Abstract Aims We examined the efficacy of exercise training for improving physical functioning and cardiopulmonary fitness in survivors of paediatric brain tumours (BTs) treated with cranial irradiation. Methods We conducted a controlled clinical trial with crossover of exercise training versus no training in the community in either a group or combined group/home setting. A volunteer sample of 28 children treated with cranial irradiation for brain tumours completed training (mean age = 11.53 years; mean time since diagnosis = 5.25 years). end-points were physical functioning assessed by four subtests from the Bruininks–Oseretsky Test of motor performance (BOT-2) and pro-rated work rate from a cycle ergometer. Linear mixed modelling was used to evaluate time, training, training setting, and carryover effects. Results Adherence to training was 84%. Performance on the BOT-2 was below average for all assessments. However, training resulted in improvement in bilateral coordination (F (1, 30) = 6.59, p = 0.02), irrespective of training setting and improved performance was maintained even approximately 12°weeks after training had ended (F (1, 24) = 9.60, p = 0.005). Training resulted in increased pro-rated work rate for participants in the group training setting only (F (1, 25) = 4.57, p = 0.04) and these participants maintained their improved work rate approximately 12°weeks after training had ended (F (1, 20) = 8.38, p = 0.01). Conclusion Exercise training improves physical functioning and fitness in paediatric BT survivors. Exercise interventions that ameliorate adverse physical effects and promote health in long-term survivors are highly recommended in this vulnerable population. ( ClinicalTrials.gov , NCT01944761 ).