Rhabdomyosarcoma (RMS), the most common malignant soft tissue tumor in children, has
several histologic subtypes that influence treatment and predict patient outcomes.
Assistance with histologic classification for pathologists as well as discovery of
optimized predictive biomarkers is needed. A convolutional neural network for RMS
histology subtype classification was developed using digitized pathology images from
80 patients collected at time of diagnosis. A subsequent embryonal rhabdomyosarcoma
(eRMS) prognostic model was also developed in a cohort of 60 eRMS patients. The RMS
classification model reached a performance of an area under the receiver operating
curve of 0.94 for alveolar rhabdomyosarcoma and an area under the receiver operating
curve of 0.92 for eRMS at slide level in the test data set (n = 192). The eRMS prognosis model separated the patients into predicted high- and
low-risk groups with significantly different event-free survival outcome (likelihood
ratio test; P = 0.02) in the test data set (n = 136). The predicted risk group is significantly associated with patient event-free
survival outcome after adjusting for patient age and sex (predicted high- versus low-risk
group hazard ratio, 4.64; 95% CI, 1.05–20.57; P = 0.04). This is the first comprehensive study to develop computational algorithms
for subtype classification and prognosis prediction for RMS histopathology images.
Such models can aid pathology evaluation and provide additional parameters for risk
stratification.
To read this article in full you will need to make a payment
Purchase one-time access:
Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online accessOne-time access price info
- For academic or personal research use, select 'Academic and Personal'
- For corporate R&D use, select 'Corporate R&D Professionals'
Subscribe:
Subscribe to The American Journal of PathologyAlready a print subscriber? Claim online access
Already an online subscriber? Sign in
Register: Create an account
Institutional Access: Sign in to ScienceDirect
References
- Cancer Incidence and Survival Among Children and Adolescents: United States SEER Program, 1975-1995.National Cancer Institute, Bethesda, MD1999
- Classification of rhabdomyosarcomas and related sarcomas: pathologic aspects and proposal for a new classification--an Intergroup Rhabdomyosarcoma Study.Cancer. 1995; 76: 1073-1085
- WHO Classification of Tumours of Soft Tissue and Bone.International Agency for Research on Cancer, Lyon, France2013
- Pathology of childhood rhabdomyosarcoma: a consensus opinion document from the Children's Oncology Group, European Paediatric Soft Tissue Sarcoma Study Group, and the Cooperative Weichteilsarkom Studiengruppe.Pediatr Blood Cancer. 2021; 68: e28798
- Rhabdomyosarcoma: a new classification scheme related to prognosis.Arch Pathol Lab Med. 1992; 116: 847-855
- Orbital rhabdomyosarcomas and related tumors in childhood: relationship of morphology to prognosis—an Intergroup Rhabdomyosarcoma Study.Med Pediatr Oncol. 1997; 29: 51-60
- Intergroup rhabdomyosarcoma study-IV: results for patients with nonmetastatic disease.J Clin Oncol. 2001; 19: 3091-3102
- PAX3-FKHR and PAX7-FKHR gene fusions are prognostic indicators in alveolar rhabdomyosarcoma: a report from the Children's Oncology Group.J Clin Oncol. 2002; 20: 2672-2679
- Fusion gene-negative alveolar rhabdomyosarcoma is clinically and molecularly indistinguishable from embryonal rhabdomyosarcoma.J Clin Oncol. 2010; 28: 2151-2158
- PAX-FOXO1 fusion status drives unfavorable outcome for children with rhabdomyosarcoma: a Children's Oncology Group report.Pediatr Blood Cancer. 2013; 60: 1411-1417
- The prognostic significance of anaplasia in childhood rhabdomyosarcoma: a report from the Children's Oncology Group.Eur J Cancer. 2021; 143: 127-133
- Pathology image analysis using segmentation deep learning algorithms.Am J Pathol. 2019; 189: 1686-1698
- Deep learning.Nature. 2015; 521: 436-444
- Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning.Nat Med. 2018; 24: 1559-1567
- Classification of breast cancer histology images using convolutional neural networks.PLoS One. 2017; 12: e0177544
- Imagenet classification with deep convolutional neural networks.Adv Neural Inf Process Syst. 2012; 25: 1097-1105
- Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer.JAMA. 2017; 318: 2199-2210
- Impact of deep learning assistance on the histopathologic review of lymph nodes for metastatic breast cancer.Am J Surg Pathol. 2018; 42: 1636-1646
- Detecting cancer metastases on gigapixel pathology images.arXiv. 2017; ([Preprint])https://doi.org/10.48550/arXiv.1703.02442
- Automatic breast cancer grading of histopathological images.IEEE, 2008: 3052-3055 (2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society)
- Assessment of algorithms for mitosis detection in breast cancer histopathology images.Med Image Anal. 2015; 20: 237-248
- Understanding of a convolutional neural network.IEEE, 2017: 1-6 (2017 International Conference on Engineering and Technology (ICET), Aug. 21-23, 2017)
- A survey on deep learning in medical image analysis.Med Image Anal. 2017; 42: 60-88
- A comprehensive survey on transfer learning.Proc IEEE. 2020; 109: 43-76
- An investigation of aggregated transfer learning for classification in digital pathology.(Proc. SPIE 10956, Medical Imaging 2019: Digital Pathology)2019 (109560U)
- Large scale tissue histopathology image classification, segmentation, and visualization via deep convolutional activation features.BMC Bioinformatics. 2017; 18: 281
- Convolutional neural networks for histopathology image classification: training vs. using pre-trained networks.IEEE, 2017: 1-6 (2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA))
- Rhabdomyosarcoma histology classification using ensemble of deep learning networks.(Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics, Association for Computing Machinery: Virtual Event, USA) ACM, 2020 (Article 24)
- Rethinking the inception architecture for computer vision.IEEE, 2016: 2818-2826 (Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition)
- Transfer learning on fused multiparametric MR images for classifying histopathological subtypes of rhabdomyosarcoma.Comput Med Imaging Graph. 2018; 65: 167-175
- Deep feature learning for soft tissue sarcoma classification in MR images via transfer learning.Expert Syst Appl. 2019; 120: 116-127
- Solo or ensemble? Choosing a CNN architecture for melanoma classification, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops.IEEE, 2019
Article Info
Publication History
Published online: April 03, 2022
Accepted:
March 3,
2022
Footnotes
Supported by the Children's Cancer Fund of Dallas (P.L.), the QuadW Foundation (P.L.), the NIH grants NCI National Clinical Trials Network (NCTN) Operations Center grant U10CA180886 (P.L.), NCTN SDC grant U10CA180899 (P.L.), Children's Oncology Group (COG) Biospecimen Bank grant U24CA196173 (P.L.), U01CA249245 (G.X.), R01GM140012 (G.X.), R01GM141519 (G.X.), R01DE030656 (G.X.), P30CA142543 (Y.X.), and R35GM136375 (Y.X.), and the Cancer Prevention and Research Institute of Texas grants RP190107 (G.X.) and RP180805 (Y.X.).
Disclosures: None declared.
Identification
Copyright
© 2022 American Society for Investigative Pathology. Published by Elsevier Inc. All rights reserved.
