AI-Designed Lung Fibrosis Drug Lowered Predicted Biological Age in Small Trial Analysis

Rentosertib, Insilico Medicine’s AI-designed drug for idiopathic pulmonary fibrosis, lowered patients’ predicted biological age on six separate proteomic aging clocks, according to a new analysis of its phase 2a trial published in Nature Biotechnology.

The work is a secondary analysis, not a new trial. Of the 71 patients in the 12-week study, 42 consented to proteomic profiling and gave blood samples at baseline and at weeks two, four and 12, The Next Web reported. The researchers measured 2,841 proteins on an Olink panel and scored them against six clocks built with different methods, using 55,319 UK Biobank profiles as the reference population. The clocks came from independent teams at Harvard, Oxford, Beijing and Insilico, The Decoder reported.

All six clocks predicted lower biological age in the treated arms, while the placebo group showed minimal change or slight increases, the paper says. Of 54 comparisons per arm, 21 reached statistical significance, concentrated at week four.

Which doses moved the clocks

The 30 mg twice-daily regimen gave the most consistent signal and was the only arm picked up by both the chronological-age clocks and the mortality-trained clocks. The 60 mg once-daily group, which had the biggest lung-function gain in the original trial, showed significant reductions of 2.71 to 3.46 years on all four chronological clocks at week four but no significant change on the two mortality clocks. Insilico’s own summary describes reductions of roughly three to four years and up to six on certain clocks, which TNW noted is the upper end of the range.

In the underlying trial, whose primary endpoint was safety, the 60 mg once-daily group gained 98.4 ml of forced vital capacity while the placebo group lost 20.3 ml. Seven patients stopped treatment because of liver toxicity, four of them while also taking nintedanib, the existing standard of care, according to TNW.

“This is the first study that shows, very clearly, that predicted biological age can be reduced,” Vadim Gladyshev, a Harvard Medical School professor who helped build one of the clocks, told The New York Times. He also pointed to the small sample and said aging clocks are not always reliable.

Lung disease or aging?

The authors say proteomic clocks alone can’t separate effects on aging from effects on the disease. Idiopathic pulmonary fibrosis is an inflammatory condition that shows up in blood proteins, so a drug that improves the lungs could move the same markers. The team tried to address that indirectly with pathway analyses, which found possible anti-aging shifts in senescence and metabolic processes alongside the drug’s anti-fibrotic activity. The drug has not been tested in healthy people.

“What convinces me is not the size of the effect but the agreement, because these models share neither their features nor their training data,” Michael Levitt, the 2013 chemistry Nobel laureate, said in comments reported by TNW.

Rentosertib entered phase 3 for the lung disease in July. Insilico deposited the proteomic data with the China National Center for Bioinformation and released its analysis pipelines as open-source software, so outside researchers can check the results.