Virtual reality-based cognitive assessment: An exploratory pilot study including patients with cancer-related...

Cancer-related cognitive impairment (CRCI), or "chemo brain," can adversely affect quality of life and potentially survival for patients with cancer. CRCI can especially impact older adults with cancer. Cognitive assessment remains challenging in a busy oncology practice, and virtual reality (VR) based assessment is an emerging option. VR-based assessment utilizes an immersive, multisensory virtual reality environment to evaluate cognitive function. One such platform includes the Virtuleap Enhance application. Our goal was to compare Virtuleap Enhance to standard cognitive assessments.

This was a single-center pilot study that included healthy participants and those with CRCI. Standard assessments included Montreal Cognitive Assessment (MOCA), Trail Making Test (TMT), Word Recall, and Stroop tasks. Subjects were assessed for three sessions, at least one month apart. Virtuleap Enhance applications included React (task switching, response inhibition), Memory Wall (short-term memory), Magic Deck (long-term memory), and Odd Egg (deductive reasoning). Statistical analysis included multiple imputed linear mixed effects regression model with inverse probability weighting. Significance was defined as p < 0.05.

The overall cohort included 54 participants, including 35% experiencing CRCI; the mean age was 65 years. Those with CRCI had higher depression, stress, and anxiety scores with lower MOCA scores compared to healthy controls. Among the CRCI group, there were reliable correlations between the React task and the MOCA and Stroop tests while other VR tasks (Odd Egg, Memory) showed no reliable association. There was significant change over time for the React and Odd Egg tasks.

VR showed reliable correlation with traditional cognitive assessments for some comparisons but not others. Some change over time was noted for certain VR tasks. This could be due to VR assessment measuring subtle changes in chemo brain that traditional testing may not necessarily capture. Future work based on lessons learned from this initial study may help clarify this possibility.