Discussion
In this systematic review 28 articles were identified investigating the potential impact of an intervention on cognitive functioning in pediatric brain tumor patients. The included articles described a variety of lifestyle, cognitive, and pharmacological possibilities.
First, exercise training had a beneficial impact on processing speed and long-term memory, and on anxiety and emotional control13,40. However, none of the lifestyle interventions reported on effect size. The positive effects of a exercise intervention were greater if the activities occurred in a group setting, whereas they were not amplified by adding attentional tasks to the exercise training13,34. For in-group training, it is unclear whether the social stimulation caused additional improvement of neurocognition or rather it had led to more intensive physical training35,40. Overall, the included studies consolidated the idea that exercise training is easily accessible, safe, and that it can be initiated at any moment after diagnosis13,35,40. The benefits for patients with a PBT are likely to be even more evident if exercise training is integrated into a sustainable healthy lifestyle, rather than being used as a short-term intervention34. Incorporating exercise training in a healthy lifestyle may be crucial, not only for its potential neurocognitive benefits but also for many other health-related factors that may be compromised in patients with a PBT2,35,40–42.
In contrast to the low number of physical exercise studies, most of the existing studies focused on cognitive interventions. These approaches either aim to target and improve one specific neurocognitive domain, or they are designed to teach compensatory strategies. Based on the reported effect sizes most promising outcomes were found for Cogmed in the domains of WM, executive functioning and symbolic WM; for CRP in the domains of learning strategies, inattention, and attention and hyperactivity symptoms; and for CRC in the domains of processing speed and visual memory 17,26–28,30,37,38. Other cognitive interventions also showed a positive impact on neurocognition however, their effect sizes were moderate or small, or they were not reported43. Except for neurofeedback training, this showed no effect on neurocognition16. It is important to critically reflect on how cognitive impact is measured after a cognitive intervention. If the intervention efficacy is assessed using cognitive tests that overlap with the content of the training exercises, the training effect can be limited to the specific tool-specific skill, rather than a generalizable effect44,45. Therefore, to increase ecological validity of the intervention results, it is essential to incorporate a more broadened neurocognitive assessment battery as well as patient- or parent-reported outcomes in daily life46.
Regarding pharmacological possibilities, the current literature described two pharmacological compounds for their potential mitigation of NCI, namely MPH and donepezil. MPH showed to improve attention in pediatric cancer patients with attentional problems. Nevertheless, reported effect sizes are moderate to small and one study did not confirm the improvement of attention21. Other domains such as academic achievement and processing speed appeared to improve as well, these results were inconsistent across different studies as well. Disparities in study outcomes may be attributable to different neurocognitive assessment batteries. Furthermore, results may not be generalizable to patients with a PBT that have a NCI in domains other than attention, and long-term safety and duration of effects is still uncertain20,22,47. Donepezil was found to be effective in irradiated patients with a PBT for both performance and proxy-reported measures of executive functioning. However, similar to MPH, there is no indication whether long-term safety can be guaranteed and what neurocognitive effects would be39. Notwithstanding the promising results found for the abovementioned psychostimulants, multiple side effects were reported. By contrast, cognitive interventions did not report specific side effects and only one fall was reported in the lifestyle intervention studies, and therefore provide initial good alternatives with limited potential side effects.
Based on the results of the included studies, it may be concluded that a tendency towards positive effects of neurocognitive interventions on NCI in patients with a PBT exists. Nevertheless, some limitations must be considered. A large methodological variability exists across included studies, decreasing comparability and generalizability of the results to the entire population of patients with a PBT. Also, study populations were relatively small and effect size was not always reported. Measurement of long-term effects in the studies is limited and the impact of the intervention on lifestyle and daily-life functioning should be considered. Additionally, each type of intervention comes with their own advantages and disadvantages. For instance, digital interventions are increasingly accessible, hence they are susceptible for technical and ethical problems, and more parental involvement is required48. Conversely, therapist-delivered interventions are time-, labor- and cost-intensive30,35. Though, in therapist-delivered interventions, a therapist can better estimate how the patient is doing and their progress, in order to adapt the intervention in time. Also, personal contact is highly important in efficacy. Furthermore, even though implementing cognitive interventions during treatment may be particularly challenging for the pediatric patient during this stressful period, early mitigation might yield larger effect sizes than post-treatment interventions38,49–51.
Altogether, the heterogeneity of intervention results and the inherent variability among patients with a PBT may suggest that a one-size-fits-all cognitive intervention may not be achievable. A patient-tailored implementation of neurocognitive interventions may be more beneficial4,49. Therefore, it is advised that these interventions should be contextualized alongside the prevention-based neurocognitive follow-up model2. Such model was proposed by Jacola et al. (2021), in which a primary intervention level is intended for all patients with a PBT. This level can encompass interventions like lifestyle- and behavioral interventions, psychoeducation, social engagement, improving school attendance and school engagement2,4,41. A second and a third intervention level can target patients with more specific neurocognitive risk factors and with more severe NCI. These patients may need more intense and targeted support, based on individual neurocognitive strengths and weaknesses as well as the patients’ resilience, fatigue and mental capacity2. Interventions such as school-based accommodations, teaching organizational strategies, time management, planning, cognitive remediation and problem-solving training, may be appropriate for these patients2,4,6,25.
Limitations
This systematic review comprehensively reviewed the existing intervention studies in pediatric brain tumors. Still, some limitations need to be noted. First, articles were included if PBTs participated, also if other diagnoses were included. This methodological choice increases the extent of literature regarding possible interventions in patients with a PBT. However, it could also contribute to a more heterogeneous outcome and could complicate generalizability to the PBT population in specific (vs. pediatric oncology patients). Furthermore, we were unable to perform a meta-analysis given the large heterogeneity of different study designs and neurocognitive assessment instruments. A consensus on the neurocognitive assessment instruments and the optimal timing of assessments would facilitate comparison between different intervention methods and different studies 2. Finally, by manual reference tracking, one additional study was identified, We cannot exclude the possibility there may be other remaining articles which were not detected.