Search

The prediction of microsatellite instability (MSI) using deep learning (DL) techniques could have significant benefits, including reducing cost and increasing MSI testing of colorectal cancer (CRC) patients. Nonetheless, batch effects or systematic biases are not well characterized in digital histology models and lead to overoptimistic estimates of model performance. Methods to not only palliate but to directly abrogate biases are needed. We present a multiple bias rejecting DL system based on adversarial networks for the prediction of MSI in CRC from tissue microarrays (TMAs), trained and validated in 1788 patients from EPICOLON and HGUA. The system consists of an end-to-end image preprocessing module that tile samples at multiple magnifications and a tissue classification module linked to the bias-rejecting MSI predictor. We detected three biases associated with the learned representations of a baseline model: the project of origin of samples, the patient's spot and the TMA glass where each spot was placed. The system was trained to directly avoid learning the batch effects of those variables. The learned features from the bias-ablated model achieved maximum discriminative power with respect to the task and minimal statistical mean dependence with the biases. The impact of different magnifications, types of tissues and the model performance at tile vs patient level is analyzed. The AUC at tile level, and including all three selected tissues (tumor epithelium, mucin and lymphocytic regions) and 4 magnifications, was 0.87 +/- 0.03 and increased to 0.9 +/- 0.03 at patient level. To the best of our knowledge, this is the first work that incorporates a multiple bias ablation technique at the DL architecture in digital pathology, and the first using TMAs for the MSI prediction task.

There is limited scientific evidence available to stratify the risk of developing metachronous colorectal cancer after resection of colonic polyps and to determine surveillance intervals and is mostly based on observational studies. However, while awaiting further evidence, the criteria of endoscopic follow-up needs to be unified in our setting. Therefore, the Spanish Association of Gastroenterology, the Spanish Society of Family and Community Medicine, the Spanish Society of Digestive Endoscopy, and the Colorectal Cancer Screening Group of the Spanish Society of Epidemiology, have written this consensus document, which is included in chapter 10 of the "Clinical Practice Guideline for Diagnosis and Prevention of Colorectal Cancer. 2018 Update". Important developments will also be presented as regards the previous edition published in 2009. First of all, situations that require and do not require endoscopic surveillance are established, and the need of endoscopic surveillance of individuals who do not present a special risk of metachronous colon cancer is eliminated. Secondly, endoscopic surveillance recommendations are established in individuals with serrated polyps. Finally, unlike the previous edition, endoscopic surveillance recommendations are given in patients operated on for colorectal cancer. At the same time, it represents an advance on the European guideline for quality assurance in colorectal cancer screening, since it eliminates the division between intermediate risk group and high risk group, which means the elimination of a considerable proportion of colonoscopies of early surveillance. Finally, clear recommendations are given on the absence of need for follow-up in the low risk group, for which the European guidelines maintained some ambiguity. (C) 2018 Elsevier Espana, S.L.U. All rights reserved.

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.

Background: Patients with ulcerative colitis (UC) have an increased risk of colorectal cancer. Scarce data regarding the development of adenomas in these patients are available both for normal and colitic mucosa. Objective: The objective of this article is to evaluate the prevalence of adenomatous polyps and associated risk factors in patients with UC. Methods: Patients with UC were identified from the databases of two tertiary referral centers. Medical, endoscopic and histologic reports were reviewed. Results: A total of 403 patients were included (53% male; 33% extensive colitis) and 1065 colonoscopies (median per patient, 2) were recorded and analyzed. Seventy-four adenomas in 47 patients (11.7%) and three cases of colorectal cancer were found during a median follow-up of 6.3 years. The cumulative risk of colorectal adenoma was 4.7%, 16.7%, 23.6% and 34.4% at 10, 20, 30 and 40 years from UC diagnosis, respectively. The cumulative risk of developing metachronous colorectal adenoma was 66.7%, 87.9%, and 90.9% at 5, 10, and 15 years from first adenoma detection. Older age at UC diagnosis and longer disease duration were independent risk factors for colorectal adenoma development. Conclusions: The prevalence of colorectal adenomas among UC patients seems to be higher than previously reported, although lower than in the background population.

BACKGROUND & AIMS: We investigated whether patients with multiple serrated polyps, but not meeting the World Health Organization criteria for serrated polyposis syndrome, and their relatives have similar risks for colorectal cancer (CRC) as those diagnosed with serrated polyposis. METHODS: We collected data from patients with more than 10 colonic polyps, recruited in 2008-2009 from 24 hospitals in Spain for a study of causes of multiple colonic polyps. We analyzed data from 53 patients who met the criteria for serrated polyposis and 145 patients who did not meet these criteria, but who had more than 10 polyps throughout the colon, of which more than 50% were serrated. We calculated age-and sex-adjusted standardized incidence ratios (SIRs) for CRC in both groups, as well as in their first-degree relatives. RESULTS: The prevalence of CRC was similar between patients with confirmed serrated polyposis and multiple serrated polyps (odds ratio, 1.35; 95% confidence interval [CI], 0.64-2.82; P=.40). The SIR for CRC in patients with serrated polyposis (0.51; 95% CI, 0.01-2.82) did not differ significantly from the SIR for CRC in patients with multiple serrated polyps (0.74; 95% CI, 0.20-1.90; P=.70). The SIR for CRC also did not differ significantly between first-degree relatives of these groups (serrated polyposis: 3.28, 95% CI, 2.16-4.77; multiple serrated polyps: 2.79, 95% CI, 2.10-3.63; P =.50). Kaplan-Meier analysis showed no differences in the incidence of CRC between groups during the follow-up period (log-rank, 0.6). CONCLUSIONS: The risk of CRC in patients with multiple serrated polyps who do not meet the criteria for serrated polyposis, and in their first-degree relatives, is similar to that of patients diagnosed with serrated polyposis.

Background: There is no information on the impact of age and gender on the diagnostic yield of different positivity thresholds for the fecal immunochemical test for hemoglobin (FIT). Objectives: To evaluate the performance of this test at distinct positivity cut-offs in a population -based colorectal cancer (CRC) screening program. Methods: CRC detection rate (DR), and analysis of resources were evaluated retrospectively, at different cut-offs of FIT (20, 25, 30, 35 and 40 tig Hb/g) respect to a reference value (15 tig Hb/g), according to age and gender, in a screening population of 10,611 participants of the ColonPrev study (Quintero. NEJM 2013). Results: At the reference cut-off value, 36 CRC and 252 advanced adenomas (AA) were diagnosed. Increasing the cut-off in women 60 years increasing cut-off at 25-30 tig Hb/g. In men >60 years, all increases in the cut-off affected the CRC DR, especially when the cut-off was increased from 35 to 40 tig Hb/g (CRC miss rate 25%). Conclusions: To improve the performance of FIT in CRC screening programs, FIT cut-offs could be individualized by age and gender. (C) 2016 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.

BACKGROUND & AIMS: We compared the ability of biennial fecal immunochemical testing (FIT) and one-time sigmoidoscopy to detect colon side-specific advanced neoplasms in a population-based, multicenter, nationwide, randomized controlled trial. METHODS: We identified asymptomatic men and women, 50-69 years old, through community health registries and randomly assigned them to groups that received a single colonoscopy examination or biennial FIT. Sigmoidoscopy yield was simulated from results obtained from the colonoscopy group, according to the criteria proposed in the UK Flexible Sigmoidoscopy Trial for colonoscopy referral. Patients who underwent FIT and were found to have >= 75 ng hemoglobin/mL were referred for colonoscopy. Data were analyzed from 5059 subjects in the colonoscopy group and 10,507 in the FIT group. The main outcome was rate of detection of any advanced neoplasm proximal to the splenic flexure. RESULTS: Advanced neoplasms were detected in 317 subjects (6.3%) in the sigmoidoscopy simulation group compared with 288 (2.7%) in the FIT group (odds ratio for sigmoidoscopy, 2.29; 95% confidence interval, 1.93-2.70; P = .0001). Sigmoidoscopy also detected advanced distal neoplasia in a higher percentage of patients than FIT (odds ratio, 2.61; 95% confidence interval, 2.20-3.10; P = .0001). The methods did not differ significantly in identifying patients with advanced proximal neoplasms (odds ratio, 1.17; 95% confidence interval, 0.78-1.76; P = .44). This was probably due to the lower performance of both strategies in detecting patients with proximal lesions (sigmoidoscopy detected these in 19.1% of patients and FIT in 14.9% of patients) vs distal ones (sigmoidoscopy detected these in 86.8% of patients and FIT in 33.5% of patients). Sigmoidoscopy, but not FIT, detected proximal lesions in lower percentages of women (especially those 50-59 years old) than men. CONCLUSIONS: Sigmoidoscopy and FIT have similar limitations in detecting advanced proximal neoplasms, which depend on patients' characteristics; sigmoidoscopy underperforms for women 50-59 years old. Screening strategies should be designed on the basis of target population to increase effectiveness and cost-effectiveness. ClinicalTrials.govnumber: NCT00906997

Colorectal cancer (CRC) is the second leading cause of cancer-related death among men and women in Western countries. Once a tumour develops, a differentiated prognosis could be determined by lifestyle habits or inherited and somatic genetic factors. Finding such prognostic factors will be helpful in order to identify cases with a shorter survival or at a higher risk of recurrence that may benefit from more intensive treatment and follow-up surveillance. Sixteen CRC genetic susceptibility variants were directly genotyped in a cohort of 1235 CRC patients recruited by the EPICOLON Spanish consortium. Univariate Cox and multivariate regression analyses were performed taking as primary outcomes overall survival (OS), disease-free survival and recurrence-free interval. Genetic variants rs9929218 at 16q22.1 and rs10795668 at 10p14 may have an effect on OS. The G allele of rs9929218 was linked with a better OS [GG genotype, genotypic model: hazard ratio (HR) 0.65, 95% confidence interval (CI) 0.450.93, P 0.0179; GG/GA genotypes, dominant model: HR 0.66, 95% CI 0.470.94, P 0.0202]. Likewise, the G allele of rs10795668 was associated with better clinical outcome (GG genotype, genotypic model: HR 0.73, 95% CI 0.531.01, P 0.0570; GA genotype, genotypic model: HR 0.66, 95% CI 0.470.92, P 0.0137; GG/GA genotypes, dominant model: HR 0.68, 95% CI 0.500.94, P 0.0194). In conclusion, CRC susceptibility variants rs9929218 and rs10795668 may exert some influence in modulating patients survival and they deserve to be further tested in additional CRC cohorts in order to confirm their potential as prognosis or predictive biomarkers.

Objective: to determine the prevalence of Helicobacter pylori infection in patients having undergone gastrectomy for non-neoplastic disease who later developed gastric stump cancer. Material and methods: retrospective study of all patients with partial gastrectomy for non-malignant peptic disease who were submitted to an endoscopic exploration between 1995 and 2001. A comparison was made of major clinical and histological characteristics, and the presence of Helicobacter pylori among patients with and without gastric cancer in the stomach remnant. Results: a total of 73 patients were studied in this period. Fifteen patients (20.5%) had remnant-stump gastric cancer. All but one were adenocarcinomas (71% intestinal and 29% diffuse, respectively). The average time between diagnosis of gastric cancer and previous gastrectomy was 32 (14-48) years. There was a higher detection rate of Helicobacter pylori in patients with cancer in the gastric remnant (100 vs. 81.5%, respectively, p < 0.07). No relationship was seen between type of gastric reconstruction (Billroth I or II) and rate of Helicobacter pylori detection. Conclusions: Helicobacter pylori infection is frequent in patients with previous gastrectomy for non-neoplastic disease. The results of the study suggest that Helicobacter pylori infection may play a role in gastric stump cancer. Objetivo: determinar la prevalencia de la infección por Helicobacter pylori en pacientes gastrectomizados por enfermedad no neoplásica, y que han desarrollado posteriormente cáncer gástrico. Material y métodos: estudio retrospectivo con reclutamiento de todos los pacientes con gastrectomía parcial por enfermedad péptica benigna que han sido sometidos a una exploración endoscópica entre 1995-2001. Se ha realizado una comparación de las principales características clínicas e histológicas y de la presencia de Helicobacter pylori en los pacientes con y sin cáncer del remanente gástrico. Resultados: se han estudiado un total de 73 pacientes en este periodo. Se han encontrado 15 pacientes (20,5%) con cáncer en el remanente gástrico, 14 adenocarcinomas (71% tipo intestinal y 29% tipo difuso) y un linfoma. El tiempo transcurrido entre el diagnóstico de cáncer gástrico y la gastrectomía previa ha sido de 32 (14-48) años. Se ha detectado un alto porcentaje de infección por Helicobacter pylori (100% en los pacientes con cáncer vs. 81,5% en los pacientes sin cáncer, p < 0,07). No se ha observado relación entre el tipo de reconstrucción gástrica (Billroth I o II) y el porcentaje de infección por Helicobacter pylori. Conclusiones: la infección por Helicobacter pylori es frecuente en pacientes gastrectomizados por patología benigna. Los resultados de este estudio sugieren que la infección por Helicobacter pylori podría jugar un papel en el cáncer gástrico.

Liver biopsies with a main histological diagnosis of steatosis were selected from 3,422 liver biopsies carried out in our department between January 1995 and December 1998. Patients with known risk factors for steatosis, such as excessive alcohol consumption, hepatitis C infection, treatment with amiodarone, perhexiline maleate, tamoxifen, antiviral drugs (didanosine, zidovudine) methotrexate, sodium valproate or total parenteral nutrition, Wilson's disease and organ transplant were subsequently excluded. Of the 43 liver biopsies finally included in the study, 23 showed simple steatosis and 20 steatohepatitis. Eighty-one per cent of the patients were male (mean age of 44 years) and the majority were asymptomatic. The most frequent indication for liver biopsy was hypertransaminasemia. No differences were observed between the two groups in terms of frequency and severity of classical risk factors for steatosis (diabetes mellitus, dyslipemia and obesity). Thirty-five percent of patients with steatohepatitis and 26% of those with simple steatosis had none of these risk factors. Patients with steatohepatitis were older than those with simple steatosis. They presented more severe symptomatology, the degree of steatosis was more intense and laboratory investigations showed greater alterations. These results suggest that simple steatosis and nonalcoholic steatohepatitis are two different phases of the same disease. The difficulty in clinical differentiation justifies carrying out liver biopsy, especially in patients with more severe symptomatology whose laboratory results show greater alterations, since these patients present more marked histological lesions, are at risk of developing liver cirrhosis and require therapy.

Gastrointestinal stromal tumors form a small percentage of digestive tract tumors. They have recently been the subject of a new histopathologic classification made possible by immunohistochemical techniques. A series of criteria has been proposed which enables the identification of the most malignant and clinically aggressive tumors. We present five patients who were diagnosed with gastrointestinal stromal tumor, each of which had distinct characteristics. The different diagnostic tests performed, the difficulties of reaching a diagnosis as well as the treatment and distinct behavior of these tumors are discussed.

The incidence of de novo neoplasms was analyzed in 340 patients with liver transplantation who survived more than 2 months post transplantation. Sixteen (4.7%) patients developed a new tumor following transplantation. The most frequent tumor observed was a lymphoma which was detected in four patients (1.2%). In three of the four lymphomas histologic diagnosis of non Hodgkin phenotype B lymphoma was confirmed and in three patients the central nervous system was involved. The remaining tumors consisted of two cases of adenocarcinoma of the colon, papillary carcinoma of the urinary bladder and ductal breast cancer (0.6%) for each of these tumors and one case of cervical cancer, adenocarcinoma of the small intestine, Kaposi sarcoma, laryngeal carcinoma, pharyngeal carcinoma and malignant melanoma (0.3% for each tumor). None of the patients developed more than one tumor. The mean time to the appearance of the tumors was 28 months (range: 3-52 months). These results suggest that de novo neoplasms in patients with liver transplantation are relatively frequent, particularly lymphoma.