Predictive Oncology Inc has developed a scientifically validated artificial intelligence (AI) platform to expedite early drug discovery and support drug development. The company recently reported that the platform can predict with 92% accuracy whether a tumor sample will respond to a certain drug compound. In just 8 weeks, their analysis of publicly available data sets on oncology drugs that have either been abandoned or discontinued by large pharmaceutical companies led to development of a registry of select drug candidates with promising mechanisms of action for further clinical testing. The predictive model measured 92 combinations of laboratory experiments on patient tumor samples to make an additional 964 confident predictions, covering a total of 79% of all possible experiments. This led to identification of repurposed drug candidates with potential efficacy in treating colon and ovarian cancer. The AI-driven technology was able to predict an additional 10 times the number of measured experiments with in silico vs in vitro research, eliminating at least 18 months of lab testing. In ongoing research, the company is also using its AI platform to identify a set of FDA-approved drugs for other cancers that show promising activity in ovarian cancer.
High level
These results from Predictive Oncology demonstrate the increasing role of AI and machine learning platforms in drug discovery and development. Use of AI may enable faster and cheaper development of new therapeutics compared with traditional early drug discovery. This proprietary technology facilitates informed selection of drug/tumor type combinations for subsequent in vitro testing. Additionally, this technology may enable repurposing of approved or abandoned drugs for additional indications, creating new opportunities for researchers and manufacturers.
Ground level
AI and machine learning technologies such as that used by Predictive Oncology may decrease the time to drug discovery and development, helping to make more effective new or repurposed treatments available to patients sooner.
