Oral Presentation International Celiac Disease Symposium 2026

A Duodenal Pre Gluten-Challenge Interferon-γ/Antigen-Presentation State Predicts Susceptibility to Gluten-Induced Mucosal Injury in Celiac Disease (143326)

Aline Pesi 1 , Valeriia Dotsenko 2 , Ralf Mohrbacher 3 , Markku Mäki 4 , Keijo Viiri 4 5 , Detlef Schuppan 1 6
  1. University Medicine Mainz, Mainz, RHEINLAND-PFALZ, Germany
  2. Department of Virology, University of Helsinki, Helsinki, Finland
  3. Dr. Falk Pharma GmbH, Freiburg, Germany
  4. Celiac Disease Research Center, Faculty of Medicine and Health Technology, Tampere University and Tampere University Hospital, Tampere, Finland
  5. Faculty of Biochemistry and Molecular Medicine, University of Oulu, Oulu, Finland
  6. Division of Gastroenterology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, USA

Background: Patients with well-controlled celiac disease (CeD) show markedly heterogeneous mucosal injury during standardized gluten exposure. We asked whether duodenal mucosa already carries a molecular signature that predicts subsequent susceptibility to gluten-induced damage.

Methods: We performed a post hoc longitudinal analysis of paired duodenal biopsies from placebo-treated CEC-3 clinical study participants undergoing 6 weeks of gluten challenge (3 g per day, n=24; 1,2). Baseline duodenal bulk RNA-seq obtained on a gluten-free diet was modeled in relation to post-challenge villus height-to-crypt depth ratio (VH:CrD), adjusting for baseline VH:CrD, using continuous limma-voom modeling. An immune/regulatory candidate space of >500 curated genes was interrogated, and convergence was assessed with the externally defined HALLMARK_INTERFERON_GAMMA_RESPONSE set. As an orthogonal sensitivity analysis, baseline expression within this Hallmark set was tested against individual histologic change (ΔVH:CrD=post−pre).

Results: Baseline transcriptomic variation was associated with subsequent mucosal injury despite clinical control on a gluten-free diet. In the primary baseline-adjusted analysis, TAP1, PSMB9, HLA-B, and IRF9 met the FDR<0.10 threshold for association with lower post-challenge VH:CrD, within a broader IFN-γ/antigen-processing program also comprising HLA-A, IRF1, BST2, B2M, IFI27, PARP14, and CCL5. The ΔVH:CrD sensitivity analysis reproduced this directionally coherent program. Strongest deterioration-associated effects were observed for HLA-B (β=−0.35), IRF9 (β=−0.33), and PSMB9 (β=−0.32), with concordant recurrence of IFI27, HLA-A, BST2, TAP1, B2M, IRF1, PARP14, and CCL5. Opposing positive associations included TRIM14, CFB, and MVP, suggesting a distinct resilience-associated state.

Conclusions: Clinically controlled CeD mucosa retains pre-existing molecular heterogeneity that marks subsequent vulnerability to gluten. Convergence across baseline-adjusted and change-score analyses identifies an IFN-γ/antigen-presentation axis as a robust candidate marker of mucosal susceptibility and a potential stratification framework for gluten-challenge studies.

  1. Schuppan D, Mäki M, Lundin KEA, Isola J, Friesing-Sosnik T, Taavela J, Popp A, Koskenpato J, Langhorst J, Hovde Ø, Lähdeaho ML, Fusco S, Schumann M, Török HP, Kupcinskas J, Zopf Y, Lohse AW, Scheinin M, Kull K, Biedermann L, Byrnes V, Stallmach A, Jahnsen J, Zeitz J, Mohrbacher R, Greinwald R; CEC-3 Trial Group. A Randomized Trial of a Transglutaminase 2 Inhibitor for Celiac Disease. N Engl J Med. 2021 Jul 1;385(1):35-45. doi: 10.1056/NEJMoa2032441. PMID: 34192430.
  2. Dotsenko V, Tewes B, Hils M, Pasternack R, Isola J, Taavela J, Popp A, Sarin J, Huhtala H, Hiltunen P, Zimmermann T, Mohrbacher R, Greinwald R, Lundin KEA, Schuppan D, Mäki M, Viiri K; CEC-3 Investigators. Transcriptomic analysis of intestine following administration of a transglutaminase 2 inhibitor to prevent gluten-induced intestinal damage in celiac disease. Nat Immunol. 2024 Jul;25(7):1218-1230. doi: 10.1038/s41590-024-01867-0. Epub 2024 Jun 24. PMID: 38914866; PMCID: PMC11224021.