Abstract 5215: Computer based “virtual tumor” simulation identifies novel use for existing drugs - implication for personalized cancer therapy

• Published in: Cancer research (Chicago, Ill.) Vol. 73; no. 8_Supplement; p. 5215
• Main Authors: Vali, Shireen, Kapoor, Shweta, Talawdekar, Anay, Kumar, Ansu, Sultana, Zeba, Gishizky, Mikhail L., Fernandes, Pradeep, Abbasi, Taher
• Format: Journal Article
• Language: English
• Published: 15.04.2013
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2013-5215

Abstract Background: Genomic tumor profiling studies have demonstrated that most cancer cells embody a multitude of genomic aberrations that may contribute to the tumor phenotype. However, development and use of targeted molecular therapies tend to focus on the effects of single genomic alterations. This is due in part to our inability to integrate and understand the contribution of multiple mutations giving rise to the tumor phenotype. The ability to develop new therapies and select treatments for individual patients would be greatly aided by the availability of a predictive technology capable of integrating and determining the effects of multiple genomic aberrations on the tumor phenotype. This patient specific customization can allow for screening for sensitivity and resistance toward specific interventions. Methods: Cellworks has developed a computer pharmacodynamic simulation model of epithelial tumors (virtual tumor) that capture the signaling and metabolic pathways for cancer relevant phenotypes and bio-marker endpoints. This model integrates clinical and preclinical research insights and contains over 25,000 crosstalk interactions between 6000 bio-molecules. Using this system, individual tumor profiles from clinical samples and derived cell lines are overlaid on a control epithelial baseline to create a “personalized” virtual tumor. In this study a “digital drug library” comprised of 75 molecularly targeted drugs from different indications were evaluated in combinations and at different doses (more than 10,000 simulation studies) for their ability to inhibit growth of mutant KRAS tumors. Therapeutic “hits” were shortlisted based on impact on the apoptotic and survival pathways. The top “hits” were evaluated for their ability to inhibit growth of tumor cells in vitro. Results: Using the computer simulation based screen, we identified two drugs, that are approved for non-oncology indications, that when used in combination, can inhibit the growth of KRAS driven tumors: CW229 (anti-inflammatory drug) and CW178 (metabolism target drug). The predictions were confirmed using KRAS mutant cells lines, H1299 and HCT116, in vitro. Both CW229 and CW178 show low to moderate impact on cell viability as single agents in vitro. However, the combination of CW229 + CW178 shows a greater than additive effect, inhibiting cell growth by over 80%. Data from additional preclinical studies will be presented. Conclusions: Using our “virtual tumor” technology and methodology we identified a novel use for two non oncology drugs, CW229 and CW178, as a potential combination therapy for the treatment of KRAS mutant tumors. In addition this study demonstrates how integration of patient genomic profiles into a “personalized virtual tumor” can guide development of new therapies and aid clinicians in selecting the most appropriate therapeutic intervention for individual patients. Citation Format: Shireen Vali, Shweta Kapoor, Anay Talawdekar, Ansu Kumar, Zeba Sultana, Mikhail L. Gishizky, Pradeep Fernandes, Taher Abbasi. Computer based “virtual tumor” simulation identifies novel use for existing drugs - implication for personalized cancer therapy. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5215. doi:10.1158/1538-7445.AM2013-5215