[Validation of a new interactive software monitoring a controlled-flow infusion pump for cisplatin dosage regimen adjustment]

Abstract

Adaptive dosing of cisplatin (CDDP) results in reduced haematological and renal toxicity but it has never been clearly shown that it affects the tumoral response rate. Before undertaking a clinical randomized study of CDDP monitoring versus standard dose, a comparative study was performed between a new software dedicated to the interactive adjustments--the AJI software--and the APIS software for clinical pharmacokinetics which incorporates a bayesian procedure and a population information computed according to a three compartment model. CDDP was administered by continuous infusion at variable rates with a controlled flow pump during four days in order to reach the target of 1.3 mg/L at the end of the first day, and to maintain this level during the whole treatment. This study was carried out on two groups of patients. Group 1 (12 patients; 27 courses) received CDDP with sequential flow rates in order to obtain a population information to be used with AJI. For patients in group 2 (14 patients; 26 courses), doses (flows) were adapted in a prospective study using the AJI software two to three times the first day, then daily. They could have been adapted (retrospective study) after platin pharmacokinetic parameters identification by APIS. The dosage recommendations proposed by APIS the first day, from 12 hours to 24 hours, to reach the target of 1.3 mg/L at 24 hours, and then daily up to D4 (AAPT at Di) to maintain this level were compared to those which were really administered during the interactive treatment (DA at Di). There were no statistically significant difference for D2, D4 and for the total dose (118.7 +/- 20.1 mg and 118.5 +/- 45.1 mg). The difference was statistically significant for D1 and D3 (P < 0.05). The inter-individual variability was less important with AJI (CV = 16.9% for DA total) than with APIS (CV = 38.0% for ADAPT total). The platin pharmacokinetic parameters identified the first day by APIS were not statistically different from those identified from the whole treatment for clearance (5.92 and 5.63 l/d) and Vtotal 87.8 and 93.1 L); the difference was statistically significant for Vinitial (34.7 and 42.5 L; P < 0.05) and the terminal half-life (13.1 and 15.5 days; P < 0.05).