1
|
Salari E, Chen X, Wynne JF, Qiu RLJ, Roper J, Shu HK, Yang X. Prediction of early recurrence of adult-type diffuse gliomas following radiotherapy using multi-modal magnetic resonance images. Med Phys 2024. [PMID: 39221589 DOI: 10.1002/mp.17382] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/22/2024] [Revised: 08/12/2024] [Accepted: 08/19/2024] [Indexed: 09/04/2024] Open
Abstract
BACKGROUND Adult-type diffuse gliomas are among the central nervous system's most aggressive malignant primary neoplasms. Despite advancements in systemic therapies and technological improvements in radiation oncology treatment delivery, the survival outcome for these patients remains poor. Fast and accurate assessment of tumor response to oncologic treatments is crucial, as it can enable the early detection of recurrent or refractory gliomas, thereby allowing timely intervention with life-prolonging salvage therapies. PURPOSE Radiomics is a developing field with great potential to improve medical image interpretation. This study aims to apply a radiomics-based predictive model for classifying response to radiotherapy within the first 3 months post-treatment. METHODS Ninety-five patients were selected from the Burdenko Glioblastoma Progression Dataset. Tumor regions were delineated in the axial plane on contrast-enhanced T1(CE T1W) and T2 fluid-attenuated inversion recovery (T2_FLAIR) magnetic resonance imaging (MRI). Hand-crafted radiomic (HCR) features, including first- and second-order features, were extracted using PyRadiomics (3.7.6) in Python (3.10). Then, recursive feature elimination with a random forest (RF) classifier was applied for feature dimensionality reduction. RF and support vector machine (SVM) classifiers were built to predict treatment outcomes using the selected features. Leave-one-out cross-validation was employed to tune hyperparameters and evaluate the models. RESULTS For each segmented target, 186 HCR features were extracted from the MRI sequence. Using the top-ranked radiomic features from a combination of CE T1W and T2_FLAIR, an optimized classifier achieved the highest averaged area under the curve (AUC) of 0.829 ± 0.075 using the RF classifier. The HCR features of CE T1W produced the worst outcomes among all models (0.603 ± 0.024 and 0.615 ± 0.075 for RF and SVM classifiers, respectively). CONCLUSIONS We developed and evaluated a radiomics-based predictive model for early tumor response to radiotherapy, demonstrating excellent performance supported by high AUC values. This model, harnessing radiomic features from multi-modal MRI, showed superior predictive performance compared to single-modal MRI approaches. These results underscore the potential of radiomics in clinical decision support for this disease process.
Collapse
Affiliation(s)
- Elahheh Salari
- Department of Radiation Oncology and Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia, USA
| | - Xuxin Chen
- Department of Radiation Oncology and Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia, USA
| | - Jacob Frank Wynne
- Department of Radiation Oncology and Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia, USA
| | - Richard L J Qiu
- Department of Radiation Oncology and Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia, USA
| | - Justin Roper
- Department of Radiation Oncology and Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia, USA
| | - Hui-Kuo Shu
- Department of Radiation Oncology and Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia, USA
| | - Xiaofeng Yang
- Department of Radiation Oncology and Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia, USA
| |
Collapse
|
2
|
Campion A, Iv M. Brain Tumor Imaging: Review of Conventional and Advanced Techniques. Semin Neurol 2023; 43:867-888. [PMID: 37963581 DOI: 10.1055/s-0043-1776765] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2023]
Abstract
Approaches to central nervous system (CNS) tumor classification and evaluation have undergone multiple iterations over the past few decades, in large part due to our growing understanding of the influence of genetics on tumor behavior and our refinement of brain tumor imaging techniques. Computed tomography and magnetic resonance imaging (MRI) both play a critical role in the diagnosis and monitoring of brain tumors, although MRI has become especially important due to its superior soft tissue resolution. The purpose of this article will be to briefly review the fundamentals of conventional and advanced techniques used in brain tumor imaging. We will also highlight the applications of these imaging tools in the context of commonly encountered tumors based on the most recently updated 2021 World Health Organization (WHO) classification of CNS tumors framework.
Collapse
Affiliation(s)
- Andrew Campion
- Department of Radiology (Neuroradiology), Stanford University, Stanford, California
| | - Michael Iv
- Department of Radiology (Neuroradiology), Stanford University, Stanford, California
| |
Collapse
|
3
|
Kamepalli H, Kalaparti V, Kesavadas C. Imaging Recommendations for the Diagnosis, Staging, and Management of Adult Brain Tumors. Indian J Med Paediatr Oncol 2023. [DOI: 10.1055/s-0042-1759712] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 03/08/2023] Open
Abstract
AbstractNeuroimaging plays a pivotal role in the clinical practice of brain tumors aiding in the diagnosis, genotype prediction, preoperative planning, and prognostication. The brain tumors most commonly seen in adults are extra-axial lesions like meningioma, intra-axial lesions like gliomas and lesions of the pituitary gland. Clinical features may be localizing like partial seizures, weakness, and sensory disturbances or nonspecific like a headache. On clinical suspicion of a brain tumor, the primary investigative workup should focus on imaging. Other investigations like fundoscopy and electroencephalography may be performed depending on the clinical presentation. Obtaining a tissue sample after identifying a brain tumor on imaging is crucial for confirming the diagnosis and planning further treatment. Tissue sample may be obtained by techniques such as stereotactic biopsy or upfront surgery. The magnetic resonance (MR) imaging protocol needs to be standardized and includes conventional sequences like T1-weighted (T1W) imaging with and without contrast, T2w imaging, fluid-attenuated axial inversion recovery, diffusion-weighted imaging (DWI), susceptibility-weighted imaging, and advanced imaging sequences like MR perfusion and MR spectroscopy. Various tumor characteristics in each of these sequences can help us narrow down the differential diagnosis and also predict the grade of the tumor. Multidisciplinary co-ordination is needed for proper management and care of brain tumor patients. Treatment protocols need to be adapted and individualized for each patient depending on the age, general condition of the patient, histopathological characteristics, and genotype of the tumor. Treatment options include surgery, radiotherapy, and chemotherapy. Imaging also plays a vital role in post-treatment follow-up. Sequences like DWI, MR perfusion, and MR spectroscopy are useful to distinguish post-treatment effects like radiation necrosis and pseudoprogression from true recurrence. Radiological reporting of brain tumor images should follow a structured format to include all the elements that could have an impact on the treatment decisions in patients.
Collapse
Affiliation(s)
- HariKishore Kamepalli
- Department of Imaging Sciences and Interventional Radiology, Sree Chitra Tirunal Institute of Medical Sciences and Technology, Trivandrum, Kerala, India
| | - Viswanadh Kalaparti
- Department of Imaging Sciences and Interventional Radiology, Sree Chitra Tirunal Institute of Medical Sciences and Technology, Trivandrum, Kerala, India
| | - Chandrasekharan Kesavadas
- Department of Imaging Sciences and Interventional Radiology, Sree Chitra Tirunal Institute of Medical Sciences and Technology, Trivandrum, Kerala, India
| |
Collapse
|
4
|
Nave O. A mathematical model for treatment using chemo-immunotherapy. Heliyon 2022; 8:e09288. [PMID: 35520602 PMCID: PMC9065634 DOI: 10.1016/j.heliyon.2022.e09288] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/20/2021] [Revised: 10/18/2021] [Accepted: 04/12/2022] [Indexed: 11/15/2022] Open
Abstract
In this study, we investigated a mathematical model for chemoimmunotherapy (a combination of chemotherapy and immunotherapy) for brain cancer. In most cases, the standard protocol for cancer treatment is fixed in terms of treatment time intervals and dosages. We offer a wide range of non-fixed protocols, which essentially vary in terms of time intervals and dosages (i.e., personalised medicine). The functions that describe this treatment are explicit and analytical. Hence, the parameters of the function can be easily changed and a new protocol can be obtained. We compared different protocols and obtained an optimal solution. In addition, we applied the singular perturbed vector field (SPVF) method to determine the hierarchy of the system of equations, which enabled us to identify the equilibrium points of the mathematical model and investigate their stability.
Collapse
Affiliation(s)
- Ophir Nave
- Department of Mathematics, Jerusalem College of Technology, Israel
| |
Collapse
|
5
|
Zhu J, Teolis S, Biassou N, Tabb A, Jabin PE, Lavi O. Tracking the Adaptation and Compensation Processes of Patients' Brain Arterial Network to an Evolving Glioblastoma. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2022; 44:488-501. [PMID: 32750811 DOI: 10.1109/tpami.2020.3008379] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/11/2023]
Abstract
The brain's vascular network dynamically affects its development and core functions. It rapidly responds to abnormal conditions by adjusting properties of the network, aiding stabilization and regulation of brain activities. Tracking prominent arterial changes has clear clinical and surgical advantages. However, the arterial network functions as a system; thus, local changes may imply global compensatory effects that could impact the dynamic progression of a disease. We developed automated personalized system-level analysis methods of the compensatory arterial changes and mean blood flow behavior from a patient's clinical images. By applying our approach to data from a patient with aggressive brain cancer compared with healthy individuals, we found unique spatiotemporal patterns of the arterial network that could assist in predicting the evolution of glioblastoma over time. Our personalized approach provides a valuable analysis tool that could augment current clinical assessments of the progression of glioblastoma and other neurological disorders affecting the brain.
Collapse
|
6
|
Metabolic Tumor Microenvironment Characterization of Contrast Enhancing Brain Tumors Using Physiologic MRI. Metabolites 2021; 11:metabo11100668. [PMID: 34677383 PMCID: PMC8537028 DOI: 10.3390/metabo11100668] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/10/2021] [Revised: 09/27/2021] [Accepted: 09/28/2021] [Indexed: 11/16/2022] Open
Abstract
The tumor microenvironment is a critical regulator of cancer development and progression as well as treatment response and resistance in brain neoplasms. The available techniques for investigation, however, are not well suited for noninvasive in vivo characterization in humans. A total of 120 patients (59 females; 61 males) with newly diagnosed contrast-enhancing brain tumors (64 glioblastoma, 20 brain metastases, 15 primary central nervous system (CNS) lymphomas (PCNSLs), and 21 meningiomas) were examined with a previously established physiological MRI protocol including quantitative blood-oxygen-level-dependent imaging and vascular architecture mapping. Six MRI biomarker maps for oxygen metabolism and neovascularization were fused for classification of five different tumor microenvironments: glycolysis, oxidative phosphorylation (OxPhos), hypoxia with/without neovascularization, and necrosis. Glioblastoma showed the highest metabolic heterogeneity followed by brain metastasis with a glycolysis-to-OxPhos ratio of approximately 2:1 in both tumor entities. In addition, glioblastoma revealed a significant higher percentage of hypoxia (24%) compared to all three other brain tumor entities: brain metastasis (7%; p < 0.001), PCNSL (8%; p = 0.001), and meningioma (8%; p = 0.003). A more aggressive biological brain tumor behavior was associated with a higher percentage of hypoxia and necrosis and a lower percentage of remaining vital tumor tissue and aerobic glycolysis. The proportion of oxidative phosphorylation, however, was rather similar (17–26%) for all four brain tumor entities. Tumor microenvironment (TME) mapping provides insights into neurobiological differences of contrast-enhancing brain tumors and deserves further clinical cancer research attention. Although there is a long roadmap ahead, TME mapping may become useful in order to develop new diagnostic and therapeutic approaches.
Collapse
|
7
|
Neuroimaging in the Era of the Evolving WHO Classification of Brain Tumors, From the AJR Special Series on Cancer Staging. AJR Am J Roentgenol 2021; 217:3-15. [PMID: 33502214 DOI: 10.2214/ajr.20.25246] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/18/2022]
Abstract
The inclusion of molecular and genetic information with histopathologic information defines the framework for brain tumor classification and grading. This framework is reflected in the major restructuring of the WHO brain tumor classification system in 2016 and in numerous subsequent proposed updates reflecting ongoing developments in understanding the impact of tumor genotype on classification and grading. This incorporation of molecular and genetic features improves tumor diagnosis and prediction of tumor behavior and response to treatment. Neuroimaging is essential for the noninvasive assessment of pretreatment tumor grading and for identification and determination of therapeutic efficacy. Use of conventional neuroimaging and physiologic imaging techniques, such as diffusion- and perfusion-weighted MRI, can increase diagnostic confidence before and after treatment. Although the use of neuroimaging to consistently determine tumor genetics is not yet robust, promising developments are on the horizon. Given the complexity of the brain tumor microenvironment, the development and implementation of a standardized reporting system can aid in conveying to radiologists, referring providers, and patients important information about brain tumor response to treatment. The purpose of this article is to review the current state and role of neuroimaging in this continuously evolving field.
Collapse
|
8
|
Haider AS, van den Bent M, Wen PY, Vogelbaum MA, Chang S, Canoll PD, Horbinski CM, Huse JT. Toward a standard pathological and molecular characterization of recurrent glioma in adults: a Response Assessment in Neuro-Oncology effort. Neuro Oncol 2021; 22:450-456. [PMID: 31844891 DOI: 10.1093/neuonc/noz233] [Citation(s) in RCA: 30] [Impact Index Per Article: 10.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/13/2022] Open
Abstract
Regardless of subtype, diffuse gliomas of adulthood are characterized by inexorable progression through treatment. Cancer recurrence in the context of therapy is by no means unique to gliomas. For many tumors residing outside the central nervous system (CNS), tissue-based analyses are routinely employed to document the molecular and cellular features of disease recurrence. Such interventions are inconsistently applied for gliomas, however, and lack rigorous standardization when they are. While many of the reasons underlying these discrepancies reflect pragmatic realities inherent to CNS disease, the suboptimal employment of histological and molecular assessment at recurrence nevertheless represents a missed opportunity to proactively guide patient management and increase knowledge. Herein, we address this quandary by pairing a succinct description of the histological, biological, and molecular characteristics of recurrent glioma with recommendations for how to better standardize and implement quality pathological assessment into patient management. We hope this review will prompt thoughtful revision of standard operating procedures to maximize the utility of glioma re-biopsy.
Collapse
Affiliation(s)
- Ali S Haider
- Departments of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Martin van den Bent
- Department of Neurology, Erasmus MC Cancer Institute, Rotterdam, the Netherlands
| | - Patrick Y Wen
- Center for Neuro-Oncology, Dana-Farber/Brigham and Women's Cancer Center and Harvard Medical School, Boston, Massachusetts, USA
| | - Michael A Vogelbaum
- Departments of Neurosurgery and Neuro-Oncology, Moffitt Cancer Center, Tampa, Florida, USA
| | - Susan Chang
- Department of Neurological Surgery, University of California San Francisco, San Francisco, California, USA
| | - Peter D Canoll
- Departments of Pathology and Cell Biology, Columbia University Medical Center, New York, New York, USA
| | - Craig M Horbinski
- Departments of Pathology, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, USA.,Neurological Surgery, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, USA
| | - Jason T Huse
- Pathology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA.,Translational Molecular Pathology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| |
Collapse
|