Population pharmacokinetics of Ganciclovir following Valganciclovir in solid organ transplant recipients infected by cytomegalovirus
Caldes A2 , Colom H1 , Garrido MJ3, Trocˇniz I3, Gilvernet S2, Armendariz Y2, Cendrˇs JM1, Peraire C1 , Grinyˇ JM2
1Pharmacy and Pharmaceutical Technology Department. School of Pharmacy, University of Barcelona. Spain. 2Nefrology Service, Hospital Universitario de Bellvitge, Barcelona. Spain. 3Pharmacy and Pharmaceutical Technology Department. School of Pharmacy, University of Pamplona, Spain.
Objectives: Cytomegalovirus (CMV) is a leading cause of disease in immunocompromised subjects, such as solid organ transplant recipients. Valganciclovir (VGC), an ester prodrug of ganciclovir, was developed to offer an alternative to long-term intravenous (iv) and low oral (po) bioavailability of ganciclovir (GCV). The aim of this study was to establish the population pharmacokinetics of GCV after iv GCV followed by po VGC as treatment of CMV infection in solid organ transplant (SOT) recipients, and explore the influence of patient covariates on drug disposition.
Methods: : 20 SOT patients (kidney (n=11), liver (n=4) and heart n=5) were recruited for this study. Demographic and biochemical data were recorded. 5 mg/kg/12 h of GCV for 5 days as a 1-hour iv infusion, followed by po VGC doses (900 mg/12 h), for 15 days, were administered. In both cases doses were adjusted by estimated creatinine clearance (CRCL). Blood samples were collected at steady-state over 12 h post-dose. A population pharmacokinetic (PK) analysis was performed using NONMEM VI. The final population model was validated through bootstrapping (n=200) by means of PsN-Toolkit (1).
Results: The PK of GCV after VGC, was best described by a two compartment open model with 1st order absorption. Interindividual variability (IIV) was included in total plasma clearance CL (41%), central distribution volume V1 (46%), absorption rate constant KA (69%) and bioavailability F (25%). Residual error was a combined error model (additive: 0.46 mg/L; proportional: 14.4%) with IIV (34%). The FOCE estimation method was used with interaction. CRCL and body weight (WGT) normalised by mean values described part of IIV in CL. The final population PK parameters were: CL=7.58*(CRCL/56.3)*(WGT/66.8); V1=31.8 L; distribution volume of the peripheral compartment V2=32.3 L; intercompartmental clearance CLD1=10.2 L/h; F=0.83; KA=0.90 h-1; and absorption lag time, LT=0.38 h. Mean values from the bootstrap analysis were close to the parameter estimate from the original data set.
Conclusions: A population PK model for GCV, after GCV iv/VGC po, has been developed. It incorporates measure of renal function and body weigt to predict total drug clearance. Validation of this model with external patients should be performed in order to assess the suitability of further VGC therapeutic drug monitoring.
 Lindbom L et al. Comput Methods Programs Biomed 1999;58:51-64.