Laurent Claret (1), Marylore Chenel (6), Chadi Saba (6), Marie-Jeanne Pierrat (6), Valérie Agrapart (6), Renata Robert (6), Jean-Charles Soria (2), Filippo DeBraud (3), Ratislav Bahleda (2), Barbara Adamo (5), Roberta Cereda (4), and Josep Tabernero (5), René Bruno (1)
(1) Pharsight a Certara Company, Marseille, France ; (2) Institut Gustave Roussy, Villejuif, France ; (3) European Institute of Oncology, Milano, Italy ; (4) EOS S.p.A, Milano, Italy ; (5) Vall d'Hebron University Hospital, Barcelona, Spain; (6) Institut de Recherche Internationale Servier, France
Objectives: Lucitanib is a multikinase inhibitor targeting FGFR1-2, VEGFR1-3 and PDGFRA/B showing a strong antitumor and antiangiogeneic activity in subcutaneous xenograft models using several human tumor cell lines. Lucitanib is undergoing a Phase I/IIa trial to determine the maximum tolerated dose and optimal dosing schedule in patients with advanced solid tumors. The drug is given 15 mg once a day (OD) with a reduction of 5 mg when dose limiting toxicities (DLT) occur. The objective of this project is to develop exposure response models for time course of graded (NCI-CTC grade) DLTs (proteinuria, asthenia) and tumor response (RECIST) as function of dosing history to simulate alternative dosing strategies that could help improving safety and efficacy.
Methods: The endpoints were analyzed independently with proportional odds ratio models. The models describe the transition probabilities of proteinuria asthenia grades and response categories as a function of dose and time. Patient dosing histories are accounted for in a virtual compartment with a first-order elimination rate driving transition probabilities. The probabilities are dependent on the preceding stage through a first-order Markov element (1-3). Multiple replications (200) of 200 patients receiving alternative dosing strategies were simulated with either continuous or intermittent dosing with various starting doses. The impact of dose reductions of 2.5 mg instead of 5 mg is also assessed. Mean delivered dose intensities across patients and days and proportion of patients with at least one grade 2 or 3 proteinuria or asthenia and responders are computed at replication level.
Results: Asthenia occurs twice faster than proteinuria (half-life of 3 vs. 6 days) and the dynamics of tumor response is slower (half-life ~9 days). Tumor response in breast cancer patients seems to be superior compared to other tumor types (p=0.004). A dose reduction of 2.5 mg would achieve slightly higher dose intensities and similar event rates than 5 mg. Scenarios with higher dose intensity are predicting better response. Drug holiday scenarios (15 mg 3 week on, 1 week off) would decrease probability of response with no clear impact on safety. Continuous administrations of 12.5 mg or 15 mg for 2 cycles followed by 10 mg would provide good performances.
Conclusions: The models compared strategies untested in the ongoing phase I/II. The results will guide the choice of the dose and administration schedule in future studies.
References:
[1] Zingmark PH, Kågedal M, Karlsson MO. Modelling a spontaneously reported side effect by use of a Markov mixed-effects model. J Pharmacokinet Pharmacodyn. 2005;32(2):261-81.
[2] Girard P, Blaschke TF, Kastrissios H, Sheiner LB. A Markov mixed effect regression model for drug compliance. Stat Med. 1998;17(20):2313-33.
[3] Hénin E, You B, VanCutsem E, Hoff PM, Cassidy J, Twelves C, Zuideveld KP, Sirzen F, Dartois C, Freyer G, Tod M, Girard P. A dynamic model of hand-and-foot syndrome in patients receiving capecitabine. Clin Pharmacol Ther. 2009;85(4):418-25.
Reference: PAGE 23 (2014) Abstr 3279 [www.page-meeting.org/?abstract=3279]
Poster: Drug/Disease modeling - Oncology