Exercise and the oncologic brain: the evidence is no longer anecdotal (network meta-analysis)

Exercise improves cognitive function in cancer patients: network meta-analysis of 27 RCTs and 2,809 patients shows consistent short- and long-term benefit.

Blog · ppj.es

Exercise and the oncologic brain: the evidence is no longer anecdotal (network meta-analysis)

Published on 19 August 2026

🇪🇸 Leer este artículo en español

Title: Exercise and the oncological brain: the evidence is no longer anecdotal (network meta-analysis) Type: B (clinical / supportive care) Tags: exercise, cognition, oncology, rehabilitation, network meta-analysis PMID: 42604722 DOI: 10.1016/j.apmr.2026.07.032 Journal: Archives of Physical Medicine and Rehabilitation (2026) Category: Oncology

⚖️ Transparency notice: this article was written with AI assistance and reviewed by the author, a medical oncologist.

27 | Randomized controlled trials 2,809 | Participants with cancer SMD 0.44 | Subjective cognition, short term SMD 0.27–0.31 | Working memory

The finding

Network meta-analysis of 27 RCTs (2,809 participants) in Arch Phys Med Rehabil (Cheng et al. 2026): exercise-based interventions improve cognitive function in cancer patients, both short-term and long-term [PMID 42604722].

The numbers

Domain | Short term (SMD, 95% CI) | Long term (SMD, 95% CI) Subjective cognitive function | 0.44 (0.27–0.61) | 0.37 (0.16–0.57) Executive function | −0.28 (−0.43 to −0.14) | −0.32 (−0.56 to −0.08) Working memory | 0.27 (0.14–0.41) | 0.31 (0.06–0.55) Processing speed | −0.16 (−0.31 to −0.01) | −0.31 (−0.55 to −0.07)

(Positive SMD = improvement on scales where higher is better; negative in executive function/processing speed = also improvement, because those are measured as time/latency.)

1. The magnitude is modest but consistent. SMD ~0.3-0.4 is a small-to-medium effect — real, but not a miracle. Sufficient to recommend it as part of supportive care.

2. Heterogeneity of modalities. The network meta-analysis compares types of exercise (aerobic, resistance, combined, mind-body), but the RCTs are small and comparability is limited. SUCRA gives ranking, not certainty.

3. Publication bias and self-report. “Subjective function” depends on the patient; objective (executive, memory) is more solid but also with small samples.

4. Broad oncological population (not only breast cancer) — generalizable, but does not say which subgroup benefits most.

It’s another piece of the non-pharmacological supportive care puzzle: exercise helps not only with fatigue and cardio, but also with cognition (post-treatment “brain fog” is real and frequent). Connects with NutriCheck and with the ICI fatigue post: oncological support must prescribe exercise, not just monitor. SUCRA will likely rank combined (aerobic + resistance) at the top — useful for recommending a specific routine.

Reference: Cheng W, et al. Arch Phys Med Rehabil. 2026. doi:10.1016/j.apmr.2026.07.032. PMID: 42604722.

— This analysis was generated by ANGIE (Always Next to Guide, Inspire and Empower), an artificial intelligence system with SOUL profiles, designed by Dr. Javier Pumares Pérez.

Connect on LinkedIn · ppj.es

Disclaimer: this article is educational and informational in nature and reflects the personal opinion of the author. It does not constitute medical advice nor replace the assessment of a healthcare professional. If you have a health concern, consult your physician.