Sergey Klimov1,2, Yue Xue3, Arkadiusz Gertych4,5,6, Rondell P Graham7, Yi Jiang2, Shristi Bhattarai1, Stephen J Pandol8, Emad A Rakha9, Michelle D Reid10, Ritu Aneja1. 1. Department of Biology, Georgia State University, Atlanta, GA, United States. 2. Department of Mathematics and Statistics, Georgia State University, Atlanta, GA, United States. 3. Department of Pathology, Northwestern University, Chicago, IL, United States. 4. Department of Surgery, Cedars-Sinai Medical Center, Los Angeles, CA, United States. 5. Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, Los Angeles, CA, United States. 6. Faculty of Biomedical Engineering, Silesian University of Technology, Zabrze, Poland. 7. Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, United States. 8. Department of Medicine, Cedars-Sinai Medical Center, Los Angeles, CA, United States. 9. Department of Cellular Pathology, University of Nottingham, Nottingham, United Kingdom. 10. Department of Pathology, Emory University, Atlanta, GA, United States.
Abstract
BACKGROUND: The prognosis of patients with pancreatic neuroendocrine tumors (PanNET), the second most common type of pancreatic cancer, varies significantly, and up to 15% of patients develop metastasis. Although certain morphological characteristics of PanNETs have been associated with patient outcome, there are no available morphology-based prognostic markers. Given that current clinical histopathology markers are unable to identify high-risk PanNET patients, the development of accurate prognostic biomarkers is needed. Here, we describe a novel machine learning, multiclassification pipeline to predict the risk of metastasis using morphological information from whole tissue slides. METHODS: Digital images from surgically resected tissues from 89 PanNET patients were used. Pathologist-annotated regions were extracted to train a convolutional neural network (CNN) to identify tiles consisting of PanNET, stroma, normal pancreas parenchyma, and fat. Computationally annotated cancer or stroma tiles and patient metastasis status were used to train CNN to calculate a region based metastatic risk score. Aggregation of the metastatic probability scores across the slide was performed to predict the risk of metastasis. RESULTS: The ability of CNN to discriminate different tissues was high (per-tile accuracy >95%; whole slide cancer regions Jaccard index = 79%). Cancer and stromal tiles with high evaluated probability provided F1 scores of 0.82 and 0.69, respectively, when we compared tissues from patients who developed metastasis and those who did not. The final model identified low-risk (n = 76) and high-risk (n = 13) patients, as well as predicted metastasis-free survival (hazard ratio: 4.71) after adjusting for common clinicopathological variables, especially in grade I/II patients. CONCLUSION: Using slides from surgically resected PanNETs, our novel, multiclassification, deep learning pipeline was able to predict the risk of metastasis in PanNET patients. Our results suggest the presence of prognostic morphological patterns in PanNET tissues, and that these patterns may help guide clinical decision making.
BACKGROUND: The prognosis of patients with pancreatic neuroendocrine tumors (PanNET), the second most common type of pancreatic cancer, varies significantly, and up to 15% of patients develop metastasis. Although certain morphological characteristics of PanNETs have been associated with patient outcome, there are no available morphology-based prognostic markers. Given that current clinical histopathology markers are unable to identify high-risk PanNET patients, the development of accurate prognostic biomarkers is needed. Here, we describe a novel machine learning, multiclassification pipeline to predict the risk of metastasis using morphological information from whole tissue slides. METHODS: Digital images from surgically resected tissues from 89 PanNET patients were used. Pathologist-annotated regions were extracted to train a convolutional neural network (CNN) to identify tiles consisting of PanNET, stroma, normal pancreas parenchyma, and fat. Computationally annotated cancer or stroma tiles and patient metastasis status were used to train CNN to calculate a region based metastatic risk score. Aggregation of the metastatic probability scores across the slide was performed to predict the risk of metastasis. RESULTS: The ability of CNN to discriminate different tissues was high (per-tile accuracy >95%; whole slide cancer regions Jaccard index = 79%). Cancer and stromal tiles with high evaluated probability provided F1 scores of 0.82 and 0.69, respectively, when we compared tissues from patients who developed metastasis and those who did not. The final model identified low-risk (n = 76) and high-risk (n = 13) patients, as well as predicted metastasis-free survival (hazard ratio: 4.71) after adjusting for common clinicopathological variables, especially in grade I/II patients. CONCLUSION: Using slides from surgically resected PanNETs, our novel, multiclassification, deep learning pipeline was able to predict the risk of metastasis in PanNET patients. Our results suggest the presence of prognostic morphological patterns in PanNET tissues, and that these patterns may help guide clinical decision making.
Authors: Laura H Tang; Brian R Untch; Diane L Reidy; Eileen O'Reilly; Deepti Dhall; Lily Jih; Olca Basturk; Peter J Allen; David S Klimstra Journal: Clin Cancer Res Date: 2015-10-19 Impact factor: 12.531
Authors: James C Yao; Manal Hassan; Alexandria Phan; Cecile Dagohoy; Colleen Leary; Jeannette E Mares; Eddie K Abdalla; Jason B Fleming; Jean-Nicolas Vauthey; Asif Rashid; Douglas B Evans Journal: J Clin Oncol Date: 2008-06-20 Impact factor: 44.544
Authors: Joshua S Hill; James T McPhee; Theodore P McDade; Zheng Zhou; Mary E Sullivan; Giles F Whalen; Jennifer F Tseng Journal: Cancer Date: 2009-02-15 Impact factor: 6.860
Authors: Antonio Viúdez; Filipe L F Carvalho; Zahra Maleki; Marianna Zahurak; Daniel Laheru; Alejandro Stark; Nilofer S Azad; Christopher L Wolfgang; Stephen Baylin; James G Herman; Ana De Jesus-Acosta Journal: Oncotarget Date: 2016-05-03
Authors: Khalid AbdulJabbar; Shan E Ahmed Raza; Rachel Rosenthal; Mariam Jamal-Hanjani; Selvaraju Veeriah; Ayse Akarca; Tom Lund; David A Moore; Roberto Salgado; Maise Al Bakir; Luis Zapata; Crispin T Hiley; Leah Officer; Marco Sereno; Claire Rachel Smith; Sherene Loi; Allan Hackshaw; Teresa Marafioti; Sergio A Quezada; Nicholas McGranahan; John Le Quesne; Charles Swanton; Yinyin Yuan Journal: Nat Med Date: 2020-05-27 Impact factor: 53.440
Authors: Scott Doyle; Michael D Feldman; Natalie Shih; John Tomaszewski; Anant Madabhushi Journal: BMC Bioinformatics Date: 2012-10-30 Impact factor: 3.169
Authors: Jaakko S Knuutila; Pilvi Riihilä; Antti Karlsson; Mikko Tukiainen; Lauri Talve; Liisa Nissinen; Veli-Matti Kähäri Journal: Sci Rep Date: 2022-06-14 Impact factor: 4.996