Gastric Precancerous Diseases Research Articles

Clinical significance of CD155 expression and correlation with cellular components of tumor microenvironment in gastric adenocarcinoma.

CD155 is recently emerging as a promising target in malignancies. However, the relationship between CD155 expression and tumor microenvironment (TME) cell infiltration in gastric adenocarcinoma (GAC) has rarely been clarified. We measured CD155 expression in specimens of gastric precancerous disease and GAC by immunohistochemistry. The association of CD155 expression with GAC progression and cells infiltration in TME was evaluated through 268 GAC tissues and public dataset analysis. We showed that the expression of CD155 was positively correlated with the pathological development of gastric precancerous disease (r = 0.521, P < 0.0001). GAC patients with high CD155 expression had a poorer overall survival (P = 0.033). Moreover, CD155 expression correlated with aggressive clinicopathological features including tumor volume, tumor stage, lymph node involvement, and cell proliferation (P <0.05). Remarkably, CD155 expression positively related to the infiltration of CD68+ macrophages in TME (P = 0.011). Meanwhile, the positive correlation was observed between CD155 and CD31 (P = 0.026). In addition, patients with high CD155 expression combined with low CD3, CD4, CD8, IL-17, IFN-γ or CD19 expression as well as those with high CD155 and α-SMA expression showed significantly worse overall survival (P < 0.05). CD155 may play a pivotal role in the development of GAC through both immunological and non-immunological mechanisms and be expected to become a novel target of immunotherapy in GAC patients.

Classification of precancerous lesions based on fusion of multiple hierarchical features

PurposeTo investigate an identification method for precancerous gastric cancer based on the fusion of superficial features and deep features of gastroscopic images. The purpose of this study is to make most use of superficial features and deep features to provide clinicians with clinical decision support to assist the diagnosis of precancerous gastric diseases and reduce the workload of doctors. MethodsAccording to the nature of gastroscopic images, 75-dimensional shallow features were manually designed, including histogram features, texture features and high-order features of the image; then, based on the constructed convolutional neural networks such as ResNet and GoogLeNet, before the output layer. A fully connected layer is added as the deep feature of the image. In order to ensure consistent feature weights, the number of neurons in the fully connected layer is designed to be 75 dimensions. Therefore, the superficial and deep features of the image are concatenated, and a machine learning classifier is used to identify gastric polyps, there are three types of gastric precancerous diseases such as gastric polyps, gastric ulcers and gastric erosions. ResultsA dataset with 420 images was collected for each disease, and divided into a training set and a test set with a ratio of 5:1, and then based on the dataset, three methods, such as traditional machine learning, deep learning, and feature fusion, were used respectively. For model training and testing of traditional machine learning and feature fusion, SVM, RF and BP neural network are used as the classification results of the classifier. For deep learning, the GoogLeNet, ResNet, and ResNeXt were implemented. The test results of the model on the test set show that the recognition accuracy of the proposed feature fusion method reaches (SVM: 85.18%; RF: 83.42%; BPNN: 85.18%), which is better than the traditional machine learning method (SVM: 80.17%; RF: 82.37%; BPNN: 84.12%) and the deep learning method (GoogLeNet: 82.54%; ResNet-18: 81.67%; ResNet-50: 81.67%; ResNeXt-50: 82.11%), which proves that this method has obvious advantages. ConclusionThis study provides a new strategy for the identification of precancerous gastric cancer, improving the efficiency and accuracy of precancerous gastric cancer identification, and hopes to provide substantial practical help for the identification of gastric precancerous diseases.

Study on molecular biological mechanism of Chinese herbal medicines for the treatment of gastric precancerous lesions based on data mining and network pharmacology.

To explore the molecular biological mechanisms of Chinese herbal medicines for the treatment of gastric precancerous lesions by data mining and network pharmacology. The keywords "gastric precancerous lesions""gastric precancerous disease""gastric mucosal intraepithelial neoplasia""gastric mucosal heterogeneous hyperplasia""gastric precancerous state""chronic gastritis, atrophic""combined Chinese and Western medicine""Chinese medicine therapy""efficacy evaluation" "randomized controlled trial"were searched in China Journal Full-text Database, Wanfang Data, VIP database, PubMed and Embase from 2001 to 2021. The information was extracted from the literature which met the inclusion and exclusion criteria, and the database was constructed to identify the high-frequency herbal medicines. The top six Chinese herbal medicines were analyzed by the network pharmacology methods, including the acquisition of herbs compounds and gastric precancerous lesions targets using Pharmacology Database and Analysis Platform and GeneCards databases, construction of protein-protein interaction network, and screening of core targets, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of core targets through Metascape platform, etc., to elucidate their active components, targets and pathways. A total of 482 compound prescriptions with 603 herbal medicines were included, and the top 6 herbal medicines with higher application frequency were Ume plum (63.35%), Curcuma longa (58.54%), Paeonia lactiflora (54.06%), Salvia miltiorrhiza (49.92%), Rhizoma alba (46.43%), and Astragalus membranaceus (45.44%). The results of the network pharmacological analysis showed that the active ingredients were 4 types from Ume plum, 3 from Curcuma longa, 9 from Paeonia lactiflora, 13 from Salvia miltiorrhiza, 7 from Astragalus alba, and 9 from Astragalus; 77 predicted targets were in Ume plum, 11 in Curcuma longa, 33 in Paeonia lactiflora, 58 in Salvia miltiorrhiza, 65 in Astragalus alba and 89 in Astragalus; and 98 crossover genes were obtained after these targets were compared with the disease genes, among which HSP90AA1, AKT1, TP53, STAT3, MAPK1 and TNF had higher relevance to the treatment of gastric precancerous lesions. The results of the GO enrichment analysis showed that the active ingredients of high frequency Chinese medicine mostly acted through biological processes such as response to inorganic substance, response to hormone, gland development, positive regulation of cell migration, positive regulation of cell motility, etc. The targets include cellular components such as vesicle lumen, secretory granule lumen, cytoplasmic vesicle lumen, transcription regulator complex, and with molecular functions such as kinase binding, protein kinase binding and DNA-binding transcription factor binding. The results of the KEGG pathway enrichment analysis showed that Paeonia lactiflora, Ulmus lucidus, Salvia miltiorrhiza and Astragalus mainly act through the cancer pathway and PI3K-AKT pathway; Curcuma longa and Rhizoma alba mainly act through the cancer pathway and proteoglycans in cancer, and all six herbs were involved in the cancer pathway and five herbs are involved in the PI3K-AKT pathway. In this study, we obtained the top 6 high-frequency Chinese herbal medicines in the treatment of gastric precancerous lesions by data mining method, and revealed that their mechanisms are involved in cell proliferation, differentiation, immunity, inflammation and other processes mainly through cancer pathway, PI3K-AKT signaling pathway, proteoglycans in cancer.

Content‐based gastric image retrieval using convolutional neural networks

AbstractThe endoscopy procedure has demonstrated great efficiency in detecting stomach lesions, with extensive numbers of endoscope images produced globally each day. The content‐based gastric image retrieval (CBGIR) system has demonstrated substantial potential in gastric image analysis. Gastric precancerous diseases (GPD) have higher prevalence in gastric cancer patients. Thus, effective intervention is crucial at the GPD stage. In this paper, a CBGIR method is proposed using a modified ResNet‐18 to generate binary hash codes for a rapid and accurate image retrieval process. We tested several popular models (AlexNet, VGGNet and ResNet), with ResNet‐18 determined as the optimum option. Our proposed method was valued using a GPD data set, resulting in a classification accuracy of 96.21 ± 0.66% and a mean average precision of 0.927 ± 0.006, outperforming other state‐of‐art conventional methods. Furthermore, we constructed a Gastric‐Map (GM) based on feature representations in order to visualize the retrieval results. This work has great auxiliary significance for endoscopists in terms of understanding the typical GPD characteristics and improving aided diagnosis.

PGC-MG7 combination could be used as a follow-up panel for monitoring dynamical progression of gastric precancerous diseases.

ObjectiveThe aim of this study was to investigate the value of the combined expression of the gastric mucosal differentiation protein pepsinogen C (PGC) and gastric cancer (GC)-associated antigen MG7 for the diagnosis of GC and prediction of the development from precancerous conditions to GC.MethodsThe gastric mucosal biopsies of 285 subjects enrolled from a region with a high incidence of GC were obtained and histopathologically examined. Subjects testing negative for GC (n=208) were followed up from 1998 to 2015. The levels of PGC and MG7 in the biopsies were determined by immunohistochemistry.ResultsPGC was positive in 91.4% of the non-atrophic gastritis, 26.5% of the atrophic gastritis, and 0% of the GC. MG7 was positive in 15.0% of the non-atrophic gastritis, 82.4% of the atrophic gastritis, and 94.8% of the GC. The non-atrophic gastritis group was predominantly “PGC+MG7−”. The atrophic gastritis and GC groups were predominantly “PGC−MG7+”. The rate of GC in subjects with “PGC−MG7+” staining was 113.4-fold higher [95% confidence interval (95% CI): 15.3−869.4, P<0.001] than that in subjects with other staining patterns. The sensitivity and specificity of the “PGC−MG7+” pattern were 92.2% and 78.8% for the detection of GC and 77.2% and 97.9% for GC and precancerous disease, respectively. In the follow-up cohort of non-GC subjects, the risk of developing GC was higher in those with the “PGC−MG7+” staining pattern.ConclusionsOur data suggest that the “PGC−MG7+” pattern can be employed as a useful follow-up panel for detecting individuals with a high risk of GC, and the dynamic assessment of the follow-up panel needs multi-centre large-scale validation in the future.

Fine-tuning Pre-trained Convolutional Neural Networks for Gastric Precancerous Disease Classification on Magnification Narrow-band Imaging Images

Gastric cancer(GC) is the fourth leading cause of cancer death worldwide. To prevent the occurrence of advanced GCs, there is a need for immediate detection and treatment of gastric precancerous and early cancerous lesions. Magnification endoscopy with narrow-band imaging (M-NBI) system as an advanced diagnostic imaging technology is widely used in evaluating gastric lesion types, which can interpret gastric lesion characteristics by enhancing contrasts between vessels and mucosal surfaces. Based on microvascular morphologies presented on M-NBI images, physicians can manually diagnose gastric lesions; but this is a tough work for unexperienced doctors and it is lacking of objectivity. In this study, we propose a transfer learning framework by fine-tuning pre-trained convolutional neural networks (CNNs) to classify gastric M-NBI images into three classes: chronic gastritis (CGT), low grade neoplasia (LGN) and early gastric cancer (EGC). The method we choose is used to compare with three kinds of traditional handcraft texture feature extraction methods and CNN models trained directly by our dataset. Results show that the performance of fine-tuned CNNs outperforms traditional handcraft features and trained CNNs. Experiments also illustrate that ResNet50 can achieve 0.96 accuracy, 0.92, 0.91 and 0.99 f1-scores for classifying M-NBI images into CGT, LGN and EGC. In conclusion, the proposed framework is suit for multi-classification tasks of gastric M-NBI images.

Correlation between negative expression of pepsinogen C and a series of phenotypic markers of gastric cancer in different gastric diseases.

Gastric tumorigenesis is a multistep process initiated by chronic superficial gastritis (SG), followed by atrophic gastritis (AG), then intestinal metaplasia (IM), and finally by dysplasia and adenocarcinoma according to the Correa model. Pepsinogen C (PGC) decreases gradually during progression of cancer, which makes PGC an ideal negative marker for GC. To explore the correlation between PGC and other positive tumor markers in different gastric diseases, we observed the expression of PGC, MG7‐Ag, MMP9, NM23, Ki‐67, and E‐cadherin by immunohistochemistry, quantitative RT‐PCR, and immunoblot analysis. Our results showed that in SG, PGC was highly expressed while malignant phenotype markers were rarely expressed. In contrast with SG, malignant phenotype markers were highly expressed while the positive rate of PGC reached only 1.44% in GC. So there was no coexpression of PGC and malignant phenotype markers in SG or GC tissues. Only in the AG group, which is well‐known to be gastric precancerous disease, coexpression of PGC and malignant phenotype markers was detected. Our results suggested that the expression of PGC in AG was negatively correlated with that of MG7‐Ag and MMP9. Of all AG, those with low expression of PGC and high expression of MG7‐Ag and MMP9 may possess a greater potential of malignant transformation. Combined detection of negative marker PGC and positive markers MG7‐Ag and MMP9 could be used as a potential follow‐up panel for monitoring dynamical progression of AG and improving the detection efficiency of high‐risk individuals of gastric cancer, and then taking necessary interventions on the target population.

Gastric precancerous diseases classification using CNN with a concise model.

Gastric precancerous diseases (GPD) may deteriorate into early gastric cancer if misdiagnosed, so it is important to help doctors recognize GPD accurately and quickly. In this paper, we realize the classification of 3-class GPD, namely, polyp, erosion, and ulcer using convolutional neural networks (CNN) with a concise model called the Gastric Precancerous Disease Network (GPDNet). GPDNet introduces fire modules from SqueezeNet to reduce the model size and parameters about 10 times while improving speed for quick classification. To maintain classification accuracy with fewer parameters, we propose an innovative method called iterative reinforced learning (IRL). After training GPDNet from scratch, we apply IRL to fine-tune the parameters whose values are close to 0, and then we take the modified model as a pretrained model for the next training. The result shows that IRL can improve the accuracy about 9% after 6 iterations. The final classification accuracy of our GPDNet was 88.90%, which is promising for clinical GPD recognition.

Serum OPN expression for identification of gastric cancer and atrophic gastritis and its influencing factors.

BackgroundMost studies have found that osteopontin (OPN) expression level is related to the poor prognosis of gastric cancer. However, few studies have examined the relationship between OPN expression and gastric precancerous diseases, and the potential role of OPN in the formation and development of GC. We investigated the relationships between serum OPN levels and the risks of gastric cancer (GC) and its precancerous disease, to explore the diagnostic efficacy of serum OPN level for GC and atrophic gastritis and its influencing factors.MethodsA total of 1,452 patients were enrolled, including 609 with mild superficial gastritis (SG), 594 with atrophic gastritis (AG) and 249 with GC. The levels of serum OPN and serum Helicobacter pylori IgG antibody were detected by enzyme-linked immunosorbent assay.ResultsSerum OPN levels increased from mild SG (1.99±1.91 ng/ml) to AG (2.37±2.27 ng/ml) to GC (5.94±4.52 ng/ml) (P≤0.002), along with increasing severity of gastric disease. OPN levels were significantly higher in patients with GC compared with the non-cancer population (2.17±2.10, P<0.0001). Serum OPN level was positively correlated with age and was higher in men than women, but was not correlated with H. pylori infection status. The area under the receiver operating characteristic curve was 0.805, the optimal cutoff was 2.56 ng/ml and the sensitivity and specificity were 74.3% and 71.8%, respectively, for the ability of serum OPN to discriminate GC.ConclusionsSerum OPN expression was closely related to the risks of GC and AG, and it might be a useful marker for the discrimination of GC. OPN level was positively correlated with age and male sex, but was not affected by H. pylori infection, and it was promoted by smoking and drinking, in patients with mild SG.

MiRNA‐199a‐3p: A potential circulating diagnostic biomarker for early gastric cancer

Tumor-associated miRNAs have been detected in serum or plasma. We investigated whether plasma miRNA-199a-3p could be a potential circulating biomarker for early gastric cancer (EGC). By using real-time qRT-PCR, the expression of miRNA-199a-3p were compared between these pre-operative plasmas from 30 EGC patients and 70 healthy controls, and between these pre-operative and post-operative plasmas. Further validation was on an independent set of plasmas from 50 EGC patients. The expression of miRNA-199a-3p (47.5 ± 6.5) in plasma in EGC patients was significantly higher than that from healthy controls (13.9 ± 2.7, P < 0.001) and gastric precancerous diseases (GPD) patients (19.2 ± 2.5, P = 0.004), respectively. Furthermore, the expression levels of miRNA-199a-3p (11.8 ± 2.9, P = 0.012) in the post-operative plasmas were significantly reduced when compared to the pre-operative plasmas. With respect of clinicopathological characteristics, the expression of miRNA-199a-3p in plasma was not associated with the depth of tumor invasion. Moreover, the AUC of the expression of miRNA-199a-3p in plasma for EGC diagnosis was 0.818, which was significantly higher than that of combined tumor markers (0.556). The sensitivity, specificity and accuracy of miRNA-199a-3p expression in plasma for EGC diagnosis were 76%, 74%, and 75%, respectively. Plasma miRNA-199a-3p could be a novel potential diagnostic biomarkers for EGC.

MiRNA-199a-3p in Plasma as a Potential Diagnostic Biomarker for Gastric Cancer

MicroRNA (miRNA) has been shown the potential of cancer diagnosis. We investigated whether plasma miRNA expression could discriminate between patients with and without gastric cancer. This study was divided into three steps: (1) miRNA microarray profiling on plasma samples from 20 gastric cancer patients and 20 healthy controls; (2) miRNA selection by real-time qRT-PCR on 30 pairs of plasma from patients and controls; and (3) qRT-PCR validation on an independent set of plasma from 180 gastric cancer patients, 80 healthy controls, and 20 patients with gastric precancerous diseases. Of the 959 human miRNAs analyzed by microarray, 37 up-regulated miRNAs and seven down-regulated miRNAs were found in gastric cancer plasma. Of the seven discrepant miRNAs validated on the plasma from 30 gastric cancer patients and 30 healthy controls, both miRNA-199a-3p and miRNA-151-5p were significantly elevated (p < 0.05) and were significantly reduced after surgery (p < 0.05) in gastric cancer patients. Further large-scale validation showed that these two miRNAs expressions in plasma were significantly higher in gastric cancer patients than healthy controls and patients with gastric precancerous diseases, respectively. However, only the expression of miRNA-199a-3p in plasma was significantly associated with tumor invasion and with lymph node metastasis and tumor, node, metastasis stage. This marker yielded an area under the receiver operating characteristic curve area of 0.837 with 80 % sensitivity and 74 % specificity in discriminating gastric cancer patients from healthy controls. In gastric cancer tissue, miRNA-199a-3p was expressed in the cytoplasm of tumor cells. miRNA-199a-3p in plasma could be a novel potential diagnostic biomarker for gastric cancer detection.