P08.49 The value of neurocognitive testing in multimodal response assessment in patients with glioblastoma multiforme

AbstractBackground:Due to several possible clinical and radiological pitfalls in response assessment, Tumorboard decision (TB) in patients with glioblastoma can sometimes be challenging. Early detection of recurrence in glioblastoma multiforme (GBM) is a crucial key factor for a prompt and effective second line therapy. Standard MRI sequences are the main diagnostic tool for follow up and the diagnosis of possible tumor recurrence. Additional MRI sequences, such as MRI-Diffusion and MRI-Perfusion imaging, or 18F-FET-PET (PET) imaging, may provide useful information in ambiguous cases of suspected recurrence. We performed a study addressing the question, whether additional neurocognitive assessment (NA) is in line with imaging results, as well as the TB.Methods:We consecutively enclosed twenty-five patients with histologically newly diagnosed GBM in a retrospective pilot study, counting for eighty-one (n=81) multimodal follow up (FU) test results. All patients were selected after receiving consecutive FU, comprising of MRI, PET and NA, using basically NeuroCogFX as neuropsychological test battery. Results were analysed for statistically significant correlations among themselves and over time. A consistency analysis was performed to demonstrate the impact of MRI, PET and NA in terms of distinguishing stable from progressive disease in an interdisciplinary neurooncological tumour-board.Results:A statistically significant and high correlation was found for all testing procedures (MRI, PET and NA) among themselves and in relation to the TB (all p>0.01). Also over time, the TB maintained the statistically significant correlation with all testing procedures (all p>0.01). NA correlated with all testing procedures significantly (TB: r=0.69, CI 0.55–0.79; MRI: r=0.35, CI 0.13–0.55; PET10: r=0.44, CI 0.23–0.61; PET60: r=0.48, CI 0.28–0.64, all p>0.01) in general. The correlations with MRI and PET increased over time, reaching a statistical significance at last FU. Consistency of test procedures increased over time. In case of inconsistent results in multimodal testing, NA and PET matched more frequently with the TB than MRI.Conclusion:As a novel finding, a strong correlation of NA with the TB in general and with other testing modalities during consecutive FU could be demonstrated. These results may suggest that NA is not inferior as compared to MRI and PET in terms of correlation with the TB. Whenever multimodal testing revealed inconsistent results, NA and PET were more frequently in accordance with the TB than MRI. Our data encourages further research to improve NA in multimodal response assessment.