JGC van Hasselt(1,2), K van Calsteren(3), ADR Huitema(2), L Heyns(3), S Han(3), M Mhallem(3), JHM Schellens(1,4), JH Beijnen(2,4), F Amant(3)
(1) Department of Clinical Pharmacology, Netherlands Cancer Institute, Netherlands; (2) Department of Pharmacy & Pharmacology, Slotervaart Hospital, Netherlands; (3) Obstetrics and Gynecology, University Hospitals Leuven, Leuven; (4) Faculty of Science, Department of Pharmaceutical Sciences, Utrecht University, Netherlands.
Introduction: Physiological changes during pregnancy may affect pharmacokinetics (PK) [1]. Studies investigating changes in PK may therefore be needed to assess if dose adjustments are necessary. The evaluation of changes in PK of anti-cancer (AC) agents during pregnancy is of specific relevance, as the narrow therapeutic window of such agents may have severe consequences, either due to increased toxicity or decreased efficacy. However, the conduct of PK studies during pregnancy is constrained by practical and ethical limitations such as low numbers of available subjects, therefore requiring informative trials that are analyzed in an efficient fashion. The aims of this project were to: i) estimate changes in PK for four AC agents and the impact on current dosing guidelines; ii) evaluate if a semi-physiological approach could be used to predict expected changes in PK.
Methods: A nonlinear mixed effect modelling approach was used to estimate gestational effects. Subsequently, we simulated the impact of estimated changes on dosing regimens. Limited D-optimal sampling designs were derived for all treatments that could be expected. Designs were optimized taking into account that each treatment could consist of multiple drugs of interest, and, expected changes in PK were leveraged in a semi-physiological fashion including renal function [2] and body composition [3].
Results: For doxorubicin, we estimated increases of 23% (RSE 24%) and 32% (RSE 15%) for the initial (V1) and terminal (V3) volumes of distribution respectively, with no identifiable change in clearance (CL). For epirubicin we estimated 10% (RSE 9%) increase in CL, 55% (RSE 13%) increase in V1 and 208% (RSE 25%) increase in V2. For docetaxel we estimated 19% increase (RSE 7%) in CL, while V1, V2 and V3 changed by 7% (RSE 12%), 37% (RSE 18%) and -9.7% (RSE 15%) respectively. For paclitaxel we estimated 30% (RSE 41%) increase in CL, , while V1, V2 and V3 changed by -15.4% (RSE 70%), 48% (RSE 34%) and 28% (RSE 29%) respectively.
Limited optimal sampling designs were successfully developed for all expected AC treatments, with a minimum of 4-5 samples required to obtain adequate PK parameter estimates.
Conclusion: Gestational effects of pregnancy on PK parameters of doxorubicin, epirubicin, docetaxel and paclitaxel were quantified, and associated dose adjustments were computed. Moreover we demonstrated how physiological data can be leveraged into a PK model in order to obtain informative clinical study designs.
References:
[1] Anderson. Clin Pharmacokin. 2005;44(10):989.
[2] Davison et al Br J Obst Gyn. 1975;82(5):374-81.
[3] Hytten et al J Obst Gyn. 1966;73(4):553-61.
Reference: PAGE 22 () Abstr 2715 [www.page-meeting.org/?abstract=2715]
Poster: Oncology