scholarly journals Deep learning detects virus presence in cancer histology

2019 ◽  
Author(s):  
Jakob Nikolas Kather ◽  
Jefree Schulte ◽  
Heike I. Grabsch ◽  
Chiara Loeffler ◽  
Hannah Muti ◽  
...  
Keyword(s):  
Deep Learning ◽  
Epstein Barr Virus ◽  
Human Cancer ◽  
Low Cost ◽  
External Validation ◽  
Area Under The Curve ◽  
Oncogenic Viruses ◽  
Barr Virus ◽  
Molecular Tests ◽  
Epstein Barr

AbstractOncogenic viruses like human papilloma virus (HPV) or Epstein Barr virus (EBV) are a major cause of human cancer. Viral oncogenesis has a direct impact on treatment decisions because virus-associated tumors can demand a lower intensity of chemotherapy and radiation or can be more susceptible to immune check-point inhibition. However, molecular tests for HPV and EBV are not ubiquitously available.We hypothesized that the histopathological features of virus-driven and non-virus driven cancers are sufficiently different to be detectable by artificial intelligence (AI) through deep learning-based analysis of images from routine hematoxylin and eosin (HE) stained slides. We show that deep transfer learning can predict presence of HPV in head and neck cancer with a patient-level 3-fold cross validated area-under-the-curve (AUC) of 0.89 [0.82; 0.94]. The same workflow was used for Epstein-Barr virus (EBV) driven gastric cancer achieving a cross-validated AUC of 0.80 [0.70; 0.92] and a similar performance in external validation sets. Reverse-engineering our deep neural networks, we show that the key morphological features can be made understandable to humans.This workflow could enable a fast and low-cost method to identify virus-induced cancer in clinical trials or clinical routine. At the same time, our approach for feature visualization allows pathologists to look into the black box of deep learning, enabling them to check the plausibility of computer-based image classification.

scholarly journals Viral Causes of Lymphoma: The History of Epstein-Barr Virus and Human T-Lymphotropic Virus 1

2017 ◽  
Vol 8 ◽  
pp. 1178122X1773177 ◽  
Author(s):  
Daniel Esau
Keyword(s):  
Epstein Barr Virus ◽  
Human Cancer ◽  
Human Herpesvirus ◽  
Kaposi Sarcoma ◽  
Cancer Type ◽  
Oncogenic Viruses ◽  
T Lymphotropic Virus ◽  
Barr Virus ◽  
History Of ◽  
Epstein Barr

In 1964, Epstein, Barr, and Achong published a report outlining their discovery of viral particles in lymphoblasts isolated from a patient with Burkitt lymphoma. The Epstein-Barr virus (EBV) was the first human cancer virus to be described, and its discovery paved the way for further investigations into the oncogenic potential of viruses. In the decades following the discovery of EBV, multinational research efforts led to the discovery of further viral causes of various human cancers. Lymphomas are perhaps the cancer type that is most closely associated with oncogenic viruses: infection with EBV, human T-lymphotropic virus 1 (HTLV-1), human immunodeficiency virus (HIV), Kaposi sarcoma-associated herpesvirus/human herpesvirus 8, and hepatitis C virus have all been associated with lymphomagenesis. Lymphomas have also played an important role in the history of oncoviruses, as both the first human oncovirus (EBV) and the first human retrovirus (HTLV-1) were discovered through isolates taken from patients with unique lymphoma syndromes. The history of the discovery of these 2 key oncoviruses is presented here, and their impact on further medical research, using the specific example of HIV research, is briefly discussed.

scholarly journals Detecting immunotherapy-sensitive subtype in gastric cancer using histologic image-based deep learning

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Munetoshi Hinata ◽  
Tetsuo Ushiku
Keyword(s):  
Deep Learning ◽  
Data Augmentation ◽  
External Validation ◽  
Area Under The Curve ◽  
The Cancer Genome Atlas ◽  
Molecular Testing ◽  
Favorable Result ◽  
Gastric Cancers ◽  
Barr Virus ◽  
Epstein Barr

AbstractImmune checkpoint inhibitor (ICI) therapy is widely used but effective only in a subset of gastric cancers. Epstein–Barr virus (EBV)-positive and microsatellite instability (MSI) / mismatch repair deficient (dMMR) tumors have been reported to be highly responsive to ICIs. However, detecting these subtypes requires costly techniques, such as immunohistochemistry and molecular testing. In the present study, we constructed a histology-based deep learning model that aimed to screen this immunotherapy-sensitive subgroup efficiently. We processed whole slide images of 408 cases of gastric adenocarcinoma, including 108 EBV, 58 MSI/dMMR, and 242 other subtypes. Many images generated by data augmentation of the learning set were used for training convolutional neural networks to establish an automatic detection platform for EBV and MSI/dMMR subtypes, and the test sets of images were used to verify the learning outcome. Our model detected the subgroup (EBV + MSI/dMMR tumors) with high accuracy in test cases with an area under the curve of 0.947 (0.901–0.992). This result was slightly better than when EBV and MSI/dMMR tumors were detected separately. In an external validation cohort including 244 gastric cancers from The Cancer Genome Atlas database, our model showed a favorable result for detecting the “EBV + MSI/dMMR” subgroup with an AUC of 0.870 (0.809–0.931). In addition, a visualization of the trained neural network highlighted intraepithelial lymphocytosis as the ground for prediction, suggesting that this feature is a discriminative characteristic shared by EBV and MSI/dMMR tumors. Histology-based deep learning models are expected to be used for detecting EBV and MSI/dMMR gastric cancers as economical and less time-consuming alternatives, which may help to effectively stratify patients who respond to ICIs.

Association of normal oral mucosa with DNA of the main types of oncogenic viruses

2020 ◽  
pp. 60-63
Author(s):  
S.I. Kutukova ◽  
A.B. Chukhlovin ◽  
A.I. Yaremenko ◽  
Yu.V. Ivaskova ◽  
A.Ya. Razumova ◽  
...  
Keyword(s):  
Oral Mucosa ◽  
Epstein Barr Virus ◽  
Oncogenic Viruses ◽  
Viral Dna ◽  
Dna Viruses ◽  
Normal Oral Mucosa ◽  
Barr Virus ◽  
Ebv Dna ◽  
Epstein Barr ◽  
Hpv 18

The aim of the study was to assess the prevalence of DNA viruses (HSV I and II, CMV, EBV, HPV6.11, HPV16 and HPV18) in the native oral mucosa of healthy volunteers (n=50; 30 men (60.0%), 20 women (40.0%); 25—74 years, median age — 55.0 years (95% CI 47.60-56.76)). All samples of the normal oral mucosa were detected by real-time PCR to detect viral DNA. The majority of the examined — 76% (33/50) — revealed the DNA: one type of viral DNA in 17 (38.00%) of the examined, a combination of the two types in 14 (28.00%). In the normal oral mucosa, DNA of Epstein-Barr virus was significantly more often detected: 15 (30.00%) (p = 0.0276) and human papilloma viruses 27 (54.00%) (p <0.0001), especially HPV-18 (24 (48.00%)): mono-association in 9 (18.00%) examined and in 7 (14.00%) in combination with EBV DNA (p = 0.0253).

Epstein-Barr virus as the cause of a human cancer

Nature ◽  
1978 ◽  
Vol 274 (5673) ◽  
pp. 740-740 ◽  
Author(s):  
M. A. Epstein
Keyword(s):  
Epstein Barr Virus ◽  
Human Cancer ◽  
Barr Virus ◽  
Epstein Barr

scholarly journals Methodological Approaches to Study Extracellular Vesicle miRNAs in Epstein–Barr Virus-Associated Cancers

2018 ◽  
Vol 19 (9) ◽  
pp. 2810 ◽  
Author(s):  
Li Sun ◽  
David Meckes
Keyword(s):  
Epstein Barr Virus ◽  
Biomarker Discovery ◽  
Immune Escape ◽  
Human Cancer ◽  
Biological Fluids ◽  
Extracellular Vesicle ◽  
Cellular Mirnas ◽  
Barr Virus ◽  
Diagnosis And Prognosis ◽  
Epstein Barr

Epstein Barr-virus (EBV) was the first virus identified to be associated with human cancer in 1964 and is found ubiquitously throughout the world’s population. It is now established that EBV contributes to the development and progression of multiple human cancers of both lymphoid and epithelial cell origins. EBV encoded miRNAs play an important role in tumor proliferation, angiogenesis, immune escape, tissue invasion, and metastasis. Recently, EBV miRNAs have been found to be released from infected cancer cells in extracellular vesicles (EVs) and regulate gene expression in neighboring uninfected cells present in the tumor microenvironment and possibly at distal sites. As EVs are abundant in many biological fluids, the viral and cellular miRNAs present within EBV-modified EVs may serve as noninvasion markers for cancer diagnosis and prognosis. In this review, we discuss recent advances in EV isolation and miRNA detection, and provide a complete workflow for EV purification from plasma and deep-sequencing for biomarker discovery.

Imaging and therapy for Epstein-Barr virus-associated gastric cancer

2007 ◽  
Vol 25 (18_suppl) ◽  
pp. 4644-4644
Author(s):  
D. Fu ◽  
J. Chong ◽  
C. Foss ◽  
J. Fox ◽  
S. Wang ◽  
...  
Keyword(s):  
Epstein Barr Virus ◽  
Tumor Regression ◽  
Human Cancer ◽  
Nucleoside Analogs ◽  
Scid Mice ◽  
Tumor Xenografts ◽  
Barr Virus ◽  
Bortezomib Treatment ◽  
Epstein Barr

4644 Background: Epstein-Barr virus (EBV) has been identified in a wide variety of malignancies, including gastric carcinomas. The virus encodes kinases that phosphorylate nucleoside analogs such as 2’-deoxy-2’-fluoro-5-iodo-1-beta-D- arabinofuranosyluracil (FIAU). We hypothesized that it might be possible to use the viral enzyme to specifically concentrate [125I]FIAU or [131I] FIAU in tumor cells harboring virus and thus deliver imaging and therapeutic radiation. Bortezomib is a potent stimulator of viral kinase expression in EBV tumor cell lines. Methods: We imaged lytic induction in vivo and evaluated the effect of [131I] FIAU on human cancer xenografts in SCID mice. These include a tumor line engineered to constitutively express the EBV thymide kinase (EBVTK), and a control engineered with a sham vector (SHAM), as well one EBV-associated human gastric tumor (KT tumor). Mice were treated with buffer, bortezomib (2μg/g), or radiolabeled FIAU or radiolabeled FIAU and bortezomib in combination. For imaging, mice, [125I]-FIAU and SPECT/CT were used. For therapy, 131I-FIAU was used and tumor dimensions were monitored with calipers. Results: SPECT/CT imaging with [125I]-FIAU of tumor-bearing SCID mice showed selective concentration of radiotracer in tumor tissue in EBVTK (3/3) and in EBV-associated KT tumors (3/3) when animals were pretreated with bortezomib. Treatment with buffer had no effect on 3 EBVTK tumors and 3 SHAM tumors all of which increased in volume. Treatment with 1.6 mCi of [131I]-FIAU alone led to tumor response in 3/3 mice with EBVTK tumors and 0/3 mice with SHAM tumors. Treatment with [131I]-FIAU alone had no effect on EBV KT tumor xenografts (0/3) and all tumors increased in volume. Treatment with bortezomib induced modest responses in all KT tumors. However, treatment with bortezomib and [131I]-FIAU led to marked tumor regression (>80%) in EBV-associated KT tumors (3/3). Conclusions: Treatment with bortezomib leads to selective concentration of radiolabeled FIAU in the EBV-associated tumor xenografts. In combination with [131I]-FIAU it leads to tumor regression. No significant financial relationships to disclose.

Pathogenic Role of Epstein-Barr Virus in Human Cancer

Intervirology ◽  
1995 ◽  
Vol 38 (3-4) ◽  
pp. 214-220 ◽  
Author(s):  
Kenzo Takada ◽  
Norio Shimizu ◽  
Akiko Tanabe-Tochikura ◽  
Yasuyuki Kuroiwa
Keyword(s):  
Epstein Barr Virus ◽  
Human Cancer ◽  
Pathogenic Role ◽  
Barr Virus ◽  
Epstein Barr

scholarly journals Long Non-coding RNAs in Gammaherpesvirus Infections: Their Roles in Tumorigenic Mechanisms

2021 ◽  
Vol 11 ◽  
Author(s):  
Wen Liu ◽  
Yan Zhang ◽  
Bing Luo
Keyword(s):  
Epstein Barr Virus ◽  
Oncogenic Viruses ◽  
Regulate Gene Expression ◽  
Barr Virus ◽  
Dna And Rna ◽  
Tumor Diagnostics ◽  
Non Coding Rnas ◽  
Epstein Barr ◽  
Underlying Mechanisms ◽  
Posttranscriptional Level

Long non-coding RNAs (lncRNAs) regulate gene expression at the epigenetic, transcriptional, or posttranscriptional level by interacting with protein, DNA, and RNA. Emerging evidence suggests that various lncRNAs are abnormally expressed and play indispensable roles in virus-triggered cancers. Besides, a growing number of studies have shown that virus-encoded lncRNAs participate in tumorigenesis. However, the functions of most lncRNAs in tumors caused by oncogenic viruses and their underlying mechanisms remain largely unknown. In this review, we summarize current findings regarding lncRNAs involved in cancers caused by Epstein–Barr virus (EBV) and Kaposi’s sarcoma herpesvirus (KSHV). Additionally, we discuss the contribution of lncRNAs to tumor occurrence, development, invasion, and metastasis; the roles of lncRNAs in key signaling pathways and their potential as biomarkers and therapeutic targets for tumor diagnostics and treatment.

Sign in / Sign up

Export Citation Format

Share Document