Ethnic Difference In Mycophenolate Mofetil Pharmacokinetics Between Chinese And Caucasian Healthy Subjects

Mycophenolate mofetil (MMF; Cellcept(?)), a morpholinoethyl ester of the immunosuppressive agent mycophenolic acid (MPA), is now the most prescribed drug for prevention of acute allograft rejection in solid organ transplantation patients. After oral administration, MMF is rapidly and completely hydrolyzed into MPA and then extensively glucuroni dated by uridine diphosphate glucuronosyltransferase enzymes (UGTs) to the primary pharmacologically inactive metabolite,7-Omycophenolic acid glucuronide (MPAG). The glucuronide metabolite is transported to the gut via the bile or excreted into the urine and eliminated. MPAG present in the bile is subject to deglucuronidation where it is converted back to MPA and reabsorbed into the systemic circulation through enterohepatic recirculation (EHC). In China, MMF was approved for the prophylaxis of acute rejection of kidney transplantation in 1997. The package insert for MMF recommended dose regimen is 1.0 g bid, in combination with cyclosporine or tacrolimus and corticosteroid, which is the same as the US label. However, in clinical practice, the dose of MMF is lower than that on the label. The common dose used in Chinese and other Asian kidney transplant patients (Thailand, China.Hongkong, Korea and Singapore) is around 0.5-0.75 g bid. There was no clinical trial to study the ethnic difference in the pharmacokinetics (PK) of MMF between Chinese and Caucasians. The objective of this study was to assess race difference of MPA and MPAG PK among Chinese and Caucasian healthy subjects by population PK analysis, and to evaluate other covariates.Data were pooled from eight clinical studies. Five studies were performed on Chinese and the other three were conducted on Caucasians. Data on dose, sampling schedule, MPA and MPAG plasma concentrations, mealtimes and demographic data from the eight studies were extracted and pooled in one dataset. To describe the physiologic process of EHC, a population PK model for MPA and MPAG that included an EHC component was considered. The analysis was performed by non-linear mixed effect model (NONMEM). Exponential errors following a log-normal distribution were used for the description of the inter-individual variability of the parameters. The first-order conditional estimation with interaction method (FOCE-I) was used to estimate the population PK parameters. The effects of the covariates of race, weight, creatinine clearance and other physiological variables on the PK parameters were evaluated. The stability and predictability of the final model was confirmed by internal goodness-of-fit (GOF), cross-validation,visual predictive check (VPC) and normalized predictive distribution error (NPDE) as well as external evaluation pooled data obtained from the published studies.A total of 3060 plasma concentration measurements (1810 MPA and 1250 MPAG) from eight studies with a total of 80 Han Chinese and 52 Caucasian healthy subjects were used for population PK model development. There were statistically significant differences in weight, age, body surface area (BSA), body mass index (BMI) and creatinine clearance (CLcr) between Chinese and Caucasians by one-way analysis of variance (P<0.01).The PK of MPA and MPAG were best described with a 5-chain compartment model including a gallbladder compartment for enterohepatic recirculation, a gastrointestinal compartment, a MPA central compartment, a MPA peripheral compartment and a MPAG central compartment. The absorption process of MPA was best described with a transit model. The number of transit compartments was 19.2, and the mean transit time (MTT) was 0.190h. Multiple peaks of MPA and MPAG due to EHC process in study cohorts were well described by the final model. Weight on apparent clearance of MPA (CL/FMPA), apparent central distribution volume of MPA (V2/F), apparent inter-compartmental clearance of MPA (Q/F), apparent clearance of MPAG (CL/FMPAG) and apparent central distribution volume of MPAG (V4/F) were identified as covariates. Race showed significant correlations with apparent clearance and volume of distribution of MPAG, but not of MPA. Apparent clearance of MPAG was 1.36 L·h-1 and 1.90 L·h-1 for a 70 kg Chinese and Caucasian, respectively, and apparent clearance of MPA was 11.5 L·h-1 for a 70 kg healthy subject. The population model for the CL/FMPA and CL/FMPAG are listed below.RACE was equal to 1 in Chinese, otherwise RACE was equal to 1.40. This also means that CL/FMPAG was greater in Caucasians than in Chinese healthy subjects. Race has no effect on CL/FMPA.In the VPC, most of the observations are well contained within the 90% prediction interval and were symmetrically distributed around the median, indicating good predictive performance of the final model. The mean prediction error of cross-validation was close to that of the original data. The NPDE indicated the good predictive ability of the model. Based on the literature search,25 reports for Chinese and Caucasian healthy subjects which included 324 MPA and 94 MPAG samples were screened as validation dataset for external model evaluation. The median prediction error (MDPE) from the validation dataset was -8.1% and -0.2% for MPA and MPAG, respectively. The median absolute prediction error (MDAE) from the validation dataset was 15.8% and 7.3% for MPA and MPAG, respectively.This population PK analysis showed that no difference was found in MPA clearance, but the clearance and volume of distribution of MPAG in Caucasian healthy subjects was 1.40-fold and 1.43-fold higher than Chinese, respectively. The effect of weight was also identified as a significant covariate on the clearance, central distribution volume, intercompartmental clearance of MPA and clearance, central distribution volume of MPAG. Both internal and external evaluations indicated that the final model had better fitness and predictive performance.