Effects of non-pharmacological interventions in patients with cancer-related cognitive impairment: A meta-analysis

Cancer-related cognitive impairment is a significant public-health issue, and non-pharmacological therapies are widely used as primary treatments. However, the comparative effectiveness of these interventions remains uncertain.

We included randomized controlled trials of non-pharmacological interventions for cancer-related cognitive impairment identified through searches of PubMed, Web of Science, EMBASE, the Cochrane Central Register of Controlled Clinical Trials, Chinese National Knowledge Infrastructure, and SinoMed through February 2025. Extracted data comprised study details, participant characteristics, intervention types, and outcomes. Treatment effects were quantified as standardized mean difference with 95 % confidence interval and pooled using random-effects models. We analyzed both subjective and objective cognitive outcomes, including attention, executive function, memory, processing speed, and language. The protocol was registered on PROSPERO (CRD 42023457681).

A total of 60 eligible studies involving 5129 patients were identified. Based on intervention characteristics, these studies were categorized into six categories: cognitive behavioral therapy-based interventions, cognitive training, physical activity, mindfulness-based interventions, multi-modal management, and acupuncture. Significant improvements in subjective cognition were associated with cognitive behavioral therapy-based interventions (7 comparisons, standardized mean difference: 0.53, 95 % confidence interval: 0.22 to 0.84, I² = 58 %), physical activity (8 comparisons, standardized mean difference: 0.86, 95 % confidence interval: 0.39 to 1.14, I² = 81 %), and mindfulness-based interventions (12 comparisons, standardized mean difference: 0.43, 95 % confidence interval: 0.15 to 0.72, I² = 66 %). Regarding objective cognition, significant improvements were observed for cognitive behavioral therapy-based interventions (1 comparison, standardized mean difference: 0.73, 95 % confidence interval: 0.18 to 1.28), cognitive training (7 comparisons, standardized mean difference: 1.23, 95 % confidence interval: 0.61 to 1.85, I² = 93 %), mindfulness-based interventions (1 comparison, standardized mean difference: 0.48, 95 % confidence interval: 0.01 to 0.96), and acupuncture (3 comparisons, standardized mean difference: 1.19, 95 % confidence interval: 0.06 to 2.32, I² = 95 %).

Current randomized evidence indicates that several non-pharmacological interventions can improve subjective and/or objective cognition in adults with cancer-related cognitive impairment. However, substantial between-study heterogeneity and generally low certainty of evidence limit confidence in these estimates. Clinicians should tailor intervention selection to individual patient needs and the clinical context. Moreover, further, better-powered, well-reported and standardized trials are needed.