Tranchand B1, Pobel C2, Ardiet CJ1, Parola JN2, Ménard G2
1Laboratoire de Pharmacocinétique, Centre Léon-Bérard, Lyon, France, 2 Service de Pharmacie Clinique, Centre Hospitalier, Saintes, France
Aim of the study: The marked inter- and intrapatient variability of the pharmacokinetics of a number of antineoplastic agents is well known. Moreover, we can presuppose that there is, at some extent, some variability degree of neutropenia following melphalan therapy, and that this variability is in some fashion related to the administered dose of drug. In the present study, we tried to quantify the pharmacodynamic activity of melphalan using a cubic spline function, and to establish a pharmacokinetic/pharmacodynamic relationship. Indeed, it would be useful to predict the myelosuppression amplitude from a pharmacokinetic parameter, in order to individualise the dosage regimen, giving the maximum tolerated dose without aplasia.
Data: Data came from a Phase I study based upon melphalan AUC escalation, in place of usual dose units. Interpatient AUC escalation was carried out between cohorts of 3 patients. Intrapatient dose modifications were carried out according to the toxicity observed in the previous course. The first level was 30 AUC units and the escalation steps were 15 AUC units.
Patients and treatment: 19 patients with ovarian carcinoma entered the study (60 courses, 1 to 6 courses per patient). Melphalan was given as single therapy. Each course consisted of two 5 min regular injections at Day 1 (test-dose), and Day 2 (to reach the target Area Under the Curve : AUC). Total AUC (AUCday1 + AUCday2) were escalated in intra- and interpatient manner up to the maximum tolerated AUC value. Blood counts were repeated on days 0, 1, 7, 14, 20 and 27 after each course. Covariates included were body weight (range [42; 80]), body surface area (BSA) (range [1.31; 1.80]), creatinemia (Cr) (range : [45; 173]), course number, PN baseline count (BASE) (range [1600;7200]). Melphalan AUC values ranged from 15 to 118 mg.l-1.min.
Model: The population model consisted of
– a cubic spline function (« structural model »),
– a covariate model relating parameters of the structural model to variables of predictive values: neutrophil baseline count and melphalan Area Under Curve (AUC).
– a variance model.
Data analysis was performed with NONMEM (using First Order method) under Visual-NM. Spline calculations (bspline subroutines from D. Verotta) in NONMEM were obtained (after preliminary investigations) using 3 internal breakpoints. The basic structural model was adapted from a Control Stream from M. Karlsson et al, who studied the leukopenic effect of etoposide (Clin Pharmacol Ther, 1995, 57, 325-334).
Results: For the population used to build the model, the estimated typical response profile is characterised by: -half maximal effect after therapy AUC50 of 55.6 mg.ml-1.min.
-Emax(t) at 25% of total duration 0.634
at 50% of total duration 0.941
at 75% of total duration 0.845
at 100% of total duration 0.209
Variability in AUC50 is 40% and analysis of conditional estimates versus covariates suggest a potential relationship between neutrophil baseline count and AUC50.
So, such a population model, after fully validating with new patients, could be used to describe response versus time and predict risk of toxicity after melphalan therapy.
Reference: PAGE 6 (1997) Abstr 669 [www.page-meeting.org/?abstract=669]
Poster: poster