Objective Serrated polyposis syndrome (SPS) is associated with an increased colorectal cancer (CRC) risk, although the magnitude of the risk remains uncertain. Whereas intensive endoscopic surveillance for CRC prevention is advised, predictors that identify patients who have high CRC risk remain unknown. We performed a multicentre nationwide study aimed at describing the CRC risk in patients with SPS and identifying clinicopathological predictors independently associated with CRC. Design From March 2013 through September 2014, patients with SPS were retrospectively recruited at 18 Spanish centres. Data were collected from medical, endoscopy and histopathology reports. Multivariate logistic regression was performed to identify CRC risk factors. Results In 296 patients with SPS with a median follow-up time of 45 months (IQR 26–79.7), a median of 26 (IQR 18.2–40.7) serrated polyps and 3 (IQR 1–6) adenomas per patient were detected. Forty-seven patients (15.8%) developed CRC at a mean age of 53.9±12.8, and 4 out of 47 (8.5%) tumours were detected during surveillance (cumulative CRC incidence 1.9%). Patients with >2 sessile serrated adenomas/polyps (SSA/Ps) proximal to splenic flexure and ≥1 proximal SSA/P with high-grade dysplasia were independent CRC risk factors (incremental OR=2, 95% CI 1.22 to 3.24, p=0.006). Patients with no risk factors showed a 55% decrease in CRC risk (OR=0.45, 95% CI 0.24 to 0.86, p=0.01). Conclusions Patients with SPS have an increased risk of CRC, although lower than previously published. Close colonoscopy surveillance in experienced centres show a low risk of developing CRC (1.9% in 5 years). Specific polyp features (SSA/P histology, proximal location and presence of high-grade dysplasia) should be used to guide clinical management.

Background Serrated polyposis syndrome (SPS) has been associated with an increased risk of colorectal cancer (CRC). Accordingly, intensive surveillance with annual colonoscopy is advised. The aim of this multicenter study was to describe the risk of advanced lesions in SPS patients undergoing surveillance, and to identify risk factors that could guide the prevention strategy. Methods From March 2013 to April 2015, 296 patients who fulfilled criteria I and/or III for SPS were retrospectively recruited at 18 centers. We selected patients in whom successful clearing colonoscopy had been performed and who underwent subsequent endoscopic surveillance. Advanced neoplasia was defined as CRC, advanced adenoma, or advanced serrated lesion that were ≥ 10 mm and/or with dysplasia. Cumulative incidence of advanced neoplasia was calculated and independent predictors of advanced neoplasia development were identified. Results In 152 SPS patients a total of 315 surveillance colonoscopies were performed (median 2, range 1 – 7). The 3-year cumulative incidence of CRC and advanced neoplasia were 3.1 % (95 % confidence interval [CI] 0 – 6.9) and 42.0 % (95 %CI 32.4 – 51.7), respectively. Fulfilling both I + III criteria and the presence of advanced serrated lesions at baseline colonoscopy were independent predictors of advanced neoplasia development (odds ratio [OR] 1.85, 95 %CI 1.03 – 3.33, P = 0.04 and OR 2.62, 95 %CI 1.18 – 5.81, P = 0.02, respectively). During follow-up, nine patients (5.9 %) were referred for surgery for invasive CRC (n = 4, 2.6 %) or because of polyp burden (n = 5, 3.3 %). After total colectomy, 17.9 % patients developed advanced neoplasia in the retained rectum. Conclusions Patients with SPS have a substantial risk of developing advanced neoplasia under endoscopic surveillance, whereas CRC incidence is low. Personalized endoscopic surveillance based on polyp burden and advanced serrated histology could help to optimize prevention in patients with SPS. ^* These authors contributed equally to the work

Digital single-operator cholangioscopy (D-SOC) has enhanced the ability to diagnose indeterminate biliary strictures (BSs). Pilot studies using artificial intelligence (AI) models in D-SOC demonstrated promising results. Our group aimed to develop a convolutional neural network (CNN) for the identification and morphological characterization of malignant BSs in D-SOC. A total of 84,994 images from 129 D-SOC exams in two centers (Portugal and Spain) were used for developing the CNN. Each image was categorized as either a normal/benign finding or as malignant lesion (the latter dependent on histopathological results). Additionally, the CNN was evaluated for the detection of morphologic features, including tumor vessels and papillary projections. The complete dataset was divided into training and validation datasets. The model was evaluated through its sensitivity, specificity, positive and negative predictive values, accuracy and area under the receiver-operating characteristic and precision-recall curves (AUROC and AUPRC, respectively). The model achieved a 82.9% overall accuracy, 83.5% sensitivity and 82.4% specificity, with an AUROC and AUPRC of 0.92 and 0.93, respectively. The developed CNN successfully distinguished benign findings from malignant BSs. The development and application of AI tools to D-SOC has the potential to significantly augment the diagnostic yield of this exam for identifying malignant strictures.

To reduce pancreatic cancer mortality, a paradigm shift in cancer screening is needed. Our group pioneered the use of low-coherence enhanced backscattering (LEBS) spectroscopy to predict the presence of pancreatic cancer by interrogating the duodenal mucosa. A previous ex vivo study (n = 203) demonstrated excellent diagnostic potential: sensitivity, 95%; specificity, 71%; and accuracy, 85%. The objective of the current case-control study was to evaluate this approach in vivo. We developed a novel endoscope-compatible fiber-optic probe to measure LEBS in the periampullary duodenum of 41 patients undergoing upper endoscopy. This approach enables minimally invasive detection of the ultrastructural consequences of pancreatic field carcinogenesis. The LEBS parameters and optical properties were significantly altered in patients harboring adenocarcinomas (including early-stage) throughout the pancreas relative to healthy controls. Test performance characteristics were excellent with sensitivity = 78%, specificity = 85%, and accuracy = 81%. Moreover, the LEBS prediction rule was not confounded by patients' demographics. We demonstrate the feasibility of in vivo measurement of histologically normal duodenal mucosa to predict the presence of adenocarcinoma throughout the pancreas. This represents the next step in establishing duodenal LEBS analysis as a prescreening technique that identifies clinically asymptomatic patients who are at elevated risk of PC.

INTRODUCTION: Endoscopic ultrasound (EUS) allows for characterization and biopsy of pancreatic lesions. Pancreatic cystic neoplasms (PCN) include mucinous (M-PCN) and nonmucinous lesions (NM-PCN). Pancreatic ductal adenocarcinoma (P-DAC) is the commonest pancreatic solid lesion (PSL), followed by pancreatic neuroendocrine tumor (P-NET). Although EUS is preferred for pancreatic lesion evaluation, its diagnostic accuracy is suboptimal. This multicentric study aims to develop a convolutional neural network (CNN) for detecting and distinguishing PCN (namely M-PCN and NM-PCN) and PSL (particularly P-DAC and P-NET). METHODS: A CNN was developed with 378 EUS examinations from 4 international reference centers (Centro Hospitalar Universitário São João, Hospital Universitario Puerta de Hierro Majadahonda, New York University Hospitals, Hospital das Clínicas Faculdade de Medicina da Universidade de São Paulo). About 126.000 images were obtained—19.528 M-PCN, 8.175 NM-PCN, 64.286 P-DAC, 29.153 P-NET, and 4.858 normal pancreas images. A trinary CNN differentiated normal pancreas tissue from M-PCN and NM-PCN. A binary CNN distinguished P-DAC from P-NET. The total data set was divided into a training and testing data set (used for model's evaluation) in a 90/10% ratio. The model was evaluated through its sensitivity, specificity, positive and negative predictive values, and accuracy. RESULTS: The CNN had 99.1% accuracy for identifying normal pancreatic tissue, 99.0% and 99.8% for M-PCN and NM-PCN, respectively. P-DAC and P-NET were distinguished with 94.0% accuracy. DISCUSSION: Our group developed the first worldwide CNN capable of detecting and differentiating the commonest PCN and PSL in EUS images, using examinations from 4 centers in 2 continents, minimizing the impact of the demographic bias. Larger multicentric studies are needed for technology implementation.