Increasing the probability of technical success in drug development using AI and patient heterogeneity

Oncology drug development makes up almost half of the annual R&D expenditure of biopharmaceutical companies. Only 10% of drugs advanced to the clinical stage are successfully brought to market, with half of all failures attributed to lack of clinical efficacy. This means that approximately $60 billion dollars are spent annually on unsuccessful drugs.1,2

Contributing to these unchecked costs - and the complexity of drug discovery and development - is the fact that the typical timeline for moving from exhaustive basic research to regulatory approval is anywhere from ten to fifteen years. Most drug failures occur in Phase II and Phase III clinical trials when drugs move from immortalized cell lines and animal testing to human patients. One of the primary reasons for these late-stage failures is that the heterogeneity of human subjects is not introduced until clinical trials are already underway.