Due to the significance of combination therapies in cancer, identifying effective multi-drug treatments is often a key aspect of drug development. However, combinations exponentially increase the number of regimens to examine, adding complexity to an already protracted and largely unsuccessful process. Another layer of challenge is added with functional screening. Although patient derived biological models more closely represent the human environment, limited biological material confines the number of treatments for ex vivo testing. To address this, we developed Optim.AI, a small data driven analytics platform which utilizes efficiently designed experiments. The resulting phenotypic response datasets then expand to hundred of thousands of treatment data points, which predictively rank all single and combination drug treatments within a set drug panel. The application and feasibility of Optim.AI been tested across several cancer indications and drug classes, highlighting its potential to streamline drug development and aid guided cancer patient treatment.
Discover Optim.AI, the small data AI platform for combinatorial functional precision medicine. Read the poster to explore its methods, results, and impact.