Insilico's AI-Designed IPF Drug Lowered Biological Age Markers

An AI-designed drug for a lung disease made trial patients look 2.7 to 3.5 years biologically younger on chronological aging clocks at week four, according to a proteomic reanalysis of Insilico Medicine's Phase IIa trial published September 7 in Nature Biotechnology and reported by Bloomberg.

The reanalysis measured 2,841 blood proteins from 42 of the 71 idiopathic pulmonary fibrosis patients enrolled across 21 sites in China, then scored the samples against six independently developed proteomic clocks (ProtAge, two OrganAge variants, PAC, ipfP3GPT and PAOPAC), benchmarked to 55,319 UK Biobank profiles. All six clocks moved in the same direction, and 21 of 54 statistical comparisons reached significance across arms and timepoints, The Next Web reported.

The drug, rentosertib, is a TNIK inhibitor whose target was picked by Insilico's PandaOmics platform and whose molecule came out of its Chemistry42 generative chemistry platform.

The paper is careful about what it can prove. The authors write that the study "cannot yet separate slower ageing from a treated lung," meaning the biological-age reduction could be a downstream effect of easing fibrosis rather than a systemic anti-aging effect. Michael Levitt, the 2013 Chemistry Nobel laureate, told Unite.AI that "what convinces me is not the size of the effect but the agreement" across the independent clocks.

On the lung side the numbers were cleaner. Forced vital capacity rose by 98.4 ml in the 30mg twice-daily arm versus a 20.3 ml decline on placebo. Seven patients discontinued for liver toxicity, four of them also taking the standard-of-care fibrosis drug nintedanib. Rentosertib is now in a Phase III IPF trial. It lands alongside 59 other healthcare AI stories we've tracked over the last 90 days.