A whole new control panel in a machine everyone thought they understood
In the quiet precision of a Cambridge laboratory, scientists have found a way to silence an enzyme that has long driven one of the most aggressive blood cancers. By blocking METTL3—a molecular switch that corrupts the body's RNA instructions and unleashes uncontrolled cell growth—a new compound called STM2457 has shown it can halt acute myeloid leukaemia in its tracks, extending life in animal models without apparent harm. It is a discovery that opens not just a new treatment, but an entirely new class of cancer medicine, built on biological machinery that science had scarcely thought to challenge.
- Acute myeloid leukaemia strikes over 3,000 people in the UK each year, moving fast through the blood and bone marrow and demanding immediate, often brutal intervention.
- The cancer's engine—an overactive enzyme called METTL3—had gone untargeted for decades, hiding within a class of RNA-modifying proteins that cancer researchers had largely overlooked.
- STM2457 acts as a direct brake on METTL3, triggering programmed death in leukaemic cells and dramatically slowing cancer progression in mouse models, all without detectable toxic side-effects.
- The results, published in Nature and backed by Cancer Research UK and the Wellcome Trust, mark the first time this entire category of enzyme has been successfully weaponised against disease.
- Clinical trials of refined successor molecules could begin as early as 2022, with researchers confident the same strategy may prove effective against multiple cancer types beyond leukaemia.
In laboratory dishes and in mice, a molecule called STM2457 stopped acute myeloid leukaemia from advancing—and the animals treated with it lived significantly longer than those left untreated. For the Cambridge team behind the work, it felt like the opening of a door.
AML develops when the bone marrow's production of white blood cells goes wrong. Normally, DNA is read into RNA, which guides the building of healthy cells. But in AML patients, an enzyme called METTL3 becomes overactive, chemically corrupting RNA in ways that produce abnormal, rapidly multiplying cells. The disease affects around 3,100 people in the UK each year and moves with dangerous speed.
Professor Tony Kouzarides and his team had been studying METTL3 since 2017. STM2457 was their answer: a molecule that acts as a brake on the enzyme, preventing it from converting healthy RNA into cancer-driving instructions. Tested on patient-derived cancer cells, it slowed growth and triggered apoptosis—programmed cell death. In mouse models, it reduced leukaemic cells in the bone marrow and spleen, extended survival, and caused no signs of toxicity.
What made the work genuinely novel was its target. RNA-modifying enzymes like METTL3 had never before been used as a point of attack against cancer—an entire class of biological machinery left unexploited. Dr Konstantinos Tzelepis, who led much of the experimental work, said clinical trials of successor molecules could begin as early as 2022, and that the approach might extend to other cancer types entirely.
Published in Nature and supported by Cancer Research UK, the Wellcome Trust, and Storm Therapeutics—a Cambridge spinout connected to Kouzarides' lab—the research marks a moment when fundamental science became a potential weapon, one that may soon be tested in human patients.
In laboratory dishes and in mice, a new molecule stopped acute myeloid leukaemia in its tracks. The compound, called STM2457, blocked the action of an enzyme that had been driving the cancer forward—and the mice treated with it lived significantly longer than untreated controls. For researchers at Cambridge, it was a moment that felt like the opening of a door.
Acute myeloid leukaemia is a blood cancer that strikes about 3,100 people in the UK each year. It develops when something goes wrong in the bone marrow's production of white blood cells. Normally, the body has a precise system for this: DNA is read and converted into RNA, which then guides the creation of proteins that build and maintain cells. But in AML patients, an enzyme called METTL3 becomes overactive. It chemically alters the RNA in ways that produce abnormal white blood cells—cells that multiply rapidly and aggressively, crowding out healthy blood production. The disease moves fast and demands immediate treatment.
Professor Tony Kouzarides and his team at Cambridge had been studying METTL3 since 2017, when they first mapped out how central this enzyme was to the disease's development. Now they had taken the next step: they had found a way to stop it. The molecule STM2457 acts as a brake on METTL3, preventing it from doing the chemical work that transforms healthy RNA into cancer-driving instructions.
When the researchers tested STM2457 on cancer cells grown from AML patients in culture, the results were clear. The drug reduced how fast the cancer cells multiplied. More than that, it triggered apoptosis—a kind of programmed cell death—killing the leukaemic cells outright. In mouse models of the disease, where human AML cells had been transplanted into the animals, the effect was even more striking. The drug slowed the expansion of the cancer cells, extended how long the mice lived, and reduced the number of leukaemic cells in their bone marrow and spleen. Crucially, the mice showed no signs of toxicity from the treatment itself. Their weight remained stable. No organ damage appeared.
What made this work genuinely novel was that no one had ever targeted this particular mechanism before. The RNA-modifying enzymes that METTL3 belongs to had been largely invisible to cancer researchers—a whole class of biological machinery that had never been weaponized against disease. "This is the beginning of a new era for cancer therapeutics," Kouzarides said. Dr Konstantinos Tzelepis, who led much of the experimental work, was more specific about what came next. Clinical trials of successor molecules could begin as early as 2022, he said. And the researchers believed the approach—targeting these RNA-modifying enzymes—could extend far beyond leukaemia. Other cancers might be vulnerable to the same strategy.
The work was published in Nature and supported by Cancer Research UK, the Wellcome Trust, and patient advocacy groups focused on leukaemia. Storm Therapeutics, a Cambridge spinout company connected to Kouzarides' lab, was involved in the development. What had begun as fundamental research into how cancer cells work had now become a potential weapon—one that could soon be tested in human patients.
Citas Notables
This is the beginning of a new era for cancer therapeutics— Professor Tony Kouzarides, University of Cambridge
This is a brand-new field of research for cancer and the first drug-like molecule of its type to be developed— Dr Konstantinos Tzelepis, Milner Therapeutics Institute
La Conversación del Hearth Otra perspectiva de la historia
Why does this enzyme matter so much? What makes METTL3 special compared to other proteins in the cancer cell?
It's a gatekeeper. Most cancer research targets the genes themselves—the DNA mutations that go wrong. But METTL3 works one step downstream. It edits the instructions after they've been read from DNA. If you think of DNA as a cookbook, METTL3 is rewriting the recipes as they're being used. In leukaemia, it's rewriting them in ways that produce abnormal cells.
And no one had tried to stop it before?
Not in cancer treatment, no. These RNA-modifying enzymes were known to exist, but they weren't on anyone's radar as drug targets. It's like discovering a whole new control panel in a machine everyone thought they understood.
The mice lived longer. But how much longer? What does that translate to in human terms?
The study doesn't give exact numbers, but the effect was significant enough that the researchers felt confident moving toward human trials. In cancer research, when you see that kind of lifespan extension in mice with no toxicity, it's a signal worth pursuing.
What about side effects? The mice didn't get sick from the drug itself?
That's what surprised them. No weight loss, no organ damage, nothing. The drug seemed to hit the cancer cells without poisoning the healthy ones. That's rare in cancer treatment.
So what happens now? Do they just start giving this to patients?
Not yet. They need to test successor molecules—refined versions of STM2457—in actual human trials. That's the real test. What works in a petri dish and in mice doesn't always work in people. But the timeline they mentioned—trials as early as 2022—suggests they're confident enough to move forward quickly.
And they think this could work for other cancers too?
That's the bigger bet. If METTL3 is driving acute myeloid leukaemia, it might be driving other blood cancers or solid tumours the same way. One drug target, potentially many diseases. That's why they're calling it a new era.