Insilico says its AI-designed lung drug lowered biological age markers in a 42-patient trial

Blood samples from a small lung-disease trial have been run through six independently built ageing clocks, and all six returned the same answer: the patients on the drug looked biologically younger than the patients on placebo.

“More lifetimes than humanity lost in all the wars ever fought. If you manage to add 3 years to everyone’s life, the drug should be able to significantly extend the healthy portion of life as well, translating into trillions of dollars in productivity and savings”, said Alex Zhavoronkov, PhD, founder and co-CEO of Insilico Medicine.

The analysis, published in Nature Biotechnology by Insilico Medicine, concerns rentosertib, a TNIK inhibitor for idiopathic pulmonary fibrosis that the company designed using generative AI.

The work is a secondary analysis rather than a new trial. Of the 71 patients in rentosertib’s phase 2a study, 42 consented to proteomic profiling, giving samples at baseline and at weeks two, four, and twelve.

Some 2,841 proteins were measured on an Olink panel and scored against six clocks built by different groups using different methods, with 55,319 UK Biobank profiles as the reference population.

The largest effect appeared at week four in the 30mg twice-daily arm, where the chronological clocks put the treated patients between 2.7 and 3.5 years younger than expected.

Insilico’s own summary of the work describes roughly three to four years, and up to six on certain clocks, which is the upper end of the range rather than its centre.

The authors are direct about the problem with all of it. The study, they write, cannot yet separate slower ageing from a treated lung.

Idiopathic pulmonary fibrosis is itself an inflammatory disease that shows up in blood proteins, so a drug that improves the lung would be expected to move the same markers without touching ageing anywhere else.

They attempted to address this by showing that the 30mg twice-daily arm reversed age-associated protein patterns while placebo drifted along a normal ageing trajectory, which is suggestive rather than conclusive.

The disease is why anyone is trying. Idiopathic pulmonary fibrosis scars the lungs progressively and without a known cause, and the two approved antifibrotics, nintedanib and pirfenidone, slow the decline in lung function rather than halt or reverse it.

A drug that produced a measured gain in forced vital capacity rather than a slower loss would be a meaningful result on its own terms, entirely separately from anything to do with ageing.

The clocks themselves are real research instruments, not marketing. Proteomic ageing scores of this kind predict mortality and the risk of common age-related disease across large and diverse populations, and organ-specific versions have been validated on similar cohorts.

What they are not is a regulatory endpoint. No medicines regulator licenses a drug on a change in proteomic age.

Michael Levitt, the 2013 chemistry Nobel laureate, framed his interest carefully.

“Six proteomic clocks from six independent groups, applied to the same 42 patients, all reported a younger biological age in the treated arms. What convinces me is not the size of the effect but the agreement, because these models share neither their features nor their training data,” he said.

That is the study that would settle whether anything here generalises beyond sick lungs.

The trial underneath the analysis was modest, and its primary endpoint was safety. Across 21 Chinese sites over 12 weeks, adverse events occurred at similar rates in every arm, and the 60mg once-daily group gained 98.4ml of forced vital capacity against a 20.3ml decline on placebo.

Seven patients discontinued because of liver toxicity, four of them while also taking nintedanib, the existing standard of care.

Rentosertib entered phase 3 in July, which is where the efficacy question gets answered for the lung, though not for ageing. Insilico deposited the proteomic data with the China National Center for Bioinformation and released its analysis pipelines as open-source software, which makes the claim checkable by people who did not write it.

For a field where AI drug discovery has produced more announcements than approvals, the deposited data and open pipelines are arguably the more interesting result here. The biological age number is a hypothesis with a sample size of 42.