Ten minutes of cycling changed 1,300 genes—but that doesn't mean exercise cures cancer
Headlines claimed a breakthrough. The science tells a stranger, more interesting story.
The man posting on Reddit had been given no viable treatment options. Lung cancer, diagnosed in 2014. Surgery in 2016 that failed to remove the tumours. By early 2017, growths had spread to all five lobes of his lungs. No chemotherapy would work. No immunotherapy either.
So he started running.
First came 5K races. Then he taught himself to swim. By 2019, he was competing in sprint triathlons—a man in his sixties with terminal cancer, racing strangers half his age.
"Tumours started shrinking," he wrote. "Oncologist said it was my crazy sense of humour."
His story surfaced in a discussion about new research from Newcastle University, where scientists had discovered something peculiar. When they took blood from people who had just exercised intensely for ten minutes and dripped it onto bowel cancer cells in a laboratory dish, more than 1,300 genes changed their behaviour. Genes that drive aggressive tumour growth went quiet. Genes involved in repairing damaged DNA woke up.
The BBC headline announced that short workouts "can halt bowel cancer." But the same week, researchers in Virginia revealed that ultramarathon runners—people who had completed at least five marathons or two ultramarathons—were developing precancerous colon growths at rates perhaps ten times higher than expected for their age.
One study suggested exercise might save you. The other hinted it might harm you. Both made headlines. Neither told the whole story.
The blood that fights cancer
What Dr Sam Orange and his team at Newcastle actually discovered deserves attention, even if the headlines mangled it. They recruited thirty overweight men, had them cycle hard for roughly ten minutes, then drew blood samples before and after. The post-exercise blood contained elevated levels of thirteen proteins, including interleukin-6—a molecule that plays complex roles in inflammation, metabolism and tissue repair.
Here is where the experiment grew interesting. The researchers bathed bowel cancer cells in serum extracted from this blood. The cells responded. Genes supporting mitochondrial energy production switched on, making the cells metabolise oxygen more efficiently. Genes linked to rapid, uncontrolled proliferation switched off. Most striking: the blood activated a gene called PNKP, which repairs broken DNA. Cancer, at its root, is a disease of accumulating genetic damage. Anything that helps cells fix their broken code matters.
"Exercise doesn't just benefit healthy tissues," Orange observed. "It sends powerful signals through the bloodstream that can directly influence thousands of genes in cancer cells."
The finding opens a window onto something researchers have long suspected but struggled to explain mechanically. We have known for decades that physically active people develop bowel cancer at rates roughly twenty per cent lower than sedentary people. Large studies, thousands of participants, consistent results across populations. But why? What does movement do inside the body that might hold malignancy at bay?
Orange's work offers a hypothesis: exercise changes the composition of blood in ways that make the internal environment less hospitable to cancer. Not through any single molecule, but through a coordinated shift in the body's signalling—a kind of systemic message that reaches every tissue the blood touches.
The chasm between petri dish and patient
The trouble is that cancer cells in a dish are not cancer in a person. They lack immune system surveillance. They lack the tangled architecture of a real tumour, with its blood vessels and support cells and chemical gradients. They lack the thousand feedback loops that make living biology so maddening to predict.
Research on laboratory cancer models has a dismal translation record. Perhaps ninety per cent of drugs that show promise in preclinical testing fail when tried in actual patients. One review in the journal Cells noted bluntly that cancer models rank among "the least predictive" in all of medical research.
This does not make Newcastle's findings worthless. Mechanistic studies like this generate hypotheses worth testing. They identify pathways that might matter. They give future researchers targets to pursue. But they do not—cannot—prove that a ten-minute bike ride prevents or treats cancer in living humans.
The BBC headline made no such distinction. Nor did most other coverage. The Reddit discussion that followed proved more scientifically literate than the journalism. "This is terrible science reporting," wrote one commenter, noting the tiny sample size and the gulf between in vitro observation and clinical claim. "Didn't expect such misleading sensationalism from the BBC."
When anonymous internet commenters outperform professional science journalists in critical reading, something has gone wrong.
When the fittest get sick
Timothy Cannon, an oncologist in Virginia, started noticing something disturbing around 2022. Young ultramarathon runners kept showing up in his clinic. Fit. Lean. Exceptionally active. And dying of advanced colon cancer in their thirties and forties.
Three cases in quick succession prompted him to investigate. He recruited one hundred marathon and ultramarathon runners between thirty-five and fifty—all apparently healthy, none with known genetic predispositions—and gave them colonoscopies.
The results, presented at the American Society of Clinical Oncology meeting in 2025, unsettled him. Forty-one per cent had at least one adenoma. Fifteen per cent had advanced adenomas—precancerous growths with serious probability of progression. For comparison, perhaps one to two per cent of people their age would normally have such lesions.
"I did not imagine that fifteen per cent would have advanced adenomas," Cannon said.
His hypothesis centres on something runners know well: the phenomenon called runner's colitis. During prolonged intense exercise, blood diverts from the gut to working muscles. The intestines experience repeated cycles of oxygen deprivation, then re-oxygenation—a kind of physiological whiplash that causes inflammation and, often, bleeding. Distance runners routinely dismiss bloody stools after long runs as normal. Cannon suspects this repeated intestinal stress might, in susceptible individuals, promote exactly the kind of cellular damage that leads to cancer.
The study has not been peer-reviewed. Critics note the small sample size and the absence of a matched control group. David Rubin, chief of gastroenterology at the University of Chicago, pointed out that the causation might run backwards—perhaps people take up extreme running because someone they loved had cancer, and family history drives both the exercise and the elevated risk.
But even sceptics acknowledge something real in the data. "It tells us there's a signal here," said David Lieberman of Oregon Health and Science University. "We wouldn't have expected these rates of high-risk adenomas in an age group like this."
The shape of the curve
These two studies appear to contradict each other. Ten-minute workouts protect against bowel cancer. Hundred-mile runs promote it. How can both be true?
The answer lies in a concept toxicologists call hormesis—a biphasic dose-response relationship where small amounts of a stressor produce benefits while large amounts cause harm. The principle echoes throughout biology. A glass of wine relaxes; a bottle daily destroys the liver. Brief fasting triggers cellular cleanup; prolonged starvation kills. Moderate sun exposure generates vitamin D; excessive exposure generates melanoma.
Exercise traces the same curve. At one extreme sits the couch—associated with elevated cancer risk, heart disease, metabolic dysfunction. As activity increases, health improves, reaching an optimal zone of moderate regular movement. But keep pushing rightward, into territory where the body experiences repeated extreme stress without adequate recovery, and the relationship may invert.
"Physical inactivity or strenuous exercise bouts increase the risk of infection," notes research in Ageing Research Reviews, "while moderate exercise up-regulates the immune system."
Newcastle examined the left side of this curve: what happens when sedentary people start moving. Virginia examined the right side: what happens when extreme athletes push beyond their bodies' adaptive capacity. The studies do not conflict. They illuminate different regions of the same underlying biology.
The Reddit commenter whose tumours shrank found himself somewhere in the middle—pushing hard enough to trigger beneficial adaptations, but not so hard as to overwhelm his body's repair mechanisms. His oncologist attributed the improvement to humour. The more likely explanation involves the molecular signalling that Orange's team observed in their petri dishes, playing out inside a living body over months and years.
The mystery we haven't solved
All of this unfolds against a backdrop of genuine alarm. Bowel cancer rates have nearly doubled in adults under fifty since the early 1990s, even as they have declined in older populations. Twenty per cent of diagnoses now occur in patients under fifty-five—double the rate from 1995. Colorectal cancer has become the leading cause of cancer death in American men under fifty. Many of those affected have no family history, no genetic syndrome, no obvious explanation.
"We don't understand a lot about the causes, the biology, or how to prevent early onset of the disease," admitted Phil Daschner of the National Cancer Institute.
Diet, obesity, microbiome disruption, childhood antibiotics, environmental chemicals—researchers have investigated them all without finding answers. The Newcastle study offers one small piece of mechanistic understanding. The Virginia study offers another. Neither solves the puzzle. Both add texture to a picture that remains frustratingly incomplete.
What should a person actually do with this information?
The evidence supporting moderate regular exercise remains among the strongest in preventive medicine. Move your body most days. Get your heart rate up. You do not need to run ultramarathons; you probably should not. The optimal dose appears to be somewhere in the middle of the curve—enough stress to trigger adaptation, not so much that adaptation fails.
The Newcastle study tells us something interesting about why this might work. It does not tell us anything we needed to know before lacing up our trainers and heading out the door. The headlines claimed a breakthrough. The science, as usual, delivered something more modest and more honest: a hypothesis worth testing, a mechanism worth exploring, another small step in the long work of understanding how our bodies protect themselves from disease.
That man on Reddit, the one with tumours in all five lobes of his lungs, is still racing. His story proves nothing that would satisfy a statistician. But it captures something the studies, taken together, also suggest: that somewhere between doing nothing and doing everything lies a zone where exercise becomes medicine—and that finding it may be less about following headlines than about listening to the body that carries us through each ordinary, extraordinary day.