Analysis of heterogeneity in T2-weighted MR images can differentiate pseudoprogression from progression in glioblastoma

Synergistic Effects of Cryotherapy and Radiotherapy in Glioblastoma Treatment: Evidence from a Murine Model

BACKGROUND/OBJECTIVES

Cryotherapy involves the insertion of cryoprobes into tumors to induce cell destruction through exposure to extremely low temperatures over several minutes. This localized treatment modality may enhance the efficacy of established therapies, such as radiotherapy, particularly for glioblastomas. Our study aimed to provide proof-of-concept for the efficacy of combining cryotherapy and radiotherapy in the treatment of subcutaneous murine brain tumors (GL-261) in immunocompetent C57BL/6 mice.

METHODS

Tumor growth, survival and response were evaluated using MRI and histological analysis.

RESULTS

Partial cryotherapy alone showed no therapeutic efficacy. However, combining cryotherapy with radiotherapy significantly potentiated treatment outcomes. A statistically significant survival benefit was observed in the combined therapy group compared to control, cryotherapy and radiotherapy groups. Notably, 40% of mice receiving the combined treatment exhibited complete responses, with no detectable tumor cells on MRI or histological analysis. Furthermore, MRI-based monitoring revealed that the Apparent Diffusion Coefficient (ADC) map could predict complete response 14 days post-treatment, unlike caliper-based measurements.

CONCLUSIONS

These findings suggest that cryotherapy may enhance radiotherapy efficacy, resulting in complete tumor regression in 4 out of 10 cases. ADC distribution may serve as a predictive marker for therapeutic response. However, given the limitations of the model, further studies in orthotopic models are needed to validate these findings and assess their clinical relevance.

Magnetic Resonance Imaging Sequences and Technologies in Adaptive Radiation Therapy

Radiation therapy is essential in both curative and palliative treatments for most cancers. However, traditional radiation therapy workflows using computed tomography (CT) simulation-based planning and cone beam CT image guidance face several technical challenges, including limited tumor visibility and daily fluctuations in tumor size and shape. Magnetic resonance imaging (MRI) guided linear accelerators (MR-Linacs) address these issues by enabling precise visualization of changes in tumor position and morphologic changes, as well as changes in surrounding organs-at-risk. The hybrid MR-Linac systems combine MRI with linear accelerator technology, offering enhanced soft tissue visualization and the potential for adaptive radiation therapy (ART). This narrative review provides a comprehensive introduction to MR guided ART technologies, covering protocol optimization with appropriate pulse sequence selection and parameter adjustment for clinical implementations on various disease sites.

Spatiotemporal sphingosine-1-phosphate receptor 3 expression within the cerebral vasculature after ischemic stroke

Sphingosine-1-phosphate receptors (S1PRs) are promising therapeutic targets in cardiovascular disease, including ischemic stroke. However, important spatiotemporal information for alterations of S1PR expression is lacking. Here, we investigated the role of S1PR3 in ischemic stroke in rodent models and patient samples. We show that S1PR3 is acutely upregulated in perilesional reactive astrocytes after stroke, and that stroke volume and behavioral deficits are improved in mice lacking S1PR3. Further, we find that administration of an S1PR3 antagonist at 4-h post-stroke, but not at later timepoints, improves stroke outcome. Lastly, we observed higher plasma S1PR3 concentrations in experimental stroke and in patients with ischemic stroke. Together, our results establish S1PR3 as a potential drug target and biomarker in ischemic stroke.

Glioblastoma and radiotherapy: A multicenter AI study for Survival Predictions from MRI (GRASP study)

BACKGROUND

The aim was to predict survival of glioblastoma at 8 months after radiotherapy (a period allowing for completing a typical course of adjuvant temozolomide), by applying deep learning to the first brain MRI after radiotherapy completion.

METHODS

Retrospective and prospective data were collected from 206 consecutive glioblastoma, isocitrate dehydrogenase -wildtype patients diagnosed between March 2014 and February 2022 across 11 UK centers. Models were trained on 158 retrospective patients from 3 centers. Holdout test sets were retrospective (n = 19; internal validation), and prospective (n = 29; external validation from 8 distinct centers). Neural network branches for T2-weighted and contrast-enhanced T1-weighted inputs were concatenated to predict survival. A nonimaging branch (demographics/MGMT/treatment data) was also combined with the imaging model. We investigated the influence of individual MR sequences; nonimaging features; and weighted dense blocks pretrained for abnormality detection.

RESULTS

The imaging model outperformed the nonimaging model in all test sets (area under the receiver-operating characteristic curve, AUC P = .038) and performed similarly to a combined imaging/nonimaging model (P > .05). Imaging, nonimaging, and combined models applied to amalgamated test sets gave AUCs of 0.93, 0.79, and 0.91. Initializing the imaging model with pretrained weights from 10 000s of brain MRIs improved performance considerably (amalgamated test sets without pretraining 0.64; P = .003).

CONCLUSIONS

A deep learning model using MRI images after radiotherapy reliably and accurately determined survival of glioblastoma. The model serves as a prognostic biomarker identifying patients who will not survive beyond a typical course of adjuvant temozolomide, thereby stratifying patients into those who might require early second-line or clinical trial treatment.

Magnetic resonance neurography in spinal cord injury: Imaging findings and clinical significance

BACKGROUND AND PURPOSE

It is unknown whether changes to the peripheral nervous system following spinal cord injury (SCI) are relevant for functional recovery or the development of neuropathic pain below the level of injury. Magnetic resonance neurography (MRN) at 3 T allows detection and localization of structural and functional nerve damage. This study aimed to combine MRN and clinical assessments in individuals with chronic SCI and nondisabled controls.

METHODS

Twenty participants with chronic SCI and 20 controls matched for gender, age, and body mass index underwent MRN of the L5 dorsal root ganglia (DRG) and the sciatic nerve. DRG volume, sciatic nerve mean cross-sectional area (CSA), fascicular lesion load, and fractional anisotropy (FA), a marker for functional nerve integrity, were calculated. Results were correlated with clinical assessments and nerve conduction studies.

RESULTS

Sciatic nerve CSA and lesion load were higher (21.29 ± 5.82 mm2 vs. 14.08 ± 4.62 mm2 , p < 0.001; and 8.70 ± 7.47% vs. 3.60 ± 2.45%, p < 0.001) in individuals with SCI compared to controls, whereas FA was lower (0.55 ± 0.11 vs. 0.63 ± 0.08, p = 0.022). DRG volumes were larger in individuals with SCI who suffered from neuropathic pain compared to those without neuropathic pain (223.7 ± 53.08 mm3 vs. 159.7 ± 55.66 mm3 , p = 0.043). Sciatic MRN parameters correlated with electrophysiological results but did not correlate with the extent of myelopathy or clinical severity of SCI.

CONCLUSIONS

Individuals with chronic SCI are subject to a decline of structural peripheral nerve integrity that may occur independently from the clinical severity of SCI. Larger volumes of DRG in SCI with neuropathic pain support existing evidence from animal studies on SCI-related neuropathic pain.

Glioblastoma Pseudoprogression Discrimination Using Multiparametric Magnetic Resonance Imaging, Principal Component Analysis, and Supervised and Unsupervised Machine Learning

BACKGROUND

One of the most frequent phenomena in the follow-up of glioblastoma is pseudoprogression, present in up to half of cases. The clinical usefulness of discriminating this phenomenon through magnetic resonance imaging and nuclear medicine has not yet been standardized; in this study, we used machine learning on multiparametric magnetic resonance imaging to explore discriminators of this phenomenon.

METHODS

For the study, 30 patients diagnosed with IDH wild-type glioblastoma operated on at both study centers in 2011-2020 were selected; 15 patients corresponded to early tumor progression and 15 patients to pseudoprogression. Using unsupervised learning, the number of clusters and tumor segmentation was recorded using gap-stat and k-means method, adjusting to voxel adjacency. In a second phase, a class prediction was carried out with a multinomial logistic regression supervised learning method; the outcome variables were the percentage of assignment, class overrepresentation, and degree of voxel adjacency.

RESULTS

Unsupervised learning of the tumor in its diagnosis shows up to 14 well-differentiated tumor areas. In the supervised learning phase, there is a higher percentage of assigned classes (P < 0.01), less overrepresentation of classes (P < 0.01), and greater adjacency (55% vs. 33%) in cases of true tumor progression compared with pseudoprogression.

CONCLUSIONS

True tumor progression preserves the multidimensional characteristics of the basal tumor at the voxel and region of interest level, resulting in a characteristic differential pattern when supervised learning is used.

Unveiling the Intricacies: A Comprehensive Review of Magnetic Resonance Imaging (MRI) Assessment of T2-Weighted Hyperintensities in the Neuroimaging Landscape

T2-weighted hyperintensities in neuroimaging represent areas of heightened signal intensity on magnetic resonance imaging (MRI) scans, holding crucial importance in neuroimaging. This comprehensive review explores the T2-weighted hyperintensities, providing insights into their definition, characteristics, clinical relevance, and underlying causes. It highlights the significance of these hyperintensities as sensitive markers for neurological disorders, including multiple sclerosis, vascular dementia, and brain tumors. The review also delves into advanced neuroimaging techniques, such as susceptibility-weighted and diffusion tensor imaging, and the application of artificial intelligence and machine learning in hyperintensities analysis. Furthermore, it outlines the challenges and pitfalls associated with their assessment and emphasizes the importance of standardized protocols and a multidisciplinary approach. The review discusses future directions for research and clinical practice, including the development of biomarkers, personalized medicine, and enhanced imaging techniques. Ultimately, the review underscores the profound impact of T2-weighted hyperintensities in shaping the landscape of neurological diagnosis, prognosis, and treatment, contributing to a deeper understanding of complex neurological conditions and guiding more informed and effective patient care.

Artificial Intelligence Analysis Using MRI and PET Imaging in Gliomas: A Narrative Review

The lack of early detection and a high rate of recurrence/progression after surgery are defined as the most common causes of a very poor prognosis of Gliomas. The developments of quantification systems with special regards to artificial intelligence (AI) on medical images (CT, MRI, PET) are under evaluation in the clinical and research context in view of several applications providing different information related to the reconstruction of imaging, the segmentation of tissues acquired, the selection of features, and the proper data analyses. Different approaches of AI have been proposed as the machine and deep learning, which utilize artificial neural networks inspired by neuronal architectures. In addition, new systems have been developed using AI techniques to offer suggestions or make decisions in medical diagnosis, emulating the judgment of radiologist experts. The potential clinical role of AI focuses on the prediction of disease progression in more aggressive forms in gliomas, differential diagnosis (pseudoprogression vs. proper progression), and the follow-up of aggressive gliomas. This narrative Review will focus on the available applications of AI in brain tumor diagnosis, mainly related to malignant gliomas, with particular attention to the postoperative application of MRI and PET imaging, considering the current state of technical approach and the evaluation after treatment (including surgery, radiotherapy/chemotherapy, and prognostic stratification).

Convolutional Neural Networks for Glioma Segmentation and Prognosis: A Systematic Review

Deep learning (DL) is poised to redefine the way medical images are processed and analyzed. Convolutional neural networks (CNNs), a specific type of DL architecture, are exceptional for high-throughput processing, allowing for the effective extraction of relevant diagnostic patterns from large volumes of complex visual data. This technology has garnered substantial interest in the field of neuro-oncology as a promising tool to enhance medical imaging throughput and analysis. A multitude of methods harnessing MRI-based CNNs have been proposed for brain tumor segmentation, classification, and prognosis prediction. They are often applied to gliomas, the most common primary brain cancer, to classify subtypes with the goal of guiding therapy decisions. Additionally, the difficulty of repeating brain biopsies to evaluate treatment response in the setting of often confusing imaging findings provides a unique niche for CNNs to help distinguish the treatment response to gliomas. For example, glioblastoma, the most aggressive type of brain cancer, can grow due to poor treatment response, can appear to grow acutely due to treatment-related inflammation as the tumor dies (pseudo-progression), or falsely appear to be regrowing after treatment as a result of brain damage from radiation (radiation necrosis). CNNs are being applied to separate this diagnostic dilemma. This review provides a detailed synthesis of recent DL methods and applications for intratumor segmentation, glioma classification, and prognosis prediction. Furthermore, this review discusses the future direction of MRI-based CNN in the field of neuro-oncology and challenges in model interpretability, data availability, and computation efficiency.

Advanced Magnetic Resonance Imaging in the Evaluation of Treated Glioblastoma: A Pictorial Essay

MRI plays a key role in the evaluation of post-treatment changes, both in the immediate post-operative period and during follow-up. There are many different treatment's lines and many different neuroradiological findings according to the treatment chosen and the clinical timepoint at which MRI is performed. Structural MRI is often insufficient to correctly interpret and define treatment-related changes. For that, advanced MRI modalities, including perfusion and permeability imaging, diffusion tensor imaging, and magnetic resonance spectroscopy, are increasingly utilized in clinical practice to characterize treatment effects more comprehensively. This article aims to provide an overview of the role of advanced MRI modalities in the evaluation of treated glioblastomas. For a didactic purpose, we choose to divide the treatment history in three main timepoints: post-surgery, during Stupp (first-line treatment) and at recurrence (second-line treatment). For each, a brief introduction, a temporal subdivision (when useful) or a specific drug-related paragraph were provided. Finally, the current trends and application of radiomics and artificial intelligence (AI) in the evaluation of treated GB have been outlined.

Artificial intelligence in the radiomic analysis of glioblastomas: A review, taxonomy, and perspective

Radiological imaging techniques, including magnetic resonance imaging (MRI) and positron emission tomography (PET), are the standard-of-care non-invasive diagnostic approaches widely applied in neuro-oncology. Unfortunately, accurate interpretation of radiological imaging data is constantly challenged by the indistinguishable radiological image features shared by different pathological changes associated with tumor progression and/or various therapeutic interventions. In recent years, machine learning (ML)-based artificial intelligence (AI) technology has been widely applied in medical image processing and bioinformatics due to its advantages in implicit image feature extraction and integrative data analysis. Despite its recent rapid development, ML technology still faces many hurdles for its broader applications in neuro-oncological radiomic analysis, such as lack of large accessible standardized real patient radiomic brain tumor data of all kinds and reliable predictions on tumor response upon various treatments. Therefore, understanding ML-based AI technologies is critically important to help us address the skyrocketing demands of neuro-oncology clinical deployments. Here, we provide an overview on the latest advancements in ML techniques for brain tumor radiomic analysis, emphasizing proprietary and public dataset preparation and state-of-the-art ML models for brain tumor diagnosis, classifications (e.g., primary and secondary tumors), discriminations between treatment effects (pseudoprogression, radiation necrosis) and true progression, survival prediction, inflammation, and identification of brain tumor biomarkers. We also compare the key features of ML models in the realm of neuroradiology with ML models employed in other medical imaging fields and discuss open research challenges and directions for future work in this nascent precision medicine area.

Advances in the In Vivo Quantitative and Qualitative Imaging Characterization of Gliomas

Gliomas are the most common and aggressive intra-axial primary tumours of the central nervous system (CNS), arising from glial cells [...]

Machine learning imaging applications in the differentiation of true tumour progression from treatment‐related effects in brain tumours: A systematic review and meta‐analysis

Introduction

Chemotherapy and radiotherapy can produce treatment‐related effects, which may mimic tumour progression. Advances in Artificial Intelligence (AI) offer the potential to provide a more consistent approach of diagnosis with improved accuracy. The aim of this study was to determine the efficacy of machine learning models to differentiate treatment‐related effects (TRE), consisting of pseudoprogression (PsP) and radiation necrosis (RN), and true tumour progression (TTP).

Methods

The systematic review was conducted in accordance with PRISMA‐DTA guidelines. Searches were performed on PubMed, Scopus, Embase, Medline (Ovid) and ProQuest databases. Quality was assessed according to the PROBAST and CLAIM criteria. There were 25 original full‐text journal articles eligible for inclusion.

Results

For gliomas: PsP versus TTP (16 studies, highest AUC = 0.98), RN versus TTP (4 studies, highest AUC = 0.9988) and TRE versus TTP (3 studies, highest AUC = 0.94). For metastasis: RN vs. TTP (2 studies, highest AUC = 0.81). A meta‐analysis was performed on 9 studies in the gliomas PsP versus TTP group using STATA. The meta‐analysis reported a high sensitivity of 95.2% (95%CI: 86.6–98.4%) and specificity of 82.4% (95%CI: 67.0–91.6%).

Conclusion

TRE can be distinguished from TTP with good performance using machine learning‐based imaging models. There remain issues with the quality of articles and the integration of models into clinical practice. Future studies should focus on the external validation of models and utilize standardized criteria such as CLAIM to allow for consistency in reporting.

High-Grade Glioma Treatment Response Monitoring Biomarkers: A Position Statement on the Evidence Supporting the Use of Advanced MRI Techniques in the Clinic, and the Latest Bench-to-Bedside Developments. Part 1: Perfusion and Diffusion Techniques

Objective

Summarize evidence for use of advanced MRI techniques as monitoring biomarkers in the clinic, and highlight the latest bench-to-bedside developments.

Methods

Experts in advanced MRI techniques applied to high-grade glioma treatment response assessment convened through a European framework. Current evidence regarding the potential for monitoring biomarkers in adult high-grade glioma is reviewed, and individual modalities of perfusion, permeability, and microstructure imaging are discussed (in Part 1 of two). In Part 2, we discuss modalities related to metabolism and/or chemical composition, appraise the clinic readiness of the individual modalities, and consider post-processing methodologies involving the combination of MRI approaches (multiparametric imaging) or machine learning (radiomics).

Results

High-grade glioma vasculature exhibits increased perfusion, blood volume, and permeability compared with normal brain tissue. Measures of cerebral blood volume derived from dynamic susceptibility contrast-enhanced MRI have consistently provided information about brain tumor growth and response to treatment; it is the most clinically validated advanced technique. Clinical studies have proven the potential of dynamic contrast-enhanced MRI for distinguishing post-treatment related effects from recurrence, but the optimal acquisition protocol, mode of analysis, parameter of highest diagnostic value, and optimal cut-off points remain to be established. Arterial spin labeling techniques do not require the injection of a contrast agent, and repeated measurements of cerebral blood flow can be performed. The absence of potential gadolinium deposition effects allows widespread use in pediatric patients and those with impaired renal function. More data are necessary to establish clinical validity as monitoring biomarkers. Diffusion-weighted imaging, apparent diffusion coefficient analysis, diffusion tensor or kurtosis imaging, intravoxel incoherent motion, and other microstructural modeling approaches also allow treatment response assessment; more robust data are required to validate these alone or when applied to post-processing methodologies.

Conclusion

Considerable progress has been made in the development of these monitoring biomarkers. Many techniques are in their infancy, whereas others have generated a larger body of evidence for clinical application.

High-Grade Glioma Treatment Response Monitoring Biomarkers: A Position Statement on the Evidence Supporting the Use of Advanced MRI Techniques in the Clinic, and the Latest Bench-to-Bedside Developments. Part 2: Spectroscopy, Chemical Exchange Saturation, Multiparametric Imaging, and Radiomics

Objective

To summarize evidence for use of advanced MRI techniques as monitoring biomarkers in the clinic, and to highlight the latest bench-to-bedside developments.

Methods

The current evidence regarding the potential for monitoring biomarkers was reviewed and individual modalities of metabolism and/or chemical composition imaging discussed. Perfusion, permeability, and microstructure imaging were similarly analyzed in Part 1 of this two-part review article and are valuable reading as background to this article. We appraise the clinic readiness of all the individual modalities and consider methodologies involving machine learning (radiomics) and the combination of MRI approaches (multiparametric imaging).

Results

The biochemical composition of high-grade gliomas is markedly different from healthy brain tissue. Magnetic resonance spectroscopy allows the simultaneous acquisition of an array of metabolic alterations, with choline-based ratios appearing to be consistently discriminatory in treatment response assessment, although challenges remain despite this being a mature technique. Promising directions relate to ultra-high field strengths, 2-hydroxyglutarate analysis, and the use of non-proton nuclei. Labile protons on endogenous proteins can be selectively targeted with chemical exchange saturation transfer to give high resolution images. The body of evidence for clinical application of amide proton transfer imaging has been building for a decade, but more evidence is required to confirm chemical exchange saturation transfer use as a monitoring biomarker. Multiparametric methodologies, including the incorporation of nuclear medicine techniques, combine probes measuring different tumor properties. Although potentially synergistic, the limitations of each individual modality also can be compounded, particularly in the absence of standardization. Machine learning requires large datasets with high-quality annotation; there is currently low-level evidence for monitoring biomarker clinical application.

Conclusion

Advanced MRI techniques show huge promise in treatment response assessment. The clinical readiness analysis highlights that most monitoring biomarkers require standardized international consensus guidelines, with more facilitation regarding technique implementation and reporting in the clinic.

Treatment Response and Prognosis Evaluation in High-Grade Glioma: An Imaging Review Based on MRI

In recent years, the development of advanced magnetic resonance imaging (MRI) technology and machine learning (ML) have created new tools for evaluating treatment response and prognosis of patients with high-grade gliomas (HGG); however, patient prognosis has not improved significantly. This is mainly due to the heterogeneity between and within HGG tumors, resulting in standard treatment methods not benefitting all patients. Moreover, the survival of patients with HGG is not only related to tumor cells, but also to noncancer cells in the tumor microenvironment (TME). Therefore, during preoperative diagnosis and follow-up treatment of patients with HGG, noninvasive imaging markers are needed to characterize intratumoral heterogeneity, and then to evaluate treatment response and predict prognosis, timeously adjust treatment strategies, and achieve individualized diagnosis and treatment. In this review, we summarize the research progress of conventional MRI, advanced MRI technology, and ML in evaluation of treatment response and prognosis of patients with HGG. We further discuss the significance of the TME in the prognosis of HGG patients, associate imaging features with the TME, indirectly reflecting the heterogeneity within the tumor, and shifting treatment strategies from tumor cells alone to systemic therapy of the TME, which may be a major development direction in the future. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY STAGE: 4.

Bibenzyl synthesis in Cannabis sativa L

This study focuses on the biosynthesis of a suite of specialized metabolites from Cannabis that are known as the 'bibenzyls'. In planta, bibenzyls accumulate in response to fungal infection and various other biotic stressors; however, it is their widely recognized anti-inflammatory properties in various animal cell models that have garnered recent therapeutic interest. We propose that these compounds are synthesized via a branch point from the core phenylpropanoid pathway in Cannabis, in a three-step sequence. First, various hydroxycinnamic acids are esterified to acyl-coenzyme A (CoA) by a member of the 4-coumarate-CoA ligase family (Cs4CL4). Next, these CoA esters are reduced by two double-bond reductases (CsDBR2 and CsDBR3) that form their corresponding dihydro-CoA derivatives from preferred substrates. Finally, the bibenzyl backbone is completed by a polyketide synthase that specifically condenses malonyl-CoA with these dihydro-hydroxycinnamoyl-CoA derivatives to form two bibenzyl scaffolds: dihydropiceatannol and dihydroresveratrol. Structural determination of this 'bibenzyl synthase' enzyme (CsBBS2) indicates that a narrowing of the hydrophobic pocket surrounding the active site evolved to sterically favor the non-canonical and more flexible dihydro-hydroxycinnamoyl-CoA substrates in comparison with their oxidized relatives. Accordingly, three point mutations that were introduced into CsBBS2 proved sufficient to restore some enzymatic activity with an oxidized substrate, in vitro. Together, the identification of this set of Cannabis enzymes provides a valuable contribution to the growing 'parts prospecting' inventory that supports the rational metabolic engineering of natural product therapeutics.

Imaging Biomarkers of Glioblastoma Treatment Response: A Systematic Review and Meta-Analysis of Recent Machine Learning Studies

OBJECTIVE

Monitoring biomarkers using machine learning (ML) may determine glioblastoma treatment response. We systematically reviewed quality and performance accuracy of recently published studies.

METHODS

Following Preferred Reporting Items for Systematic Reviews and Meta-Analysis: Diagnostic Test Accuracy, we extracted articles from MEDLINE, EMBASE and Cochrane Register between 09/2018-01/2021. Included study participants were adults with glioblastoma having undergone standard treatment (maximal resection, radiotherapy with concomitant and adjuvant temozolomide), and follow-up imaging to determine treatment response status (specifically, distinguishing progression/recurrence from progression/recurrence mimics, the target condition). Using Quality Assessment of Diagnostic Accuracy Studies Two/Checklist for Artificial Intelligence in Medical Imaging, we assessed bias risk and applicability concerns. We determined test set performance accuracy (sensitivity, specificity, precision, F1-score, balanced accuracy). We used a bivariate random-effect model to determine pooled sensitivity, specificity, area-under the receiver operator characteristic curve (ROC-AUC). Pooled measures of balanced accuracy, positive/negative likelihood ratios (PLR/NLR) and diagnostic odds ratio (DOR) were calculated. PROSPERO registered (CRD42021261965).

RESULTS

Eighteen studies were included (1335/384 patients for training/testing respectively). Small patient numbers, high bias risk, applicability concerns (particularly confounding in reference standard and patient selection) and low level of evidence, allow limited conclusions from studies. Ten studies (10/18, 56%) included in meta-analysis gave 0.769 (0.649-0.858) sensitivity [pooled (95% CI)]; 0.648 (0.749-0.532) specificity; 0.706 (0.623-0.779) balanced accuracy; 2.220 (1.560-3.140) PLR; 0.366 (0.213-0.572) NLR; 6.670 (2.800-13.500) DOR; 0.765 ROC-AUC.

CONCLUSION

ML models using MRI features to distinguish between progression and mimics appear to demonstrate good diagnostic performance. However, study quality and design require improvement.

Effective Detection and Monitoring of Glioma Using [18F]FPIA PET Imaging

BACKGROUND

Reprogrammed cellular metabolism is a cancer hallmark. In addition to increased glycolysis, the oxidation of acetate in the citric acid cycle is another common metabolic phenotype. We have recently developed a novel fluorine-18-labelled trimethylacetate-based radiotracer, [18F]fluoro-pivalic acid ([18F]FPIA), for imaging the transcellular flux of short-chain fatty acids, and investigated whether this radiotracer can be used for the detection of glioma growth.

METHODS

We evaluated the potential of [18F]FPIA PET to monitor tumor growth in orthotopic patient-derived (HSJD-GBM-001) and cell line-derived (U87, LN229) glioma xenografts, and also included [18F]FDG PET for comparison. We assessed proliferation (Ki-67) and the expression of lipid metabolism and transport proteins (CPT1, SLC22A2, SLC22A5, SLC25A20) by immunohistochemistry, along with etomoxir treatment to provide insights into [18F]FPIA uptake.

RESULTS

Longitudinal PET imaging showed gradual increase in [18F]FPIA uptake in orthotopic glioma models with disease progression (p < 0.0001), and high tumor-to-brain contrast compared to [18F]FDG (p < 0.0001). [18F]FPIA uptake correlated positively with Ki-67 (p < 0.01), SLC22A5 (p < 0.001) and SLC25A20 (p = 0.001), and negatively with CPT1 (p < 0.01) and SLC22A2 (p < 0.01). Etomoxir reduced [18F]FPIA uptake, which correlated with decreased Ki-67 (p < 0.05).

CONCLUSIONS

Our findings support the use of [18F]FPIA PET for the detection and longitudinal monitoring of glioma, showing a positive correlation with tumor proliferation, and suggest transcellular flux-mediated radiotracer uptake.

Machine learning-based radiomic evaluation of treatment response prediction in glioblastoma

AIM

To investigate machine learning based models combining clinical, radiomic, and molecular information to distinguish between early true progression (tPD) and pseudoprogression (psPD) in patients with glioblastoma.

MATERIALS AND METHODS

A retrospective analysis was undertaken of 76 patients (46 tPD, 30 psPD) with early enhancing disease following chemoradiotherapy for glioblastoma. Outcome was determined on follow-up until 6 months post-chemoradiotherapy. Models comprised clinical characteristics, O⁶-methylguanine-DNA methyltransferase (MGMT) promoter methylation status, and 307 quantitative imaging features extracted from enhancing disease and perilesional oedema masks on early post-chemoradiotherapy contrast-enhanced T1-weighted imaging, T2-weighted imaging (T2WI), and apparent diffusion coefficient (ADC) maps. Feature selection was performed within bootstrapped cross-validated recursive feature elimination with a random forest algorithm. Naive Bayes five-fold cross-validation was used to validate the final model.

RESULTS

Top selected features included age, MGMT promoter methylation status, two shape-based features from the enhancing disease mask, three radiomic features from the enhancing disease mask on ADC, and one radiomic feature from the perilesional oedema mask on T2WI. The final model had an area under the receiver operating characteristics curve (AUC) of 0.80, sensitivity 78.2%, specificity 66.7%, and accuracy of 73.7%.

CONCLUSION

Incorporating a machine learning-based approach using quantitative radiomic features from standard-of-care magnetic resonance imaging (MRI), in combination with clinical characteristics and MGMT promoter methylation status has a complementary effect and improves model performance for early prediction of glioblastoma treatment response.

MR-Guided Radiotherapy for Brain and Spine Tumors

MRI is the standard modality to assess anatomy and response to treatment in brain and spine tumors given its superb anatomic soft tissue contrast (e.g., T1 and T2) and numerous additional intrinsic contrast mechanisms that can be used to investigate physiology (e.g., diffusion, perfusion, spectroscopy). As such, hybrid MRI and radiotherapy (RT) devices hold unique promise for Magnetic Resonance guided Radiation Therapy (MRgRT). In the brain, MRgRT provides daily visualizations of evolving tumors that are not seen with cone beam CT guidance and cannot be fully characterized with occasional standalone MRI scans. Significant evolving anatomic changes during radiotherapy can be observed in patients with glioblastoma during the 6-week fractionated MRIgRT course. In this review, a case of rapidly changing symptomatic tumor is demonstrated for possible therapy adaptation. For stereotactic body RT of the spine, MRgRT acquires clear isotropic images of tumor in relation to spinal cord, cerebral spinal fluid, and nearby moving organs at risk such as bowel. This visualization allows for setup reassurance and the possibility of adaptive radiotherapy based on anatomy in difficult cases. A review of the literature for MR relaxometry, diffusion, perfusion, and spectroscopy during RT is also presented. These techniques are known to correlate with physiologic changes in the tumor such as cellularity, necrosis, and metabolism, and serve as early biomarkers of chemotherapy and RT response correlating with patient survival. While physiologic tumor investigations during RT have been limited by the feasibility and cost of obtaining frequent standalone MRIs, MRIgRT systems have enabled daily and widespread physiologic measurements. We demonstrate an example case of a poorly responding tumor on the 0.35 T MRIgRT system with relaxometry and diffusion measured several times per week. Future studies must elucidate which changes in MR-based physiologic metrics and at which timepoints best predict patient outcomes. This will lead to early treatment intensification for tumors identified to have the worst physiologic responses during RT in efforts to improve glioblastoma survival.

Unraveling response to temozolomide in preclinical GL261 glioblastoma with MRI/MRSI using radiomics and signal source extraction

Glioblastoma is the most frequent aggressive primary brain tumor amongst human adults. Its standard treatment involves chemotherapy, for which the drug temozolomide is a common choice. These are heterogeneous and variable tumors which might benefit from personalized, data-based therapy strategies, and for which there is room for improvement in therapy response follow-up, investigated with preclinical models. This study addresses a preclinical question that involves distinguishing between treated and control (untreated) mice bearing glioblastoma, using machine learning techniques, from magnetic resonance-based data in two modalities: MRI and MRSI. It aims to go beyond the comparison of methods for such discrimination to provide an analytical pipeline that could be used in subsequent human studies. This analytical pipeline is meant to be a usable and interpretable tool for the radiology expert in the hope that such interpretation helps revealing new insights about the problem itself. For that, we propose coupling source extraction-based and radiomics-based data transformations with feature selection. Special attention is paid to the generation of radiologist-friendly visual nosological representations of the analyzed tumors.

Glioblastoma post-operative imaging in neuro-oncology: current UK practice (GIN CUP study)

OBJECTIVES

MRI remains the preferred imaging investigation for glioblastoma. Appropriate and timely neuroimaging in the follow-up period is considered to be important in making management decisions. There is a paucity of evidence-based information in current UK, European and international guidelines regarding the optimal timing and type of neuroimaging following initial neurosurgical treatment. This study assessed the current imaging practices amongst UK neuro-oncology centres, thus providing baseline data and informing future practice.

METHODS

The lead neuro-oncologist, neuroradiologist and neurosurgeon from every UK neuro-oncology centre were invited to complete an online survey. Participants were asked about current and ideal imaging practices following initial treatment.

RESULTS

Ninety-two participants from all 31 neuro-oncology centres completed the survey (100% response rate). Most centres routinely performed an early post-operative MRI (87%, 27/31), whereas only a third performed a pre-radiotherapy MRI (32%, 10/31). The number and timing of scans routinely performed during adjuvant TMZ treatment varied widely between centres. At the end of the adjuvant period, most centres performed an MRI (71%, 22/31), followed by monitoring scans at 3 monthly intervals (81%, 25/31). Additional short-interval imaging was carried out in cases of possible pseudoprogression in most centres (71%, 22/31). Routine use of advanced imaging was infrequent; however, the addition of advanced sequences was the most popular suggestion for ideal imaging practice, followed by changes in the timing of EPMRI.

CONCLUSION

Variations in neuroimaging practices exist after initial glioblastoma treatment within the UK. Multicentre, longitudinal, prospective trials are needed to define the optimal imaging schedule for assessment.

KEY POINTS

• Variations in imaging practices exist in the frequency, timing and type of interval neuroimaging after initial treatment of glioblastoma within the UK. • Large, multicentre, longitudinal, prospective trials are needed to define the optimal imaging schedule for assessment.

Machine Learning Model to Predict Pseudoprogression Versus Progression in Glioblastoma Using MRI: A Multi-Institutional Study (KROG 18-07)

Simple Summary

Even after the introduction of a standard regimen consisting of concurrent chemoradiotherapy and adjuvant temozolomide, patients with glioblastoma multiforme mostly experience disease progression. Clinicians often encounter a situation where they need to distinguish progressive disease from pseudoprogression after treatment. We tried to investigate the feasibility of machine learning algorithm to distinguish pseudoprogression from progressive disease. In multi-institutional dataset, the developed machine learning model showed an acceptable performance. This algorithm involving MRI data and clinical features could help making decision during patients’ disease course. For the practical use, we calibrated the machine learning model to offer the probability of pseudoprogression to clinicians, then we constructed the web-based user interface to access the model.

Abstract

Some patients with glioblastoma show a worsening presentation in imaging after concurrent chemoradiation, even when they receive gross total resection. Previously, we showed the feasibility of a machine learning model to predict pseudoprogression (PsPD) versus progressive disease (PD) in glioblastoma patients. The previous model was based on the dataset from two institutions (termed as the Seoul National University Hospital (SNUH) dataset, N = 78). To test this model in a larger dataset, we collected cases from multiple institutions that raised the problem of PsPD vs. PD diagnosis in clinics (Korean Radiation Oncology Group (KROG) dataset, N = 104). The dataset was composed of brain MR images and clinical information. We tested the previous model in the KROG dataset; however, that model showed limited performance. After hyperparameter optimization, we developed a deep learning model based on the whole dataset (N = 182). The 10-fold cross validation revealed that the micro-average area under the precision-recall curve (AUPRC) was 0.86. The calibration model was constructed to estimate the interpretable probability directly from the model output. After calibration, the final model offers clinical probability in a web-user interface.

IDH1 mutation prediction using MR-based radiomics in glioblastoma: comparison between manual and fully automated deep learning-based approach of tumor segmentation

PURPOSE

This study aimed to determine whether MR-based radiomics of glioblastoma can predict the isocitrate dehydrogenase-1 (IDH1) mutation status and compare predictive performances between manual and fully automatic deep-learning segmentations.

METHOD

Forty-five glioblastoma patients with pretreatment T2-weighted MRIs were retrospectively evaluated. Manual segmentations of glioblastoma and peri-tumoral edema were trained via a deep neural network (V-Net). An independent external cohort of 137 glioblastoma patients from the Cancer Imaging Archive was also included (test set 1, n = 46; test set 2, n = 91). Test set 1-without known IDH1 status-was used to calculate dice similarity coefficients (DSC) between the two segmentation methods (manual & V-Net). From test set 2, all-relevant radiomic features were selected via a random forest-based wrapper algorithm for IDH1 prediction. Receiver operating characteristics (ROC) curves with areas under the curve (AUC) were plotted as performance metrics for both methods.

RESULTS

Among 136 patients (45 and 91 patients from our institution and test set 2, respectively), 17 patients (11.2 %) had IDH1 mutations. The mean DSC of test set 1 was 0.78 ± 0.14 (range, 0.34-0.94). A subset of 9 all-relevant features (8.4 %, 9/107) was selected. V-Net segmentation of the test set 2 yielded similar performance in predicting IDH1 mutation as compared to manual segmentation (V-Net AUC = 0.86 vs. manual AUC = 0.90). The optimal cut-point threshold of AUC yielded 86.8 % accuracy for manual segmentation and 75.8 % for V-Net segmentation.

CONCLUSIONS

V-Net showed robust segmentation capability of glioblastoma on T2-weighted MRI. All-relevant radiomics features from both segmentation methods yielded a similar performance in IDH1 prediction.

Machine learning and glioma imaging biomarkers

AIM

To review how machine learning (ML) is applied to imaging biomarkers in neuro-oncology, in particular for diagnosis, prognosis, and treatment response monitoring.

MATERIALS AND METHODS

The PubMed and MEDLINE databases were searched for articles published before September 2018 using relevant search terms. The search strategy focused on articles applying ML to high-grade glioma biomarkers for treatment response monitoring, prognosis, and prediction.

RESULTS

Magnetic resonance imaging (MRI) is typically used throughout the patient pathway because routine structural imaging provides detailed anatomical and pathological information and advanced techniques provide additional physiological detail. Using carefully chosen image features, ML is frequently used to allow accurate classification in a variety of scenarios. Rather than being chosen by human selection, ML also enables image features to be identified by an algorithm. Much research is applied to determining molecular profiles, histological tumour grade, and prognosis using MRI images acquired at the time that patients first present with a brain tumour. Differentiating a treatment response from a post-treatment-related effect using imaging is clinically important and also an area of active study (described here in one of two Special Issue publications dedicated to the application of ML in glioma imaging).

CONCLUSION

Although pioneering, most of the evidence is of a low level, having been obtained retrospectively and in single centres. Studies applying ML to build neuro-oncology monitoring biomarker models have yet to show an overall advantage over those using traditional statistical methods. Development and validation of ML models applied to neuro-oncology require large, well-annotated datasets, and therefore multidisciplinary and multi-centre collaborations are necessary.

Structural Nerve Remodeling at 3-T MR Neurography Differs between Painful and Painless Diabetic Polyneuropathy in Type 1 or 2 Diabetes

Background The pathophysiologic mechanisms underlying painful symptoms in diabetic polyneuropathy (DPN) are poorly understood. They may be associated with MRI characteristics, which have not yet been investigated. Purpose To investigate correlations between nerve structure, load and spatial distribution of nerve lesions, and pain in patients with DPN. Materials and Methods In this prospective single-center cross-sectional study, participants with type 1 or 2 diabetes volunteered between June 2015 and March 2018. Participants underwent 3-T MR neurography of the sciatic nerve with a T2-weighed fat-suppressed sequence, which was preceded by clinical and electrophysiologic tests. For group comparisons, analysis of variance or the Kruskal-Wallis test was performed depending on Gaussian or non-Gaussian distribution of data. Spearman correlation coefficients were calculated for correlation analysis. Results A total of 131 participants (mean age, 62 years ± 11 [standard deviation]; 82 men) with either type 1 (n = 45) or type 2 (n = 86) diabetes were evaluated with painful (n = 64), painless (n = 37), or no (n = 30) DPN. Participants who had painful diabetic neuropathy had a higher percentage of nerve lesions in the full nerve volume (15.2% ± 1.6) than did participants with nonpainful DPN (10.4% ± 1.7, P = .03) or no DPN (8.3% ± 1.7; P < .001). The amount and extension of T2-weighted hyperintense nerve lesions correlated positively with the neuropathy disability score (r = 0.37; 95% confidence interval [CI]: 0.21, 0.52; r = 0.37; 95% CI: 0.20, 0.52, respectively) and the neuropathy symptom score (r = 0.41; 95% CI: 0.25, 0.55; r = 0.34; 95% CI: 0.17, 0.49, respectively). Negative correlations were found for the tibial nerve conduction velocity (r = -0.23; 95% CI: -0.44, -0.01; r = -0.37; 95% CI: -0.55, -0.15, respectively). The cross-sectional area of the nerve was positively correlated with the neuropathy disability score (r = 0.23; 95% CI: 0.03, 0.36). Negative correlations were found for the tibial nerve conduction velocity (r = -0.24; 95% CI: -0.45, -0.01). Conclusion The amount and extension of T2-weighted hyperintense fascicular nerve lesions were greater in patients with painful diabetic neuropathy than in those with painless diabetic neuropathy. These results suggest that proximal fascicular damage is associated with the evolution of painful sensory symptoms in diabetic polyneuropathy. © RSNA, 2019 Online supplemental material is available for this article.

Conventional MRI radiomics in patients with suspected early- or pseudo-progression

Background

After radiochemotherapy, 30% of patients with early worsening MRI experience pseudoprogression (Psp) which is not distinguishable from early progression (EP). We aimed to assess the diagnostic value of radiomics in patients with suspected EP or Psp.

Methods

Radiomics features (RF) of 76 patients (53 EP and 23 Psp) retrospectively identified were extracted from conventional MRI based on four volumes-of-interest. Subjects were randomly assigned into training and validation groups. Classification model (EP versus Psp) consisted of a random forest algorithm after univariate filtering. Overall (OS) and progression-free survivals (PFS) were predicted using a semi-supervised principal component analysis, and forecasts were evaluated using C-index and integrated Brier scores (IBS).

Results

Using 11 RFs, radiomics classified patients with 75.0% and 76.0% accuracy, 81.6% and 94.1% sensitivity, 50.0% and 37.5% specificity, respectively, in training and validation phases. Addition of MGMT promoter status improved accuracy to 83% and 79.2%, and specificity to 63.6% and 75%. OS model included 14 RFs and stratified low- and high-risk patients both in the training (hazard ratio [HR], 3.63; P = .002) and the validation (HR, 3.76; P = .001) phases. Similarly, PFS model stratified patients during training (HR, 2.58; P = .005) and validation (HR, 3.58; P = .004) phases using 5 RF. OS and PFS forecasts had C-index of 0.65 and 0.69, and IBS of 0.122 and 0.147, respectively.

Conclusions

Conventional MRI radiomics has promising diagnostic value, especially when combined with MGMT promoter status, but with moderate specificity. In addition, our results suggest a potential for predicting OS and PFS.

Analysis of peritumoral hyperintensity on pre-operative T2-weighted MR images in glioblastoma: Additive prognostic value of Minkowski functionals

Objectives

The extent of peritumoral tumor cell infiltrations in glioblastoma contributes to poor prognosis. We aimed to assess additive prognostic value of Minkowski functionals in analyzing heterogeneity of peritumoral hyperintensity on T2WI in glioblastoma patients.

Methods

Clinical data (age, sex, extent of surgical resection), O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status and pre-operative T2WI of 113 pathologically confirmed glioblastoma patients (from our institution, n = 61; from the Cancer Imaging Archive, n = 52) were retrospectively reviewed. The patients were randomly grouped into a training set (n = 80) and a test set (n = 33). Peritumoral T2 hyperintensity was manually segmented and Minkowski functionals—a texture analysis method capturing heterogeneity of MR images—were computed as a function of 11 grayscale thresholds. The Cox proportional hazards models were fitted with clinical variables, Minkowski functionals features as well as both combined. The risk prediction performances of the Minkowski functionals and combined models were validated on a separate test dataset. The sex-specific survival difference of the entire cohort was analyzed according to MGMT methylation status via Kaplan-Meier survival curves.

Results

Thirty-three Minkowski features (11 area, 11 perimeter and 11 genus) for each patient were acquired giving a total of 3729 features. Cox regression models fitted with clinical data, Minkowski features, and both combined had incremental concordance indices of 0.577 (P = 0.02), 0.706 (P = 0.02) and 0.714 (P = 0.01) respectively. The prediction error rate of the combined model—having clinical and Minkowski features—was lower than that of Minkowski functionals model (0.135 and 0.161, respectively) when validated on a test dataset. No sex-specific survival difference was found according to MGMT methylation status (male, P = 0.2; female, P = 0.22).

Conclusions

Minkowski functionals features computed from peritumoral hyperintensity can capture heterogeneity of glioblastoma on T2WI and have additive prognostic value in predicting survival, demonstrating their potential in complementing currently available prognostic parameters.

Serial T2-Weighted Magnetic Resonance Images Acquired on a 1.5 Tesla Magnetic Resonance Linear Accelerator Reveal Radiomic Feature Variation in Organs at Risk: An Exploratory Analysis of Novel Metrics of Tissue Response in Prostate Cancer

"Delta-radiomics" investigates variations in quantitative image metrics over time and can yield important clinical information. We hypothesized that in patients undergoing active radiation therapy (RT) for prostate cancer (PCa), there would exist observable variation in the quantitative metrics that describe the T₂-weighted (T_2W) intensity histogram in the prostate and surrounding organs at risk (OAR) over time. We investigated the feasibility of acquisition and subsequent analysis of the delta-radiomic profiles of these regions of interest (ROI) in serial T_2W magnetic resonance (MR) images obtained on a 1.5 Tesla (T) Magnetic Resonance Linear Accelerator (MRL). Principally, we sought to illustrate the significance of longitudinal radiomic data acquisition for tissue response monitoring and provide a framework for future hypothesis driven research.

Patients with PCa undergoing treatment with RT were compiled from an ongoing prospective observational imaging trial using a 1.5 T MRL (NCT30500081). Contiguous axial slices of prostate parenchyma were contoured and temporally normalized to sections of Sartorius muscle which served as a control. Similarly, contiguous sections of rectal and bladder wall adjacent to the prostate were contoured and temporally normalized to regions of these organs further removed from the planning target volume (PTV). First order statistical descriptors of the T_2W intensity histogram were extracted and evaluated for changes over time using linear mixed effects regression modeling and post-hoc contrasts. Benjamini-Hochberg corrections were employed to reduce the effects of multiple testing and control for the false discovery rate (FDR).

Four patients with a median age of 69 comprised this exploratory cohort. One patient had low-risk disease, two had intermediate (one favorable, one unfavorable), and one had high risk disease. Three out of four patients underwent definitive radiation to 75.6 Gray (Gy) in 42 fractions and one received hypofractionated therapy to a total dose of 70 Gy over 28 fractions, and all received treatment on a conventional linear accelerator. The most significant acute toxicity event was grade 2 GU dysfunction observed in two patients. Follow up ranged from 1 month to 10 months post treatment, and no long-term complications were reported in patients who completed treatment at least one month prior. Bladder wall adjacent to the prostate demonstrated significant variation in the mean and median metric values after the first week of treatment. In addition, rectal wall adjacent to the prostate exhibited significant variation in the mean, median, and standard deviation metric values by the second week of treatment. No significant variation in any radiomic feature was observed in the Sartorius control.

This exploratory study is one of the earliest examining the delta-radiomic characteristics of the T_2W intensity histogram in OAR extracted from images acquired on a 1.5 T MRL in patients actively being treated with RT for PCa. We demonstrated a feasible approach to longitudinal radiomic data acquisition providing limitless opportunity for future research. Analysis of the delta-radiomic profiles in OAR revealed significant variation in metrics after only one week of RT in bladder and rectal wall adjacent to the prostate. These findings must be further investigated and validated with expanded data sets with long-term follow up and correlation to clinical outcomes including toxicity and tumor control.

Perfusion MRI in treatment evaluation of glioblastomas: Clinical relevance of current and future techniques

Treatment evaluation of patients with glioblastomas is important to aid in clinical decisions. Conventional MRI with contrast is currently the standard method, but unable to differentiate tumor progression from treatment-related effects. Pseudoprogression appears as new enhancement, and thus mimics tumor progression on conventional MRI. Contrarily, a decrease in enhancement or edema on conventional MRI during antiangiogenic treatment can be due to pseudoresponse and is not necessarily reflective of a favorable outcome. Neovascularization is a hallmark of tumor progression but not for posttherapeutic effects. Perfusion-weighted MRI provides a plethora of additional parameters that can help to identify this neovascularization. This review shows that perfusion MRI aids to identify tumor progression, pseudoprogression, and pseudoresponse. The review provides an overview of the most applicable perfusion MRI methods and their limitations. Finally, future developments and remaining challenges of perfusion MRI in treatment evaluation in neuro-oncology are discussed. Level of Evidence: 3 Technical Efficacy: Stage 4 J. Magn. Reson. Imaging 2019;49:11-22.

Pseudoprogression of brain tumors

This review describes the definition, incidence, clinical implications, and magnetic resonance imaging (MRI) findings of pseudoprogression of brain tumors, in particular, but not limited to, high-grade glioma. Pseudoprogression is an important clinical problem after brain tumor treatment, interfering not only with day-to-day patient care but also the execution and interpretation of clinical trials. Radiologically, pseudoprogression is defined as a new or enlarging area(s) of contrast agent enhancement, in the absence of true tumor growth, which subsides or stabilizes without a change in therapy. The clinical definitions of pseudoprogression have been quite variable, which may explain some of the differences in reported incidences, which range from 9-30%. Conventional structural MRI is insufficient for distinguishing pseudoprogression from true progressive disease, and advanced imaging is needed to obtain higher levels of diagnostic certainty. Perfusion MRI is the most widely used imaging technique to diagnose pseudoprogression and has high reported diagnostic accuracy. Diagnostic performance of MR spectroscopy (MRS) appears to be somewhat higher, but MRS is less suitable for the routine and universal application in brain tumor follow-up. The combination of MRS and diffusion-weighted imaging and/or perfusion MRI seems to be particularly powerful, with diagnostic accuracy reaching up to or even greater than 90%. While diagnostic performance can be high with appropriate implementation and interpretation, even a combination of techniques, however, does not provide 100% accuracy. It should also be noted that most studies to date are small, heterogeneous, and retrospective in nature. Future improvements in diagnostic accuracy can be expected with harmonization of acquisition and postprocessing, quantitative MRI and computer-aided diagnostic technology, and meticulous evaluation with clinical and pathological data.

LEVEL OF EVIDENCE

3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018.