Background
Cancer is the leading cause of death by disease in children and adolescents aged 5–15 years [
1]. Advances in early diagnosis and improved treatment approaches have led to an increase in long-term survival rates of up to 80% [
2]. However, survival of pediatric cancer comes at a price: An increasing amount of literature indicates that the most frequent pediatric cancer diagnoses are associated with late effects including neurocognitive deficits and intellectual decline (e.g., [
3‐
5]).
Cognitive functioning in pediatric cancer is affected by complex interactions between several factors such as age at onset or treatment modality (for review see [
6]). For example, younger age at diagnosis and at treatment of central nervous system (CNS) tumor is associated with greater cognitive problems indicating that sequelae of cancer and treatment depends on the developmental status of the child [
7]. The most common treatment for pediatric cancer include surgery, chemotherapy, and radiation therapy. All of these treatments do not solely target malignant cancer cells, but entail harmful effects to multiple organ systems, including the CNS [
4]. Treatment related side effects such as neurotoxicity [
8,
9] seem to be particularly harmful to specific cognitive processes such as the executive functions (EF) [
10‐
12].
EF are of particular importance for academic achievement with executive dysfunction having far reaching consequences on survivors’ scholastic career and overall quality of life [
13,
14]. There are three core EF (inhibition, shifting and working memory) that build the basis for higher order cognitive functioning such as planning or problem solving [
15,
16]. Impairments in EF at an early stage after diagnosis will put the patients at risk to academically fall behind peers [
3,
17,
18]. Thus, research recently focuses on developing adequate intervention and rehabilitation programs with the aim to alleviate cognitive impairments, facilitate the return to school and improve the long-term quality of life of pediatric cancer survivors (e.g., [
4]). Although the neurocognitive impairments in pediatric cancer survivors imply the necessity to intervene as early as possible [
6,
18], until now only very few intervention studies in this population are available.
Cognitive training programs have been used to address core cognitive deficits in a variety of individuals with developmental difficulties. Home-based, computerized cognitive training carries a minor burden because it can be completed flexibly at any time at home without adverse side effects [
19]. Computerized cognitive training (e.g., Cogmed RM® [
20]) is often based on a core cognitive function such as i.e. working memory and follows the assumptions that through repeated and intensive practice cognitive capacity can be increased. Several studies investigated the efficacy of such a working memory training in children and adolescents with atypical development such as attention deficit hyperactivity disorder (e.g., [
20,
21]), traumatic brain injury, stroke [
22,
23] or very preterm-born children [
24]. There is supporting evidence for improvements on tasks that resemble the training task (near-transfer effects) following working memory training. However, transfer effects to other untrained cognitive domains (far transfer effects) are currently subject to a controversial debate (for more information see [
25‐
28]).
Studies investigating working memory training in pediatric cancer survivors revealed promising results (for review see [
5]). Hardy and colleagues showed significant improvement in visual working memory and in parent-rated learning problems in a pilot study with 20 pediatric cancer survivors after working memory training when compared to an active control group undergoing a non-adaptive intervention [
29]. Working memory trainings, such as Cogmed RM®, were found to be feasible and a viable option to address cognitive late effects among pediatric cancer survivors [
19,
30]. Although there are first encouraging results regarding working memory training, yet data are too limited to form “best practice” guidelines [
5].
Physical exercise seems to be another promising approach to foster cognitive performance. Many studies indicate that physical exercise can have positive effects on a range of cognitive functions in typically developing children and adolescents [
31‐
33]. Regularly performed physical exercise can alter brain functions responsible for cognition and behavior [
31,
34,
35]. In particular, EF seem to benefit from physical training (e.g., [
36‐
38]). Recently, there has been an increasing interest in qualitative factors of physical exercise such as cognitive engagement (e.g., [
39‐
41]), because they likely influence cognition in a positive way [
32,
39,
42]. A recent study was able to demonstrate that a 6-week cognitively demanding sports game intervention for school children, but not a pure endurance training, yielded significant intervention effects on shifting, a core dimension of EF [
43]. The underlying assumption is that cognitively engaging physical activities also train brain regions that are used to control higher order cognition [
34,
42,
44]. Hence, physical exercise should ideally not only challenge the body but also the mind.
An innovative combination of a physically and cognitively demanding training at home can be achieved with exergaming [
45]. Exergaming is a portmanteau of “exercise” and “gaming” [
46], which enables individuals to physically interact with a virtual environment. In a gamified fashion, the individual has for example to avoid obstacles without touching them by jumping from left to right. The quantitative (intensity, duration = e.g., faster obstacles) and qualitative physical exercise characteristics (modality = e.g., jumping, running) can be modulated, allowing to go “beyond simply moving to moving with thought” [
47]. Up to date, there are few studies investigating the relationship between exergames and cognition [
37,
41,
48‐
51]. In pediatric cancer survivors, first positive results of physical exercise on quality of life, body composition and physical activity have been described [
52,
53]. However, high quality studies including larger samples are needed [
52,
54]. Although exergaming enables a training under highly controlled conditions at home, to our knowledge, no study to date examined the impact of physical exercise or exergaming on cognitive performance in pediatric cancer survivors.
The literature suggests that training related changes in brain structure and function can occur with cognitive and physical training [
35,
55]. Although there seems to be evidence that they might facilitate a positive effect on cognition, the underlying mechanisms remain unclear. Brain imaging therefore might be a valuable method to identify neuronal effects following working memory training or physical exercise. In adults, studies on working memory training and physical exercise revealed brain changes in structure and function (for reviews see [
35,
55]). There is, evidence for increases of brain activation (stronger neural response and greater task related activity), decreases of brain activation (increased neuronal efficiency) or a combination of both [
55]. Furthermore, there seems to be an increased functional connectivity for example in the default mode network and in the executive network following physical exercise [
56]. Studies on both, working memory training and physical exercise, detected an increase in cerebral blood flow following training [
35,
55]. Besides functional changes, structural changes in gray or white matter have been found. However, there seems to be no clear single pattern of results regarding the neural effects of cognitive and physical training, pointing towards highly dynamic plastic processes underlying cognitive change [
55].
In pediatric cancer survivors, only very few studies on working memory training or physical exercise are available. A study on working memory training suggests alterations in the functional network of working memory [
30] while physical exercise has an impact on white matter and hippocampal volume [
57]. However, the relationship between cognitive or physical training and neuroplasticity in pediatric cancer is far from understood. Therefore, more studies are needed to investigate the neural mechanisms of training.
Study aims and hypotheses
The purpose of this study is twofold: First, the availability of well-designed, child-friendly, and evidence-based training is of major clinical relevance and will contribute to the prevention of a further decline of cognitive functions and scholastic problems. Therefore, this study will compare the efficacy of two different trainings aiming to foster cognitive performance in pediatric cancer survivors. As primary outcome, we hypothesize that both trainings (computerized working memory training and exergaming) lead to improvements in core EF performance (inhibition, shifting and working memory) compared to a control condition. As secondary outcome, near and far transfer effects of both trainings are expected immediately after the training and at 3-months follow-up.
Second, the detection of training-induced changes in brain structure and function will give insight into the training related plasticity of the child’s brain. As further secondary outcome, the relationship between training related cognitive change and training related change in brain structure and function will be examined.
Discussion
Survivors of pediatric cancer frequently experience cancer-related cognitive sequelae [
3,
4]. Despite steadily improved medical approaches, pediatric cancer is often followed by lifelong cognitive constraints, leading to significant academic and professional limitations and thus a reduced quality of life [
4]. Scientific evidence for the potential negative cognitive impact of chemotherapy is emerging and seem alarming [
12,
86]).
Cancer treatment has long known to be associated with harmful effects to multiple organ systems, including the CNS. The cognitive impairments caused by the therapeutic interventions (surgery, chemotherapy and/or radiation) may be caused by therapeutic interventions, because cancer treatment might damage healthy cells. They have been associated with a vulnerability of higher-order cognitive processes such as EF and in particular attention and working memory (e.g., [
4,
6,
9,
17]). The question of interest therefore is whether early interventions can ameliorate the extent to which these late effects impair cognitive functioning of pediatric cancer survivors [
6].
As discussed in the introductory section, some studies support the efficacy of computer based working memory training programs for children with attention and working memory deficits (e.g., [
20,
21]. Studies on physical exercise also seem to yield cognitive improvements in children and adolescents (e.g., [
31‐
33]). However, until now only very few intervention studies with pediatric cancer survivors have been published. Therefore, the investigation of different training methods in a group of children in need of support is of major importance. When administered early after treatment, such training programs might have the power to ameliorate or even prevent EF problems and thus academic failure upon return to school.
There are several advantages emerging from computerized interventions. Both trainings used in the present study (Cogmed RM® and exergaming) are comparably easy to implement and enable highly controlled conditions. In addition, they offer a direct form of reward, which occurs during and immediately after training via the feedback from the computer and the number of points scored. The trainings are presented in a child-appropriate form, which might be a promising approach to foster EF performance. Moreover, an advantage of computerized interventions is the adaptivity. The level is adjusted continuously, in order to create an optimal challenging training and avoid mental underload. Besides promising first results, our study seems to be the first to examine the effects of a physical exercise on cognitive functions in pediatric cancer survivors. In addition, it is the first study comparing cognitive and physical interventions in a population of pediatric cancer survivors.
A few limitations need to be mentioned, which each and altogether might affect study results. First of all, the recruitment of the study participants will take place in two specialized pediatric units in Switzerland, the assessments will take place at one unit only. This ensures quality and standardization, but might result in smaller sample size as participants have to travel to the pediatric unit. To overcome this issue, small incentives will be offered to all participants and travel costs will be reimbursed. Second, study participation includes three assessments at the hospital as well as a supervised first training session and the intervention phase, where participants have to train at home. Therefore, the study design itself could introduce selection bias, indicating that motivated children and adolescents and participants from parents with high engagement are more likely to participate. Although, in contrast, a computerized home-based training can rather be regarded as a low-threshold intervention, it might be that the motivation to participate could have an influence on study results.