Simulated clinical deployment of fully automatic deep learning for clinical prostate MRI assessment.

OBJECTIVES: To simulate clinical deployment, evaluate performance, and establish quality assurance of a deep learning algorithm (U-Net) for detection, localization, and segmentation of clinically significant prostate cancer (sPC), ISUP grade group ≥ 2, using bi-parametric MRI. METHODS: In 2017, 284 consecutive men in active surveillance, biopsy-naïve or pre-biopsied, received targeted and extended systematic MRI/transrectal US-fusion biopsy, after examination on a single MRI scanner (3 T). A prospective adjustment scheme was evaluated comparing the performance of the Prostate Imaging Reporting and Data System (PI-RADS) and U-Net using sensitivity, specificity, predictive values, and the Dice coefficient. RESULTS: In the 259 eligible men (median 64 [IQR 61-72] years), PI-RADS had a sensitivity of 98% [106/108]/84% [91/108] with a specificity of 17% [25/151]/58% [88/151], for thresholds at ≥ 3/≥ 4 respectively. U-Net using dynamic threshold adjustment had a sensitivity of 99% [107/108]/83% [90/108] (p > 0.99/> 0.99) with a specificity of 24% [36/151]/55% [83/151] (p > 0.99/> 0.99) for probability thresholds d3 and d4 emulating PI-RADS ≥ 3 and ≥ 4 decisions respectively, not statistically different from PI-RADS. Co-occurrence of a radiological PI-RADS ≥ 4 examination and U-Net ≥ d3 assessment significantly improved the positive predictive value from 59 to 63% (p = 0.03), on a per-patient basis. CONCLUSIONS: U-Net has similar performance to PI-RADS in simulated continued clinical use. Regular quality assurance should be implemented to ensure desired performance. KEY POINTS: • U-Net maintained similar diagnostic performance compared to radiological assessment of PI-RADS ≥ 4 when applied in a simulated clinical deployment. • Application of our proposed prospective dynamic calibration method successfully adjusted U-Net performance within acceptable limits of the PI-RADS reference over time, while not being limited to PI-RADS as a reference. • Simultaneous detection by U-Net and radiological assessment significantly improved the positive predictive value on a per-patient and per-lesion basis, while the negative predictive value remained unchanged.

Methylome analyses of three glioblastoma cohorts reveal chemotherapy sensitivity markers within DDR genes.

BACKGROUND: Gliomas evade current therapies through primary and acquired resistance and the effect of temozolomide is mainly restricted to methylguanin-O6-methyltransferase promoter (MGMT) promoter hypermethylated tumors. Further resistance markers are largely unknown and would help for better stratification. METHODS: Clinical data and methylation profiles from the NOA-08 (104, elderly glioblastoma) and the EORTC 26101 (297, glioblastoma) studies and 398 patients with glioblastoma from the Heidelberg Neuro-Oncology center have been analyzed focused on the predictive effect of DNA damage response (DDR) gene methylation. Candidate genes were validated in vitro. RESULTS: Twenty-eight glioblastoma 5'-cytosine-phosphat-guanine-3' (CpGs) from 17 DDR genes negatively correlated with expression and were used together with telomerase reverse transcriptase (TERT) promoter mutations in further analysis. CpG methylation of DDR genes shows highest association with the mesenchymal (MES) and receptor tyrosine kinase (RTK) II glioblastoma subgroup. MES tumors have lower tumor purity compared to RTK I and II subgroup tumors. CpG hypomethylation of DDR genes TP73 and PRPF19 correlated with worse patient survival in particular in MGMT promoter unmethylated tumors. TERT promoter mutation is most frequent in RTK I and II subtypes and associated with worse survival. Primary glioma cells show methylation patterns that resemble RTK I and II glioblastoma and long term established glioma cell lines do not match with glioblastoma subtypes. Silencing of selected resistance genes PRPF19 and TERT increase sensitivity to temozolomide in vitro. CONCLUSION: Hypomethylation of DDR genes and TERT promoter mutations is associated with worse tumor prognosis, dependent on the methylation cluster and MGMT promoter methylation status in IDH wild-type glioblastoma.

T2/FLAIR-mismatch sign for noninvasive detection of IDH-mutant 1p/19q non-codeleted gliomas: validity and pathophysiology.

BACKGROUND: This study aimed to assess the validity and pathophysiology of the T2/FLAIR-mismatch sign for noninvasive identification of isocitrate dehydrogenase (IDH)-mutant 1p/19q non-codeleted glioma. METHODS: Magnetic resonance imaging scans from 408 consecutive patients with newly diagnosed glioma (113 lower-grade gliomas and 295 glioblastomas) were evaluated for the presence of T2/FLAIR-mismatch sign by 2 independent reviewers. Sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were calculated to assess the performance of the T2/FLAIR-mismatch sign for identifying IDH-mutant 1p/19q non-codeleted tumors. An exploratory analysis of differences in contrast-enhancing tumor volumes, apparent diffusion coefficient (ADC) values, and relative cerebral blood volume (rCBV) values in IDH-mutant gliomas with versus without the presence of a T2/FLAIR-mismatch sign (as well as analysis of spatial differences within tumors with the presence of a T2/FLAIR-mismatch sign) was performed. RESULTS: The T2/FLAIR-mismatch sign was present in 12 cases with lower-grade glioma (10.6%), all of them being IDH-mutant 1p/19q non-codeleted tumors (sensitivity = 10.9%, specificity = 100%, PPV = 100%, NPV = 3.0%, accuracy = 13.3%). There was a substantial interrater agreement to identify the T2/FLAIR-mismatch sign (Cohen's kappa = 0.75 [95% CI, 0.57-0.93]). The T2/FLAIR-mismatch sign was not identified in any other molecular subgroup, including IDH-mutant glioblastoma cases (n = 5). IDH-mutant gliomas with a T2/FLAIR-mismatch sign showed significantly higher ADC (P < .0001) and lower rCBV values (P = .0123) as compared to IDH-mutant gliomas without a T2/FLAIR-mismatch sign. Moreover, in IDH-mutant gliomas with T2/FLAIR-mismatch sign the ADC values were significantly lower in the FLAIR-hyperintense rim as compared to the FLAIR-hypointense core of the tumor (P = .0005). CONCLUSIONS: This study confirms the high specificity of the T2/FLAIR-mismatch sign for noninvasive identification of IDH-mutant 1p/19q non-codeleted gliomas; however, sensitivity is low and applicability is limited to lower-grade gliomas. Whether the higher ADC and lower rCBV values in IDH-mutant gliomas with a T2/FLAIR-mismatch sign (as compared to those without) translate into a measurable prognostic effect requires investigation in future studies. Moreover, spatial differences in ADC values between the core and rim of tumors with a T2/FLAIR-mismatch sign potentially reflect specific distinctions in tumor cellularity and microenvironment.

Noninvasive Characterization of Tumor Angiogenesis and Oxygenation in Bevacizumab-treated Recurrent Glioblastoma by Using Dynamic Susceptibility MRI: Secondary Analysis of the European Organization for Research and Treatment of Cancer 26101 Trial.

Background Relevance of antiangiogenic treatment with bevacizumab in patients with glioblastoma is controversial because progression-free survival benefit did not translate into an overall survival (OS) benefit in randomized phase III trials. Purpose To perform longitudinal characterization of intratumoral angiogenesis and oxygenation by using dynamic susceptibility contrast agent-enhanced (DSC) MRI and evaluate its potential for predicting outcome from administration of bevacizumab. Materials and Methods In this secondary analysis of the prospective randomized phase II/III European Organization for Research and Treatment of Cancer 26101 trial conducted between October 2011 and December 2015 in 596 patients with first recurrence of glioblastoma, the subset of patients with availability of anatomic MRI and DSC MRI at baseline and first follow-up was analyzed. Patients were allocated into those administered bevacizumab (hereafter, the BEV group; either bevacizumab monotherapy or bevacizumab with lomustine) and those not administered bevacizumab (hereafter, the non-BEV group with lomustine monotherapy). Contrast-enhanced tumor volume, noncontrast-enhanced T2 fluid-attenuated inversion recovery (FLAIR) signal abnormality volume, Gaussian-normalized relative cerebral blood volume (nrCBV), Gaussian-normalized relative blood flow (nrCBF), and tumor metabolic rate of oxygen (nTMRO2) was quantified. The predictive ability of these imaging parameters was assessed with multivariable Cox regression and formal interaction testing. Results A total of 254 of 596 patients were evaluated (mean age, 57 years ± 11; 155 men; 161 in the BEV group and 93 in non-BEV group). Progression-free survival was longer in the BEV group (3.7 months; 95% confidence interval [CI]: 3.0, 4.2) compared with the non-BEV group (2.5 months; 95% CI: 1.5, 2.9; P = .01), whereas OS was not different (P = .15). The nrCBV decreased for the BEV group (-16.3%; interquartile range [IQR], -39.5% to 12.0%; P = .01), but not for the non-BEV group (1.2%; IQR, -17.9% to 23.3%; P = .19) between baseline and first follow-up. An identical pattern was observed for both nrCBF and nTMRO2 values. Contrast-enhanced tumor and noncontrast-enhanced T2 FLAIR signal abnormality volumes decreased for the BEV group (-66% [IQR, -83% to -35%] and -33% [IQR, -71% to -5%], respectively; P < .001 for both), whereas they increased for the non-BEV group (30% [IQR, -17% to 98%], P = .001; and 10% [IQR, -13% to 82%], P = .02, respectively) between baseline and first follow-up. None of the assessed MRI parameters were predictive for OS in the BEV group. Conclusion Bevacizumab treatment decreased tumor volumes, angiogenesis, and oxygenation, thereby reflecting its effectiveness for extending progression-free survival; however, these parameters were not predictive of overall survival (OS), which highlighted the challenges of identifying patients that derive an OS benefit from bevacizumab. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Dillon in this issue.

Radiomics and Deep Learning from Research to Clinical Workflow: Neuro-Oncologic Imaging.

Imaging plays a key role in the management of brain tumors, including the diagnosis, prognosis, and treatment response assessment. Radiomics and deep learning approaches, along with various advanced physiologic imaging parameters, hold great potential for aiding radiological assessments in neuro-oncology. The ongoing development of new technology needs to be validated in clinical trials and incorporated into the clinical workflow. However, none of the potential neuro-oncological applications for radiomics and deep learning has yet been realized in clinical practice. In this review, we summarize the current applications of radiomics and deep learning in neuro-oncology and discuss challenges in relation to evidence-based medicine and reporting guidelines, as well as potential applications in clinical workflows and routine clinical practice.

Validation of diffusion MRI phenotypes for predicting response to bevacizumab in recurrent glioblastoma: post-hoc analysis of the EORTC-26101 trial.

BACKGROUND: This study validated a previously described diffusion MRI phenotype as a potential predictive imaging biomarker in patients with recurrent glioblastoma receiving bevacizumab (BEV). METHODS: A total of 396/596 patients (66%) from the prospective randomized phase II/III EORTC-26101 trial (with n = 242 in the BEV and n = 154 in the non-BEV arm) met the inclusion criteria with availability of anatomical and diffusion MRI sequences at baseline prior treatment. Apparent diffusion coefficient (ADC) histograms from the contrast-enhancing tumor volume were fitted to a double Gaussian distribution and the mean of the lower curve (ADClow) was used for further analysis. The predictive ability of ADClow was assessed with biomarker threshold models and multivariable Cox regression for overall survival (OS) and progression-free survival (PFS). RESULTS: ADClow was associated with PFS (hazard ratio [HR] = 0.625, P = 0.007) and OS (HR = 0.656, P = 0.031). However, no (predictive) interaction between ADClow and the treatment arm was present (P = 0.865 for PFS, P = 0.722 for OS). Independent (prognostic) significance of ADClow was retained after adjusting for epidemiological, clinical, and molecular characteristics (P ≤ 0.02 for OS, P ≤ 0.01 PFS). The biomarker threshold model revealed an optimal ADClow cutoff of 1241*10-6 mm2/s for OS. Thereby, median OS for BEV-patients with ADClow ≥ 1241 was 10.39 months versus 8.09 months for those with ADClow < 1241 (P = 0.004). Similarly, median OS for non-BEV patients with ADClow ≥ 1241 was 9.80 months versus 7.79 months for those with ADClow < 1241 (P = 0.054). CONCLUSIONS: ADClow is an independent prognostic parameter for stratifying OS and PFS in patients with recurrent glioblastoma. Consequently, the previously suggested role of ADClow as predictive imaging biomarker could not be confirmed within this phase II/III trial.

Impact of slice thickness on clinical utility of automated Alberta Stroke Program Early Computed Tomography Scores.

OBJECTIVES: The clinical utility of electronically derived ASPECTS (e-ASPECTS) to quantify signs of acute ischemic infarction could be demonstrated in multiple studies. Here, we aim to clinically validate the impact of CT slice thickness (ST) on the performance of e-ASPECTS software. METHODS: A consecutive series of n = 258 patients (06/2016 and 01/2019) with middle cerebral artery occlusion and subsequent treatment with mechanical thrombectomy was analyzed. The e-ASPECTS score and acute infarct volumes were calculated from baseline non-contrast CT with a software using 1-mm slice thickness (ST) (defined as ground truth) and axial reconstructions with 2-10-mm ST and correlated with baseline stroke severity (NIHSS) as well as clinical outcome (mRS) using logistic regressions. RESULTS: In comparison with the ground truth, significant differences were seen in e-ASPECTS scores with ST > 6 mm (p ≤ 0.031) and infarct volumes with ST > 4 mm (p ≤ 0.001). There was a significant correlation of lower e-ASPECTS and higher acute infarct volumes with increasing baseline NIHSS values for all ST (p ≤ 0.001, respectively), with values derived from 1 mm yielding the highest correlation for both parameters (rho, - 0.38 and 0.31, respectively). Similarly, lower e-ASPECTS and higher acute infarct volumes from all ST were significantly associated with poor outcome after 90 days (p ≤ 0.05, respectively) with values derived from 1-mm ST yielding the highest effects for both parameters (OR, 0.69 [95% CI 0.50-0.88] and 1.27 [95% CI 1.10-1.50], respectively). CONCLUSIONS: The e-ASPECTS software generates robust values for e-ASPECTS and acute infarct volumes when using ST ≤ 4 mm with ST = 1 mm yielding the best performance for predicting baseline stroke severity and clinical outcome after 90 days. KEY POINTS: • Clinical utility of automatically derived ASPECTS from computed tomography scans was shown in patients with acute ischemic stroke and treatment with mechanical thrombectomy. • Thin slices (= 1 mm) had the highest clinical utility in comparison with thicker slices (2-10 mm) by having the strongest correlation with baseline stroke severity and independent effects on clinical outcome after 90 days. • Automatically calculated acute infarct volumes possess clinical utility beyond ASPECTS and should be considered in future studies.

Heterogeneity of response to immune checkpoint blockade in hypermutated experimental gliomas.

Intrinsic malignant brain tumors, such as glioblastomas are frequently resistant to immune checkpoint blockade (ICB) with few hypermutated glioblastomas showing response. Modeling patient-individual resistance is challenging due to the lack of predictive biomarkers and limited accessibility of tissue for serial biopsies. Here, we investigate resistance mechanisms to anti-PD-1 and anti-CTLA-4 therapy in syngeneic hypermutated experimental gliomas and show a clear dichotomy and acquired immune heterogeneity in ICB-responder and non-responder tumors. We made use of this dichotomy to establish a radiomic signature predicting tumor regression after pseudoprogression induced by ICB therapy based on serial magnetic resonance imaging. We provide evidence that macrophage-driven ICB resistance is established by CD4 T cell suppression and Treg expansion in the tumor microenvironment via the PD-L1/PD-1/CD80 axis. These findings uncover an unexpected heterogeneity of response to ICB in strictly syngeneic tumors and provide a rationale for targeting PD-L1-expressing tumor-associated macrophages to overcome resistance to ICB.

Automated volumetric assessment with artificial neural networks might enable a more accurate assessment of disease burden in patients with multiple sclerosis.

OBJECTIVES: Patients with multiple sclerosis (MS) regularly undergo MRI for assessment of disease burden. However, interpretation may be time consuming and prone to intra- and interobserver variability. Here, we evaluate the potential of artificial neural networks (ANN) for automated volumetric assessment of MS disease burden and activity on MRI. METHODS: A single-institutional dataset with 334 MS patients (334 MRI exams) was used to develop and train an ANN for automated identification and volumetric segmentation of T2/FLAIR-hyperintense and contrast-enhancing (CE) lesions. Independent testing was performed in a single-institutional longitudinal dataset with 82 patients (266 MRI exams). We evaluated lesion detection performance (F1 scores), lesion segmentation agreement (DICE coefficients), and lesion volume agreement (concordance correlation coefficients [CCC]). Independent evaluation was performed on the public ISBI-2015 challenge dataset. RESULTS: The F1 score was maximized in the training set at a detection threshold of 7 mm3 for T2/FLAIR lesions and 14 mm3 for CE lesions. In the training set, mean F1 scores were 0.867 for T2/FLAIR lesions and 0.636 for CE lesions, as compared to 0.878 for T2/FLAIR lesions and 0.715 for CE lesions in the test set. Using these thresholds, the ANN yielded mean DICE coefficients of 0.834 and 0.878 for segmentation of T2/FLAIR and CE lesions in the training set (fivefold cross-validation). Corresponding DICE coefficients in the test set were 0.846 for T2/FLAIR lesions and 0.908 for CE lesions, and the CCC was ≥ 0.960 in each dataset. CONCLUSIONS: Our results highlight the capability of ANN for quantitative state-of-the-art assessment of volumetric lesion load on MRI and potentially enable a more accurate assessment of disease burden in patients with MS. KEY POINTS: • Artificial neural networks (ANN) can accurately detect and segment both T2/FLAIR and contrast-enhancing MS lesions in MRI data. • Performance of the ANN was consistent in a clinically derived dataset, with patients presenting all possible disease stages in MRI scans acquired from standard clinical routine rather than with high-quality research sequences. • Computer-aided evaluation of MS with ANN could streamline both clinical and research procedures in the volumetric assessment of MS disease burden as well as in lesion detection.

Classification of Cancer at Prostate MRI: Deep Learning versus Clinical PI-RADS Assessment.

Background Men suspected of having clinically significant prostate cancer (sPC) increasingly undergo prostate MRI. The potential of deep learning to provide diagnostic support for human interpretation requires further evaluation. Purpose To compare the performance of clinical assessment to a deep learning system optimized for segmentation trained with T2-weighted and diffusion MRI in the task of detection and segmentation of lesions suspicious for sPC. Materials and Methods In this retrospective study, T2-weighted and diffusion prostate MRI sequences from consecutive men examined with a single 3.0-T MRI system between 2015 and 2016 were manually segmented. Ground truth was provided by combined targeted and extended systematic MRI-transrectal US fusion biopsy, with sPC defined as International Society of Urological Pathology Gleason grade group greater than or equal to 2. By using split-sample validation, U-Net was internally validated on the training set (80% of the data) through cross validation and subsequently externally validated on the test set (20% of the data). U-Net-derived sPC probability maps were calibrated by matching sextant-based cross-validation performance to clinical performance of Prostate Imaging Reporting and Data System (PI-RADS). Performance of PI-RADS and U-Net were compared by using sensitivities, specificities, predictive values, and Dice coefficient. Results A total of 312 men (median age, 64 years; interquartile range [IQR], 58-71 years) were evaluated. The training set consisted of 250 men (median age, 64 years; IQR, 58-71 years) and the test set of 62 men (median age, 64 years; IQR, 60-69 years). In the test set, PI-RADS cutoffs greater than or equal to 3 versus cutoffs greater than or equal to 4 on a per-patient basis had sensitivity of 96% (25 of 26) versus 88% (23 of 26) at specificity of 22% (eight of 36) versus 50% (18 of 36). U-Net at probability thresholds of greater than or equal to 0.22 versus greater than or equal to 0.33 had sensitivity of 96% (25 of 26) versus 92% (24 of 26) (both P > .99) with specificity of 31% (11 of 36) versus 47% (17 of 36) (both P > .99), not statistically different from PI-RADS. Dice coefficients were 0.89 for prostate and 0.35 for MRI lesion segmentation. In the test set, coincidence of PI-RADS greater than or equal to 4 with U-Net lesions improved the positive predictive value from 48% (28 of 58) to 67% (24 of 36) for U-Net probability thresholds greater than or equal to 0.33 (P = .01), while the negative predictive value remained unchanged (83% [25 of 30] vs 83% [43 of 52]; P > .99). Conclusion U-Net trained with T2-weighted and diffusion MRI achieves similar performance to clinical Prostate Imaging Reporting and Data System assessment. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Padhani and Turkbey in this issue.

Automated brain extraction of multisequence MRI using artificial neural networks.

Brain extraction is a critical preprocessing step in the analysis of neuroimaging studies conducted with magnetic resonance imaging (MRI) and influences the accuracy of downstream analyses. The majority of brain extraction algorithms are, however, optimized for processing healthy brains and thus frequently fail in the presence of pathologically altered brain or when applied to heterogeneous MRI datasets. Here we introduce a new, rigorously validated algorithm (termed HD-BET) relying on artificial neural networks that aim to overcome these limitations. We demonstrate that HD-BET outperforms six popular, publicly available brain extraction algorithms in several large-scale neuroimaging datasets, including one from a prospective multicentric trial in neuro-oncology, yielding state-of-the-art performance with median improvements of +1.16 to +2.50 points for the Dice coefficient and -0.66 to -2.51 mm for the Hausdorff distance. Importantly, the HD-BET algorithm, which shows robust performance in the presence of pathology or treatment-induced tissue alterations, is applicable to a broad range of MRI sequence types and is not influenced by variations in MRI hardware and acquisition parameters encountered in both research and clinical practice. For broader accessibility, the HD-BET prediction algorithm is made freely available (www.neuroAI-HD.org) and may become an essential component for robust, automated, high-throughput processing of MRI neuroimaging data.

Can Virtual Contrast Enhancement in Brain MRI Replace Gadolinium?: A Feasibility Study.

OBJECTIVES: Gadolinium-based contrast agents (GBCAs) have become an integral part in daily clinical decision making in the last 3 decades. However, there is a broad consensus that GBCAs should be exclusively used if no contrast-free magnetic resonance imaging (MRI) technique is available to reduce the amount of applied GBCAs in patients. In the current study, we investigate the possibility of predicting contrast enhancement from noncontrast multiparametric brain MRI scans using a deep-learning (DL) architecture. MATERIALS AND METHODS: A Bayesian DL architecture for the prediction of virtual contrast enhancement was developed using 10-channel multiparametric MRI data acquired before GBCA application. The model was quantitatively and qualitatively evaluated on 116 data sets from glioma patients and healthy subjects by comparing the virtual contrast enhancement maps to the ground truth contrast-enhanced T1-weighted imaging. Subjects were split in 3 different groups: enhancing tumors (n = 47), nonenhancing tumors (n = 39), and patients without pathologic changes (n = 30). The tumor regions were segmented for a detailed analysis of subregions. The influence of the different MRI sequences was determined. RESULTS: Quantitative results of the virtual contrast enhancement yielded a sensitivity of 91.8% and a specificity of 91.2%. T2-weighted imaging, followed by diffusion-weighted imaging, was the most influential sequence for the prediction of virtual contrast enhancement. Analysis of the whole brain showed a mean area under the curve of 0.969 ± 0.019, a peak signal-to-noise ratio of 22.967 ± 1.162 dB, and a structural similarity index of 0.872 ± 0.031. Enhancing and nonenhancing tumor subregions performed worse (except for the peak signal-to-noise ratio of the nonenhancing tumors). The qualitative evaluation by 2 raters using a 4-point Likert scale showed good to excellent (3-4) results for 91.5% of the enhancing and 92.3% of the nonenhancing gliomas. However, despite the good scores and ratings, there were visual deviations between the virtual contrast maps and the ground truth, including a more blurry, less nodular-like ring enhancement, few low-contrast false-positive enhancements of nonenhancing gliomas, and a tendency to omit smaller vessels. These "features" were also exploited by 2 trained radiologists when performing a Turing test, allowing them to discriminate between real and virtual contrast-enhanced images in 80% and 90% of the cases, respectively. CONCLUSIONS: The introduced model for virtual gadolinium enhancement demonstrates a very good quantitative and qualitative performance. Future systematic studies in larger patient collectives with varying neurological disorders need to evaluate if the introduced virtual contrast enhancement might reduce GBCA exposure in clinical practice.

Imaging necrosis during treatment is associated with worse survival in EORTC 26101 study.

OBJECTIVE: Imaging necrosis on MRI scans was assessed and compared to outcome measures of the European Organisation for Research and Treatment of Cancer 26101 phase III trial that compared single-agent lomustine with lomustine plus bevacizumab in patients with progressive glioblastoma. METHODS: MRI in this post hoc analysis was available for 359 patients (lomustine = 127, lomustine + bevacizumab = 232). First, imaging necrosis at baseline being formally measurable (>10 × 10 mm, given 2 slices) was assessed. At weeks 6 and 12 of treatment, it was analyzed whether this necrosis remained stable or increased >25% calculated by 2 perpendicular diameters or whether necrosis developed de novo. Univariate and multivariate associations of baseline necrosis with overall survival (OS) and progression-free survival (PFS) were tested by log-rank test. Hazard ratios (HR) with 95% confidence interval were calculated by Cox model. RESULTS: Imaging necrosis at baseline was detected in 191 patients (53.2%) and was associated with worse OS and PFS in univariate, but not in multivariate analysis. Baseline necrosis was predictive for OS in the lomustine-only group (HR 1.46, p = 0.018). At weeks 6 and 12 of treatment, increase of baseline necrosis and de novo necrosis were strongly associated with worse OS and PFS in univariate and multivariate analysis (PFS both p < 0.001, OS univariate p < 0.001, multivariate p = 0.0046). CONCLUSION: Increase of and new development of imaging necrosis during treatment is a negative prognostic factor for patients with progressive glioblastoma. These data call for consideration of integrating the assessment of imaging necrosis as a separate item into the MRI response assessment criteria.

Automated quantitative tumour response assessment of MRI in neuro-oncology with artificial neural networks: a multicentre, retrospective study.

BACKGROUND: The Response Assessment in Neuro-Oncology (RANO) criteria and requirements for a uniform protocol have been introduced to standardise assessment of MRI scans in both clinical trials and clinical practice. However, these criteria mainly rely on manual two-dimensional measurements of contrast-enhancing (CE) target lesions and thus restrict both reliability and accurate assessment of tumour burden and treatment response. We aimed to develop a framework relying on artificial neural networks (ANNs) for fully automated quantitative analysis of MRI in neuro-oncology to overcome the inherent limitations of manual assessment of tumour burden. METHODS: In this retrospective study, we compiled a single-institution dataset of MRI data from patients with brain tumours being treated at Heidelberg University Hospital (Heidelberg, Germany; Heidelberg training dataset) to develop and train an ANN for automated identification and volumetric segmentation of CE tumours and non-enhancing T2-signal abnormalities (NEs) on MRI. Independent testing and large-scale application of the ANN for tumour segmentation was done in a single-institution longitudinal testing dataset from the Heidelberg University Hospital and in a multi-institutional longitudinal testing dataset from the prospective randomised phase 2 and 3 European Organisation for Research and Treatment of Cancer (EORTC)-26101 trial (NCT01290939), acquired at 38 institutions across Europe. In both longitudinal datasets, spatial and temporal tumour volume dynamics were automatically quantified to calculate time to progression, which was compared with time to progression determined by RANO, both in terms of reliability and as a surrogate endpoint for predicting overall survival. We integrated this approach for fully automated quantitative analysis of MRI in neuro-oncology within an application-ready software infrastructure and applied it in a simulated clinical environment of patients with brain tumours from the Heidelberg University Hospital (Heidelberg simulation dataset). FINDINGS: For training of the ANN, MRI data were collected from 455 patients with brain tumours (one MRI per patient) being treated at Heidelberg hospital between July 29, 2009, and March 17, 2017 (Heidelberg training dataset). For independent testing of the ANN, an independent longitudinal dataset of 40 patients, with data from 239 MRI scans, was collected at Heidelberg University Hospital in parallel with the training dataset (Heidelberg test dataset), and 2034 MRI scans from 532 patients at 34 institutions collected between Oct 26, 2011, and Dec 3, 2015, in the EORTC-26101 study were of sufficient quality to be included in the EORTC-26101 test dataset. The ANN yielded excellent performance for accurate detection and segmentation of CE tumours and NE volumes in both longitudinal test datasets (median DICE coefficient for CE tumours 0·89 [95% CI 0·86-0·90], and for NEs 0·93 [0·92-0·94] in the Heidelberg test dataset; CE tumours 0·91 [0·90-0·92], NEs 0·93 [0·93-0·94] in the EORTC-26101 test dataset). Time to progression from quantitative ANN-based assessment of tumour response was a significantly better surrogate endpoint than central RANO assessment for predicting overall survival in the EORTC-26101 test dataset (hazard ratios ANN 2·59 [95% CI 1·86-3·60] vs central RANO 2·07 [1·46-2·92]; p<0·0001) and also yielded a 36% margin over RANO (p<0·0001) when comparing reliability values (ie, agreement in the quantitative volumetrically defined time to progression [based on radiologist ground truth vs automated assessment with ANN] of 87% [266 of 306 with sufficient data] compared with 51% [155 of 306] with local vs independent central RANO assessment). In the Heidelberg simulation dataset, which comprised 466 patients with brain tumours, with 595 MRI scans obtained between April 27, and Sept 17, 2018, automated on-demand processing of MRI scans and quantitative tumour response assessment within the simulated clinical environment required 10 min of computation time (average per scan). INTERPRETATION: Overall, we found that ANN enabled objective and automated assessment of tumour response in neuro-oncology at high throughput and could ultimately serve as a blueprint for the application of ANN in radiology to improve clinical decision making. Future research should focus on prospective validation within clinical trials and application for automated high-throughput imaging biomarker discovery and extension to other diseases. FUNDING: Medical Faculty Heidelberg Postdoc-Program, Else Kröner-Fresenius Foundation.

Risk factors of intracranial hemorrhage after mechanical thrombectomy of anterior circulation ischemic stroke.

PURPOSE: Intracranial hemorrhage (ICH) is a potentially severe complication after mechanical thrombectomy (MT). Here, we investigated risk factors for the occurrence of any and symptomatic ICH after MT due to large-vessel occlusion of the anterior circulation. METHODS: Consecutive patients with acute ischemic anterior circulation stroke with large-vessel occlusion undergoing MT were analyzed. ICH was categorized according to the Heidelberg Bleeding Classification. Forty-three procedural and clinical parameters were analyzed using univariate tests and multivariate logistic regressions. RESULTS: Of 612 patients, any ICH was detected in 195 (31.9%), while 27 (4.4%) developed a symptomatic ICH. Infarct size > 1/3 of vascular territory in control imaging (OR 2.18, 95% CI 1.45-3.21), higher serum glucose levels (OR 1.23 for change of 15 units mg/dL, 95% CI 1.10-1.39), and higher thrombectomy maneuver count (OR 1.21, 95% CI 1.11-1.32) were significantly associated with a higher risk of developing any ICH compared to no ICH. Wake-up strokes (OR 3.99, 95% CI 1.38-11.60), transfer from an external clinic (OR 3.04, 95% CI 1.24-7.48), and higher serum glucose levels (OR 1.22 for change of 15 units mg/dL, 95% CI 1.05-1.42) were revealed as independent risk factors for development of symptomatic ICH compared to no symptomatic ICH. Patients with no infarct demarcation (OR 0.10, 95% CI 0.01-0.80) and complete recanalization (OR 0.57, 95% CI 0.37-0.86) showed a lower risk of developing any ICH. CONCLUSION: Wake-up strokes and patients who are treated within a drip-and-ship concept are especially vulnerable for symptomatic ICH, while complete recanalization, contrary to subtotal recanalization only, was revealed as a protective factor against ICH.

Radiomic Machine Learning for Characterization of Prostate Lesions with MRI: Comparison to ADC Values.

Purpose To compare biparametric contrast-free radiomic machine learning (RML), mean apparent diffusion coefficient (ADC), and radiologist assessment for characterization of prostate lesions detected during prospective MRI interpretation. Materials and Methods This single-institution study included 316 men (mean age ± standard deviation, 64.0 years ± 7.8) with an indication for MRI-transrectal US fusion biopsy between May 2015 and September 2016 (training cohort, 183 patients; test cohort, 133 patients). Lesions identified by prospective clinical readings were manually segmented for mean ADC and radiomics analysis. Global and zone-specific random forest RML and mean ADC models for classification of clinically significant prostate cancer (Gleason grade group ≥ 2) were developed on the training set and the fixed models tested on an independent test set. Clinical readings, mean ADC, and radiomics were compared by using the McNemar test and receiver operating characteristic (ROC) analysis. Results In the test set, radiologist interpretation had a per-lesion sensitivity of 88% (53 of 60) and specificity of 50% (79 of 159). Quantitative measurement of the mean ADC (cut-off 732 mm2/sec) significantly reduced false-positive (FP) lesions from 80 to 60 (specificity 62% [99 of 159]) and false-negative (FN) lesions from seven to six (sensitivity 90% [54 of 60]) (P = .048). Radiologist interpretation had a per-patient sensitivity of 89% (40 of 45) and specificity of 43% (38 of 88). Quantitative measurement of the mean ADC reduced the number of patients with FP lesions from 50 to 43 (specificity 51% [45 of 88]) and the number of patients with FN lesions from five to three (sensitivity 93% [42 of 45]) (P = .496). Comparison of the area under the ROC curve (AUC) for the mean ADC (AUCglobal = 0.84; AUCzone-specific ≤ 0.87) vs the RML (AUCglobal = 0.88, P = .176; AUCzone-specific ≤ 0.89, P ≥ .493) showed no significantly different performance. Conclusion Quantitative measurement of the mean apparent diffusion coefficient (ADC) improved differentiation of benign versus malignant prostate lesions, compared with clinical assessment. Radiomic machine learning had comparable but not better performance than mean ADC assessment. © RSNA, 2018 Online supplemental material is available for this article.

Voxel-wise radiogenomic mapping of tumor location with key molecular alterations in patients with glioma.

Background: This study aims to evaluate the impact of tumor location on key molecular alterations on a single voxel level in patients with newly diagnosed glioma. Methods: A consecutive series of n = 237 patients with newly diagnosed glioblastoma and n = 131 patients with lower-grade glioma was analyzed. Volumetric tumor segmentation was performed on preoperative MRI with a semi-automated approach and images were registered to the standard Montreal Neurological Institute 152 space. Using a voxel-based lesion symptom mapping (VLSM) analysis, we identified specific brain regions that were associated with tumor-specific molecular alterations. We assessed a predefined set of n = 17 molecular characteristics in the glioblastoma cohort and n = 2 molecular characteristics in the lower-grade glioma cohort. Permutation adjustment (n = 1000 iterations) was used to correct for multiple testing, and voxel t-values that were greater than the t-value in >95% of the permutations were retained in the VLSM results (α = 0.05, power > 0.8). Results: Tumor location predilection for isocitrate dehydrogenase (IDH) mutant tumors was found in both glioblastoma and lower-grade glioma cohorts, each showing a concordant predominance in the frontal lobe adjacent to the rostral extension of the lateral ventricles (permutation-adjusted P = 0.021 for the glioblastoma and 0.013 for the lower-grade glioma cohort). Apart from that, the VLSM analysis did not reveal a significant association of the tumor location with any other key molecular alteration in both cohorts (permutation-adjusted P > 0.05 each). Conclusion: Our study highlights the unique properties of IDH mutations and underpins the hypothesis that the rostral extension of the lateral ventricles is a potential location for the cell of origin in IDH-mutant gliomas.

Nonmeasurable Speckled Contrast-Enhancing Lesions Appearing During Course of Disease Are Associated With IDH Mutation in High-Grade Astrocytoma Patients.

PURPOSE: Because treatment options at progression are limited for patients with glioma, accuracy in definition of progression is pivotal. Clinically asymptomatic, newly detected, nonmeasurable, speckled contrast-enhancing lesions (SCEs) without immediate relation to prior immune therapy or radiation therapy appear relatively frequently during the course of disease in patients with glioma and challenge the definition of progression based on Response Assessment in Neuro-oncology criteria. Therefore, data characterizing these SCEs are needed for recommendations of subsequent clinical management. MATERIALS AND METHODS: Magnetic resonance imaging of 746 patients with glioma included in this study were retrospectively revised for appearance of newly detected SCEs during the course of disease. Associations with molecular and clinical baseline parameters and their prognostic impact were statistically analyzed, and frequency, natural course, and location of SCEs were described. RESULTS: SCEs occurred more frequently in World Health Organization grade 2 and 3 astrocytoma and oligodendroglial tumors and were significantly associated with isocitrate dehydrogenase mutation in World Health Organization grade 3 astrocytoma and glioblastoma. SCEs mostly remained stable or dissolved in follow-up magnetic resonance imaging, even if no new treatment was initiated. SCEs were frequently located within the tumor or tumor-associated fluid-attenuated inversion recovery abnormalities, but distant appearance also occurred. In patients with glioblastoma, SCEs were associated with a favorable prognosis, which was also observed in the subgroup of patients with glioblastoma with isocitrate dehydrogenase wildtype status. CONCLUSIONS: The data demonstrate a predominantly benign course of SCEs after their appearance and emphasize cautious definitions of progression and regular clinical and radiographic follow-up rather than premature initiation of new antitumor therapies until progression is confirmed.

Early postoperative delineation of residual tumor after low-grade glioma resection by probabilistic quantification of diffusion-weighted imaging.

OBJECTIVEIn WHO grade II low-grade gliomas (LGGs), early postoperative MRI (epMRI) may overestimate residual tumor on FLAIR sequences. Consequently, MRI at 3-6 months follow-up (fuMRI) is used for delineation of residual tumor. This study sought to evaluate if integration of apparent diffusion coefficient (ADC) maps permits an accurate estimation of residual tumor early on epMRI.METHODSFrom a consecutive cohort, 43 cases with an initial surgery for an LGG, and complete epMRI (< 72 hours after resection) and fuMRI including ADC maps, were retrospectively identified. Residual FLAIR hyperintense tumor was manually segmented on epMRI and corresponding ADC maps were coregistered. Using an expectation maximization algorithm, residual tumor segments were probabilistically clustered into areas of residual tumor, ischemia, or normal white matter (NWM) by fitting a mixture model of superimposed Gaussian curves to the ADC histogram. Tumor volumes from epMRI, clustering, and fuMRI were statistically compared and agreement analysis was performed.RESULTSMean FLAIR hyperintensity suggesting residual tumor was significantly larger on epMRI compared to fuMRI (19.4 ± 16.5 ml vs 8.4 ± 10.2 ml, p < 0.0001). Probabilistic clustering of corresponding ADC histograms on epMRI identified subsegments that were interpreted as mean residual tumor (7.6 ± 10.2 ml), ischemia (8.1 ± 5.9 ml), and NWM (3.7 ± 4.9 ml). Therefore, mean tumor quantification error between epMRI and fuMRI was significantly reduced (11.0 ± 10.6 ml vs -0.8 ± 3.7 ml, p < 0.0001). Mean clustered tumor volumes on epMRI were no longer significantly different from the fuMRI reference (7.6 ± 10.2 ml vs 8.4 ± 10.2 ml, p = 0.16). Correlation (Pearson r = 0.96, p < 0.0001), concordance correlation coefficient (0.89, 95% confidence interval 0.83), and Bland-Altman analysis suggested strong agreement between both measures after clustering.CONCLUSIONSProbabilistic segmentation of ADC maps facilitates accurate assessment of residual tumor within 72 hours after LGG resection. Multiparametric image analysis detected FLAIR signal alterations attributable to surgical trauma, which led to overestimation of residual LGG on epMRI compared to fuMRI. The prognostic value and clinical impact of this method has to be evaluated in larger case series in the future.

Radiomics, Metabolic, and Molecular MRI for Brain Tumors.

Magnetic resonance imaging plays a key role in diagnosis and treatment monitoring of brain tumors. Novel imaging techniques that specifically interrogate aspects of underlying tumor biology and biochemical pathways have great potential in neuro-oncology. This review focuses on the emerging role of 2-hydroxyglutarate-targeted magnetic resonance spectroscopy, as well as radiomics and radiogenomics in establishing diagnosis for isocitrate dehydrogenase mutant gliomas, and for monitoring treatment response and predicting prognosis of this group of brain tumor patients.