Background
Magnetic resonance imaging (MRI) and amino acid positron-emission tomography (PET) of the brain contain a vast amount of structural and functional information that can be analyzed by ...machine learning algorithms and radiomics for the use of radiotherapy in patients with malignant brain tumors.
Methods
This study is based on comprehensive literature research on machine learning and radiomics analyses in neuroimaging and their potential application for radiotherapy in patients with malignant glioma or brain metastases.
Results
Feature-based radiomics and deep learning-based machine learning methods can be used to improve brain tumor diagnostics and automate various steps of radiotherapy planning. In glioma patients, important applications are the determination of WHO grade and molecular markers for integrated diagnosis in patients not eligible for biopsy or resection, automatic image segmentation for target volume planning, prediction of the location of tumor recurrence, and differentiation of pseudoprogression from actual tumor progression. In patients with brain metastases, radiomics is applied for additional detection of smaller brain metastases, accurate segmentation of multiple larger metastases, prediction of local response after radiosurgery, and differentiation of radiation injury from local brain metastasis relapse. Importantly, high diagnostic accuracies of 80–90% can be achieved by most approaches, despite a large variety in terms of applied imaging techniques and computational methods.
Conclusion
Clinical application of automated image analyses based on radiomics and artificial intelligence has a great potential for improving radiotherapy in patients with malignant brain tumors. However, a common problem associated with these techniques is the large variability and the lack of standardization of the methods applied.
Mutations in the isocitrate dehydrogenase (IDH mut) gene have gained paramount importance for the prognosis of glioma patients. To date, reliable techniques for a preoperative evaluation of IDH ...genotype remain scarce. Therefore, we investigated the potential of O-(2-
Ffluoroethyl)-L-tyrosine (FET) PET radiomics using textural features combined with static and dynamic parameters of FET uptake for noninvasive prediction of IDH genotype. Prior to surgery, 84 patients with newly diagnosed and untreated gliomas underwent FET PET using a standard scanner (15 of 56 patients with IDH mut) or a dedicated high-resolution hybrid PET/MR scanner (11 of 28 patients with IDH mut). Static, dynamic and textural parameters of FET uptake in the tumor area were evaluated. Diagnostic accuracy of the parameters was evaluated using the neuropathological result as reference. Additionally, FET PET and textural parameters were combined to further increase the diagnostic accuracy. The resulting models were validated using cross-validation. Independent of scanner type, the combination of standard PET parameters with textural features increased significantly diagnostic accuracy. The highest diagnostic accuracy of 93% for prediction of IDH genotype was achieved with the hybrid PET/MR scanner. Our findings suggest that the combination of conventional FET PET parameters with textural features provides important diagnostic information for the non-invasive prediction of the IDH genotype.
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-
F-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.
The number of positron-emission tomography (PET) tracers used to evaluate patients with brain tumors has increased substantially over the last years. For the management of patients with brain tumors, ...the most important indications are the delineation of tumor extent (e.g., for planning of resection or radiotherapy), the assessment of treatment response to systemic treatment options such as alkylating chemotherapy, and the differentiation of treatment-related changes (e.g., pseudoprogression or radiation necrosis) from tumor progression. Furthermore, newer PET imaging approaches aim to address the need for noninvasive assessment of tumoral immune cell infiltration and response to immunotherapies (e.g., T-cell imaging). This review summarizes the clinical value of the landscape of tracers that have been used in recent years for the above-mentioned indications and also provides an overview of promising newer tracers for this group of patients.
Abstract
Over the past decades, a variety of PET tracers have been used for the evaluation of patients with brain tumors. For clinical routine, the most important clinical indications for PET imaging ...in patients with brain tumors are the identification of neoplastic tissue including the delineation of tumor extent for the further diagnostic and therapeutic management (ie, biopsy, resection, or radiotherapy planning), the assessment of response to a certain anticancer therapy including its (predictive) effect on the patients’ outcome and the differentiation of treatment-related changes (eg, pseudoprogression and radiation necrosis) from tumor progression at follow-up. To serve medical professionals of all disciplines involved in the diagnosis and care of patients with brain tumors, this review summarizes the value of PET imaging for the latter-mentioned 3 clinically relevant indications in patients with glioma, meningioma, and brain metastases.
Purpose
Perfusion-weighted MRI (PWI) and O-(2-
18
Ffluoroethyl-)-
l
-tyrosine (
18
FFET) PET are both applied to discriminate tumor progression (TP) from treatment-related changes (TRC) in patients ...with suspected recurrent glioma. While the combination of both methods has been reported to improve the diagnostic accuracy, the performance of a sequential implementation has not been further investigated. Therefore, we retrospectively analyzed the diagnostic value of consecutive PWI and
18
FFET PET.
Methods
We evaluated 104 patients with WHO grade II–IV glioma and suspected TP on conventional MRI using PWI and dynamic
18
FFET PET. Leakage corrected maximum relative cerebral blood volumes (rCBV
max
) were obtained from dynamic susceptibility contrast PWI. Furthermore, we calculated static (i.e., maximum tumor to brain ratios; TBR
max
) and dynamic
18
FFET PET parameters (i.e., Slope). Definitive diagnoses were based on histopathology (
n
= 42) or clinico-radiological follow-up (
n
= 62). The diagnostic performance of PWI and
18
FFET PET parameters to differentiate TP from TRC was evaluated by analyzing receiver operating characteristic and area under the curve (AUC).
Results
Across all patients, the differentiation of TP from TRC using rCBV
max
or
18
FFET PET parameters was moderate (AUC = 0.69–0.75;
p
< 0.01). A rCBV
max
cutoff > 2.85 had a positive predictive value for TP of 100%, enabling a correct TP diagnosis in 44 patients. In the remaining 60 patients, combined static and dynamic
18
FFET PET parameters (TBR
max
, Slope) correctly discriminated TP and TRC in a significant 78% of patients, increasing the overall accuracy to 87%. A subgroup analysis of isocitrate dehydrogenase (IDH) mutant tumors indicated a superior performance of PWI to
18
FFET PET (AUC = 0.8/< 0.62,
p
< 0.01/≥ 0.3).
Conclusion
While marked hyperperfusion on PWI indicated TP,
18
FFET PET proved beneficial to discriminate TP from TRC when PWI remained inconclusive. Thus, our results highlight the clinical value of sequential use of PWI and
18
FFET PET, allowing an economical use of diagnostic methods. The impact of an
IDH
mutation needs further investigation.
Quantitative values derived from PET brain images are of high interest for neuroscientific applications. Insufficient DT correction (DTC) can lead to a systematic bias of the output parameters ...obtained by a detailed analysis of the time activity curves (TACs). The DTC method currently used for the Siemens 3T MR BrainPET insert is global, i.e., differences in DT losses between detector blocks are not considered, leading to inaccurate DTC and, consequently, to inaccurate measurements masked by a bias. However, following careful evaluation with phantom measurements, a new block-pairwise DTC method has demonstrated a higher degree of accuracy compared to the global DTC method.
Differences between the global and the block-pairwise DTC method were studied in this work by applying several radioactive tracers. We evaluated the impact on 11CABP688, O-(2-18Ffluoroethyl)-L-tyrosine (FET), and 15OH2O TACs.
For 11CABP688, a relevant bias of between -0.0034 and -0.0053 ml/ (cm3 • min) was found in all studied brain regions for the volume of distribution (VT) when using the current global DTC method. For 18FFET-PET, differences of up to 10% were observed in the tumor-to-brain ratio (TBRmax), these differences depend on the radial distance of the maximum from the PET isocenter. For 15OH2O, differences between +4% and -7% were observed in the GM region. Average biases of -4.58%, -3.2%, and -1.2% for the regional cerebral blood flow (CBF (K1)), the rate constant k2, and the volume of distribution VT were observed, respectively. Conversely, in the white matter region, average biases of -4.9%, -7.0%, and 3.8% were observed for CBF (K1), k2, and VT, respectively.
The bias introduced by the global DTC method leads to an overestimation in the studied quantitative parameters for all applications compared to the block-pairwise method.
The observed differences between the two DTC methods are particularly relevant for research applications in neuroscientific studies as they affect the accuracy of quantitative Brain PET images.
The aim of this study was to investigate the potential of combined textural feature analysis of contrast-enhanced MRI (CE-MRI) and static O-(2-18Ffluoroethyl)-L-tyrosine (FET) PET for the ...differentiation between local recurrent brain metastasis and radiation injury since CE-MRI often remains inconclusive.
Fifty-two patients with new or progressive contrast-enhancing brain lesions on MRI after radiotherapy (predominantly stereotactic radiosurgery) of brain metastases were additionally investigated using FET PET. Based on histology (n = 19) or clinicoradiological follow-up (n = 33), local recurrent brain metastases were diagnosed in 21 patients (40%) and radiation injury in 31 patients (60%). Forty-two textural features were calculated on both unfiltered and filtered CE-MRI and summed FET PET images (20–40 min p.i.), using the software LIFEx. After feature selection, logistic regression models using a maximum of five features to avoid overfitting were calculated for each imaging modality separately and for the combined FET PET/MRI features. The resulting models were validated using cross-validation. Diagnostic accuracies were calculated for each imaging modality separately as well as for the combined model.
For the differentiation between radiation injury and recurrence of brain metastasis, textural features extracted from CE-MRI had a diagnostic accuracy of 81% (sensitivity, 67%; specificity, 90%). FET PET textural features revealed a slightly higher diagnostic accuracy of 83% (sensitivity, 88%; specificity, 75%). However, the highest diagnostic accuracy was obtained when combining CE-MRI and FET PET features (accuracy, 89%; sensitivity, 85%; specificity, 96%).
Our findings suggest that combined FET PET/CE-MRI radiomics using textural feature analysis offers a great potential to contribute significantly to the management of patients with brain metastases.
•Differentiation between brain metastasis recurrence and radiation injury is of high clinical importance.•Differentiation based on contrast-enhanced conventional MRI is often inconclusive.•Radiomics and hybrid amino acid PET/MR imaging are increasingly gaining attention in Neuro-Oncology.•We investigated the potential of combined PET/MRI radiomics analysis using MRI and FET PET in patients with brain metastases.•Combined PET/MRI radiomics allows the differentiation of brain metastasis recurrence from radiation injury with high accuracy.
The diagnostic potential of PET using the amino acid analogue O-(2-18Ffluoroethyl)-L-tyrosine (18FFET) in brain tumor diagnostics has been proven in many studies during the last two decades and is ...still the subject of multiple studies every year. In addition to standard magnetic resonance imaging (MRI), positron emission tomography (PET) using 18FFET provides important diagnostic data concerning brain tumor delineation, therapy planning, treatment monitoring, and improved differentiation between treatment-related changes and tumor recurrence. The pharmacokinetics, uptake mechanisms and metabolism have been well described in various preclinical studies. The accumulation of 18FFET in most benign lesions and healthy brain tissue has been shown to be low, thus providing a high contrast between tumor tissue and benign tissue alterations. Based on logistic advantages of F-18 labelling and convincing clinical results, 18FFET has widely replaced short lived amino acid tracers such as L-11Cmethyl-methionine (11CMET) in many centers across Western Europe. This review summarizes the basic knowledge on 18FFET and its contribution to the care of patients with brain tumors. In particular, recent studies about specificity, possible pitfalls, and the utility of 18FFET PET in tumor grading and prognostication regarding the revised WHO classification of brain tumors are addressed.