Deep learning can predict survival directly from histology in clear cell renal cell carcinoma

• Verfasst von: Wessels, Frederik [VerfasserIn]
• Verfasst von: Schmitt, Max [VerfasserIn]
• Verfasst von: Krieghoff-Henning, Eva [VerfasserIn]
• Verfasst von: Kather, Jakob Nikolas [VerfasserIn]
• Verfasst von: Nientiedt, Malin [VerfasserIn]
• Verfasst von: Kriegmair, Maximilian [VerfasserIn]
• Verfasst von: Worst, Thomas [VerfasserIn]
• Verfasst von: Neuberger, Manuel [VerfasserIn]
• Verfasst von: Steeg, Matthias [VerfasserIn]
• Verfasst von: Popovic, Zoran V. [VerfasserIn]
• Verfasst von: Gaiser, Timo [VerfasserIn]
• Verfasst von: Kalle, Christof von [VerfasserIn]
• Verfasst von: Utikal, Jochen [VerfasserIn]
• Verfasst von: Fröhling, Stefan [VerfasserIn]
• Verfasst von: Michel, Maurice Stephan [VerfasserIn]
• Verfasst von: Nuhn, Philipp [VerfasserIn]
• Verfasst von: Brinker, Titus Josef [VerfasserIn]
• Titel: Deep learning can predict survival directly from histology in clear cell renal cell carcinoma
• Verf.angabe: Frederik Wessels, Max Schmitt, Eva Krieghoff-Henning, Jakob N. Kather, Malin Nientiedt, Maximilian C. Kriegmair, Thomas S. Worst, Manuel Neuberger, Matthias Steeg, Zoran V. Popovic, Timo Gaiser, Christof von Kalle, Jochen S. Utikal, Stefan Fröhling, Maurice S. Michel, Philipp Nuhn, Titus J. Brinker
• E-Jahr: 2022
• Jahr: August 17, 2022
• Umfang: 14 S.
• Fussnoten: Gesehen am 28.10.2022
• Titel Quelle: Enthalten in: PLOS ONE
• Ort Quelle: San Francisco, California, US : PLOS, 2006
• Jahr Quelle: 2022
• Band/Heft Quelle: 17(2022), 8, Artikel-ID e0272656, Seite 1-14
• ISSN Quelle: 1932-6203
• DOI: doi:10.1371/journal.pone.0272656
• Datenträger: Online-Ressource
• Sprache: eng
• Sach-SW: Artificial intelligence
• Sach-SW: Cancer detection and diagnosis
• Sach-SW: Cancer risk factors
• Sach-SW: Histology
• Sach-SW: Malignant tumors
• Sach-SW: Metastasis
• Sach-SW: Renal cancer
• Sach-SW: Renal cell carcinoma
• K10plus-PPN: 1820320413

Abstract

For clear cell renal cell carcinoma (ccRCC) risk-dependent diagnostic and therapeutic algorithms are routinely implemented in clinical practice. Artificial intelligence-based image analysis has the potential to improve outcome prediction and thereby risk stratification. Thus, we investigated whether a convolutional neural network (CNN) can extract relevant image features from a representative hematoxylin and eosin-stained slide to predict 5-year overall survival (5y-OS) in ccRCC. The CNN was trained to predict 5y-OS in a binary manner using slides from TCGA and validated using an independent in-house cohort. Multivariable logistic regression was used to combine of the CNNs prediction and clinicopathological parameters. A mean balanced accuracy of 72.0% (standard deviation [SD] = 7.9%), sensitivity of 72.4% (SD = 10.6%), specificity of 71.7% (SD = 11.9%) and area under receiver operating characteristics curve (AUROC) of 0.75 (SD = 0.07) was achieved on the TCGA training set (n = 254 patients / WSIs) using 10-fold cross-validation. On the external validation cohort (n = 99 patients / WSIs), mean accuracy, sensitivity, specificity and AUROC were 65.5% (95%-confidence interval [CI]: 62.9-68.1%), 86.2% (95%-CI: 81.8-90.5%), 44.9% (95%-CI: 40.2-49.6%), and 0.70 (95%-CI: 0.69-0.71). A multivariable model including age, tumor stage and metastasis yielded an AUROC of 0.75 on the TCGA cohort. The inclusion of the CNN-based classification (Odds ratio = 4.86, 95%-CI: 2.70-8.75, p < 0.01) raised the AUROC to 0.81. On the validation cohort, both models showed an AUROC of 0.88. In univariable Cox regression, the CNN showed a hazard ratio of 3.69 (95%-CI: 2.60-5.23, p < 0.01) on TCGA and 2.13 (95%-CI: 0.92-4.94, p = 0.08) on external validation. The results demonstrate that the CNN’s image-based prediction of survival is promising and thus this widely applicable technique should be further investigated with the aim of improving existing risk stratification in ccRCC.