1
|
Zhu FY, Sun YF, Yin XP, Zhang Y, Xing LH, Ma ZP, Xue LY, Wang JN. Using machine learning-based radiomics to differentiate between glioma and solitary brain metastasis from lung cancer and its subtypes. Discov Oncol 2023; 14:224. [PMID: 38055122 DOI: 10.1007/s12672-023-00837-6] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/03/2023] [Accepted: 11/22/2023] [Indexed: 12/07/2023] Open
Abstract
OBJECTIVE To establish a machine learning-based radiomics model to differentiate between glioma and solitary brain metastasis from lung cancer and its subtypes, thereby achieving accurate preoperative classification. MATERIALS AND METHODS A retrospective analysis was conducted on MRI T1WI-enhanced images of 105 patients with glioma and 172 patients with solitary brain metastasis from lung cancer, which were confirmed pathologically. The patients were divided into the training group and validation group in an 8:2 ratio for image segmentation, extraction, and filtering; multiple layer perceptron (MLP), support vector machine (SVM), random forest (RF), and logistic regression (LR) were used for modeling; fivefold cross-validation was used to train the model; the validation group was used to evaluate and assess the predictive performance of the model, ROC curve was used to calculate the accuracy, sensitivity, and specificity of the model, and the area under curve (AUC) was used to assess the predictive performance of the model. RESULTS The accuracy and AUC of the MLP differentiation model for high-grade glioma and solitary brain metastasis in the validation group was 0.992, 1.000, respectively, while the sensitivity and specificity were 1.000, 0.968, respectively. The accuracy and AUC for the MLP and SVM differentiation model for high-grade glioma and small cell lung cancer brain metastasis in the validation group was 0.966, 1.000, respectively, while the sensitivity and specificity were 1.000, 0.929, respectively. The accuracy and AUC for the MLP differentiation model for high-grade glioma and non-small cell lung cancer brain metastasis in the validation group was 0.982, 0.999, respectively, while the sensitivity and specificity were 0.958, 1.000, respectively. CONCLUSION The application of machine learning-based radiomics has a certain clinical value in differentiating glioma from solitary brain metastasis from lung cancer and its subtypes. In the HGG/SBM and HGG/NSCLC SBM validation groups, the MLP model had the best diagnostic performance, while in the HGG/SCLC SBM validation group, the MLP and SVM models had the best diagnostic performance.
Collapse
Affiliation(s)
- Feng-Ying Zhu
- Department of Radiology, Affiliated Hospital of Hebei University, No.212 of Yuhua Road, Lianchi District, Baoding, 071000, China
| | - Yu-Feng Sun
- College of Electronic Information Engineering, Hebei University, Baoding, 071002, China
| | - Xiao-Ping Yin
- Department of Radiology, Affiliated Hospital of Hebei University, No.212 of Yuhua Road, Lianchi District, Baoding, 071000, China
| | - Yu Zhang
- Department of Radiology, Affiliated Hospital of Hebei University, No.212 of Yuhua Road, Lianchi District, Baoding, 071000, China
| | - Li-Hong Xing
- Department of Radiology, Affiliated Hospital of Hebei University, No.212 of Yuhua Road, Lianchi District, Baoding, 071000, China
| | - Ze-Peng Ma
- Department of Radiology, Affiliated Hospital of Hebei University, No.212 of Yuhua Road, Lianchi District, Baoding, 071000, China
| | - Lin-Yan Xue
- College of Quality and Technical Supervision, Hebei University, No.180 of Wusi Road, Lianchi District, Baoding, 071002, China.
| | - Jia-Ning Wang
- Department of Radiology, Affiliated Hospital of Hebei University, No.212 of Yuhua Road, Lianchi District, Baoding, 071000, China.
| |
Collapse
|
2
|
Yan Q, Li F, Cui Y, Wang Y, Wang X, Jia W, Liu X, Li Y, Chang H, Shi F, Xia Y, Zhou Q, Zeng Q. Discrimination Between Glioblastoma and Solitary Brain Metastasis Using Conventional MRI and Diffusion-Weighted Imaging Based on a Deep Learning Algorithm. J Digit Imaging 2023; 36:1480-1488. [PMID: 37156977 PMCID: PMC10406764 DOI: 10.1007/s10278-023-00838-5] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/17/2023] [Revised: 04/13/2023] [Accepted: 04/18/2023] [Indexed: 05/10/2023] Open
Abstract
This study aims to develop and validate a deep learning (DL) model to differentiate glioblastoma from single brain metastasis (BM) using conventional MRI combined with diffusion-weighted imaging (DWI). Preoperative conventional MRI and DWI of 202 patients with solitary brain tumor (104 glioblastoma and 98 BM) were retrospectively obtained between February 2016 and September 2022. The data were divided into training and validation sets in a 7:3 ratio. An additional 32 patients (19 glioblastoma and 13 BM) from a different hospital were considered testing set. Single-MRI-sequence DL models were developed using the 3D residual network-18 architecture in tumoral (T model) and tumoral + peritumoral regions (T&P model). Furthermore, the combination model based on conventional MRI and DWI was developed. The area under the receiver operating characteristic curve (AUC) was used to assess the classification performance. The attention area of the model was visualized as a heatmap by gradient-weighted class activation mapping technique. For the single-MRI-sequence DL model, the T2WI sequence achieved the highest AUC in the validation set with either T models (0.889) or T&P models (0.934). In the combination models of the T&P model, the model of DWI combined with T2WI and contrast-enhanced T1WI showed increased AUC of 0.949 and 0.930 compared with that of single-MRI sequences in the validation set, respectively. And the highest AUC (0.956) was achieved by combined contrast-enhanced T1WI, T2WI, and DWI. In the heatmap, the central region of the tumoral was hotter and received more attention than other areas and was more important for differentiating glioblastoma from BM. A conventional MRI-based DL model could differentiate glioblastoma from solitary BM, and the combination models improved classification performance.
Collapse
Affiliation(s)
- Qingqing Yan
- Department of Radiology, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Jinan, China
- Shandong First Medical University, Jinan, China
| | - Fuyan Li
- Department of Radiology, Shandong Provincial Hospital affiliated to Shandong First Medical University, Jinan, China
| | - Yi Cui
- Department of Radiology, Qilu Hospital of Shandong University, Jinan, China
| | - Yong Wang
- Shandong Cancer Hospital and Institute, Shandong First Medical University, Shandong Academy of Medical Sciences, Jinan, China
| | - Xiao Wang
- Department of Radiology, Jining NO.1 People's Hospital, Jining, China
| | - Wenjing Jia
- Department of Radiology, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Jinan, China
| | - Xinhui Liu
- Department of Radiology, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Jinan, China
| | - Yuting Li
- Department of Radiology, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Jinan, China
| | - Huan Chang
- Department of Radiology, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Jinan, China
| | - Feng Shi
- Shanghai United Imaging Intelligence, Shanghai, China
| | - Yuwei Xia
- Shanghai United Imaging Intelligence, Shanghai, China
| | - Qing Zhou
- Shanghai United Imaging Intelligence, Shanghai, China
| | - Qingshi Zeng
- Department of Radiology, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Jinan, China.
| |
Collapse
|
3
|
Paisana E, Cascão R, Custódia C, Qin N, Picard D, Pauck D, Carvalho T, Ruivo P, Barreto C, Doutel D, Cabeçadas J, Roque R, Pimentel J, Miguéns J, Remke M, Barata JT, Faria CC. UBE2C promotes leptomeningeal dissemination and is a therapeutic target in brain metastatic disease. Neurooncol Adv 2023; 5:vdad048. [PMID: 37215954 PMCID: PMC10195208 DOI: 10.1093/noajnl/vdad048] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 05/24/2023] Open
Abstract
Background Despite current improvements in systemic cancer treatment, brain metastases (BM) remain incurable, and there is an unmet clinical need for effective targeted therapies. Methods Here, we sought common molecular events in brain metastatic disease. RNA sequencing of thirty human BM identified the upregulation of UBE2C, a gene that ensures the correct transition from metaphase to anaphase, across different primary tumor origins. Results Tissue microarray analysis of an independent BM patient cohort revealed that high expression of UBE2C was associated with decreased survival. UBE2C-driven orthotopic mouse models developed extensive leptomeningeal dissemination, likely due to increased migration and invasion. Early cancer treatment with dactolisib (dual PI3K/mTOR inhibitor) prevented the development of UBE2C-induced leptomeningeal metastases. Conclusions Our findings reveal UBE2C as a key player in the development of metastatic brain disease and highlight PI3K/mTOR inhibition as a promising anticancer therapy to prevent late-stage metastatic brain cancer.
Collapse
Affiliation(s)
- Eunice Paisana
- Instituto de Medicina Molecular João Lobo Antunes, Faculdade de Medicina da Universidade de Lisboa; Av. Prof. Egas Moniz, 1649-028, Lisboa, Portugal
| | - Rita Cascão
- Instituto de Medicina Molecular João Lobo Antunes, Faculdade de Medicina da Universidade de Lisboa; Av. Prof. Egas Moniz, 1649-028, Lisboa, Portugal
| | - Carlos Custódia
- Instituto de Medicina Molecular João Lobo Antunes, Faculdade de Medicina da Universidade de Lisboa; Av. Prof. Egas Moniz, 1649-028, Lisboa, Portugal
| | - Nan Qin
- Department of Pediatric Oncology, Hematology and Clinical Immunology, Heinrich Heine University Düsseldorf, Medical Faculty, and University Hospital Düsseldorf; Moorenstraße 5, 40225 Düsseldorf, Germany
- German Cancer Consortium (DKTK), Partner Site Essen/Düsseldorf, Düsseldorf, Germany
- Moorenstraße 5, 40225 Düsseldorf, Germany
| | - Daniel Picard
- Department of Pediatric Oncology, Hematology and Clinical Immunology, Heinrich Heine University Düsseldorf, Medical Faculty, and University Hospital Düsseldorf; Moorenstraße 5, 40225 Düsseldorf, Germany
- German Cancer Consortium (DKTK), Partner Site Essen/Düsseldorf, Düsseldorf, Germany
- Moorenstraße 5, 40225 Düsseldorf, Germany
| | - David Pauck
- Department of Pediatric Oncology, Hematology and Clinical Immunology, Heinrich Heine University Düsseldorf, Medical Faculty, and University Hospital Düsseldorf; Moorenstraße 5, 40225 Düsseldorf, Germany
- German Cancer Consortium (DKTK), Partner Site Essen/Düsseldorf, Düsseldorf, Germany
- Moorenstraße 5, 40225 Düsseldorf, Germany
| | - Tânia Carvalho
- Instituto de Medicina Molecular João Lobo Antunes, Faculdade de Medicina da Universidade de Lisboa; Av. Prof. Egas Moniz, 1649-028, Lisboa, Portugal
| | - Pedro Ruivo
- Instituto de Medicina Molecular João Lobo Antunes, Faculdade de Medicina da Universidade de Lisboa; Av. Prof. Egas Moniz, 1649-028, Lisboa, Portugal
| | - Clara Barreto
- Instituto de Medicina Molecular João Lobo Antunes, Faculdade de Medicina da Universidade de Lisboa; Av. Prof. Egas Moniz, 1649-028, Lisboa, Portugal
| | - Delfim Doutel
- Anatomic Pathology Department, Instituto Português de Oncologia Francisco Gentil, R. Prof. Lima Basto, 1099-023, Lisboa, Portugal
| | - José Cabeçadas
- Anatomic Pathology Department, Instituto Português de Oncologia Francisco Gentil, R. Prof. Lima Basto, 1099-023, Lisboa, Portugal
| | - Rafael Roque
- Neurology Department, Laboratory of Neuropathology, Hospital de Santa Maria, Centro Hospitalar Universitário Lisboa Norte (CHULN), Av. Prof. Egas Moniz, 1649-028, Lisboa, Portugal
| | - José Pimentel
- Neurology Department, Laboratory of Neuropathology, Hospital de Santa Maria, Centro Hospitalar Universitário Lisboa Norte (CHULN), Av. Prof. Egas Moniz, 1649-028, Lisboa, Portugal
| | - José Miguéns
- Department of Neurosurgery, Hospital de Santa Maria, Centro Hospitalar Universitário Lisboa Norte (CHULN), Av. Prof. Egas Moniz, 1649-028, Lisboa, Portugal
| | - Marc Remke
- Department of Pediatric Oncology, Hematology and Clinical Immunology, Heinrich Heine University Düsseldorf, Medical Faculty, and University Hospital Düsseldorf; Moorenstraße 5, 40225 Düsseldorf, Germany
- German Cancer Consortium (DKTK), Partner Site Essen/Düsseldorf, Düsseldorf, Germany
- Moorenstraße 5, 40225 Düsseldorf, Germany
| | | | - Claudia C Faria
- Corresponding Author: Claudia C. Faria, Instituto de Medicina Molecular João Lobo Antunes, Edifício Egas Moniz, Faculdade de Medicina da Universidade de Lisboa, Av. Professor Egas Moniz, Lisboa, 1649-028, Portugal ()
| |
Collapse
|
4
|
Chen J, Williams M, Huang Y, Si S. Identifying Topics and Evolutionary Trends of Literature on Brain Metastases Using Latent Dirichlet Allocation. Front Mol Biosci 2022; 9:858577. [PMID: 35720132 PMCID: PMC9201447 DOI: 10.3389/fmolb.2022.858577] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/20/2022] [Accepted: 04/01/2022] [Indexed: 11/13/2022] Open
Abstract
Research on brain metastases kept innovating. We aimed to illustrate what topics the research focused on and how it varied in different periods of all the studies on brain metastases with topic modelling. We used the latent Dirichlet allocation model to analyse the titles and abstracts of 50,176 articles on brain metastases retrieved from Web of Science, Embase and MEDLINE. We further stratified the articles to find out the topic trends of different periods. Our study identified that a rising number of studies on brain metastases were published in recent decades at a higher rate than all cancer articles. Overall, the major themes focused on treatment and histopathology. Radiotherapy took over the first and third places in the top 20 topics. Since the 2010’s, increasing attention concerned about gene mutations. Targeted therapy was a popular topic of brain metastases research after 2020.
Collapse
Affiliation(s)
- Jiarong Chen
- Clinical Experimental Center, Jiangmen Key Laboratory of Clinical Biobanks and Translational Research, Jiangmen Central Hospital, Jiangmen, China
- Department of Oncology, Jiangmen Central Hospital, Jiangmen, China
- Computational Oncology Group, Department of Surgery and Cancer, Imperial College London, London, United Kingdom
- *Correspondence: Jiarong Chen, ; Shijing Si,
| | - Matt Williams
- Computational Oncology Group, Department of Surgery and Cancer, Imperial College London, London, United Kingdom
- Department of Radiotherapy, Charing Cross Hospital, Imperial College Healthcare NHS Trust, London, United Kingdom
| | - Yanming Huang
- Clinical Experimental Center, Jiangmen Key Laboratory of Clinical Biobanks and Translational Research, Jiangmen Central Hospital, Jiangmen, China
| | - Shijing Si
- Duke University, Durham, NC, United States
- *Correspondence: Jiarong Chen, ; Shijing Si,
| |
Collapse
|
5
|
|
6
|
McKay MJ. Brain metastases: increasingly precision medicine-a narrative review. ANNALS OF TRANSLATIONAL MEDICINE 2021; 9:1629. [PMID: 34926673 PMCID: PMC8640905 DOI: 10.21037/atm-21-3665] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Subscribe] [Scholar Register] [Received: 07/14/2021] [Accepted: 10/12/2021] [Indexed: 12/13/2022]
Abstract
Objective To broadly review the modern management of brain metastases. Background Brain metastases are the commonest neurological manifestation of cancer and a major cause of morbidity in cancer patients. Brain metastases are increasing in frequency, as a result of longer life expectancy of cancer patients, more sensitive methods for brain metastasis detection and an ageing population. The proportional incidence of brain metastases according to cancer of origin, from greatest to least, is lung cancer, melanoma, renal, breast and colorectal cancers. Patients with lung cancer and melanoma are most likely to have brain metastases at diagnosis. Brain metastases cause a variety of symptoms, depending on their size and location, whether they cause mass effect and oedema, compression of the brain parenchyma, or focal neurological deficits. The major differential diagnoses of brain metastases include primary tumours and vascular/inflammatory lesions. Prognosis is dependent on the site, number and volume of lesions, the patients’ performance status, age and the activity and extent of extracranial disease. Methods English literature articles in PubMed from 1950 to June 2021 were reviewed. Article bibliographies provided further references. Conclusions Treatment of brain metastasis patients has moved from considering them as a homogenous population of patients, to individualised treatment. In those brain metastases patients of satisfactory performance status with a solitary lesion, especially one in a non-eloquent/accessible area causing significant mass effect and/or raised intracranial pressure or for whom the diagnosis is in doubt (histology needed), surgical resection is usually the treatment of choice. For multiple brain metastases, radiotherapy with or without systemic therapies are usually employed. For relatively fit patients with limited numbers of brain metastases (e.g., 4 or less), stereotactic radiosurgery is standard of care. Current clinical trials are testing the efficacy of stereotactic treatment alone for >4 brain metastases (although it is increasingly used for such patients in many centres) as well as integration of local therapies with targeted and immunological therapies in appropriately selected cases. In certain circumstances, cranial irradiation can be omitted.
Collapse
Affiliation(s)
- Michael Jerome McKay
- Northern Cancer Service, North West Cancer Centre, Burnie, Tasmania, Australia.,The University of Tasmania, Rural Clinical School, Northwest Regional Hospital, Burnie, Tasmania, Australia
| |
Collapse
|