Consensus statements on the current landscape of artificial intelligence applications in endoscopy, addressing roadblocks, and advancing artificial intelligence in gastroenterology.

BACKGROUND AND AIMS: The American Society for Gastrointestinal Endoscopy (ASGE) AI Task Force along with experts in endoscopy, technology space, regulatory authorities, and other medical subspecialties initiated a consensus process that analyzed the current literature, highlighted potential areas, and outlined the necessary research in artificial intelligence (AI) to allow a clearer understanding of AI as it pertains to endoscopy currently. METHODS: A modified Delphi process was used to develop these consensus statements. RESULTS: Statement 1: Current advances in AI allow for the development of AI-based algorithms that can be applied to endoscopy to augment endoscopist performance in detection and characterization of endoscopic lesions. Statement 2: Computer vision-based algorithms provide opportunities to redefine quality metrics in endoscopy using AI, which can be standardized and can reduce subjectivity in reporting quality metrics. Natural language processing-based algorithms can help with the data abstraction needed for reporting current quality metrics in GI endoscopy effortlessly. Statement 3: AI technologies can support smart endoscopy suites, which may help optimize workflows in the endoscopy suite, including automated documentation. Statement 4: Using AI and machine learning helps in predictive modeling, diagnosis, and prognostication. High-quality data with multidimensionality are needed for risk prediction, prognostication of specific clinical conditions, and their outcomes when using machine learning methods. Statement 5: Big data and cloud-based tools can help advance clinical research in gastroenterology. Multimodal data are key to understanding the maximal extent of the disease state and unlocking treatment options. Statement 6: Understanding how to evaluate AI algorithms in the gastroenterology literature and clinical trials is important for gastroenterologists, trainees, and researchers, and hence education efforts by GI societies are needed. Statement 7: Several challenges regarding integrating AI solutions into the clinical practice of endoscopy exist, including understanding the role of human-AI interaction. Transparency, interpretability, and explainability of AI algorithms play a key role in their clinical adoption in GI endoscopy. Developing appropriate AI governance, data procurement, and tools needed for the AI lifecycle are critical for the successful implementation of AI into clinical practice. Statement 8: For payment of AI in endoscopy, a thorough evaluation of the potential value proposition for AI systems may help guide purchasing decisions in endoscopy. Reliable cost-effectiveness studies to guide reimbursement are needed. Statement 9: Relevant clinical outcomes and performance metrics for AI in gastroenterology are currently not well defined. To improve the quality and interpretability of research in the field, steps need to be taken to define these evidence standards. Statement 10: A balanced view of AI technologies and active collaboration between the medical technology industry, computer scientists, gastroenterologists, and researchers are critical for the meaningful advancement of AI in gastroenterology. CONCLUSIONS: The consensus process led by the ASGE AI Task Force and experts from various disciplines has shed light on the potential of AI in endoscopy and gastroenterology. AI-based algorithms have shown promise in augmenting endoscopist performance, redefining quality metrics, optimizing workflows, and aiding in predictive modeling and diagnosis. However, challenges remain in evaluating AI algorithms, ensuring transparency and interpretability, addressing governance and data procurement, determining payment models, defining relevant clinical outcomes, and fostering collaboration between stakeholders. Addressing these challenges while maintaining a balanced perspective is crucial for the meaningful advancement of AI in gastroenterology.

Incidence and factors associated with stent dysfunction and pancreatitis after gastroduodenal stenting for malignant gastric outlet obstruction.

Background and study aims Endoscopic gastroduodenal stent (GDS) deployment is currently a standard treatment for malignant gastric outlet obstruction (mGOO) in patients with limited life expectancy; however, stent dysfunction (SD) and complicated pancreatitis often occur after GDS deployment. We investigated incidence and contributing factors of SD and complicated pancreatitis. Patients and methods We retrospectively reviewed 203 patients who underwent initial GDS deployment for palliation of mGOO symptoms between October 2017 and July 2022, including 109 who underwent GDS deployment across the duodenal papilla (sub-cohort). Results SDs, including tumor ingrowth (n = 26), kinking (n = 14), and migration (n = 13), occurred in 68 patients (33.5%). Cumulative SD incidence was 41.1% (95% confidence interval, 32.6-49.4%). SD incidence increased to 0.4%, 0.16%, and 0.06% per day at < 8, 8-16, and>16 weeks, respectively. On multivariate analysis, Niti-S pyloric/duodenal stent deployment (sub-distribution hazard ratio [sHR] 0.26, P = 0.01) and survival length ≥ 90 days (sHR 2.5, P = 0.01) were respectively identified as favorable and risk factors significantly associated with SD. Pancreatitis developed in 14 patients (12.8%) in the sub-cohort, which had significantly higher parenchymal diameter ( P < 0.01) and lower main pancreatic duct (MPD) caliber ( P < 0.01) than the non-pancreatitis cohort. On multivariate analysis, MPD caliber < 3 mm independently predicted pancreatitis (odds ratio 6.8, P = 0.03). Conclusions Deployment of the Niti-S pyloric/duodenal stent, with conformability even for angulated strictures, significantly reduced the incidence of SD. Stent selection, life expectancy, and MPD caliber should be taken into consideration during decision-making for GDS deployment for mGOO.

International Digestive Endoscopy Network consensus on the management of antithrombotic agents in patients undergoing gastrointestinal endoscopy.

Antithrombotic agents, including antiplatelet agents and anticoagulants, are widely used in Korea because of the increasing incidence of cardiocerebrovascular disease and the aging population. The management of patients using antithrombotic agents during endoscopic procedures is an important clinical challenge. The clinical practice guidelines for this issue, developed by the Korean Society of Gastrointestinal Endoscopy, were published in 2020. However, new evidence on the use of dual antiplatelet therapy and direct anticoagulant management has emerged, and revised guidelines have been issued in the United States and Europe. Accordingly, the previous guidelines were revised. Cardiologists were part of the group that developed the guideline, and the recommendations went through a consensus-reaching process among international experts. This guideline presents 14 recommendations made based on the Grading of Recommendations, Assessment, Development, and Evaluation methodology and was reviewed by multidisciplinary experts. These guidelines provide useful information that can assist endoscopists in the management of patients receiving antithrombotic agents who require diagnostic and elective therapeutic endoscopy. It will be revised as necessary to cover changes in technology, evidence, or other aspects of clinical practice.

Proportions and characteristics of interval cancer in annual fecal immunochemical test screening and postcolonoscopy colorectal cancer: Results from a Japanese multicenter prospective study using questionnaires, the C-DETECT study.

OBJECTIVES: There are several types of colorectal cancer (CRC) according to the detection methods and intervals, including interval CRC (iCRC) and postcolonoscopy CRC (PCCRC). We aimed to examine their proportions and characteristics. METHODS: We conducted a multicenter prospective study using questionnaires in Japan ("C-DETECT study"), in which differences in CRC characteristics according to detection methods and intervals were examined from consecutive adult patients. Because the annual fecal immunochemical test (FIT) was used in population-based screening, the annual FIT-iCRC was assessed. RESULTS: In total, 1241 CRC patients (1064 with invasive CRC) were included. Annual FIT-iCRC (a), 3-year PCCRC (b), and CRC detected within 1 year after a positive FIT with noncompliance to colonoscopy (c) accounted for 4.5%, 7.0%, and 3.9% of all CRCs, respectively, and for 3.9%, 5.4%, and 4.3% of invasive CRCs, respectively. The comparison among these (a, b, c) and other CRCs (d) demonstrated differences in the proportions of ≥T2 invasion ([a] 58.9%, [b] 44.8%, [c] 87.5%, [d] 73.0%), metastasis ([a] 33.9%, [b] 21.8%, [c] 54.2%, [d] 43.9%), right-sided CRC ([a] 42.9%, [b] 40.2%, [c] 18.8%, [d] 28.6%), and female sex ([a] 53.6%, [b] 49.4%, [c] 27.1%, [d] 41.6%). In metastatic CRC, (a) and (b) showed a higher proportions of BRAF mutations ([a] [b] 12.0%, [c] [d] 3.1%). CONCLUSIONS: Annual FIT-iCRC and 3-year PCCRC existed in nonnegligible proportions. They were characterized by higher proportions of right-sided tumors, female sex, and BRAF mutations. These findings suggest that annual FIT-iCRC and 3-year PCCRC may have biological features different from those of other CRCs.

Treatment Decision for Locally Resected T1 Colorectal Carcinoma-Verification of the Japanese Guideline Criteria for Additional Surgery Based on Long-Term Clinical Outcomes.

INTRODUCTION: To verify the value of the pathological criteria for additional treatment in locally resected pT1 colorectal carcinoma (CRC) which have been used in the Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines since 2009. METHODS: We enrolled 4,719 patients with pT1 CRC treated at 27 institutions between July 2009 and December 2016 (1,259 patients with local resection alone [group A], 1,508 patients with additional surgery after local resection [group B], and 1,952 patients with surgery alone [group C]). All 5 factors of the JSCCR guidelines (submucosal resection margin, tumor histologic grade, submucosal invasion depth, lymphovascular invasion, and tumor budding) for lymph node metastasis (LNM) had been diagnosed prospectively. RESULTS: Any of the risk factors were present in 3,801 patients. The LNM incidence was 10.3% (95% confidence interval 9.3-11.4) in group B/C patients with risk factors, whereas it was 1.8% (95% confidence interval 0.4-5.2) in those without risk factors ( P < 0.01). In group A, the incidence of recurrence was 3.4% in patients with risk factors, but it was only 0.1% in patients without risk factors ( P < 0.01). The disease-free survival rate of group A patients classified as risk positive was significantly worse than those of groups B and C patients. However, the 5-year disease-free survival rate in group A patients with no risk was 99.2%. DISCUSSION: Our large-scale real-world multicenter study demonstrated the validity of the JSCCR criteria for pT1 CRC after local resection, especially regarding favorable outcomes in patients with low risk of LNM.

Comparison of the efficacy and tolerability of elobixibat plus sodium picosulfate with magnesium citrate and split-dose 2-L polyethylene glycol with ascorbic acid for bowel preparation before outpatient colonoscopy: a study protocol for the multicentre, randomised, controlled E-PLUS trial.

BACKGROUND: Sodium picosulfate (SP)/magnesium citrate (MC) and polyethylene glycol (PEG) plus ascorbic acid are recommended by Western guidelines as laxative solutions for bowel preparation. Clinically, SP/MC has a slower post-dose defaecation response than PEG and is perceived as less cleansing; therefore, it is not currently used for major bowel cancer screening preparation. The standard formulation for bowel preparation is PEG; however, a large dose is required, and it has a distinctive flavour that is considered unpleasant. SP/MC requires a small dose and ensures fluid intake because it is administered in another beverage. Therefore, clinical trials have shown that SP/MC is superior to PEG in terms of acceptability. We aim to compare the novel bowel cleansing method (test group) comprising SP/MC with elobixibat hydrate and the standard bowel cleansing method comprising PEG plus ascorbic acid (standard group) for patients preparing for outpatient colonoscopy. METHODS: This phase III, multicentre, single-blind, noninferiority, randomised, controlled, trial has not yet been completed. Patients aged 40-69 years will be included as participants. Patients with a history of abdominal or pelvic surgery, constipation, inflammatory bowel disease, or severe organ dysfunction will be excluded. The target number of research participants is 540 (standard group, 270 cases; test group, 270 cases). The primary endpoint is the degree of bowel cleansing (Boston Bowel Preparation Scale [BBPS] score ≥ 6). The secondary endpoints are patient acceptability, adverse events, polyp/adenoma detection rate, number of polyps/adenomas detected, degree of bowel cleansing according to the BBPS (BBPS score ≥ 8), degree of bowel cleansing according to the Aronchik scale, and bowel cleansing time. DISCUSSION: This trial aims to develop a "patient-first" colon cleansing regimen without the risk of inadequate bowel preparation by using both elobixibat hydrate and SP/MC. TRIAL REGISTRATION: Japan Registry of Clinical Trials (jRCT; no. s041210067; 9 September 2021; https://jrct.niph.go.jp/ ), protocol version 1.5 (May 1, 2023).

Efficacy of endoscopic submucosal resection with a ligation device for small rectal neuroendocrine tumor: study protocol of a multicenter open-label randomized control trial (BANDIT trial).

BACKGROUND: Endoscopic resection is widely accepted as a local treatment for rectal neuroendocrine tumors sized ≤ 10 mm. However, there is no consensus on the best method for the endoscopic resection of rectal neuroendocrine tumors. As a simplified endoscopic procedure, endoscopic submucosal resection with a ligation device (ESMR-L) indicates a histologically complete resection rate comparable to that of endoscopic submucosal dissection (ESD). We hypothesized that ESMR-L than ESD would be preferred for rectal neuroendocrine tumors. Hence, this trial aimed to verify whether ESMR-L is non-inferior to ESD in terms of histologically complete resection rate. METHODS: This is a prospective, open-label, multicenter, non-inferiority, randomized controlled trial of two parallel groups, conducted at the Shizuoka Cancer Center and 31 other institutions in Japan. Patients with a lesion endoscopically diagnosed as a rectal neuroendocrine tumor ≤ 10 mm are eligible for inclusion. A total of 266 patients will be recruited and randomized to undergo either ESD or ESMR-L. The primary endpoint is the rate of en bloc resection with histologically tumor-free margins (R0 resection). Secondary endpoints include en bloc resection rate, procedure time, adverse events, hospitalization days, total devices and agents cost, adverse event rate between groups with and without resection site closure, outcomes between expert and non-expert endoscopists, and factors associated with R0 resection failure. The sample size is determined based on the assumption that the R0 resection rate will be 95.2% in the ESD group and 95.3% in the ESMR-L group, with a non-inferiority margin of 8%. With a one-sided significance level of 0.05 and a power of 80%, 226 participants are required. Assuming a dropout rate of 15%, 266 patients will be included in this study. DISCUSSION: This is the first multicenter randomized controlled trial comparing ESD and ESMR-L for the R0 resection of rectal neuroendocrine tumors ≤ 10 mm. This will provide valuable information for standardizing endoscopic resection methods for rectal neuroendocrine tumors. TRIAL REGISTRATION: Japan Registry of Clinical Trials, jRCTs042210124. Registered on Jan 6, 2022.

Extensive antibody search with whole spectrum black-box optimization.

In designing functional biological sequences with machine learning, the activity predictor tends to be inaccurate due to shortage of data. Top ranked sequences are thus unlikely to contain effective ones. This paper proposes to take prediction stability into account to provide domain experts with a reasonable list of sequences to choose from. In our approach, multiple prediction models are trained by subsampling the training set and the multi-objective optimization problem, where one objective is the average activity and the other is the standard deviation, is solved. The Pareto front represents a list of sequences with the whole spectrum of activity and stability. Using this method, we designed VHH (Variable domain of Heavy chain of Heavy chain) antibodies based on the dataset obtained from deep mutational screening. To solve multi-objective optimization, we employed our sequence design software MOQA that uses quantum annealing. By applying several selection criteria to 19,778 designed sequences, five sequences were selected for wet-lab validation. One sequence, 16 mutations away from the closest training sequence, was successfully expressed and found to possess desired binding specificity. Our whole spectrum approach provides a balanced way of dealing with the prediction uncertainty, and can possibly be applied to extensive search of functional sequences.

Long-term outcomes after non-curative endoscopic resection for esophageal squamous cell carcinoma followed by additional chemoradiotherapy.

Endoscopic resection (ER) of esophageal squamous cell carcinoma (ESCC) is evaluated pathologically, and additional treatment is recommended for cases resulting in non-curative resection, defined as pMM with lymphovascular invasion (LVI), pSM, or positive vertical margin. This study aimed to assess long-term outcomes and risk factors for recurrence in patients with ESCC treated with non-curative ER followed by additional chemoradiotherapy (CRT). We retrospectively reviewed the clinical courses of patients who underwent non-curative ER followed by additional CRT for ESCCs between August 2007 and December 2017. Recurrence rates and risk factors for recurrence were analyzed. Among 97 patients with non-curative ER, 73 underwent additional CRT. With a median follow-up period of 71 months, recurrences were observed in 10 (14%) of 73 patients, with a median interval of 24.5 (1-59 months). The 3- and 5-year recurrence-free survival were 89 and 85%, respectively, and the 3- and 5-year overall survival rates were 96 and 91%, respectively. Multivariate analysis showed that lymphatic invasion was an independent risk factor for recurrence in patients with non-curative ESCC receiving additional CRT. Among the 10 patients with recurrence, 4, 3, 2, and 1 underwent surgery, chemotherapy, supportive care, and CRT, respectively. Notably, all four patients who underwent surgery survived, regardless of regional and/or distant lymph node metastasis. Lymphatic invasion is an independent risk factor for the recurrence of non-curative ESCCs. Careful follow-up is required for at least 5 years after ER with additional CRT.

Endoscopist-related factors affecting adenoma detection during colonoscopy: Data from the J-SCOUT study.

OBJECTIVES: Colonoscopy withdrawal times are associated with the adenoma detection rate (ADR). However, the relationship between ADR and cecal insertion time has been inadequately characterized. We aimed to evaluate endoscopist-related factors involved in the ADR, including the average individual colonoscopy insertion and withdrawal times. METHODS: This observational study used a colonoscopy database with pathology data from routine clinical practice in Japanese institutions. The odds ratios (OR) of endoscopist-related factors related to ADRs were examined using a generalized linear mixed model. RESULTS: Of the 186,293 colonoscopies performed during the study period, 47,705 colonoscopies by 189 endoscopists in four hospitals were analyzed for ADR. The overall ADR was 38.3% (95% confidence interval [CI] 37.8, 38.7). Compared to endoscopists with mean cecal insertion times of <5 min, the OR of ADR for those with mean cecal insertion times of 5-9, 10-14, and ≥15 min were 0.84 (95% CI 0.71, 0.99), 0.68 (95% CI 0.52, 0.90), and 0.45 (95% CI 0.25, 0.78), respectively. Compared to endoscopists with mean withdrawal times of <6 min, the OR of ADR for those with mean withdrawal times of 6-9, 10-14, and ≥15 min were 1.38 (95% CI 1.03, 1.85), 1.48 (95% CI 1.09, 2.02), and 1.68 (95% CI 1.04, 2.61), respectively. There were no significant differences in ADRs by endoscopist specialty, gender, or the total number of examinations performed. CONCLUSION: Individual mean colonoscopy insertion time was associated with ADR and might be considered as a colonoscopy quality indicator as well as withdrawal time.

Hyperplastic polyp-like adenoma: a subtype of colonic adenoma with a proliferative zone confined to the lower two-thirds of the crypt.

BACKGROUND: The proliferative zone of colonic adenomas is confined to the upper third of the crypt or is scattered along its entire axis. In contrast, there are unusual adenomas with proliferative zones confined to the lower two-thirds of the crypt. We investigated the frequency and endoscopic features of adenomas with lower proliferative zones. METHODS: We retrospectively reviewed consecutive patients who underwent colonoscopies between September 2022 and March 2023 at the Toyoshima Endoscopy Clinic. Colorectal polyps were endoscopically assessed using the Japan Narrow-Band Imaging Expert Team (JNET) classification. All resected polyps were histologically examined, and the proliferative zone locations were assessed in the adenomas. RESULTS: The frequency of adenomas with a lower proliferative zone was 1.8% (44/2420) in adenomas. Among these adenomas, JNET type 1 incidence was 43.2% (19/44), which was significantly higher than that in adenomas with a normal proliferative zone. Adenomas with a lower proliferative zone were diminutive (mean size: 2.5 mm) and prone to develop in the proximal colon. CONCLUSION: Colonic adenomas with proliferative zones confined to the lower two-thirds of the crypt often appear as diminutive, hyperplastic polyps.

An Analysis of Delayed Bleeding in Cases of Colorectal Endoscopic Submucosal Dissection Due to Types of Direct Oral Anticoagulants in Japan.

BACKGROUND & AIMS: Reported rates of delayed bleeding (DB) after endoscopic resection using direct oral anticoagulants (DOACs) are high and heterogeneous. This large-scale multicenter study analyzed cases of DB after colorectal endoscopic submucosal dissection related to various types of DOACs in Japan (the ABCD-J study) with those associated with warfarin. METHODS: We retrospectively reviewed 1019 lesions in patients treated with DOACs and 459 lesions in patients treated with warfarin among 34,455 endoscopic submucosal dissection cases from 47 Japanese institutions between 2012 and 2021. The DB rate (DBR) with each DOAC was compared with that with warfarin. Risk factors for DB in patients treated with DOACs or warfarin were also investigated. RESULTS: The mean tumor sizes in the DOAC and warfarin groups were 29.6 ± 14.0 and 30.3 ± 16.4 mm, respectively. In the DOAC group, the DBR with dabigatran (18.26%) was significantly higher than that with apixaban (10.08%, P = .029), edoxaban (7.73%, P = .001), and rivaroxaban (7.21%, P < .001). Only rivaroxaban showed a significantly lower DBR than warfarin (11.76%, P = .033). In the multivariate analysis, heparin bridging therapy (odds ratio [OR], 2.18; 95% confidence interval [CI], 1.27-3.73, P = .005), rectal location (2.01, 1.28-3.16, P = .002), and procedure time ≥55 minutes (2.43, 1.49-3.95, P < .001) were significant risk factors for DB in the DOAC group. The DB risk in the DOAC group (OR, (95% CI)) was 2.13 (1.30-3.50) and 4.53 (2.52-8.15) for 1 and 2 significant risk factors, respectively. CONCLUSIONS: Dabigatran was associated with a higher DBR than other DOACs, and only rivaroxaban was associated with a significantly lower DBR than warfarin.

Endoscopic Removal of Premalignant Lesions Reduces Long-Term Colorectal Cancer Risk: Results From the Japan Polyp Study.

BACKGROUND & AIMS: To date, no regional evidence of long-term colorectal cancer (CRC) risk reduction after endoscopic premalignant lesion removal has been established. We aimed to analyze this over a long-term follow-up evaluation. METHODS: This was a prospective cohort study of participants from the Japan Polyp Study conducted at 11 Japanese institutions. Participants underwent scheduled follow-up colonoscopies after a 2-round baseline colonoscopy process. The primary outcome was CRC incidence after randomization. The observed/expected ratio of CRC was calculated using data from the population-based Osaka Cancer Registry. Secondary outcomes were the incidence and characteristics of advanced neoplasia (AN). RESULTS: A total of 1895 participants were analyzed. The mean number of follow-up colonoscopies and the median follow-up period were 2.8 years (range, 1-15 y) and 6.1 years (range, 0.8-11.9 y; 11,559.5 person-years), respectively. Overall, 4 patients (all males) developed CRCs during the study period. The observed/expected ratios for CRC in all participants, males, and females, were as follows: 0.14 (86% reduction), 0.18, and 0, respectively, and 77 ANs were detected in 71 patients (6.1 per 1000 person-years). Of the 77 ANs detected, 31 lesions (40.3%) were laterally spreading tumors, nongranular type. Nonpolypoid colorectal neoplasms (NP-CRNs), including flat (<10 mm), depressed, and laterally spreading, accounted for 59.7% of all detected ANs. Furthermore, 2 of the 4 CRCs corresponded to T1 NP-CRNs. CONCLUSIONS: Endoscopic removal of premalignant lesions, including NP-CRNs, effectively reduced CRC risk. More than half of metachronous ANs removed by surveillance colonoscopy were NP-CRNs. The Japan Polyp Study: University Hospital Medical Information Network Clinical Trial Registry: University Hospital Medical Information Network Clinical Trial Registry, C000000058; cohort study: UMIN000040731.

Short-term and long-term outcomes of submucosal dissection for residual or recurrent colorectal tumors after endoscopic resection: Analysis of a multicenter prospective study.

OBJECTIVES: We previously demonstrated that a favorable long-term prognosis indicated that endoscopic submucosal dissection (ESD) could be the standard treatment for large colorectal epithelial neoplasms, but the usefulness of ESD for local residual or recurrent tumors with submucosal fibrosis has not been fully demonstrated. The aim of the present study was to assess the usefulness of ESD for local residual or recurrent colorectal tumors. METHODS: We conducted a nationwide multicenter prospective study to evaluate the outcomes of ESD for colorectal tumors. In this post hoc analysis, a total of 54 local residual or recurrent colorectal tumors in 54 patients were included, and we analyzed the short-term and long-term outcomes of ESD for these lesions. RESULTS: The median size of the lesions was 16.0 (interquartile range [IQR] 11-25) mm. ESD was completed in 53 cases (98.1%) with a median procedure time of 65.0 min, but it was discontinued in one case because of submucosal cancer invasion. En bloc resection was achieved in 52 cases (96.3%), whereas R0 resection was achieved in 45 cases (83.3%). Intraoperative perforation was observed in four cases (7.4%) and delayed perforation in one (1.9%), but all cases could be managed conservatively. Delayed bleeding was not observed. There were no significant differences in short-term outcomes between the rectal and colonic lesions. There was no recurrence of the tumor during the median follow-up period of 60 (IQR 50-64) months. CONCLUSION: An analysis of our multicenter prospective study suggests that ESD is an effective salvage management for local residual or recurrent colorectal lesions.

A discrimination model by machine learning to avoid gastrectomy for early gastric cancer.

Aim: Gastrectomy is recommended for patients with early gastric cancer (EGC) because the possibility of lymph node metastasis (LNM) cannot be completely denied. The aim of this study was to develop a discrimination model to select patients who do not require surgery using machine learning. Methods: Data from 382 patients who received gastrectomy for gastric cancer and who were diagnosed with pT1b were extracted for developing a discrimination model. For the validation of this discrimination model, data from 140 consecutive patients who underwent endoscopic resection followed by gastrectomy, with a diagnosis of pT1b EGC, were extracted. We applied XGBoost to develop a discrimination model for clinical and pathological variables. The performance of the discrimination model was evaluated based on the number of cases classified as true negatives for LNM, with no false negatives for LNM allowed. Results: Lymph node metastasis was observed in 95 patients (25%) in the development cohort and 11 patients (8%) in the validation cohort. The discrimination model was developed to identify 27 (7%) patients with no indications for additional surgery due to the prediction of an LNM-negative status with no false negatives. In the validation cohort, 13 (9%) patients were identified as having no indications for additional surgery and no patients with LNM were classified into this group. Conclusion: The discrimination model using XGBoost algorithms could select patients with no risk of LNM from patients with pT1b EGC. This discrimination model was considered promising for clinical decision-making in relation to patients with EGC.