Murray, Joseph A.

Abstract Background Celiac disease (CeD) is a T-cell mediated enteropathy driven by gluten in genetically susceptible individuals. Mucosal inflammation persists for years even when treated with a gluten-free diet (GFD). Although patients may be advised to increase dietary fibre, some fibres are poorly tolerated leading, potentially, to suboptimal fibre intake and hampering dietary recommendations. Aims Our project aims to investigate whether and how the supplementation of two fibre types with distinct functional profiles affects the recovery of gluten-immunopathology in a mouse model of gluten-sensitivity. We hypothesize that: (1) HylonVII, but not inulin, accelerates mucosal recovery by increasing microbial carbolytic activity and production of short-chain fatty acids (SCFA); (2) fibre intake is decreased in CeD patients, leading to decreased SCFA production. Methods NOD-DQ8 mice were immunized to gliadin and then placed on a gluten-containing diet. Three weeks later, mice were returned to GFD supplemented with or without HylonVII or inulin to 15% total fibre, for 6, 10, or 12 weeks. Recovery from gluten-induced pathology was evaluated by counting CD3+ intraepithelial lymphocytes (IELs) in villi tips and measuring villus-to-crypt (V/C) ratios in the small intestine. Adult CeD patients (newly diagnosed n=6 and GFD-treated n=7) and healthy controls (n=5) completed a food frequency questionnaire (Victoria DQES v2) and provided fecal samples for evaluation of SCFA production. Fecal SCFA were measured with targeted single ion monitoring gas-chromatography/mass-spectrometry (Agilent) as markers of fibre metabolism. Results In gluten immunized mice, 3 weeks of wheat diet led to increased CD3+ IEL counts and decreased V/C ratios (pampersand:003C0.01), which remained abnormal after 6 weeks of GFD. Mice consuming GFD+inulin, but not HylonVII, normalized CD3+ IELs and V/C ratios, had higher fecal SCFAs (pampersand:003C0.05) (primarily acetate), and a higher carbolytic-proteolytic metabolite ratio (pampersand:003C0.05) compared with mice consuming a GFD alone for six weeks. Only 1/7 patients with CeD on GFD met the minimum recommendation for fibre intake (25 g/day, Health Canada), compared with 5/11 participants on gluten-containing diets. SCFA production was impaired in active but not in treated CeD patients (2563 vs 3744 μg/g, pampersand:003C0.01). Conclusions In gluten-immunized NOD-DQ8 mice, GFD+inulin accelerates mucosal recovery, which is associated with increased fecal SCFA. In our pilot clinical data, patients on a GFD did not meet the recommendation for fibre intake although they were capable of metabolizing fibre to produce SCFA that could facilitate mucosal recovery. Together these results suggest that CeD on GFD may benefit from increasing fibre, such as inulin, during the GFD. Funding Agencies CIHRCeliac Canada

Serologic measures of tissue transglutaminase (tTG) immunoglobulin A (IgA) and deamidated gliadin peptide (DGP) IgA and immunoglobulin G (IgG) are hallmark tests utilized when diagnosing individuals for celiac disease (CeD) and for monitoring adherence to a gluten-free diet (GFD), currently the only available treatment for CeD. We address two issues in this study: (i) the relapse to seropositivity for CeD patients who resume a gluten containing diet and (ii) the correlation between two different tTG-IgA assays near the upper limit of normal (ULN) designated thresholds. Regarding the first issue, often a suspected CeD individual is put back on a gluten diet to return to their serologic levels. However, we show it requires a substantial amount of gluten for serology to return to a positive level. For example, in one study of 22 patients treated with placebo and taking 84 g of gluten over 6 weeks, only two converted from seronegative to seropositive for tTG-IgA. Regarding the second topic, we compare the relationship for different serologic assays, namely tTG-IgA AB (recombinant, ULN = 4 units/mL) vs. tTG-IgA (non-recombinant, ULN = 20 units). There is a strong correlation between both measurements as evidenced by a Pearson coefficient of R = 0.8584; however, we observed that the cross-correlation in terms of sensitivity and specificity improved substantially by using an ULN value of three instead of four for the tTG-IgA AB (recombinant) assay. This result suggests that assay thresholds used for initial diagnosis in patients who have not yet started a GFD may need to be adjusted for monitoring and in the setting of a diagnostic gluten challenge.

Goals: We aimed to evaluate symptom outcomes in those on a gluten-free diet during the 5 years after diagnosis. Background: Celiac disease is common; however, little is known about long-term symptom outcomes. Study: We performed a retrospective chart review on individuals with celiac disease followed at a tertiary referral center between 2012 and 2018. To minimize bias, strict inclusion/exclusion criteria were utilized. Only those with definitive biopsy-proven celiac disease, on a gluten-free diet, and with systematic follow-up were included. The standardized care at this center reduced the risk that decisions on testing and follow-up visits were determined by symptom status. Summary statistics were computed and generalized linear models with a logit link were used to associate the proportion of symptomatic visits with various covariates using R statistical programming. Results: Of the 1023 records reviewed, 212 met inclusion/exclusion criteria; 146 (69%) were female and the mean age at diagnosis was 43 (range: 11 to 84 y old). During follow-up, over 50% remained symptomatic, with many having the same symptoms that prompted the diagnosis. The only predictors for remaining symptomatic were female sex and younger age at diagnosis. Abnormal serology during follow-up and small bowel normalization were not predictive. Conclusions: In individuals with definitive celiac disease with systematic long-term follow-up in a Celiac Clinic, roughly half remained symptomatic despite a gluten-free diet. Many suffer from the same symptoms that prompted the diagnosis of celiac disease. Small bowel healing and abnormal serology in follow-up were not predictive of remaining symptomatic. These findings stress the importance of long-term care in celiac disease.

Background Therapy for celiac disease (CD) mainly relies on following a gluten‐free diet (GFD); however, a serum marker for gluten intake has yet to be established. Aims To evaluate the utility of alkylresorcinol concentrations for detecting gluten intake in studies of human and mouse. Methods Alkylresorcinol concentrations were compared among treated CD patients (n=34), untreated CD patients (n=36), and controls (n=33). Furthermore, 7 additional CD patients whose serum samples were available at diagnosis and after GFD were evaluated. In mice studies, alkylresorcinol concentrations were compared in the serum of 5 mice fed a regular chow and 10 mice fed lifelong with a gluten‐free chow. In addition, the effect of adding gluten on changes of alkylresorcinol concentrations was also evaluated. Results Total alkylresorcinol concentrations were significantly lower in treated CD patients (median [IQR], 3 (2–8) nmol/L), compared to untreated CD patients (median [IQR], 32 [11–74] nmol/L; P<.0001) or healthy controls (median [IQR], 54 [23–112] nmol/L; P<.0001). Moreover, alkylresorcinol concentrations in CD patients significantly decreased after introduction of a GFD (median, 34 nmol/L at diagnosis vs. 5 nmol.L after GFD, p=0.02). In the mice, median (IQR) total alkylresorcinol concentrations in serum samples of mice fed lifelong with a gluten‐free chow was 1.8 (1.6–2.3) nmol/L, which was further significantly increased to 16 (11–22) nmol/L after 8 days of feeding with the gluten free chow that had gluten added to it. (p=.008). Conclusion Serum alkylresorcinol concentrations could be a useful marker for dietary gluten in CD. Support or Funding Information No funding to report.

Goals: We aim to describe the weight loss outcomes of patients with celiac disease (CeD) taking antiobesity medications (AOMs) and compare it with the weight loss outcomes of patients without CeD taking AOMs. Background: Increasing rates of obesity and obesity-associated comorbidities have been previously reported in patients with CeD on a gluten-free diet. The effectiveness of AOMs in this population has not been previously described. Methods: In our retrospective cohort study, we matched 39 patients with treated CeD to 78 patients without CeD based on sex and AOM. We assessed the weight loss outcomes at 3, 6, and 12 months after starting the AOM in both cohorts and analyzed if there was a differential response when comparing by type of AOM [injectable glucagon-like peptide 1 (GLP-1) receptor agonists vs. oral non-GLP-1 AOMs]. Results: Both cohorts had similar baseline demographic and anthropometric characteristics. At 12 months, the CeD cohort had a nonsignificantly inferior total body weight loss percentage compared with the cohort without CeD (6.5% vs. 9.5%, P=0.13). The CeD cohort had a similar proportion of patients achieving a total body weight loss percentage of ≥5% than the cohort without CeD (72.7% vs. 72.1%, P=1.00). No significant difference was observed when comparing the weight loss outcomes of injectables (GLP-1 receptor agonists) to oral AOMs. The proportion of patients reporting side effects was similar for both groups, regardless of the type of AOM. Conclusion: Patients with CeD taking AOMs had similar weight loss outcomes to patients without CeD. Hence, AOMs can be a safe and effective therapy for weight management in patients with CeD.

SummaryBackgroundThe Th2 allergic pathway in eosinophilic oesophagitis (EoE) responds to food antigen exposure.AimTo compare the presence and temporal pattern of food antigen penetration in oesophageal mucosa in active and inactive EoE and controlsMethodsThirty‐two patients with EoE (20 active) and 10 controls were asked to eliminate all wheat and/or dairy 12, 24, 48, 72 or 96 hours before endoscopy. Immunostaining on endoscopic biopsies was performed for gliadin, casein and whey.ResultsGluten, casein and whey were detected by positive staining in 17/32 (53.1%), 21/32 (65.6%), and 30/32 (92.0%) of patients, respectively. In active vs inactive EoE, 70.0% vs 25.0% (P < 0.05), 80.0% vs 41.5%, and 90.0% vs 90.9% patients had detectable gliadin, casein and whey, respectively. Casein and whey (20.0% and 100%, respectively) but not gliadin, were present in controls. The gliadin staining density was greater in active compared to inactive disease at ≤ 24 vs >24 hours after exposure (P = 0.05) but no differences were detected when comparing active and inactive patients for casein and whey. There was greater staining density for whey than casein for all patients at ≤24 hours (mean 2.14 ± 0.91 and 1.07 ± 1.33, P = 0.02). In active EoE, IgG4 was present in 14/20 compared to one inactive patient.ConclusionThe oesophageal epithelium is selectively permeable and has relatively long dwell times for food antigens known to trigger EoE. The precise mechanism of antigen‐specific mucosal entry and the factors that determine the induction or effector trigger of the Th2 pathway activation merit further study.

Background: The renin-angiotensin-aldosterone system (RAAS) has been associated with gastrointestinal inflammation and fibrosis, suggesting that RAAS blockade may be beneficial in patients with inflammatory bowel disease. Using retrospective analysis, we aimed to compare the disease course of patients with Crohn’s disease (CD) taking two commonly prescribed classes of RAAS-blocking agents. Study: Patients with CD initiated on an angiotensin-converting enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB) between 2000 and 2016 were enrolled. Data on clinical, radiologic, and procedural surrogate markers of inflammatory bowel disease were collected in the subsequent 3, 5, and 10 years and compared with matched controls using univariate and multivariate analyses. Results: Compared with controls, patients taking ARBs had fewer instances of corticosteroid use (1.06 vs 2.88, P < 0.01) at 10 years. Patients taking ACEIs had an overall worse disease course, with more imaging studies (3.00 vs 1.75, P = 0.03) and endoscopic procedures (2.70 vs 1.78, P = 0.01) at 5 years, and more imaging studies (6.19 vs 3.50, P < 0.01), endoscopic procedures (5.91 vs 3.78, P < 0.01), and gastrointestinal operations (0.59 vs 0.18, P < 0.02) at 10 years. Results remained significant on multivariate analysis, adjusting for CD characteristics and the use of other antihypertensive medications. Conclusions: Our study provides insight into the long-term use of RAAS-blocking agents in patients with CD, suggesting that differences exist among commonly prescribed medication classes. While ACEIs were associated with an overall worse disease course at 5 and 10 years, patients taking ARBs were noted to have fewer instances of corticosteroid use at 10 years. Future large-scale studies are needed to further explore this association.