Insilico Medicine, via it's latest Press Release, has announced that it initiated a pivotal Phase III clinical trial for Rentosertib, an oral TNIK inhibitor for idiopathic pulmonary fibrosis. What makes this trial unusual is not the disease target or the trial design. It is that Rentosertib was discovered and designed entirely by artificial intelligence.
The study is a 52-week, prospective, randomised, double-blind, placebo-controlled trial expected to recruit 320 patients over the age of 40 across sites in China. If successful, Rentosertib would become the first drug to reach market approval that was both identified and designed by generative AI, a milestone the pharmaceutical industry has been working toward for the better part of a decade.
Insilico's approach is worth understanding because it differs from how drugs are typically found. In conventional drug discovery, researchers start with a biological target and screen large libraries of existing compounds for activity against it. Insilico's platform instead used generative AI to design a novel molecular structure from scratch, one that would bind to TNIK, a kinase involved in fibrotic signalling pathways. The company has described Rentosertib as a potentially first-in-class small molecule, meaning no existing drug works through the same mechanism.
The path to Phase III has been relatively fast. Insilico advanced Rentosertib from initial concept to Phase I in roughly 30 months, a timeline that would be difficult to match with traditional methods. Alex Aliper, Insilico's co-founder and president, is scheduled to discuss the company's AI-driven pipeline at the AI for Good Global Summit in Geneva on July 7-10.
Idiopathic pulmonary fibrosis is a devastating condition with limited treatment options and a median survival of three to five years after diagnosis. The unmet medical need is substantial, which partly explains why regulators and investors have followed Insilico's progress with such close attention. IPF carries a poor prognosis and current therapies can slow but not reverse lung scarring.
The trial will face the same rigorous standards as any other Phase III study. The fact that AI played a role in the drug's invention does not lower the evidentiary bar. But if Rentosertib succeeds, it will mark a turning point not just for Insilico but for the entire field of AI-driven pharmaceutical research, offering proof that algorithms can do more than assist scientists: they can originate medicines.