1
|
Tam LT, Yeom KW, Wright JN, Jaju A, Radmanesh A, Han M, Toescu S, Maleki M, Chen E, Campion A, Lai HA, Eghbal AA, Oztekin O, Mankad K, Hargrave D, Jacques TS, Goetti R, Lober RM, Cheshier SH, Napel S, Said M, Aquilina K, Ho CY, Monje M, Vitanza NA, Mattonen SA. MRI-based radiomics for prognosis of pediatric diffuse intrinsic pontine glioma: an international study. Neurooncol Adv 2021; 3:vdab042. [PMID: 33977272 PMCID: PMC8095337 DOI: 10.1093/noajnl/vdab042] [Citation(s) in RCA: 6] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/19/2022] Open
Abstract
Background Diffuse intrinsic pontine gliomas (DIPGs) are lethal pediatric brain tumors. Presently, MRI is the mainstay of disease diagnosis and surveillance. We identify clinically significant computational features from MRI and create a prognostic machine learning model. Methods We isolated tumor volumes of T1-post-contrast (T1) and T2-weighted (T2) MRIs from 177 treatment-naïve DIPG patients from an international cohort for model training and testing. The Quantitative Image Feature Pipeline and PyRadiomics was used for feature extraction. Ten-fold cross-validation of least absolute shrinkage and selection operator Cox regression selected optimal features to predict overall survival in the training dataset and tested in the independent testing dataset. We analyzed model performance using clinical variables (age at diagnosis and sex) only, radiomics only, and radiomics plus clinical variables. Results All selected features were intensity and texture-based on the wavelet-filtered images (3 T1 gray-level co-occurrence matrix (GLCM) texture features, T2 GLCM texture feature, and T2 first-order mean). This multivariable Cox model demonstrated a concordance of 0.68 (95% CI: 0.61–0.74) in the training dataset, significantly outperforming the clinical-only model (C = 0.57 [95% CI: 0.49–0.64]). Adding clinical features to radiomics slightly improved performance (C = 0.70 [95% CI: 0.64–0.77]). The combined radiomics and clinical model was validated in the independent testing dataset (C = 0.59 [95% CI: 0.51–0.67], Noether’s test P = .02). Conclusions In this international study, we demonstrate the use of radiomic signatures to create a machine learning model for DIPG prognostication. Standardized, quantitative approaches that objectively measure DIPG changes, including computational MRI evaluation, could offer new approaches to assessing tumor phenotype and serve a future role for optimizing clinical trial eligibility and tumor surveillance.
Collapse
Affiliation(s)
- Lydia T Tam
- Stanford University School of Medicine, Stanford, California, USA.,Department of Radiology, Lucile Packard Children's Hospital, Stanford University School of Medicine, Stanford, California, USA
| | - Kristen W Yeom
- Stanford University School of Medicine, Stanford, California, USA.,Department of Radiology, Lucile Packard Children's Hospital, Stanford University School of Medicine, Stanford, California, USA
| | - Jason N Wright
- Department of Radiology, Seattle Children's Hospital, Seattle, Washington, USA.,Harborview Medical Center, Seattle, Washington, USA
| | - Alok Jaju
- Department of Medical Imaging, Ann and Robert H. Lurie Children's Hospital of Chicago, Chicago, Illinois, USA
| | - Alireza Radmanesh
- Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA
| | - Michelle Han
- Stanford University School of Medicine, Stanford, California, USA.,Department of Radiology, Lucile Packard Children's Hospital, Stanford University School of Medicine, Stanford, California, USA
| | - Sebastian Toescu
- University College London, Great Ormond Street Institute of Child Health, London, UK
| | - Maryam Maleki
- Department of Radiology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
| | - Eric Chen
- Departments of Clinical Radiology & Imaging Sciences, Riley Children's Hospital, Indiana University, Indianapolis, Indiana, USA
| | - Andrew Campion
- Department of Radiology, Lucile Packard Children's Hospital, Stanford University School of Medicine, Stanford, California, USA
| | - Hollie A Lai
- Department of Radiology, CHOC Children's Hospital, Orange, California, USA.,University of California, Irvine, California, USA
| | - Azam A Eghbal
- Department of Radiology, CHOC Children's Hospital, Orange, California, USA.,University of California, Irvine, California, USA
| | - Ozgur Oztekin
- Department of Neuroradiology, Bakircay University, Cigli Education and Research Hospital, Izmir, Turkey.,Department of Neuroradiology, Health Science University, Tepecik Education and Research Hospital, Izmir, Turkey
| | - Kshitij Mankad
- University College London, Great Ormond Street Institute of Child Health, London, UK.,Department of Radiology, Great Ormond Street Hospital for Children, London, UK
| | - Darren Hargrave
- University College London, Great Ormond Street Institute of Child Health, London, UK
| | - Thomas S Jacques
- University College London, Great Ormond Street Institute of Child Health, London, UK
| | - Robert Goetti
- Department of Medical Imaging, The Children's Hospital at Westmead, The University of Sydney, Westmead, Australia
| | - Robert M Lober
- Department of Neurosurgery, Dayton Children's Hospital, Wright State University Boonshoft School of Medicine, Dayton, Ohio, USA
| | - Samuel H Cheshier
- Department of Neurosurgery, University of Utah School of Medicine, Salt Lake City, Utah, USA
| | - Sandy Napel
- Department of Radiology, Stanford University, Stanford, California, USA
| | - Mourad Said
- Radiology Department Centre International Carthage Médicale, Monastir, Tunisia
| | - Kristian Aquilina
- University College London, Great Ormond Street Institute of Child Health, London, UK
| | - Chang Y Ho
- Departments of Clinical Radiology & Imaging Sciences, Riley Children's Hospital, Indiana University, Indianapolis, Indiana, USA
| | - Michelle Monje
- Stanford University School of Medicine, Stanford, California, USA.,Department of Neurology and Neurological Sciences, Stanford University, Stanford, California, USA
| | - Nicholas A Vitanza
- Division of Pediatric Hematology/Oncology, Department of Pediatrics, Seattle Children's Hospital, Seattle, Washington, USA.,Ben Towne Center for Childhood Cancer Research, Seattle Children's Research Institute, Seattle, Washington, USA
| | - Sarah A Mattonen
- Department of Medical Biophysics, Western University, London, Onatrio, Canada.,Department of Oncology, Western University, London, Ontario, Canada
| |
Collapse
|
3
|
Li L, London NR, Chen X. Malignant Mucosal Melanoma of the Eustachian Tube With Extension Into the Ipsilateral External Ear Canal: A Case Report and Review of the Literature. EAR, NOSE & THROAT JOURNAL 2020; 100:730S-733S. [PMID: 32070121 DOI: 10.1177/0145561320904813] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022] Open
Abstract
Mucosal melanoma arising in the middle ear or eustachian tube is uncommon. We present a patient with hearing loss and otalgia found to have mucosal melanoma which occurred in the eustachian tube with extension into the middle ear cavity and external ear canal. Otologic clinics was consulted and biopsy of the mass located at the external canal was performed to ascertain the pathological diagnosis. The patient refused immunotherapy and surgery instead of undergoing radiotherapy and died from hepatic metastasis 8 months later. The mucosal melanoma originated from the eustachian tube with extension into the external ear canal is exceedingly rare, and the differential diagnosis should be considered for tumors in external ear canal.
Collapse
Affiliation(s)
- Lifeng Li
- Department of Otolaryngology-Head and Neck Surgery, 117902Beijing Tongren Hospital, Capital Medical University, Beijing, China
| | - Nyall R London
- Department of Otolaryngology-Head & Neck Surgery, 1501Johns Hopkins School of Medicine, Baltimore, MA, USA.,National Institute on Deafness and Other Communication Disorders, NIH, Bethesda, MA, USA
| | - Xiaohong Chen
- Department of Otolaryngology-Head and Neck Surgery, 117902Beijing Tongren Hospital, Capital Medical University, Beijing, China
| |
Collapse
|
4
|
Maxwell AK, Takeda H, Gubbels SP. Primary Middle Ear Mucosal Melanoma: Case Report and Comprehensive Literature Review of 21 Cases of Primary Middle Ear and Eustachian Tube Melanoma. Ann Otol Rhinol Laryngol 2018; 127:856-863. [PMID: 30103615 DOI: 10.1177/0003489418793154] [Citation(s) in RCA: 12] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/03/2023]
Abstract
OBJECTIVE To present a case of primary middle ear mucosal melanoma and perform a comprehensive literature review of middle ear and eustachian tube mucosal melanoma. PATIENT A 61-year-old female presented with no prior history of melanoma and 3 months of aural fullness. A middle ear mass demonstrated primary mucosal melanoma. The mass extended from mesotympanum into hypotympanum, epitympanum, protympanum, eustachian tube, and mastoid antrum. Additionally, a nonenhancing expansile lesion of the petrous apex was noted on magnetic resonance imaging. INTERVENTION Subtotal temporal bone resection with transotic approach to the petrous abnormality was performed. Postoperative adjuvant radiation and immunotherapy were given. RESULTS Five months postoperatively, the patient developed an isolated cutaneous metastatic focus treated with wide local excision and has had no further evidence of recurrence to date. Upon comprehensive literature review, patients with primary middle ear melanomas (n = 10) present with otorrhea (50%), aural fullness (40%), and hearing loss (30%) most commonly, while hearing loss (81.8%) and aural fullness (54.5%) were the most common presenting symptoms for eustachian tube melanomas (n = 11). Patients were treated with combinations of surgery, radiation, and/or chemotherapy. Middle ear melanoma demonstrated particularly poor outcomes, with 70% mortality, 20% local recurrence, and 50% distant metastasis, whereas eustachian tube origin demonstrated 9.1%, 18.2%, and 36.4%, respectively. CONCLUSIONS Middle ear and eustachian tube mucosal melanomas are exceedingly rare, with middle ear melanomas demonstrating a worse prognosis. Multimodality therapy is commonly used; however, outcomes are poor, with high mortality among affected patients.
Collapse
Affiliation(s)
- Anne K Maxwell
- 1 Department of Otolaryngology, University of Colorado School of Medicine, Aurora, Colorado, USA
| | - Hiroki Takeda
- 1 Department of Otolaryngology, University of Colorado School of Medicine, Aurora, Colorado, USA
| | - Samuel P Gubbels
- 1 Department of Otolaryngology, University of Colorado School of Medicine, Aurora, Colorado, USA
| |
Collapse
|
6
|
Wei C, Sirikanjanapong S, Lieberman S, Delacure M, Martiniuk F, Levis W, Wang BY. Primary mucosal melanoma arising from the eustachian tube with CTLA-4, IL-17A, IL-17C, and IL-17E upregulation. EAR, NOSE & THROAT JOURNAL 2013; 92:36-40. [PMID: 23354891 DOI: 10.1177/014556131309200112] [Citation(s) in RCA: 9] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022] Open
Abstract
Primary malignant melanoma arising from the eustachian tube is extremely rare. We report the case of a 63-year-old white man who presented with a 1-month history of left-sided hearing loss and aural fullness. Flexible fiberoptic laryngoscopy detected a blue-purple mass that appeared to arise from the left lateral nasopharynx. Computed tomography demonstrated an enhancing mass arising from an orifice of the left eustachian tube. The tumor was debulked endoscopically and was confirmed to have originated in the left eustachian tube. Histologically, the tumor was made up of heavily pigmented pleomorphic spindle cells with frequent mitoses. The tumor cells were immunohistochemically positive for S-100 protein, HMB-45, Melan-A, and PNL-2. The final diagnosis was a mucosal malignant melanoma. We also performed a nested polymerase chain reaction assay for several genes of interest, including CTLA-4, IL-17A, IL-17B, IL-17C, IL-17D, IL-17E, IL-17F, PLZF, Foxp3, RORγt, CD27, and CD70. These genes have been studied mainly in cutaneous melanomas, especially for the development of immunotherapy, but only very limited studies have been done on mucosal melanomas. Our investigation found upregulation of CTLA-4, IL-17A, IL-17C, and IL-17E. Based on our finding of CTLA-4 upregulation, it may be suggested that our patient might have had low antitumor immunity and that he might have benefited from CTLA-4 blockade. On the other hand, upregulation of IL-17A and IL-17E might reflect increased antitumor immunity, which could suggest that patients with a mucosal melanoma might benefit from immunomodulators associated with the effect of Th17. These genes also have great potential to help melanoma patients obtain tailored treatment, and they can be used as biomarkers for predicting prognosis.
Collapse
Affiliation(s)
- Calvin Wei
- Department of Otolaryngology, New York University School of Medicine, New York, NY, USA
| | | | | | | | | | | | | |
Collapse
|
9
|
Lükens A, Dimartino E, Günther RW, Krombach GA. Functional MR imaging of the eustachian tube in patients with clinically proven dysfunction: correlation with lesions detected on MR images. Eur Radiol 2011; 22:533-8. [PMID: 21993981 DOI: 10.1007/s00330-011-2303-3] [Citation(s) in RCA: 17] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/27/2011] [Revised: 09/11/2011] [Accepted: 09/27/2011] [Indexed: 10/16/2022]
Abstract
OBJECTIVES To visualise the function of the Eustachian tube by MRI and assess the effect of surrounding lesions. METHODS Using 1.5 T MRI, 32 Eustachian tubes of 16 patients with clinically proven dysfunction (clinical symptoms, such as autophonia and fullness and non-opening at tympanometry) were investigated. For depiction of the anatomy and associated abnormalities, coronal and axial T2-weighted turbo spin echo sequences and a T1-weighted-2D-gradient echo sequences were acquired. Eustachian tube opening was analysed with real time turbo-gradient echo sequences during a Valsalva manoeuvre. RESULTS Dysfunction, that is, failure of opening of the Eustachian tube as shown by MRI, correlated in all patients with tympanometry. Lesions detected included nasopharyngeal carcinoma in 4 patients. Mucosal swelling in the paranasal sinus was present in 9 patients. One patient had postoperative defects. In two patients both pharyngotympanic tubes did not open although no structural abnormalities could be depicted. CONCLUSIONS The opening of the Eustachian tubes during the Valsalva manoeuvre is assessable on MRI. Failure of opening may be due to swelling of the mucosa and may be also associated with tumours. MRI assessment may be helpful in patients with chronic otitis media, where the information about tubal function is important in preoperative planning. KEY POINTS •Eustachian tube opening during Valsalva manoeuvre can be visualised with MRI •Lesions hampering tubal opening can be delineated at the same MRI examination •Functional MRI of the Eustachian tubes might determine the cause of dysfunction.
Collapse
Affiliation(s)
- Anna Lükens
- Department of Diagnostic Radiology, University Hospital, University of Technology (RWTH) Aachen, Aachen, Germany
| | | | | | | |
Collapse
|