Search

Background/Objectives: Artificial intelligence (AI)-assisted endoscopic ultrasonography (EUS) diagnostic tools have shown excellent performance in diagnosing gastric mesenchymal tumors. This study aimed to assess whether incorporating clinical and endoscopic factors into AI-assisted EUS classification models based on digital image analysis could improve the diagnostic performance of AI-assisted EUS diagnostic tools. Methods: We retrospectively analyzed the data of 464 patients who underwent both EUS and surgical resection of gastric mesenchymal tumors, including 294 gastrointestinal stromal tumors (GISTs), 52 leiomyomas, and 41 schwannomas. AI-assisted classification models for GISTs and non-GIST tumors were developed utilizing clinical and endoscopic factors and digital EUS image analysis. Results: Regarding the baseline EUS classification models, the area under the receiver operating characteristic (AUC) values of the logistic regression, decision tree, random forest, K-nearest neighbor (KNN), and support vector machine (SVM) models were 0.805, 0.673, 0.781, 0.740, and 0.791, respectively. Using the new classification models incorporating clinical and endoscopic factors into the baseline classification models, the AUC values of the logistic regression, decision tree, random forest, KNN, and SVM models increased to 0.853, 0.715, 0.896, 0.825, and 0.794, respectively. In particular, the random forest and KNN models exhibited significant improvement in performance in Delong’s test (both p < 0.001). Conclusion: The diagnostic performance of the AI-assisted EUS classification models improved when clinical and endoscopic factors were incorporated. Our results provided direction for developing new AI-assisted EUS models for gastric mesenchymal tumors.

Duodenal neuroendocrine tumors (NETs) are rare subepithelial tumors that arise from the neuroendocrine cells beneath the epithelial layer. However, an accurate histopathological diagnosis is difficult when tissue samples are obtained using conventional endoscopic forceps biopsy alone. This study aimed to evaluate the magnifying endoscopy with narrow-band imaging (ME-NBI) findings of duodenal NETs. We retrospectively analyzed a database of 22 duodenal NETs from 21 patients who underwent ME-NBI between January 2011 and June 2022. The ME-NBI, endosonographic, and histopathologic findings of duodenal NETs were analyzed. Nineteen lesions were located in the bulb, two were located in the superior duodenal angle, and one was located in the second portion of the duodenum. Eighteen lesions (82%) had IIa morphology, and nine (41%) had central depression on the surface. On endoscopic ultrasonography, almost all lesions (20/22, 91%) were located in the second and/or third layers, and the median tumor size was 6 mm. During ME-NBI, the microsurface pattern was regular in 18 lesions (82%) and absent in 4 (18%). The microvascular pattern was regular in 17 lesions (77%), irregular in 4 (18%), and absent in 1 (5%). Thickened subepithelial vessels were observed in 15 (68%) lesions. There was no difference in tumor size according to the presence or absence of thickened subepithelial vessels (6.1 ± 1.8 mm vs. 5.9 ± 3.8 mm, p = 0.860). In conclusion, the characteristic ME-NBI findings of duodenal NETs were regular microsurface and microvascular patterns and the presence of thickened subepithelial vessels. These ME-NBI features may be useful for differentiating duodenal NETs from other duodenal subepithelial lesions.

Exosomal miRNAs have been studied in various cancers as minimally invasive biomarkers. This study aimed to investigate the potential of exosomal microRNAs (miRNAs) as biomarkers for esophageal squamous cell carcinoma (ESCC). Exosomes were isolated from cultures of esophageal epithelial cell and ESCC cell lines using ExoDisc, and exosomal miRNAs were detected via miRNA sequencing. Of the differentially expressed 14 miRNAs, the top 2 up-regulated miRNAs (miR-205-5p and miR-429) and top 2 down-regulated miRNAs (miR-375-3p and miR-483-3p) were selected as ESCC target miRNAs. Four selected exosomal miRNAs were validated in the plasma of 20 healthy controls (HCs) and 40 ESCC patients via quantitative reverse transcription-polymerase chain reaction. The expression of plasma exosomal miR-205-5p and miR-429 significantly increased, while that of plasma exosomal miR-375-3p was significantly reduced in ESCC patients compared to that in HCs. At cut-off values of 5.04, 2.564, and 0.136, the sensitivity and specificity for the diagnosis of ESCC were 72.5% and 70.0% for miR-205-5p, 60.0% and 60.0% for miR-429, and 65.0% and 65.0% for miR-375-3p, respectively. Based on the exosomal miRNAs identified in ESCC cell lines, our study demonstrated that plasma exosomal miR-205-5p, miR-429, and miR-375-3p could serve as potential biomarkers for ESCC diagnosis.

We aimed to evaluate whether adding a sustained-release (SR) formula of mosapride to proton-pump inhibitors (PPIs) would be more effective in controlling symptoms than PPI alone in patients with gastroesophageal reflux disease (GERD). Sixty patients with heartburn and/or regurgitation were randomly assigned to two groups: mosapride SR 15 mg combined with esomeprazole 20 mg once daily (ME group) and esomeprazole 20 mg once daily alone (E group). The primary endpoint was the complete-resolution rate of GERD symptoms after eight-week medication, and the secondary endpoints were the complete-resolution rate of GERD symptoms after four-week medication, symptom-improvement rates ≥ 50% after four- and eight-week medication, and change in reflux-disease-questionnaire (RDQ) and GERD-health-related quality-of-life (GERD-HRQL) scores from baseline at four- and eight-week medication. No significant differences in complete-symptom-resolution rates at eight weeks and four weeks or in the changes in RDQ and GERD-HRQL scores from baseline at four- and eight-week medication were observed between the ME and E groups. The symptom-improvement rate of ≥50% after four and eight weeks was comparable between both groups. Adding mosapride SR to esomeprazole in patients with GERD provides no additional benefits in controlling GERD symptoms.

Background and Aims: The clinical significance of circulating tumor cells (CTCs) and TWIST expression in CTCs remains unelucidated in patients with gastric cancer (GC). Here, we evaluated CTCs and TWIST expression in CTCs and explored their correlation with prognosis in patients with metastatic GC. Methods: Peripheral blood samples were prospectively obtained from 31 patients with metastatic GC between September 2017 and December 2018, prior to treatment. CTCs were detected using a centrifugal microfluidic system and CTCs positive for TWIST immunostaining were defined as TWIST (+) CTCs. Results: CTCs and TWIST (+) CTCs were detected in 25 (80.6%) and 24 (77.4%) of the 31 patients, respectively. CTC count in patients with first diagnosis of metastatic cancer tended to be higher than that in those with recurrent metastatic cancer, but TWIST (+) CTC count was not different between the two groups. There was no difference in CTC and TWIST (+) CTC counts according to histopathologic type, peritoneal dissemination, hematogenous metastasis, serum tumor makers, or response to first-line chemotherapy. Patients with CTCs > 7.5/7.5 mL of blood showed shorter overall survival (OS) than those with CTCs ≤ 7.5/7.5 mL of blood (p = 0.049). Additionally, patients with TWIST (+) CTCs > 2.5/7.5 mL of blood tended to show shorter OS than those with TWIST (+) CTCs ≤ 2.5/7.5 mL of blood (p = 0.105). Conclusions: Our study demonstrated that high levels of CTCs and TWIST (+) CTCs were associated with worse OS.

Background and Aims: Endoscopic ultrasonography (EUS) is a useful diagnostic modality for evaluating gastric mesenchymal tumors; however, differentiating gastrointestinal stromal tumors (GISTs) from benign mesenchymal tumors such as leiomyomas and schwannomas remains challenging. For this reason, we developed a convolutional neural network computer-aided diagnosis (CNN-CAD) system that can analyze gastric mesenchymal tumors on EUS images. Methods: A total of 905 EUS images of gastric mesenchymal tumors (pathologically confirmed GIST, leiomyoma, and schwannoma) were used as a training dataset. Validation was performed using 212 EUS images of gastric mesenchymal tumors. This test dataset was interpreted by three experienced and three junior endoscopists. Results: The sensitivity, specificity, and accuracy of the CNN-CAD system for differentiating GISTs from non-GIST tumors were 83.0%, 75.5%, and 79.2%, respectively. Its diagnostic specificity and accuracy were significantly higher than those of two experienced and one junior endoscopists. In the further sequential analysis to differentiate leiomyoma from schwannoma in non-GIST tumors, the final diagnostic accuracy of the CNN-CAD system was 75.5%, which was significantly higher than that of two experienced and one junior endoscopists. Conclusions: Our CNN-CAD system showed high accuracy in diagnosing gastric mesenchymal tumors on EUS images. It may complement the current clinical practices in the EUS diagnosis of gastric mesenchymal tumors.

Background and Aims: Endoscopic resection (ER) for submucosal invasive colorectal cancer (T1 CRC) can be grouped as curative ER (C-ER) and non-curative ER (NC-ER). Little is known about the long-term outcomes of patients in these two groups. Therefore, we have evaluated the long-term outcomes in endoscopically resected T1 CRC patients in C-ER and NC-ER groups. Methods: We conducted a retrospective study on 220 patients with T1 CRC treated with ER from January 2007 to December 2017. First, we investigated the long-term outcomes (5-year overall survival [OS] and recurrence-free survival [RFS]) in the C-ER group (n = 49). In the NC-ER group (n = 171), we compared long-term outcomes between patients who underwent additional surgical resection (ASR) (n = 117) and those who did not (surveillance-only, n = 54). Results: T1 CRC patients in the C-ER and NC-ER groups had a median follow-up of 44 (interquartile range 32–69) months. There was no risk of tumor recurrence and cancer-related deaths in patients with C-ER. In the NC-ER group, the 5-year OS rates were 75.3% and 92.6% in the surveillance-only and ASR subgroups, respectively. The hazard ratio (HR) for ASR in NC-ER vs. surveillance-only in NC-ER was statistically insignificant. However, RFS rates were significantly different between the ASR (97.2%) and surveillance-only (84.0%) subgroups. Multivariate analysis indicated a submucosal invasion depth (SID) of >2500 µm and margin positivity to be associated with recurrence. Conclusions: The surveillance-only approach can be considered as an alternative surgical option for T1 CRCs in selected patients undergoing NC-ER.

Objectives: The incidence of Crohn's disease and the number of associated surgeries are increasing in Korea. This study investigated the effect of azathioprine/6-mercaptopurine (6-MP) and TNF-α antagonists on abdominal and perianal surgery in Korean patients with Crohn's disease. Design: A retrospective cohort study. Setting: Data from the Crohn's Disease Clinical Network and Cohort (CONNECT) were used. Patients with confirmed Crohn's disease between 1982 and 2008 from 32 hospitals in the Republic of Korea were enrolled. The effect of azathioprine/6-MP on abdominal and perianal surgery was analysed using logistic regression analysis adjusting for age and sex. Participants: In total, 1161 Crohn's disease patients were included in the Republic of Korea in the surgery (n = 462, male = 339, female = 123) and control groups (n = 699, male = 484, female = 215). Results: In total, 1161 patients were selected, with 462 patients who underwent abdominal (n = 245) or perianal surgery (n = 217). The preoperative usage rates of azathioprine/6-MP were 18.8% and 65.1% (p < 0.0001) in the surgery and control groups, respectively. The preoperative usage rates of TNF-α antagonists were 7.1% and 23.3% (p < 0.0001) in the surgery and control groups, respectively. A multivariate analysis revealed that the preoperative use of azathioprine/6-MP had an odds ratio of 0.094 for all surgeries (95% confidence interval [CI]: 0.070–0.127, p < 0.0001), 0.131 for abdominal surgery (95% CI: 397–1.599, p < 0.0001), and 0.059 for perianal surgery (95% CI: 0.038–0.091, p < 0.0001). The preoperative use of TNF-α antagonists had an odds ratio of 0.225 for all surgeries (95% CI: 0.151–0.335, p < 0.0001), 0.403 for abdominal surgery (95% CI: 0.261–0.623, p < 0.0001), and 0.064 for perianal surgery (95% CI: 0.026–0.160, p < 0.001). Strengths of this study: The study presents new evidence of the reduced risk of surgery following azathioprine use in Crohn's disease patients. Limitations of this study (1) This was not a controlled prospective study. (2) There was a selection bias specific to the CONNECT cohort. (3) The combination or sequential use of azathioprine/6-MP and TNF-α antagonists was not excluded. Conclusion: Azathioprine/6-MP is significantly associated with a reduced risk of abdominal and perianal surgery in Korean patients with Crohn's disease. [ABSTRACT FROM AUTHOR]

Background/Objectives: Artificial intelligence (AI)-assisted endoscopic ultrasonography (EUS) diagnostic tools have shown excellent performance in diagnosing gastric mesenchymal tumors. This study aimed to assess whether incorporating clinical and endoscopic factors into AI-assisted EUS classification models based on digital image analysis could improve the diagnostic performance of AI-assisted EUS diagnostic tools. Methods: We retrospectively analyzed the data of 464 patients who underwent both EUS and surgical resection of gastric mesenchymal tumors, including 294 gastrointestinal stromal tumors (GISTs), 52 leiomyomas, and 41 schwannomas. AI-assisted classification models for GISTs and non-GIST tumors were developed utilizing clinical and endoscopic factors and digital EUS image analysis. Results: Regarding the baseline EUS classification models, the area under the receiver operating characteristic (AUC) values of the logistic regression, decision tree, random forest, K-nearest neighbor (KNN), and support vector machine (SVM) models were 0.805, 0.673, 0.781, 0.740, and 0.791, respectively. Using the new classification models incorporating clinical and endoscopic factors into the baseline classification models, the AUC values of the logistic regression, decision tree, random forest, KNN, and SVM models increased to 0.853, 0.715, 0.896, 0.825, and 0.794, respectively. In particular, the random forest and KNN models exhibited significant improvement in performance in Delong's test (both p < 0.001). Conclusion: The diagnostic performance of the AI-assisted EUS classification models improved when clinical and endoscopic factors were incorporated. Our results provided direction for developing new AI-assisted EUS models for gastric mesenchymal tumors.