Japanese Upper GI Symptoms Compared With Iranian and Canadian Patients Presenting

Objective The purpose of the trial is to compare upper GI symptoms and endoscopy findings in Canada with Japan and Iran, and correlate this with the upper GI microbiome.

The primary outcomes is the bacterial microbiome in patients with FD compared to control patients without upper GI symptoms and a normal endoscopy. The investigators will do a planned subgroup analysis according to country of recruitment as well as sex and age based subgroups (<or =50 years versus >50 years)

Secondary outcomes are

1. The microbiome in esophagitis patients compared to patients without esophagitis. To assess the oral nitrate reducing bacteria on the dorsum of the tongue of erosive GERD patients, patients with non-erosive GERD and comparing them with those of the normal controls enrolled in the JUICE study.
2. The microbiome in H. pylori negative peptic ulcer disease (PUD) patients compared to H. pylori positive PUD patients and asymptomatic normal endoscopy controls.
3. The microbiome in those with anxiety patients versus controls without anxiety in the FD group.
4. The microbiome in those with depression versus controls without depression in the FD group.
5. The microbiome in those with anxiety patients versus controls without anxiety in the patients without upper GI symptoms and a normal endoscopy.
6. The microbiome in those with depression versus controls without depression in the patients without upper GI symptoms and a normal endoscopy
7. Comparison of upper GI pathology (inflammation, dysplasia, malignancy) and the microbiome in the Canadian versus the Japanese populations.

Methodology Study population, inclusion \& exclusion criteria This is a prospective international cohort study in Canada, Japan, and Iran. The investigators will recruit adult patients (≥18 years of age) who are undergoing esophago-gastro duodenoscopy (EGD) for any reason and are consenting to be enrolled in the study during the study period. The control group will be patients with no or minimal upper GI symptoms who are referred for EGD. Exclusion criteria will be the patients who cannot speak English in Canada, who cannot speak Japanese in Japan, and cannot speak Persian in Iran.

Methods Written consent will be obtained from all participants. The consent by clinician for medical reasons includes standard endoscopy and gastric biopsies for assessing H.pylori status. The research consent contains saliva collection with specific device called "Oracol plus®" and tissue brushing at the oesophagus, stomach (body and antrum) and distal duodenum. If the patient is eligible for this study, the investigators will record demographics (age, sex, education, race, past medical history, medication taking - especially PPIs and antibiotics), and the main reason for endoscopy. Patients will complete questionnaires; upper GI symptoms measured by Short Form Leeds Dyspepsia Questionnaire (SFLDQ), quality of life measured by EQ-5D, and anxiety and depression measured by The Hospital Anxiety and Depression Scale (HADS). Then saliva collection device; Oracol plus® (Malvern Medical Developments, Worchester, UK) will be applied to collect saliva. The device is designed to be used in a similar way to a toothbrush. Saliva is collected by rubbing the sponge swab firmly along the gum (at the base of the teeth if present) as well as the dorsum of the tongue until the sponge is wet, this takes about 1 minute. Once the saliva has been taken it is secure within the container, a microtube is incorporated within the device so that the saliva is centrifuged directly into the micro tube provided. This reduces the risk of aerosol contamination. This process is aimed to document the microbiomes in oral cavity in order to confirm the resident microbiomes in upper GI tract.

The oral nitrate reductase activity (NRA) measurement

Considering that erosive GERD is rather infrequent in Japan, this part of the study will be performed in Hamilton and Tehran as follows:

Twenty erosive GERD patients, twenty non-erosive GERD patients enrolled in JUICE from Canada and twenty controls with no upper GI symptoms are enrolled after informed consent is obtained. In addition to the procedures in described in JUICE, the following is done for these groups:

1. An Oracol plus® swab is brushed over the gums and the dorsum of the tongue for 60 seconds as described in the main protocol. It is processed in the same way and the microbiome study is done in the same way described for the JUICE protocol proper.
2. NRA is measured with consecutive oral cavity samplings at one minute intervals for three minutes as follows: cases and controls are asked to fast overnight and avoid high-nitrate foods (including lettuce, cabbage, celery, turnip, beetroot, radish, and rhubarb) for at least 8 h. Mouth NRA assay is done between 7:30 and 10:00 am. On the morning of the test, the individual is initially asked to rinse his/her mouth with 22 ml of deionized, sterile water. Then he/she is asked to hold 22 ml of a 10 mg nitrate-N/L solution in the mouth for 3 min. The subject is instructed to mix the solution in the mouth at the beginning and every 1 min over the 3 min, and 1.5 ml of the solution incubated in the mouth is sampled with a sterile syringe after each mixing (i.e., at the beginning and at 1-min intervals). The rest of the solution is discarded. Each sample is immediately transferred into an Eppendorf tube containing 56.25 microliter 0.8 M NaOH (to destroy vitamin C and top further nitrite production). Then 18.0 microliter 1.4 M ZnSO4 is added to each tube, shaken for 2 s and left at room temperature for 30 min. The samples are then centrifuged at 13,000g at 25℃ for 7 min. This treatment removes the proteins which could possibly interfere with the Griess reaction used for nitrite measurement. Nitrite is measured by adding 250 microliter of Griess reagent A [0.2% (w/v) sulfanilamide in 5% (v/v) concentrated H3PO4] to 500 microliter of the supernatant of the centrifuged sample in a 1,000-microliter cuvette and reading absorbance at 540 nm using a spectrophotometer (WPA BiowaveII, UK). This reading is named ''A1.'' Then, 250 microliter of Griess reagent B [0.02% (w/v) N-(1-naphtyl)-ethylenediamine-dihydrochloride solution] is added to the cuvette. The cuvette is then kept in the dark at room temperature for 10 min and the absorbance was again read at 540 nm and the reading named ''A2.'' The value (A2-A1) is read against a standard curve developed by testing known concentrations of NaNO2 and the final nitrite concentration determined. A 500-microliter sample of the de-ionized sterile water is treated exactly like the saliva sample in each instance and serves as blank for quality control. Nitrite values at 0, 1, 2, and 3 min are plotted against time and the slope of the trend line calculated. This slope represents the nitrite formed per person per minute. The slope is multiplied by 22 (the volume of nitrate solution) to calculate the ''nitrate reductase activity'' and reported as microgram nitrite-N formed per person per minute.

Definitions:

For practical purposes, GERD is defined clinically according to the Montreal summit as troublesome regurgitation or heartburn occurring at least once a week. This definition will be used to enroll the non-erosive patients as well as excluding GERD in controls. It will also be used to assess the erosive patients clinically. The controls need not only to be asymptomatic (according to the above-mentioned definition), but also to have no gross endoscopic sign of reflux.

Esophagogastroduodenoscopy and sampling for the microbiome EGD will be performed and a sterile sheathed brush will be used to collect gut microbiota. For each patient, tissue brushing will be performed first from the the middle and lower esophagus with one brush, then the stomach (greater and lesser curvature of body and antrum) with a second brush, the duodenal cap and distual duodenum with a third brush. This will be conducted as the endoscope is introduced and in each case the collection will be obtained by gently brushing the appropriate area. The material collected on the brush surface will be re-suspended into RNA later in a sterile tube. Samples were allowed to incubate at room temperature for up to 2 hours, then frozen and stored at -80 °C. Japanese frozen microbiome samples will ship or bring to McMaster Genome Centre by one of investigators. The Iranian samples will be processed in Tehran and the extracted DNA samples will be transferred to McMaster for sequencing. The last part i.e. the analysis of the sequenced DNA samples can either be done in Tehran or at McMaster. Two biopsies each from antrum and body of stomach will be obtained to assess H. pylori status as a standard of care of both cohorts.

Microbiome analysis DNA extraction and molecular profiling will be carried out using standard protocols. Bacterial community profiling of the 16S rRNA gene will be carried out using paired end reads of the V3 region. Triplicate amplifications will be pooled for 150 or 250 nt paired-end Illumina sequencing in the McMaster Genome Center. This approach provides overlapping sequence reads of the V3 region, which can be used for correcting poor quality base calls and increasing sequencing accuracy. Multiplexing the run with barcoded primers will provide approximately 25,000 sequence reads per sample. The sequence data will be processed by an in-house bioinformatics pipeline that incorporates quality filtering with cutadap PandaSeq, AbundantOTU+, mothur, and QIIME. Output will include clustered sequences in operational taxonomic units (OTUs) using AbundantOTU+ and taxonomic assignments using the RDP classifier classifier using the Greengenes training set. Microbiome analysis will include α-diversity metrics for each sample and β-diversity measures (weighted and unweighted unifrac, Bray-Curtis, nonmetric multidimensional scaling) and other statistical analysis using QIIME25 and PhyloSeq. Functional properties of the microbiota will be inferred using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) and further compared using Statistical

Sample size The investigators plan to recruit 500 new patients referred for upper GI endoscopy in Canada (McMaster University) and 500 in Japan (Division of Gastroenterology, Tohoku University Hospital) and 500 from Iran (Tehran University). This is based on preliminary data from 20 patients that have had assessment of the microbiome in the esophagus, stomach and duodenum and assumes that the odds ratio for the operational taxonomic unit will be 2 fold between organic disease and healthy control (EGD with no GI symptoms e.g. for evaluation of anemia where no abnormality is found) and assumes that there will be at least 50 healthy controls and 50 patients with a given disease. 500 patients will be needed as the investigators will be evaluating gastric ulcer, FD, esophagitis separately and will need to adjust for multiple testing.

For NRA measurement, the investigators need 18 patients among GERD, NERD and control arm. This assumes a Z-alpha of 1.96 (two-sided), 1-beta=0.8416, Sigma of 0.99, and mean difference of 0.9310.

Statistical analysis Baseline descriptive data will be analyzed and reported as means and standard deviations for continuous variables, and percentage and frequency for categorical variables. Dichotomous variables will be analyzed suing Chi squared (x2) test and continuous using Student's t test (for two groups) and ANOVA (for more than two groups) unless numbers are < 50 in a group when the investigators will use Mann Whitney U test for two groups and Kruskal-Wallis for more than two groups. SPSS version 21.0 (SPSS (Japan) Co., Ltd., Tokyo, Japan) for Windows systems will be used with differences considered significant at the <0.05 level for the primary outcomes and <0.01 for the secondary to adjust for multiple testing.