Hitesh Mistry (1), David Orrell (1), Eric Fernandez (1), Frances Brightman (1), Deborah Guest (1), Linda Collins (2), Avinash Gupta (2), Mark Middleton (2), Christophe Chassagnole (1)
(1) Physiomics PLC, Magdalen Centre, Oxford Science Park, Oxford, OX4 4GA, UK. (2) Department of Oncology, University of Oxford, Churchill Hospital, Oxford, OX3 7LE, UK.
Objectives: The translation of results from animal to man is a key phase in oncology drug development. Being able to determine at which doses key biopsy measurements should be taken, or we should expect to start seeing efficacy, is important for successful evaluation of a new drug within early clinical development. Furthermore, being able to accurately translate combination schedules from mouse to man would provide significant cost savings and advance clinical development. Here we present a case study highlighting the potential applicability of Virtual Tumour Clinical in translational science.
Methods: We have developed a mathematical model of a tumour cell population called the Virtual Tumour, which has been used extensively to predict the efficacy of single drug or drug combination treatment in preclinical studies1–4. We have now extended and adapted our preclinical model to predict efficacy in the clinic, thereby creating the ‘Virtual Tumour Clinical’.
Results: Here we show two sets of results highlighting the translational predictivity of Virtual Tumour Clinical. The first example highlights the back-translational capabilities of the model with vemurafenib; we train the model to clinical data and assess whether we can predict the outcome of xenograft studies. The second example looks at using the model for forward translation. Here we train the model to preclinical monotherapy data only for docetaxel and selumetinib, and assess whether we can predict the efficacy of both arms of a recent phase 2 trial assessing the combination versus docetaxel monotherapy5.
Conclusions: Virtual Tumour Clinical was able to accurately back translate the effects of vemurafenib: predict xenograft changes in tumour volume using clinical data to calibrate the model. Virtual Tumour Clinical also made accurate predictions on the clinical efficacy in both arms of a phase 2 trial of docetaxel and selumetinib.
References:
[1] Fernandez, E. et al. Modeling the sequence-sensitive gemcitabine-docetaxel combination using the Virtual Tumor. in AACR 102nd Annu. Meet. Orlando FL (2011).
[2] Fernandez, E., Orrell, D., Brightman, F., Fell, D. & Chassagnole, C. Modeling ionizing radiation exposure in vitro and in vivo using the Virtual Tumor. (2013).
[3] Orrell, D. & Fernandez, E. Using predictive mathematical models to optimise the scheduling of anti-cancer drugs. in Innov. Pharm. Technol. 33, 58–62 (2010).
[4] Orrell, D. et al. Predicting the effect of combination schedules on xenograft tumor using the Virtual Tumor. in Proc. Am. Assoc. Cancer Res. (2011).
[5] Gupta, A. et al. DOC-MEK: A double blind randomized phase 2 trial of docetaxel with or without selumetinib in wild-type BRAF advanced melanoma. Ann. Oncol. (2014).
Reference: PAGE 23 () Abstr 3112 [www.page-meeting.org/?abstract=3112]
Poster: Drug/Disease modeling - Oncology