Population pharmacokinetic modelling of unbound and total plasma concentrations of oxaliplatin administered by hepatic arterial infusion to patients with liver-metastases.
Garrido M.J.1, Viudez A.2, Rodríguez J.2, Zárate R.2, Navarro I.3, Romero E.1, García-Foncillas J.2, Trocóniz I.F.1
1.-Department of Pharmacy and Pharmaceutical Technology. School of Pharmacy; 2.-Department of Oncology. University Clinic of Navarra. 3.- Department of Chemistry, School of Sciences. University of Navarra. Pamplona. Spain
Background: Oxaliplatin, a third generation platinum (Pt) complex, is active in metastatic colorectal cancer. The pharmacokinetic profile of ultrafilterable oxaliplatin has been investigated by several authors, and was described by a three or two compartment open models, depending on the sampling time and the analytical method to Pt quantification [1]. Although in many of these studies Pt levels have been measured as total and free plasma levels, the pharmacokinetic analyses of total and free levels have been performed separately [2]. Therefore, the aim of this study was to develop a population pharmacokinetic model describing simultaneously unbound and total plasma concentrations of oxaliplatin after intra-hepatic administration in patients with metastatic cancer.
Methods: Seventeen patients with liver metastases were treated with 100 mg/m2 of oxaliplatin administered by the hepatic artery in 3-h infusion, followed by 500 mg/m2 cetuximab intravenously infused for 2-h, and oral capecitabine administered every 12h for a week with dose escalation schedule depending on prior-dose-level toxicity. Blood samples were collected according to a sparse sampling with 3-4 samples per cycle and during 1 o 2 cycles.
Population pharmacokinetic analysis using log-transformation data, was performed with the first-order conditional estimation (FOCE) method with interaction in NONMEM program (version VI) [3]. Graphical diagnostics, representations and generation of simulation were done with the programs Xpose 4 and S-PLUS 6.2. [4].
Results: Total and free oxaliplatin plasma concentrations could be described by a two compartment model incorporating a non-linear plasma protein bindingh. Estimates for BMAX, maximal binding capacity, and KD, the concentration at half-maximal binding, were 4.4 (0.03) and 0.15 (0.18) mg/L, respectively. The values for other population parameters such as CL, total plasma clearance and V1, apparent central volume distribution, were according to those previously reported in the literature [1,5]. Interindividual variability could be estimated on V1 and KD and was 21 and 51 %, respectively. Although body mass index, age, hematocrit and capecitabine dose were selected from the GAM approach as significant covariates, none of these were found to have a significant effect on the PK characteristics once they were incorporated in the NONMEM model (P> 0.01).
Conclusion: This study shows that time profiles of total and unbound oxaliplatin plasma concentrations could be simultaneously described with a two compartment model associated with a non-linear binding to plasma components.
References:
[1] G.Bastian, A. Barrail, S. Urien. Anti-Cancer Drugs (2003), 14: 817-824.
[2] M.A. Graham, G.F. Lockwood, D. Greenslade, S. Brienza et al. Clin Cancer Res (2000), 6: 1205-18.
[3] S.L. Beal and L.B. Sheiner. NONMEM User's Guides. San Francisco. University of California, 1998.
[4] E.N. Jonsson and M.O. Karlsson. Comput Methods Programs Biomed (1999) 58: 51-64.
[5] J.P. Delord, A. Umlil, R. Guimbaud et al. Cancer Chemother Pharmacol (2003) 51: 127-131