AI-designed drug candidate reverses biological age in clinical study
Insilico Medicine has reported findings from a Phase IIa clinical trial of rentosertib showing consistent reductions in predicted biological age among treated patients. Researchers analyzed longitudinal Olink proteomic data using six independently developed proteomic aging clocks. All six models indicated a decrease in predicted biological age following treatment, with the strongest effects observed in the 30 mg twice-daily group at Week 4. Several models suggested approximately 3–4 years of biological age reversal, while one indicated a reduction of up to six years.
The study also reported promising dose-dependent improvements in Forced Vital Capacity (FVC), a key measure of lung function that typically declines with age. These results broadly corresponded with the biological age changes identified through proteomic analysis, although the dose producing the greatest lung-function improvement differed from the dose associated with the strongest age-reversal signal.
Published in Nature Biotechnology, the research provides a clinical assessment of a drug candidate developed through an AI-driven approach that combines an AI-identified target with an AI-designed molecule. Insilico identified TNIK as a potential target linked to both aging biology and fibrosis and subsequently used its generative chemistry platform, Chemistry42, to develop rentosertib, a small-molecule TNIK inhibitor.
The analysis included serum proteomic profiles from 42 clinical-trial participants covering 2,841 proteins. Despite using different methodologies and training approaches, all six aging clocks consistently showed lower predicted biological age in rentosertib-treated participants compared with those receiving placebo.
Comparison with 55,319 UK Biobank profiles further indicated that rentosertib shifted protein-expression patterns away from typical age-related changes. The treatment also appeared to affect biological pathways involved in cellular senescence, growth-factor signaling, antioxidant activity, and cholesterol metabolism.
These findings suggest that rentosertib may have biological effects extending beyond its potential benefits for pulmonary fibrosis. However, the results remain an early clinical proof of concept, and additional research is needed to establish whether changes in proteomic aging measures can translate into meaningful improvements in healthspan or lifespan.
The study also highlights the potential value of incorporating aging biomarkers into conventional clinical trials. Using such biomarkers as exploratory endpoints, followed by formal validation and qualification, could provide a new strategy for identifying and developing potential geroprotective therapies more efficiently.
