It now costs more than $800 million to develop a new drug. But what if pharmaceutical companies had a way to predict which experimental drugs will ultimately get FDA approval, giving them the confidence to invest money in them, and which drugs will ultimately fail, allowing them to cut their losses early?
In the February issue of Nature Reviews Drug Discovery, researchers from the Children's Hospital Boston Informatics Program (CHIP) present a forecasting model that may increase the efficiency of drug R&D and save hundreds of millions of dollars per new drug. They also argue that more data sharing by the drug industry particularly of "negative" data would greatly improve the accuracy of forecasting and benefit industry and patients alike, allowing more medical discoveries to be brought to the bedside.
Asher Schachter, MD, MMSc, MS, and Marco Ramoni, PhD, both of CHIP, constructed a Bayesian network model to calculate the probability that a given new drug would pass successfully through Phase III trials and receive New Drug Application (NDA) approval. Their approach differs from convention in modeling populations of drugs rather than populations of patients. They used publicly available safety and efficacy data for about 500 successful and failed new drugs, broken down by therapeutic category, then confirmed the validity of their model by testing it with a group of cancer drugs whose fates are already known.
To gauge the model's potential economic impact, Schachter and Ramoni then performed a pharmaco-economic analysis in collaboration with Stan Finkelstein, MD, Senior Research Scientist at the MIT Sloan School of Management. This analysis, using summary data on industry-reported expenditures and revenues, indicated that application of the model would reduce mean capitalized expenditures by an average of $283 million per successful new drug (from $727 to $444), and increase revenues by an average of $160 million per Phase III trial
Contact: Anna Gonski
Children's Hospital Boston