Original title: Exploring proteins as intermediates between physical activity and cancer in UK biobank
Lucy Goudswaard, Matthew Lee, Lisa Hobson, Rebecca Richmond +6 · September 2026
Being physically active lowers the risk of getting several cancers, but scientists do not fully understand why. Researchers wanted to see if blood proteins act as biological messengers that carry out this protective effect. They studied activity tracking data, blood protein measurements, and cancer records from thousands of people. They checked whether exercise changes specific protein levels and whether those changed proteins then alter cancer risk. They found that physical activity was observationally linked to lower rates of breast, colorectal, and endometrial cancer, and gene-based tests showed exercise causally protects against breast, colorectal, and prostate cancer. Exercise was tied to 436 blood proteins, and 44 of those were also linked to cancer risk. However, deeper genetic tests showed that these proteins, including one named TNFSRF13B, largely do not drive these protective effects. Instead, the proteins mainly serve as markers showing that someone is active.
Imagine you are picking a lock with a big keyring. Jogging shakes the keyring and rattles dozens of keys. One of those keys might open the lock, but you cannot tell by the rattling alone. In this paper, researchers checked whether the keys that rattle when you jog are the ones that open the cancer lock. They found that even though many keys rattle, none of them are the right key for the cancer lock.
Exercise clearly lowers the risk of several cancers, but we don't know why. This study tested whether proteins in the blood are the missing link. It found that while exercise alters hundreds of proteins, those changes don't seem to cause the cancer protection. That means we need to look beyond these common protein markers to understand how exercise works, which matters for designing better prevention strategies.
Observational associations between physical activity and cancer were attenuated after adjusting for body mass index, suggesting possible confounding. MR analysis for a key protein, TNFSRF13B, showed conflicting evidence for its effect on breast cancer. Overall, the study did not find strong evidence that any circulating protein serves as a causal intermediate, though this may reflect limited statistical power or the specific proteins measured. The analysis is limited to UK Biobank and four cancer types.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Exploring proteins as intermediates between physical activity and cancer in UK biobank — Lucy Goudswaard, Matthew Lee, Lisa Hobson et al. — medRxiv — doi:10.64898/2026.09.10.26362738
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.