Integrating network reconstruction with mechanistic modeling to predict cancer therapies

Signal transduction networks (STNs) are often rewired in cancerous cells. Effective cancer treatment requires identifying and repairing these harmful alterations. We developed a computational framework which can identify these aberrations and predict potential targets for intervention. It reconstructs network models of STNs from noisy and incomplete perturbation response data, and then uses the reconstructed networks to develop mechanistic models of STNs for predicting potential treatments. As a proof of principle, we analysed a perturbation dataset targeting Epidermal Growth Factor Receptor (EGFR) and Insulin like 2 Growth Factor 1 Receptor (IGF1R) pathways in a panel of colorectal cancer (CRC) cells, revealing cell line specific STN rewiring. Specifically, we found that the feedback inhibition of IRS1 by p70S6K is associated with resistance to EGF receptor (EGFR) inhibition, and disrupting this feedback may restore sensitivity to EGFR inhibitors in CRC cells. These findings were experimentally validated in vitro and in zebrafish (Danio rerio) xenografts.