Pathophysiological Implications of Variability in Blood Tacrolimus Levels in Pediatric and Adolescent Kidney Transplant Recipients

Improving long-term kidney graft survival has been the focus of many studies in recent years. Immune-mediated graft injury due to insufficient immunosuppression is a major contributor to graft loss, and it is often associated with the formation of de novo HLA donor-specific antibodies (DSAs). Low immunosuppressive drug concentrations and fluctuations in levels may be due to several factors, including genetic polymorphisms that affect drug absorption and metabolism, feeding problems, intestinal malabsorption, and interaction with foods or other medications. The challenge of nonadherence in transplant recipients, in particular of the adolescent age group, is well known to pediatric nephrologists and other clinicians caring for these patients, and it is probably the most frequent cause of inadequate immunosuppression (1). In this issue of CJASN, Piburn et al. (2) characterize the baseline pattern of tacrolimus intrapatient variability in pediatric kidney transplant recipients and then examine its relationship with graft outcomes. This retrospective study included all pediatric kidney-only transplants performed at a single center between 2004 and 2018. All available tacrolimus levels measured were collected from the electronic health record, and intrapatient variability was calculated from levels obtained during the 6-month period prior to each level. A subset of patients transplanted after 2010 had routine monitoring of de novo DSAs performed as part of surveillance screening as well as when clinically indicated. The authors found that mean tacrolimus coefficient of variation (reflecting fluctuations in drug concentrations) was high during the first few months after transplantation as expected, probably due to frequent changes in target trough levels and dosing, as well as medication interactions. The coefficient of variation stabilized at 10 months after transplantation, with a median value of 30% (interquartile range, 21%– 41%). Thirty-one recipients of 220 in the outcome cohort (14%) developed C1q-binding de novo DSAs >10 months after transplantation. A significantly higher tacrolimus coefficient of variation was found in these patients compared with those who did not develop antibodies (38% versus 28%). The risk of forming de novo DSAs was higher in patients with a tacrolimus coefficient of variation above 30%, and the association was strongest in the top quartile (intrapatient variability >41%) both for C1q-binding de novo DSAs and for anti-HLA class 2 antibodies. One of the strengths of the study, which examines data from 426 patients, is the large number of tacrolimus levels analyzed—over 31,000 in total and a median of 64 levels per patient. Higher variability was seen in the youngest patients, attributed to feeding intolerance and frequent viral infections, but surprisingly, it was not demonstrated in the adolescents, who are known to be prone to nonadherence. The study has limitations, which the authors list in the discussion section, including the retrospective nature and the "real-world" setting of the study. Tacrolimus levels were measured in different laboratories by various methods, and information is lacking regarding the actual timing of trough levels, the frequency of testing, and concomitant medications. On the other hand, these limitations are probably more likely to bias toward the null hypothesis, and the link shown between tacrolimus intrapatient variability and de novo DSA formation seems, therefore, to be robust. Real-world data reflect clinical practice, complement evidence obtained from high-quality clinical trials, and may be of particular interest when examining the effects of patient nonadherence. Tacrolimus is an integral part of chronic immunosuppression in solid organ transplantation and has become almost universal in the treatment of pediatric and adolescent kidney transplant recipients. Previous studies have attempted to tease out the effect of tacrolimus exposure and graft outcomes: whether low exposure (persistently low drug levels) or intrapatient variability (fluctuations of drug levels) is associated with antibody production, acute rejection episodes, lower graft function, and graft failure. In a study of 803 adult kidney graft recipients, high tacrolimus coefficient of variation was associated with higher risk of graft loss in patients exposed to low drug levels but not in those not exposed to low tacrolimus levels (3). Maintaining high levels of tacrolimus has been shown to reduce formation of de novo DSAs in patients with HLA class 2 molecular mismatch (4). A large pediatric study showed that both low tacrolimus exposure and high intrapatient variability are associated with graft loss (5). The link between tacrolimus intrapatient variability and de novo DSA formation in pediatric kidney transplant recipients has been reported previously in smaller studies, and this study supports this observation (6). C1q-binding de novo DSAs are associated with biopsy-proven acute rejection and graft loss compared with C1q-negative DSAs in pediatric kidney transplantation, underscoring the importance of these antibodies for graft outcomes (7). Questions remain on the effect of other immunosuppressive agents and drug combinations on the development of de novo DSAs, acute rejection, and graft loss. Therapeutic drug monitoring of mycophenolic acid (MPA), the active metabolite of mycophenolate mofetil, is not performed in most centers. In one pediatric study, the intrapatient variability of MPA was higher than that of tacrolimus (8). However, data are mixed as to the effect of MPA levels and intrapatient variability on DSA formation and donor-reactive T lymphocytes (9). The study by Piburn et al. (2) was not equipped to evaluate whether corticosteroids modified the effect of tacrolimus on de novo DSA formation, as only patients with a high immunologic risk were prescribed prednisone (2). A multidisciplinary team approach to the long-term care of pediatric, adolescent, and young adult kidney transplant recipients should include repeated assessment of impediments to adherence to immunosuppression. These might be a combination of medical insurance issues, socioeconomic factors, side effects of medications, scheduling problems, and behavioral issues. Interventions aimed at reducing these barriers decrease the risk of acute rejection in this population (10). Conversion to a once daily extended release tacrolimus formulation has been shown to reduce tacrolimus intrapatient variation in adolescents and young adults, but further study is necessary to determine the effects on graft outcomes in this age group (11). Extended release formulations are not yet approved in patients under the age of 18. Patient education regarding drug and food interactions and care during diarrheal illness is also important in maintaining low tacrolimus coefficient of variation and, consequently, stable immunosuppression. Beyond monitoring of immunosuppressive drug level and variability, there remains the biologic effect of the treatment, which may differ significantly between individuals. Quantifying the degree of suppression of the immune response in organ transplantation remains elusive and is the focus of many studies. Clinical assessment of medication adherence should be an integral part of all routine clinic visits of pediatric and adolescent kidney transplant recipients. Incorporating the tacrolimus intrapatient coefficient of variation as a parameter in the patient electronic health record, easily accessible to the clinician during patient visits, could add a valuable tool to patient management. Early detection of insufficient immunosuppression and timely intervention could prevent the formation of de novo DSAs and adverse graft outcomes. Disclosures The author has nothing to disclose. Funding None.