Phase 2 study data indicates ‘dose-dependent improvement in lung function’ for TNIK-targeting therapeutic.
Generative AI-powered longevity biotech Insilico Medicine has reported positive preliminary results from its Phase 2a clinical trial for ISM001-055, a novel first-in-class drug for the treatment of idiopathic pulmonary fibrosis (IPF). The company says the trial data shows its AI-designed drug has a favorable safety profile and a dose-dependent improvement in lung function, as measured by forced vital capacity (FVC), a key metric in IPF treatment.
IPF is a chronic, progressive disease marked by irreversible scarring of lung tissue, which leads to severe impairment in respiratory function. Affecting roughly 5 million people worldwide, the condition typically presents in individuals between the ages of 60 and 70, and has a median survival rate of 3 to 4 years after diagnosis.
ISM001-055 targets TNIK (Traf2- and Nck-interacting kinase), a protein involved in pathological fibrosis of the lungs, which contributes to the progressive decline in lung function in IPF patients. Insilico’s proprietary AI platform played a pivotal role in the discovery and design of ISM001-055, leveraging deep generative models and machine learning to identify TNIK as a novel therapeutic target in IPF. The drug’s development was detailed in a March 2024 paper in Nature Biotechnology, which described the preclinical evaluations and initial clinical studies that supported its advancement into later-stage testing.
The 12-week Phase 2a clinical trial enrolled 71 patients across 21 sites in China, with participants receiving either a placebo or varying doses of the drug: 30 mg once or twice daily, or 60 mg once daily. The trial’s primary endpoint was safety, and secondary efficacy endpoints included changes in lung function. The results indicated that the drug was well-tolerated across all dose levels, with the highest dose group (60 mg once daily) showing the most significant improvement in FVC.
“These results are very encouraging, particularly the dose-dependent response in FVC,” said trial investigator Dr Toby M Maher, an expert in interstitial lung disease. “IPF is a devastating disease, and seeing improvements in lung function over just 12 weeks of treatment is a promising indication that ISM001-055 may provide a new therapeutic option for patients.”
Validation of AI drug discovery?
Insilico’s leadership hailed the data as a potentially significant breakthrough, not only in IPF treatment but also as a validation of AI-driven drug discovery.
“This study result represents a critical milestone in AI-powered drug discovery and in my life to date,” said Dr Alex Zhavoronkov, co-CEO of Insilico. “While we expected the drug to be safe, we did not expect to see such a clear dose-dependent efficacy signal after such a short dosing period. IPF is a very diverse disease, and it is very rare to see improvement in FVC. With our novel TNIK inhibitor, we attempted to go after what we think is a common mechanism in fibrotic diseases and in aging to maximize indication expansion potential.”
“I am excited to see that ISM001-055 demonstrated obvious clinical efficacy in IPF patients in only 3-months treatment,” added Dr Feng Ren, co-CEO and CSO of Insilico. “While preliminary, this clinical data is certainly encouraging, and provides the clinical validation of AI-powered drug R&D for both novel target and novel molecule.”
Insilico now plans to engage regulatory authorities to discuss a more extensive Phase 2b trial. This next phase is expected to involve longer treatment periods and larger patient cohorts to further evaluate the drug’s efficacy and safety. In addition to the trial conducted in China, a parallel Phase 2a study is currently underway in the United States, with patient enrollment ongoing. The company says drug’s ability to target a common mechanism in fibrotic diseases also raises the possibility of expanding its use to other conditions beyond IPF, opening doors for broader therapeutic applications.
Scientific leaders also weighed in on the news and its wider implications for AI drug discovery and design.
“Last year, I presented a lecture on how generative AI can help with end-to-end drug discovery from disease modeling and target identification to generation of novel drugs with the desired properties and purposing it to a specific disease,” said Michael Levitt, PhD, the 2013 Nobel Laureate in Chemistry. “I used Insilico’s TRAF2 and NCK-interacting kinase (TNIK) inhibitor as a case study going from 0 to Phase 1. The fact that this same drug demonstrated efficacy in addition to safety in a Phase 2a study is extraordinary and represents a true first in this new era of AI-powered drug discovery.”
“With all the hype that surrounds the potential of generative AI in drug discovery and many other potential applications, it is thrilling see to the dose dependence of ISM001-055 in Insilico Medicine’s Phase 2a IPF clinical trial,” said the renowned geneticist Charles Cantor, PhD.
