Annick Rousseau, Aurélie Prémaud, Pierre Marquet
Department of Pharmacology and Toxicology, University Hospital, Limoges, France
Background: In the early post-grafting period, the pharmacokinetic profiles of immunosuppressive drugs are complex. The time-blood concentration curves of cyclosporine (CsA) frequently exhibit delayed absorption and secondary peaks. Complex absorption profiles and time-dependent clearance were reported for mycophenolate (MPA), whose pharmacokinetic profiles also frequently present multiple peaks. Clinical evidence advocates for the rapid achievement of sufficient and stable exposure of these drugs in the early post-transplantation period, which may require dose adjustment based on pharmacokinetic modeling (1).
Objectives: To develop a pharmacokinetic model able to fit the complex pharmacokinetic profiles of CsA (Neoral®) and MPA (Cellcept®) in the early post-grafting period in renal transplant patients.
Methods: Pharmacokinetic evaluation was performed in 20 patients who had undergone de novo renal transplantation and received an immunosuppressive therapy including corticosteroids, MPA (Cellcept®) and CsA (Neoral®). In each patient, 11 blood samples were collected on the 3rd day of treatment, immediately before (T0) the morning dose and 0.33, 0.67, 1, 1.5, 2, 3, 4 6 9 and 12 hours after dosing. The pharmacokinetic analyses were performed using nonlinear mixed effect modeling with the NONMEM program version V (2) through the Visual-NM graphical interface (3).
Results: The different profiles of CsA could be adequately described using a model taking into account a discontinuous absorption (4). Briefly, this model assumes that (i) absorption from the stomach is negligible; (ii) a given oral dose is absorbed sequentially in two portions, (iii) at the same absorption site (i.e. a gut compartment). So, after oral administration a first fraction (f) of the dose is transferred from the stomach to the gut while the transfer of the remaining fraction (1-f) is delayed. This is described by first-order processes with lag-times, whereas absorption from the gut compartment is described by a first order rate constant. Six patients presented pharmacokinetic profiles with two peaks, while the 14 others had apparently a single, sometimes delayed and wide peak. A one-compartment model with first order elimination in association with the double absorption model best fitted the data, as shown by a highly significant decrease in the objective function (OF) (110 points less than with one- or two-compartment models with first-order absorption and a lag time). Residual variability consisted in a combined additional and proportional error of respectively 21 µg/L (close to the analytical limit of quantification of 20 µg/L) and 13.1%. The First-Order Conditional Estimation (FOCE) interaction method was used.
For mycophenolate, a two-compartment model with first order elimination in association with the double absorption model better fitted the kinetic data (OF = 398) than an absorption model based either on a first order process with lag time (OF = 450) or on Weibull distribution (OF = 436). However, the residual variability remained fairly high, consisting in a combined additional and proportional error of respectively 0.6 mg/L (i.e. 6 time the limit of quantification) and 18 %. Interestingly, in the only previously reported population pharmacokinetic study of MPA, few structural pharmacokinetic models including non-linear absorption processes were tested, i.e. the Emax model, Weibull function and a dual, sequential first-order absorption process. The authors concluded that these models did not improve the fit as compared to a two-compartment model with first order absorption with a lag-time and first-order elimination (5).
Conclusion: The pharmacokinetic model developed in the present study is suitable for the analysis of CsA population pharmacokinetics in the first days after renal grafting. It reliably described the pharmacokinetics of cyclosporine whether a double peak was present or not on the concentration-time profiles. For mycophenolate, this multi-exponential absorption model provided a better fit than the other models tested, though improvements are still needed. The influence of covariates, enterohepatic circulation and of a potential timedependentelimination process are still to be investigated.
References:
(1) Levy G, Burra P, Cavallari A, Duvoux C, Lake J, Mayer AD, Mies S, Pollard SG, Varo E, Villamil F, Johnston A. Improved clinical outcomes for liver transplant recipients using cyclosporine monitoring based on 2-hr post-dose levels (C2). Transplantation. 2002 ;73(6):953-9.
(2) Beal SL, Sheiner LB. NonMEM Users Guide (I-VIII). Hanover, Maryland: GloboMax, 1989-1998.
(3) Gomeni R. Visual-NM user’s manual. Research Development Population Pharmacokinetics. Montpellier France, 1998.
(4) Yin OQP, Tomlinson B, Show AHL, Show MSS. A modified two-portion absorption model to describe double-peak absorption profiles of ranitidine. Clin Pharmacokinet 2003; 42(2):179-92.
Reference: PAGE 13 () Abstr 497 [www.page-meeting.org/?abstract=497]
Poster: poster