Can exercise lower your cancer risk? For most cancers, the evidence points one way

Being more physically active is associated with a lower risk of many, though not all, cancers — and for at least one, melanoma, the association runs the other way. That is the honest shape of the observational evidence. Separately, a 2025 randomised trial has now shown that structured exercise after treatment for colon cancer improves survival, which is a stronger kind of evidence than the risk associations can offer, but it answers a different question: what exercise does after a diagnosis, not whether it prevents one.

What the large pooled study found

The most detailed look at activity and cancer incidence pooled 12 prospective US and European cohorts into a single analysis of 1.44 million participants, among whom 186,932 cancers were recorded [s1]. Researchers compared people at the 90th percentile of leisure-time physical activity with those at the 10th, and estimated hazard ratios for 26 cancer types [s1].

High versus low activity was associated with a lower risk of 13 cancers [s1]. The largest associations were for oesophageal adenocarcinoma (hazard ratio 0.58), liver (0.73), lung (0.74) and kidney (0.77) [s1]. More common cancers showed smaller but still significant associations: colon (0.84), rectal (0.87), bladder (0.87) and breast (0.90) [s1]. Endometrial cancer sat at 0.79 [s1].

Two cancers went the opposite way. Activity was associated with a higher risk of malignant melanoma (hazard ratio 1.27) and a slightly higher risk of prostate cancer (1.05) [s1]. The melanoma finding almost certainly reflects sun exposure during outdoor activity rather than exercise itself — a reminder that these are associations from observational data, not proof of cause.

Why the associations are believable but not proof

The study's strength is its size and its handling of two obvious confounders. Because active people tend to be leaner, and leanness independently lowers the risk of several cancers, the researchers re-ran the analysis adjusting for body-mass index [s1]. Adjustment modestly weakened the associations, but 10 of the 13 inverse associations remained statistically significant [s1]. Associations were generally similar in overweight or obese and normal-weight participants, and — apart from lung cancer — similar regardless of smoking history [s1].

That consistency is what makes the direction of the finding credible. What it cannot do is prove that activity itself lowers risk. People who exercise differ from people who do not in diet, alcohol, income and health-seeking behaviour, and no observational study fully removes those differences. The authors frame their result as supporting broad generalisability, not as a licence to promise prevention [s1].

The trial that raises the evidence bar

For years the exercise-and-cancer field lacked what it most needed: a randomised trial. The CHALLENGE trial, published in 2025, provides one, though for survivors rather than for prevention [s2]. It randomly assigned 889 patients with resected colon cancer who had finished adjuvant chemotherapy either to a three-year structured exercise programme or to health-education materials alone [s2].

At a median follow-up of 7.9 years, disease-free survival was significantly longer in the exercise group, with a hazard ratio for recurrence, new cancer or death of 0.72 (95% confidence interval 0.55 to 0.94) [s2]. Five-year disease-free survival was 80.3% with exercise versus 73.9% with health education, a difference of 6.4 percentage points [s2]. Overall survival also favoured exercise (hazard ratio for death 0.63), with eight-year survival of 90.3% versus 83.2% [s2]. The programme was not free of downside: musculoskeletal adverse events occurred in 18.5% of the exercise group versus 11.5% [s2].

How to read the two together

The distinction matters. The pooled cohort speaks to primary prevention — whether active people get fewer cancers — and finds a consistent but non-randomised association that holds for most cancer types and reverses for melanoma [s1]. The CHALLENGE trial speaks to what happens after a colon-cancer diagnosis, and there the evidence is now level-1: a structured programme lengthened survival [s2]. One is a strong association; the other is cause and effect, in a specific group.

Neither result tells an individual reader what to do, and this article does not. What they jointly support is modest and consistent: physical activity tracks with lower risk across a wide range of cancers, the melanoma exception is real and probably about sun rather than movement, and for people already treated for colon cancer, structured exercise now has trial evidence behind it. The number of servings, minutes or sessions that any one person should aim for is a clinical question, not one these studies answer.

Sources

• [s1] Association of Leisure-Time Physical Activity With Risk of 26 Types of Cancer in 1.44 Million Adults — JAMA Internal Medicine, 1 June 2016. https://doi.org/10.1001/jamainternmed.2016.1548
• [s2] Structured Exercise after Adjuvant Chemotherapy for Colon Cancer — New England Journal of Medicine, 3 July 2025. https://doi.org/10.1056/NEJMoa2502760