TL;DR — Key Takeaways
- Rentosertib, an AI-designed drug being studied for idiopathic pulmonary fibrosis, also showed signs of potentially reducing biological age.
- Six proteomic aging clocks detected shifts toward a younger predicted biological age in patients receiving the drug compared with placebo.
- The strongest effects appeared after four weeks, particularly in patients taking 30 milligrams twice daily.
Along the way of evaluating how well an AI-designed drug called rentosertib could treat a chronic lung disease, researchers noticed something else happening in the clinical trial. The drug showed signs that it might also be putting the brakes on biological aging.
The study of rentosertib as a treatment for idiopathic pulmonary fibrosis, or IPF, a progressive disease in which the lungs become scarred and stiff, is continuing. But now researchers have another intriguing reason to study the drug: Six different “aging clocks” that analyzed proteins in patients’ blood all pointed in the same direction, toward a younger predicted biological age.
That does not mean the patients actually became younger, or that rentosertib has been proved to slow human aging. But the consistency across six independently developed clocks was striking enough that researchers say the possibility of an anti-aging effect deserves further investigation.
The findings, published Sept. 7 in Nature Biotechnology, grew out of a 12-week phase 2a clinical trial of rentosertib, an experimental drug developed by Insilico Medicine using AI.
Rentosertib is a small-molecule drug that inhibits a protein called TNIK, which has been linked to fibrosis and several biological processes associated with aging. AI was used both to identify TNIK as a potential therapeutic target and to help design the drug intended to block it.
The original trial involved 71 patients with IPF at 21 sites in China. For the aging study, researchers analyzed blood samples from 42 participants, with an average age of about 67, who had provided samples at four points during the trial.
Those samples contained thousands of proteins, and within them, researchers went looking for signs of aging. Chronological age is easy. It is simply the number of years since you were born, but biological age is more complicated.
Two people can both be 67, for example, while their bodies may show very different degrees of aging. One may have healthier blood vessels, organs and metabolism than the other. Scientists have spent years trying to measure those differences by developing biological “clocks” that look for molecular changes associated with getting older.
Some aging clocks examine chemical changes to DNA. The six used in the rentosertib study are known as proteomic clocks because they analyze proteins circulating in the blood.
Proteins perform much of the body’s day-to-day work, helping regulate everything from metabolism and inflammation to immune responses and tissue repair. As we age, the levels of certain proteins tend to rise or fall in patterns.
Researchers can analyze those patterns and compare them with data from large groups of people of known ages. Algorithms then estimate a person’s biological age based on the protein profile.
Think of it as looking under the hood of a car. The odometer tells you how many miles it has traveled. Examining the engine, belts, hoses and fluids tells you something different: how much wear and tear the machine has actually experienced.
In this study, all six clocks detected shifts toward lower predicted biological age among patients receiving rentosertib compared with those receiving a placebo. The strongest and most consistent effects appeared at four weeks, particularly among patients taking 30 milligrams twice daily. Five of the six clocks showed an effect for that dosage at that point in the trial.
The researchers then went a step further, working to eliminate an inherent possibility, that the clocks were simply registering that patients’ lungs were improving and that tilted the biological register. To investigate, researchers compared protein changes in the rentosertib patients with normal age-related protein changes found among more than 55,000 older participants in the UK Biobank.
There remain significant questions. The aging analysis involved only 42 people, all of whom had IPF. The trial lasted just 12 weeks, and proteomic clocks are still an emerging method of measuring biological aging. The researchers acknowledge that they cannot completely disentangle the drug’s effects on fibrosis from its possible effects on aging. Confirming an anti-aging effect would require additional research, potentially including studies involving people without IPF.
So rentosertib is a long way from becoming a longevity pill, but the study offers a glimpse at another potentially profound use of AI in medicine.
AI is increasingly being used not just to analyze scans or predict disease, but to help scientists invent drugs. Algorithms can sift through enormous amounts of biological and chemical data to identify promising targets, predict how molecules might interact with them and help design entirely new compounds.
Rentosertib is one example. Researchers at the University of California, Irvine, and GATC Health have also used an AI-assisted approach to develop a potential drug targeting the serotonin 5-HT2A receptor for opioid use disorder. Other AI-assisted drugs are being investigated for inflammatory bowel disease and other conditions.