Background: Individuals affected by childhood cancer can have cognitive dysfunction that persists into adulthood and negatively affects quality of life. In this study, we aimed to evaluate the effects of physical activity and exercise on cognitive function among individuals affected by childhood cancer.
Methods: In this systematic review and meta-analysis, we searched seven databases (CINAHL Plus, Cochrane Library, Embase, MEDLINE, PsycINFO, SPORTDiscus, and Web of Science) and two clinical trial registries (ClinicalTrials.gov and the International Clinical Trials Registry Platform) for randomised controlled trials (RCTs) and non-randomised studies of interventions (NRSIs) published (or registered) from database inception to Jan 30, 2022, with no language restrictions. We included studies that compared the effects of physical activity or exercise interventions with controls (no intervention or usual care) on cognitive function among individuals diagnosed with any type of cancer at age 0-19 years. Two reviewers (JDKB and FR) independently screened records for eligibility and searched references of the selected studies; extracted study-level data from published reports; and assessed study risk of bias of RCTs and NRSIs using the Cochrane risk of bias tool for randomised trials (RoB 2) and Risk Of Bias In Non-randomised Studies-of Interventions (ROBINS-I) tools, certainty of the evidence using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) approach, and any adverse events. We used intention-to-treat data and unpublished data if available. Cognitive function was assessed by standardised cognitive performance measures (primary outcome) and by validated patient-reported measures (secondary outcome). A random-effects meta-analysis model using the inverse-variance and Hartung-Knapp methods was used to calculate pooled estimates (Hedges' g) and 95% CI values. We estimated the heterogeneity variance by the restricted maximum likelihood method and calculated I2 values to measure heterogeneity. We examined funnel plots and used Egger's regression test to assess for publication bias. This study is registered with PROSPERO, CRD42021261061.
Findings: We screened 12 425 titles and abstracts, which resulted in full-text assessment of 131 potentially relevant reports. We evaluated 22 unique studies (16 RCTs and six NRSIs) with data on 1277 individuals affected by childhood cancer and low-to-moderate risk of bias. Of the 1277 individuals, 674 [52·8%] were male and 603 [47·2%] were female; median age at study start was 12 (IQR 11-14) years, median time since the end of cancer treatment was 2·5 (IQR -1·1 to 3·0) years, and median intervention period was 12 [IQR 10-24] weeks. There was moderate-quality evidence that, compared with control, physical activity and exercise improved cognitive performance measures (five RCTs; Hedges' g 0·40 [95% CI 0·07-0·73], p=0·027; I2=18%) and patient-reported measures of cognitive function (13 RCTs; Hedges' g 0·26 [0·09-0·43], p=0·0070; I2=40%). No evidence of publication bias was found. Nine mild adverse events were reported.
Interpretation: There is moderate-certainty evidence that physical activity and exercise improves cognitive function among individuals affected by childhood cancer, which supports the use of physical activity for managing cancer-related cognitive impairment.
Funding: Research Impact Fund of Research Grants Council of the Hong Kong University Grants Committee (R7024-20) and Seed Fund for Basic Research of the University of Hong Kong.
Copyright: © 2022 Published by Elsevier Ltd. All rights reserved.
Copyright © 2023 Elsevier Ltd. All rights reserved.