Diane-Charlotte Imbs(1), Raouf El Cheikh(1), Joseph Ciccolini(1), Severine Mollard(1), Sarah Giacometti(1), Dominique Barbolosi(1), Sebastien Benzekry(2)
(1) SMARTc Unit, Pharmacokinetics Laboratory, Inserm S_911 CRO2 Aix Marseille University and La Timone University Hospital of Marseille, (2) Modelling Control, and Computations Unit, Institut National de Recherche en Informatique et en Automatique, Bordeaux, France
Objectives: To develop a biologically-based mathematical model that is capable of describing the impact of adding bevacizumab (B) to chemotherapy (C) for non-small cell lung cancer treatment (NSCLC). The aim is to assess, through in silico simulations, the benefits of the additional administration of bevacizumab and to look for the best treatment modality.
Methods: In a nonclinical proof-of-concept (POC) study, immunocompromised NSCLC-bearing mice (H460-Luc+ xenograft, n=12 per group) were treated according to three different sequences associating bevacizumab and pemetrexed – cisplatin: B and C concomitantly and two sequential schedules with a lag time of 4 days between the two treatments (B before C and C before B) administered every 14 days. Tumor growth follow-up was done for each group, by bioluminescence imaging and compared with a control group (without treatment).
A pharmacokinetic/pharmacodynamic (PK/PD) model was built to describe the tumor growth observed experimentally for each of the four groups. The growth was assumed to follow a Gompertzian law and the model includes the effects of B on the quality of tumor vasculature (1). Parameters were estimated by fitting the model to the tumor growth experimental data according to a population approach (non-linear mixed effects models) using Monolix 4.3.3 Lixoft. Different time lags for the sequential administration were tested in silico to find the optimal time-lag for the treatment.
Results: The POC study in NSCLC confirmed the superiority of the alternative schedule (i.e., sequential administration of B given before C) in terms of survival and tumor growth compared to other sequences, with a lower tumor growth (-54.8%) and an increase in survival of 24.9% after 50 days of treatment.
The PK/PD model was able to reproduce the tumor growth data, and identify significantly higher tumor reduction in the sequential B/C group (p-value = 0.035).
Conclusion: Preclinical POC has shown evidence for superiority of sequential administration of B and C over the standard concomitant schedule. The biological-based mathematical model can be used as an in silico tool in order to optimize and individualize treatment regimen for the combination of bevacizumab and chemotherapy in NSCLC.
References:
[1] Benzekry S, Chapuisat G, Ciccolini J, Erlinger A, Hubert F. A new mathematical model for optimizing the combination between antiangiogenic and cytotoxic drugs in oncology. Comptes Rendus Math. janv 2012;350(1–2):23-8.
Reference: PAGE 25 (2016) Abstr 5940 [www.page-meeting.org/?abstract=5940]
Poster: Drug/Disease modeling - Oncology