Biomarkers Offer Early Detection of Active Celiac Disease in Pre-Disposed Children

Biomarkers that Predict Which Pre-Disposed Children Will Develop Celiac Disease wpIn children with asymptomatic potential celiac disease (PCD), a panel of seven serum proteomic biomarkers can predict which individuals will go on to develop villous atrophy (VA), according to investigators.

  • Will Pass, GI Hepatology News  1

Given that PCD patients present with positive serology and intact duodenal architecture, these findings may provide a much-needed tool for identifying patients who are more likely to benefit from early dietary interventions, lead author Renata Auricchio, MD, PhD, of the University of Naples Federico II, Italy, and colleagues reported.

  • “PCD offers the unique opportunity to observe the progression of gluten-induced tissue damage in celiac disease,” the investigators wrote in Gastroenterology. “These patients recognize gluten and produce specific autoantibodies, but have not developed intestinal damage.”

The Study

  • 31 children with asymptomatic PCD, positive serology and intact duodenal architecture who were eating a gluten-containing diet.
  • Serum samples from each child were analyzed for the relative abundance of 92 inflammation-linked proteins using a proximity extension immunoassay.
  • Statistical analyses, including partial least squares discriminant and linear discriminant analyses, were then applied to identify which proteins were associated with the development of villous atrophy (VA.0

Results

  • After a mean follow-up period of 5.85 years, 14 participants developed VA (ie, celiac disease), while the remaining 17 remained asymptomatic.
  • Panel analysis revealed that specific inflammatory proteins, including interleukin  were able to significantly distinguish between symptomatic and asymptomatic cases in a linear discriminant model.
  • Specifically (IL)–20, IL-2, sirtuin 2 (SIRT2), leukemia inhibitory factor (LIF), IL-22 receptor subunit a1, cystatin D (CST5), IL-17 receptor A, IL-15 receptor subunit a (RA), CUB domain–containing protein 1 (CDCP1), and IL-14, were 1.23- to 1.76-fold higher in children who developed VA. Among these, seven proteins — CDCP1, IL-2, LIF, IL10RA, SIRT2, CST5, and IL-4

This panel of seven proteins achieved a predictive accuracy of 96.8% in identifying children at risk of VA.

Additionally, bioinformatics pathway analysis confirmed that the broader set of proteins is involved in the positive regulation of JAK-STAT signaling (involving IL-22 receptor subunit a1, IL-4, IL-20, IL10RA, LIF, and IL-2), inflammatory responses (IL-4, IL-20, LIF, and IL-2), and processes such as tyrosine phosphorylation, leukocyte differentiation, IgG isotype switching, and protein phosphorylation regulation.

These findings suggest that gluten-induced inflammation may already be active in early stages of the disease, including the initial phases of leukocyte differentiation, according to the investigators.

  • “Over a long follow-up on a gluten-containing diet, only 40% of these patients [who developed celiac disease] progressed to VA.
  • Notably, 25%-30% of children with PCD even stop producing anti–tissue transglutaminase antibodies, and the others keep on producing autoantibodies but preserve a normal intestinal mucosa.
  • Considering these data, the decision to address a patient with PCD on a gluten-free diet at time of diagnosis is quite critical.”

The researchers noted that this new model, with accuracy exceeding 95%, is well suited for routine use because of its practicality and reliability.

  • “Our previous model, based mainly on small intestinal mucosa features, moved a step toward the prediction of outcome but still required a mucosal biopsy, and the accuracy of prediction was not greater than 80%, which is somewhat uncertain for a lifelong clinical decision. In contrast, the present model “appears to be sufficient to immediately suggest a gluten-free diet in children with PCD, who are almost certainly committed to developing VA.”

The investigators called for long-term studies to validate their findings in other cohorts, including adult populations.This study was supported by the TIMID project and Inflammation in Human Early Life: Targeting Impacts on Life Course Health (INITIALISE) by the Horizon Europe Program of the European Union. The investigators disclosed no conflicts of interest.