How Anti-cancer Targeted Therapies Used in Combination Interact? Analysis with a Semi-mechanistic Model of Minimal Signalling Networks
Arnaud Quelin, Michel Tod, Emilie Henin
EMR 3738 CTO, UCBL-HCL Faculté de Médecine Lyon-Sud, Université Lyon 1, Oullins, France
Objectives: Targeted agents are specific inhibitors of signalling and metabolic pathways used in oncology to counteract tumour proliferation and angiogenesis. They were developed to offer an efficient and safe alternative or complement to cytotoxic chemotherapies [1]. Nevertheless, their efficacy may be lower than expected due to cross-talks between signalling cascades, resulting in the over-activation of one pathway after the blockage of another. There is a rationale to use agents in combination, targeting several pathways at once. This work aims to theoretically characterize interactions of targeted agents used in combination in minimal signalling networks and the impact on the therapeutic response using a semi-mechanistic pharmacodynamic model.
Methods: Three enzymes X, Y, Z (in either inactive or active state) interacting on either serial or parallel pathway with or without feedback loops were considered. Two drugs D1 and D2, given as continuous infusions could affect the activation of X and Y enzymes.
In each case, the system and the action of both drugs on the system were described by 6 ordinary differential equations. Response surfaces were computed to quantitatively evaluate the magnitude of the combined effect. Sensitivity analyses were performed to identify the most influential parameters. Simulations, graphics and analyses were performed using R software.
Results: 32 different pathway structures were considered (128 with feedback loops). including 16 "coherent" cases where both drugs trigger the same overall effect ;and 16 "non-coherent" cases where drugs trigger opposite overall effect. Non trivial response surfaces were explored at steady state enabling theoretical determination of the best drug combinations and ranges of neutral combinations for which no difference with the baseline activity level exist. When respectively negative or positive feedback loops are added to the network, oscillations or increasing steepness of the response were observed, as previously reported in [2].
Conclusions: Even in a minimal setting (only 3 enzymes), the pathway structure may result in non-trivial response. It is therefore determinant to understand the signalling network and the eventual cross-talks in order to optimise how targeted agents should be combined.
References:
[1] T. Bagnyukova, Chemotherapy and signalling: How can targeted therapies supercharge cytotoxic agents ? , Cancer. Biol. Ther. 2010 November 1; 10(9): 839:853.
[2] B. N. Kholodenko, Negative feedback and ultrasensitivity can bring about oscillations in the mitogen-activated protein kinase cascades, Eur. J. Biochem. 2000; 267: 1583:1588.
