For generations, cancer's cruelest companion has been cachexia — a wasting that steals the body from within, indifferent to how much a patient eats or fights. Researchers at KAIST in South Korea have now traced this destruction to a precise signal in the brainstem, where a protein released by tumors instructs the body to consume itself, and they have developed an RNA-based therapy to silence that instruction at its source. In mice with advanced cachexia, the treatment raised survival from 20 to 90 percent — a result that places human clinical trials on the horizon by 2030, and with them, the p
KAIST team develops RNA therapy to block cancer cachexia via brain pathway
The brain receives an order to consume itself. Block the receiver, and the order never arrives.
Why does the brain matter here? Cancer is in the body—why would blocking a brain signal stop muscle loss?
Because the cancer cell is clever. It doesn't directly destroy muscle. Instead, it sends a chemical messenger—GDF15—that travels to the brainstem and essentially tells the body to consume itself. The brain is the command center. Block the receiver, and the order never arrives.
So the mice that survived—were they actually healthier, or just alive longer?
Both. They regained muscle mass and fat. Their metabolism, which had broken down, started working again. It's not just survival on life support; it's actual recovery of function.
Why hasn't anyone tried this before?
The connection between GDF15 and GFRAL in the brainstem is relatively recent science. And developing an RNA therapy precise enough to target just that one receptor—that's a technical challenge that required the right tools and the right team.
What happens in 2030 when they start human trials?
That's when we find out if what works in mice translates to people. The mice study is proof of principle. Humans are far more complex. But if it works, this becomes something oncologists give alongside chemotherapy, not instead of it.
Could this help other wasting diseases?
Possibly. Any condition where the body is receiving a signal to break itself down might benefit. But cancer cachexia is the immediate target because it's so common and so destructive to treatment outcomes.
The Pulse
- Cancer cachexia quietly kills between half and four-fifths of all cancer patients, draining muscle and fat even when nutrition is adequate and making chemotherapy harder to survive.
- The culprit is a tumor-produced protein, GDF15, that travels to the brainstem and hijacks a receptor called GFRAL — effectively ordering the body to dismantle itself from the inside.
- Previous treatments only addressed hunger, leaving the underlying wasting signal untouched; the KAIST team instead targeted the receptor itself using antisense oligonucleotides that silence the gene before the destructive protein can bind.
- In mice already deep into cachexia, the therapy restored muscle and fat mass, recovered metabolic function, and lifted survival rates from 20 percent to 90 percent by day 50 of the study.
- With findings published in Cell Reports Medicine and a clinical trial target set for 2030, the therapy is being positioned as an adjuvant treatment — one that could preserve patients' bodies while existing cancer drugs do their work.
For generations, cancer's cruelest companion has been cachexia — a wasting that steals the body from within, indifferent to how much a patient eats or fights. Researchers at KAIST in South Korea have now traced this destruction to a precise signal in the brainstem, where a protein released by tumors instructs the body to consume itself, and they have developed an RNA-based therapy to silence that instruction at its source. In mice with advanced cachexia, the treatment raised survival from 20 to 90 percent — a result that places human clinical trials on the horizon by 2030, and with them, the possibility that wasting need no longer be accepted as cancer's inevitable toll.
Cancer cachexia is a quiet killer. It steals from patients even when they eat enough — a relentless wasting that affects between half and four-fifths of everyone diagnosed with cancer. Muscle and fat break down on their own schedule, indifferent to nutrition. Chemotherapy becomes harder to tolerate. Until now, doctors could only offer appetite stimulants, which addressed the symptom while leaving the underlying machinery of destruction untouched.
A team at KAIST, South Korea's advanced science institute, has found where that machinery lives: in the brain. Professors Minho Shong and Jinkuk Kim discovered that as cancer progresses, tumors flood the body with a protein called GDF15, which travels to the brainstem and binds to a receptor called GFRAL. Once bound, it sends a single, devastating instruction — stop eating, start consuming your own reserves. The body obeys.
The team's solution was to silence the receiver. Using antisense oligonucleotides — RNA-based molecules that selectively switch off gene expression — they blocked GFRAL production before the protein could ever find its target. Without the receptor, the cancer's wasting signal has nowhere to land.
Tested in mice already deep into advanced cachexia, the results were striking: treated animals recovered substantial muscle and fat mass, restored deteriorating metabolic function, and by day 50, 90 percent were still alive. In the untreated group, only 20 percent survived. Crucially, the therapy worked even after the disease had significantly progressed — mirroring the real-world moment when patients typically seek help.
Published in Cell Reports Medicine on July 27, the research emerges from KAIST's Graduate School of Medical Science and Engineering and faculty startup THOR Therapeutics. Professor Shong has outlined a clear path forward: accelerated preclinical work and manufacturing development, with human clinical trials targeted for 2030. The ambition is to deploy this as an adjuvant therapy — used alongside existing cancer treatments to preserve the body, sustain treatment tolerance, and improve survival. For millions of patients each year, it is the first real signal that wasting need not be cancer's inevitable companion.
Cancer cachexia is a quiet killer. It steals from patients even when they eat enough—a relentless wasting that affects between half and four-fifths of all people diagnosed with cancer. The body breaks down muscle and fat on its own schedule, indifferent to nutrition. Chemotherapy becomes harder to tolerate. Survival rates drop. Until now, doctors could only offer appetite stimulants, which addressed the symptom but left the underlying machinery of destruction untouched.
A team at KAIST, South Korea's advanced science institute, has identified where that machinery lives: in the brain. Researchers led by Professor Minho Shong and Professor Jinkuk Kim discovered that cancer cells produce a protein called GDF15 in large quantities as the disease progresses. This protein travels to the brainstem, where it binds to a receptor called GFRAL. Once bound, it sends a signal that tells the body to stop eating and start consuming its own muscle and fat reserves. The body obeys. The patient declines.
The solution, the team realized, was to silence the receiver. They developed a therapy using antisense oligonucleotides—RNA-based molecules that can selectively turn off the gene responsible for producing GFRAL. Without the receptor, the cancer's signal cannot land. The wasting signal has nowhere to go.
The researchers tested this approach in mice that already had advanced cachexia. The results were striking. Treated animals showed substantial recovery of muscle and fat mass. Their metabolic function, which had deteriorated, began to restore itself. By day 50 of the study, 90 percent of the treated mice were still alive. In the untreated group, only 20 percent survived. The therapy worked even after the disease had progressed significantly—a crucial finding, since patients typically seek treatment after symptoms have already taken hold.
What makes this different from existing drugs is the target. Previous treatments tried to make patients hungry again. This therapy addresses the root cause: the signal itself. By blocking GFRAL production at the RNA level—the intermediate step between genetic code and protein—the researchers shut down the cascade before it can begin. The body no longer receives the instruction to waste away.
The team, which includes researchers from KAIST's Graduate School of Medical Science and Engineering and the faculty startup THOR Therapeutics, published their findings in Cell Reports Medicine on July 27. Professor Shong outlined an ambitious timeline: preclinical research and manufacturing systems will advance without delay, with human clinical trials targeted to begin by 2030. The goal is to position this therapy as an adjuvant treatment—something used alongside existing cancer drugs to preserve muscle, maintain metabolic function, and ultimately improve both quality of life and survival.
For the millions of cancer patients who experience cachexia each year, this represents a shift in how the disease might be managed. Rather than accepting wasting as an inevitable part of cancer's progression, doctors may soon have a tool that addresses the signal driving it. The brain pathway that cancer exploits may finally have a way to be blocked.
Notable Quotes
This study is significant in that it directly targeted a key receptor in the brainstem at the RNA level to suppress the root cause of cancer cachexia, rather than only temporarily boosting appetite.— Professor Minho Shong, KAIST