Gluten found to trigger immune response at very low doses

Very small amounts of gluten can trigger measurable immune activation in patients with celiac disease, even at doses below current international food-labeling thresholds, according to the results of a randomized controlled trial published in Gastroenterology.

In a double-blind, placebo-controlled dose-response study of 51 adults with biopsy-confirmed celiac disease on a gluten-free diet, investigators in Australia found that doses as low as 3 mg induced immune responses in some patients. Symptoms did not reliably distinguish gluten exposure from placebo.

For the trial, researchers evaluated single-dose gluten challenges ranging from 1,000 mg down to 1 mg. Participants at a single tertiary center in Brisbane underwent three challenges each, spaced 4 weeks apart, with immune activation defined as at least a twofold increase in serum interleukin-2 (IL-2) within 6 hours, reported A. James M. Daveson, MBBS, Director of the Celiac Disease and Immune Health Research Program at Wesley Research Institute in Auchenflower, Queensland, Australia, and colleagues.

Dose-dependent immune activation

Gluten exposure produced a clear dose-response relationship in IL-2 levels, peaking at 4 hours. At higher doses, most patients showed immune activation: 83% responded at 610 mg and 56% at 1,000 mg. However, responses were also observed at lower doses, including 36% at 90 mg, 27% at 8 mg, and 17% at both 13 mg and 3 mg.

When the researchers used interval-censored modeling, the estimated dose triggering immune activation in 10% of patients (eliciting dose, ED₁₀) was 2.4 mg, and in 5% (ED₀₅) was 0.8 mg. The median eliciting dose (ED₅₀) was 111 mg. These findings suggest that immune activation can occur at gluten exposures below the 20 parts-per-million thresholds used in the US and Europe, which corresponds to roughly 3–5 mg per typical serving.

Symptoms do not track with exposure

Despite measurable immune activation, patient-reported symptoms were not dose dependent and did not differ from placebo at any gluten dose. Symptom severity scores increased modestly after challenges overall, but there was no association between gluten dose and symptoms using multiple validated tools, including the Global Symptom Survey and Patient Gastrointestinal Symptoms survey.

This disconnect was evident even at higher doses, indicating that patients cannot reliably detect low-level gluten exposure based on symptoms alone.

Practical implications

For gastroenterologists, the findings highlight that current “gluten-free” labeling standards may not fully prevent immune activation in some patients with celiac disease. The study also reinforces that symptom-based assessment is insufficient for monitoring dietary adherence or inadvertent exposure, particularly at low doses. Objective biomarkers such as IL-2 may offer a more sensitive tool for detecting immune activation in both clinical practice and trials.

The prospective study enrolled adults aged 25–75 years who had maintained a gluten-free diet for more than two years. Participants completed 153 total challenges across four sequential cohorts, with adaptive dose de-escalation and rigorous placebo control.

The researchers noted limitations of the study, including its single-center design, small sample size, and the exclusion of children and certain genetic subgroups, including patients homozygous for HLA-DQ2.5. It also assessed acute immune responses rather than long-term outcomes such as mucosal injury.

The study was funded by Wesley Research Institute and Coeliac Australia. Dr. Daveson and several coauthors reported consulting roles, research funding, or patents related to celiac disease therapies.

_________________________________________________________________________________

GI & Hepatology News invited Dr. Daveson to elaborate on the study findings.

Why does this study matter? Global food labeling regulations vary around the world and are largely based on a study from 2007, where 50 mg of gluten a day resulted in a deterioration in people with celiac disease, though firm conclusions could not be drawn about doses of gluten below this level. Our study was the first to look at whether small amounts of gluten result in immune activation at very low doses — even below 10 mg and 50 mg. What we found is that immune activation was seen down to 3 mg.

When you had all the data in front of you, was there a finding, or perhaps more than one, that surprised you? At higher doses of gluten (3,000 mg to 6,000 mg), people with celiac disease on a strict gluten-free diet can develop pronounced nausea and vomiting when exposed n, making it easier for them to recognize that they have ingested gluten. What was surprising in this study was that there was no difference in symptoms between the low doses and placebo, suggesting that people with celiac disease may not be able to recognize when they are inadvertently exposed to very small amounts of gluten. This is important for people with celiac disease who struggle to normalize their small bowel histology when trying to maintain a strict gluten-free diet. It may also have implications for the current FDA approval process of novel therapies for celiac disease, which currently require not only a demonstrable improvement in small bowel histology, though also an improvement in symptoms.

How might the findings influence clinical practice? They may influence global food labeling regulations regarding what is defined as “gluten-free” by regulatory bodies. It also answers a common question among people with celiac disease: what happens when they are inadvertently exposed to small amounts of gluten. It may also help explain why some studies suggest that patients take many years to enter mucosal remission despite adhering to a strict gluten-free diet.

Is there anything else you’d like to say about this work? Global food labeling is based on a concentration which can be difficult to understand. We were also able to calculate an eliciting dose, (EDp), where p indicates that dose is unsafe (in terms of IL-2 release) for p% of the population. An ED₀₅ would be safe in terms of IL-2 release for 95% of the population. In the United States and Europe, consuming 150–250 g of “gluten-free” food under the 20 ppm (or 20 mg/kg) standard may result in 3-5 mg of gluten exposure, exceeding the ED₁₀ (2.4 mg) estimated from our acute IL‑2 endpoint. A comparable intake under the Australian and New Zealand “no detectable gluten” requirement (that sits at a threshold of ~3 ppm or 3 mg/kg) would be ~0.45-0.75 mg, below the ED₀₅ (0.8 mg) estimate and therefore predicted to be tolerated by most individuals with respect to acute immune activation.