Article
Towards an optimised clinical neuropsychological assessment for glioblastoma patients – concept and interim analysis of a prospective multi-centre study
Hin zu einer optimierten klinisch-neuropsychologischen Testung bei Glioblastompatienten – Konzept und Interimanalyse einer prospektiven, multizentrischen Studie
Search Medline for
Authors
Published: | June 26, 2020 |
---|
Outline
Text
Objective: Neurocognitive deficits are a common and burdening sequela of brain lesions often remain undetected due to the lack of time-efficient, easily administrable and reliable tests, validated for this specific patient collective. We set up this study to investigate the use of a comprehensive neurocognitive test battery in a representative cohort of glioblastoma patients with respect to
- (i) clinical feasibility,
- (ii) tumour-location-related test sensitivity and specificity,
- (iii) test-parallel-test-reliability and to assess the influence of factors like the neurosurgical intervention on specific cognitive domains.
Methods: A total of n=250 previously untreated patients with suspected single supratentorial glioblastoma lesions will be allocated to this multicenter study throughout Germany. MRI scans, parallel neurocognitive test versions and patient surveys (SF-12, BDI) are performed pre- and postoperatively as well as every three months until the date of first tumour recurrence. Assessment time and motivation of the patients to participate in repeated testing are assessed as feasibility measures. As an addition to the calculation of sensitivity and specificity, a voxel-based lesion symptom mapping is performed. Intraclass correlation coefficients (ICC) are calculated as a measure of test-parallel-test-reliability. The average postoperative test results are compared to the paired preoperative baseline assessments.
Results: The interim analysis based on n=128 glioblastoma patients shows encouraging results with regard to the logistical feasibility (mean administration time: 39 ± 10 min) and usually excellent acceptance by the patients (8/10 on the numeric rating scale). Reliability was good to excellent (ICC=0.6–0.9), depending on the specific test, except for the short form of the trail-making-test C (ICC=0.51).
Conclusion: The study introduces an easily administrable, reliable and validated tool for a comprehensive assessment of neurocognitive functions in glioblastoma patients, offering the opportunity of longitudinal follow-up due to five parallel versions. Moreover, the final study results will allow to tailor location-specific (shortened) test sets in order to further improve the clinical feasibility.