Frances Brightman, David Orrell, Eric Fernandez, Christophe Chassagnole
Physiomics plc
Objectives: Immunotherapy has recently developed into a highly active area of anticancer drug development. The field is dominated by immune-checkpoint blockers, which counteract the suppression of the immune response that is often observed in cancer. While early results for monotherapies are promising, the real potential of immunotherapy agents could be in combining them together or with other anticancer treatments. However, there is currently no rational basis on which to select optimal dosing regimens or combination schedules, and a clear unmet need for predictive tools to aid this process[1,2].
Methods: Physiomics has developed a preclinical and a clinical ‘Virtual Tumour’ (“VT”) technology that can predict how a tumour will respond to drug exposure. The VT technology integrates PK and PD effects and models the way individual cells behave within a tumour population. These agent-based methods are particularly suitable for modelling not only tumour cells, but also other cell populations – such as those involved in the immune response – and interactions between cells. Here we describe our recent development and application of the VT technology for modelling preclinical efficacy of immune-checkpoint blockers, with a focus on agents targeting the PD-1/PD-L1 axis. The VT platform has been extended by the addition of an immunotherapy module, which has been developed and calibrated using data taken from the literature. This module captures the mechanisms by which the immunotherapy activates the antitumor immune response and synergizes with conventional anticancer therapies.
Results:Through a preclinical case study derived from the literature, we demonstrate that the extended VT can be applied to model the efficacy of an anti-PD-L1 antibody in syngeneic mouse xenografts, both alone and in combination with irradiation. The model was calibrated using published PK data for the antibody[3], and tumour growth inhibition data for the monotherapies in two in vivo models (TUBO and MC38)[4]. As in a blind validation study, the calibrated model was then used to predict the respective combination efficacies. The predictions were validated against the published experimental data[4], and found to accurately reflect the synergy of the combination treatment.
Conclusions: Through this case study, we demonstrate that our enhanced VT capability represents the first step towards a ground-breaking tool for optimizing dosing and scheduling of immunotherapy, both alone and in combination with conventional anticancer therapies.
References:
[1] Pardoll, D. M. The blockade of immune checkpoints in cancer immunotherapy. Nat. Rev. Cancer 12, 252–264 (2012). [2] Lesterhuis, W. J., et al. Cancer immunotherapy – revisited. Nat. Rev. Drug Discov. 10, 591–600 (2011). [3] Contreras-Sandoval, A. M. et al. PK-PD modelling of an anti-PD-monoclonal antibody. PAGE meeting, Alicante, Spain (2014). [4] Deng, L. et al. Irradiation and anti–PD-L1 treatment synergistically promote antitumor immunity in mice. J. Clin. Invest. 124, 687-695 (2014).
Reference: PAGE 24 (2015) Abstr 3561 [www.page-meeting.org/?abstract=3561]
Poster: Drug/Disease modeling - Oncology