Results: Among 1979 patients included in the analysis, 2 months was the earliest accurate timepoint to determine High patients (85% of High patients identified at 2 months remained High long-term, Brier score = 0.06). Multivariate analysis revealed that High patients determined at 2 months (n = 499) had a significantly higher risk of allograft loss (cause-specific hazard ratio [CS-HR] = 2.00, 95% confidence interval [CI] = 1.48-2.69) and allograft rejection (CS-HR = 1.71, 95% CI = 1.15-2.54) than Low patients after adjustment for confounding factors. Moreover, allograft function was lower in High patients (46.7 vs. 52.9 ml/min, at 3 years, P < 0.0001) with a higher proportion of chronic vascular lesions at 1 year.

Conclusion: C0/D is a simple and pragmatic tool capable of identifying patients at risk of rejection and allograft failure as early as the second month posttransplantation.

Keywords: allograft survival; kidney transplantation; tacrolimus toxicity.

Graphical abstract

Figure 1

(a) Evaluation of tacrolimus trough level among groups defined from T = 2-months during the first 3 years posttransplantation. (b) Evaluation of C0/D ratio among groups defined from T = 2 months during the first 3 years posttransplantation. (c) Density plot of all measured C0/D values among groups defined from the second month posttransplantation. C0/D, tacrolimus trough-level / total daily dose.

Figure 2

(a) Long-term confounder-adjusted graft survival from T = 2 months posttransplantation according to the patient’s status (High vs. Low) estimated from the weighted Kaplan-Meier estimator. (b) Long-term confounder-adjusted graft survival from T = 2 months posttransplantation according to the patient’s status (High vs. Variable) estimated from the weighted Kaplan-Meier estimator. (c) Long-term confounder-adjusted graft survival from T = 2 months posttransplantation according to the patient’s status (Variable vs. Low) estimated from the weighted Kaplan-Meier estimator.

Figure 3

(a) Long-term confounder-adjusted death-censored occurrence of rejection from T = 2 months posttransplantation according to the patient’s status (High vs. Low) estimated from the weighted Kaplan-Meier estimator. (b) Long-term confounder-adjusted occurrence of de novo DSA from T = 2 months posttransplantation according to the patient’s status (High vs. Low) estimated from the weighted Kaplan-Meier estimator. DSA, donor-specific antibody.

Figure 4

(a) Long-term confounder-adjusted occurrence of severe infectious complications from T = 2 months posttransplantation according to the patient’s status (High vs. Low) estimated from the weighted Kaplan-Meier estimator. (b) Long-term confounder-adjusted occurrence of CMV viremia from T = 2 months posttransplantation according to the patient’s status (High vs. Low) estimated from the weighted Kaplan-Meier estimator. (c) Long-term confounder-adjusted occurrence of BKV infection from T = 2 months posttransplantation according to the patient’s status (High vs. Low) estimated from the weighted Kaplan-Meier estimator. (d) Long-term confounder-adjusted occurrence of posttransplant diabetes mellitus from T = 2 months posttransplantation according to the patient’s status (High vs. Low) estimated from the weighted Kaplan-Meier estimator. BKV, BK polyomavirus; CMV, cytomegalovirus.

Figure 5

(a) Evaluation of allograft function estimated by eGFR (MDRD) among groups during the first years censored by allograft failure. (b) Evaluation of allograft function estimated by eGFR (MDRD) among groups during the first years censored by allograft failure and occurrence of a rejection episode. (c) Occurrence of severe chronic injuries in patients from High, Variable, and Low groups defined from the second month posttransplantation, in for cause and protocolar biopsies at 3 and 12 months posttransplantation. eGFR, estimated glomerular filtration rate; MDRD, modification of diet in renal disease.