Search Algorithms for Drug Combinations: Extending Approved Cancer Therapies

Salgomed, Inc., Del Mar CA

Investigators

Linked publications, trials & patents

Abstract

DESCRIPTION (provided by applicant): Search algorithms for drug combinations: Extending approved cancer therapies. Combination drug therapy is commonly used to enhance efficacy and overcome drug resistance in cancer, but at present the choice of drugs is based on empirical clinical experience alone. Testing multi-drug therapies in a systematic way is hampered by the large number of possible choices for drugs and doses. Therefore, a new approach for discovering effective drug combinations is needed. This project is based on the hypothesis that search algorithms can quickly explore the experimental space of combinations to find efficient therapies. Thus far, advanced mathematical algorithms and high-throughput screening have not been integrated into commercially available platforms able to explore multi-drug combinations. Our goal is to develop an integrated computational and experimental platform implementing in vitro search algorithms for drug combinations. We will start with algorithms commonly used in engineering and physics that iteratively search for optimal therapeutic interventions. In the implementation of the algorithms, we will integrate biological information from genomics and gene expression data, and predictions from explicit signaling networks models to improve convergence and performance. We will focus on an approved EGFR inhibitor, and we will validate our approach by including this drug in combinations with up to three other compounds from libraries of approved drugs and of kinase inhibitors. The proposed platform will be able to optimize combinations of drugs for selective killing of cancer cell lines and limited toxicity on normal cells lines. Protein kinases are central to cellular signaling and are highly investigated as targets of therapeutic agents because they contain activating mutations in many cancers. The specific pattern of oncogenic kinase mutations varies among cancers and individual patients, creating opportunities for selective action and personalization of combinations.