Artificial intelligence companies are running virtual drug trials to raise the chance that experimental medicines succeed in costly human studies. The global biopharmaceutical industry spends about $140 billion a year on clinical testing, yet only around 12 percent of candidates gain regulatory approval, a rate that has barely changed for decades. Firms such as BioinvestGPT and QuantHealth say their simulations help decide which programs to advance and how to value acquisition targets. Human trials take years; some AI simulations finish in a month or less. BioinvestGPT correctly predicted the outcomes of five of six high-profile trials it analysed in advance, including the recent late-stage failure of Novartis’ del-desiran. The platform uses DNA sequencing to build virtual patients matching trial criteria and then tests a model of the drug. QuantHealth’s chief medical officer said the industry should focus on running fewer trials that are likely to fail.
Artificial intelligence companies are running virtual drug trials to raise the chance that experimental medicines succeed in costly human studies. The global biopharmaceutical industry spends about $140 billion a year on clinical testing, yet only around 12 percent of candidates gain regulatory approval, a rate that has barely changed for decades. Firms such as BioinvestGPT and QuantHealth say their simulations help decide which programs to advance and how to value acquisition targets. Human trials take years; some AI simulations finish in a month or less. BioinvestGPT correctly predicted the outcomes of five of six high-profile trials it analysed in advance, including the recent late-stage failure of Novartis’ del-desiran. The platform uses DNA sequencing to build virtual patients matching trial criteria and then tests a model of the drug. QuantHealth’s chief medical officer said the industry should focus on running fewer trials that are likely to fail.