Quantitative imaging biomarkers in neuro-oncology

Improving performance in colorectal cancer histology decomposition using deep and ensemble machine learning

In routine colorectal cancer management, histologic samples stained with hematoxylin and eosin are commonly used. Nonetheless, their potential for defining objective biomarkers for patient stratification and treatment selection is still being explored. The current gold standard relies on expensive and time-consuming genetic tests. However, recent research highlights the potential of convolutional neural networks (CNNs) to facilitate the extraction of clinically relevant biomarkers from these readily available images. These CNN-based biomarkers can predict patient outcomes comparably to golden standards, with the added advantages of speed, automation, and minimal cost. The predictive potential of CNN-based biomarkers fundamentally relies on the ability of CNNs to accurately classify diverse tissue types from whole slide microscope images. Consequently, enhancing the accuracy of tissue class decomposition is critical to amplifying the prognostic potential of imaging-based biomarkers. This study introduces a hybrid deep transfer learning and ensemble machine learning model that improves upon previous approaches, including a transformer and neural architecture search baseline for this task. We employed a pairing of the EfficientNetV2 architecture with a random forest classification head. Our model achieved 96.74% accuracy (95% CI: 96.3%-97.1%) on the external test set and 99.89% on the internal test set. Recognizing the potential of these models in the task, we have made them publicly available.

Improved accuracy in colorectal cancer tissue decomposition through refinement of established deep learning solutions

Hematoxylin and eosin-stained biopsy slides are regularly available for colorectal cancer patients. These slides are often not used to define objective biomarkers for patient stratification and treatment selection. Standard biomarkers often pertain to costly and slow genetic tests. However, recent work has shown that relevant biomarkers can be extracted from these images using convolutional neural networks (CNNs). The CNN-based biomarkers predicted colorectal cancer patient outcomes comparably to gold standards. Extracting CNN-biomarkers is fast, automatic, and of minimal cost. CNN-based biomarkers rely on the ability of CNNs to recognize distinct tissue types from microscope whole slide images. The quality of these biomarkers (coined 'Deep Stroma') depends on the accuracy of CNNs in decomposing all relevant tissue classes. Improving tissue decomposition accuracy is essential for improving the prognostic potential of CNN-biomarkers. In this study, we implemented a novel training strategy to refine an established CNN model, which then surpassed all previous solutions . We obtained a 95.6% average accuracy in the external test set and 99.5% in the internal test set. Our approach reduced errors in biomarker-relevant classes, such as Lymphocytes, and was the first to include interpretability methods. These methods were used to better apprehend our model's limitations and capabilities.

Imaging biomarkers associated with extra-axial intracranial tumors: a systematic review

Extra-axial brain tumors are extra-cerebral tumors and are usually benign. The choice of treatment for extra-axial tumors is often dependent on the growth of the tumor, and imaging plays a significant role in monitoring growth and clinical decision-making. This motivates the investigation of imaging biomarkers for these tumors that may be incorporated into clinical workflows to inform treatment decisions. The databases from Pubmed, Web of Science, Embase, and Medline were searched from 1 January 2000 to 7 March 2022, to systematically identify relevant publications in this area. All studies that used an imaging tool and found an association with a growth-related factor, including molecular markers, grade, survival, growth/progression, recurrence, and treatment outcomes, were included in this review. We included 42 studies, comprising 22 studies (50%) of patients with meningioma; 17 studies (38.6%) of patients with pituitary tumors; three studies (6.8%) of patients with vestibular schwannomas; and two studies (4.5%) of patients with solitary fibrous tumors. The included studies were explicitly and narratively analyzed according to tumor type and imaging tool. The risk of bias and concerns regarding applicability were assessed using QUADAS-2. Most studies (41/44) used statistics-based analysis methods, and a small number of studies (3/44) used machine learning. Our review highlights an opportunity for future work to focus on machine learning-based deep feature identification as biomarkers, combining various feature classes such as size, shape, and intensity. Systematic Review Registration: PROSPERO, CRD42022306922.

Prognostic prediction based on histopathologic features of tumor microenvironment in colorectal cancer

Purpose

To automatically quantify colorectal tumor microenvironment (TME) in hematoxylin and eosin stained whole slide images (WSIs), and to develop a TME signature for prognostic prediction in colorectal cancer (CRC).

Methods

A deep learning model based on VGG19 architecture and transfer learning strategy was trained to recognize nine different tissue types in whole slide images of patients with CRC. Seven of the nine tissue types were defined as TME components besides background and debris. Then 13 TME features were calculated based on the areas of TME components. A total of 562 patients with gene expression data, survival information and WSIs were collected from The Cancer Genome Atlas project for further analysis. A TME signature for prognostic prediction was developed and validated using Cox regression method. A prognostic prediction model combined the TME signature and clinical variables was also established. At last, gene-set enrichment analysis was performed to identify the significant TME signature associated pathways by querying Gene Ontology database and Kyoto Encyclopedia of Genes and Genomes database.

Results

The deep learning model achieved an accuracy of 94.2% for tissue type recognition. The developed TME signature was found significantly associated to progression-free survival. The clinical combined model achieved a concordance index of 0.714. Gene-set enrichment analysis revealed the TME signature associated genes were enriched in neuroactive ligand-receptor interaction pathway.

Conclusion

The TME signature was proved to be a prognostic factor and the associated biologic pathways would be beneficial to a better understanding of TME in CRC patients.

A scoping review of neurodegenerative manifestations in explainable digital phenotyping

Neurologists nowadays no longer view neurodegenerative diseases, like Parkinson's and Alzheimer's disease, as single entities, but rather as a spectrum of multifaceted symptoms with heterogeneous progression courses and treatment responses. The definition of the naturalistic behavioral repertoire of early neurodegenerative manifestations is still elusive, impeding early diagnosis and intervention. Central to this view is the role of artificial intelligence (AI) in reinforcing the depth of phenotypic information, thereby supporting the paradigm shift to precision medicine and personalized healthcare. This suggestion advocates the definition of disease subtypes in a new biomarker-supported nosology framework, yet without empirical consensus on standardization, reliability and interpretability. Although the well-defined neurodegenerative processes, linked to a triad of motor and non-motor preclinical symptoms, are detected by clinical intuition, we undertake an unbiased data-driven approach to identify different patterns of neuropathology distribution based on the naturalistic behavior data inherent to populations in-the-wild. We appraise the role of remote technologies in the definition of digital phenotyping specific to brain-, body- and social-level neurodegenerative subtle symptoms, emphasizing inter- and intra-patient variability powered by deep learning. As such, the present review endeavors to exploit digital technologies and AI to create disease-specific phenotypic explanations, facilitating the understanding of neurodegenerative diseases as "bio-psycho-social" conditions. Not only does this translational effort within explainable digital phenotyping foster the understanding of disease-induced traits, but it also enhances diagnostic and, eventually, treatment personalization.

Glioma survival prediction from whole-brain MRI without tumor segmentation using deep attention network: a multicenter study

OBJECTIVES

To develop and validate a deep learning model for predicting overall survival from whole-brain MRI without tumor segmentation in patients with diffuse gliomas.

METHODS

In this multicenter retrospective study, two deep learning models were built for survival prediction from MRI, including a DeepRisk model built from whole-brain MRI, and an original ResNet model built from expert-segmented tumor images. Both models were developed using a training dataset (n = 935) and an internal tuning dataset (n = 156) and tested on two external test datasets (n = 194 and 150) and a TCIA dataset (n = 121). C-index, integrated Brier score (IBS), prediction error curves, and calibration curves were used to assess the model performance.

RESULTS

In total, 1556 patients were enrolled (age, 49.0 ± 13.1 years; 830 male). The DeepRisk score was an independent predictor and can stratify patients in each test dataset into three risk subgroups. The IBS and C-index for DeepRisk were 0.14 and 0.83 in external test dataset 1, 0.15 and 0.80 in external dataset 2, and 0.16 and 0.77 in TCIA dataset, respectively, which were comparable with those for original ResNet. The AUCs at 6, 12, 24, 26, and 48 months for DeepRisk ranged between 0.77 and 0.94. Combining DeepRisk score with clinicomolecular factors resulted in a nomogram with a better calibration and classification accuracy (net reclassification improvement 0.69, p < 0.001) than the clinical nomogram.

CONCLUSIONS

DeepRisk that obviated the need of tumor segmentation can predict glioma survival from whole-brain MRI and offers incremental prognostic value.

KEY POINTS

• DeepRisk can predict overall survival directly from whole-brain MRI without tumor segmentation. • DeepRisk achieves comparable accuracy in survival prediction with deep learning model built using expert-segmented tumor images. • DeepRisk has independent and incremental prognostic value over existing clinical parameters and IDH mutation status.

Diagnostic yield of simultaneous dynamic contrast-enhanced magnetic resonance perfusion measurements and [18F]FET PET in patients with suspected recurrent anaplastic astrocytoma and glioblastoma

Purpose

Both amino acid positron emission tomography (PET) and magnetic resonance imaging (MRI) blood volume (BV) measurements are used in suspected recurrent high-grade gliomas. We compared the separate and combined diagnostic yield of simultaneously acquired dynamic contrast-enhanced (DCE) perfusion MRI and O-(2-[¹⁸F]-fluoroethyl)-L-tyrosine ([¹⁸F]FET) PET in patients with anaplastic astrocytoma and glioblastoma following standard therapy.

Methods

A total of 76 lesions in 60 hybrid [¹⁸F]FET PET/MRI scans with DCE MRI from patients with suspected recurrence of anaplastic astrocytoma and glioblastoma were included retrospectively. BV was measured from DCE MRI employing a 2-compartment exchange model (2CXM). Diagnostic performances of maximal tumour-to-background [¹⁸F]FET uptake (TBR_max), maximal BV (BV_max) and normalised BV_max (nBV_max) were determined by ROC analysis using 6-month histopathological (n = 28) or clinical/radiographical follow-up (n = 48) as reference. Sensitivity and specificity at optimal cut-offs were determined separately for enhancing and non-enhancing lesions.

Results

In progressive lesions, all BV and [¹⁸F]FET metrics were higher than in non-progressive lesions. ROC analyses showed higher overall ROC AUCs for TBR_max than both BV_max and nBV_max in both lesion-wise (all lesions, p = 0.04) and in patient-wise analysis (p < 0.01). Combining TBR_max with BV metrics did not increase ROC AUC. Lesion-wise positive fraction/sensitivity/specificity at optimal cut-offs were 55%/91%/84% for TBR_max, 45%/77%/84% for BV_max and 59%/84%/72% for nBV_max. Combining TBR_max and best-performing BV cut-offs yielded lesion-wise sensitivity/specificity of 75/97%. The fraction of progressive lesions was 11% in concordant negative lesions, 33% in lesions only BV positive, 64% in lesions only [¹⁸F]FET positive and 97% in concordant positive lesions.

Conclusion

The overall diagnostic accuracy of DCE BV imaging is good, but lower than that of [¹⁸F]FET PET. Adding DCE BV imaging did not improve the overall diagnostic accuracy of [¹⁸F]FET PET, but may improve specificity and allow better lesion-wise risk stratification than [¹⁸F]FET PET alone.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00259-022-05917-3.

High-Resolution Diffusion-Weighted Imaging of C6 Glioma on a 7T BioSpec MRI Scanner: Correlation of Tumor Cellularity and Nuclear-to-Cytoplasmic Ratio with Apparent Diffusion Coefficient

RATIONALE AND OBJECTIVES

To determine the association of the apparent diffusion coefficient (ADC) with quantitative cellularity and the nuclear-to-cytoplasmic ratio in C6 glioma.

MATERIALS AND METHODS

Animal models bearing C6 gliomas underwent MR scans with T1 rapid acquisition with relaxation enhancement (RARE), T2 RARE, and high-resolution diffusion-weighted imaging sequences. For each model, three consecutive sections were used to draw regions of interest (ROIs) and measure ADC values; the middle section was localized in the plane with the maximal solid tumor area. The minimal, mean, and maximal ADC values were recorded for each ROI. GFAP-immunostained sections coregistered with ADC measurements were used to calculate tumor cellularity and the nuclear-to-cytoplasmic (N/C) ratio. Spearman's correlation was used to assess the relationship between ADC values and quantitative tumor cellularity as well as N/C ratios with a significance level of p < 0.05.

RESULTS

Thirty-three sections from 11 glioma-bearing rats were analyzed. The median values of the minimal, mean, and maximal ADC were 0.443 × 10^-3, 0.744 × 10^-3, and 1.140 × 10^-3 mm²/s, respectively. The median cellularity and N/C ratio were 2151.234 per 0.025 mm² and 0.857, respectively. The minimal, mean, and maximal ADCs were all significantly associated with cellularity, with correlation coefficients of -0.712 (p < 0.001), -0.631 (p < 0.001), and -0.460 (p = 0.007), respectively. The minimal and mean ADC had significant negative relationships with the N/C ratio, with correlation coefficients of -0.565 (p =  0.001) and -0.426 (p = 0.013), respectively.

CONCLUSION

The minimal ADC correlated well with cellularity and N/C ratios in C6 glioma and may be used as a biomarker of these two pathological features.

Interpretable Diagnosis for Whole-Slide Melanoma Histology Images Using Convolutional Neural Network

At present, deep learning-based medical image diagnosis had achieved high performance in several diseases. However, the black-box nature of the convolutional neural network (CNN) limits their role in diagnosis. In this study, a novel interpretable diagnosis pipeline using the CNN model was proposed. Furthermore, a sizeable melanoma database that contains 841 digital whole-slide images (WSIs) was built to train and evaluate the model. The model achieved strong melanoma classification ability (0.962 areas under the receiver operating characteristic, 0.887 sensitivity, and 0.925 specificity). Moreover, the proposed model outperformed the existing schemes in terms of accuracy that is 20 pathologists (0.933 vs 0.732 accuracy). Finally, the gradient-weighted class activation mapping (Grad-CAM) method was used to show the inner logic of the proposed model and its feasibility to improve diagnosis process in healthcare. The mechanism of feature heat maps which is visualized through a saliency mapping has demonstrated that features learned or extracted by the proposed model are compatible with the accepted pathological features. Conclusively, the proposed model provides a rapid and accurate diagnosis by locating the distinctive features of melanoma to build doctors' trust in the CNNs' diagnosis results.

BOLD Asynchrony Elucidates Tumor Burden in IDH-Mutated Gliomas

BACKGROUND

Gliomas comprise the most common type of primary brain tumor, are highly invasive, and often fatal. IDH-mutated gliomas are particularly challenging to image and there is currently no clinically accepted method for identifying the extent of tumor burden in these neoplasms. This uncertainty poses a challenge to clinicians who must balance the need to treat the tumor while sparing healthy brain from iatrogenic damage. The purpose of this study was to investigate the feasibility of using resting-state blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI) to detect glioma-related asynchrony in vascular dynamics for distinguishing tumor from healthy brain.

METHODS

Twenty-four stereotactically localized biopsies were obtained during open surgical resection from ten treatment-naïve patients with IDH-mutated gliomas who received standard of care preoperative imaging as well as echo-planar resting-state BOLD fMRI. Signal intensity for BOLD asynchrony and standard of care imaging was compared to cell counts of total cellularity (H&E), tumor density (IDH1 & Sox2), cellular proliferation (Ki67), and neuronal density (NeuN), for each corresponding sample.

RESULTS

BOLD asynchrony was directly related to total cellularity (H&E, p = 4 x 10 -5), tumor density (IDH1, p = 4 x 10 -5; Sox2, p = 3 x 10 -5), cellular proliferation (Ki67, p = 0.002), and as well as inversely related to neuronal density (NeuN, p = 1 x 10 -4).

CONCLUSIONS

Asynchrony in vascular dynamics, as measured by resting-state BOLD fMRI, correlates with tumor burden and provides a radiographic delineation of tumor boundaries in IDH-mutated gliomas.

Clinical quantitative MRI and the need for metrology

MRI has been an essential diagnostic tool in healthcare for several decades. It offers unique insights into most tissues without the need for ionising radiation. Historically, MRI has been predominantly used qualitatively, images are formed to allow visual discrimination of tissues types and pathologies, rather than providing quantitative measurements. Increasingly, quantitative MRI (qMRI) is also finding clinical application, where images provide the basis for physical measurements of, e.g. tissue volume measures and represent aspects of tissue composition and microstructure. This article reviews some common current research and clinical applications of qMRI from the perspective of measurement science. qMRI not only offers additional information for radiologists, but also the opportunity for improved harmonisation and calibration between scanners and as such it is well-suited to large-scale investigations such as clinical trials and longitudinal studies. Realising these benefits, however, presents a new kind of technical challenge to MRI practioners. When measuring a parameter quantitatively, it is crucial that the reliability and reproducibility of the technique are well understood. Strictly speaking, a numerical result of a measurement is meaningless unless it is accompanied by a description of the associated measurement uncertainty. It is therefore necessary to produce not just estimates of physical properties in a quantitative image, but also their associated uncertainties. As the process of determining a physical property from the raw MR signal is complicated and multistep, estimation of uncertainty is challenging and there are many aspects of the MRI process that require validation. With the clinical implementation of qMRI techniques and its continued expansion, there is a clear and urgent need for metrology in this field.

A Position Statement on the Utility of Interval Imaging in Standard of Care Brain Tumour Management: Defining the Evidence Gap and Opportunities for Future Research

OBJECTIV E

To summarise current evidence for the utility of interval imaging in monitoring disease in adult brain tumours, and to develop a position for future evidence gathering while incorporating the application of data science and health economics.

METHODS

Experts in 'interval imaging' (imaging at pre-planned time-points to assess tumour status); data science; health economics, trial management of adult brain tumours, and patient representatives convened in London, UK. The current evidence on the use of interval imaging for monitoring brain tumours was reviewed. To improve the evidence that interval imaging has a role in disease management, we discussed specific themes of data science, health economics, statistical considerations, patient and carer perspectives, and multi-centre study design. Suggestions for future studies aimed at filling knowledge gaps were discussed.

RESULTS

Meningioma and glioma were identified as priorities for interval imaging utility analysis. The "monitoring biomarkers" most commonly used in adult brain tumour patients were standard structural MRI features. Interval imaging was commonly scheduled to provide reported imaging prior to planned, regular clinic visits. There is limited evidence relating interval imaging in the absence of clinical deterioration to management change that alters morbidity, mortality, quality of life, or resource use. Progression-free survival is confounded as an outcome measure when using structural MRI in glioma. Uncertainty from imaging causes distress for some patients and their caregivers, while for others it provides an important indicator of disease activity. Any study design that changes imaging regimens should consider the potential for influencing current or planned therapeutic trials, ensure that opportunity costs are measured, and capture indirect benefits and added value.

CONCLUSION

Evidence for the value, and therefore utility, of regular interval imaging is currently lacking. Ongoing collaborative efforts will improve trial design and generate the evidence to optimise monitoring imaging biomarkers in standard of care brain tumour management.

Utilizing Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) to Analyze Interstitial Fluid Flow and Transport in Glioblastoma and the Surrounding Parenchyma in Human Patients

Background: Glioblastoma (GBM) is the deadliest and most common brain tumor in adults, with poor survival and response to aggressive therapy. Limited access of drugs to tumor cells is one reason for such grim clinical outcomes. A driving force for therapeutic delivery is interstitial fluid flow (IFF), both within the tumor and in the surrounding brain parenchyma. However, convective and diffusive transport mechanisms are understudied. In this study, we examined the application of a novel image analysis method to measure fluid flow and diffusion in GBM patients. Methods: Here, we applied an imaging methodology that had been previously tested and validated in vitro, in silico, and in preclinical models of disease to archival patient data from the Ivy Glioblastoma Atlas Project (GAP) dataset. The analysis required the use of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), which is readily available in the database. The analysis results, which consisted of IFF flow velocity and diffusion coefficients, were then compared to patient outcomes such as survival. Results: We characterized IFF and diffusion patterns in patients. We found strong correlations between flow rates measured within tumors and in the surrounding parenchymal space, where we hypothesized that velocities would be higher. Analyzing overall magnitudes indicated a significant correlation with both age and survival in this patient cohort. Additionally, we found that neither tumor size nor resection significantly altered the velocity magnitude. Lastly, we mapped the flow pathways in patient tumors and found a variability in the degree of directionality that we hypothesize may lead to information concerning treatment, invasive spread, and progression in future studies. Conclusions: An analysis of standard DCE-MRI in patients with GBM offers more information regarding IFF and transport within and around the tumor, shows that IFF is still detected post-resection, and indicates that velocity magnitudes correlate with patient prognosis.

Characterization of dysregulated glutamine metabolism in human glioma tissue with 1H NMR

Gliomas are one of the most common types of brain tumors. Given low survival and high treatment resistance rates, particularly for high grade gliomas, there is a need for specific biomarkers that can be used to stratify patients for therapy and monitor treatment response. Recent work has demonstrated that metabolic reprogramming, often mediated by inflammation, can lead to an upregulation of glutamine as an energy source for cancer cells. As a result, glutamine pathways are an emerging pharmacologic target. The goal of this pilot study was to characterize changes in glutamine metabolism and inflammation in human glioma samples and explore the use of glutamine as a potential biomarker. ¹H high-resolution magic angle spinning nuclear magnetic resonance spectra were acquired from ex vivo glioma tissue (n = 16, grades II–IV) to quantify metabolite concentrations. Tumor inflammatory markers were quantified using electrochemiluminescence assays. Glutamate, glutathione, lactate, and alanine, as well as interleukin (IL)-1β and IL-8, increased significantly in samples from grade IV gliomas compared to grades II and III (p ≤ .05). Following dimension reduction of the inflammatory markers using probabilistic principal component analysis, we observed that glutamine, alanine, glutathione, and lactate were positively associated with the first inflammatory marker principal component. Our findings support the hypothesis that glutamine may be a key marker for glioma progression and indicate that inflammation is associated with changes in glutamine metabolism. These results motivate further in vivo investigation of glutamine as a biomarker for tumor progression and treatment response.

Correlations between DTI-derived metrics and MRS metabolites in tumour regions of glioblastoma: a pilot study

Introduction

Specific correlations among diffusion tensor imaging (DTI)-derived metrics and magnetic resonance spectroscopy (MRS) metabolite ratios in brains with glioblastoma are still not completely understood.

Patients and methods

We made retrospective cohort study. MRS ratios (choline-to-N-acetyl aspartate [Cho/NAA], lipids and lactate to creatine [LL/Cr], and myo-inositol/creatine [mI/Cr]) were correlated with eleven DTI biomarkers: mean diffusivity (MD), fractional anisotropy (FA), pure isotropic diffusion (p), pure anisotropic diffusion (q), the total magnitude of the diffusion tensor (L), linear tensor (Cl), planar tensor (Cp), spherical tensor (Cs), relative anisotropy (RA), axial diffusivity (AD) and radial diffusivity (RD) at the same regions: enhanced rim, peritumoral oedema and normal-appearing white matter. Correlational analyses of 546 MRS and DTI measurements used Spearman coefficient.

Results

At the enhancing rim we found four significant correlations: FA ⇔ LL/Cr, Rs = -.364, p = .034; Cp ⇔ LL/Cr, Rs = .362, p = .035; q ⇔ LL/Cr, Rs = -.349, p = .035; RA ⇔ LL/Cr, Rs = -.357, p = .038. Another ten pairs of significant correlations were found in the peritumoral edema: AD ⇔ LL/Cr, AD ⇔ mI/Cr, MD ⇔ LL/Cr, MD ⇔ mI/Cr, p ⇔ LL/Cr, p ⇔ mI/ Cr, RD ⇔ mI/Cr, RD ⇔ mI/Cr, L ⇔ LL/Cr, L ⇔ mI/Cr.

Conclusions

DTI and MRS biomarkers answer different questions; peritumoral oedema represents the biggest challenge with at least ten significant correlations between DTI and MRS that need additional studies. The fact that DTI and MRS measures are not specific of one histologic type of tumour broadens their application to a wider variety of intracranial pathologies.

Imaging growth as a predictor of grade of malignancy and aggressiveness of IDH-mutant and 1p/19q-codeleted oligodendrogliomas in adults

BACKGROUND

We quantified the spontaneous imaging growth rate of oligodendrogliomas. We assessed whether (i) it discriminates between World Health Organization (WHO) grade II and grade III oligodendrogliomas, and (ii) grade III oligodendrogliomas with neo-angiogenesis are associated with more fast growth rates (≥8 mm/y).

METHODS

This work employed a retrospective bicentric cohort study (2010-2016) of adult patients harboring a newly diagnosed supratentorial oligodendroglioma, isocitrate dehydrogenase (IDH) mutant and 1p/19q codeleted (WHO 2016 classification), with a minimum of 2 available MRIs before any treatment (minimum 6-week interval) to measure the spontaneous tumor growth rate.

RESULTS

We included 108 patients (age 44.7 ± 14.1 y, 60 males). The tumor growth rate was higher in grade III oligodendrogliomas with neo-angiogenesis (n = 37, median 10.4 mm/y, mean 10.0 ± 6.9) than in grade III oligodendrogliomas with increased mitosis count only (cutoff ≥6 mitoses, n = 18, median 3.9 mm/y, mean 4.5 ± 3.2; P = 0.004), and higher than in grade II oligodendrogliomas (n = 53, median 2.3 mm/y, mean 2.8 ± 2.2; P < 0.001). There was increased prevalence of fast tumor growth rates in grade III oligodendrogliomas with neo-angiogenesis (54.1%) compared with grade III oligodendrogliomas with increased mitosis count only (11.1%; P < 0.001), and in grade II oligodendrogliomas (0.0%; P < 0.001). The tumor growth rate trends did not differ between centers (P = 0.121). Neo-angiogenesis (P < 0.001) and mitosis count at ≥9 (P = 0.013) were independently associated with tumor growth rates ≥8 mm/year. A tumor growth rate ≥8 mm/year was the only predictor independently associated with shorter progression-free survival (P = 0.041).

CONCLUSIONS

The spontaneous tumor growth rate recapitulates oligodendroglioma aggressiveness, permits identification of grade III oligodendrogliomas preoperatively when ≥8 mm/year, and questions the grading by mitosis count.

Biological washout modelling for in-beam PET: rabbit brain irradiation by 11C and 15O ion beams

Positron emission tomography (PET) has been used for dose verification in charged particle therapy. The causes of washout of positron emitters by physiological functions should be clarified for accurate dose verification. In this study, we visualized the distribution of irradiated radioactive beams, ¹¹C and ¹⁵O beams, in the rabbit whole-body using our original depth-of-interaction (DOI)-PET prototype to add basic data for biological washout effect correction. Time activity curves of the irradiated field and organs were measured immediately after the irradiations. All data were corrected for physical decay before further analysis. We also collected expired gas of the rabbit during beam irradiation and the energy spectrum was measured with a germanium detector. Irradiated radioactive beams into the brain were distributed to the whole body due to the biological washout process, and the implanted ¹¹C and ¹⁵O ions were concentrated in the regions which had high blood volume. The ¹¹C-labelled ¹¹CO₂ was detected in expired gas under the ¹¹C beam irradiation, while no significant signal was detected under the ¹⁵O beam irradiation as a form of C¹⁵O₂. Results suggested that the implanted ¹¹C ions form molecules that diffuse out to the whole body by undergoing perfusion, then, they are incorporated into the blood-gas exchange in the respiratory system. This study provides basic data for modelling of the biological washout effect.

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.

Antibody-guided nanomedicines as novel breakthrough therapeutic, diagnostic and theranostic tools

Recent advances in nanotechnology, such as the development of various types of nanoparticles and hybrid nanomaterials, have revolutionized nanomedicine. The small size, customizable surface, enhanced solubility, and multi-functionality endow the nanoparticles with an ability to interact with complex cellular and biological functions in new ways. Furthermore, these systems can deliver drugs to specific tissues and provide a targeted therapy. For this purpose, different categories of molecules, particularly antibodies, have been used as ligands. Antibody-conjugated nanomaterials can significantly enhance the efficiency of nanomedicines, especially in the field of cancer. This review is focused on three major medical applications of antibody-conjugated nanomaterials, namely, therapeutic, diagnostic and theranostic applications. To provide comprehensive information on the topic and an overview of these hybrid nanomaterials for biomedical applications, a brief summary of nanomaterials and antibodies is given. Moreover, the review has depicted the potential applications of antibody-conjugated nanomaterials in different fields and their capabilities to empower nanomedicine, particularly in relation to the treatment and detection of malignancies.

Predicting survival from colorectal cancer histology slides using deep learning: A retrospective multicenter study

BACKGROUND

For virtually every patient with colorectal cancer (CRC), hematoxylin-eosin (HE)-stained tissue slides are available. These images contain quantitative information, which is not routinely used to objectively extract prognostic biomarkers. In the present study, we investigated whether deep convolutional neural networks (CNNs) can extract prognosticators directly from these widely available images.

METHODS AND FINDINGS

We hand-delineated single-tissue regions in 86 CRC tissue slides, yielding more than 100,000 HE image patches, and used these to train a CNN by transfer learning, reaching a nine-class accuracy of >94% in an independent data set of 7,180 images from 25 CRC patients. With this tool, we performed automated tissue decomposition of representative multitissue HE images from 862 HE slides in 500 stage I-IV CRC patients in the The Cancer Genome Atlas (TCGA) cohort, a large international multicenter collection of CRC tissue. Based on the output neuron activations in the CNN, we calculated a "deep stroma score," which was an independent prognostic factor for overall survival (OS) in a multivariable Cox proportional hazard model (hazard ratio [HR] with 95% confidence interval [CI]: 1.99 [1.27-3.12], p = 0.0028), while in the same cohort, manual quantification of stromal areas and a gene expression signature of cancer-associated fibroblasts (CAFs) were only prognostic in specific tumor stages. We validated these findings in an independent cohort of 409 stage I-IV CRC patients from the "Darmkrebs: Chancen der Verhütung durch Screening" (DACHS) study who were recruited between 2003 and 2007 in multiple institutions in Germany. Again, the score was an independent prognostic factor for OS (HR 1.63 [1.14-2.33], p = 0.008), CRC-specific OS (HR 2.29 [1.5-3.48], p = 0.0004), and relapse-free survival (RFS; HR 1.92 [1.34-2.76], p = 0.0004). A prospective validation is required before this biomarker can be implemented in clinical workflows.

CONCLUSIONS

In our retrospective study, we show that a CNN can assess the human tumor microenvironment and predict prognosis directly from histopathological images.

Automated Modular Magnetic Resonance Imaging Clinical Decision Support System (MIROR): An Application in Pediatric Cancer Diagnosis

BACKGROUND

Advances in magnetic resonance imaging and the introduction of clinical decision support systems has underlined the need for an analysis tool to extract and analyze relevant information from magnetic resonance imaging data to aid decision making, prevent errors, and enhance health care.

OBJECTIVE

The aim of this study was to design and develop a modular medical image region of interest analysis tool and repository (MIROR) for automatic processing, classification, evaluation, and representation of advanced magnetic resonance imaging data.

METHODS

The clinical decision support system was developed and evaluated for diffusion-weighted imaging of body tumors in children (cohort of 48 children, with 37 malignant and 11 benign tumors). Mevislab software and Python have been used for the development of MIROR. Regions of interests were drawn around benign and malignant body tumors on different diffusion parametric maps, and extracted information was used to discriminate the malignant tumors from benign tumors.

RESULTS

Using MIROR, the various histogram parameters derived for each tumor case when compared with the information in the repository provided additional information for tumor characterization and facilitated the discrimination between benign and malignant tumors. Clinical decision support system cross-validation showed high sensitivity and specificity in discriminating between these tumor groups using histogram parameters.

CONCLUSIONS

MIROR, as a diagnostic tool and repository, allowed the interpretation and analysis of magnetic resonance imaging images to be more accessible and comprehensive for clinicians. It aims to increase clinicians' skillset by introducing newer techniques and up-to-date findings to their repertoire and make information from previous cases available to aid decision making. The modular-based format of the tool allows integration of analyses that are not readily available clinically and streamlines the future developments.

Diagnostic Value of Fractal Analysis for the Differentiation of Brain Tumors Using 3-Tesla Magnetic Resonance Susceptibility-Weighted Imaging

BACKGROUND

Susceptibility-weighted imaging (SWI) of brain tumors provides information about neoplastic vasculature and intratumoral micro- and macrobleedings. Low- and high-grade gliomas can be distinguished by SWI due to their different vascular characteristics. Fractal analysis allows for quantification of these radiological differences by a computer-based morphological assessment of SWI patterns.

OBJECTIVE

To show the feasibility of SWI analysis on 3-T magnetic resonance imaging to distinguish different kinds of brain tumors.

METHODS

Seventy-eight patients affected by brain tumors of different histopathology (low- and high-grade gliomas, metastases, meningiomas, lymphomas) were included. All patients underwent preoperative 3-T magnetic resonance imaging including SWI, on which the lesions were contoured. The images underwent automated computation, extracting 2 quantitative parameters: the volume fraction of SWI signals within the tumors (signal ratio) and the morphological self-similar features (fractal dimension [FD]). The results were then correlated with each histopathological type of tumor.

RESULTS

Signal ratio and FD were able to differentiate low-grade gliomas from grade III and IV gliomas, metastases, and meningiomas (P < .05). FD was statistically different between lymphomas and high-grade gliomas (P < .05). A receiver-operating characteristic analysis showed that the optimal cutoff value for differentiating low- from high-grade gliomas was 1.75 for FD (sensitivity, 81%; specificity, 89%) and 0.03 for signal ratio (sensitivity, 80%; specificity, 86%).

CONCLUSION

FD of SWI on 3-T magnetic resonance imaging is a novel image biomarker for glioma grading and brain tumor characterization. Computational models offer promising results that may improve diagnosis and open perspectives in the radiological assessment of brain tumors.

ABBREVIATIONS

FD, fractal dimensionSR, signal ratioSWI, susceptibility-weighted imaging.

Diagnostic Accuracy of Neuroimaging to Delineate Diffuse Gliomas within the Brain: A Meta-Analysis

BACKGROUND

Brain imaging in diffuse glioma is used for diagnosis, treatment planning, and follow-up.

PURPOSE

In this meta-analysis, we address the diagnostic accuracy of imaging to delineate diffuse glioma.

DATA SOURCES

We systematically searched studies of adults with diffuse gliomas and correlation of imaging with histopathology.

STUDY SELECTION

Study inclusion was based on quality criteria. Individual patient data were used, if available.

DATA ANALYSIS

A hierarchic summary receiver operating characteristic method was applied. Low- and high-grade gliomas were analyzed in subgroups.

DATA SYNTHESIS

Sixty-one studies described 3532 samples in 1309 patients. The mean Standard for Reporting of Diagnostic Accuracy score (13/25) indicated suboptimal reporting quality. For diffuse gliomas as a whole, the diagnostic accuracy was best with T2-weighted imaging, measured as area under the curve, false-positive rate, true-positive rate, and diagnostic odds ratio of 95.6%, 3.3%, 82%, and 152. For low-grade gliomas, the diagnostic accuracy of T2-weighted imaging as a reference was 89.0%, 0.4%, 44.7%, and 205; and for high-grade gliomas, with T1-weighted gadolinium-enhanced MR imaging as a reference, it was 80.7%, 16.8%, 73.3%, and 14.8. In high-grade gliomas, MR spectroscopy (85.7%, 35.0%, 85.7%, and 12.4) and ¹¹C methionine-PET (85.1%, 38.7%, 93.7%, and 26.6) performed better than the reference imaging.

LIMITATIONS

True-negative samples were underrepresented in these data, so false-positive rates are probably less reliable than true-positive rates. Multimodality imaging data were unavailable.

CONCLUSIONS

The diagnostic accuracy of commonly used imaging is better for delineation of low-grade gliomas than high-grade gliomas on the basis of limited evidence. Improvement is indicated from advanced techniques, such as MR spectroscopy and PET.

Multiparametric MRI and [18F]Fluorodeoxyglucose Positron Emission Tomography Imaging Is a Potential Prognostic Imaging Biomarker in Recurrent Glioblastoma

Purpose/objectives

Multiparametric advanced MR and [¹⁸F]fluorodeoxyglucose (FDG)-positron emission tomography (PET) imaging may be important biomarkers for prognosis as well for distinguishing recurrent glioblastoma multiforme (GBM) from treatment-related changes.

Methods/materials

We retrospectively evaluated 30 patients treated with chemoradiation for GBM and underwent advanced MR and FDG-PET for confirmation of tumor progression. Multiparametric MRI and FDG-PET imaging metrics were evaluated for their association with 6-month overall (OS) and progression-free survival (PFS) based on pathological, radiographic, and clinical criteria.

Results

17 males and 13 females were treated between 2001 and 2014, and later underwent FDG-PET at suspected recurrence. Baseline FDG-PET and MRI imaging was obtained at a median of 7.5 months [interquartile range (IQR) 3.7–12.4] following completion of chemoradiation. Median follow-up after FDG-PET imaging was 10 months (IQR 7.2–13.0). Receiver-operator characteristic curve analysis identified that lesions characterized by a ratio of the SUV_max to the normal contralateral brain (SUV_max/NB index) >1.5 and mean apparent diffusion coefficient (ADC) value of ≤1,400 × 10⁻⁶ mm²/s correlated with worse 6-month OS and PFS. We defined three patient groups that predicted the probability of tumor progression: SUV_max/NB index >1.5 and ADC ≤1,400 × 10⁻⁶ mm²/s defined high-risk patients (n = 7), SUV_max/NB index ≤1.5 and ADC >1,400 × 10⁻⁶ mm²/s defined low-risk patients (n = 11), and intermediate-risk (n = 12) defined the remainder of the patients. Median OS following the time of the FDG-PET scan for the low, intermediate, and high-risk groups were 23.5, 10.5, and 3.8 months (p < 0.01). Median PFS were 10.0, 4.4, and 1.9 months (p = 0.03). Rates of progression at 6-months in the low, intermediate, and high-risk groups were 36, 67, and 86% (p = 0.04).

Conclusion

Recurrent GBM in the molecular era is associated with highly variable outcomes. Multiparametric MR and FDG-PET biomarkers may provide a clinically relevant, non-invasive and cost-effective method of predicting prognosis and improving clinical decision making in the treatment of patients with suspected tumor recurrence.

Brain Tumors: An Update on Clinical PET Research in Gliomas

A previous review published in 2012 demonstrated the role of clinical PET for diagnosis and management of brain tumors using mainly FDG, amino acid tracers, and ¹⁸F-fluorothymidine. This review provides an update on clinical PET studies, most of which are motivated by prediction of prognosis and planning and monitoring of therapy in gliomas. For FDG, there has been additional evidence supporting late scanning, and combination with ¹³N ammonia has yielded some promising results. Large neutral amino acid tracers have found widespread applications mostly based on ¹⁸F-labeled compounds fluoroethyltyrosine and fluorodopa for targeting biopsies, therapy planning and monitoring, and as outcome markers in clinical trials. ¹¹C-alpha-methyltryptophan (AMT) has been proposed as an alternative to ¹¹C-methionine, and there may also be a role for cyclic amino acid tracers. ¹⁸F-fluorothymidine has shown strengths for tumor grading and as an outcome marker. Studies using ¹⁸F-fluorocholine (FCH) and ⁶⁸Ga-labeled compounds are promising but have not yet clearly defined their role. Studies on radiotherapy planning have explored the use of large neutral amino acid tracers to improve the delineation of tumor volume for irradiation and the use of hypoxia markers, in particular ¹⁸F-fluoromisonidazole. Many studies employed the combination of PET with advanced multimodal MR imaging methods, mostly demonstrating complementarity and some potential benefits of hybrid PET/MR.

Application of Dynamic Contrast-Enhanced MRI Parameters for Differentiating Squamous Cell Carcinoma and Malignant Lymphoma of the Oropharynx

OBJECTIVE

The purpose of this study was to investigate the usefulness of histogram analysis of dynamic contrast-enhanced MRI (DCE-MRI) parameters for the differentiation of squamous cell carcinoma (SCC) and malignant lymphoma of the oropharynx.

MATERIALS AND METHODS

Pretreatment DCE-MRI was performed in 21 patients with pathologically confirmed oropharyngeal SCC and six patients with malignant lymphoma. DCE-MRI parameter maps including the volume transfer constant (K(trans)), flux rate constant (kep), and extravascular extracellular volume fraction (ve) based on the Tofts model were obtained. Enhancing tumors were manually segmented on each slice of the parameter maps, and the data were collected to obtain a histogram for the entire tumor volume. The Wilcoxon rank sum test was used to compare the histogram parameters of each DCE-MRI-derived variable of oropharyngeal SCC and lymphoma.

RESULTS

Histogram analysis of K(trans) and ve maps revealed that the median and mode of K(trans) were significantly higher in SCC than in lymphoma (p = 0.039 and 0.032, respectively), and the mode, skewness, and kurtosis of ve were significantly different in SCC than in lymphoma (p = 0.046, 0.039, and 0.032, respectively). On ROC analysis, the kurtosis of ve had the best discriminative value for distinguishing between oropharyngeal SCC and lymphoma (AUC, 0.865; cutoff value, 2.60; sensitivity, 83.3%; specificity, 90.5%).

CONCLUSION

Our preliminary evidence using histogram analysis of DCE-MRI parameters based on the whole tumor volume suggests that it might be useful for differentiating SCC from malignant lymphoma of the oropharynx.

Imaging Tumor Vascularity and Response to Anti-Angiogenic Therapy Using Gaussia Luciferase

We developed a novel approach to assess tumor vascularity using recombinant Gaussia luciferase (rGluc) protein and bioluminescence imaging. Upon intravenous injection of rGluc followed by its substrate coelenterazine, non-invasive visualization of tumor vascularity by bioluminescence imaging was possible. We applied this method for longitudinal monitoring of tumor vascularity in response to the anti-angiogenic drug tivozanib. This simple and sensitive method could be extended to image blood vessels/vasculature in many different fields.

Clinical Intravoxel Incoherent Motion and Diffusion MR Imaging: Past, Present, and Future

The concept of diffusion magnetic resonance (MR) imaging emerged in the mid-1980s, together with the first images of water diffusion in the human brain, as a way to probe tissue structure at a microscopic scale, although the images were acquired at a millimetric scale. Since then, diffusion MR imaging has become a pillar of modern clinical imaging. Diffusion MR imaging has mainly been used to investigate neurologic disorders. A dramatic application of diffusion MR imaging has been acute brain ischemia, providing patients with the opportunity to receive suitable treatment at a stage when brain tissue might still be salvageable, thus avoiding terrible handicaps. On the other hand, it was found that water diffusion is anisotropic in white matter, because axon membranes limit molecular movement perpendicularly to the nerve fibers. This feature can be exploited to produce stunning maps of the orientation in space of the white matter tracts and brain connections in just a few minutes. Diffusion MR imaging is now also rapidly expanding in oncology, for the detection of malignant lesions and metastases, as well as monitoring. Water diffusion is usually largely decreased in malignant tissues, and body diffusion MR imaging, which does not require any tracer injection, is rapidly becoming a modality of choice to detect, characterize, or even stage malignant lesions, especially for breast or prostate cancer. After a brief summary of the key methodological concepts beyond diffusion MR imaging, this article will give a review of the clinical literature, mainly focusing on current outstanding issues, followed by some innovative proposals for future improvements.

Application of iron oxide nanoparticles in glioma imaging and therapy: from bench to bedside

Gliomas are the most common primary brain tumors and have a very dismal prognosis. However, recent advancements in nanomedicine and nanotechnology provide opportunities for personalized treatment regimens to improve the poor prognosis of patients suffering from glioma. This comprehensive review starts with an outline of the current status facing glioma. It then provides an overview of the state-of-the-art applications of iron oxide nanoparticles (IONPs) to glioma diagnostics and therapeutics, including MR contrast enhancement, drug delivery, cell labeling and tracking, magnetic hyperthermia treatment and magnetic particle imaging. It also addresses current challenges associated with the biological barriers and IONP design with an emphasis on recent advances and innovative approaches for glioma targeting strategies. Opportunities for future development are highlighted.

Whole-brain spectroscopic MRI biomarkers identify infiltrating margins in glioblastoma patients

BACKGROUND

The standard of care for glioblastoma (GBM) is maximal safe resection followed by radiation therapy with chemotherapy. Currently, contrast-enhanced MRI is used to define primary treatment volumes for surgery and radiation therapy. However, enhancement does not identify the tumor entirely, resulting in limited local control. Proton spectroscopic MRI (sMRI), a method reporting endogenous metabolism, may better define the tumor margin. Here, we develop a whole-brain sMRI pipeline and validate sMRI metrics with quantitative measures of tumor infiltration.

METHODS

Whole-brain sMRI metabolite maps were coregistered with surgical planning MRI and imported into a neuronavigation system to guide tissue sampling in GBM patients receiving 5-aminolevulinic acid fluorescence-guided surgery. Samples were collected from regions with metabolic abnormalities in a biopsy-like fashion before bulk resection. Tissue fluorescence was measured ex vivo using a hand-held spectrometer. Tissue samples were immunostained for Sox2 and analyzed to quantify the density of staining cells using a novel digital pathology image analysis tool. Correlations among sMRI markers, Sox2 density, and ex vivo fluorescence were evaluated.

RESULTS

Spectroscopic MRI biomarkers exhibit significant correlations with Sox2-positive cell density and ex vivo fluorescence. The choline to N-acetylaspartate ratio showed significant associations with each quantitative marker (Pearson's ρ = 0.82, P < .001 and ρ = 0.36, P < .0001, respectively). Clinically, sMRI metabolic abnormalities predated contrast enhancement at sites of tumor recurrence and exhibited an inverse relationship with progression-free survival.

CONCLUSIONS

As it identifies tumor infiltration and regions at high risk for recurrence, sMRI could complement conventional MRI to improve local control in GBM patients.

Monoclonal antibody-conjugated superparamagnetic iron oxide nanoparticles for imaging of epidermal growth factor receptor-targeted cells and gliomas

The objective of this study was to successfully synthesize epidermal growth factor receptor monoclonal antibody-conjugated superparamagnetic iron oxide nanoparticles (EGFRmAb-SPIONs) and explore their biocompatibility and potential applications as a targeted magnetic resonance imaging (MRI) contrast agent for the EGFR-specific detection of brain glioma in vivo. After conjugation of EGFRmAb with SPIONs, the magnetic characteristics of EGFRmAb-SPIONs were investigated. Thereafter, the targeting abilities of EGFRmAb-SPIONs with MRI were qualitatively and quantitatively assessed in EGFR-positive C6 glioma cells in vitro and in a Wistar rat model bearing C6 glioma in vivo. Furthermore, the preliminary biocompatibility and toxicity of EGFRmAb-SPIONs were evaluated in normal rats through hematology assays and histopathologic analyses. Statistical analysis was performed using one-way analysis of variance and Student t-test, with a significance level of p < .05. From the results of EGFRmAb-SPION characterizations, the average particle size was 10.21 nm and the hydrodynamic diameter was 161.5 ± 2.12 nm. The saturation magnetization was 55 emu/g·Fe, and T2 relaxivity was 92.73 s-1mM-1 in distilled water. The preferential accumulation of the EGFRmAb-SPIONs within glioma and subsequent MRI contrast enhancement were demonstrated both in vitro in C6 cells and in vivo in rats bearing C6 glioma. After intravenous administration of EGFRmAb-SPIONs, T2-weighted MRI of the rat model with brain glioma exhibited an apparent hypointense region within glioma from 2 to 48 hours. The maximal image contrast was reached at 24 hours, where the signal intensity decreased and the R2 value increased by 30% compared to baseline. However, T2-weighted imaging of the rat model administered with SPIONs showed no visible signal changes within the tumor over the same time period. Moreover, no evident toxicities in vitro and in vivo with EGFRmAb-SPIONs were clearly identified based on the laboratory examinations. EGFRmAb-SPIONs could potentially be employed as a targeted contrast agent in the molecule-specific diagnosis of brain glioma in MRI.

Pretargeted Positron Emission Tomography Imaging That Employs Supramolecular Nanoparticles with in Vivo Bioorthogonal Chemistry

A pretargeted oncologic positron emission tomography (PET) imaging that leverages the power of supramolecular nanoparticles with in vivo bioorthogonal chemistry was demonstrated for the clinically relevant problem of tumor imaging. The advantages of this approach are that (i) the pharmacokinetics (PKs) of tumor-targeting and imaging agents can be independently altered via chemical alteration to achieve the desired in vivo performance and (ii) the interplay between the two PKs and other controllable variables confers a second layer of control toward improved PET imaging. In brief, we utilized supramolecular chemistry to synthesize tumor-targeting nanoparticles containing transcyclooctene (TCO, a bioorthogonal reactive motif), called TCO⊂SNPs. After the intravenous injection and subsequent concentration of the TCO⊂SNPs in the tumors of living mice, a small molecule containing both the complementary bioorthogonal motif (tetrazine, Tz) and a positron-emitting radioisotope ((64)Cu) was injected to react selectively and irreversibly to TCO. High-contrast PET imaging of the tumor mass was accomplished after the rapid clearance of the unreacted (64)Cu-Tz probe. Our nanoparticle approach encompasses a wider gamut of tumor types due to the use of EPR effects, which is a universal phenomenon for most solid tumors.

Principles of pharmacotherapy

This chapter will review the challenges in pharmacotherapy in primary brain tumors that include the presence of the blood-brain barrier, a blood-tumor barrier, active drug efflux pumps, and high plasma protein binding of agents. The approaches to improve the delivery of drugs to the brain will be discussed. Often the management of brain tumors involves the use of corticosteroids and enzyme-inducing antiseizure medications that can have significant drug interactions that may impact the efficacy or toxicity of drugs used to treat these patients. Various techniques used to assess drug distribution to the brain will be reviewed.

Correlation of (18)F-fluoroethyl tyrosine positron-emission tomography uptake values and histomorphological findings by stereotactic serial biopsy in newly diagnosed brain tumors using a refined software tool

Background

Magnetic resonance imaging (MRI) is the standard neuroimaging method to diagnose neoplastic brain lesions, as well as to perform stereotactic biopsy surgical planning. MRI has the advantage of providing structural anatomical details with high sensitivity, though histological specificity is limited. Although combining MRI with other imaging modalities, such as positron-emission tomography (PET), has proven to increment specificity, exact correlation between PET threshold uptake ratios (URs) and histological diagnosis and grading has not yet been described.

Objectives

The aim of this study was to correlate exactly the histopathological criteria of the biopsy site to its PET uptake value with high spatial resolution (mm³), and to analyze the diagnostic value of PET using the amino acid O-(2-[¹⁸F]fluoroethyl)-l-tyrosine (¹⁸F-FET) PET in patients with newly diagnosed brain lesions in comparison to histological findings obtained from stereotactic serial biopsy.

Patients and methods

A total of 23 adult patients with newly diagnosed brain tumors on MRI were enrolled in this study. Subsequently to diagnoses, all patients underwent a ¹⁸F-FET PET-guided stereotactic biopsy, using an original newly developed software module, which is presented here. Conventional MRI, stereotactic computed tomography series, and ¹⁸F-FET PET images were semiautomatically fused, and hot-spot detection was performed for target planning. UR was determined using the uptake value from the biopsy sites in relation to the contralateral frontal white matter. UR values ≥1.6 were considered positive for glioma. High-grade glioma (HGG) was suspected with URs ≥3.0, while low-grade glioma (LGG) was suspected with URs between 1.6 and 3.0. Stereotactic serial biopsies along the trajectory at multiple sites were performed in millimeter steps, and the FET URs for each site were correlated exactly with a panel of 27 different histopathological markers. Comparisons between FET URs along the biopsy trajectories and the histological diagnoses were made with Pearson product-moment correlation coefficients. Analysis of variance was performed to test for significant differences in maximum UR between different tumor grades.

Results

A total of 363 biopsy specimens were taken from 23 patients by stereotactic serial biopsies. Histological examination revealed eight patients (35%) with an LGG: one with a World Health Organization (WHO)-I lesion and seven with a WHO-II lesion. Thirteen (57%) patients revealed an HGG (two with a WHO-III and three with a WHO-IV tumor), and two patients (9%) showed a process that was neither HGG nor LGG (group X or no-grade group). The correlation matrix between histological findings and the UR revealed five strong correlations. Low cell density in tissue samples was found to have a significant negative correlation with the measured cortical uptake rate (r=−0.43, P=0.02), as well as moderate cell density (r=−0.48, P=0.02). Pathological patterns of proliferation (r=0.37, P=0.04), GFAP (r=0.37, P=0.04), and Olig2 (r=0.36, P=0.05) showed a significant positive correlation with cortical URs. Analysis of variance tests showed a significant difference between the LGG and the HGG groups (F=8.27, P<0.002), but no significant differences when differentiating between the X group and the HGG (P=0.2)/LGG (P=0.8) groups, nor between the no-grade group and the WHO-I group.

Conclusion

¹⁸F-FET PET is a valuable tool, as it allows the differentiation of HGGs from LGGs. Its use is not limited to preoperative evaluation; it may also refine biopsy targeting and improve tumor delimitation for radiotherapy. Histology is still necessary, and remains the gold standard for definitive diagnosis of brain lesions.

Simultaneous evaluation of brain tumour metabolism, structure and blood volume using [(18)F]-fluoroethyltyrosine (FET) PET/MRI: feasibility, agreement and initial experience

PURPOSE

Both [(18)F]-fluoroethyltyrosine (FET) PET and blood volume (BV) MRI supplement routine T1-weighted contrast-enhanced MRI in gliomas, but whether the two modalities provide identical or complementary information is unresolved. The aims of the study were to investigate the feasibility of simultaneous structural MRI, BV MRI and FET PET of gliomas using an integrated PET/MRI scanner and to assess the spatial and quantitative agreement in tumour imaging between BV MRI and FET PET.

METHODS

A total of 32 glioma patients underwent a 20-min static simultaneous PET/MRI acquisition on a Siemens mMR system 20 min after injection of 200 MBq FET. The MRI protocol included standard structural MRI and dynamic susceptibility contrast (DSC) imaging for BV measurements. Maximal relative tumour FET uptake (TBRmax) and BV (rBVmax), and Dice coefficients were calculated to assess the quantitative and spatial congruence in the tumour volumes determined by FET PET, BV MRI and contrast-enhanced MRI.

RESULTS

FET volume and TBRmax were higher in BV-positive than in BV-negative scans, and both VOLBV and rBVmax were higher in FET-positive than in FET-negative scans. TBRmax and rBVmax were positively correlated (R (2) = 0.59, p < 0.001). FET and BV positivity were in agreement in only 26 of the 32 patients and in 42 of 63 lesions, and spatial congruence in the tumour volumes as assessed by the Dice coefficients was generally poor with median Dice coefficients exceeding 0.1 in less than half the patients positive on at least one modality for any pair of modalities. In 56 % of the patients susceptibility artefacts in DSC BV maps overlapped the tumour on MRI.

CONCLUSION

The study demonstrated that although tumour volumes determined by BV MRI and FET PET were quantitatively correlated, their spatial congruence in a mixed population of treated glioma patients was generally poor, and the modalities did not provide the same information in this population of patients. Combined imaging of brain tumour metabolism and perfusion using hybrid PET/MR systems may provide complementary information on tumour biology, but the potential clinical value remains to be determined in future trials.

The 18-kDa mitochondrial translocator protein in gliomas: from the bench to bedside

The 18-kDa mitochondrial translocator protein (TSPO) is known to be highly expressed in several types of cancer, including gliomas, whereas expression in normal brain is low. TSPO functions in glioma are still incompletely understood. The TSPO can be quantified pre-operatively with molecular imaging making it an ideal candidate for personalized treatment of patient with glioma. Studies have proposed to exploit the TSPO as a transporter of chemotherapics to selectively target tumour cells in the brain. Our studies proved that positron emission tomography (PET)-imaging can contribute to predict progression of patients with glioma and that molecular imaging with TSPO-specific ligands is suitable to stratify patients in view of TSPO-targeted treatment. Finally, we proved that TSPO in gliomas is predominantly expressed by tumour cells.

Lessons from anti-vascular endothelial growth factor and anti-vascular endothelial growth factor receptor trials in patients with glioblastoma

Treatment of glioblastoma (GBM), the most common primary malignant brain tumor in adults, remains a significant unmet need in oncology. Historically, cytotoxic treatments provided little durable benefit, and tumors recurred within several months. This has spurred a substantial research effort to establish more effective therapies for both newly diagnosed and recurrent GBM. In this context, antiangiogenic therapy emerged as a promising treatment strategy because GBMs are highly vascular tumors. In particular, GBMs overexpress vascular endothelial growth factor (VEGF), a proangiogenic cytokine. Indeed, many studies have demonstrated promising radiographic response rates, delayed tumor progression, and a relatively safe profile for anti-VEGF agents. However, randomized phase III trials conducted to date have failed to show an overall survival benefit for antiangiogenic agents alone or in combination with chemoradiotherapy. These results indicate that antiangiogenic agents may not be beneficial in unselected populations of patients with GBM. Unfortunately, biomarker development has lagged behind in the process of drug development, and no validated biomarker exists for patient stratification. However, hypothesis-generating data from phase II trials that reveal an association between increased perfusion and/or oxygenation (ie, consequences of vascular normalization) and survival suggest that early imaging biomarkers could help identify the subset of patients who most likely will benefit from anti-VEGF agents. In this article, we discuss the lessons learned from the trials conducted to date and how we could potentially use recent advances in GBM biology and imaging to improve outcomes of patients with GBM who receive antiangiogenic therapy.

Surgical management of high-grade glioma: a standard of care

Neurosurgical oncology for intrinsic glioma is evolving rapidly. It must be patient-centered, consultant-led and research-orientated. The value of specialist neurosurgical engagement is becoming more widely recognized. Detailed evaluation tailored to each patient is essential before the surgical admission, in conjunction with clinical oncology input. Medical optimization, collation of magnetic resonance datasets for preoperative planning and providing an informed explanation of the proposed management and its alternatives are all part of the neurosurgeon's remit. Meticulous microsurgical technique during surgery utilizing modern neuronavigation and physiological monitoring are integral components of the specialist armamentarium. A clear understanding of the rationale for surgical intervention, including its place alongside radiotherapy and chemotherapy, informs surgical decision-making. Recognition and understanding of these issues are driving the evolution of neurosurgical management of high-grade glioma. New challenges are emerging and need to be critically evaluated in robustly designed clinical trials.

Imaging biomarkers in primary brain tumours

We are getting used to referring to instrumentally detectable biological features in medical language as "imaging biomarkers". These two terms combined reflect the evolution of medical imaging during recent decades, and conceptually comprise the principle of noninvasive detection of internal processes that can become targets for supplementary therapeutic strategies. These targets in oncology include those biological pathways that are associated with several tumour features including independence from growth and growth-inhibitory signals, avoidance of apoptosis and immune system control, unlimited potential for replication, self-sufficiency in vascular supply and neoangiogenesis, acquired tissue invasiveness and metastatic diffusion. Concerning brain tumours, there have been major improvements in neurosurgical techniques and radiotherapy planning, and developments of novel target drugs, thus increasing the need for reproducible, noninvasive, quantitative imaging biomarkers. However, in this context, conventional radiological criteria may be inappropriate to determine the best therapeutic option and subsequently to assess response to therapy. Integration of molecular imaging for the evaluation of brain tumours has for this reason become necessary, and an important role in this setting is played by imaging biomarkers in PET and MRI. In the current review, we describe most relevant techniques and biomarkers used for imaging primary brain tumours in clinical practice, and discuss potential future developments from the experimental context.

Antiangiogenic therapy for glioblastoma: current status and future prospects

Glioblastoma is characterized by high expression levels of proangiogenic cytokines and microvascular proliferation, highlighting the potential value of treatments targeting angiogenesis. Antiangiogenic treatment likely achieves a beneficial impact through multiple mechanisms of action. Ultimately, however, alternative proangiogenic signal transduction pathways are activated, leading to the development of resistance, even in tumors that initially respond. The identification of biomarkers or imaging parameters to predict response and to herald resistance is of high priority. Despite promising phase II clinical trial results and patient benefit in terms of clinical improvement and longer progression-free survival, an overall survival benefit has not been demonstrated in four randomized phase III trials of bevacizumab or cilengitide in newly diagnosed glioblastoma or cediranib or enzastaurin in recurrent glioblastoma. However, future studies are warranted. Predictive markers may allow appropriate patient enrichment, combination with chemotherapy may ultimately prove successful in improving overall survival, and novel agents targeting multiple proangiogenic pathways may prove effective.

Use of CT perfusion to discriminate between brain metastases from different primaries

Thirty-six metastases in 22 patients were studied prospectively using computed tomography perfusion. Regions of interests were drawn around: the enhancing part of the tumor, necrotic central part, periphery, peritumoral edema, and normal white matter. Cerebral blood volume, cerebral blood flow, and mean transit time were calculated for each zone. The enhancing part of the tumor significantly differed from the other zones in 11 of 12. Metastases of different primaries can be differentiated from one another with statistically significance (P<.05) by at least one perfusion parameter in 57% of cases.

Vessel caliber--a potential MRI biomarker of tumour response in clinical trials

Our understanding of the importance of blood vessels and angiogenesis in cancer has increased considerably over the past decades, and the assessment of tumour vessel calibre and structure has become increasingly important for in vivo monitoring of therapeutic response. The preferred method for in vivo imaging of most solid cancers is MRI, and the concept of vessel-calibre MRI has evolved since its initial inception in the early 1990s. Almost a quarter of a century later, unlike traditional contrast-enhanced MRI techniques, vessel-calibre MRI remains widely inaccessible to the general clinical community. The narrow availability of the technique is, in part, attributable to limited awareness and a lack of imaging standardization. Thus, the role of vessel-calibre MRI in early phase clinical trials remains to be determined. By contrast, regulatory approvals of antiangiogenic agents that are not directly cytotoxic have created an urgent need for clinical trials incorporating advanced imaging analyses, going beyond traditional assessments of tumour volume. To this end, we review the field of vessel-calibre MRI and summarize the emerging evidence supporting the use of this technique to monitor response to anticancer therapy. We also discuss the potential use of this biomarker assessment in clinical imaging trials and highlight relevant avenues for future research.

Perfusion magnetic resonance imaging: a comprehensive update on principles and techniques

Perfusion is a fundamental biological function that refers to the delivery of oxygen and nutrients to tissue by means of blood flow. Perfusion MRI is sensitive to microvasculature and has been applied in a wide variety of clinical applications, including the classification of tumors, identification of stroke regions, and characterization of other diseases. Perfusion MRI techniques are classified with or without using an exogenous contrast agent. Bolus methods, with injections of a contrast agent, provide better sensitivity with higher spatial resolution, and are therefore more widely used in clinical applications. However, arterial spin-labeling methods provide a unique opportunity to measure cerebral blood flow without requiring an exogenous contrast agent and have better accuracy for quantification. Importantly, MRI-based perfusion measurements are minimally invasive overall, and do not use any radiation and radioisotopes. In this review, we describe the principles and techniques of perfusion MRI. This review summarizes comprehensive updated knowledge on the physical principles and techniques of perfusion MRI.

Glioma grading using apparent diffusion coefficient map: application of histogram analysis based on automatic segmentation

The accurate diagnosis of glioma subtypes is critical for appropriate treatment, but conventional histopathologic diagnosis often exhibits significant intra-observer variability and sampling error. The aim of this study was to investigate whether histogram analysis using an automatically segmented region of interest (ROI), excluding cystic or necrotic portions, could improve the differentiation between low-grade and high-grade gliomas. Thirty-two patients (nine low-grade and 23 high-grade gliomas) were included in this retrospective investigation. The outer boundaries of the entire tumors were manually drawn in each section of the contrast-enhanced T1 -weighted MR images. We excluded cystic or necrotic portions from the entire tumor volume. The histogram analyses were performed within the ROI on normalized apparent diffusion coefficient (ADC) maps. To evaluate the contribution of the proposed method to glioma grading, we compared the area under the receiver operating characteristic (ROC) curves. We found that an ROI excluding cystic or necrotic portions was more useful for glioma grading than was an entire tumor ROI. In the case of the fifth percentile values of the normalized ADC histogram, the area under the ROC curve for the tumor ROIs excluding cystic or necrotic portions was significantly higher than that for the entire tumor ROIs (p < 0.005). The automatic segmentation of a cystic or necrotic area probably improves the ability to differentiate between high- and low-grade gliomas on an ADC map.

In vivo chemical exchange saturation transfer imaging allows early detection of a therapeutic response in glioblastoma

Glioblastoma multiforme (GBM), which account for more than 50% of all gliomas, is among the deadliest of all human cancers. Given the dismal prognosis of GBM, it would be advantageous to identify early biomarkers of a response to therapy to avoid continuing ineffective treatments and to initiate other therapeutic strategies. The present in vivo longitudinal study in an orthotopic mouse model demonstrates quantitative assessment of early treatment response during short-term chemotherapy with temozolomide (TMZ) by amide proton transfer (APT) imaging. In a GBM line, only one course of TMZ (3 d exposure and 4 d rest) at a dose of 80 mg/kg resulted in substantial reduction in APT signal compared with untreated control animals, in which the APT signal continued to increase. Although there were no detectable differences in tumor volume, cell density, or apoptosis rate between groups, levels of Ki67 (index of cell proliferation) were substantially reduced in treated tumors. In another TMZ-resistant GBM line, the APT signal and levels of Ki67 increased despite the same course of TMZ treatment. As metabolite changes are known to occur early in the time course of chemotherapy and precede morphologic changes, these results suggest that the APT signal in glioma may be a useful functional biomarker of treatment response or degree of tumor progression. Thus, APT imaging may serve as a sensitive biomarker of early treatment response and could potentially replace invasive biopsies to provide a definitive diagnosis. This would have a major impact on the clinical management of patients with glioma.

PORTR: Pre-operative and post-recurrence brain tumor registration

We propose a new method for deformable registration of pre-operative and post-recurrence brain MR scans of glioma patients. Performing this type of intra-subject registration is challenging as tumor, resection, recurrence, and edema cause large deformations, missing correspondences, and inconsistent intensity profiles between the scans. To address this challenging task, our method, called PORTR, explicitly accounts for pathological information. It segments tumor, resection cavity, and recurrence based on models specific to each scan. PORTR then uses the resulting maps to exclude pathological regions from the image-based correspondence term while simultaneously measuring the overlap between the aligned tumor and resection cavity. Embedded into a symmetric registration framework, we determine the optimal solution by taking advantage of both discrete and continuous search methods. We apply our method to scans of 24 glioma patients. Both quantitative and qualitative analysis of the results clearly show that our method is superior to other state-of-the-art approaches.