Downloaded from https://journals.lww.com/ajg by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3ojN+4MJbriOWz0uU9QLoQLRFB151soihz8D2ogfK+qk= on 05/21/2019 © 2016 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY Abstracts S695 Methods: We sequenced the V3—V4 region of the 16S ribosomal RNA gene extracted from gastric cor- pus biopsies obtained during routine clinical endoscopies at the national university hospital in Leon, Nicaragua. Using the Illumina MiSeq platform and QIIME processing pipeline, we obtained microbiota measures of alpha diversity, beta diversity, and individual taxa-both phyla and operational taxonomic units (OTUs). We compared these measures between groups using the Kruskal—Wallis test. Results: Of 81 subjects, 30 had premalignant lesions and 51 did not, with characteristics described in Table 1. Firmicutes was the dominant phylum in both groups, but only composed 43% of sequences in the premalignant group, compared to 68% of non—premalignant sequences (Figure 1). The difference was related to the Proteobacteria phylum (35% of OTUs in premalignant and 18% in non—premaligant), pre- dominantly from Hp. No differences in alpha or beta diversity existed between groups however, whether calculated with Hp reads included or excluded (Figure 2). Alpha diversity was correlated with number of Hp counts (p<0.001). One OTU was found to be significantly less prevalent in premalignant than non— premalignant samples. The OTU belonged to the Bacillales order of the Firmicutes phylum (p=0.006). Conclusion: In this study of the gastric corpus microbiome, premalignant lesions were associated with increasing Hp density and a corresponding decrease in Firmicutes, but otherwise limited differences in microbiota compared to stomachs with minimal or non—atrophic gastritis on histology. Additional areas warranting study include the antrum microbiome and the microbiome in higher risk premalignant lesions (e.g., incomplete gastric intestinal metaplasia or dysplasia). 1216_A Figure 1 1216_B Figure 2. Summary of phyla detected in gastric biopsy specimens with gastric intestinal metaplasia (GIM), multifocal atrophic gastritis (MAG), non—atrophic gastritis (NAG), and “normal” histology, with H. pylori OTUs included. 1217 The American Neurogastroenterology and Motility Society Gastroparesis Cardinal Symptom Index—Daily Diary (ANMS GCSI—DD): Psychometric Evaluation in Patients With Idiopathic or Diabetic Gastroparesis Category Award (Stomach) Presidential Poster Award Dennis A. Revicki, PhD 1 , Rebecca Speck, PhD 1 , Sara Gleeson, MPH 1 , Jorge Puelles, PharmD, MSc, MBA 2 , Braden Kuo, MD 3 , Michael Camilleri, MD 4 , Henry P. Parkman, MD 5 . 1. Evidera, Bethesda, MD; 2. Takeda Pharmaceuticals, London, England, United Kingdom; 3. Massachusetts General Hospital, Boston, MA; 4. Mayo Clinic, Rochester, MN; 5. Temple University Hospital, Philadelphia, PA Introduction: The ANMS GCSI—DD was developed to meet FDA recommendations for use as a Patient Reported Outcome endpoint for therapeutic trials in gastroparesis. The purpose of this study was to 1215 Marginal Ulcers Are Less Responsive to Proton Pump Inhibitors Compared to Gastric Ulcers: A Case—Control Study Zhuo Z. Geng, MD, Itegbemie Obaitan, MD, Aasma Shaukat, MD. University of Minnesota, Minneapolis, MN Introduction: Marginal ulcers are a well—known complication after gastrectomy associated surgeries. Proton pump inhibitors (PPIs) are commonly used to treat marginal ulcers, but their effectiveness has not been well studied. Our objective was to conduct a retrospective case—control study to assess the effectiveness of PPI treatment of marginal ulcers in comparison to gastric ulcers and elucidate the risk factors affecting healing of marginal ulcers. Methods: The study was conducted at Minneapolis Veterans Affairs medical center from year 2000 to 2018. We included patients with marginal ulcers diagnosed at endoscopy who were treated with PPIs and had a follow—up endoscopy to check for healing. Each marginal ulcer patient was age—matched with two to three controls with gastric ulcers. We extracted data on patient characteristics and the responsive- ness to PPIs. Data were analyzed using logistic regression. Results: A total of 20 patients with marginal ulcers and 50 patients with gastric ulcers were included in the study. The baseline characteristics of participants were shown in Table 1. There was no statistically significant difference between the two groups. Among patients with marginal ulcers, the average age was 66 years and 85% were men; 40% were smokers and 60% had chronic nonsteroidal anti—inflammatory drugs (NSAIDs) use. The most common surgeries were Roux—en—Y gastric bypass (45%), and the most common surgery indications were obesity (30%). When compared to gastric ulcers, patients with mar- ginal ulcers were significantly less responsive to PPI treatment in healing ulcers (Odds Ratio=0.14; 95% Cl: 0.04—0.49, p=0.002). Age, endoscopy indications, ulcer size, PPI regimen, H pylori status, NSAIDs use, current smoking status, heavy alcohol use, diabetes and cardiovascular diseases were not indepen- dent risk factors affecting healing of marginal ulcers. Conclusion: Marginal ulcers are less responsive to PPIs compared to gastric ulcers, and about 50% of marginal ulcers will not heal with PPI treatment. Our study was limited by sample size. Further studies with large sample size are warranted to further investigate effective medical treatments for marginal ulcers and to identify risk factors impeding ulcer healing. 1215 Figure 1. Baseline Characteristics of the Study Population 1216 The Microbiota of Gastric Premalignant Lesions: 16S Sequencing Data in a Central American Population Michael Dougherty, MD, MSCR 1 , Andrea Azcarate—Peril, PhD 2 , Samuel Vilchez, PhD 3 , Carla Rivera, MD 3 , Javier Pastora, MD 3 , Douglas Morgan, MD, MPH 4 . 1. University of North Carolina School of Medicine, Durham, NC; 2. University of North Carolina School of Medicine, Chapel Hill, NC; 3. National Autonomous University of Nicaragua, Leon, Nicaragua; 4. Vanderbilt University Medical Center, Nashville, TN Introduction: Gastric cancer develops through a cascade, from gastritis to atrophic gastritis and gastric intestinal metaplasia, to dysplasia and adenocarcinoma. Helicobacter pylori (Hp) plays a prominent role in this cascade, contributing to the high cancer incidence in regions such as Central America, but the role of other bacteria is less defined. We compared the gastric microbiota of patients with and without premalignant lesions (atrophic gastritis or intestinal metaplasia).