Predicting Survival After Hepatocellular Carcinoma Resection Using Deep Learning on Histological Slides.
Journal
Hepatology (Baltimore, Md.)
ISSN: 1527-3350
Titre abrégé: Hepatology
Pays: United States
ID NLM: 8302946
Informations de publication
Date de publication:
12 2020
12 2020
Historique:
received:
29
10
2019
revised:
23
12
2019
accepted:
09
02
2020
pubmed:
29
2
2020
medline:
5
5
2021
entrez:
29
2
2020
Statut:
ppublish
Résumé
Standardized and robust risk-stratification systems for patients with hepatocellular carcinoma (HCC) are required to improve therapeutic strategies and investigate the benefits of adjuvant systemic therapies after curative resection/ablation. In this study, we used two deep-learning algorithms based on whole-slide digitized histological slides (whole-slide imaging; WSI) to build models for predicting survival of patients with HCC treated by surgical resection. Two independent series were investigated: a discovery set (Henri Mondor Hospital, n = 194) used to develop our algorithms and an independent validation set (The Cancer Genome Atlas [TCGA], n = 328). WSIs were first divided into small squares ("tiles"), and features were extracted with a pretrained convolutional neural network (preprocessing step). The first deep-learning-based algorithm ("SCHMOWDER") uses an attention mechanism on tumoral areas annotated by a pathologist whereas the second ("CHOWDER") does not require human expertise. In the discovery set, c-indices for survival prediction of SCHMOWDER and CHOWDER reached 0.78 and 0.75, respectively. Both models outperformed a composite score incorporating all baseline variables associated with survival. Prognostic value of the models was further validated in the TCGA data set, and, as observed in the discovery series, both models had a higher discriminatory power than a score combining all baseline variables associated with survival. Pathological review showed that the tumoral areas most predictive of poor survival were characterized by vascular spaces, the macrotrabecular architectural pattern, and a lack of immune infiltration. This study shows that artificial intelligence can help refine the prediction of HCC prognosis. It highlights the importance of pathologist/machine interactions for the construction of deep-learning algorithms that benefit from expert knowledge and allow a biological understanding of their output.
Sections du résumé
BACKGROUND AND AIMS
Standardized and robust risk-stratification systems for patients with hepatocellular carcinoma (HCC) are required to improve therapeutic strategies and investigate the benefits of adjuvant systemic therapies after curative resection/ablation.
APPROACH AND RESULTS
In this study, we used two deep-learning algorithms based on whole-slide digitized histological slides (whole-slide imaging; WSI) to build models for predicting survival of patients with HCC treated by surgical resection. Two independent series were investigated: a discovery set (Henri Mondor Hospital, n = 194) used to develop our algorithms and an independent validation set (The Cancer Genome Atlas [TCGA], n = 328). WSIs were first divided into small squares ("tiles"), and features were extracted with a pretrained convolutional neural network (preprocessing step). The first deep-learning-based algorithm ("SCHMOWDER") uses an attention mechanism on tumoral areas annotated by a pathologist whereas the second ("CHOWDER") does not require human expertise. In the discovery set, c-indices for survival prediction of SCHMOWDER and CHOWDER reached 0.78 and 0.75, respectively. Both models outperformed a composite score incorporating all baseline variables associated with survival. Prognostic value of the models was further validated in the TCGA data set, and, as observed in the discovery series, both models had a higher discriminatory power than a score combining all baseline variables associated with survival. Pathological review showed that the tumoral areas most predictive of poor survival were characterized by vascular spaces, the macrotrabecular architectural pattern, and a lack of immune infiltration.
CONCLUSIONS
This study shows that artificial intelligence can help refine the prediction of HCC prognosis. It highlights the importance of pathologist/machine interactions for the construction of deep-learning algorithms that benefit from expert knowledge and allow a biological understanding of their output.
Types de publication
Journal Article
Validation Study
Langues
eng
Sous-ensembles de citation
IM
Pagination
2000-2013Commentaires et corrections
Type : CommentIn
Type : CommentIn
Informations de copyright
© 2020 by the American Association for the Study of Liver Diseases.
Références
Llovet JM, Zucman-Rossi J, Pikarsky E, Sangro B, Schwartz M, Sherman M, et al. Hepatocellular carcinoma. Nat Rev Dis Primers 2016;2:16018.
Goldblum JR, Lamps LW, McKenney JK, Myers JL, Ackerman LV, Rosai J. Rosai and Ackerman’s Surgical Pathology, 11th ed. Philadelphia, PA: Elsevier; 2018:2 volumes (xiv, 2142 pages).
Niazi MKK, Parwani AV, Gurcan MN. Digital pathology and artificial intelligence. Lancet Oncol 2019;20:e253-e261.
Ehteshami Bejnordi B, Veta M, Johannes van Diest P, van Ginneken B, Karssemeijer N, Litjens G, et al. Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast Cancer. JAMA 2017;318:2199-2210.
Coudray N, Ocampo PS, Sakellaropoulos T, Narula N, Snuderl M, Fenyo D, et al. Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning. Nat Med 2018;24:1559-1567.
Yu KH, Zhang C, Berry GJ, Altman RB, Re C, Rubin DL, et al. Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features. Nat Commun 2016;7:12474.
Mobadersany P, Yousefi S, Amgad M, Gutman DA, Barnholtz-Sloan JS, Velazquez Vega JE, et al. Predicting cancer outcomes from histology and genomics using convolutional networks. Proc Natl Acad Sci U S A 2018;115:E2970-E2979.
Kather JN, Krisam J, Charoentong P, Luedde T, Herpel E, Weis CA, et al. Predicting survival from colorectal cancer histology slides using deep learning: a retrospective multicenter study. PLoS Med 2019;16:e1002730.
Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012;2:401-404.
Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal 2013;6:pl1.
Cancer Genome Atlas Research Network. Comprehensive and integrative genomic characterization of hepatocellular carcinoma. Cell 2017;169:1327-1341.e23.
Davidson-Pilon C, Kalderstam J, Zivich P, Kuhn B, Fiore-Gartland A, et al. CamDavidsonPilon/lifelines: v0.22.9 (Version v0.22.9). Zenodo. 2019 Oct 30. http://doi.org/10.5281/zenodo.3523175.
Davidson-Pilon C, Kalderstam J, Zivich P, Kuhn B, Fiore-Gartland A, Abdeali JK. CamDavidsonPilon/lifelines: v0.22.9 (Version v0.22.9). Zenodo. 2019 Oct 30. http://doi.org/10.5281/zenodo.3523175.
He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. arXiv 2015;arXiv:1512.03385 [cs.CV].
Zucman-Rossi J, Villanueva A, Nault JC, Llovet JM. Genetic landscape and biomarkers of hepatocellular carcinoma. Gastroenterology 2015;149:1226-1239.e4.
Price WN. Big data and black-box medical algorithms. Sci Transl Med 2018;10:eaao5333.
Calderaro J, Couchy G, Imbeaud S, Amaddeo G, Letouze E, Blanc JF, et al. Histological subtypes of hepatocellular carcinoma are related to gene mutations and molecular tumour classification. J Hepatol 2017;67;727-738.
Ziol M, Pote N, Amaddeo G, Laurent A, Nault JC, Oberti F, et al. Macrotrabecular-massive hepatocellular carcinoma: a distinctive histological subtype with clinical relevance. Hepatology 2018;68:103-112.
Fang JH, Zhou HC, Zhang C, Shang LR, Zhang L, Xu J, et al. A novel vascular pattern promotes metastasis of hepatocellular carcinoma in an epithelial-mesenchymal transition-independent manner. Hepatology 2015;62:452-465.
Renne SL, Woo HY, Allegra S, Rudini N, Yano H, Donadon M, et al. Vessels encapsulating tumor clusters (VETC) is a powerful predictor of aggressive hepatocellular carcinoma. Hepatology 2020;71:183-195.
Calderaro J, Ziol M, Paradis V, Zucman-Rossi J. Molecular and histological correlations in liver cancer. J Hepatol 2019;71:616-630.
Villa E, Critelli R, Lei B, Marzocchi G, Camma C, Giannelli G, et al. Neoangiogenesis-related genes are hallmarks of fast-growing hepatocellular carcinomas and worst survival. Results from a prospective study. Gut 2016;65:861-869.
MacSween RNM, Burt AD, Portmann B, Ferrell LD. MacSween’s Pathology of the Liver, 6th ed. Edinburgh, UK: Churchill Livingstone; 2011:770-798.