AI-designed lung disease drug lowers biological age markers in early trial

Rentosertib is an AI-designed TRAF2- and NCK-interacting kinase (TNIK) inhibitor. TNIK is a protein enzyme found throughout the body that plays an important role in cell growth and brain health.

Past studies have linked overactive TNIK activity to certain health conditions, including certain cancers, fibrotic diseases like idiopathic pulmonary fibrosis (IPF), and metabolic disorders.

For this study, researchers collected blood samples from 42 participants taking rentosertib at different times over a 12-week period. The participants had a confirmed IPF diagnosis, and otherwise were not healthy adults.

Scientists measured the concentrations of more than 2,800 different proteins in the blood samples, which were then applied to six different types of proteomic aging clocks — a test that uses protein patterns and machine learning to measure a person’s biological age, rather than their calendar age.

The six proteomic aging clocks included:

• ipfP3GPT (developed by Insilico Medicine, which created rentosertib)
• OrganAge (chronological and mortality variants)
• PAC (Proteomic Aging Clock)
• PAOPAC (Proteome-Aware Organ Proxy Aging Clock)
• ProtAge

At the study’s conclusion, researchers found that study participants taking rentosertib experienced reversal of predicted biological aging through six different models of proteomic aging clocks.

The “peak effect” reportedly took place at week 4, with participants showing about a 3 to 4-year reversal in biological age, and up to six years in a specific aging clock.

Dung Trinh, MD, an internist for MemorialCare Medical Group and chief medical officer of the Healthy Brain Clinic in Irvine, CA — who was not involved in this study — said these findings are exciting, but need to be interpreted cautiously.

“The study found that six different proteomic aging clocks all shifted toward a younger predicted biological age in patients treated with rentosertib, which is intriguing because the signal was consistent across multiple models,” Trinh told Medical News Today.

“However, this does not prove that the drug reverses human aging. The study was small, involved patients with idiopathic pulmonary fibrosis, and lasted only 12 weeks, so some of the biomarker changes may reflect improvement in the underlying disease.”
— Dung Trinh, MD

“I see it as a promising signal that we may be getting better at measuring how treatments affect biology associated with aging, but there is not yet evidence that patients can be made biologically younger,” Trinh added.

MNT also spoke with Zeeshan Khan, MD, chief of geriatric medicine at Hackensack Meridian Jersey Shore University Medical Center in New Jersey, who told MNT that while the findings are exciting, particularly because multiple measures of biological aging moved in the same direction.

However, a younger score on an aging clock is not the same as proving we have reversed aging.

“What ultimately matters is whether patients live healthier and more functional lives,” Khan, who likewise was not involved in this study, explained.

“In geriatrics, our goal isn’t simply to extend one’s life span, it’s to extend their healthspan. That is to say, we want to increase the years that people remain healthy, independent, and cognitively/physically functional.”
— Zeeshan Khan, MD

“If we are able to safely target biological processes that contribute to multiple age-related diseases at the same time, the impact could be much greater than treating each disease only after it develops,” he added.

As rentosertib is a drug that was discovered and designed using generative AI, MNT askedPeter Gliebus, MD, chief of neurology and director of cognitive and behavioral neurology at Marcus Neuroscience Institute, a part of Baptist Health South Florida, if this is where the future of drug development is going.

Gliebus said he does think artificial intelligence will play an increasingly important role in drug development, as it has the ability to analyze enormous amounts of biological and molecular data much more rapidly than traditional approaches, and potentially identify drug targets or compounds that researchers might otherwise overlook.

“Rentosertib is an interesting example because AI was used both to identify TNIK as a potential therapeutic target and to help design the molecule that inhibits it,” he explained.

“The earlier development program reportedly progressed from target discovery to a preclinical candidate in approximately 18 months. That illustrates the potential for AI to make parts of drug discovery faster and more efficient,” he continued.

AI won’t be replacing humans

“At the same time, AI will not replace the need for physicians, scientists or rigorous clinical trials. A computer can help identify a promising molecule, but researchers still have to determine whether that molecule is safe, whether it actually works in humans and whether its benefits outweigh its risks. I see AI as a powerful tool that could accelerate drug development rather than something that replaces the traditional scientific process.”
— Peter Gliebus, MD