P06.05.a the Viability of Subjective Measures of Cognition as Proxies for Objective Measures

Abstract BACKGROUND Patients with IDH-mutated gliomas are at risk of impaired cognition, related to the tumor itself and given treatment. As survival rates improve, cognitive functioning and health related quality of life (HRQoL) have become more important. HRQoL is often assessed through self-reports and can include subjective measures of cognition. Cognition can also be measured by objective cognitive testing. We explored if self-reported cognition could reflect objective testing, and act as proxies. MATERIAL AND METHODS In total 66 patients with IDH-mutated glioma assessed between 2017-2024 were included. Patients underwent cognitive testing and completed the HADS, EORTC QLQ-C30 and BN20 questionnaires, before and at one-year following surgery. From the questionnaires, the functional scales: Cognitive functioning (concentration, memory) and Communication deficit(word-finding, communication- and speaking) were used. Objective cognitive measures were chosen based on the scales and consisted of the Digit span test (concentration), RAVLT (memory), Boston Naming Test (word-finding), phonemic- and semantic fluency (communication and speaking). From the neuropsychological testing, composite scores and changes in composite scores were correlated with corresponding EORTC scales and changes in scale scores. In addition, individual tests and items of the EORTC questionnaires were examined. RESULTS The EORTC cognitive functioning (r=0.38, p=0.002) as well as the single item Concentration (r=-0.31, p=0.01) within this scale were significantly correlated to the Digit span test forward. Cognitive functioning was also related to the RAVLT memory (r=0.28, p=0.036) as well as the single item Memory, both at one year (r=0.30, p=0.023). No other cognitive test scores showed significant correlation to the cognitive scale scores, nor single items within this scale. The EORTC Communication deficit were correlated to the cognitive composite test scores, preoperatively (r=0.40, p=0.002) and at one year (r=0.49, p0.001), and regarding changes over time (r=0.39, p=0.003). All single language test scores also showed significant correlation (r=0.30-0.36, p0.05) preoperatively, and at one year for the semantic fluency (r=0.58, p0.001) and phonemic fluency (r=0.40, p=0.002). Multiple items were correlated to objective test measures (r=0.27-0.61), semantic fluency being the strongest correlated. CONCLUSION There were few significant and mostly weak correlations between subjective and objective measures of cognitive function. The communication domain showed more significant associations between objective performance and subjective experience than the cognition domain. However, apart from semantic fluency, most correlations remained relatively weak. The EORTC does not seem to be a viable proxy for the more objective measures.