Finding breakthrough therapies in cancer research

Should a pharmaceutical company enter the final stage (phase III) of the drug development process? In current oncology research, late-stage clinical trials fail about half the time, letting the colossal investment made up to that point go to waste. Therefore, management consulting firm Catenion teamed up with idalab to develop a predictive risk model for the probability of success of phase III studies based on a variety of public and proprietary data sources. The predictive risk model outperforms academic approaches and is a powerful tool for pharmaceutical companies to allocate spending on cancer research more efficiently. At the same time, it enables doctors to make better and more ethical decisions as to whether enrol patients in clinical trials.

 

Situation

  • Success rate for phase III trials in cancer research is 45%, with costs ~$20M-200M for phase III alone
  • Huge unmet therapeutic need, many cancers are still poorly understood
  • R&D pipeline value is the main asset in pharma

Approach

  • Development of a predictive mathematical model for phase III success, based on an integration of public and expert input
  • Definition of meaningful predictive metrics:
    • Accumulated knowledge and strength of results from previous phases
    • Historical success rates within the indication and track record of clinical investigators

Impact

  • High predictive power (~70%), outperforming state of the art, black-box academic models

Back to cases overview