Tacrolimus (Tac), a calcineurin inhibitor, is the cornerstone of immunosuppressive therapy in solid organ transplantation. This drug is very effective in preventing acute rejection, but clinical dose adjustment remains difficult due to its narrow therapeutic window and high intra- and inter-patient pharmacokinetic (PK) and pharmacodynamic (PD) variability.1,2 Perhaps the greatest challenge is to find a therapy that adapts to the individual characteristics of the patient achieving a good balance between efficacy and toxicity.3 For thirty years, dose adjustment of Tac in clinical practice has been done empirically according to the drug's summary of product characteristics. The calculation of the initial dose of Tac after solid organ transplantation is based on the patient's body weight. Subsequent doses are adjusted based on the pre-dose morning concentration (C0) of Tac for therapeutic drug monitoring (TDM) at the discretion of each centre.3

Tac metabolism and PK profiles are becoming increasingly important. The individual's rate of Tac metabolism has been linked to outcomes after transplantation, further exacerbating the problem of inter-individual differences in dose-response relationships.2 Exposure to Tac has been evaluated based on genetic polymorphisms of cytochrome P450 (CYP3A4 and CYP3A5), which modifies the metabolism of Tac.4–10 In order to define the population based on the pharmacogenetic characteristics of the cluster and the differences in the metabolisation of Tac, three phenotypes have been described; fast, intermediate and slow metabolisers. Fast metabolisers have low levels of Tac, requiring at least twice the dose of slow metabolisers.6,9,11

The concentration/dose approach (C/D) has been analysed as a simple tool to estimate the metabolism of Tac.12–18 In these studies, a low C/D ratio (C/D < 1.05 (ng/mL)/mg in kidney transplant recipients; <1.09 (ng/mL)/mg in liver transplant recipients) indicates rapid metabolism of Tac and it is associated with a decreased renal function, a higher rate of biopsy-proven nephrotoxicity and more frequent switching to an alternative immunosuppressive regimen than patients with a higher C/D ratio.12,14–16,19–22

Both strategies, the study of polymorphisms and the C/D ratio, focus on Tac metabolism and are suggested as additional tools for personalising the dose of Tac in transplant patients. The goal is to personalise Tac treatment and limit the time that transplant recipients are under- or over-exposed to Tac, in order to minimise the risks of rejection and toxicity, respectively. This study evaluated both strategies to identify whether the stratification of patients according to the PK of C/D corresponds to their classification according to the metaboliser profile of the cluster CYP3A4/5, and whether there is any association between either classification and under- or over-exposure to Tac, intra-individual variability or the outcome of the transplanted organ.

The study included a total of 425 patients who received a kidney transplant from 2013 to 2020 (including both years) (Fig. 1). The study was carried out in accordance with the Declaration of Helsinki and with the approval of the local ethics committee of the Hospital Universitario de Bellvitge, Spain.

Fig. 1.

Flowchart of the patients included in the study.

(0.54MB).

Patients included were taking a formulation of Tac every 12 h and had a DNA sample available-*. Those who were switched to other 24-h formulations of Tac were excluded. Patients who died or suffered graft loss in the first month after kidney transplantation and those who had incomplete biochemical data for measuring the concentration/dose ratio (C/D) during follow-up were also excluded. The usual immunosuppressive therapy they received consisted of antithymocyte globulins (ATG) or basiliximab as induction therapy, followed by Tac and mycophenolate mofetil (MMF) and steroids as maintenance immunosuppression. Tac concentration and dose was monitored on days 7 and 15, and at one, three, six and twelve months post-transplant.

Demographic data on the donor and recipient, type and number of transplants, aetiology of kidney disease, panel reactive anti-HLA antibodies (PRA) and histocompatibility in HLA A, B and DR antigens, were collected. Serum creatinine (μmol/l) levels were collected at day five, seven and 15, and at one, three, six and 1twelve months after transplantation. The presence of delayed graft function and biopsy-proven acute rejection was assessed following histological classification according to the Banff Classification (2018).23 Delayed graft function was defined as the need for dialysis sessions during the first postoperative week. Graft loss was defined as return to dialysis or re-transplantation.

Metabolism of tacrolimus according to the C/D ratio

Following the guidelines described in the article by Thölking et al.,12 the Tac metabolism rate was determined at months one, three and six after kidney transplantation by dividing the blood Tac concentration by the corresponding daily dose of Tac (D).

C/D = pre-dose blood Tac levels (ng/mL)/Tac daily dose (mg).

The mean of the C/D ratio for months one, three and six was used to categorise the three patient groups according to their Tac metabolism rate (fast, intermediate and slow). Patients were evenly distributed into three groups. Intermediate metabolisers were grouped around the median of the C/D ratio (1.33–2.156). Patients with a C/D ratio <1.33 (ng/mL)/mg were defined as fast metabolisers and ≥2.156 (ng/mL)/mg as slow metabolisers (Fig. 2).

Fig. 2.

Histogram of the distribution of the variable mean of the Tac C/D (ng/mL)/mg values at 1, 3 and 6 months post-transplant. Patients with C/D < 1.33 (ng/mL)/mg (percentile 33.3%) were considered fast metabolisers, C/D ≥ 2.156 (ng/mL)/mg (percentile 66.7%) slow metabolisers and C/D 1.33-2.156, intermediate metabolisers.

(0.18MB).

Maintaining a good immunosuppressive regimen, particularly in the early stages after kidney transplantation, is crucial for ensuring a good long-term prognosis for the graft. The high intra- and inter-patient variability of Tac make correct dosage adjustment challenging. Knowing the metaboliser phenotype of the transplant recipient can be helpful.

In recent years, several studies have proposed calculating the C/D ratio as an estimate of the transplant patient's Tac metabolisation rate.12–22 However, there is no agreed criterion for C/D studies. Thölking et al.12 divided the population into three groups according to the percentiles of their cohort, defining a C/D ratio below 1.05 (ng/mL)/mg for fast metabolisers, from 1.05 to 1.55 (ng/mL)/mg for intermediate metabolisers, and above 1.55 (ng/mL)/mg for slow metabolisers. The value of 1.05 (ng/mL)/mg to define the fast metaboliser population has also been used in other studies, although with variations in the study population.14,19,21 In our cohort, the median of the C/D ratio was higher than that reported by Thölking12 (1.74 and 1.29, respectively). If we had classified our patients according to the values proposed by Thölking et al.,12 the distribution of patients according to their metabolisation would not have been equal (53 fast, 81 intermediate and 191 slow). For that reason, we classified the patients according to the limit values established in the tertiles (C/D < 1.33 [ng/mL]/mg fast metabolisers, C/D 1.33–2.156 [(ng/mL]/mg intermediate metabolisers and C/D ≥ 2.156 [ng/mL]/mg slow metabolisers). Other authors have also used this strategy to establish the interval limit values.17,18 These discrepancies in the median values for C/D obtained in the different studies could be explained both by differences in the ethnicity of the patients included in the studies, and by the different therapeutic ranges established at each hospital.

In an attempt to establish the metaboliser groups, there are C/D studies in the literature that classify patients into two groups (fast and non-fast metabolisers)14,17,18,21,22 and others with three groups (fast, intermediate and slow). Also, and more importantly, there are discrepancies in both the number of C/D measurements to take into consideration for calculating the ratio and when they should be made throughout the post-transplant period.12,14,17,21 In our study, the C/D values increased over time, just as the dose of Tac decreased. Consequently, depending on the strategy used to establish the C/D ratio, the same patient could belong to different metaboliser groups.

By definition, the phenotype represents the result of the interaction between the individual's genotype and their relationship with the environment. To introduce genetic terms into the metabolisation of Tac, the polymorphisms of its metaboliser enzymes have to be taken into account, primarily CYP3A4 and CYP3A5.6,9,26 Following the genetic criterion, the cohort of patients in this study was divided into three groups with an uneven number of patients (82 fast metabolisers, 310 intermediate and 33 slow metabolisers) as a consequence of the low prevalence of the alleles CYP3A4*22 and CYP3A5*1 in the Caucasian population.6,27 In our study, fast metaboliser patients required at least twice the dose of slow metabolisers to maintain blood Tac levels in the therapeutic range, corroborating previously reported results.9,26 The dose differences led to a significantly different C/D ratio among the three groups defined according to fast, intermediate or slow genotype based on the cluster of CYP3A4 and CYP3A5 polymorphisms.

We compared the proportion of patients in each metaboliser group established according to genotype that coincided with the homonymous group defined according to the C/D ratio. The results showed that not all patients considered fast metabolisers according to genotype had low C/D ratios. In turn, the same discordant results were observed when analysing the intermediate population and the slow population with both criteria. To understand this discrepancy, taking into account that the rate of clearance of a drug slows down with age, the age variable was considered as a possible factor to explain these differences. The results showed that the age of the patients in the groups established according to C/D ratio was significantly different, with young patients having lower values and older patients having higher values, as previously reported in the literature.12,14,17 When we analysed the ages within each metaboliser group established according to genotype, we found a clear tendency for older patients to have high C/D ratios and vice versa. In line with other studies,11,28–30 these findings reinforce the idea that, apart from the genetic criteria, other variables (adherence to treatment, diarrhoea, dose of corticosteroids, plasma albumin, haematocrit) should also be taken into consideration for the personalised adjustment of Tac. To integrate all factors that influence Tac exposure, population pharmacokinetics (PopPK) models have been developed that include CYP3A genetic polymorphisms along with other factors that also affect Tac pharmacokinetics (age, plasma albumin, body surface area, co-medication and haematocrit).1,31,32

Periods in which patients are under- or over-exposed to Tac increase the risk of rejection and toxicity, respectively. With the two classification criteria of the metaboliser phenotype of the patients analysed in this study, the fast metaboliser patients showed a higher percentage of underexposure to Tac than the other groups. Patients considered slow metabolisers had a higher percentage of overexposure, which could be explained by their lower rate of total drug clearance. Slow metabolisers according to C/D also had greater intra-individual variability, which could be inherent to the difficulty of adjusting the dose in these patients. This trend was not seen in the case of slow metabolisers according to genotype, possibly because the sample size was smaller and the patients more homogeneous, compared to the slow metabolisers according to C/D, who may have different metaboliser genotypes.

Lastly, we investigated the influence of the metaboliser groups on the outcome of the graft over the first year post-transplant, considering creatinine levels and the incidences of delayed graft function, acute rejection and graft loss. No differences were detected between metaboliser groups regardless of the classification criterion. In previous studies that took genotype into account, no differences were found during the first year after transplantation.9 Studies that analysed longer-term outcomes of patients (two to five years) found that patients who are fast metabolisers may have a higher incidence of nephrotoxicity,19 worse kidney function,12,16,18,22 and poorer graft survival.14 However, there are studies that found no significant differences between metaboliser groups in relation to long-term graft progress,15,17 probably due to their classification system. Bartlett et al.17 divided the population into two groups (fast and non-fast metabolisers) with a cut-off C/D ratio of 2.04 (ng/mL)/mg. This value is higher than that of the rest of the studies, which consider fast metabolisers to be those with a C/D ratio of around 1.05 (ng/mL)/mg. Bartmann et al.15 considered C/D ratios from the immediate post-transplant period when there is more variability. It would be interesting for our study to do a long-term follow-up of the patients in order to analyse whether the incidence of events long-term is the same in the metaboliser groups established according to the two proposed classification criteria.

Knowing the metaboliser group according to genotype gives us the advantage of adapting the Tac dosage according to the transplant patient's metaboliser phenotype from the beginning, improving Tac exposure. Monitoring of the changes in the Tac dose over the course of the post-transplant follow-up period has resulted in the recommendation that fast metabolisers require 60% higher doses than intermediate metabolisers, while slow metabolisers require 20% less than intermediates. In contrast, although simpler, the classification of patients according to the C/D ratio has the limitation of requiring data to be gathered during the course of the patient's post-transplant follow-up and, consequently, it cannot be used as a tool to assist dosing in the immediate period after transplant when dose adjustment most needs to be fine-tuned. However, the C/D classification would help us identify which patients are most at risk in the long term.

In conclusion, in our cohort of patients, the determination of their metaboliser phenotype both according to genotype and according to the C/D ratio made it possible to distinguish between patients according to their exposure to Tac, without finding differences in adverse events for the graft in the first year after kidney transplantation. In fact, the combination of the two classification criteria would probably be a good tool for managing Tac dosage for transplant patients. PopPK modelling of Tac incorporating these parameters could be a step forward towards personalised therapy in organ transplantation, by fine-tuning the adjustment of the Tac maintenance dose.

Key concepts

• 1

  The cluster study of the CYP3A4 and CYP3A5 polymorphisms has defined three populations that exhibit differences in Tac clearance. Patients carrying the functional allele *1 require at least twice as many doses as those who are not carriers.

• 2

  The study of the C/D ratio divides the population into tertiles to define the metabolism of Tac. It is proposed as an additional tool for monitoring the adjustment of Tac in transplant patients.

• 3

  It is important to use new tools to be able to personalise Tac dose adjustment and limit the time in which transplant recipients are under- or over-exposed, in order to minimise the risks of rejection and toxicity, respectively. Population pharmacokinetics (popPK) models that include different factors that may explain Tac variability can help to optimise personalised Tac dose adjustment post-transplant.

This study was funded with a grant from the Instituto de Salud Carlos III [National Institute of Health Carlos III] and the Spanish Ministry of Health and Consumer Affairs (PI18/01740), Sociedad Española de Nefrología [Spanish Society of Nephrology] (17PSJ026), and the European Regional Development Fund (ERDF) co-financed by ERDF funds from the programme "European Regional Development Fund (ERDF), a way of making Europe".

The authors declare that they have no conflicts of interest.