Search

Background and Aims We previously identified a 5 methylated DNA marker (MDM) panel for the detection of nonendoscopic Barrett’s esophagus (BE). In this study, we aimed to recalibrate the performance of the 5 MDM panel using a simplified assay in a training cohort, validate the panel in an independent test cohort, and explore the accuracy of an MDM panel with only 3 markers. Methods Participants were recruited from 3 medical centers. The sponge on a string device (EsophaCap; CapNostics, Concord, NC, USA) was swallowed and withdrawn, followed by endoscopy, in BE cases and control subjects. A 5 MDM panel was blindly assayed using a simplified assay. Random forest modeling analysis was performed, in silico cross-validated in the training set, and then locked down, before test set analysis. Results The training set had 199 patients: 110 BE cases and 89 control subjects, and the test set had 89 patients: 60 BE cases and 29 control subjects. Sensitivity of the 5 MDM panel for BE diagnosis was 93% at 90% specificity in the training set and 93% at 93% specificity in the test set. Areas under the receiver operating characteristic curves were .96 and .97 in the training and test sets, respectively. Model accuracy was not influenced by age, sex, or smoking history. Multiple 3 MDM panels achieved similar accuracy. Conclusions A 5 MDM panel for BE is highly accurate in training and test sets in a blinded multisite case-control analysis using a simplified assay. This panel may be reduced to only 3 MDMs in the future. (Clinical trial registration number: NCT 02560623.)

Background and Aims Locally advanced pancreatic cancer (LAPC) often causes obstruction. Verteporfin photodynamic therapy (PDT) can feasibly “debulk” the tumor more safely than noncurative surgery and has multiple advantages over older PDT agents. We aimed to assess the feasibility of EUS-guided verteporfin PDT in ablating nonresectable LAPC. Methods Adults with LAPC with adequate biliary drainage were prospectively enrolled. Exclusion criteria were significant metastatic disease burden, disease involving >50% duodenal or major artery circumference, and recent treatment with curative intent. CT was obtained between days –28 to 0. On day 0, verteporfin .4 mg/kg was infused 60 to 90 minutes before EUS, during which a diffuser was positioned in the tumor and delivered light at 50 J/cm for 333 seconds. CT was obtained on day 2, with adverse event monitoring occurring on days 1, 2, and 14. The primary outcome was presence of necrosis. Results Of 8 patients (62.5% men, mean age 65 ± 7.9 years) included in the study, 5 were staged at T3, 2 at T2, and 1 at T1. Most (n = 4) had primary lesions in the pancreatic head. Mean pretrial tumor diameter was 33.3 ± 13.4 mm. On day 2 CT, 5 lesions demonstrated a zone of necrosis measuring a mean diameter of 15.7 ± 5.5 mm; 3 cases did not develop necrosis. No adverse events were noted during the procedure or postprocedure observation period (days 1-3), and no changes in patient-reported outcomes were noted. Conclusions In this pilot study, EUS-guided verteporfin PDT is feasible and shows promise as a minimally invasive ablative therapy for LAPC in select patients. Tumor necrosis is visible within 48 hours after treatment. Patient enrollment and data collection are ongoing. (Clinical trial registration number: NCT03033225.)

Endoscopic submucosal dissection (ESD) is used for treating early luminal GI cancers. ESD is the standard of care in Asia, where it results in multiday admissions for observation. The goal of our study was to evaluate the safety and feasibility of same-day discharge (SDD) after ESD.This is a retrospective cohort study of adults who underwent similar esophageal ESD with a Clutch Cutter device (DP2618DT; Fujifilm) at the Mayo Clinic (Rochester, Minn, USA) from 2017 to 2019 with a single endoscopist. The primary end point was postprocedural adverse events within 7 days of ESD.Of 96 patients (75% male, mean age, 70 ± 10.3 years) undergoing a total of 140 ESDs, 85 were SDD versus 55 admissions. Of the 55 admits, 53 were discharged within 24 hours, whereas 2 were admitted for 2 to 3 days for reasons unrelated to the ESD procedure. Admissions were more likely to have a history of antiplatelet/anticoagulant use (56.4% vs 34.1%; P = .01) and higher mean American Society of Anesthesiologists (ASA) score (3.2 vs 2.9; P = .007). Admissions had larger resections (28.6 vs 20.1 mm; P .0001) with longer procedural durations (103.4 vs 62 minutes; P .0001). Among SDDs, no intraprocedural or postprocedural adverse events were seen. Among admissions, 1 (1.8% vs 0%; P = .39) experienced intraprocedural bleeding requiring endoscopic intervention, 1 required transfusion before discharge within 24 hours of ESD (1.8% vs 0%; P = .39), and 1 required rehospitalization and endoscopic intervention within 7 days to address an active bleed along the resection margin (1.8% vs 0%; P = .39).SDD after esophageal ESD is safe and feasible. An experienced endoscopist can determine if SDD can be considered in patients with ASA physical classification status ≤2 who undergo resections off antiplatelet/anticoagulant therapy and do not experience intraprocedural adverse events.

Background Volumetric laser endomicroscopy (VLE) is an advanced imaging modality used in Barrett's esophagus (BE) surveillance. VLE interpretation relies on identification of specific features associated with BE dysplasia and can be enhanced by intelligent real-time image segmentation (IRIS). The diagnostic performance of established VLE interpretation systems has not been validated. This study uses a dataset of histology-matched regions of interest (ROI) to determine the diagnostic performance of VLE with and without IRIS enhancement. Methods VLE scans with ROIs laser-marked for targeted biopsy and matching histopathology were performed in patients undergoing BE surveillance. ROIs were rated independently for BE dysplasia by 3 blinded VLE experts using established VLE interpretation systems. Following a washout period, ROIs were rated using IRIS adapted to the VLE interpretation systems. Inter-rater agreement, diagnostic performance, and optimal diagnostic thresholds were calculated for each scoring system. Results A total of 71 laser-marked ROIs from 44 patients (mean age: 68 years) and BE length of 5 cm were included. About 45.1% of ROIs contained dysplasia. Inter-rater agreement was uniformly very good or excellent and only the surface intensity scoring agreement was enhanced by IRIS. The scoring systems’ AUC ranged from 0.59 to 0.74 and 0.60 to 0.75 for VLE with and without IRIS, respectively. When diagnostic thresholds of the existing scoring systems were optimized, the adjusted AUC ranged from 0.70 to 0.78. Conclusions This is the first study to validate and optimize the diagnostic performance of established VLE interpretation systems for BE dysplasia. IRIS image enhancement adapted to these interpretation systems demonstrated similar diagnostic performance.

Background and Aims Treatment of large esophageal neoplasia is gradually evolving from piecemeal to en bloc resections. Endoscopic submucosal dissection (ESD) is known to achieve more complete resections than piecemeal EMR for large lesions, yet it remains underused in the West because of technical and safety concerns with traditional electrosurgical knives. We aimed to evaluate a novel endoscopic articulating knife used with ESD (ESD-AR) to determine its safety and efficacy for large esophageal neoplasms in comparison with EMR. Methods We retrospectively studied clinically indicated cases of ESD-AR and EMR for esophageal lesions that were 15 mm or greater. All EMR cases had at least 3 simultaneous EMRs to adequately compare resection area. Rates of perforation, GI bleeding, technical performance, and pre- and postendoscopic resection diagnoses were evaluated. Results Seventy-two ESD-AR and 72 widespread EMR cases were evaluated for Barrett’s esophagus (56%), adenocarcinoma (36%), squamous nodularity (2%), and squamous cell carcinoma (6%). There were no statistical differences in age, sex, Barrett’s esophagus length, and lesion or resection size between the 2 groups. No perforations occurred. Two adverse events were recorded with ESD-AR and none with EMR (3% vs 0%, P = .50); these were associated with anticoagulation use (P = .04) and greater resection area (P = .02). There were more upgraded diagnoses post-ESD versus EMR (27% vs 12%, P = .05). Conclusions ESD-AR by an experienced endoscopist has a comparable safety profile with widespread EMR for large esophageal neoplasia and may have advantages for diagnostic staging.

BACKGROUND & AIMS: Recommended surveillance intervals after complete eradication of intestinal metaplasia (CE-IM) after endoscopic eradication therapy (EET) are largely not evidence-based. Using recurrence rates in a multicenter international Barrett’s esophagus (BE) CE-IM cohort, we aimed to generate optimal intervals for surveillance. METHODS: Patients with dysplastic BE undergoing EET and achieving CE-IM from prospectively maintained databases at 5 tertiary-care centers in the United States and the United Kingdom were included. The cumulative incidence of recurrence was estimated, accounting for the unknown date of actual recurrence that lies between the dates of current and previous endoscopy. This cumulative incidence of recurrence subsequently was used to estimate the proportion of patients with undetected recurrence for various surveillance intervals over 5 years. Intervals were selected that minimized recurrences remaining undetected for more than 6 months. Actual patterns of post-CE-lM follow-up evaluation are described. RESULTS: A total of 498 patients (with baseline low-grade dysplasia, 115 patients; high-grade dysplasia [HGD], 288 patients; and intramucosal adenocarcinoma [IMCa], 95 patients) were included. Any recurrence occurred in 27.1% and dysplastic recurrence occurred in 8.4% over a median of 2.6 years of follow-up evaluation. For pre-ablation HGD/IMCa, intervals of 6, 12, 18, and 24 months, and then annually, resulted in no patients with dysplastic recurrence undetected for more than 6 months, comparable with current guideline recommendations despite a 33% reduction in the number of surveillance endoscopies. For pre-ablation low-grade dysplasia, intervals of 1, 2, and 4 years balanced endoscopic burden and undetected recurrence risk. CONCLUSIONS: Lengthening post-CE-IM surveillance intervals would reduce the endoscopic burden after CE-IM with comparable rates of recurrent HGD/IMCa. Future guidelines should consider reduced surveillance frequency.

Background Pre-operative staging of pancreatic adenocarcinoma guides clinical decision making. Limited data indicate that metastasis to celiac ganglia (CG) correlates with poor prognosis. We investigated feasibility and safety of endoscopic ultrasound fine needle aspiration (EUS-FNA) detection of CG metastasis and its impact upon tumor stage, resectability, and survival in pancreatic ductal adenocarcinoma (PDAC). Patients We reviewed our prospectively maintained EUS and cytopathology databases to identify patients with FNA proven CG metastasis in patients with PDAC from 2004 to 2017. Clinical demographics, EUS, CT, MRI, cytopathology, cancer stage, and resectability data were analyzed. Survival of PDAC patients with CG metastasis was compared to the expected survival of PDAC patients of similar stage as reported by the United States National Cancer Database. Results Twenty-one patients with PDAC [median age 73 (IQR63-78); 14 (67%) female)], had CG metastasis confirmed by cytopathologic assessment. CG metastasis resulted in tumor upstaging relative to other EUS findings and cross sectional imaging findings in 12 (57%) and 15 (71%) patients, and converted cancers from resectable to unresectable relative to EUS and cross sectional imaging in 7 (37%) and 7 (37%) patients, respectively. In patients with PDAC, the survival of patients with CG metastasis was not significantly different from the overall survival (hazard ratio 0.71; 95% confidence interval 0.44, 1.13; p = 0.15). Conclusions EUS-FNA may safely identify CG metastases. While CG metastasis upstaged and altered the resectability status among this cohort of patients with PDAC, the survival data with regard to PDAC suggest that this may be misguided.

Although esophageal adenocarcinoma is an uncommon cancer, it is often fatal unless found at an early disease stage. Screening to discover early stage disease for its only known precursor, Barrett esophagus, is currently the only solution to improve survival. Multiple strategies have been suggested for screening including identification of a higher risk population using epidemiological factors such as sex, age, symptoms of reflux, and presence of central obesity. In addition, new nonendoscopic methods of screening have been created including sponges, balloons, breath tests, and tests on various bodily fluids. All of these new developments were created in order to improve the efficacy and cost-effectiveness of screening. New technologies may allow for a low-cost, noninvasive screening technology that is highly acceptable to healthy patients at risk for esophageal cancer.

Background and Aims Seattle protocol forceps biopsies (FB) are currently recommended for surveillance in Barrett’s esophagus (BE) but limited by sampling error and lack of adherence. Wide-area transepithelial sampling with 3D analysis (WATS-3D) is reported to increase BE dysplasia detection. We assessed the incremental yield and clinical significance of WATS-3D for dysplasia detection over FB in a systematic review and meta-analysis. Methods We queried major scientific databases for studies using WATS-3D and FB from 2000-2020. Primary outcome was the incremental yield of WATS-3D detected dysplasia (defined as a composite of indefinite for dysplasia [IND], low- and high-grade dysplasia [LGD, HGD] and esophageal adenocarcinoma [EAC]) over FB. Secondary outcomes were incremental yields of HGD/EAC, and rate of reconfirmation of WATS-3D dysplasia on subsequent FB. Results Meta-analysis of 7 eligible studies demonstrated that FB diagnosed dysplasia in 15.9% of cases, whereas incremental yield with WATS-3D was 7.2% (95% CI, 3.9%-11.5%; I2=92.1%). Meta-analysis of 6 studies demonstrated that FB diagnosed HGD/EAC in 2.3% of patients, whereas incremental yield with WATS-3D was 2.1% (95% CI, 0.4%-5.3%; I2=92.7%). Notably, WATS-3D was negative in 62.5% of cases where FB identified dysplasia. Two studies reported reconfirmation of WATS-3D dysplasia with FB histology in only 20 patients. Conclusion WATS-3D increases dysplasia detection, however, the clinical significance of this increased dysplasia detection remains uncertain. Data from endoscopic follow up to ascertain FB histology in patients with dysplasia based solely on WATS-3D are needed to determine the optimal clinical application and significance of WATS only dysplasia.

BACKGROUND AND AIMS: Endoscopic resection (ER) is an important component of the endoscopic treatment of Barrett’s Esophagus (BE) with dysplasia and intramucosal adenocarcinoma (IMC). ER can be performed by cap assisted endoscopic mucosal resection (cEMR) or endoscopic submucosal dissection (ESD). We compared the histological outcomes of ESD versus cEMR, followed by ablation. METHODS: We queried a prospectively maintained database of all patients undergoing cEMR and ESD followed by ablation at our institution from January 2006 to March 2020 and abstracted relevant demographic and clinical data. Our primary outcomes included the rate of complete remission of dysplasia (CRD): absence of dysplasia on surveillance histology and complete remission of intestinal metaplasia (CRIM): absence of intestinal metaplasia. Our secondary outcome included complication rates. RESULTS: We included 537 patients in the study: 456 who underwent cEMR and 81 who underwent ESD. The cumulative probabilities of CRD at 2 years were 75.8% and 85.6% in the cEMR and ESD groups (p< 0.01). Independent predictors of CRD were: ESD (HR: 2.38; p0.05). The only independent predictor of CRIM was a shorter BE segment (HR: 1.16; p < 0.01). CONCLUSIONS: BE patients with dysplasia or IMC undergoing ESD reach CRD at higher rates than those treated with cEMR, though CRIM rates at two years and complication rates were similar between the two groups.

Background and Aims Wide-area transepithelial sampling (WATS) with computer-assisted 3-dimensional analysis is a sampling technique that combines abrasive brushing of the Barrett's esophagus (BE) mucosa followed by neural network analysis to highlight abnormal-appearing cells. Methods We performed a randomized trial of referred BE patients undergoing surveillance at 16 medical centers. Subjects received either biopsy sampling followed by WATS or WATS followed by biopsy sampling. The primary outcome was rate of detection of high-grade dysplasia/esophageal adenocarcinoma (HGD/EAC) using WATS in conjunction with biopsy sampling compared with biopsy sampling alone using standard histopathologic criteria. Secondary aims included evaluating neoplasia detection rates based on the procedure order (WATS vs biopsy sampling first), of each procedure separately, and the additional time required for WATS. Results One hundred sixty patients (mean age, 63.4 years; 76% men; 95% white) completed the trial. The median circumferential and maximal BE extents were 1.0 cm (interquartile range: .0-5.0) and 4.0 cm (interquartile range, 2.0-8.0), respectively. The diagnostic yield for biopsy sampling alone was as follows: HGD/EAC, 7 (4.4%); low-grade dysplasia (LGD), 28 (17.5%); nondysplastic BE (NDBE), 106 (66.25%); and no BE, 19 (11.9%). The addition of WATS to biopsy sampling yielded an additional 23 cases of HGD/EAC (absolute increase, 14.4%; 95% confidence interval, 7.5%-21.2%). Among these 23 patients, 11 were classified by biopsy sampling as NDBE and 12 as LGD/indefinite for dysplasia (IND); 14 received biopsy sampling first and 9 WATS first (not significant) and most (n = 21; 91.7%) had a prior dysplasia history. WATS added an average of 4.5 minutes to the procedure. Conclusion Results of this multicenter, prospective, randomized trial demonstrate that the use of WATS in a referral BE population increases the detection of HGD/EAC. (Clinical trial registration number: NCT03008980.)

In this narrative review, invited by the Editors of Gastroenterology, we summarize recent advances in the field of gastrointestinal endoscopy. We have chosen articles published primarily in the past 2-3 years. Although a thorough literature review was performed for each topic, the nature of the article is subjective and systematic and is based on the authors' experience and expertise regarding articles we believed were most likely to be of high clinical and scientific importance.

Background and Aims Malignant vascular invasion usually results from gross direct infiltration from a primary tumor and impacts cancer staging, prognosis, and therapy. However, patients may also develop a remote malignant thrombi (RMT), defined as a malignant intravascular thrombus located remote and noncontiguous to the primary tumor. Our aim was to compare EUS, CT, and magnetic resonance imaging (MRI) findings of RMT and to explore the potential impact on cancer staging. Methods Patients with RMT were identified from a prospectively maintained EUS database. Retrospective chart review was performed to obtain EUS, CT/MRI, clinical, and outcome data. Results A median of 3 FNAs (range, 1-8) was obtained from RMT in 17 patients (60 ± 14.1 years, 56% men) between April 2003 and August 2016, with the finding of malignant cytology in 12 patients (70.6%; 10 positive, 2 suspicious). CT/MRI detected the RMT in 5 patients (29.4%), 4 of whom had positive or suspicious EUS-FNA cytology. Among the 8 newly diagnosed pancreatic adenocarcinoma (PaC) patients, CT did not detect the RMT in 5 (63%), of whom 3 patients had positive or suspicious intravascular EUS-FNA cytology. For newly diagnosed PaC patients (n = 8), the EUS-FNA diagnosis of a biopsy specimen–proven RMT upstaged 3 patients (37.5%) and converted 2 patients (25%) from CT resectable to unresectable disease. No adverse events were reported. The mean follow-up was 18.9 ± 27.7 months. Conclusions Our study demonstrates the ability and potential safety of intravascular FNA to detect radiographically occult RMT, which substantially impacts cancer staging and resectability.

In the last decade, radiofrequency ablation in combination with endoscopic mucosal resection has simplified and improved the treatment of Barrett's esophagus. These treatments not only reduced the progression of dysplastic Barrett's esophagus to esophageal adenocarcinoma but also decreased treatment-related complications. More recent data from larger series with extended follow-up periods are emerging to refine expectations in patients treated with radiofrequency ablation. Although most patients achieve eradication of neoplasia and intestinal metaplasia, in the long-term a substantial portion of patients develop recurrent disease. This article provides an updated review of radiofrequency ablation efficacy, complications, and durability.

Gastric adenocarcinoma, esophageal adenocarcinoma, and esophageal squamous cell carcinoma are among the most prevalent and deadly of malignancies worldwide. Screening and prevention programs will be critical to finally improving outcomes in these diseases. For gastric adenocarcinoma, screening in high-risk populations has significantly reduced mortality. More research is needed on screening high-risk individuals in low-risk populations. For esophageal adenocarcinoma, work is needed to develop efficient and effective techniques in mass screening programs. For most Western populations, current screening is not cost effective. Avoiding environmental risk factors is critical to reducing the incidence of this deadly illness.

Summary Background Treatment of dysplastic Barrett's oesophagus prevents progression to adenocarcinoma; however, the optimal diagnostic strategy for Barrett's oesophagus is unclear. The Cytosponge-trefoil factor 3 (TFF3) is a non-endoscopic test for Barrett's oesophagus. The aim of this study was to investigate whether offering this test to patients on medication for gastro-oesophageal reflux would increase the detection of Barrett's oesophagus compared with standard management. Methods This multicentre, pragmatic, randomised controlled trial was done in 109 socio-demographically diverse general practice clinics in England. Randomisation was done both at the general practice clinic level (cluster randomisation) and at the individual patient level, and the results for each type of randomisation were analysed separately before being combined. Patients were eligible if they were aged 50 years or older, had been taking acid-suppressants for symptoms of gastro-oesophageal reflux for more than 6 months, and had not undergone an endoscopy procedure within the past 5 years. General practice clinics were selected by the local clinical research network and invited to participate in the trial. For cluster randomisation, clinics were randomly assigned (1:1) by the trial statistician using a computer-generated randomisation sequence; for individual patient-level randomisation, patients were randomly assigned (1:1) by the general practice clinics using a centrally prepared computer-generated randomisation sequence. After randomisation, participants received either standard management of gastro-oesophageal reflux (usual care group), in which participants only received an endoscopy if required by their general practitioner, or usual care plus an offer of the Cytosponge-TFF3 procedure, with a subsequent endoscopy if the procedure identified TFF3-positive cells (intervention group). The primary outcome was the diagnosis of Barrett's oesophagus at 12 months after enrolment, expressed as a rate per 1000 person-years, in all participants in the intervention group (regardless of whether they had accepted the offer of the Cytosponge-TFF3 procedure) compared with all participants in the usual care group. Analyses were intention-to-treat. The trial is registered with the ISRCTN registry, ISRCTN68382401, and is completed. Findings Between March 20, 2017, and March 21, 2019, 113 general practice clinics were enrolled, but four clinics dropped out shortly after randomisation. Using an automated search of the electronic prescribing records of the remaining 109 clinics, we identified 13 657 eligible patients who were sent an introductory letter with 14 days to opt out. 13 514 of these patients were randomly assigned (per practice or at the individual patient level) to the usual care group (n=6531) or the intervention group (n=6983). Following randomisation, 149 (2%) of 6983 participants in the intervention group and 143 (2%) of 6531 participants in the usual care group, on further scrutiny, did not meet all eligibility criteria or withdrew from the study. Of the remaining 6834 participants in the intervention group, 2679 (39%) expressed an interest in undergoing the Cytosponge-TFF3 procedure. Of these, 1750 (65%) met all of the eligibility criteria on telephone screening and underwent the procedure. Most of these participants (1654 [95%]; median age 69 years) swallowed the Cytosponge successfully and produced a sample. 231 (3%) of 6834 participants had a positive Cytosponge-TFF3 result and were referred for an endoscopy. Patients who declined the offer of the Cytosponge-TFF3 procedure and all participants in the usual care group only had an endoscopy if deemed necessary by their general practitioner. During an average of 12 months of follow-up, 140 (2%) of 6834 participants in the intervention group and 13 (

Background and Aims After endoscopic eradication of Barrett’s esophagus (BE), recurrence of intestinal metaplasia at the gastroesophageal junction (GEJIM) is common. The clinical significance of this finding is unclear. We assessed whether recurrent GEJIM is associated with increased risk of subsequent dysplasia and whether endoscopic treatment lowers this risk. Methods A retrospective, multicenter, cohort study was performed of treated BE patients who achieved complete eradication of intestinal metaplasia (IM). Postablation follow-up was performed at standard intervals. Recurrent GEJIM was defined as nondysplastic IM on gastroesophageal junction biopsy specimens without endoscopic evidence of BE. Patients were categorized as “never-GEJIM,” “GEJIM-observed,” or “GEJIM-treated.” Endoscopic treatment for recurrent GEJIM was at the endoscopists’ discretion. The primary outcome was dysplasia recurrence. Analyses were performed using log-rank tests and Cox proportional hazards modeling. Results Six hundred thirty-three patients were analyzed; median follow-up was 47 months (interquartile range, 24-69). Most patients (81%) had high-grade dysplasia or intramucosal adenocarcinoma before treatment. Dysplasia recurrence was 2.2% per year. GEJIM-observed patients had the lowest rate of recurrence (.6%/y) followed by GEJIM-treated (2.2%/y) and never-GEJIM (2.6%/y) (log-rank P = .07). In multivariate analyses, compared with never-GEJIM, the risk of dysplasia recurrence was significantly lower in GEJIM-observed patients (adjusted hazard ratio, .19; 95% confidence interval, .05-.81) and not different in GEJIM-treated patients (adjusted hazard ratio, .81; 95% confidence interval, .39-1.67). Older age and longer initial BE length were independently associated with recurrence. Conclusions Recurrent GEJIM after endoscopic eradication of BE was not associated with an increased risk of subsequent dysplasia. Future studies are warranted to determine if observation is appropriate for this finding.

Background and Aims Detection and characterization of focal liver lesions (FLLs) is key for optimizing treatment for patients who may have a primary hepatic cancer or metastatic disease to the liver. This is the first study to develop an EUS-based convolutional neural network (CNN) model for the purpose of identifying and classifying FLLs. Methods A prospective EUS database comprising cases of FLLs visualized and sampled via EUS was reviewed. Relevant still images and videos of liver parenchyma and FLLs were extracted. Patient data were then randomly distributed for the purpose of CNN model training and testing. Once a final model was created, occlusion heatmap analysis was performed to assess the ability of the EUS-CNN model to autonomously identify FLLs. The performance of the EUS-CNN for differentiating benign and malignant FLLs was also analyzed. Results A total of 210,685 unique EUS images from 256 patients were used to train, validate, and test the CNN model. Occlusion heatmap analyses demonstrated that the EUS-CNN model was successful in autonomously locating FLLs in 92.0% of EUS video assets. When evaluating any random still image extracted from videos or physician-captured images, the CNN model was 90% sensitive and 71% specific (area under the receiver operating characteristic [AUROC], 0.861) for classifying malignant FLLs. When evaluating full-length video assets, the EUS-CNN model was 100% sensitive and 80% specific (AUROC, 0.904) for classifying malignant FLLs. Conclusions This study demonstrated the capability of an EUS-CNN model to autonomously identify FLLs and to accurately classify them as either malignant or benign lesions.

Background & Aims It is a challenge to detect dysplasia in Barrett’s esophagus (BE) and esophageal adenocarcinomas (EACs) are missed in 25%–33% of cases. The neoplasia detection rate (NDR), defined as the rate of high-grade dysplasia (HGD) or EAC detection during initial surveillance endoscopy, has been proposed as a quality metric for endoscopic evaluation of patients with BE. However, current estimates are from referral center cohorts, which might overestimate NDR. Effects on rates of missed dysplasia are also unknown. We analyzed data from a large cohort of patients with BE to estimate the NDR and factors associated with it, and assess the effects of the NDR on the rate of missed dysplasia. Methods We analyzed data from 1066 patients the Rochester Epidemiology Project-linked medical record system, a population-based cohort of patients with BE (confirmed by review of the endoscopic and histologic reports) from 11 southeastern Minnesota counties from 1991 through 2019. Biopsies reported to contain dysplasia were confirmed by expert gastrointestinal pathologists. The NDR was calculated as the rate of HGD or EAC detected by histologic analyses of biopsies collected during the first surveillance endoscopy. Patients without HGD or EAC at their initial endoscopy (n=391) underwent repeat endoscopy within 12 months; HGD or EAC detected at the repeat endoscopy were considered to be missed on index endoscopy. Factors associated with NDR and missed dysplasia were identified using univariate and multivariate logistic regression models. Results The NDR was 4.9% (95% CI, 3.8–6.4); 3.1% of patients had HGD, 1.8% had EAC, and 10.6% had low-grade dysplasia. Factors associated with higher rates of detection of neoplasia included older age, male sex, smoking, increasing length of BE, and surveillance endoscopies by gastroenterologists. This NDR was associated with a substantially lower rate of missed dysplasia (13%). Conclusions In an analysis of 1066 patients with BE in a population-based cohort, we found a lower NDR and lower rate of missed dysplasia than previously reported. NDR may have value as a quality metric in BE surveillance if validated in other cohorts.

Background and Aims The treatment of submucosal (T1b) esophageal adenocarcinoma (EAC) remains in evolution, with some evidence supporting endoscopic management of low-risk lesions. Using a multicenter cohort, we evaluated outcomes of patients with T1b EAC and predictors of survival. Methods Patients diagnosed between 2001 and 2016 with T1b EAC were identified from 3 academic medical centers in the United States. Demographic, clinical, and outcome data were collected. Outcomes studied were overall and cancer-free survival. Cox proportional hazards models were constructed to assess independent predictors of survival. Results One hundred forty-one patients were included, of whom 68 (48%) underwent esophagectomy and 73 (52%) were treated endoscopically. Most patients (85.8%) had high-risk histologic features. Thirty-day operative mortality was 2.9%. Median follow-up in the esophagectomy and endoscopic cohorts was 49.4 and 43.4 months, respectively. Patients treated endoscopically were older with higher comorbidity scores, with 46 (63%) achieving histologic remission. Nineteen patients (26.0%) also received chemoradiation. Five-year overall survival rates in the surgical and endoscopic cohorts were 89% and 59%, respectively, whereas 5-year cancer-free survival rates were 92% and 69%. Presence of high-risk histologic features was associated with reduced overall survival. Conclusions In this large multicenter study of patients with T1b EAC, esophagectomy was associated with improved overall but not cancer-free survival. High-risk histologic features were associated with poorer survival.