|
 
 |
| REVIEW |
|
| Year : 2018 | Volume
: 1
| Issue : 1 | Page : 9-15 |
|
Current concepts of imaging genomics in glioma
Ru-Jia Wang1, Gui-Quan Shen2, Mark S Shiroishi3, Bo Gao4
1 Department of Radiology, The Affiliated Hospital of Guizhou Medical University, Guiyang, Guizhou; Department of Radiology, Yantai Yuhuangding Hospital, Yantai, Shandong, China
2 Department of Radiology, The Affiliated Hospital of Guizhou Medical University, Guiyang, Guizhou, China
3 Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, United States
4 Department of Radiology, Yantai Yuhuangding Hospital, Yantai, Shandong, China
| Date of Web Publication | 28-Feb-2018 |
Correspondence Address:
Dr. Bo Gao
Department of Radiology, Yantai Yuhuangding Hospital, Yantai 264000, Shandong
China
 Source of Support: None, Conflict of Interest: None  | 2 |
DOI: 10.4103/glioma.glioma_1_18
Gliomas are the most common primary brain tumors, the grading of which is associated with biological behavior and prognosis. With rapid advances in our understanding of glioma biology and a move toward personalized medicine, it is important to obtain a greater understanding of these tumors, beyond that provided by conventional neuroimaging. “Imaging genomics” is an emerging field which has an immense potential to greatly expand our understanding of glioma behavior. This article reviews the relationship between the radiologic and genomic features of gliomas, which may aid in personalizing patient treatment.
Keywords: Glioblastoma, glioma, imaging genomics, magnetic resonance imaging
How to cite this article:
Wang RJ, Shen GQ, Shiroishi MS, Gao B. Current concepts of imaging genomics in glioma. Glioma 2018;1:9-15 |
| Introduction | |  |
According to the central brain tumor registry of the United States, gliomas are the most common malignant tumor in the central nervous system, accounting for 27% of total number of tumors in central nervous system (CNS). Forty-six percent of gliomas are glioblastomas (GBM), which is one of the most common malignancies.[1] In 2016, World Health Organization (WHO) reclassified the tumors of the central nervous system, which includes diffuse astrocytic tumors, oligodendroglial tumors, ependymal tumors, choroid plexus tumors, neuronal and mixed neuronal-glial tumors, tumors of pineal region, embryonal tumors, tumors of the cranial, paraspinal nerves, and other tumors.[2] Different tumor types and grades result in markedly different survival rates, and tumor heterogeneity is a key feature that determines the impact of early diagnosis for a given individual. WHO IV glioma, also known as GBM, is one of the most malignant and lethal tumors. The incidence is most common in the elderly population, and its prognosis is very poor with 5-year survival rates of <5%. GBM patients typically present with symptoms that include headache, vomiting, changes of cognition, personality disorder, gait ataxic, uracratia, hemiplegia, logagnosia, visual impairment, and epilepsy.[3] The standard of care therapy usually involves a combination of surgery and radiotherapy, followed by temozolomide (TMZ). Despite the treatment, the overall survival (OS) of GBM patients is very poor.[4] Tumor location, clinical symptoms, and the risk factors should be taken into consideration to determine the optimal clinical management of patients.[5] Recommendations for glioma treatment protocols have been put forward by the European Association for Neuro-Oncology,[6] which focus on the link between clinical, pathological, radiological, and surgical characteristics of gliomas. The location of solid tumors limits the invasive biopsy procedure. Given this, medical imaging could be a potential noninvasive strategy to characterize such intratumoral heterogeneity. Automatic computer technology involved in imaging segmentation, feature extraction, and analysis of imaging features in lesions such as glioma is called radiomics. Though radiomics is a potentially powerful technique that can greatly expand the role of conventional imaging methods from being purely diagnostic to those that can play a role in personalized medicine, much further validation is needed.[7],[8] One area where radiomics can be particularly informative is in the assessment of tumor genetics, an approach commonly referred to as imaging genetics.[9],[10]
| Genetics of Glioblastom | | |
Histopathologic descriptions have remained unchanged for decades; however, genetic alterations, such as those involving isocitratedehydrogenase-1 (IDH-1) mutation, O6-methylguanine-DNA-methyltransferase, methylation status, and epidermal growth factor receptor (EGFR) amplification, can provide more insight into tumor biology and different clinical outcomes.[11] The Cancer Genome Atlas More Details did a genome-wide study in which the mutations of MGMT had been identified in GBM tumor samples.[12] IDH-1/2 mutations have been described in low-grade glioma and secondary GBM, while only approximately 5% of primary GBMs have IDH mutations. Olar and Aldape [13] emphasized the association of molecular classification of GBM with personalized treatment. The process of IDH-1/2 mutations includes the origin of glioma cells with changes in the activity of some enzymes, which synthesize 2-hydroxyglutarate (a potential oncometabolite) instead of NADPH. It is not clear whether the oncogenic mechanism of IDH mutations involves alterations in hydroxylases, redox potential, cell metabolism, or gene expression. IDH mutations result in increased methylation of gliomas.[14] IDH mutation appears to be an early molecular event in the process of tumorigenesis.[15] It is vital for neuroradiologists to understand the molecular changes involved in GBM tumorigenesis and its relation to imaging characteristics, if we are to work toward improving patient outcomes.[16] A schematic diagram illustrating the impact of genetic changes on tumor biology is shown in [Figure 1]. | Figure 1: A chart reflects from genetic changes to pathologic features in glioblastomas. Genetic alterations are influenced by cellular microenvironment which leads to the alteration of cell and then cell alteration is displayed in the pathologic features
Click here to view |
| Classification of Glioma | | |
Primary brain tumors, including gliomas, are histopathologically typed and graded according to the WHO tumor classification. This divides tumors into Grades I–IV which indicate the increasing histologic characteristics of malignancy and inform clinical management decisions. More recent work suggests that molecular data can differentiate tumors of the same WHO grade and morphologic type and therefore hold promise toward the goal of personalized medicine.[17] With respect to this, the 2016 WHO tumor classification has now incorporated some molecular characteristics into their categorization of CNS tumors. Four molecular biomarkers have been incorporated into the revised WHO tumor classification and these are central to diagnosing and treating gliomas: IDH-1 mutation, MGMT promoter methylation, 1p/19q codeletion, and histone H3-K27M mutation. This effort is hoped to improve the early diagnosis of CNS tumors and individualize patient treatment to improve prognosis.[2] GBM is the most common malignant intracranial tumor.[18] In previous studies, primary and secondary GBMs are largely indistinguishable by imaging, but they differ in their genetic profiles,[19] for example, the mutation status of IDH-1.[20],[21] The classification of glioma according to the mutation status of IDH-1 is presented in a schematic diagram [Figure 2]. | Figure 2: Classification of gliomas according to the mutation status of isocitratedehydrogenase1
Click here to view |
| Radiological Imaging Characteristics of Gliomas | | |
Conventional magnetic resonance imaging (MRI) technique for brain tumor patients consists of T1-weighted, T2-weighted, T2-weighted fluid-attenuated inversion recovery (FLAIR), and contrast-enhanced T1-weighted sequences. Imaging features including nonhomogeneous enhancement, peritumoral edema, necrosis, solitary supratentorial infiltrative, and meningeal/ependymal infiltration lesions can be seen in GBM.[22],[23] Other sequences such as diffusion-weighted imaging (DWI) can assess tumor cell density and distinguish a cystic brain tumor from a brain abscess.[24] Diffusion tensor imaging (DTI) is a form of diffusion MRI which depicts the 3D shape of water diffusion in the brain. It may be useful to characterize disruption and damage of white matter fiber bundles caused by glioma and tumor angiogenesis,[25] similar to the damage caused in the corticospinal tract area resulting in motor dysfunction.[26] Shan and Wang [27] found that the fractional anisotropy value in the magnetic resonance DTI can be applied to distinguish between low-grade and high-grade glioma, with a specificity and sensitivity of 35.7% and 94.1%, respectively.[27] Magnetic resonance spectroscopy is another advanced MRI technique that provides metabolic information of the brain.[28],[29] Susceptibility-weighted imaging (SWI) is sensitive to venous vessels, blood components (e.g., metabolic products after hemorrhage), calcification, iron deposition, and so on. Relative cerebral blood volume from dynamic susceptibility contrast MRI could serve as a noninvasive tool for grading glioma [Figure 3].[30] SWI seems to be a valuable noninvasive tool in both evaluating the gliomas and differentiating between the low-grade glioma and high-grade glioma.[31] | Figure 3: A 45-year-old man with an astrocytoma (World Health Organization II Grade) of the right parietal lobe. (A-C) T1-weighted postcontrast magnetic resonance spectroscopy. It reflects the numerical range of N-acetyl aspartate (NAA) choline (Cho) and creatine (Cr). (D) Susceptibility-weighted imaging. Linear blooming artifact may indicate minor hemorrhage. (E and F) Diffusion tensor imaging shows that the right corticospinal tract shows mild displacement due to infiltrative edema
Click here to view |
| Imaging Genomics | | |
Imaging genomics is a novel field which aims to integrate radiological and molecular features, by examining the relationship between radiological and histological features. The radiological findings such as location of the tumor, the size of the tumor, hypointensity/hyperintensity on T1-/T2-weighted images, and any associated edema have demonstrated some obvious association with genomic features. Many studies believe that the imaging genomics may extend our views of radiology and molecular phenotype and serve as a tool for identifying subtypes of glioma, benefiting the patients.[32] Imaging genomics is an emerging technique that seeks to leverage large datasets to identify predictive and prognostic biomarkers in patients with diseases such as GBM [Figure 4]. Imaging genetics may allow evaluation of the active oncogenic pathway during GBM diagnosis and treatment.[9] Several molecular biomarkers (including IDH-1, MGMT, EGFR) may be associated with OS in GBM.[21],[33],[34] Grant et al.[35] noted that MGMT promoter methylation, 1p/19q codeletion, and IDH-1 mutations are all useful molecular biomarkers to gain a better understanding of glioma status. Besides, Colen et al.[32] found that MRI biomarkers could distinguish GBM phenotypes related to specific molecular pathways. Imaging genomic would thus provide a method for detecting particular genomic phenotype. | Figure 4: Imaging genomics: the relationship between radiology and gene phenotypes
Click here to view |
Isocitratedehydrogenase-1
IDH-1 mutation status has been associated with the diagnosis, treatment, and prognosis of glioma. In 2016, Yu et al.[36] described a noninvasive quantitative imaging genomics method to reveal IDH-1 status in Grade II glioma. The method includes image segmentation, high-throughput feature extraction, radiomics sequencing, feature selection, and classification. They collected and analyzed T2-weighted fluid attenuated inversion recovery (T2-FLAIR) of 110 patients with Grade II gliomas. The results showed that with an accuracy of 0.80, sensitivity of 0.83, and specificity of 0.74, the radiomics method employed could effectively estimate the IDH-1 mutation status noninvasively.
The relationship between magnetic resonance imaging performance and isocitratedehydrogenase-1
In previous studies, it was found that the high-grade gliomas with IDH-1 mutations pose longer survival, compared to those with the IDH-1 wild-type gene.[21],[33],[37],[38] Others have explored the use of machine-learning to predict IDH-1 genotype in high-grade gliomas based on clinical variables and multimodal features extracted from conventional MRI (including precontrast and postcontrast T1-weighted imaging, T2-weighted imaging, and DWI).[39] One hundred and twenty Grade III and IV primary glioma patients were included in this retrospective study. By means of utilizing a random forest algorithm to build models to predict IDH-1 genotype, nonredundant features were integrated with clinical material. Anatomical location served as an important factor which implicates a specific precursor cell in the period of tumor growth. The results were obtained with an accuracy of 86% (area under the curve (AUC) = 0.8830) in the training cohort and 89% (AUC = 0.9231) in the validation cohort, in their model. Similarly, Yu et al.[40] found the association between the anatomical location and IDH-1 mutation status in low-grade gliomas. Sonoda et al.[41] also found that IDH-1 mutation status is related to position and imaging feature in anaplastic gliomas. One hundred and twenty-two patients with anaplastic astrocytoma were evaluated in the study, and the results demonstrated that IDH-1 mutation was more likely to occur in tumors located in the frontal lobe. Other work focusing on low-grade gliomas has revealed that IDH-1 wild-type gliomas are associated with an ill-defined tumor boundary.[42] Moreover, a significant association of IDH-1 mutations with preoperative seizures was established by Yang et al.[39]
Analysis of isocitratedehydrogenase-1 mutation status by voxel-based lesion symptom mapping
Voxel-based lesion symptom mapping (VLSM) is an advanced neuroimaging technique which correlates the anatomical location of brain abnormalities and a particular genetic phenotype. VLSM analysis by Yuan et al.[43] discovered that the p53 mutation, MGMT promoter methylation, wild-type IDH-1, and EGFR overexpression usually appeared in the GBM located in the periventricular region of the left hemisphere and were associated with shorter survival. The team considers that the tumor location would be a significant factor in imaging features which can enhance the understanding of the GBM, improving its diagnostic and prognostic value.
Analysis of isocitratedehydrogenase-1 by deep learning-based radiomics
Li et al.[44] utilized deep learning-based radiomics which could extract characteristics from multimodality MRI to predict the mutation status of IDH-1 in low-grade glioma. By using this technique, the area under the operating characteristic curve (AUC) for IDH-1 estimation was improved to 92%, compared to 86% for the normal radiomics method. Nie et al.[45] assessed multimodal imaging to predict OS by 3D deep learning and concluded that MRI and DTI are important to predict OS in patients.
MGMT
The O6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation has been associated with a better prognosis and TMZ response in GBM patients.[46] If the status of the MGMT promoter methylation is known, it may be possible to use the information to individualize patients' treatment. Unfortunately, it may not always be possible to know this due to the necessity of an invasive biopsy or surgery.
The relationship between magnetic resonance imaging performance and MGMT
Mulholland et al.[47] found several valuable parameters including edema/necrosis volume ratio, tumor/necrosis volume ratio, edema volume, and tumor location, which could predict the status of MGMT promoter methylation in GBM. Rundle-Thiele et al.[48] found that MGMT methylation promoter might be an early molecular phenomenon in astrocytic and oligodendroglial tumors.
Wang et al.[49] used VLSM to correlate the anatomical location with MGMT protein expression in GBM. They showed that the low expression of MGMT protein often appeared in the right temporoparietal lobe, whereas, high expression of MGMT protein was often situated in the left frontal lobe.
Colen et al.[28] studied the relationship between the necrotic volume of GBM and molecular genotype of sex-specific apoptosis, by using MRI data. The volume of necrosis was lower in the female patients than in male patients, with female patients demonstrating longer survival duration than male. The result supports the notion that cell apoptosis based on specific molecular pathways differed by sex. Further, Nicolasjilwan et al.[50] showed that the clinical factors, imaging characteristics, and genomic data could better predict OS of GBM patients.
The analysis of MGMT by positron emission tomography
In 2017, Bosnyák et al.[33] found that the tumoral amino acid uptake and molecular markers were associated with OS of GBM patients. In their work, 21 patients underwent MRI examinations and positron emission tomography (PET) scanning with alpha [C-11]-L-methyl-tryptophan (AMT) before surgery. They found that several imaging factors including T1/T2 contrast volume, tumoral tryptophan uptake, PET-based metabolic tumor volume, and kinetic variables were associated with molecular biomarkers and OS. Unmethylated MGMT promoter was associated with larger metabolic volume and lower tumor/cortex AMT unidirectional uptake ratios than methylated MGMT promoter. Furthermore, patients with high tryptophan uptake on PET experienced improved survival.
| Analysis of Magnetic Resonance Imaging Texture | | |
Carrillo et al.[51] investigated the ability of imaging correlates to act as available biomarkers for molecularly defined GBM subtype. They confirmed that not only IDH-1 mutation but also MGMT promoter methylation are related to longer patient survival. Yang et al.[52] showed that the texture features from MRI can predict molecular subtypes and 12-month OS status of GBM patients. In their work, 82 GBM patients underwent manual segmentation of their tumors using MRI information including postcontrast T1-weighted imaging (T1WI) and T2-FLAIR. A series of texture features were extracted which consist of 48 segmentation-based fractal texture analysis features, 576 histogram of oriented gradients features, 44 run-length matrix features, 256 local binary patterns features, and 52 Haralick features from MRI, as shown in [Figure 5]. | Figure 5: A 67-year-old man who was diagnosed anaplastic oligodendroglioma. (A) T1-weighted postcontrast magnetic resonance sequence, region of interest was drew in the picture which includes the mass. (B) The magnetic resonance imaging texture analysis consists of about 300 texture parameters by MaZda (version 4.7, The Technical University of Lodz, Institute of Electronics, http://www.eletel.p.lodz.pl/mazda/)
Click here to view |
In 2014, Brynolfsson et al.[53] investigated the relationship between the texture of apparent diffusion coefficient (ADC) images and DWI pretreatment MRIs and found that ADC texture could potentially act as a biomarker to determine therapeutic effects and prognosis. Kinoshita et al.[54] examined textural features on T2-weighted images using parameters such as Shannon entropy and Prewitt filtering. Fifty glioma patients were studied, the results of which revealed an association of IDH-1 mutated type gliomas with larger Shannon entropy on T2-weighted images than IDH-1 wild-type gliomas. However, no relation was established between IDH-1 mutated type gliomas and IDH-1 wild-type gliomas in Edge median values using Prewitt filtering. Korfiatis et al.[55] also found that textural features could act as potential imaging markers to predict the status of MGMT methylation in GBM. The investigators recruited 155 GBM patients whose MGMT methylation status was available. Co-occurrence and run length texture features (correlation, energy, entropy, and local intensity) that originated from T2-weighted images were calculated to predict the status of MGMT methylation by making use of support vector machines and random forest classifiers.
In 2015, Itakura et al.[56] investigated MRI characteristics to identify the genetic phenotypes of GBM with distinct molecular pathways. One hundred and twenty-one patients were involved in this study, the MR images of whom were analyzed for the shape, texture, and edge sharpness of each lesion. The image features could identify the enriched c-kit and FOXA pathways, while also determining differential probabilities of survival, indicating diagnostic and prognostic importance.
From the above results, we can see that the imaging features-based texture extraction can predict the molecular subtypes and survival status in GBM, and MRI texture analysis could be used as a novel tool to improve the accuracy of glioma grading.
| Conclusion | | |
With the development of imaging genomics, we have improved our understanding of glioma biology that might not easily be obtained hisopathologically or through conventional neuroimaging. As a result, it is apparent that neuroradiologists should be better acquainted with glioma genetics, genomics, and epigenetics. In future, it is still a challenge for us to determine molecular phenotypes of glioma using imaging characteristics, which could individualize patient treatment and potentially alter outcomes. The challenges include tumor masking, feature extraction, feature selection, data analysis, and clinical application and not forget computer use. With the emerging era of imaging genomics, it is necessary for expertise from multiple disciplines to work together to devise individualized treatment plan that would improve the prognosis of patients.
Financial support and sponsorship
This study was financially supported by the research grant from Natural Science Foundation of Shandong Province (Grant No. ZR2014HL084) and National Natural Science Foundation of China (Grant No. 81471645).
Conflicts of interest
There are no conflicts of interest.
| References | | |
| 1. | Ostrom QT, Gittleman H, Fulop J, Liu M, Blanda R, Kromer C, et al. CBTRUS statistical report: Primary brain and central nervous system tumors diagnosed in the united states in 2008-2012. Neuro Oncol 2015;17 Suppl 4:iv1-62.  |
| 2. | Louis DN, Perry A, Reifenberger G, von Deimling A, Figarella-Branger D, Cavenee WK, et al. The 2016 world health organization classification of tumors of the central nervous system: A summary. Acta Neuropathol 2016;131:803-20. |
| 3. | Penfold C, Joannides AJ, Bell J, Walter FM. Diagnosing adult primary brain tumours: Can we do better? Br J Gen Pract 2017;67:278-9. |
| 4. | Ostrom QT, Gittleman H, Farah P, Ondracek A, Chen Y, Wolinsky Y, et al. CBTRUS statistical report: Primary brain and central nervous system tumors diagnosed in the united states in 2006-2010. Neuro Oncol 2013;15 Suppl 2:ii1-56. |
| 5. | Rajesh Y, Pal I, Banik P, Chakraborty S, Borkar SA, Dey G, et al. Insights into molecular therapy of glioma: Current challenges and next generation blueprint. Acta Pharmacol Sin 2017;38:591-613. |
| 6. | Weller M, van den Bent M, Tonn JC, Stupp R, Preusser M, Elizabeth C, et al. European association for neuro-oncology (EANO) guideline on the diagnosis and treatment of adult astrocytic and oligodendroglial gliomas. Lancet Oncol 2017;18:e315-29. |
| 7. | Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout RG, Granton P, et al. Radiomics: Extracting more information from medical images using advanced feature analysis. Eur J Cancer 2012;48:441-6. |
| 8. | Bourgier C, Colinge J, Aillères N, Fenogliettod P, Brenguesabcd M, Pèlegrind A, et al. Radiomics: Definition and clinical development. Cancer Radiothér 2015;19:532-7. (Article in French). |
| 9. | Pope WB. Genomics of brain tumor imaging. Neuroimaging Clin N Am 2015;25:105-19. |
| 10. | Zinn PO, Mahmood Z, Elbanan MG, Colen RR. Imaging genomics in gliomas. Cancer J 2015;21:225-34. |
| 11. | Adamson C, Kanu OO, Mehta AI, Di C, Lin N, Mattox AK, et al. Glioblastoma multiforme: A review of where we have been and where we are going. Expert Opin Investig Drugs 2009;18:1061-83. |
| 12. | Cancer Genome Atlas Research Network. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature 2008;455:1061-8. |
| 13. | Olar A, Aldape KD. Using the molecular classification of glioblastoma to inform personalized treatment. J Pathol 2014;232:165-77. |
| 14. | Cohen AL, Holmen SL, Colman H. IDH1 and IDH2 mutations in gliomas. Curr Neurol Neurosci Rep 2013;13:345. |
| 15. | Dimitrov L, Hong CS, Yang C, Zhuang Z, Heiss JD. New developments in the pathogenesis and therapeutic targeting of the IDH1 mutation in glioma. Int J Med Sci 2015;12:201-13. |
| 16. | Belden CJ, Valdes PA, Ran C, Pastel DA, Harris BT, Fadul CE, et al. Genetics of glioblastoma: A window into its imaging and histopathologic variability. Radiographics 2011;31:1717-40. |
| 17. | Smits M, van den Bent MJ. Imaging correlates of adult glioma genotypes. Radiology 2017;284:316-31. |
| 18. | Agnihotri S, Burrell KE, Wolf A, Jalali S, Hawkins C, Rutka JT, et al. Glioblastoma, a brief review of history, molecular genetics, animal models and novel therapeutic strategies. Arch Immunol Ther Exp (Warsz) 2013;61:25-41. |
| 19. | Waddington CH. The epigenotype 1942. Int J Epidemiol 2012;41:10-3. |
| 20. | Ohgaki H, Kleihues P. The definition of primary and secondary glioblastoma. Clin Cancer Res 2013;19:764-72. |
| 21. | Parsons DW, Jones S, Zhang X, Lin JC, Leary RJ, Angenendt P, et al. An integrated genomic analysis of human glioblastoma multiforme. Science 2008;321:1807-12. |
| 22. | Malikova H, Koubska E, Weichet J, Klener J, Rulseh A, Liscak R, et al. Can morphological MRI differentiate between primary central nervous system lymphoma and glioblastom? Cancer Imaging 2016;16:40. |
| 23. | Ko CC, Tai MH, Li CF, Chen TY, Chen JH, Shu G, et al. Differentiation between glioblastoma multiforme and primary cerebral lymphoma: Additional benefits of quantitative diffusion-weighted MR imaging. PLoS One 2016;11:e0162565. |
| 24. | Behin A, Hoang-Xuan K, Carpentier AF, Delattre JY. Primary brain tumours in adults. Lancet 2003;361:323-31. |
| 25. | Davanian F, Faeghi F, Shahzadi S, Farshifar Z. Diffusion tensor imaging for glioma grading: Analysis of fiber density index. Basic Clin Neurosci 2017;8:13-8. |
| 26. | Gao B, Shen X, Shiroishi MS, Pang M, Li Z, Yu B, et al. Apilot study of pre-operative motor dysfunction from gliomas in the region of corticospinal tract: Evaluation with diffusion tensor imaging. PLoS One 2017;12:e0182795. |
| 27. | Shan W, Wang XL. Clinical application value of 3.0T MR diffusion tensor imaging in grade diagnosis of gliomas. Oncol Lett 2017;14:2009-14. |
| 28. | Colen RR, Wang J, Singh SK, Gutman DA, Zinn PO. Glioblastoma: Imaging genomic mapping reveals sex-specific oncogenic associations of cell death. Radiology 2015;275:215-27. |
| 29. | He T, Qiu T, Wang X, Gui H, Wang X, Hu Q, et al. Multivoxel magnetic resonance spectroscopy identifies enriched foci of cancer stem-like cells in high-grade gliomas. Onco Targets Ther 2017;10:195-203. |
| 30. | Aydin O, Buyukkaya R, Hakyemez B. Susceptibility imaging in glial tumor grading; using 3 tesla magnetic resonance (MR) system and 32 channel head coil. Pol J Radiol 2017;82:179-87. |
| 31. | Li X, Zhu Y, Kang H, Zhang Y, Liang H, Wang S, et al. Glioma grading by microvascular permeability parameters derived from dynamic contrast-enhanced MRI and intratumoral susceptibility signal on susceptibility weighted imaging. Cancer Imaging 2015;15:4. |
| 32. | Colen RR, Vangel M, Wang J, Gutman DA, Hwang SN, Wintermark M, et al. Imaging genomic mapping of an invasive MRI phenotype predicts patient outcome and metabolic dysfunction: A TCGA glioma phenotype research group project. BMC Med Genomics 2014;7:30. |
| 33. | Bosnyák E, Michelhaugh SK, Klinger NV, Kamson DO, Barger GR, Mittal S, et al. Prognostic molecular and imaging biomarkers in primary glioblastoma. Clin Nucl Med 2017;42:341-7. |
| 34. | Weller M, Pfister SM, Wick W, Hegi ME, Reifenberger G, Stupp R, et al. Molecular neuro-oncology in clinical practice: A new horizon. Lancet Oncol 2013;14:e370-9. |
| 35. | Grant R, Kolb L, Moliterno J. Molecular and genetic pathways in gliomas: The future of personalized therapeutics. CNS Oncol 2014;3:123-36. |
| 36. | Yu J, Shi Z, Lian Y, Li Z, Liu T, Gao Y, et al. Noninvasive IDH1 mutation estimation based on a quantitative radiomics approach for grade II glioma. Eur Radiol 2017;27:3509-22. |
| 37. | Hartmann C, Hentschel B, Wick W, Capper D, Felsberg J, Simon M, et al. Patients with IDH1 wild type anaplastic astrocytomas exhibit worse prognosis than IDH1-mutated glioblastomas, and IDH1 mutation status accounts for the unfavorable prognostic effect of higher age: Implications for classification of gliomas. Acta Neuropathol 2010;120:707-18. |
| 38. | Kramář F, Minárik M, Benešová L, Halková T, Netuka D, Bradáč O, et al. IDH1/2 Mutation and MGMT Promoter Methylation-the relevant survival predictors in czech patients with brain gliomas. Folia Biol (Praha) 2016;62:194-202. |
| 39. | Yang Y, Mao Y, Wang X, Liu Y, Liang R, Lu J, et al. Mapping genetic factors in high-grade glioma patients. Clin Neurol Neurosurg 2016;150:159-63. |
| 40. | Yu J, Shi Z, Ji C, Lian Y, Wang Y, Chen L, et al. Anatomical location differences between mutated and wild-type isocitrate dehydrogenase 1 in low-grade gliomas. Int J Neurosci 2017;127:873-80. |
| 41. | Sonoda Y, Shibahara I, Kawaguchi T, Saito R, Kanamori M, Watanabe M, et al. Association between molecular alterations and tumor location and MRI characteristics in anaplastic gliomas. Brain Tumor Pathol 2015;32:99-104. |
| 42. | Mazurowski MA, Clark K, Czarnek NM, Shamsesfandabadi P, Peters KB, Saha A. Radiogenomics of lower-grade glioma: Algorithmically-assessed tumor shape is associated with tumor genomic subtypes and patient outcomes in a multi-institutional study with The Cancer Genome Atlas data. J Neurooncol 2017;133:27-35. |
| 43. | Yuan Y, Yunhe M, Xiang W, Yanhui L, Ruofei L, Jiewen L, et al. Mapping genetic factors in high-grade glioma patients. Clin Neurol Neurosurg 2016;150:159-63. |
| 44. | Li Z, Wang Y, Yu J, Guo Y, Cao W. Deep learning based radiomics (DLR) and its usage in noninvasive IDH1 prediction for low grade glioma. Sci Rep 2017;7:5467. |
| 45. | Nie D, Zhang H, Adeli E, Liu L, Shen D 3D deep learning for multi-modal imaging-guided survival time prediction of brain tumor patients. Med Image Comput Comput Assist Interv 2016;9901:212-20. |
| 46. | Kanas VG, Zacharaki EI, Thomas GA, Zinn PO, Megalooikonomou V, Colen RR, et al. Learning MRI-based classification models for MGMT methylation status prediction in glioblastoma. Comput Methods Programs Biomed 2017;140:249-57. |
| 47. | Mulholland S, Pearson DM, Hamoudi RA, Malley DS, Smith CM, Weaver JM, et al. MGMT CpG island is invariably methylated in adult astrocytic and oligodendroglial tumors with IDH1 or IDH2 mutations. Int J Cancer 2012;131:1104-13. |
| 48. | Rundle-Thiele D, Day B, Stringer B, Fay M, Martin J, Jeffree RL, et al. Using the apparent diffusion coefficient to identifying MGMT promoter methylation status early in glioblastoma: Importance of analytical method. J Med Radi Sci 2015;62:92-8. |
| 49. | Wang Y, Fan X, Zhang C, Zhang T, Peng X, Li S, et al. Anatomical specificity of O6-methylguanine DNA methyltransferase protein expression in glioblastomas. J Neurooncol 2014;120:331-7. |
| 50. | Nicolasjilwan M, Hu Y, Yan C, Meerzaman D, Holderc D, Gutmand D, et al. Addition of MR imaging features and genetic biomarkers strengthens glioblastoma survival prediction in TCGA patients. J Neuroradiol 2015;42:212-21. |
| 51. | Carrillo JA, Lai A, Nghiemphu PL, Kim HJ, Phillips HS, Kharbanda S, et al. Relationship between tumor enhancement, edema, IDH1 mutational status, MGMT promoter methylation, and survival in glioblastoma. AJNR Am J Neuroradiol 2012;33:1349-55. |
| 52. | Yang D, Rao G, Martinez J, Veeraraghavan A, Rao A. Evaluation of tumor-derived MRI-texture features for discrimination of molecular subtypes and prediction of 12-month survival status in glioblastoma. Med Phys 2015;42:6725-35. |
| 53. | Brynolfsson P, Nilsson D, Henriksson R, Hauksson J, Karlsson M, Garpebring A, et al. ADC texture – An imaging biomarker for high-grade glioma? Med Phys 2014;41:101903. |
| 54. | Kinoshita M, Sakai M, Arita H, Shofuda T, Chiba Y, Kagawa N, et al. Introduction of high throughput magnetic resonance T2-weighted image texture analysis for WHO grade 2 and 3 gliomas. PLoS One 2016;11:e0164268. |
| 55. | Korfiatis P, Kline TL, Coufalova L, Lachance DH, Parney IF, Carter RE, et al. MRI texture features as biomarkers to predict MGMT methylation status in glioblastomas. Med Phys 2016;43:2835-44. |
| 56. | Itakura H, Achrol AS, Mitchell LA, Loya JJ, Liu T, Westbroek EM, et al. Magnetic resonance image features identify glioblastoma phenotypic subtypes with distinct molecular pathway activities. Sci Transl Med 2015;7:303ra138. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
|
Search Pubmed for