Society: AGA

Introduction Interleukin-15 (IL-15) is an important cytokine involved in tissue homeostasis and inflammation, and a potent regulator of innate and adaptive immune responses. IL-15 expression has been shown to be dysregulated in several gastro-intestinal inflammatory diseases, including celiac disease (CeD) and eosinophilic esophagitis (EoE). Preclinical studies in animal models with genetic loss or gain of function of IL-15, or using antibodies neutralizing IL-15, have suggested therapeutic potential of blocking IL-15 for CeD and EoE. CALY-002 is an IgG1k monoclonal antibody that neutralizes with equally high potency free and IL-15Ra-complexed IL-15.
Methodology A first-in-human study evaluating intravenously administered (IV) CALY-002 in single ascending doses (SAD) in healthy volunteers and multiple ascending doses (MAD) in subjects with CeD and EOE (NCT04593251) is ongoing. The SAD part of the study has been completed investigating the dose range of 0.01 - 10 mg/kg CALY-002 as 60 minutes infusion. Safety and pharmacokinetics were the primary endpoints. The pharmacodynamic effects of CALY-002 were assessed by analyzing peripheral blood leukocyte subsets using flow cytometry.
Results Adverse events were balanced in incidence and severity between CALY-002 and placebo-dosed subjects. No serious adverse events or adverse events above grade 2 occurred; marked safety observations in blood laboratory results, vital signs, ECGs, or physical examination were absent. Plasma CALY-002 concentration increased dose-proportionally with an average half-life of 22 days (Figure 1). CALY-002 induced, starting from Day 28, a 25-30% decrease in circulating NK cells starting from the anticipated pharmacologically active dose of 0.3 mg/kg confirming non-clinical modeling estimates (Figure 2). There was no observed effect on other lymphocyte subsets including T, B and Treg cells, or monocytes.
Conclusion In single doses up to 10 mg/kg, CALY-002 was well tolerated and demonstrated PK characteristics typical for an IgG monoclonal antibody. In contrast to an earlier report on effects of IL-15 blockade on peripheral blood NK cell numbers in humans, CALY-002 treatment did result in decreased NK cells numbers. The observed pharmacological activity at and above 0.3 mg/kg is consistent with the neutralization of the known homeostatic and anti-apoptotic effects of IL-15 on NK cells as also reported for JAK inhibitors. Multiple dosing for 8 weeks in gluten-challenged subjects with CeD and for 12 weeks in subjects with EoE is ongoing.

<b><u>Mean CALY-002 Plasma Concentration-Time Plots by Dose Cohort</u></b><br /> Figure shows plasma concentration levels of CALY-002 (mean standard deviation, semi-logarithmic scale - Y axis) across time (day - X axis) after single IV administration of CALY-002. Each symbol represents a different dose group, as indicated.

Mean CALY-002 Plasma Concentration-Time Plots by Dose Cohort
Figure shows plasma concentration levels of CALY-002 (mean standard deviation, semi-logarithmic scale - Y axis) across time (day - X axis) after single IV administration of CALY-002. Each symbol represents a different dose group, as indicated.

<b><u>Percent Change from Baseline of Blood NK Cell Counts upon a Single IV Dose of CALY-002</u></b><br /> Figure shows heatmap representation of the mean percent change from baseline (color code on the right side of the figure) of blood NK cells determined as CD3-CD16+CD56+ cells by flow cytometry, after single IV injection of CALY-002. The X axis represents placebo or each CALY-002 dose group. The Y axis represents study days and end of study (EOS) samples.

Percent Change from Baseline of Blood NK Cell Counts upon a Single IV Dose of CALY-002
Figure shows heatmap representation of the mean percent change from baseline (color code on the right side of the figure) of blood NK cells determined as CD3-CD16+CD56+ cells by flow cytometry, after single IV injection of CALY-002. The X axis represents placebo or each CALY-002 dose group. The Y axis represents study days and end of study (EOS) samples.

Background: Celiac disease (CeD) is a common disorder. The global prevalence of CeD is about 1% (0.5-2.5%), with some variations across different countries. However, the total population-based prevalence in most countries is not known. Studies in Western populations point to a high ratio of undiagnosed to diagnosed CeD cases (4:1 to 7:1) - a phenomenon known as “the coeliac iceberg”. The aim of this study was to settle the total prevalence of CeD, including the undiagnosed cases, in an adult general population.

Methods: The study was based on the fourth Trøndelag health study (HUNT4), which took place from August 2017 to February 2019, including 56,042 adult (>20 years of age) inhabitants in Trøndelag country, Norway (54% response rate). Serum samples from 54,566 participants in HUNT4 were analyzed with a newly developed serological assay for simultaneously measurements of IgA and IgG anti transglutaminase 2 (anti-TG2) at Oslo University Hospital from June 2019 to October 2021. To validate the serological results, all seropositive participants were invited to clinical examination with upper endoscopy and small bowel biopsies from the horizontal part and bulb at Levanger Hospital. The diagnosis of CD was confirmed by histopathological findings corresponding to Marsh grade 3. To identify the seronegative participants with a previous CeD diagnosis, the Norwegian Patient Register and the participants medical records were searched and the diagnose verified.

Results: A total of 1049 individuals were seropositive, corresponding to a rate of 1.9% (1049/54,566). By March 2022, 750 seropositive individuals have been invited to the clinical examination and 686 of these have attended endoscopy with biopsies. Among these attendees, a new previously unknown CeD diagnosis was confirmed in 350 individuals. In addition, 51 seropositive individuals had already a CeD diagnosis. From the register and medical record searches, 374 previously known and seronegative CeD cases were identified. In total, 775 individuals with CeD have been identified so far, corresponding to a total populations-based prevalence of CeD of 1.4% (774/56,042). The ratio of previously diagnosed to previously undiagnosed CeD cases was 1.2 (425/350).

Conclusion: The total population-based biopsy confirmed prevalence of CeD in an adult general population in Norway, including both previously unknown and known CeD cases, was higher than the average global figures. The amount of newly diagnosed cases was high and shows that CeD is still an underdiagnosed condition, but the ratio was not as high as reported elsewhere.

Background
Celiac disease (CeD) is an immune-mediated enteropathy associated with HLA-DQ2/DQ8. Recent evidence suggests HLA-DQ7 might also trigger an immune response to gluten in some people, presenting clinically as CeD. To investigate this, we identified individuals with celiac haplotyping who were HLA-DQ2^-/DQ8^-and compared the DQ7⁺ and DQ7^- subsets.

Methods
We conducted a retrospective review of HLA tests performed as part of the evaluation for CeD at our institution from 1/2016 to 06/2022. Patients without a DQ2 or DQ8 allele were included, and those with HLA-DQ7: DQB1*0301/0304 were compared to those without. Demographics, symptoms, serology (anti-tTG IgA, anti-DGP IgA/IgG), duodenal histology, and response to a gluten-free diet (GFD) were extracted from the medical record. The subgroup with clinical features favoring a diagnosis of CeD (i.e., elevated serology and/or duodenal villous atrophy (VA)) was classified as suspected CeD. Those with other causes of seronegative enteropathy (i.e., small intestinal bacterial overgrowth, Crohn’s disease, HIV, medication-induced, etc.) were identified and excluded. Relative risk (RR) of CeD among HLA-DQ7 individuals was calculated.

Results
We identified 1,120 patients with HLA testing performed due to suspicion of CeD; 433 (39%) were DQ2.5, DQ2.2, and DQ8 negative. Of these, 216 (50%) were DQ7⁺ (n=172, 80% female). One-third of DQ7⁺ patients (n=70/216) reported starting a GFD before their CeD diagnostic testing. Nevertheless, 18/216 (8%) displayed findings suggestive of CeD; 7 (3%) had positive serology only (tTG n=2, DGP n=7), 9 (4%) had seronegative VA, and 2 (11%) had both positive serology and VA (Table 1). Additionally, 8/18 (44%) reported symptom improvement after following a GFD.

There were 217 (19%) non-DQ2, DQ8, or DQ7 patients. Of these, 6 (3%) had positive serology (Table 1). Three seronegative patients had VA. However, they had coexisting disorders (common variable immunodeficiency, autoimmune enteropathy+ lymphocytic colitis, and collagenous colitis not responsive to GFD), which could account for the histological findings; they were excluded from the suspected CeD cohort. Only 1/6 reported clinical response to a GFD. Overall, HLA DQ2^-/DQ8^-/DQ7^- subjects displayed fewer findings compatible with CeD than HLA DQ7.

Among HLA-DQ2^-DQ8^-patients, HLA-DQ7 conferred a three-fold increased risk of having signs of CeD compared to the non-HLA DQ2/DQ8/DQ7 cohort [RR 3.01(1.22,7.45 95%CI)].

Conclusion
Our data support an association between HLA-DQ7 and CeD in a cohort of non-HLA DQ2/DQ8 patients. While further investigations to determine whether HLA-DQ7 is a predisposing factor are warranted, awareness of this possible association might lead to prompt identification and management of CeD cases that might otherwise be missed, as no workup for CeD is currently recommended in patients found to be HLA-DQ2^-/DQ8^-.

<b>Table 1. Clinical presentation among patients with characteristics of CeD within a non-HLA DQ2/DQ8 cohort </b>

Table 1. Clinical presentation among patients with characteristics of CeD within a non-HLA DQ2/DQ8 cohort

Background: Celiac disease (CeD) is a chronic inflammatory disorder triggered and maintained by ingestion of gluten and which has no approved therapeutic treatment. KAN-101 is an investigational therapy composed of a liver-targeting glycosylation signature conjugated to a gliadin-derived peptide designed to induce tolerance as a therapeutic approach to the treatment of CeD. Previously KAN-101 was shown to be safe and tolerated and demonstrated induction of functional tolerance of gliadin-specific T cells in CeD patients in the Ph1 clinical trial.
Aim: The Assessment of KAN-101 in CeD (ACeD) study evaluated PK and immune biomarker responses following a GC to assess induction of immune tolerance in CeD patients who received multiple doses of KAN-101 or placebo (NCT04248855).
Methods: DQ2.5+ CeD patients on a gluten free diet were randomized 3:1 to receive three intravenous (IV) infusions of KAN-101 at 0.15, 0.3, or 0.6 mg/kg or placebo on days 1, 4 and 7, followed by a 3-day oral GC (9 grams/day) starting on day 15. Plasma samples were collected on day 1 and 7 for pharmacokinetic analysis of KAN-101. PBMC samples collected pre-dose, day 15 pre-GC and day 21 were analyzed by flow cytometry for HLA-DQ2.5 α-I and α-II gliadin-tetramer positive T cells and surface phenotype markers. Plasma samples collected pre- and post-GC were analyzed for cytokines using multiplex immunoassays.
Results: Following IV infusion, KAN-101 was cleared rapidly from systemic circulation within approximately 6 hours across all dose levels. KAN-101 exhibited non-linear PK behavior with increasing dose, consistent with target-mediated drug disposition. Following KAN-101 administration, an increase in tetramer positive (Tet+) CD4+ T cell populations was observed on day 15 prior to GC in KAN-101-treated patients (Fig 1), however, the Tet+ T cells exhibited a surface phenotype consistent with T cell anergy, including programmed cell-death 1 (PD-1, or CD279) and lymphocyte activation gene 3 (LAG-3, or CD223) and restimulation of samples did not result in activation. Placebo patients experienced an increase in Tet+ cells following GC (day 21) while most KAN-101-treated patients showed no increase. Profiling of plasma samples taken within 6 hours of the GC demonstrate a dose-dependent modulation of cytokines, including IL-2, by KAN-101 but not placebo.
Conclusions: While the clearance of KAN-101 after administration occurs within hours, the pharmacodynamic effects can be observed for at least 3 weeks after the start of treatment. Gliadin-specific T cells are increased following KAN-101 treatment but display an exhausted phenotype, are functionally anergic and do not expand further after GC. In patients receiving placebo, Tet+ CD4+ cells are only observed following GC and respond upon antigen re-stimulation. Furthermore, KAN-101 modulated the plasma cytokine response to GC.

Effect of KAN-101 on HLA-DQ2.5 α-I and α-II gliadin-tetramer positive T cells. PBMC samples were analyzed by flow cytometry prior to dosing (day 1), one week after completion of dosing but prior to GC (day 15) and 6 days after initiation of GC (day 21). Data from the 0.6mg/kg cohort are unavailable due to technical issues with the assay.

Effect of KAN-101 on HLA-DQ2.5 α-I and α-II gliadin-tetramer positive T cells. PBMC samples were analyzed by flow cytometry prior to dosing (day 1), one week after completion of dosing but prior to GC (day 15) and 6 days after initiation of GC (day 21). Data from the 0.6mg/kg cohort are unavailable due to technical issues with the assay.

Background:
Celiac disease is an autoimmune condition occurring in individuals with hypersensitive reactions towards wheat gluten. The inflammatory reaction damages the small bowel mucosa, which manifests histologically as blunting of the normally elongated villi. Histopathological examination of small bowel biopsies is considered the “gold standard” for assessment of villous morphology and length. However, this traditional method is subjected to significant undersampling and interpretative errors. Here, we present a comprehensive 3D tissue analysis of small bowel biopsies using multi-resolution open-top light-sheet (OTLS) microscopy to image the entire biopsies and quantify the 3D morphology of individual villi.
Methods: Human small bowel biopsy samples were obtained fresh from surgical resection specimens and fixed in 4% PFA for 24-48 hours. Samples were stained with the fluorescent nucleic acid stain ToPro3 and optically cleared using an iDISCO-based protocol. Regions of biopsies >1 mm3 in volume were imaged with an OTLS microscope. Villi were segmented across the complete 3D dataset using Aivia (Leica Microsystems). Python and R were used to skeletonise villous structures and generate quantitative measurements.
Results: Entire human small bowel biopsy samples were stained, cleared, and imaged without the need to section the tissue. Overall, this 3D approach visualized >99% more tissue than a single 5 micron section mounted on a glass slide. We have developed quantitative approaches to measure 3D morphology and length of individual villi in human small bowel samples. 3D villi were segmented and processed using customized python scripts to skeletonise the structures to enable measurement of length in each villous structure in the specimen. 3D measurements of villi varied substantially throughout the biopsy ranging from 363.01 μm to 540.61 μm (mean = 456.14 μm, SD ± 52.17 μm).
Conclusion: We analyzed human small bowel biopsies in 3D using our OTLS microscope and quantification techniques to provide a comprehensive assessment of villous morphology. This novel method enables us to identify and quantify pertinent parameters such as the average length of each villous structure across the whole biopsy. Quantitative analysis demonstrated that villous heights show substantial variability throughout the sample, which is magnified by variability caused by the plane of sectioning and subjective interpretation. These 3D methods may enable us to develop a more accurate scoring system which correlates better with clinical celiac disease severity. Imaging and analysis of additional samples, including samples from celiac patients, are being planned to quantify the benefits of the 3D approach.

<b>We calculated the length (um) of each villous structure (n=21) in a small bowel biopsy. Min = 363.01 um, max = 540.61 um, mean = 465.14 um, with a standard deviation = 52.17 um. </b>

We calculated the length (um) of each villous structure (n=21) in a small bowel biopsy. Min = 363.01 um, max = 540.61 um, mean = 465.14 um, with a standard deviation = 52.17 um.

Introduction
Celiac disease (CD) is a chronic small intestinal immune-mediated enteropathy with many different manifestations. There are studies suggesting a link between CD and adverse maternal and neonatal outcomes. We aimed to evaluate the effect of CD on pregnancy outcomes using a large database.

Methods
We conducted a retrospective cohort study using the National Inpatient Sample (NIS) of all pregnant women who delivered between the years 2015-2019. Using the 10th version of International Classification of Diseases codes (ICD-10), we identified pregnant patients who had CD and those who did not. Univariate logistic regression was used to estimate the maternal and fetal outcomes.

Results
Of the 12,039,222 deliveries in the databse, there were 10,555 births in women with CD, with an overall prevalence of 0.09%. Baseline characteristics and demographics are listed in Table 1. Pregnant women with CD were more likely to be white and older compared to pregnant women without CD. Pregnant women with CD were at higher risk for numerous pregnancy and delivery related complications (Table 2), although maternal mortality did not differ [0 (0%) vs 895 (<0.1%), p=0.7]. Celiac disease moms were much less likely to have a full-term uncomplicated delivery (OR 0.15; 95% CI, 0.09-0.26) compared to moms without CD. Length of stay was higher in moms with celiac [2.65 (2.41%) vs 3 (3.34%), p <0.001] compared to moms without celiac disease . Babies of celiac moms were more likely to be small for gestational age (OR 1.26; 95% CI 1.01-1.57) and experience distress (OR 1.29; 95% CI, 1.17-1.42) during delivery. There were no significant differences in stillbirth rates in babies of celiac moms when compared to babies of non-celiac moms [15 (<0.1%) vs 11,455 (<0.1%), p=0.5]

Discussion
In our study, we found that pregnant women with CD were at increased risk of preeclampsia, placental diseases, obstetric complications, and small for gestational age babies. Undiagnosed CD should be considered in patients with these recurrent complications of pregnancy and delivery. Clinicians should have a low threshold to test for celiac disease in these patients. One of the limitations of our study was that the status of adherence to gluten-free diet was unknown; this is important since prior research shows an association between poor adherence to gluten-free diet and worse pregnancy outcomes. Further studies are needed to investigate the underlying mechanisms behind these associations.

Table 1: Baseline characteristics

Table 2: Effects of CD on maternal and neonatal outcomes.

Table 2: Effects of CD on maternal and neonatal outcomes.

Claire Jansson-Knodell

Appearances