P. L. Y. Tang1,2, M. Smits2,3, R. A. Nout1, R. Baak1, C. van Rij1, C. Slagter1, A. Swaak1, E. A. H. Warnert2, and A. Mendez Romero1; 1Department of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, Netherlands, 2Department of Radiology & Nuclear Medicine, Erasmus MC, University Medical Center Rotterdam, Rotterdam, Netherlands, 3Medical Delta, Delft, Netherlands
Purpose/Objective(s): Radiotherapy is one of the pillars in the management of glioblastoma. This tumor is notorious for extensive tumor infiltration that is not visible on conventional MRI scans. Therefore, a 15 mm clinical target volume (CTV) margin around the gross tumor volume (GTV) is recommended according to the ESTRO-EANO 2023 guidelines. This approach commonly results in substantial CTVs that can include large amounts of healthy tissue, increasing the risk of radiation-induced side effects. Amide proton transfer weighted (APTw) imaging – an MRI technique that reflects local levels of mobile proteins and peptides – has been associated with increased Ki-67 expression and cell proliferation in glioblastoma, and could aid in visualizing tumor infiltration. This offers an opportunity for more accurate target delineation through the introduction of a biological tumor volume (BTV), and reduction of the 15 mm CTV margin. In this study, the potential of integrating APTw imaging for definition of a biological GTV (GTVbio) is explored. The hypothesis is that a GTVbio, which includes information derived from APTw imaging, is larger than the conventional GTV and smaller than the conventional CTV. Materials/
Methods: APTw imaging was added to the MRI scan prior to radiotherapy of patients with glioblastoma. The construction of the GTV, CTV and radiotherapy treatment plan was based on the conventional MRI scans. A BTV, defined by hyperintense APTw signal within regions with contrast enhancement or T2w/FLAIR hyperintensity, was semi-automatically delineated via a patient-specific threshold. Thereafter, the BTV was added to the GTV to construct the GTVbio. A Wilcoxon signed-rank test was performed to compare the volume of the GTV and CTV with the volume of the GTVbio. Additionally, the dice similarity coefficient (DSC) between the GTV and the GTVbio was computed to examine the similarity between the two volumes. Results: Between June 2023 and January 2024, twelve patients (9M, 3F) with glioblastoma were included; one patient (M) was excluded from analysis as the patient did not proceed with radiation treatment. Eight patients received a total dose of 60 Gy; three patients received 40.05 Gy. The GTVbio (median volume = 62.2 mL, IQR 37.4 mL - 85.4 mL) was significantly larger than the GTV (median volume = 62.2 mL, IQR 37.4 mL – 68.1 mL), with a median increase in size of 0.87% (IQR 0.13% - 16.96%, p=0.00195), and significantly smaller than the CTV (median volume = 177.0 mL, IQR 139.5 mL – 264.3 mL), with a median reduction in size of 73.93% (IQR 63.78% - 76.52%, p=0.00098). The median DSC between GTV and GTVbio was 0.996 (IQR 0.925 - 0.999). Conclusion: In this study, the distinctive GTVbio volumes in relation to the GTV and CTV underscore the potential of APTw imaging for improved target delineation due to enhanced visualization of tumor infiltration, ultimately enabling more personalized target definition and reduced risk of radiation-induced toxicity. Future work will focus on pattern-of-failure analysis of the recurring tumor.
