1,3Llopis MC; 1Pérez-Ruixo JJ; 3Casabó VG; 4Boddy A ; 2Almenar D; 1,3Jiménez NV.
1Servicio de Farmacia y 2Servicio de Oncologia. Hospital Universitario Dr Peset. Valencia. 3Departamento de Farmacia y TecnologÃa Farmacéutica. Universidad de Valencia. 4Cancer Research Unit. Medical School. University of Newcastle
BACKGROUND. Cyclophosphamide (CP) is most commonly used in high dose chemotherapy (HDCT) prior to peripheral blood stem cell transplantation (PBSCT) in high risk breast cancer. CP is known to induce it own metabolism following repeated or continuous administration.
OBJECTIVE. To characterize the population pharmacokinetic of CP in high risk breast cancer patients.
PATIENTS AND METHODS. 53 females with high risk breast cancer diagnosis, age 46 (9.2) years, weighted 69.2 (11.8)kg, and normal renal and hepatic funcion were included in the study. Patients received a STAMP-V protocol of HDCT: cyclophosphamide 6 g/m2, thiotepa 500 mg/m2 and carboplatin 800 mg/m2 administered in intravenous continuos infusion of 96 hours and PBSCT was performed three days later. Blood samples were drawn at the 6, 24, 48, 72, 96, 97 and 99 hours since begin continuos infusion, plasma was separated and stored at -20ºC until assayed for CP. CP plasma concentration-time profiles were described by a one-compartment model with zero-order absortion and first-order inducible elimination. CP autoinduction was modeled with an enzyme turn-over model, under the assumption that Cp increases the production rate of enzyme in a linear fashion, where Scp is the slope of linear function for induction by CP. Kenz is the rate constant for first-order degradation (time-1) of the pool. To normalize the enzyme concentration to unity at baseline, the zero-order production rate of enzyme was set to Kenz (amount·time-1) , . Data analysis was performed using the nonlinear mixed effects modelling program (NONMEM). Population and individual estimes for Cl, V, SCp, Kenz, were obtained using a general nonlinear population pharmacokinetic with first order processes and differential equations (AVDAN 6 TOL=5). Interindividual variability was asumed to be proportional error model and residual variability was modelled using additive error model.
RESULTS. In the model, We identify two subpopulations, a fraction 47 % of the patient has kenz1, and that the remaining fraction 53% has kenz2 (using subroutine $MIX).
Reference: PAGE 10 (2001) Abstr 219 [www.page-meeting.org/?abstract=219]
Poster: poster