FAP-specific PET signaling shows a moderately positive correlation with relative CBV and no correlation with ADC in 13 IDH wildtype glioblastomas.

OBJECTIVES:Targeting Fibroblast Activation Protein (FAP) is a new approach for glioblastoma imaging. In a recent pilot study glioblastomas showed elevated tracer uptake with high intratumoral heterogeneity in projection on the corresponding T2w/FLAIR and contrast enhanced MRI lesions. In this study, we correlated FAP-specific signaling with apparent diffusion coefficient (ADC) and relative cerebral blood volume (rCBV) signals in MRI to further characterize the significance of FAP uptake. METHODS:Clinical PET/CT scans of 13 glioblastoma patients were performed post i. v. administration of ⁶⁸Ga-labelled-FAP-specific tracer molecules. PET- and corresponding MRI-scans were co-registrated. 3d volumetric segmentations were performed of T2w/FLAIR lesions and contrast enhancing lesions within co-registrated MRI slides. Signal intensity values of FAP-specific PET signaling, ADC and rCBV were analyzed for their pixel wise correlation in each patient. Pooled estimates of the correlation coefficients were calculated by using the Fisher z-transformation. RESULTS:FAP-specific PET signals showed a moderately positive correlation with rCBV values which is more pronounced within the T2w/FLAIR lesion (pooled correlation 0,229) than in the contrast enhancing tumor region (pooled correlation 0.09). FAP-specific PET signals showed no correlation with ADC values. CONCLUSIONS:The moderately positive correlation of FAP-specific signals with rCBV values in MRI indicates that FAP-signaling is not independent from perfusion, but also does not only reflect intratumoral perfusion differences. The missing correlation of FAP-specific signals with ADC indicates that FAP-specific imaging does not reflect cell density, but the spot-like expression of FAP in glioblastomas. The clinical value of FAP-specific imaging needs further investigation. Copyright © 2020 Elsevier B.V. All rights reserved.

KEYWORDS: {{ getKeywords(articleDetailText.words) }}