Integrated analysis of dynamic FET PET/CT parameters, histology, and methylation profiling of 44 gliomas

Röhrich, Manuel; Huang, Kristin; Schrimpf, Daniel; Albert, Nathalie L. ; Hielscher, Thomas; Deimling, Andreas von; Schüller, Ulrich; Dimitrakopoulou-Strauss, Antonia und Haberkorn, Uwe (2018): Integrated analysis of dynamic FET PET/CT parameters, histology, and methylation profiling of 44 gliomas. In: European Journal of Nuclear Medicine and Molecular Imaging, Bd. 45, Nr. 9: S. 1573-1584

Abstract

Dynamic F-18-FET PET/CT is a powerful tool for the diagnosis of gliomas.F-18-FET PET time-activity curves (TAC) allow differentiation between histological low-grade gliomas (LGG) and high-grade gliomas (HGG). Molecular methods such as epigenetic profiling are of rising importance for glioma grading and subclassification. Here, we analysed dynamic F-18-FET PET data, and the histological and epigenetic features of 44 gliomas. Dynamic F-18-FET PET was performed in 44 patients with newly diagnosed, untreated glioma: 10 WHO grade II glioma, 13 WHO grade III glioma and 21 glioblastoma (GBM). All patients underwent stereotactic biopsy or tumour resection after F-18-FET PET imaging. As well as histological analysis of tissue samples, DNA was subjected to epigenetic analysis using the Illumina 850 K methylation array. TACs, standardized uptake values corrected for background uptake in healthy tissue (SUVmax/BG), time to peak (TTP) and kinetic modelling parameters were correlated with histological diagnoses and with epigenetic signatures. Multivariate analyses were performed to evaluate the diagnostic accuracy of F-18-FET PET in relation to the tumour groups identified by histological and methylation-based analysis. Epigenetic profiling led to substantial tumour reclassification, with six grade II/III gliomas reclassified as GBM. Overlap of HGG-typical TACs and LGG-typical TACs was dramatically reduced when tumours were clustered on the basis of their methylation profile. SUVmax/BG values of GBM were higher than those of LGGs following both histological diagnosis and methylation-based diagnosis. The differences in TTP between GBMs and grade II/III gliomas were greater following methylation-based diagnosis than following histological diagnosis. Kinetic modeling showed that relative K1 and fractal dimension (FD) values significantly differed in histology- and methylation-based GBM and grade II/III glioma between those diagnosed histologically and those diagnosed by methylation analysis. Multivariate analysis revealed slightly greater diagnostic accuracy with methylation-based diagnosis. IDH-mutant gliomas and GBM subgroups tended to differ in their F-18-FET PET kinetics. The status of dynamic F-18-FET PET as a biologically and clinically relevant imaging modality is confirmed in the context of molecular glioma diagnosis.

• Dokumententyp: Zeitschriftenartikel
• Fakultät: Medizin
• Themengebiete: 600 Technik, Medizin, angewandte Wissenschaften > 610 Medizin und Gesundheit
• ISSN: 1619-7070
• Sprache: Englisch