PHARMACOKINETICS AND TOLERABILITY OF 40–0-[2-HYDROXYETHYL]RAPAMYCIN IN DE NOVO LIVER TRANSPLANT RECIPIENTS¹

Abstract

Background. 

40–0-[2-Hydroxyethyl]rapamycin (RAD), a novel macrolide with potent immunosuppressive and antiproliferative activities, prevents rejection in animal allotransplantation models. This phase I trial assessed the effects of bile diversion, administration route, and time after transplant on RAD pharmacokinetics after single-dose administration in de novo liver allograft recipients. The influence of RAD on cyclosporine (CsA) pharmacokinetics and the safety of RAD were also evaluated.

Methods. 

Twenty-six de novo liver allograft recipients were assigned to one of four treatment groups based on the presence or absence of a T tube, administration route (nasogastric or nasoduodenal), and timing of RAD administration. Patients received a single 7.5-mg RAD dose on one to three occasions in addition to CsA (Neoral^®) and corticosteroids. Steady-state cyclosporine profiles with and without RAD coadministration were evaluated.

Results. 

Recipients with bile diversion demonstrated lower peak concentration (C_max) than those without, but overall drug exposure (AUC) was not altered. C_max and AUC were not influenced by administration route. A trend towards higher C_max on postoperative day 3 than on postoperative day 1 was noted, although AUC was not altered. Single-dose RAD coadministration did not affect steady-state CsA pharmacokinetics. RAD was well tolerated and caused few drug-related adverse effects. RAD administration did not increase infection rates or produce clinically significant changes in laboratory parameters.

Conclusions. 

In de novo liver transplant recipients, the overall extent of RAD absorption was not influenced by bile diversion, administration route, or time of administration. CsA pharmacokinetics were not affected by single-dose RAD coadministration. RAD capsules administered in single doses of 7.5 mg were well tolerated and safe.

Although liver transplantation has dramatically improved clinical outcomes in patients with end-stage liver disease, complications such as acute and chronic rejection contribute to patient morbidity and graft loss. Current immunosuppressive regimens are associated with adverse effects, such as nephrotoxicity, neurotoxicity, and infections. Consequently, there is a need for novel immunosuppressive agents that increase effectiveness, but minimize long-term toxicities.

40–0-[2-Hydroxyethyl]rapamycin (RAD) is a novel macrolide with potent immunosuppressive and antiproliferative activities (¹). It is related to, but structurally different from sirolimus and, by virtue of its hydroxyethyl side chain, has enhanced bioavailability (²). RAD exerts its immunosuppressive activity at a later stage of the cell cycle than either cyclosporine (CsA) or tacrolimus, resulting in cell arrest in the late G₁ phase, just before it enters the S phase (³). RAD inhibits growth factor-dependent proliferation of vascular smooth muscle cells in addition to hematopoietic cell lines (¹). In preclinical models, RAD has shown potential to prevent acute and chronic rejection (^{1, 4, 5}) and acts in synergy with CsA (⁶). In clinical trials, RAD co-administered with CsA to kidney transplant recipients resulted in a reduction in the incidence and severity of acute rejection (⁷).

Our phase I study is the first report of RAD in orthotopic liver transplantation. The objectives were to determine the effects of external bile diversion, route of administration (i.e., nasogastric [NG] vs. nasoduodenal [ND] tube), and time after transplant on the whole-blood pharmacokinetics of RAD after single-dose administration on one, two, or three occasions. The effect of a single dose of RAD on the kinetics of CsA was also studied by determining CsA pharmacokinetic profiles just before and after RAD administration. Finally, this study evaluated the tolerability and safety of RAD.

The study population consisted of 26 de novo liver transplant recipients between 18 and 70 years of age. The study was approved by the appropriate institutional ethics board and written informed consent was obtained from all patients before enrollment. All patients were administered an immunosuppressive regimen consisting of CsA (Neoral^®) and corticosteroids (i.e., methylprednisolone or prednisone). Patients were excluded from the study if they met any of the following criteria: recipient of multiple organ transplants; evidence of graft primary nonfunction within the first 48 hr after transplantation; poor renal function (i.e., serum creatinine >300 μmol/L at baseline or serum creatinine increase by >50 μmol/L on three consecutive days before baseline); renal dialysis before transplantation; severe cardiac disease; clinically significant abnormal laboratory values at baseline; or positivity for the human immunodeficiency virus, hepatitis B surface antigen, or hepatitis C antibody.

The study consisted of three treatment periods: week 1 consisted of study days 1–7; week 5 consisted of study days 29–35; and week 6 consisted of study days 36–42. Patients were assigned to one of four treatment groups based on the presence or absence of a T tube, route of drug administration (via NG tube or ND tube), and timing of administration of the first RAD dose (Table 1). Patients were administered a single oral dose of RAD (7.5 mg) on one, two, or three occasions, depending on their respective treatment groups.

Table 1: Study design

A hard gelatin capsule formulation of RAD at dosage strengths of 0.25 and 1 mg was used. RAD capsules were opened and the contents mixed with 25 ml of water in a glass container; administered via NG tube, ND tube, or orally; and then followed by 125 ml of water. Baseline immunosuppression consisted of CsA (10–15 mg/kg per day) divided in two doses, plus corticosteroids given according to the local center’s regimen.

On the days of pharmacokinetic profiling, the study drug was administered in the morning after a 12-hr overnight fast with only water allowed. After the first dose of RAD, patients fasted (except for oral medication) for an additional 2 hr. For the second and third RAD doses, each patient ingested 100 ml of fluid/hr, from 1 hr before drug ingestion until 12 hr thereafter. Patients fasted for 4 hr after administration of the second and third doses of RAD.

RAD single-dose pharmacokinetic profiles consisted of 21 postdose samples over a 6-day period. Venous blood samples were obtained before dosing and then 0.25, 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 4, 6, 8, 10, 12, 24, 36, 48, 72, 96, 120, and 144 hr thereafter.

Steady-state CsA profiles were assessed in a subset of patients on the day before the last RAD dose (CsA alone) and again on the day of the last RAD dose (coadministration). Venous blood samples of CsA were obtained before dosing, and then 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 6, 8, 10, and 12 hr after dosing.

Concentrations of RAD in whole blood were measured by a liquid chromatography, mass spectrometric method. The assay quantification limit was 0.3 ng/ml. CsA was measured in whole blood by a radioimmunoassay method using a monoclonal antibody specific for the parent compound. The assay quantification limit was 5 ng/ml.

Standard noncompartmental pharmacokinetic parameters were derived from the blood-concentration-time profiles of RAD and CsA. For RAD, these included the peak concentration (C_max) and the time of its occurrence (t_max), the area under the concentration-time curve (AUC), and the elimination half-life (t_1/2). For steady-state CsA profiles, C_max and t_max were defined as stated previously, the predose trough concentration was designated C_min, and AUC was determined over the 12-hr dosing interval.

The influence of external bile diversion on the pharmacokinetics of RAD was addressed from two perspectives: presence/absence of a T tube in a parallel-group comparison; and open/closed T tube in a cross-over comparison. For the former evaluation, pharmacokinetic parameters from week 1 and week 5 in patients from treatment groups I (with T tube) and II (without T tube) were assessed in an analysis of variance with the following sources of variation: T-tube status, day after transplant, T-tube-by-day interaction, and patient nested within T-tube status. For the cross-over evaluation, pharmacokinetic parameters from week 5 (T tube open) and week 6 (T tube closed) in patients from treatment group I were compared with a paired Student’s t-test. The effect of administration route and time after transplant on the pharmacokinetics of RAD was assessed with an unpaired Student’s t-test.

To explore the interpatient variability and covariates of RAD pharmacokinetics, the largest pharmacokinetic data set pooled from patients treated under the same administration conditions was obtained from group I in week 6 and group II in week 5 (n=11). This represented data from oral administration without external bile diversion approximately 1 month after transplant. The interpatient coefficient of variation (%CV) was calculated for C_max and AUC as standard deviation divided by the mean×100. The influence of demographic characteristics on the pharmacokinetics of RAD was explored by conventional linear regression for age and weight and by unpaired Student’s t-test for gender. The percent contribution of weight to interpatient variability was quantified based on the coefficient of determination (r2 value) from the regression analysis.

Safety variables consisted of adverse events, rejection episodes, physical examination findings, vital signs, laboratory values, electrocardiographic data, and bile data. All patients (n=26) were included in the safety analysis, and 25 were evaluable for pharmacokinetic assessments.

There were no clinically relevant differences between treatment groups for any baseline demographic variables. The mean ages for groups I and II were 53±12 and 53±10 years, whereas, for groups III and IV, the mean ages were 40±11 and 47±17 years, respectively. In group I, 83% of the patients were men and in groups II and III, 63% and 67% of the patients were women, respectively. In group IV, 50% of the patients were male and 50% were female. The majority of patients (≥83%) in all treatment groups were white. The minimum weight for all treatment groups was 62 kg, and the maximum weight was 113 kg.

The causes of pretransplant liver disease varied among the four treatment groups. The most commonly reported causes were as follows: in group I, alcoholic cirrhosis (50%); in group II, sclerosing cholangitis (38%); in group III, primary biliary cirrhosis (33%) and alcoholic cirrhosis (33%); in group IV, cryptogenic cirrhosis (33%). There were no clinically relevant differences observed in the liver transplantation history variables. One patient in group III (4%) had hepatitis at baseline.

Single doses of RAD were well tolerated. RAD administration was not associated with any specific adverse events. Few of the adverse events were drug-related, occurring in four patients in week 1 (thrombocytopenia, vomiting, hot flashes), one patient in week 5 (anemia), and two patients in week 6 (flushing, anemia/leukopenia). Single-dose RAD did not produce clinically significant effects on hematologic values, cholesterol levels, triglyceride levels, or renal- or liver-function tests. Alterations in laboratory parameters reflected the underlying disease or were consistent with events generally seen in the postoperative course of patients who have undergone liver transplantation. Moreover, there was no increase in infectious complications after RAD administration. Infection episodes were reported in 46%, 9%, and 17% of patients in weeks 1, 5, and 6, respectively. Most infections were scattered throughout the body systems in all groups, with no predominance in any one body system or treatment group. The incidence and type of infections reported were consistent with those generally seen in the early posttransplant period in liver transplant recipients.

After a 30-day follow-up period, overall patient survival and graft survival were 96% and 92%, respectively. Four patients discontinued the study; two withdrew consent, one withdrew because of acute rejection, and one discontinued because of multiple organ failure and sepsis and subsequently died. None of these adverse events or deaths was attributable to the study drug.

This study examined the pharmacokinetics of RAD in the presence of clinical factors that may affect drug disposition in the posttransplant period in de novo liver transplant recipients. Patients who had a T tube (group I) demonstrated a significantly lower C_max than those without a T tube (group II;P =0.01, Table 2). Over the first transplant month, C_max increased significantly by 41% in recipients with a T tube and by 26% in recipients without a T tube (P =0.02). Of most importance, the overall extent of drug absorption (AUC) was not affected by the presence or absence of bile in the immediate posttransplant period (P =0.30) or at 1 month after transplant (P =0.72). Mean elimination t_1/2 was higher in recipients with a T tube in week 1, although the differences did not reach statistical significance (60±24 hr vs. 39±14 hr;P =0.10) and appeared to be influenced by two outlier values among the recipients with a T tube. t_1/2 decreased significantly by week 5 in recipients both with and without a T tube (P =0.03).

Table 2: Effect of external bile diversion on pharmacokinetic parameters of RAD^a

The intrapatient influence of external bile diversion on the pharmacokinetics of RAD was determined by a cross-over evaluation performed in patients from group I (n=6) when the T tube was open (week 5) and when the T tube was closed (week 6). To ensure adequate bile output was occurring in these patients, the T tube was opened 12 hr before pharmacokinetic assessments. Patients were only included in the evaluation if a minimum of 300 ml of bile was collected from this procedure. Results indicated that the presence or absence of bile in the gastrointestinal tract did not significantly affect the C_max (35±10 ng/ml with T tube open vs. 46±19 ng/ml with T tube closed;P =0.56) or the AUC (608±234 ng·hr/ml with T tube open vs. 608±226 ng·hr/ml with T tube closed;P =0.98) of RAD. When the T tube was closed, t_max occurred earlier than when the T tube was open (1.5 [1–2.5] hr vs. 2.5 [2–3.1] hr;P <0.05).

The influence of administration route on the pharmacokinetics of RAD was determined by a comparison of pharmacokinetic parameters after administration via ND tube (group II) and after administration via NG tube (group III) during week 1 (Table 3). The route of administration did not significantly affect the pharmacokinetic profile of RAD, including the overall extent of drug absorption.

Table 3: Effect of administration route (ND vs. NG tube) on pharmacokinetic parameters of RAD^a

Poor liver function or gastrointestinal abnormalities present in the immediate postoperative period did not appear to significantly affect RAD absorption. A comparison of recipients given RAD on postoperative day (POD) 1 (n=6) and recipients given RAD on POD 3 (n=5) demonstrated that both the rate (t_max=3 hr vs. 2 hr;P >0.05) and extent of drug exposure C_max (20±9 vs. 33±9 ng/ml;P =0.07) increased somewhat with time after transplant, although the results did not reach statistical significance. On POD 1, total drug exposure (AUC) was only 25% lower than on POD 3, indicating good RAD absorption even in the immediate postoperative period (606±420 ng·hr/ml on POD 1 vs. 811±416 ng·hr/ml on POD 3;P =0.34).

The interpatient coefficients of variability for C_max and AUC were 35% and 34%, respectively, in 11 patients (without bile drainage) assessed 1 month after transplant. Neither age nor weight was significantly correlated with these parameters, and each contributed <5% to the respective pharmacokinetic variabilities, suggesting that dosing RAD by weight may not be of any clinical benefit.

Steady-state CsA pharmacokinetic profiles were evaluated in a total of seven patients (n=3 from group I, T tube present; n=4 from group II, T tube absent). Results indicated coadministration of single-dose RAD with CsA did not significantly affect the steady-state pharmacokinetics of CsA, regardless of whether the patients had external bile diversion or not (Table 4). This finding is important because RAD is intended to be used as part of a CsA-based maintenance immunosuppressive regimen. Clearly, this needs to be confirmed at steady state for both drugs in a broader population.

Table 4: Effect of single-dose RAD administration on pharmacokinetic parameters of CsA^a

The pharmacokinetic results of this study in liver transplant recipients are consistent with those obtained in renal transplant recipients. In a single-dose pharmacokinetic study that evaluated four ascending doses of RAD in kidney transplant recipients, the 7.5-mg dose of RAD exhibited a similar overall extent of absorption as in de novo liver transplant recipients (⁸). Furthermore, RAD administration did not affect the steady-state pharmacokinetics of CsA in this single-dose study and in two multiple-dose studies in renal transplant recipients (^{2, 9}).

In conclusion, results of this first phase I study of RAD in de novo liver transplant recipients demonstrate that, during the postoperative period, when drug absorption is often problematic, RAD exhibits an excellent absorption profile. RAD is not significantly affected by external bile diversion, route of administration, or time of posttransplant administration, and our results suggest that no dose adjustments over time will be necessary in the posttransplant period. Moreover, single-dose RAD administration does not affect the pharmacokinetics of CsA, an important observation given that these two immunosuppressive agents will most likely be used concomitantly. Additionally, single doses of RAD are well tolerated and have a good safety profile. The results of this study are promising; additional trials assessing the efficacy of RAD in de novo liver transplant recipients are now ongoing.