1
|
Nakagaki R, Debsarkar SS, Kawanaka H, Aronow BJ, Prasath VBS. Deep learning-based IDH1 gene mutation prediction using histopathological imaging and clinical data. Comput Biol Med 2024; 179:108902. [PMID: 39038392 DOI: 10.1016/j.compbiomed.2024.108902] [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: 02/03/2024] [Revised: 06/27/2024] [Accepted: 07/14/2024] [Indexed: 07/24/2024]
Abstract
In the field of histopathology, many studies on the classification of whole slide images (WSIs) using artificial intelligence (AI) technology have been reported. We have studied the disease progression assessment of glioma. Adult-type diffuse gliomas, a type of brain tumor, are classified into astrocytoma, oligodendroglioma, and glioblastoma. Astrocytoma and oligodendroglioma are also called low grade glioma (LGG), and glioblastoma is also called glioblastoma multiforme (GBM). LGG patients frequently have isocitrate dehydrogenase (IDH) mutations. Patients with IDH mutations have been reported to have a better prognosis than patients without IDH mutations. Therefore, IDH mutations are an essential indicator for the classification of glioma. That is why we focused on the IDH1 mutation. In this paper, we aimed to classify the presence or absence of the IDH1 mutation using WSIs and clinical data of glioma patients. Ensemble learning between the WSIs model and the clinical data model is used to classify the presence or absence of IDH1 mutation. By using slide level labels, we combined patch-based imaging information from hematoxylin and eosin (H & E) stained WSIs, along with clinical data using deep image feature extraction and machine learning classifier for predicting IDH1 gene mutation prediction versus wild-type across cohort of 546 patients. We experimented with different deep learning (DL) models including attention-based multiple instance learning (ABMIL) models on imaging data along with gradient boosting machine (LightGBM) for the clinical variables. Further, we used hyperparameter optimization to find the best overall model in terms of classification accuracy. We obtained the highest area under the curve (AUC) of 0.823 for WSIs, 0.782 for clinical data, and 0.852 for ensemble results using MaxViT and LightGBM combination, respectively. Our experimental results indicate that the overall accuracy of the AI models can be improved by using both clinical data and images.
Collapse
Affiliation(s)
- Riku Nakagaki
- Graduate School of Engineering, Mie University, 1577 Kurima-machiya, Tsu, Mie 514-8507, Japan.
| | | | - Hiroharu Kawanaka
- Graduate School of Engineering, Mie University, 1577 Kurima-machiya, Tsu, Mie 514-8507, Japan.
| | - Bruce J Aronow
- Department of Computer Science, University of Cincinnati, OH 45221, USA; Division of Biomedical Informatics, Cincinnati Children's Hospital Medical Center, Cincinnati, OH 45229, USA; Department of Pediatrics, University of Cincinnati, OH 45267, USA; Department of Biomedical Informatics, College of Medicine, University of Cincinnati, OH 45267, USA.
| | - V B Surya Prasath
- Department of Computer Science, University of Cincinnati, OH 45221, USA; Division of Biomedical Informatics, Cincinnati Children's Hospital Medical Center, Cincinnati, OH 45229, USA; Department of Pediatrics, University of Cincinnati, OH 45267, USA; Department of Biomedical Informatics, College of Medicine, University of Cincinnati, OH 45267, USA.
| |
Collapse
|
2
|
Hajri R, Nicod-Lalonde M, Hottinger AF, Prior JO, Dunet V. Prediction of Glioma Grade and IDH Status Using 18F-FET PET/CT Dynamic and Multiparametric Texture Analysis. Diagnostics (Basel) 2023; 13:2604. [PMID: 37568967 PMCID: PMC10417545 DOI: 10.3390/diagnostics13152604] [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: 06/21/2023] [Revised: 08/01/2023] [Accepted: 08/03/2023] [Indexed: 08/13/2023] Open
Abstract
Mutations in isocitrate dehydrogenase (IDH) represent an independent predictor of better survival in patients with gliomas. We aimed to assess grade and IDH mutation status in patients with untreated gliomas, by evaluating the respective value of 18F-FET PET/CT via dynamic and texture analyses. A total of 73 patients (male: 48, median age: 47) who underwent an 18F-FET PET/CT for initial glioma evaluation were retrospectively included. IDH status was available in 61 patients (20 patients with WHO grade 2 gliomas, 41 with grade 3-4 gliomas). Time-activity curve type and 20 parameters obtained from static analysis using LIFEx© v6.30 software were recorded. Respective performance was assessed using receiver operating characteristic curve analysis and stepwise multivariate regression analysis adjusted for patients' age and sex. The time-activity curve type and texture parameters derived from the static parameters showed satisfactory-to-good performance in predicting glioma grade and IDH status. Both time-activity curve type (stepwise OR: 101.6 (95% CI: 5.76-1791), p = 0.002) and NGLDM coarseness (stepwise OR: 2.08 × 1043 (95% CI: 2.76 × 1012-1.57 × 1074), p = 0.006) were independent predictors of glioma grade. No independent predictor of IDH status was found. Dynamic and texture analyses of 18F-FET PET/CT have limited predictive value for IDH status when adjusted for confounding factors. However, they both help predict glioma grade.
Collapse
Affiliation(s)
- Rami Hajri
- Department of Diagnostic and Interventional Radiology, Lausanne University Hospital, Centre Hospitalier Universitaire Vaudois, 1011 Lausanne, Switzerland;
| | - Marie Nicod-Lalonde
- Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital, Centre Hospitalier Universitaire Vaudois, 1011 Lausanne, Switzerland; (M.N.-L.); (J.O.P.)
| | - Andreas F. Hottinger
- Department of Neurology, Lausanne University Hospital, Centre Hospitalier Universitaire Vaudois, 1011 Lausanne, Switzerland;
- Lukas Lundin & Family Brain Tumor Research Center, Lausanne University Hospital, Centre Hospitalier Universitaire Vaudois, 1011 Lausanne, Switzerland
| | - John O. Prior
- Department of Nuclear Medicine and Molecular Imaging, Lausanne University Hospital, Centre Hospitalier Universitaire Vaudois, 1011 Lausanne, Switzerland; (M.N.-L.); (J.O.P.)
| | - Vincent Dunet
- Department of Diagnostic and Interventional Radiology, Lausanne University Hospital, Centre Hospitalier Universitaire Vaudois, 1011 Lausanne, Switzerland;
| |
Collapse
|
3
|
Voon HPJ, Wong LH. Chromatin mutations in pediatric high grade gliomas. Front Oncol 2023; 12:1104129. [PMID: 36686810 PMCID: PMC9853562 DOI: 10.3389/fonc.2022.1104129] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/21/2022] [Accepted: 12/19/2022] [Indexed: 01/07/2023] Open
Abstract
Pediatric high grade gliomas (HGG) are lethal tumors which are currently untreatable. A number of recent studies have provided much needed insights into the mutations and mechanisms which drive oncogenesis in pediatric HGGs. It is now clear that mutations in chromatin proteins, particularly H3.3 and its associated chaperone complex (ATRX), are a hallmark feature of pediatric HGGs. We review the current literature on the normal roles of the ATRX/H3.3 complex and how these functions are disrupted by oncogenic mutations. We discuss the current clinical trials and pre-clinical models that target chromatin and DNA, and how these agents fit into the ATRX/H3.3 mutation model. As chromatin mutations are a relatively new discovery in pediatric HGGs, developing clear mechanistic insights are a key step to improving therapies for these tumors.
Collapse
|
4
|
Characteristics, Patterns of Care and Predictive Geriatric Factors in Elderly Patients Treated for High-Grade IDH-Mutant Gliomas: A French POLA Network Study. Cancers (Basel) 2022; 14:cancers14225509. [PMID: 36428602 PMCID: PMC9688655 DOI: 10.3390/cancers14225509] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/17/2022] [Revised: 10/27/2022] [Accepted: 11/03/2022] [Indexed: 11/12/2022] Open
Abstract
Background: Describe the characteristics, patterns of care, and predictive geriatric factors of elderly patients with IDHm high-grade glioma (HGG) included in the French POLA network. Material and Methods: The characteristics of elderly (≥70 years) patients IDHm HGG were compared to those of younger patients IDHm HGG (<70 years) and of elderly patients IDHwt HGG. Geriatric features were collected. Results: Out of 1433 HGG patients included, 119 (8.3%) were ≥70 years. Among them, 39 presented with IDHm HGG. The main characteristics of elderly IDHm HGG were different from those of elderly IDHwt HGG but similar to those of younger IDHm HGG. In contrast, their therapeutic management was different from those of younger IDHm HGG with less frequent gross total resection and radiotherapy. The median progression-free survival (PFS) and overall survival (OS) were longer for elderly patients IDHm HGG (29.3 months and 62.1 months) than elderly patients IDHwt HGG (8.3 months and 13.3 months) but shorter than those of younger patients IDHm HGG (69.1 months and not reached). Geriatric factors associated with PFS and OS were mobility, neuropsychological disorders, body mass index, and autonomy. Geriatric factors associated with PFS and OS were mobility, neuropsychological disorders, and body mass index, and autonomy. Conclusion: the outcome of IDHm HGG in elderly patients is better than that of IDHwt HGG. Geriatric assessment may be particularly important to optimally manage these patients.
Collapse
|
5
|
Mellinghoff IK, Chang SM, Jaeckle KA, van den Bent M. Isocitrate Dehydrogenase Mutant Grade II and III Glial Neoplasms. Hematol Oncol Clin North Am 2021; 36:95-111. [PMID: 34711457 DOI: 10.1016/j.hoc.2021.08.008] [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: 01/01/2023]
Abstract
Mutations in isocitrate dehydrogenase (IDH) 1 or IDH2 occur in most of the adult low-grade gliomas and, less commonly, in cholangiocarcinoma, chondrosarcoma, acute myeloid leukemia, and other human malignancies. Cancer-associated mutations alter the function of the enzyme, resulting in production of R(-)-2-hydroxyglutarate and broad epigenetic dysregulation. Small molecule IDH inhibitors have received regulatory approval for the treatment of IDH mutant (mIDH) leukemia and are under development for the treatment of mIDH solid tumors. This article provides a current view of mIDH adult astrocytic and oligodendroglial tumors, including their clinical presentation and treatment, and discusses novel approaches and challenges toward improving the treatment of these tumors.
Collapse
Affiliation(s)
- Ingo K Mellinghoff
- Memorial Sloan Kettering Cancer Center, 1275 York Avenue, New York, NY 10065, USA
| | - Susan M Chang
- Department of Neurological Surgery, University of California San Francisco, 505 Parnassus Room M 774SF, San Francisco, CA 94142-0112, USA
| | - Kurt A Jaeckle
- Department of Neurology and Oncology, Mayo Clinic Florida, Mangurian 4415, 4500 San Pablo Road, Jacksonville, FL 32224, USA
| | - Martin van den Bent
- Department of Neuro-onoclogy, Brain Tumor Center at Erasmus MC Cancer Institute, Nt-542, Dr Molenwaterplein 40, Rotterdam 3015 GD, The Netherlands.
| |
Collapse
|
6
|
Pryzbylski AL, Kollmeyer TM, Praska CE, Raghunathan A, Jentoft ME, Giannini C, Vaubel RA, Halling KC, Zheng G, DiGuardo MA, Kipp BR, Jenkins RB, Ida CM. Non-canonical IDH Mutation Frequency in IDH1-R132H-Negative Glioblastoma Patients Older Than 54 Years. J Neuropathol Exp Neurol 2021; 80:804-806. [PMID: 33550363 DOI: 10.1093/jnen/nlab004] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/13/2022] Open
Affiliation(s)
- Amber L Pryzbylski
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Thomas M Kollmeyer
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Corinne E Praska
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Aditya Raghunathan
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Mark E Jentoft
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Jacksonville, Florida, USA
| | - Caterina Giannini
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Rachael A Vaubel
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Kevin C Halling
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Gang Zheng
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Margaret A DiGuardo
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Benjamin R Kipp
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Robert B Jenkins
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| | - Cristiane M Ida
- Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
| |
Collapse
|