Histopathology-verified comparison of FET PET and contrast-enhanced MRI for tumor visualization in glioblastoma.
Abstract
2049 Background: In glioblastoma, precise visualization is a prerequisite for optimal resection strategies and accurate radiotherapy volume definition. In routine practice, contrast-enhanced MRI (CE-MRI) is most commonly used; however, it can underestimate tumor extent when viable tumor is present without blood–brain barrier disruption and therefore without contrast uptake. The present study examined whether amino acid PET with O-(2-[ 18 F]fluoroethyl)-L-tyrosine (FET) can better visualize the true distribution of tumor tissue and thereby improve preoperative assessment. Methods: In this prospective, biopsy-validated single-center study, in 61 patients with IDH-wildtype glioblastoma (44 newly diagnosed, 17 recurrent) 472 biopsies spatially correlated with preoperative FET PET and CE-MRI were histopathologically evaluated. Diagnostic metrics were calculated and compared between modalities using McNemar and Chi² tests (p = 0.05). Results: Overall, FET PET identified tumor tissue with significantly higher sensitivity than CE-MRI (91.2% vs. 55.7%; p < 0.001). When benchmarked against T2/FLAIR, FET remained more sensitive (85.3%; p = 0.012) while also demonstrating a pronounced gain in specificity (82.5% vs. 33.3%; p < 0.001). Remarkably, there was the high reliability of FET in regions without MRI contrast enhancement: biopsies taken from FET-positive areas without contrast enhancement showed a positive predictive value of 98.3%. Subgroup analyses consistently demonstrated the advantage of FET across clinical and histological strata, including newly diagnosed and recurrent disease (p < 0.001), as well as within cohorts with resection (p < 0.001; specificity p = 0.03) or biopsy (p < 0.001). Importantly, performance advantages were also observed regardless of histological tumor cell density, with FET exceeding CE-MRI in both histological strata (infiltration zone and solid tumor, both p < 0.001). Conclusions: FET PET demonstrated significant diagnostic superiority over CE-MRI for the visualization of glioblastoma tissue. Comparable positive predictive values in contrast-enhancing and non-enhancing regions indicate that FET PET detects tumor reliably, independent of the blood–brain barrier integrity. These findings support the integration of amino acid PET as an important component in target definition for both surgical and radiation planning in glioblastoma.
Article Details
Journal Info
Journal of Clinical Oncology
Lippincott Williams & Wilkins
Authors (20)
Roman Stuerzl
Department of Nuclear Medicine, LMU University Hospital, LMU Munich, Munich, Germany
Sabrina Viktoria Kirchleitner
Department of Neurosurgery, University of Munich (LMU), Munich, Germany
Julia Lorenz
Department of Neuropathology, University Hospital Regensburg, 93053 Regensburg, Germany, Regensburg, Germany
Katharina Müller
Department of Pathology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
Veit Michael Stoecklein
Department of Neurosurgery, University Hospital, LMU Munich, 81377 Munich, Germany, Munich, Germany
Stephanie Quach
Department of Neurosurgery, University Hospital OWL, Campus Bielefeld-Bethel, Bielefeld, Germany, Bielefeld, Germany
Sophia Stoecklein
5LMU University Hospital, Department of Radiology, Munich, Germany
Jens Blobner
Department of Neurosurgery LMU, Munich, Germany
Tobias Greve
Matthias Brendel
Patrick Harter
Center of Neuropathology and Prion Research, Faculty of Medicine, LMU Munich, Munich, Germany
Victoria Ruf
Center for Neuropathology and Prion Research, Faculty of Medicine, LMU Munich, Munich, Germany
Rainer Rupprecht
Department of Psychiatry and Psychotherapy, University of Regensburg, Regensburg, Germany
Florian Ringel
Darius Kalasauskas
Department of Neurosurgery, LMU University Hospita, Munich, Germany
Markus Riemenschneider
University of Regensburg, Regensburg, Germany
Niklas Thon
Department of Neurosurgery, Knappschaft University Hospital Bochum, Bochum, Germany
Louisa von Baumgarten
12Department of Neurosurgery, Ludwig Maximilian University, Munich, Germany
Joerg Tonn
Department of Neurosurgery, LMU University Hospital, Munich, Germany
Nathalie Lisa Albert
Department of Nuclear Medicine, LMU University Hospital, LMU Munich, Munich, Germany