COGTIPS (COGnitive Training In Parkinson Study): The Effect of Online Cognitive Training on Cognition and Brain Networks in Parkinson's Disease

BACKGROUND In Parkinson's disease (PD), cognitive dysfunction is frequently reported - approximately 50% of PD patients experience cognitive impairment (Litvan et al., 2011). Of these impairments, executive dysfunction is most frequently reported early in the disease trajectory (Bosboom, Stoffers, & Wolters, 2004; Muslimovic, Post, Speelman, & Schmand, 2005), while impairments in other cognitive domains (i.e. attention, episodic memory, visuospatial functions) are also highly prevalent (Bosboom et al., 2004). The majority of PD patients ultimately develops PD dementia (PD-D; Aarsland, Andersen, Larsen, Lolk, & Kragh-Sorensen, 2003; Hely, Reid, Adena, Halliday, & Morris, 2008). Moreover, about 10% of the PD patients develops PD-D every year (Aarsland & Kurz, 2010). Cognitive dysfunctions in PD have a significant negative influence on the quality of life (Klepac, Trkulja, Relja, & Babic, 2008), while treatment of these dysfunctions is in its infancy.

Cognitive training may provide a new intervention for reducing cognitive complaints and delaying the onset of mild cognitive impairment (MCI) or PD-D. This intervention has been widely studied in other diseases (Cicerone et al., 2011; Olazaran et al., 2010). Moreover, studies have provided evidence not only for behavioral influences, but also for brain connectivity and activity effects of cognitive training (Chapman et al., 2015; Castellanos et al., 2010; Subramaniam et al., 2012; Subramaniam et al., 2014; Belleville et al., 2011; Rosen, Sugiura, Kramer, Whitfield-Gabrieli, & Gabrieli, 2011). This suggests a restorative effect of cognitive training on disrupted brain networks.

In PD, cognitive dysfunction - mainly executive dysfunction - is associated with disruption of the cortico-striato-thalamo-corticale circuits by depletion of dopamine. Dysfunction of these circuits seems to disrupt several cognitive networks, which leads to cognitive dysfunction (Baggio et al., 2014). Cognitive training could counteract these disruptions by normalising activity and connectivity, and ultimately lead to a reduction of impairment. Since earlier studies in different patient populations have shown that cognitive training has lasting effects (Petrelli et al., 2015), normalising disruptions underlying cognitive impairment could prevent cognitive deterioration and therefore prevent or delay the development of PD-D.

Few studies in PD have focused on cognitive training and its neural correlates. A meta-analysis by Leung et al. (2015) showed positive effects of cognitive training on mainly 'frontal' cognitive functions (i.e. working memory, executive functions, processing speed). In addition, earlier research has described a neuroprotective effect of cognitive training on the development of MCI in PD (odds ratio: 3; Petrelli et al., 2015). Until now, however, studies have been relatively small and mainly without a controlled design - consequently, there is a need for large randomized controlled studies (Hindle, Petrelli, Clare, & Kalbe, 2013; Leung et al., 2015). Moreover, neural effects of cognitive training are largely unknown in PD. Furthermore, it is important to study the improvement of patients on daily functioning after cognitive training, rather than solely focusing on cognitive tasks and neural measures. Finally, cognitive training has been performed mainly in hospital settings, while PD patients have mobility problems - a training method suitable to perform from home is therefore needed for this population.

OBJECTIVES The study objective is primarily to measure the effect of an online cognitive training in patients with mild cognitive complaints in PD. An online training, specifically altered for PD patients (BrainGymmer) will be compared with an active comparator. In both conditions, participants will train eight weeks, three times a week during 45 minutes.