The performance of AI-discovered drugs in clinical settings is generating significant interest, particularly given the traditionally lengthy and costly drug development process, which can exceed $2.6 billion and take up to 15 years. While AI has been heralded as a transformative force in drug R&D, the actual success of AI-generated drugs remains uncertain.
Currently, only a handful of AI-discovered drugs have entered clinical trials, with those that have shown a higher success rate in phase 1 compared to conventional drugs. However, this advantage may not persist into phase 2 trials, as evidenced by setbacks like Verge Genomics’ ALS candidate, which failed to progress beyond phase 1.
Among notable AI-driven candidates, Insilico Medicine has emerged as a key player, with over 40 programs in development. Its drug rentosertib, designed through an innovative AI approach targeting TNIK inhibition, has shown promise in treating idiopathic pulmonary fibrosis (IPF) and is now advancing to phase 3 trials in China. Another candidate, garutadustat, aims to address inflammatory bowel disease and is currently in phase 2 trials.
Recursion Pharma has also made strides in AI drug development, despite facing challenges. Its drug REC-4881 has demonstrated a significant reduction in polyp burden in patients with familial adenomatous polyposis, suggesting potential for treating conditions with limited existing options. As these companies continue to navigate the complexities of drug development, the outcomes of their trials will be critical in determining the viability of AI in this space.