This is an interesting question as it goes to the question as to whether too much exercise causes health problems. The problems could be accelerated ageing.
https://www.sciencedirect.com/science/article/pii/S1877782126001037
chatGPT:
Paper
Swain et al., “Advanced adenomas among young endurance runners: A prospective hypothesis-generating prevalence study,” Cancer Epidemiology, 2026.
Summary
This paper reports a prospective, single-centre, cross-sectional prevalence study of colorectal neoplasia in young endurance runners aged 35–50. Participants were eligible if they had completed at least two ultramarathons of 50 km or longer or at least five marathons, and had not had a colonoscopy in the previous 10 years. The main aim was to estimate the prevalence of advanced adenomas, not to prove causation.
The biological rationale is that extreme endurance running can cause repeated gastrointestinal stress: splanchnic hypoperfusion, mucosal injury, ischemia-reperfusion, post-exercise bleeding, inflammation, oxidative stress, and compensatory epithelial proliferation. The authors suggest that repeated cycles of injury and repair could plausibly increase adenoma formation or progression, while acknowledging that ordinary physical activity is generally associated with lower colorectal cancer risk.
The study enrolled 100 participants, of whom 94 underwent colonoscopy and were included in the analysis. The median age was about 42, 54.3% were female, and most were White/Caucasian.
The main findings were striking:
| Finding | Result |
|---|---|
| Participants | 94 |
| Any adenoma | 39/94, 41.5% |
| Advanced adenoma | 14/94, 15.0% |
| Advanced adenoma in runners without post-exercise rectal bleeding | 6/68, 8.8% |
| Colorectal cancers | 0 |
| Post-exercise rectal bleeding | 26/94, 27.7% |
The paper compares the 15.0% advanced adenoma prevalence with a historical screening benchmark of 1.2% in asymptomatic average-risk adults aged 40–49, but repeatedly stresses that this comparison is imperfect because this runner cohort was not a matched control group and included people with post-exercise bleeding.
The lesions were mostly right-sided, especially in the ascending colon. Advanced adenomas were defined by standard criteria: size at least 10 mm, villous/tubulovillous histology, or high-grade dysplasia. Most advanced lesions were classified by size, tubulovillous features were also common, and high-grade dysplasia was rare.
The strongest clinical signal was the association with post-exercise rectal bleeding. Among those with advanced adenomas, 57.1% reported bleeding after endurance exercise, compared with 16.0% in the non-advanced adenoma group and 25.5% in those with no adenomas. Advanced adenomas were found in 30.8% of runners reporting post-exercise bleeding versus 8.8% of those not reporting bleeding.
The authors’ practical conclusion is not that endurance running has been proven to cause colorectal neoplasia. Rather, they argue that post-exercise rectal bleeding in endurance athletes should not simply be dismissed as “normal runner’s bleeding”, and that a larger controlled study is needed.
Novelty
The main novelty is that this appears to be the first prospectively enrolled prevalence study specifically assessing colorectal cancer and advanced adenomas in extreme endurance athletes. Previous literature had discussed runner’s colitis, occult gastrointestinal bleeding, ischemic colitis, microbiome changes, and occasional case reports, but not a prospective colonoscopy-based prevalence estimate in this selected young endurance-runner population.
A second novel feature is the focus on advanced adenomas as precursor lesions, rather than waiting for colorectal cancer events. That is useful because cancer incidence in this age group is low, so a cancer-endpoint study would need far larger numbers and longer follow-up.
A third useful contribution is the separation of runners with and without post-exercise rectal bleeding. Even among those denying bleeding, the advanced adenoma prevalence was 8.8%, which keeps the signal alive even after partially addressing the concern that symptomatic runners were driving the result.
The study is also interesting because the lesions were predominantly right-sided, whereas the broader rise in early-onset colorectal cancer has often been described as involving more distal and rectal disease. That anatomical difference may suggest a distinct biology or may simply reflect small-sample noise.
Critique
The study is important but should be treated as signal-generating, not definitive.
The biggest limitation is the lack of a contemporaneous matched control group. Comparing 94 self-selected endurance runners with historical screening cohorts is weak because the groups may differ in age distribution, symptoms, family history, colonoscopy indication, diet, medications, metabolic health, and willingness to undergo colonoscopy. The authors acknowledge this directly.
Selection bias is a major concern. Recruitment was through endurance events, social media, and race invitations; participants received a no-cost colonoscopy and compensation. People with bleeding, anxiety, family history, gastrointestinal symptoms, or concern about colorectal cancer may have been more likely to volunteer. This could substantially inflate the apparent prevalence.
The paper’s central mechanistic idea—recurrent ischemia-reperfusion causing epithelial injury, inflammation, repair, and mutagenesis—is plausible, but not directly tested. The study did not measure mucosal ischemia, inflammatory biomarkers, oxidative damage, microbiome changes, immune surveillance, epithelial proliferation, or mutation burden. Therefore, the mechanistic part remains speculative.
Important confounders were not adequately captured. The authors state that they did not collect detailed family history of colorectal cancer, metabolic comorbidities, medication exposures, detailed diet, objective fitness measures such as VO₂max, menstrual-cycle or hormonal-contraceptive variables, or timing of the last marathon/ultramarathon relative to colonoscopy. These omissions matter because each could affect bleeding, inflammation, adenoma risk, or exposure classification.
There is also possible detection bias. Endoscopists knew participants were in an endurance-runner study, and colonoscopy quality metrics were incompletely documented: explicit withdrawal time was recorded in only 10 reports. Pathologists were also aware of study participation. These factors could increase lesion detection or influence classification, although advanced adenoma classification is partly objective.
The finding that training-volume variables showed no clear pattern weakens a simple dose-response interpretation. If extreme running were causally driving adenoma risk, one might expect more adenomas or more advanced adenomas in those with higher lifetime running volume, more races, or more weekly mileage. The study may simply be too small or too crudely measured to detect that, but absence of a dose-response is a caution.
Finally, the result should not be overgeneralised to ordinary exercise. The paper itself emphasises that physical activity is generally associated with lower colon cancer risk, and that this study concerns a highly selected group exposed to repeated extreme gastrointestinal stress.
Bottom line
This is a provocative and clinically useful hypothesis-generating paper. Its main contribution is the observation that advanced adenomas were found in 15% of a young endurance-runner cohort, with a particularly high rate among those reporting post-exercise rectal bleeding. The paper does not prove that endurance running causes adenomas or colorectal cancer, but it does support a practical message: rectal bleeding after endurance exercise should not automatically be dismissed as benign runner’s colitis. Larger controlled studies with matched non-athlete comparators, proper colonoscopy quality metrics, family-history data, dietary and medication data, and mechanistic biomarkers are needed.