Machine learning for rhabdomyosarcoma histopathology.
Journal
Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc
ISSN: 1530-0285
Titre abrégé: Mod Pathol
Pays: United States
ID NLM: 8806605
Informations de publication
Date de publication:
09 2022
09 2022
Historique:
received:
03
12
2021
accepted:
19
03
2022
revised:
16
03
2022
pubmed:
23
4
2022
medline:
1
9
2022
entrez:
22
4
2022
Statut:
ppublish
Résumé
Correctly diagnosing a rare childhood cancer such as sarcoma can be critical to assigning the correct treatment regimen. With a finite number of pathologists worldwide specializing in pediatric/young adult sarcoma histopathology, access to expert differential diagnosis early in case assessment is limited for many global regions. The lack of highly-trained sarcoma pathologists is especially pronounced in low to middle-income countries, where pathology expertise may be limited despite a similar rate of sarcoma incidence. To address this issue in part, we developed a deep learning convolutional neural network (CNN)-based differential diagnosis system to act as a pre-pathologist screening tool that quantifies diagnosis likelihood amongst trained soft-tissue sarcoma subtypes based on whole histopathology tissue slides. The CNN model is trained on a cohort of 424 centrally-reviewed histopathology tissue slides of alveolar rhabdomyosarcoma, embryonal rhabdomyosarcoma and clear-cell sarcoma tumors, all initially diagnosed at the originating institution and subsequently validated by central review. This CNN model was able to accurately classify the withheld testing cohort with resulting receiver operating characteristic (ROC) area under curve (AUC) values above 0.889 for all tested sarcoma subtypes. We subsequently used the CNN model to classify an externally-sourced cohort of human alveolar and embryonal rhabdomyosarcoma samples and a cohort of 318 histopathology tissue sections from genetically engineered mouse models of rhabdomyosarcoma. Finally, we investigated the overall robustness of the trained CNN model with respect to histopathological variations such as anaplasia, and classification outcomes on histopathology slides from untrained disease models. Overall positive results from our validation studies coupled with the limited worldwide availability of sarcoma pathology expertise suggests the potential of machine learning to assist local pathologists in quickly narrowing the differential diagnosis of sarcoma subtype in children, adolescents, and young adults.
Identifiants
pubmed: 35449398
doi: 10.1038/s41379-022-01075-x
pii: S0893-3952(22)00274-5
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
1193-1203Commentaires et corrections
Type : ErratumIn
Informations de copyright
© 2022. The Author(s), under exclusive licence to United States & Canadian Academy of Pathology.
Références
Tinkle, C. L., Fernandez-Pineda, I., Sykes, A., Lu, Z., Hua, C.-H., Neel, M. D. et al. Nonrhabdomyosarcoma soft tissue sarcoma (NRSTS) in pediatric and young adult patients: Results from a prospective study using limited-margin radiotherapy. Cancer 123, 4419–4429 (2017).
Sangkhathat, S. Current management of pediatric soft tissue sarcomas. World J Clin Pediatr 4, 94–105 (2015).
Skapek, S. X., Ferrari, A., Gupta, A. A., Lupo, P. J., Butler, E., Shipley, J. et al. Rhabdomyosarcoma. Nat Rev Dis Primers 5, 1–1 (2019).
Spunt, S. L., Skapek, S. X. & Coffin, C. M. Pediatric nonrhabdomyosarcoma soft tissue sarcomas. Oncologist 13, 668–678 (2008).
Steliarova-Foucher, E., Colombet, M., Ries, L. A. G., Moreno, F., Dolya, A., Bray, F. et al. International incidence of childhood cancer, 2001–10: a population-based registry study. Lancet Oncol 18, 719-731 (2017).
Force, L. M., Abdollahpour, I., Advani, S. M., Agius, D., Ahmadian, E., Alahdab, F. et al. The global burden of childhood and adolescent cancer in 2017: an analysis of the Global Burden of Disease Study 2017. Lancet Oncol 20, 1211-1225 (2019).
Qualman, S. J., Coffin, C. M., Newton, W. A., Hojo, H., Triche, T. J., Parham, D. M. et al. Intergroup Rhabdomyosarcoma Study: update for pathologists. Pediatr Dev Pathol 1, 550–561 (1998).
Fouda, A., Mansour, A. & Al-Tonbary, Y. The many faces of Ewing sarcoma: Difficult to diagnose pediatric cases. Hematol Oncol Stem Cell Ther 2, 411–417 (2009).
Chen, J. & Mullen, C. A. Patterns of diagnosis and misdiagnosis in pediatric cancer and relationship to survival. J Pediatr Hematol Oncol 39, e110–e115 (2017).
Yu, K.-H., Zhang, C., Berry, G. J., Altman, R. B., Ré, C., Rubin, D. L. et al. Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features. Nat Commun 7, 12474 (2016).
Yu, K. H., Berry, G. J., Rubin, D. L., Ré, C., Altman, R. B. & Snyder, M. Association of omics features with histopathology patterns in lung adenocarcinoma. Cell Syst 5, 620–627 (2017).
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature 542, 115–118 (2017).
Coudray, N., Ocampo, P. S., Sakellaropoulos, T., Narula, N., Snuderl, M., Fenyö, D. et al. Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning. Nat Med 24, 1559–1567 (2018).
Rudzinski, E. R., Anderson, J. R., Lyden, E. R., Bridge, J. A., Barr, F. G., Gastier-Foster, J. M. et al. Myogenin, AP2β, NOS-1, and HMGA2 are surrogate markers of fusion status in rhabdomyosarcoma: a report from the soft tissue sarcoma committee of the children’s oncology group. Am J Surg Pathol 38, 654–659 (2014).
Schöffski, P., Wozniak, A., Stacchiotti, S., Rutkowski, P., Blay, J. Y., Lindner, L. H. et al. Activity and safety of crizotinib in patients with advanced clear-cell sarcoma with MET alterations: European Organization for Research and Treatment of Cancer phase II trial 90101 CREATE. Ann Oncol 28, 3000–3008 (2017).
Abraham, J., Prajapati, S. I., Nishijo, K., Schaffer, B. S., Taniguchi, E., Kilcoyne, A. et al. Evasion mechanisms to Igf1r inhibition in rhabdomyosarcoma. Mol Cancer Ther 10, 697–707 (2011).
Nishijo, K., Chen, Q. R., Zhang, L., McCleish, A. T., Rodriguez, A., Cho, M. J. et al. Credentialing a preclinical mouse model of alveolar rhabdomyosarcoma. Cancer Res 69, 2902–2911 (2009).
Abraham, J., Nunez-Alvarez, Y., Hettmer, S., Carrio, E., Chen, H. I., Nishijo, K. et al. Lineage of origin in rhabdomyosarcoma informs pharmacological response. Genes Dev 28, 1578–1591 (2014).
Kikuchi, K., Taniguchi, E., Chen, H. H., Svalina, M. N., Abraham, J., Huang, E. T. et al. Rb1 loss modifies but does not initiate alveolar rhabdomyosarcoma. Skelet Muscle 3, 27 (2013).
Vahadane, A., Peng, T., Sethi, A., Albarqouni, S., Wang, L., Baust, M. et al. Structure-preserving color normalization and sparse stain separation for histological images. IEEE Trans Med Imaging 35, 1962–1971 (2016).
Hanley, J. A. & McNeil, B. J. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology 148, 839–843 (1983).