2014 - Alicante - Spain

PAGE 2014: Drug/Disease modeling - Oncology
Jeroen Elassaiss-Schaap

Translational Pharmacokinetic/Pharmacodynamic Model of Tumor Growth Inhibition by the New Anti-PD1 Monoclonal Antibody MK-3475

Lindauer, A (1), Valiathan, C (2), Mehta, K (2), Philips, J (3), DeAlwis, D (1), de Greef, R (1), Elassaiss-Schaap, J (1)

Merck & Co. Inc. Departments of: (1) Quantitative Pharmacology & Pharmacometrics - Oss, NL and Rahway, US; (2) Scientific Informatics and EDDS IT – Boston, US and Westpoint, US; (3) Translational Immunology – Palo Alto, US

Objectives: To support dose selection of MK-3475 in the melanoma and lung cancer indications based on projection of the likely dose range associated with clinical efficacy.

Methods: MK-3475 is a monoclonal antibody targeted against the programmed-death receptor 1 and is currently under clinical development for a variety of cancers[1]. PK, receptor occupancy (RO) and tumor size data from syngeneic mice treated with an analogue of MK-3475 were used to develop an integrated PKPD model. For this, a compartmental model describing mouse PK in plasma was combined with a physiology-based tumor compartment derived from the literature [2]. The predicted concentration in tumor was linked to receptor occupancy which acted as the driver of tumor growth inhibition (TGI). The final model was translated to human by replacing the mouse PK model with a clinical PK model and system-specific parameters describing distribution and binding were substituted with the relevant human values from in vitro experiments and literature. Several different approaches of scaling the rates of tumor growth shrinkage from mouse to man were explored as scenarios in the dose-response simulations. A sensitivity analysis provided further insight into the effect of changes to parameter values on the primary response of tumor volume.

Results: Plasma PK in mice was best described by a two-compartment model with a parallel non-linear and linear clearance pathway from the central compartment. Inclusion of a feedback mechanism representing the up-regulation of PD-1-receptors in the tumor significantly improved the fit and is consistent with the mode of action of the drugThe final model adequately described the trends in the preclinical data, which was confirmed by test data not used for model development. Sensitivity analyses showed that parameters related to tumor growth, RO and receptor up-regulation were important determinants of the predicted tumor response particularly at lower doses, whereas model predictions for higher doses were less sensitive to parameter changes.

Conclusions: The translational model captured the dynamics of tumor inhibition by MK-3475, and provided important information about the biological and mathematical uncertainty in the model and the expected dose-response properties of MK-3475 in the clinic. The model was successfully applied to support the team during early development, when clinical information was still very sparse, to select a dose range for further evaluation.



References:
[1] Hamid et al., Safety and Tumor Responses with Lambrolizumab (Anti–PD-1) in Melanoma, N Engl J Med (2013) 369:134-144 [2] Shah & Betts, Towards a platform PBPK model to characterize the plasma and tissue disposition of monoclonal antibodies in preclinical species and human, JPKPD (2012) 39:67-86


Reference: PAGE 23 (2014) Abstr 3214 [www.page-meeting.org/?abstract=3214]
Poster: Drug/Disease modeling - Oncology
Top