Seaberg, Maasa

Background: Post-treatment prognosis and monitoring are critical for determining the timing of salvage treatment in glioblastoma patients but has been challenging due to difficulties differentiating progression from treatment effects in conventional images. This exploratory study aimed to establish the correlation of radiomics image features from time series of amino acid tracer 18F-DOPA PET images, with outcomes, using machine learning and dimension reduction analysis. Methods: 18F-DOPA PET images were collected for a patient cohort with wild-type IDH and unmethylated MGMT who underwent dose-escalated radiation therapy. Quantitative features were derived from the high uptake region (T/N > 2.0) in pre- and post-radiation therapy follow-up 18F-DOPA PET images. A customized workflow was utilized for pre-selecting predictive features, followed by manifold learning. Machine learning algorithms were employed to establish associations between imaging features and remaining survival (RS), defined as the time between a follow-up scan and date of death. Results: The ML models exhibited 81–83% ROC_AUC in predicting RS evaluated on an independent test dataset. A RS map is proposed for monitoring tumor alterations through serial 18F-DOPA PET scans, demonstrating superior sensitivity and better correlation with survival compared to the RANO criteria. Conclusions: Our study demonstrates that ML models utilizing FU 18F-DOPA PET images have the potential to effectively predict future survival outcomes in patients with glioblastoma treated with dose-escalated radiation therapy. The capability to assess changes in tumor over time through imaging can potentially assist in patient stratification and the selection of salvage treatments, while also aiding in distinguishing treatment effects from genuine tumor progression.

Abstract While dose escalation of radiotherapy (DERT) has failed to improve overall survival (OS) or progression-free survival (PFS) for glioblastoma in previous studies, a recent phase II clinical trial utilizing 18F-DOPA-PET-directed DERT demonstrated improved PFS in MGMT-unmethylated patients and OS in MGMT-methylated patients compared to historical controls. This planned secondary analysis sought to determine 1) how 18F-DOPA-PET changes RT volumes beyond standard MRI-planning, 2) which patients benefit most and least from this protocol, 3) which are mostly likely to experience clinically significant radionecrosis after DERT, and 4) patterns of failure after DERT. For 69 evaluable patients, median MRI-defined, PET-defined, and combined low-dose gross tumor volumes (GTV51) were 54 cc (range 9-248), 23 cc (0.4-179), and 62 cc (10-260), respectively. Median MRI-defined, PET-defined, and combined high-dose GTVs (GTV76) were 32 cc (range 4-136), 6 cc (0.1-138), and 34 cc (4-162), respectively. 18F-DOPA-PET resulted in a median volumetric expansion of 13% (0-243%) and 5% (0-217%) from MRI-defined low-dose and high-dose GTV’s, respectively. Central failures ( >95% of recurrence tumor volume) occurred within the 76 Gy, 60 Gy, and 51 Gy isodose lines in 32 (46%), 60 (87%), and 64 (93%) patients, respectively. Recursive partitioning analysis stratified patients by OS and PFS. Patients with 18F-DOPA-PET-defined GTV76 > 7.8cc, MRI-defined GTV76 > 42.7cc, and MGMT promotor-unmethylated tumors had the shortest OS, while patients with smaller PET and MRI-defined tumors had the longest OS (median 10.4 vs. 64.6 months, p< 0.001). Similarly, PFS was worst in patients with 18F-DOPA-PET-defined GTV76 > 2.17 cc who had biopsy only (median PFS 3.2 months, p< 0.001). Patients with 18F-DOPA-PET-defined GTV51 > 50 cc had the highest risk of grade 3+ radionecrosis (p< 0.001). In conclusion, larger 18F-DOPA-PET and MRI-defined tumor volumes were associated with worse outcomes, and 18F-DOPA-PET-directed DERT appears to reduce risk of central recurrence in high-dose volumes.