NIMG-06. CHARACTERIZATION OF LONG-TERM METABOLIC CHANGES OF IRRADIATED BRAIN METASTASES USING SERIAL DYNAMIC FET PET IMAGING

BACKGROUND In the present study, we characterized the long-term metabolic changes of brain metastases irradiated with stereotactic radiosurgery (SRS) by sequential dynamic PET imaging using the radiolabeled amino acid O-(2-[18F]-fluoroethyl)-L-tyrosine (FET). We hypothesized that this approach is of considerable clinical value to diagnose delayed radiation-induced changes. PATIENTS AND METHODS From 2010-2021, we retrospectively identified patients with brain metastases from solid extracranial primary tumors who (i) were treated with SRS with or without concurrent immunotherapy using checkpoint inhibitors, (ii) had equivocal or progressive MRI findings after SRS, and (iii) subsequently underwent at least two additional dynamic FET PET scans during follow-up for long-term evaluation. Mean tumor-to-brain ratios (TBR) and the dynamic FET PET parameter time-to-peak were obtained. Diagnostic performances were calculated using receiver operating characteristic curve analyses. Diagnoses were confirmed histologically or clinicoradiologically. RESULTS We identified 36 patients with 98 FET PET scans (median number, 3; range, 2-6). Concurrent to SRS, 8 patients (22%) were treated with checkpoint inhibitors. Following SRS, suspicious MRI findings occurred after a median time of 11 months (range, 2-64 months). Subsequently, FET PET scans were acquired over a median period of 13 months (range, 5-60 months). The overall median follow-up time was 26 months (range, 8-101 months). Twenty-one patients (58%) had delayed radiation-induced changes. TBRs calculated from the last available FET PET scan showed the highest accuracy (92%) to identify delayed radiation-induced changes (threshold, 1.95; P< 0.001). CONCLUSIONS FET PET has a high diagnostic accuracy for characterizing the long-term changes of irradiated brain metastases.

Prognostic value of pre-irradiation FET PET in patients with not completely resectable IDH-wildtype glioma and minimal or absent contrast enhancement

In glioma patients, complete resection of the contrast-enhancing portion is associated with improved survival, which, however, cannot be achieved in a considerable number of patients. Here, we evaluated the prognostic value of O-(2-[18F]-fluoroethyl)-L-tyrosine (FET) PET in not completely resectable glioma patients with minimal or absent contrast enhancement before temozolomide chemoradiation. Dynamic FET PET scans were performed in 18 newly diagnosed patients with partially resected (n = 8) or biopsied (n = 10) IDH-wildtype astrocytic glioma before initiation of temozolomide chemoradiation. Static and dynamic FET PET parameters, as well as contrast-enhancing volumes on MRI, were calculated. Using receiver operating characteristic analyses, threshold values for which the product of paired values for sensitivity and specificity reached a maximum were obtained. Subsequently, the prognostic values of FET PET parameters and contrast-enhancing volumes on MRI were evaluated using univariate Kaplan–Meier and multivariate Cox regression (including the MTV, age, MGMT promoter methylation, and contrast-enhancing volume) survival analyses for progression-free and overall survival (PFS, OS). On MRI, eight patients had no contrast enhancement; the remaining patients had minimal contrast-enhancing volumes (range, 0.2–5.3 mL). Univariate analyses revealed that smaller pre-irradiation FET PET tumor volumes were significantly correlated with a more favorable PFS (7.9 vs. 4.2 months; threshold, 14.8 mL; P = 0.012) and OS (16.6 vs. 9.0 months; threshold, 23.8 mL; P = 0.002). In contrast, mean tumor-to-brain ratios and time-to-peak values were only associated with a longer PFS (P = 0.048 and P = 0.045, respectively). Furthermore, the pre-irradiation FET PET tumor volume remained significant in multivariate analyses (P = 0.043), indicating an independent predictor for OS. Our results suggest that pre-irradiation FET PET parameters have a prognostic impact in this subgroup of patients.

OS03.3.A Characterization of long-term metabolic changes of irradiated brain metastases using serial dynamic FET PET imaging

BACKGROUND In the present study, we characterized the long-term metabolic changes of brain metastases irradiated with stereotactic radiosurgery (SRS) by sequential dynamic PET imaging using the radiolabeled amino acid O-(2-[18F]-fluoroethyl)-L-tyrosine (FET). We hypothesized that this approach is of considerable clinical value to diagnose delayed radiation-induced changes. MATERIAL AND METHODS From 2010–2021, we retrospectively identified patients with brain metastases from solid extracranial primary tumors who (i) were treated with SRS with or without concurrent immunotherapy using checkpoint inhibitors, (ii) had equivocal or progressive MRI findings after SRS, and (iii) subsequently underwent at least two additional dynamic FET PET scans during follow-up for long-term evaluation. Mean tumor-to-brain ratios (TBR) and the dynamic FET PET parameter time-to-peak were obtained. Diagnostic performances were calculated using receiver operating characteristic curve analyses. Diagnoses were confirmed histologically or clinicoradiologically. RESULTS We identified 36 patients with 98 FET PET scans (median number, 3; range, 2–6). Concurrent to SRS, 8 patients (22%) were treated with checkpoint inhibitors. Following SRS, suspicious MRI findings occurred after a median time of 11 months (range, 2–64 months). Subsequently, FET PET scans were acquired over a median period of 13 months (range, 5–60 months). The overall median follow-up time was 26 months (range, 8–101 months). Twenty-one patients (58%) had delayed radiation-induced changes. TBRs calculated from the last available FET PET scan showed the highest accuracy (92%) to identify delayed radiation-induced changes (threshold, 1.95; P<0.001). CONCLUSION FET PET has a high diagnostic accuracy for characterizing the long-term changes of irradiated brain metastases.

Prognostic value of pre-irradiation FET PET in patients with not completely resectable IDH-wildtype glioma and minimal or absent contrast enhancement

Purpose: In glioma patients, complete resection of the contrast-enhancing portion is associated with improved survival, which, however, cannot be achieved in a considerable number of patients. Here, we evaluated the prognostic value of O-(2-[18F]-fluoroethyl)-L-tyrosine (FET) PET in not completely resectable glioma patients with minimal or absent contrast enhancement before temozolomide chemoradiation.Methods: Dynamic FET PET scans were performed in 18 newly diagnosed patients with partially resected (n=8) or biopsied (n=10) IDH-wildtype astrocytic glioma before initiation of temozolomide chemoradiation. Static and dynamic FET PET parameters, as well as contrast-enhancing volumes on MRI, were calculated. Using receiver operating characteristic analyses, threshold values for these parameters were obtained. Subsequently, the prognostic values of FET PET parameters and contrast-enhancing volumes on MRI were evaluated using univariate Kaplan-Meier and multivariate Cox regression survival analyses for progression-free and overall survival (PFS, OS).Results: On MRI, eight patients had no contrast enhancement; the remaining patients had minimal contrast-enhancing volumes (range, 0.2-5.3 mL). Univariate analyses revealed that a smaller pre-irradiation FET PET tumor volume significantly influenced PFS (7.9 vs. 4.2 months; threshold, 14.8 mL; P=0.012) and OS (16.6 vs. 9.0 months; threshold, 23.8 mL; P=0.002). In contrast, mean tumor-to-brain ratios and time-to-peak values were only associated with a longer PFS (P=0.048 and P=0.045, respectively). Furthermore, the pre-irradiation FET PET tumor volume remained significant in multivariate analyses (P=0.043), indicating an independent predictor for OS.Conclusion: Our results suggest that pre-irradiation FET PET parameters have a prognostic impact in this subgroup of patients.

FET PET Radiomics for Differentiating Pseudoprogression from Early Tumor Progression in Glioma Patients Post-Chemoradiation

Currently, a reliable diagnostic test for differentiating pseudoprogression from early tumor progression is lacking. We explored the potential of O-(2-[18F]fluoroethyl)-L-tyrosine (FET) positron emission tomography (PET) radiomics for this clinically important task. Thirty-four patients (isocitrate dehydrogenase (IDH)-wildtype glioblastoma, 94%) with progressive magnetic resonance imaging (MRI) changes according to the Response Assessment in Neuro-Oncology (RANO) criteria within the first 12 weeks after completing temozolomide chemoradiation underwent a dynamic FET PET scan. Static and dynamic FET PET parameters were calculated. For radiomics analysis, the number of datasets was increased to 102 using data augmentation. After randomly assigning patients to a training and test dataset, 944 features were calculated on unfiltered and filtered images. The number of features for model generation was limited to four to avoid data overfitting. Eighteen patients were diagnosed with early tumor progression, and 16 patients had pseudoprogression. The FET PET radiomics model correctly diagnosed pseudoprogression in all test cohort patients (sensitivity, 100%; negative predictive value, 100%). In contrast, the diagnostic performance of the best FET PET parameter (TBRmax) was lower (sensitivity, 81%; negative predictive value, 80%). The results suggest that FET PET radiomics helps diagnose patients with pseudoprogression with a high diagnostic performance. Given the clinical significance, further studies are warranted.

NIMG-14. MACHINE LEARNING-BASED EVALUATION OF STATIC AND DYNAMIC FET-PET FOR THE DETECTION OF PSEUDOPROGRESSION IN PATIENTS WITH IDH-WILDTYPE GLIOBLASTOMA

BACKGROUND Pseudoprogression (PSP) detection in glioblastoma has important clinical implications and remains a challenging task. With the significant advances provided by machine learning (ML) in health care, we investigated the potential of ML in improving the performance of PET using O-(2-[18F]-fluoroethyl)-L-tyrosine (FET) for differentiation of tumor progression from PSP in IDH-wildtype glioblastoma. METHODS We retrospectively evaluated the PET data of patients with newly diagnosed IDH-wildtype glioblastoma following chemoradiation. All patients presented imaging findings suspected of PSP/TP on contrast-enhanced MRI. For further diagnostic evaluation, patients underwent subsequently an additional dynamic FET-PET scan. The modified Response Assessment in Neuro-Oncology (RANO) criteria served to diagnose PSP. To develop a robust ML model, we trained a Linear Discriminant Analysis (LDA)-based classifier using FET-PET derived features on a training cohort and validated the results on a separate test cohort. The results of the ML model were compared with a conventional FET-PET analysis using the receiver-operating-characteristic (ROC) curve. RESULTS Of the 44 patients included in this study, 14 patients were diagnosed with PSP. The mean (TBRmean) and maximum tumor-to-brain ratios (TBRmax) were significantly higher in the TP group as compared to the PSP group (p=0.010 and p=0.047, respectively). The area under the ROC curve (AUC) for TBRmax and TBRmean was 0.68 and 0.74, respectively. Using the LDA-based algorithm, the AUC (0.93) was significantly higher than the AUC for TBRmax. CONCLUSIONS This study shows that in IDH-wildtype glioblastoma, ML-based PSP detection leads to better diagnostic performance compared to conventional ROC analysis.

A Preliminary Study on Machine Learning-Based Evaluation of Static and Dynamic FET-PET for the Detection of Pseudoprogression in Patients with IDH-Wildtype Glioblastoma

Pseudoprogression (PSP) detection in glioblastoma remains challenging and has important clinical implications. We investigated the potential of machine learning (ML) in improving the performance of PET using O-(2-[18F]-fluoroethyl)-L-tyrosine (FET) for differentiation of tumor progression from PSP in IDH-wildtype glioblastoma. We retrospectively evaluated the PET data of patients with newly diagnosed IDH-wildtype glioblastoma following chemoradiation. Contrast-enhanced MRI suspected PSP/TP and all patients underwent subsequently an additional dynamic FET-PET scan. The modified Response Assessment in Neuro-Oncology (RANO) criteria served to diagnose PSP. We trained a Linear Discriminant Analysis (LDA)-based classifier using FET-PET derived features on a hold-out validation set. The results of the ML model were compared with a conventional FET-PET analysis using the receiver-operating-characteristic (ROC) curve. Of the 44 patients included in this preliminary study, 14 patients were diagnosed with PSP. The mean (TBRmean) and maximum tumor-to-brain ratios (TBRmax) were significantly higher in the TP group as compared to the PSP group (p = 0.014 and p = 0.033, respectively). The area under the ROC curve (AUC) for TBRmax and TBRmean was 0.68 and 0.74, respectively. Using the LDA-based algorithm, the AUC (0.93) was significantly higher than the AUC for TBRmax. This preliminary study shows that in IDH-wildtype glioblastoma, ML-based PSP detection leads to better diagnostic performance.

NIMG-05. THE T2-FLAIR MISMATCH SIGN IN IDH-MUTANT ASTROCYTOMAS - IS THERE AN ASSOCIATION WITH FET PET UPTAKE?

BACKGROUND The purpose of this study was (i) to assess the reproducibility of the previously described T2-FLAIR mismatch sign as a highly specific MR imaging marker in non-enhancing IDH-mutant, 1p/19q non-codeleted lower-grade gliomas (LGG) of the WHO grades II or III, and (ii) its association with the uptake of the radiolabeled amino acid O-(2-[18F]-fluoroethyl)-L-tyrosine (FET) in PET to further metabolically characterize that sign, which is currently poorly understood. METHODS Consecutive MRI and dynamic FET PET scans (n=134) from newly diagnosed and neuropathologically confirmed IDH-mutant LGG (n=65) and IDH-wildtype gliomas as control group (n=69) were evaluated by two independent raters to assess presence/absence of the T2-FLAIR mismatch sign as well as FET uptake. Interrater agreement was assessed using Cohen’s kappa (κ), as well as diagnostic performance (i.e., positive/negative predictive value; PPV, NPV) of the T2-FLAIR mismatch sign to identify IDH-mutant astrocytomas. RESULTS In the LGG group, 13 patients (20%) had a T2-FLAIR mismatch sign, which could be identified with a substantial interrater agreement (κ=0.75). In contrast, that sign was absent in IDH-wildtype gliomas. All 13 cases that were positive for the T2/FLAIR mismatch sign were IDH-mutant, 1p/19q non-codeleted tumors (PPV=100%, NPV=57%). Interestingly, compared to IDH-mutant gliomas without the T2-FLAIR mismatch sign, the sign was significantly (P=0.027; 10 of 13 patients) associated with a negative FET PET scan (i.e., 5 tumors with indifferent FET uptake comparable to the background activity, or FET uptake below background activity (photopenic defect) in 5 tumors). CONCLUSIONS With a robust interrater agreement, our findings are in line with previously reported findings regarding the T2-FLAIR mismatch sign. Additionally, the T2-FLAIR mismatch sign seems to be significantly related with a lack of increased FET uptake in PET, which may help to further characterize patients with that sign. Notwithstanding, the clinical relevance of this imaging constellation warrants further investigation.

OS9.6 Diagnosis of pseudoprogression using FET PET radiomics

BACKGROUND Radiomics derived from different imaging modalities is gaining increasing interest in the field of neuro-oncology. Besides MRI, amino acid PET radiomics may also improve the to date challenging, clinically relevant diagnostic problem of differentiating pseudoprogression (PsP) from tumor progression (TP). To this end, we here explored the potential of O-(2-[18F]fluoroethyl)-L-tyrosine (FET) PET radiomics to discriminate between PsP and TP. MATERIAL AND METHODS Thirty-five newly diagnosed IDH-wildtype glioblastoma patients with MRI findings suspicious for TP within 12 weeks after completion of chemoradiation with temozolomide underwent an additional dynamic FET PET scan. FET PET tumor volumes were segmented using a tumor-to-brain ratio (TBR) ≥ 1.6. The static PET parameters TBRmax and TBRmean, as well as the dynamic parameter time-to-peak (TTP), were calculated. For radiomics analysis, the number of datasets for model generation was increased using data augmentation techniques. Subsequently, 70 datasets were available for model generation. Prior to further processing, patients were randomly assigned to a discovery and a validation dataset in a ratio of 70/30, with balanced distribution of PsP and TP diagnoses. Forty-two radiomics features (4 shape-based, 6 first- and 32 second-order features) were obtained using the software LifeX (lifexsoft.org). Afterwards, a z-score transformation was performed for data normalization. For feature selection, recursive feature elimination using random forest regressors was performed. For the final model generation, the number of parameters was limited to three to avoid data overfitting. Different algorithms for model calculation were compared, and the diagnostic accuracy was assessed using leave-one-out cross-validation. Finally, the resulting models were applied to the validation dataset to evaluate model robustness. RESULTS Eighteen patients were diagnosed with TP, and 17 patients had PsP. Diagnoses were based on a neuropathological confirmation or clinicoradiological follow-up (26% and 74%, respectively). The diagnostic accuracy of the best single FET PET parameter was 75% (TBRmax). Combining TBRmax and TTP increased the diagnostic accuracy to 83%. Other combinations of static and dynamic FET PET parameters, however, did not further increase the accuracy. The highest diagnostic accuracy of 92% was achieved by a three-parameter model combining the FET PET parameter TTP with two radiomics features. The model demonstrated its robustness in the validation dataset with a diagnostic accuracy of 86%. CONCLUSION The results suggest that FET PET radiomics improves the diagnostic accuracy for discerning PsP and TP considerably. Given the clinical significance of differentiating PSP and TP, prospective multicenter studies are warranted. FUNDING Wilhelm-Sander Stiftung and the DAAD GERSS Program, Germany