An artificial intelligence system for comprehensive pathologic outcome prediction in early gastric cancer through endoscopic image analysis (with video).

BACKGROUND: Accurate prediction of pathologic results for early gastric cancer (EGC) based on endoscopic findings is essential in deciding between endoscopic and surgical resection. This study aimed to develop an artificial intelligence (AI) model to assess comprehensive pathologic characteristics of EGC using white-light endoscopic images and videos. METHODS: To train the model, we retrospectively collected 4,336 images and prospectively included 153 videos from patients with EGC who underwent endoscopic or surgical resection. The performance of the model was tested and compared to that of 16 endoscopists (nine experts and seven novices) using a mutually exclusive set of 260 images and 10 videos. Finally, we conducted external validation using 436 images and 89 videos from another institution. RESULTS: After training, the model achieved predictive accuracies of 89.7% for undifferentiated histology, 88.0% for submucosal invasion, 87.9% for lymphovascular invasion (LVI), and 92.7% for lymph node metastasis (LNM), using endoscopic videos. The area under the curve values of the model were 0.992 for undifferentiated histology, 0.902 for submucosal invasion, 0.706 for LVI, and 0.680 for LNM in the test. In addition, the model showed significantly higher accuracy than the experts in predicting undifferentiated histology (92.7% vs. 71.6%), submucosal invasion (87.3% vs. 72.6%), and LNM (87.7% vs. 72.3%). The external validation showed accuracies of 75.6% and 71.9% for undifferentiated histology and submucosal invasion, respectively. CONCLUSIONS: AI may assist endoscopists with high predictive performance for differentiation status and invasion depth of EGC. Further research is needed to improve the detection of LVI and LNM.

Gastric microbiome signature for predicting metachronous recurrence after endoscopic resection of gastric neoplasm.

BACKGROUND: Changes in gastric microbiome are associated with gastric carcinogenesis. Studies on the association between gastric mucosa-associated gastric microbiome (MAM) and metachronous gastric cancer are limited. This study aimed to identify gastric MAM as a predictive factor for metachronous recurrence following endoscopic resection of gastric neoplasms. METHOD: Microbiome analyses were conducted for 81 patients in a prospective cohort to investigate surrogate markers to predict metachronous recurrence. Gastric MAM in non-cancerous corporal biopsy specimens was evaluated using Illumina MiSeq platform targeting 16S ribosomal DNA. RESULTS: Over a median follow-up duration of 53.8 months, 16 metachronous gastric neoplasms developed. Baseline gastric MAM varied with Helicobacter pylori infection status, but was unaffected by initial pathologic diagnosis, presence of atrophic gastritis, intestinal metaplasia, or synchronous lesions. The group with metachronous recurrence did not exhibit distinct phylogenetic diversity compared with the group devoid of recurrence but showed significant difference in ß-diversity. The study population could be classified into two distinct gastrotypes based on baseline gastric MAM: gastrotype 1, Helicobacter-abundant; gastrotype 2: Akkermansia-abundant. Patients in gastrotype 2 showed higher risk of metachronous recurrence than gastrotype (Cox proportional hazard analysis, adjusted hazard ratio [95% confidence interval]: 5.10 [1.09-23.79]). CONCLUSIONS: Gastric cancer patients can be classified into two distinct gastrotype groups by their MAM profiles, which were associated with different risk of metachronous recurrence.

Real-World Application of Artificial Intelligence for Detecting Pathologic Gastric Atypia and Neoplastic Lesions.

PURPOSE: Results of initial endoscopic biopsy of gastric lesions often differ from those of the final pathological diagnosis. We evaluated whether an artificial intelligence-based gastric lesion detection and diagnostic system, ENdoscopy as AI-powered Device Computer Aided Diagnosis for Gastroscopy (ENAD CAD-G), could reduce this discrepancy. MATERIALS AND METHODS: We retrospectively collected 24,948 endoscopic images of early gastric cancers (EGCs), dysplasia, and benign lesions from 9,892 patients who underwent esophagogastroduodenoscopy between 2011 and 2021. The diagnostic performance of ENAD CAD-G was evaluated using the following real-world datasets: patients referred from community clinics with initial biopsy results of atypia (n=154), participants who underwent endoscopic resection for neoplasms (Internal video set, n=140), and participants who underwent endoscopy for screening or suspicion of gastric neoplasm referred from community clinics (External video set, n=296). RESULTS: ENAD CAD-G classified the referred gastric lesions of atypia into EGC (accuracy, 82.47%; 95% confidence interval [CI], 76.46%-88.47%), dysplasia (88.31%; 83.24%-93.39%), and benign lesions (83.12%; 77.20%-89.03%). In the Internal video set, ENAD CAD-G identified dysplasia and EGC with diagnostic accuracies of 88.57% (95% CI, 83.30%-93.84%) and 91.43% (86.79%-96.07%), respectively, compared with an accuracy of 60.71% (52.62%-68.80%) for the initial biopsy results (P<0.001). In the External video set, ENAD CAD-G classified EGC, dysplasia, and benign lesions with diagnostic accuracies of 87.50% (83.73%-91.27%), 90.54% (87.21%-93.87%), and 88.85% (85.27%-92.44%), respectively. CONCLUSIONS: ENAD CAD-G is superior to initial biopsy for the detection and diagnosis of gastric lesions that require endoscopic resection. ENAD CAD-G can assist community endoscopists in identifying gastric lesions that require endoscopic resection.

Efficacy of Quadruple-coated Probiotics in Patients With Irritable Bowel Syndrome: A Randomized, Double-blind, Placebo-controlled, Parallel-group Study.

Background/Aims: To evaluate the efficacy of quadruple-coated probiotics (gQlab) in patients with irritable bowel syndrome (IBS), focusing on sex differences and IBS subtypes. Methods: One hundred and nine Rome III-diagnosed IBS patients were randomized into either a gQlab or placebo group and received either gQlab or a placebo for 4 weeks. Participants replied to questionnaires assessing compliance, symptoms, and safety. Fecal samples were collected at 0 and 4 weeks to measure the probiotic levels using real-time quantitative polymerase chain reaction (qPCR) and to perform metagenomic analysis via 16S ribosomal DNA sequencing. The primary endpoint was the change in the overall IBS symptoms after 4 weeks of treatment. Results: Ninety-two subjects (47 and 45 in the gQlab and placebo groups, respectively) completed the study protocol. At week 4, there was a higher relief of the overall IBS symptoms in the gQlab group (P = 0.005). The overall IBS symptom improvement was statistically significant (P = 0.017) in female patients of the gQlab group compared with the placebo group. Among the IBS subtypes, constipation-predominant IBS patients showed significant relief of the overall IBS symptoms (P = 0.002). At week 4, the fecal microbiome profiles between the 2 groups did not differ, but the qPCR levels of Lactobacillus plantarum, Lactobacillus acidophilus, Lactobacillus helveticus, Bifidobacterium longum, and Bifidobacterium breve were increased in the gQlab group (P < 0.05 by repeated measures ANOVA). Conclusions: gQlab administration can improve the overall IBS symptoms, especially in female and constipation-predominant IBS patients. Further research is necessary to clarify the pathophysiology behind sex-related treatment responses in IBS patients.

The Persistence of Hypertriglyceridemia and the Risk of Early Onset Colorectal Cancer According to Tumor Subsites: A Nationwide Population-based Study.

Purpose: The incidence of early-onset colorectal cancer (EoCRC) is increasing worldwide. The association between hypertriglyceridemia (HTG) and EoCRC risk remains unclear. Materials and Methods: We conducted a nationwide cohort study of 3,340,635 individuals aged 20-49 years who underwent health checkups between 2009 and 2011 under the Korean National Health Insurance Service. HTG was defined as serum triglyceride (TG) level ≥ 150 mg/dL. According to the change in TG status, participants were categorized into persistent normotriglyceridemia (NTG, group 1), NTG to HTG (group 2), HTG to NTG (group 3), and persistent HTG (group 4) groups. The EoCRC incidence was followed up until 2019. Results: In total, 7492 EoCRC cases developed after a mean of 6.05 years of follow-up. Group 4 had the highest risk of EoCRC (adjusted hazard ratio [aHR]: 1.097, 95% confidence interval [CI]: 1.025-1.174). While the risk of rectal cancer was significantly increased in groups 3 and 4 (aHR [95% CI]: 1.236[1.076-1.419] and 1.175[1.042-1.325], respectively), no significant risk differences were observed in right colon cancer. In group 4, male sex and diabetes were associated with a further increased risk of EoCRC (aHR [95% CI]:1.149[1.082-1.221] and 1.409[1.169-1.699], respectively). In addition, there was a dose-response relationship between serum TG levels and the risk of EoCRC (p for trends < 0.0001). Conclusion: Persistent HTG increased the risk of EoCRC, which was significantly higher only for rectal cancer and marginally higher for other colonic subsites.

Association between Helicobacter pylori Infection and Eosinophilic Esophagitis.

Background/Aims: This study examined the association between eosinophilic esophagitis (EoE) and Helicobacter pylori (H. pylori) infection. Methods: A single tertiary referral center case-control study was performed. EoE patients diagnosed at Seoul National University Bundang Hospital from July 2003 to March 2022 were reviewed retrospectively. Forty-five EoE patients were included in the analysis. For each EoE patient, two age and sex-matched normal controls were selected randomly from an outpatient population who received upper gastrointestinal endoscopy. Results: Although 17 out of 90 (18.9%) controls had a H. pylori infection, only two out of 45 (4.4%) EoE patients showed evidence of a H. pylori infection. EoE was inversely associated with a H. pylori infection (odds ratio 0.20, 95% confidence interval 0.04-0.91, p=0.044). Conclusions: An inverse association was observed between H. pylori infection and EoE. Further prospective studies will be needed to validate the protective effects of H. pylori infection for EoE.

Development and Validation of Models to Predict Lymph Node Metastasis in Early Gastric Cancer Using Logistic Regression and Gradient Boosting Machine Methods.

PURPOSE: To identify important features of lymph node metastasis (LNM) and develop a prediction model for early gastric cancer (EGC) using a gradient boosting machine (GBM) method. MATERIALS AND METHODS: The clinicopathologic data of 2556 patients with EGC who underwent gastrectomy were used as training set and the internal validation set (set 1) at a ratio of 8:2. Additionally, 548 patients with EGC who underwent endoscopic submucosal dissection (ESD) as the initial treatment were included in the external validation set (set 2). The GBM model was constructed, and its performance was compared with that of the Japanese guidelines. RESULTS: LNM was identified in 12.6% (321/2556) of the gastrectomy group (training set & set 1) and 4.3% (24/548) of the ESD group (set 2). In the GBM analysis, the top five features that most affected LNM were lymphovascular invasion, depth, differentiation, size, and location. The accuracy, sensitivity, specificity, and the area under the receiver operating characteristics of set 1 were 0.566, 0.922, 0.516, and 0.867, while those of set 2 were 0.810, 0.958, 0.803, and 0.944, respectively. When the sensitivity of GBM was adjusted to that of Japanese guidelines (beyond the expanded criteria in set 1 [0.922] and eCuraC-2 in set 2 [0.958]), the specificities of GBM in sets 1 and 2 were 0.516 (95% confidence interval, 0.502-0.523) and 0.803 (0.795-0.805), while those of the Japanese guidelines were 0.502 (0.488-0.509) and 0.788 (0.780-0.790), respectively. CONCLUSION: The GBM model showed good performance comparable with the eCura system in predicting LNM risk in EGCs.