P03.21.b Identifying Residual Tumor Burden Following Glioblastoma Resection: a Comparative Analysis of Mr-based Rano Resect Classes Vs.[18f]fet-pet
Year: 2025
Type: conference-abstract
Abstract: BACKGROUND Extensive resection improves glioblastoma survival, with residual tumor volume as a key prognostic marker. The RANO resect group proposed a classification based on postoperative MRI volume. With [18F]FET-PET increasingly used for glioma delineation, we compared MRI- and PET-based volumetrics. MATERIAL AND METHODS We retrospectively analyzed IDH-wildtype glioblastoma patients who underwent [18F]FET-PET and MRI post-surgery. Two investigators assessed residual tumor volumes using PET RANO 1.0 and MRI-based RANO resect criteria, comparing both modalities. Tumor detection confidence (1-5 scale) and spatial overlap between modalities (Dice-Sørensen coefficients) were assessed. RESULTS We identified 42 patients with available MRI and [18F]FET-PET for review. Contrast-enhancing (CE+) tumor was identified in 76.2% of cases (32/42), with a median volume of 0.48 cm³ ± 4.9 cm³ (range: 0.03-31.7 cm³). Non-contrast-enhancing (CE-) tumor was present in 98% of cases (41/42), with a median volume of 3.68 cm³ ± 6.5 cm³ (range: 0.3-33.7 cm³). According to RANO Resect criteria, 21% (9/42) of patients were classified as class 1, 43% (18/42) as class 2, and 36% (15/42) as class 3. PET-positive volumes were detected in all patients. The median PET-positive volume was 13.45 cm³ ± 16.2 cm³ (range: 1.84-70.1 cm³), with a median TBRmax of 2.78 ± 1 (range: 1.84-6.14). PET-positive volumes increased with higher RANO Resect classes (Spearman r < 0.37). Spatial overlap between PET and MRI-based tumor volumes (T2 and CE-T1) was assessed using the Dice-Sørensen coefficient, revealing limited concordance and frequent PET volume predominance. Overlap was categorized into match, partial match, and mismatch. PET raters showed higher confidence (median score: 5) in detecting residual tumor compared to MRI readers (median: 4), particularly for CE- tumor components. In nearly all cases, PET-defined tumor volumes exceeded those seen on MRI, resulting in only few spatially matching tumor areas. PET raters also reported higher confidence in tumor detection (median 5) compared to MRI investigators (median 4). PET volumes increased across RANO resect classes, from 6.49 cm³ ± 6.8 cm³ in class 1 to 20.5 cm³ ± 20.8 cm³ in class 3. CONCLUSION Our study suggests that PET may detect more residual disease post-surgery than MRI, with PET volumes often exceeding those delineated by MRI indicating higher sensitivity. [18F]FET-PET may present a valuable tool to complement MRI for the stratification of patients based on the residual tumor volume and potentially also to guide local therapies.
Source: Neuro-Oncology
Authors: J Blobner, K J Mueller, R Forbrig, N Thon, S Schoenecker, P N Harter, L Baumgarten, J Tonn, P Karschnia, N L Albert
Language: English
Subfield: Genetics
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W4414799758
Cites:
Cited by:
Related to: