In mice, shutting down a single aging-related enzyme has rebuilt worn knee cartilage, restored strength to withered muscles, shielded the blood-brain barrier, and preserved memory in animals bred to develop Alzheimer’s-like damage.
The enzyme is called 15-PGDH, short for 15-hydroxyprostaglandin dehydrogenase. The Stanford scientists who study it gave its kind a new name in 2023: “gerozymes,” proteins that become more active with age and quietly dismantle the body’s repair systems.
The drug designed to block it, the MF-300 molecule, has left the laboratory. Developed by the San Diego biotech Epirium Bio, it has completed an early human safety study with no serious adverse events and is now moving into Phase 2 testing in patients with age-related muscle weakness. Its fate will help determine how far this new class of medicine can go, not just for muscles, but for joints, bones, and possibly the brain.
What Is the MF-300 Molecule?
MF-300 is a small molecule, not a peptide or a biologic, small enough to be swallowed as a once-daily pill. Epirium describes it as an oral 15-PGDH inhibitor that “reversibly occupies” the prostaglandin E2 binding site of the enzyme, making it one of the first gerozyme inhibitors to reach mid-stage human testing.
That matters because 15-PGDH’s day job is destroying prostaglandin E2 (PGE2), a fleeting, hormone-like signal that helps tissues repair themselves. PGE2 is best known, unfairly, for its role in pain and inflammation. At the low levels the body maintains naturally, it also appears to drive healing. Block the enzyme, and PGE2 lingers a little longer, quietly boosting the tissue’s ability to fix itself.
Epirium’s first target is sarcopenia, the muscle loss that comes with age. By the FDA’s estimate, up to a third of Americans over 60 are affected, and there are no approved drugs. “Over 20 million seniors” have the condition, according to the company’s CEO.
The Enzyme Scientists Call a “Gerozyme”
Aging used to be treated as a background condition of medicine: something to manage, never modify. That view is fading. Researchers have rejuvenated aged brain stem cells in rats, rebuilt worn joints in mice, and achieved results that would have sounded absurd a generation ago.
The gerozyme concept is one of the boldest entries. Helen Blau, a stem cell biologist, and Nidhi Bhutani, an orthopedic researcher, both at Stanford Medicine, use the word for enzymes whose activity climbs with age and whose work erodes tissue as it goes. Their founding example is 15-PGDH, which roughly doubles in aged muscle and in aged cartilage.
The proof runs in both directions. When the team forced young muscles to produce extra 15-PGDH, the muscles shrank and weakened. When they blocked the enzyme in old mice, muscle mass and endurance climbed. “We therefore coined the term ‘gerozyme’ to define a class of dysregulated enzymes that accelerate tissue aging,” the team wrote in a landmark paper published in Science in November 2025.
The enzyme’s own record is telling. People born with two disabled copies of the 15-PGDH gene never clear PGE2 properly, and the result is a rare inherited condition whose hallmarks include clubbed fingertips, overgrowth of bone around the long bones, thickened skin, and, in some infants, a heart duct that fails to close. Mice engineered to lack the enzyme die within days of birth for the same reason. And in many colon, lung, breast, and bladder cancers, the enzyme is switched off: there it acts as a tumor suppressor, a natural brake on the same growth-promoting prostaglandin signals that help tissues repair. Blocking that brake for years is one reason long-term safety data matter. What drives the enzyme to climb with age is still an open question.
What Blocking It Has Done So Far
What follows are animal studies and human tissue experiments, unless noted otherwise. No patient has yet been treated with these drugs for any of the diseases below.
Muscle: Strength Returns
In a 2021 study in Science, Blau’s team showed that 15-PGDH accumulates in aging muscle and that blocking it, genetically or with a drug, boosted muscle mass, strength, and exercise performance in old mice. Higher PGE2 levels improved the energy centers of cells and dialed down two pathways that chew up muscle protein.
MF-300 itself was then tested in aged mice. Dosed by mouth every other day for 12 weeks, it “reverses age-related muscle weakness” by improving muscle quality, meaning strength relative to size, with the biggest gains in fast-twitch fibers, the kind that fade most with age. Nerve-to-muscle connections improved, and treated mice even gained better bone micro-architecture, the fine structure that decides fracture risk.
Joints: Cartilage Grows Back
Osteoarthritis, the loss of the smooth hyaline cartilage that cushions joints, affects about one in five US adults and costs roughly $65 billion a year in direct medical care. No drug has ever slowed the disease, let alone reversed it. For severe cases, the only real fix is a new joint.
The Stanford Medicine-led study, published in Science in November 2025, began with a discovery: 15-PGDH doubles in the knee cartilage of old mice. Then came the intervention. In aged mice, blocking the enzyme, first system-wide and then with injections directly into the knee, thickened cartilage that age had thinned. In mice with ACL-style knee injuries, the kind that lead to arthritis in about half of patients within 15 years, a month of twice-weekly injections after injury dramatically cut the odds that arthritis developed. Treated animals walked and bore weight more normally.
The results in human tissue raised the room’s temperature. Cartilage samples taken from patients during knee replacement surgery and treated in the lab began producing new, functional cartilage.
“This gerozyme inhibitor causes a dramatic regeneration of cartilage beyond that reported in response to any other drug or intervention,” Bhutani said.
The mechanism stunned the researchers. The regeneration was not the work of stem cells. Mature cartilage cells, called chondrocytes, simply switched their gene expression back toward a younger pattern. In treated mice, a degraded and inflammatory cell population fell from 8% to 3% of sampled cells, a scar-like fibrous population fell from 16% to 8%, and a youthful, cartilage-building population jumped from 22% to 42%.
“Imagine regrowing existing cartilage and avoiding joint replacement,” Blau said.
The Brain: An Unexpected Shield
The most unexpected discoveries came from the brain. A team co-led by Andrew Pieper and Sanford Markowitz at Case Western Reserve University, University Hospitals Cleveland, and the Cleveland VA found that 15-PGDH is unusually active in the blood-brain barrier, the tightly sealed wall that protects the brain, and that its activity rises in Alzheimer’s disease, traumatic brain injury, and even normal aging.
In mice engineered to develop Alzheimer’s-like damage, blocking the enzyme protected the barrier’s structure and function, normalized markers of oxidative stress, and rescued newborn neurons that normally die off. In memory tests, treated animals performed “at levels indistinguishable from healthy controls.” For traumatic brain injury, treatment started 24 hours after the trauma and still prevented memory deficits. The study, published in the Proceedings of the National Academy of Sciences, earned a 2025 Cozzarelli Prize, one of the National Academy of Sciences’ top honors for its journal.
“Most notably, inhibiting 15-PGDH not only halts neurodegeneration but also preserves cognitive function at levels indistinguishable from healthy controls,” the authors wrote.
Two details stand out. First, the protection happened without touching amyloid plaques, the sticky protein clumps that have dominated Alzheimer’s research for decades. “Remarkably, these neuroprotective effects in AD are achieved without affecting amyloid pathology,” the authors wrote, suggesting the drug works by hardening the brain against damage rather than clearing the culprit protein. Second, the timing: the Alzheimer’s mice were treated before symptoms began, and the injured mice within a day of their trauma. Nobody has yet reversed long-established brain damage. Dead neurons do not return. Researchers have restored memory in mice before through other routes, by remodeling the scaffolding around neurons, and the open question is whether this approach can do more than protect.
The gerozyme approach is the newest entry in a longer line of research: scientists have linked leaky blood-brain barriers and brain inflammation to age-related cognitive decline, and by one MRI estimate, nearly 60% of adults over 70 have a leaky barrier. Quieting that inflammatory response has been shown to restore youthful behavior in old mice.
Parkinson’s: Protection in Three Models
In June 2026, the same Cleveland-led group, with collaborators at Seoul National University, reported in Redox Biology that 15-PGDH is elevated in the brains of people who died with Parkinson’s disease. Then they tested blocking it in three different mouse models of the disease, including one built on alpha-synuclein, the protein that clumps in patients’ neurons. Across all three, treatment protected dopamine-producing neurons and preserved movement, as measured by tests of balance and coordination.
As with Alzheimer’s, the benefit did not come from clearing the disease’s hallmark protein. Treated mice still accumulated alpha-synuclein clumps; they simply withstood them better, with less oxidative damage and calmer immune cells in the brain.
Getting drugs into the brain has always been Parkinson’s hardest problem. One 2019 trial infused its protein directly into patients’ brains through surgically implanted tubes. This approach is different: the target sits inside the brain already, and the research compound used in the studies reached it readily. Whether the oral drug does the same is one of the questions the coming years will answer.
“MF-300 has already completed human phase 1 trials,” the authors wrote. “Our results now provide the rationale to repurpose such agents for the treatment of PD.” University Hospitals Cleveland called the finding “a potential new treatment strategy” for Parkinson’s and other neurodegenerative conditions.
Beyond Muscles, Joints and the Brain
The list keeps growing. In animal studies and human tissue experiments, 15-PGDH inhibition has:
- sped recovery of the blood system after bone marrow transplant, the finding that launched the field in 2015, and mice lacking the enzyme kept better blood and gut fitness as they aged;
- promoted liver regeneration and protected the gut lining; in colitis models, Epirium’s oral inhibitors MF-300 and MF-1305 eased disease activity, inflammation, and colon damage, with MF-300 matching an antibody control;
- reduced lung scarring in models of pulmonary fibrosis, now in Epirium’s preclinical pipeline;
- protected the kidneys from acute injury in several mouse models, including from restricted blood flow, contrast dye, and sepsis;
- reduced heart scarring and stiffness in mice, and reversed the activated state of heart fibroblasts taken from patients with heart failure;
- accelerated skin wound repair in skin-cell and animal models, and encouraged hair growth in human follicle cells and cultured human follicles;
- improved bone micro-architecture in aged mice and boosted new bone formation, an effect that ran through the PGE2 receptor EP4;
- helped repair the nerve-muscle connections that weaken with age.
That breadth is exactly the gerozyme thesis: one aging-driven enzyme, sabotaging many tissues at once. It is also why Amgen paid $55 million upfront, with up to $666 million more in milestones, to buy the 15-PGDH program’s original developer, Rodeo Therapeutics, in 2021. The deal is one reason the field’s progress is worth watching so closely.
The Human Chapter Has Begun
Human data so far comes from safety trials, but it has been encouraging. In a Phase 1 study completed in 2025, 100 healthy volunteers, including adults over 65, received MF-300 or a placebo in single and multiple doses up to 800 milligrams. There were no serious adverse events; all participants finished the study; most side effects were mild. The drug absorbed reliably, supported once-daily dosing, and produced dose-dependent changes in PGE2 breakdown products, evidence that it was hitting its target. In the multiple-dose portion, PGE-MUM, the major urinary metabolite of PGE2, fell by as much as 83% at the highest dose tested, a signal worth watching because PGE-MUM tracks disease activity in inflammatory bowel disease.
In older adults, those changes were especially striking: reductions in PGE2 metabolites “consistent with levels associated with maximal gains in muscle force in aged mouse models,” plus PGE2 increases “comparable to increases observed in human muscle tissue after exercise,” CEO Alex Casdin said.
Regulators appear convinced the signal is worth chasing. In January 2026, the FDA agreed with Epirium on the design of a Phase 2b trial: about 200 patients, six months of treatment, with muscle function, strength, and quality-of-life endpoints. The company plans to seek Fast Track designation, which can speed review.
Epirium’s pipeline lists the Phase 2b as in start-up, before any patients have been dosed. A smaller open-label Phase 2 study, in which every participant receives the drug, is expected to dose its first patient in the first quarter of 2027, with 12-week data in the third quarter and 24-week data by the end of the year. The larger double-blind study is expected to begin dosing in the second quarter of 2027, with topline results in 2028.
Beyond muscle, Epirium lists osteoporosis as its next target, with human proof-of-concept data on bone biomarkers and bone mineral density expected to come out of the Phase 2b study itself. The program has also generated preclinical data in spinal muscular atrophy, inflammatory bowel disease, and idiopathic pulmonary fibrosis. Researchers in the brain field are openly calling for their drug to be tested in Parkinson’s next, and Blau’s team hopes a cartilage trial will follow.
What the Science Does, and Doesn’t, Say
Here is what the evidence shows so far.
- Reversed in animals: muscle and cartilage damage was repaired in mice, and human cartilage perked up in the lab. No human has yet had cartilage or muscle regrown by this drug; that is what trials are for.
- Protected, not resurrected, in the brain: the Alzheimer’s mice were treated before symptoms appeared, and the injured and Parkinson’s models early in the process. Nothing yet shows a reversal of long-established brain disease, and lost neurons are not coming back.
- After you stop: structurally repaired tissue should largely stay repaired, though aging itself continues. For brain diseases, progression would most likely resume once treatment stops, because the underlying proteins are still there. A lingering benefit is possible if a repaired blood-brain barrier keeps more damage out, but that is a theory, not a finding.
- The gut: the enzyme is highly active in the gut lining, and blocking it protects that barrier and eases colitis in animals.
- Safety looks clean so far: no serious side effects in Phase 1, no toxicity signals in animal studies, and the drugs block the enzyme reversibly rather than removing it. Long-term safety is still unknown, and a 200-person trial is early days.
One transparency note: science like this arrives with interests attached. Blau co-founded Epirium and holds equity in the company; Stanford licensed related patents to Epirium. The patents behind the brain studies are licensed to Amgen. The research passed peer review at major journals, but knowing who stands to gain is part of reading it well, and the authors disclose these relationships themselves.
Quick Answers
What is an oral 15-PGDH inhibitor used for?
Right now, the closest to a use is sarcopenia, age-related muscle weakness, in Phase 2 testing. If the approach works, researchers hope it could be adapted for osteoarthritis, Parkinson’s disease, Alzheimer’s disease, fibrosis, and other conditions where aging tissues fail to repair themselves.
Is there a clinical trial using a 15-PGDH inhibitor?
Yes. Epirium’s Phase 2b trial of MF-300 in sarcopenia is in start-up, with the first patients expected in 2027, following a completed Phase 1 safety study in 100 volunteers.
Is MF-300 a peptide?
No. It is often mistaken for one online, but MF-300 is a small molecule, the classic pill-style drug, not a peptide or a protein.
Can you get MF-300 today?
Not as a medicine. MF-300 has not been approved for any use, so patients cannot get it outside clinical trials. Research-grade 15-PGDH inhibitors, including one catalogued as MF-DH-300, are sold to laboratories for research use only, not for human use.
The Road Ahead
In the space of a decade, 15-PGDH has gone from an obscure enzyme to a potential master switch of tissue aging, with a drug already tested in people and studies spanning muscles, joints, bones, blood, gut, lungs, and brain. The 200-patient double-blind trial, due to report in 2028, will help decide the MF-300 molecule’s near-term fate.
The mice have already delivered the proof of concept. The humans will deliver the verdict. If the Phase 2b trial succeeds, it would put the first approved medicine for sarcopenia within reach, and open the first chapter of a far bolder idea: that the damage of aging is not destiny, but biology we can negotiate with.
