Systematic review finds supervised exercise improves strength during childhood cancer treatment

A systematic review of nutrition and physical activity interventions delivered to children and young people receiving cancer treatment found the strongest evidence for supervised physical activity. Across the studies reporting exercise capacity, 15 of 17 interventions favoured the intervention, while a positive effect direction was recorded for 11 of 12 interventions measuring muscle strength. Physical activity participation, fatigue and quality of life also generally moved in a favourable direction. Evidence for nutrition interventions was more limited and inconsistent, suggesting that improving diet quality during treatment remains an important research gap.

The review, published in Supportive Care in Cancer, included 77 articles reporting 66 original cohorts. Four studies evaluated nutrition alone, 68 evaluated physical activity alone, and four combined nutrition and physical activity; the combined interventions were described across five articles. The searches covered MEDLINE, Embase, SCOPUS and ProQuest Health and Medical Sciences from database inception through 22 March 2024, producing 8,349 records before screening. The protocol was registered on PROSPERO as CRD42024513364, and the review followed PRISMA guidelines. Investigators assessed risk of bias with the Cochrane ROB-1 tool for randomised controlled trials, the Newcastle-Ottawa Scale for non-randomised studies and an NIH tool for uncontrolled studies.

Eligible participants were children and young people aged 0 to 21 years, with at least 75 per cent of each study population diagnosed with childhood cancer. Interventions had to begin between diagnosis and completion of treatment, or before the day of a haematopoietic stem cell transplant. Diagnoses included acute lymphoblastic leukaemia, bone tumours, brain tumours and mixed cancer types. Among 62 physical activity interventions, 17 enrolled only children with acute lymphoblastic leukaemia, 37 included mixed diagnoses, three concerned bone tumours and one included brain tumours. Most intervention groups contained 10 to 50 participants, while eight studies enrolled fewer than 10. These small samples and wide differences in cancer type, treatment stage, programme design and measurement limited the strength and comparability of the conclusions.

Exercise capacity produced one of the clearest signals in the review. The direction of effect favoured physical activity interventions in 15 of 17 studies reporting this outcome, and the pattern was greater than expected by chance in the review’s sign test, with a p-value of 0.0023. Functional capacity measured with the six-minute walk test improved in two of five cohort studies: one followed six weeks of advanced robotic gait training, and the other followed eight weeks of supervised physiotherapy added to independent or parent-supported home exercises. Three of five interventions reporting cardiorespiratory fitness showed increased maximal oxygen uptake. These programmes included robotic gait training, aerobic activity delivered through video games, and combined social support with supervised aerobic, resistance and balance exercise during hospital admission.

Broader tests of physical function produced a less uniform picture. Results favoured the intervention in 12 of 16 studies, with another five showing an unclear effect direction. Four of five randomised controlled trials reported some improvement, although two found no statistically significant difference in Timed-Up-and-Go results. Those two trials used aerobic and resistance training: one provided supervised sessions three times weekly for 19 weeks, while the other combined five supervised sessions over four months with home-based exercise on three to five days each week. A non-randomised study likewise found no difference in Timed-Up-and-Go after a six-month programme delivered during hospital admissions. Treatment-related neuropathy, fatigue, inactivity and other impairments may affect several components of functional testing, making whole-body function harder to improve than isolated strength or endurance measures.

Muscle strength showed a more consistent pattern than some physical-function tests. A positive effect direction was reported in 11 of 12 interventions measuring strength, with a sign-test p-value of 0.0063. Improvements occurred in lateral-row, bench-press and leg-press repetition maximum after exercise interventions. Effects on hand-grip strength and knee extension were mixed, and several studies reported strength outcomes without blinded assessment. Every intervention associated with a statistically significant strength increase included supervised exercise sessions. The investigators considered supervision potentially useful for maintaining participation, helping children adapt activity to symptoms and supporting performance at an appropriate level, but the review could not isolate supervision as the reason for the observed improvements.

Physical activity participation, fatigue and quality of life generally showed favourable patterns, although the number and design of studies varied. Activity increased in three cohort studies, two randomised trials and one non-randomised study. The direction of effect favoured intervention in 13 of 15 studies reporting physical activity, with a sign-test p-value of 0.0074. Fatigue results likewise favoured the intervention in 13 of 15 studies, with the same p-value. Overall quality-of-life scores also leaned toward benefit, with a sign-test p-value of 0.0001. Statistically significant quality-of-life differences were reported in three randomised trials and one non-randomised study, although not every randomised trial found such a difference. These results describe direction of effect across heterogeneous studies rather than a single pooled treatment estimate.

Nutrition evidence was thinner. Six of the eight nutrition-focused studies included children with acute lymphoblastic leukaemia, and two involved mixed diagnoses. The interventions ranged from one-to-one counselling with a registered dietitian and educational workshops for children and parents to a self-guided online programme supported by a lifestyle coach. Durations ranged from one session to 12 months. Diet-quality findings were inconsistent: after three months, the mean Healthy Eating Index score increased from 44.5 with a standard deviation of 9.08 to 47.7 with a standard deviation of 7.91 out of 100, but the mean change of 3.14 was not statistically significant, with a 95 per cent confidence interval of −3.55 to 9.82 and a p-value of 0.29. A separate 12-month programme produced a statistically significant increase in the mean Diet Quality Index from 48.53 to 55.75 out of 100, while its Healthy Diet Index score did not change significantly.

Fruit and vegetable intake increased after nutrition education workshops and after dietary counselling during intensive treatment for acute lymphoblastic leukaemia. It did not change substantially after a three-month self-guided online education programme. Two studies reported higher calorie intake, and two found a greater share of energy supplied by protein at different follow-up points. Changes in anthropometry and body composition were generally not statistically significant in either the nutrition or physical activity studies. Across the review, recruitment ranged from 42.3 per cent to 100 per cent for exercise interventions and from 50 per cent to 90.9 per cent for nutrition interventions. Adherence exceeded 75 per cent in 18 of the 42 interventions that reported it, and adverse events were uncommon among studies that assessed them. Inconsistent reporting of adherence and compliance nevertheless prevented estimation of dose-response relationships for frequency, intensity, activity type and duration.

The review identified a major gap in integrated care: only four studies combined nutrition and physical activity, and their nutrition and activity outcomes were generally reported separately. The authors note that diet and activity are physiologically and behaviourally interconnected. Adequate energy and protein intake support growth, tissue repair and preservation of lean mass, while activity provides a stimulus for maintaining muscle, strength and cardiorespiratory fitness. Fatigue and functional decline can reduce participation in activity, while inactivity can worsen deconditioning and further limit movement. The evidence therefore does not establish that any particular programme is superior, nor does it prove benefits for every child. Instead, it identifies supervised physical activity as the most consistently supported intervention and nutrition counselling and education as feasible approaches whose effectiveness requires further evaluation in larger, more comparable studies.

For future trials, the review calls for standardised outcomes that extend beyond body weight and body mass index. Nutrition studies should report diet quality, protein adequacy, body composition and treatment-related nutritional toxicities alongside anthropometry. Physical activity studies should combine objective participation measures, such as accelerometry, with exercise capacity, fatigue and quality-of-life outcomes. Consistent documentation of treatment intensity, nutrition-support use, nutrition-impact symptoms, adherence and intervention dose would help clarify which components work, for which patients and under what circumstances. The authors also advocate integrated programmes incorporating healthy dietary behaviour, supervised aerobic and resistance exercise where feasible, behavioural support and family involvement, with enough flexibility to accommodate changing symptoms and treatment demands.

Subject of Research: Effects of nutrition and physical activity interventions in children and young people receiving cancer treatment

Article Title: Nutrition and physical activity interventions delivered to children and young people during cancer treatment: a systematic literature review

Article References: Ross, A. S., Bertrand, É., Seretny, M., Braakhuis, A., van Waart, H., & Lovell, A. L. (2026). Nutrition and physical activity interventions delivered to children and young people during cancer treatment: a systematic literature review. Supportive Care in Cancer, 34(10), Article 978. https://doi.org/10.1007/s00520-026-11194-9

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11194-9

Keywords: childhood cancer, paediatric oncology, physical activity, exercise training, nutrition, muscle strength, exercise capacity, supportive care, systematic review, physical, activity, interventions