Piloting a Deep Learning Model for Predicting Nuclear BAP1 Immunohistochemical Expression of Uveal Melanoma from Hematoxylin-and-Eosin Sections

Zhang, Hongrun; Kalirai, Helen; Acha-Sagredo, Amelia; Yang, Xiaoyun; Zheng, Yalin; Coupland, Sarah E.

doi:10.1167/tvst.9.2.50

Cited by 30 publications

(20 citation statements)

References 45 publications

(66 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While our results need further verification in different experimental settings, the CIBERSORTx algorithm we employed has been extensively validated with techniques such as flow cytometry, mass cytometry, immunohistochemistry, and single-cell RNA-seq by multiple different groups, providing robust data confirming its accuracy [40,[63][64][65]. Nevertheless, others have already created mRNA/miRNA/DNA-based scores with strong prognostic values in UM tumors [33,[66][67][68][69][70][71][72]. Rather than compete with them, our foremost motive for creating the cell type-based prognostic score was to select the cell types playing a prime role in the biology of UM progression Previous research has identified extensive lymphocyte infiltration as a negative prognostic factor in UM; however, the prognostic role of individual subsets was not assessed Another dominant subset of tumor-infiltrating immune cells is macrophages.…”

Section: Journal Of Immunology Researchmentioning

confidence: 75%

Bioinformatic Analysis Reveals Central Role for Tumor-Infiltrating Immune Cells in Uveal Melanoma Progression

Lachota

Lennikov

Malmberg

et al. 2021

Journal of Immunology Research

View full text Add to dashboard Cite

Tumor-infiltrating immune cells are capable of effective cancer surveillance, and their abundance is linked to better prognosis in numerous tumor types. However, in uveal melanoma (UM), extensive immune infiltrate is associated with poor survival. This study aims to decipher the role of different tumor-infiltrating cell subsets in UM in order to identify potential targets for future immunotherapeutic treatment. We have chosen the TCGA-UVM cohort as a training dataset and GSE22138 as a testing dataset by mining publicly available databases. The abundance of 22 immune cell types was estimated using CIBERSORTx. Then, to determine the significance of tumor-infiltrating cell subsets in UM, we built a multicell type prognostic signature, which was validated in the testing cohort. The created signature was built upon the negative prognostic role of CD8+ T cells and M0 macrophages and the positive role of neutrophils. Based on the created signature score, we divided the patients into low- and high-risk groups. Kaplan-Meier, Cox, and ROC analyses demonstrated superior performance of our risk score compared to either clinical or pathologic characteristics of both cohorts. Further, we found the molecular pathways associated with cancer immunoevasion and metastasis to be enriched in the high-risk group, explaining both the lack of adequate immune surveillance despite increased infiltration of CD8+ T cells as well as the higher metastatic potential. Genes associated with tryptophan metabolism (IDO1 and KYNU) and metalloproteinases were among the most differentially expressed between the high- and low-risk groups. Our correlation analyses interpreted in context of published in vitro data strongly suggest the central role of CD8+ T cells in shifting the UM tumor microenvironment towards suppressive and metastasis-promoting. Therefore, we propose further investigations of IDO1 and metalloproteinases as novel targets for immunotherapy in lymphocyte-rich metastatic UM patients.

show abstract

Section: Journal Of Immunology Researchmentioning

confidence: 75%

Bioinformatic Analysis Reveals Central Role for Tumor-Infiltrating Immune Cells in Uveal Melanoma Progression

Lachota

Lennikov

Malmberg

et al. 2021

Journal of Immunology Research

View full text Add to dashboard Cite

show abstract

“…With the rapidly growing number of digital image analysis and artificial intelligence solutions for assisting pathologists with diagnosis, treatment prediction and prognostication, further developments in this field should be expected. [37][38][39][40][41] Further, severe anaplasia correlated strongly with gain of chromosome 6p. The central part of the short arm of chromosome 6p has been reported to harbor oncogenes that are linked to tumor progression and gains at 6p have been associated with metastatic disease and poor prognosis in other cancers.…”

Section: Discussionmentioning

confidence: 99%

Gain of Chromosome 6p Correlates with Severe Anaplasia, Cellular Hyperchromasia, and Extraocular Spread of Retinoblastoma

Stålhammar

Yeung

Mendoza

et al. 2022

Ophthalmology Science

View full text Add to dashboard Cite

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

show abstract

“…In this study, a DenseNet neural network architecture outperformed three other state-of-the-art neural network architectures, as well as a support vector machine. A later study by Zhang et al reportedly outperformed these results using a larger data set [51]. Here, a ResNet18 model was used to extract feature vectors from tiles before assembling a set of feature vectors corresponding to a single slide into a feature map according to their spatial locations.…”

Section: Predicting Gene Expression and Hormone Receptor Statusmentioning

confidence: 98%

Deep Learning of Histopathological Features for the Prediction of Tumour Molecular Genetics

et al. 2021

View full text Add to dashboard Cite

Advanced diagnostics are enabling cancer treatments to become increasingly tailored to the individual through developments in immunotherapies and targeted therapies. However, long turnaround times and high costs of molecular testing hinder the widespread implementation of targeted cancer treatments. Meanwhile, gold-standard histopathological assessment carried out by a trained pathologist is widely regarded as routine and mandatory in most cancers. Recently, methods have been developed to mine hidden information from histopathological slides using deep learning applied to scanned and digitized slides; deep learning comprises a collection of computational methods which learn patterns in data in order to make predictions. Such methods have been reported to be successful in a variety of cancers for predicting the presence of biomarkers such as driver mutations, tumour mutational burden, and microsatellite instability. This information could prove valuable to pathologists and oncologists in clinical decision making for cancer treatment and triage for in-depth sequencing. In addition to identifying molecular features, deep learning has been applied to predict prognosis and treatment response in certain cancers. Despite reported successes, many challenges remain before the clinical implementation of such diagnostic strategies in the clinical setting is possible. This review aims to outline recent developments in the field of deep learning for predicting molecular genetics from histopathological slides, as well as to highlight limitations and pitfalls of working with histopathology slides in deep learning.

show abstract

Piloting a Deep Learning Model for Predicting Nuclear BAP1 Immunohistochemical Expression of Uveal Melanoma from Hematoxylin-and-Eosin Sections

Abstract: Piloting a deep learning model for predicting nuclear BAP1 immunohistochemical expression of uveal melanoma from hematoxylin-and-eosin sections.

Cited by 30 publications

References 45 publications

Bioinformatic Analysis Reveals Central Role for Tumor-Infiltrating Immune Cells in Uveal Melanoma Progression

Bioinformatic Analysis Reveals Central Role for Tumor-Infiltrating Immune Cells in Uveal Melanoma Progression

Gain of Chromosome 6p Correlates with Severe Anaplasia, Cellular Hyperchromasia, and Extraocular Spread of Retinoblastoma

Deep Learning of Histopathological Features for the Prediction of Tumour Molecular Genetics

Contact Info

Product

Resources

About