Deep Learning Predicts EBV Status in Gastric Cancer Based on Spatial Patterns of Lymphocyte Infiltration

N6-methyladenosine methylation regulates the tumor microenvironment of Epstein-Barr virus-associated gastric cancer

BACKGROUND

N6-methyladenosine (m6A) methylation modification exists in Epstein-Barr virus (EBV) primary infection, latency, and lytic reactivation. It also modifies EBV latent genes and lytic genes. EBV-associated gastric cancer (EBVaGC) is a distinctive molecular subtype of GC. We hypothesized EBV and m6A methylation regulators interact with each other in EBVaGC to differentiate it from other types of GC.

AIM

To investigate the mechanisms of m6A methylation regulators in EBVaGC to determine the differentiating factors from other types of GC.

METHODS

First, The Cancer Gene Atlas and Gene Expression Omnibus databases were used to analyze the expression pattern of m6A methylation regulators between EBVaGC and EBV-negative GC (EBVnGC). Second, we identified Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) functional enrichment of m6A-related differentially expressed genes. We quantified the relative abundance of immune cells and inflammatory factors in the tumor microenvironment (TME). Finally, cell counting kit-8 cell proliferation test, transwell test, and flow cytometry were used to verify the effect of insulin-like growth factor binding protein 1 (IGFBP1) in EBVaGC cell lines.

RESULTS

m6A methylation regulators were involved in the occurrence and development of EBVaGC. Compared with EBVnGC, the expression levels of m6A methylation regulators Wilms tumor 1-associated protein, RNA binding motif protein 15B, CBL proto-oncogene like 1, leucine rich pentatricopeptide repeat containing, heterogeneous nuclear ribonucleoprotein A2B1, IGFBP1, and insulin-like growth factor 2 binding protein 1 were significantly downregulated in EBVaGC (P < 0.05). The overall survival rate of EBVaGC patients with a lower expression level of IGFBP1 was significantly higher (P = 0.046). GO and KEGG functional enrichment analyses showed that the immunity pathways were significantly activated and rich in immune cell infiltration in EBVaGC. Compared with EBVnGC, the infiltration of activated CD4+ T cells, activated CD8+ T cells, monocytes, activated dendritic cells, and plasmacytoid dendritic cells were significantly upregulated in EBVaGC (P < 0.001). In EBVaGC, the expression level of proinflammatory factors interleukin (IL)-17, IL-21, and interferon-γ and immunosuppressive factor IL-10 were significantly increased (P < 0.05). In vitro experiments demonstrated that the expression level of IGFBP1 was significantly lower in an EBVaGC cell line (SNU719) than in an EBVnGC cell line (AGS) (P < 0.05). IGFBP1 overexpression significantly attenuated proliferation and migration and promoted the apoptosis levels in SNU719. Interfering IGFBP1 significantly promoted proliferation and migration and attenuated the apoptosis levels in AGS.

CONCLUSION

m6A regulators could remodel the TME of EBVaGC, which is classified as an immune-inflamed phenotype and referred to as a “hot” tumor. Among these regulators, we demonstrated that IGFBP1 affected proliferation, migration, and apoptosis.

N6-methyladenosine methylation regulates the tumor microenvironment of Epstein-Barr virus-associated gastric cancer

BACKGROUND

N6-methyladenosine (m6A) methylation modification exists in Epstein-Barr virus (EBV) primary infection, latency, and lytic reactivation. It also modifies EBV latent genes and lytic genes. EBV-associated gastric cancer (EBVaGC) is a distinctive molecular subtype of GC. We hypothesized EBV and m6A methylation regulators interact with each other in EBVaGC to differentiate it from other types of GC.

AIM

To investigate the mechanisms of m6A methylation regulators in EBVaGC to determine the differentiating factors from other types of GC.

METHODS

First, The Cancer Gene Atlas and Gene Expression Omnibus databases were used to analyze the expression pattern of m6A methylation regulators between EBVaGC and EBV-negative GC (EBVnGC). Second, we identified Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) functional enrichment of m6A-related differentially expressed genes. We quantified the relative abundance of immune cells and inflammatory factors in the tumor microenvironment (TME). Finally, cell counting kit-8 cell proliferation test, transwell test, and flow cytometry were used to verify the effect of insulin-like growth factor binding protein 1 (IGFBP1) in EBVaGC cell lines.

RESULTS

m6A methylation regulators were involved in the occurrence and development of EBVaGC. Compared with EBVnGC, the expression levels of m6A methylation regulators Wilms tumor 1-associated protein, RNA binding motif protein 15B, CBL proto-oncogene like 1, leucine rich pentatricopeptide repeat containing, heterogeneous nuclear ribonucleoprotein A2B1, IGFBP1, and insulin-like growth factor 2 binding protein 1 were significantly downregulated in EBVaGC (P < 0.05). The overall survival rate of EBVaGC patients with a lower expression level of IGFBP1 was significantly higher (P = 0.046). GO and KEGG functional enrichment analyses showed that the immunity pathways were significantly activated and rich in immune cell infiltration in EBVaGC. Compared with EBVnGC, the infiltration of activated CD4+ T cells, activated CD8+ T cells, monocytes, activated dendritic cells, and plasmacytoid dendritic cells were significantly upregulated in EBVaGC (P < 0.001). In EBVaGC, the expression level of proinflammatory factors interleukin (IL)-17, IL-21, and interferon-γ and immunosuppressive factor IL-10 were significantly increased (P < 0.05). In vitro experiments demonstrated that the expression level of IGFBP1 was significantly lower in an EBVaGC cell line (SNU719) than in an EBVnGC cell line (AGS) (P < 0.05). IGFBP1 overexpression significantly attenuated proliferation and migration and promoted the apoptosis levels in SNU719. Interfering IGFBP1 significantly promoted proliferation and migration and attenuated the apoptosis levels in AGS.

CONCLUSION

m6A regulators could remodel the TME of EBVaGC, which is classified as an immune-inflamed phenotype and referred to as a "hot" tumor. Among these regulators, we demonstrated that IGFBP1 affected proliferation, migration, and apoptosis.

The tumor immune composition of mismatch repair deficient and Epstein-Barr virus-positive gastric cancer: A systematic review

BACKGROUND

Gastric cancer (GC), known for its unfavorable prognosis, has been classified in four distinct molecular subtypes. These subtypes not only exhibit differences in their genome and transcriptome but also in the composition of their tumor immune microenvironment. The microsatellite instable (MSI) and Epstein-Barr virus (EBV) positive GC subtypes show clear clinical benefits from immune checkpoint blockade, likely due to a neoantigen-driven and virus-driven antitumor immune response and high expression of immune checkpoint molecule PD-L1. However, even within these subtypes response to checkpoint inhibition is variable, which is potentially related to heterogeneity in the tumor immune microenvironment (TIME) and expression of co-inhibitory molecules. We conducted a systematic review to outline the current knowledge about the immunological features on the TIME of MSI and EBV + GCs.

METHODS

A systematic search was performed in PubMed, EMBASE and Cochrane Library. All articles from the year 1990 and onwards addressing immune features of gastric adenocarcinoma were reviewed and included based on predefined in- and exclusion criteria.

RESULTS

In total 5962 records were screened, of which 139 were included that reported immunological data on molecular GC subtypes. MSI and EBV + GCs were reported to have a more inflamed TIME compared to non-MSI and EBV- GC subtypes. Compared to microsatellite stable (MSS) tumors, MSI tumors were characterized by higher numbers of CD8 + and FoxP3 + T cells, and tumor infiltrating pro- and anti-inflammatory macrophages. HLA-deficiency was most common in MSI tumors compared to other molecular GC subtypes and associated with lower T and B cell infiltrates compared to HLA-proficient tumors. EBV + was associated with a high number of CD8 + T cells, Tregs, NK cells and macrophages. Expression of PD-L1, CTLA-4, Granzyme A and B, Perforin and interferon-gamma was enriched in EBV + tumors. Overall, MSI tumors harbored a more heterogeneous TIME in terms of immune cell composition and immune checkpoints compared to the EBV + tumors.

DISCUSSION AND CONCLUSION

MSI and EBV + GCs are highly Handbook for Conducting a Literature-Based Health Assessment Using OHAT Approach for Systematic Review and Evidence Integration.; 2019pro-inflammatory immune cell populations. Although studies on the direct comparison of EBV + and MSI tumors are limited, EBV + tumors show less intra-subgroup heterogeneity compared to MSI tumors. More studies are needed to identify how Intra-subgroup heterogeneity impacts response to immunotherapy efficacy.

Epstein-Barr virus miR-BART2-5p and miR-BART11-5p regulate cell proliferation, apoptosis, and migration by targeting RB and p21 in gastric carcinoma

Epstein-Barr virus (EBV) was the first tumor virus discovered in humans and can cause various types of tumors. Molecular classification suggests that EBV-associated gastric cancer (EBVaGC) is a unique subtype of gastric cancer.EBV was also the first virus found to encode its own microRNAs. However, the functions of many miRNAs remain unknown. This study investigated the roles and targets of miR-BART2-5p (BART2-5p) and miR-BART11-5p (BART11-5p) in EBVaGC. The expression of RB and p21 in EBVaGC and EBV negative GC (EBVnGC) cells was evaluated by western blotting. Expression of BART2-5p and BART11-5p in EBVaGC cells was evaluated by droplet digital PCR. The effects of BART2-5p or BART11-5p and their potential mechanisms were further investigated using cell counting kit-8, colony formation assay, flow cytometry analysis, and transwell assay. BART2-5p and BART11-5p were abundantly expressed and RB and p21 were downregulated in EBVaGC cells. BART2-5p regulates RB and p21 expression by directly targeting them. BART11-5p regulates RB expression by directly targeting RB. Both BART2-5p and BART11-5p promoted proliferation and migration of gastric cancer cells, while inhibiting apoptosis and promoting S-phase arrest of the cell cycle. Thus, BART2-5p and BART11-5p play important roles in promoting proliferation and migration, and inhibiting apoptosis in EBVaGC by targeting RB and p21, thus providing new potential therapeutic targets for EBVaGC.

Deep learning model to predict Epstein-Barr virus associated gastric cancer in histology

The detection of Epstein-Barr virus (EBV) in gastric cancer patients is crucial for clinical decision making, as it is related with specific treatment responses and prognoses. Despite its importance, the limited medical resources preclude universal EBV testing. Herein, we propose a deep learning-based EBV prediction method from H&E-stained whole-slide images (WSI). Our model was developed using 319 H&E stained WSI (26 EBV positive; TCGA dataset) from the Cancer Genome Atlas, and 108 WSI (8 EBV positive; ISH dataset) from an independent institution. Our deep learning model, EBVNet consists of two sequential components: a tumor classifier and an EBV classifier. We visualized the learned representation by the classifiers using UMAP. We externally validated the model using 60 additional WSI (7 being EBV positive; HGH dataset). We compared the model's performance with those of four pathologists. EBVNet achieved an AUPRC of 0.65, whereas the four pathologists yielded a mean AUPRC of 0.41. Moreover, EBVNet achieved an negative predictive value, sensitivity, specificity, precision, and F1-score of 0.98, 0.86, 0.92, 0.60, and 0.71, respectively. Our proposed model is expected to contribute to prescreen patients for confirmatory testing, potentially to save test-related cost and labor.

Prediction of Epstein-Barr Virus Status in Gastric Cancer Biopsy Specimens Using a Deep Learning Algorithm

IMPORTANCE

Epstein-Barr virus (EBV)-associated gastric cancer (EBV-GC) is 1 of 4 molecular subtypes of GC and is confirmed by an expensive molecular test, EBV-encoded small RNA in situ hybridization. EBV-GC has 2 histologic characteristics, lymphoid stroma and lace-like tumor pattern, but projecting EBV-GC at biopsy is difficult even for experienced pathologists.

OBJECTIVE

To develop and validate a deep learning algorithm to predict EBV status from pathology images of GC biopsy.

DESIGN, SETTING, AND PARTICIPANTS

This diagnostic study developed a deep learning classifier to predict EBV-GC using image patches of tissue microarray (TMA) and whole slide images (WSIs) of GC and applied it to GC biopsy specimens from GCs diagnosed at Kangbuk Samsung Hospital between 2011 and 2020. For a quantitative evaluation and EBV-GC prediction on biopsy specimens, the area of each class and the fraction in total tissue or tumor area were calculated. Data were analyzed from March 5, 2021, to February 10, 2022.

MAIN OUTCOMES AND MEASURES

Evaluation metrics of predictive model performance were assessed on accuracy, recall, precision, F1 score, area under the receiver operating characteristic curve (AUC), and κ coefficient.

RESULTS

This study included 137 184 image patches from 16 TMAs (708 tissue cores), 24 WSIs, and 286 biopsy images of GC. The classifier was able to classify EBV-GC image patches from TMAs and WSIs with 94.70% accuracy, 0.936 recall, 0.938 precision, 0.937 F1 score, and 0.909 κ coefficient. The classifier was used for predicting and measuring the area and fraction of EBV-GC on biopsy tissue specimens. A 10% cutoff value for the predicted fraction of EBV-GC to tissue (EBV-GC/tissue area) produced the best prediction results in EBV-GC biopsy specimens and showed the highest AUC value (0.8723; 95% CI, 0.7560-0.9501). That cutoff also obtained high sensitivity (0.895) and moderate specificity (0.745) compared with experienced pathologist sensitivity (0.842) and specificity (0.854) when using the presence of lymphoid stroma and a lace-like pattern as diagnostic criteria. On prediction maps, EBV-GCs with lace-like pattern and lymphoid stroma showed the same prediction results as EBV-GC, but cases lacking these histologic features revealed heterogeneous prediction results of EBV-GC and non-EBV-GC areas.

CONCLUSIONS AND RELEVANCE

This study showed the feasibility of EBV-GC prediction using a deep learning algorithm, even in biopsy samples. Use of such an image-based classifier before a confirmatory molecular test will reduce costs and tissue waste.

Artificial intelligence to identify genetic alterations in conventional histopathology

Precision oncology relies on the identification of targetable molecular alterations in tumor tissues. In many tumor types, a limited set of molecular tests is currently part of standard diagnostic workflows. However, universal testing for all targetable alterations, especially rare ones, is limited by the cost and availability of molecular assays. From 2017 to 2021, multiple studies have shown that artificial intelligence (AI) methods can predict the probability of specific genetic alterations directly from conventional hematoxylin and eosin (H&E) tissue slides. Although these methods are currently less accurate than gold-standard testing (e.g. immunohistochemistry, polymerase chain reaction or next-generation sequencing), they could be used as pre-screening tools to reduce the workload of genetic analyses. In this systematic literature review, we summarize the state of the art in predicting molecular alterations from H&E using AI. We found that AI methods perform reasonably well across multiple tumor types, although few algorithms have been broadly validated. In addition, we found that genetic alterations in FGFR, IDH, PIK3CA, BRAF, TP53 and DNA repair pathways are predictable from H&E in multiple tumor types, while many other genetic alterations have rarely been investigated or were only poorly predictable. Finally, we discuss the next steps for the implementation of AI-based surrogate tests in diagnostic workflows. This article is protected by copyright. All rights reserved.