As the celiac disease field prepares for the arrival of its first approved therapies, a fundamental question remains unresolved: What exactly constitutes a diagnosis of celiac disease?
That question was the focus of a presentation at a virtual policy symposium hosted by the Society for the Study of Celiac Disease (SSCD) and the Celiac Disease Foundation. Christopher Cao, MD, director of the Celiac Disease Program at Mount Sinai Hospital, New York, discussed recommendations as chair of the multidisciplinary SSCD Diagnosis Workgroup, which examined how current diagnostic standards could affect access to future treatments, participation in clinical trials, and insurance coverage.
“The reason that this is such an important and such a timely discussion is because there are many clinical trials for the treatment of celiac disease that are ongoing,” Dr. Cao said. “It is important that we align diagnostics at this time because how we define celiac disease and how we diagnose it impacts which patients may be eligible for medications or even trials in the future.”
For decades, diagnosing celiac disease has followed a fairly straightforward process. Patients with suggestive symptoms or risk factors undergo serologic testing, followed by upper endoscopy with duodenal biopsies to confirm villous atrophy. This approach remains the standard and is reflected in a FDA draft guidance document, which states that “a diagnostic EGD with multiple biopsies is needed to establish a diagnosis of celiac disease,” Dr. Cao noted.
However, the picture has become more complex. Over the past decade, European guidelines for both children and adults have increasingly supported diagnosing celiac disease without a biopsy in selected patients who have very high tissue transglutaminase (tTG)-IgA levels and positive endomysial antibody test results. In contrast, most North American guidelines still recommend confirming the diagnosis with a biopsy.
Most gastroenterologists are familiar with the debate over biopsy-based versus non-biopsy diagnosis. Supporters of biopsy-based diagnosis argue that examining tissue samples provides clear evidence of intestinal damage, creates a baseline for tracking healing over time, and helps rule out other possible conditions. An endoscopy also ensures that patients are evaluated by a gastroenterologist with expertise in celiac disease.
Supporters of non-biopsy diagnosis point to studies showing that when tTG-IgA levels are more than 10 times the upper limit of normal and endomysial antibody tests are also positive, the diagnosis of celiac disease is correct more than 95% of the time. They argue that this approach can help patients avoid the cost, inconvenience, and small risks associated with endoscopy.
“Our role is not to tell you which approach is better,” Dr. Cao said. “It’s not to make the determination of which approach to take, but really to highlight the risks that are associated with the heterogeneity of guidelines that we see today.”
He argued that these concerns will become even more important as new therapies move closer to approval. For example, a child diagnosed with celiac disease at age 8 using European non-biopsy criteria may present later in life to an adult gastroenterologist without any biopsy-based confirmation of the disease.
“Based on heterogenous guidelines, we currently don’t have consensus for what constitutes a formal diagnosis of celiac disease,” Dr. Cao said. “There is a gap between the pediatric and adult population here.”
International differences in diagnostic standards create another challenge. Patients diagnosed under European guidelines may later move to North America, where expectations for confirming celiac disease are different. “Do they still have that same diagnosis here in the US?” Dr. Cao asked.
Insurance coverage could become another area of uncertainty. Depending on which clinical guidelines an insurer follows, patients diagnosed without a biopsy may or may not qualify for future celiac disease therapies.
Another often-overlooked group is patients who were diagnosed years ago but no longer have access to their medical records. “What happens to this population of patients?” he wondered.
For patients who have been following a gluten-free diet for many years, confirming the diagnosis can be challenging. The traditional approach often requires a gluten challenge, followed by repeat blood tests and a biopsy, which many patients are unwilling to undergo.
To address these challenges, the SSCD Diagnosis Workgroup proposed moving away from strict diagnostic categories and instead focusing on the level of certainty behind a diagnosis. Rather than viewing celiac disease as simply present or absent, clinicians could assess how strongly the available evidence supports the diagnosis.
Under the unpublished proposed model, patients with biopsy-confirmed celiac disease and those who meet established non-biopsy diagnostic criteria would be considered to have a high level of diagnostic certainty. Patients with a documented past diagnosis but no available records would be classified as having moderate diagnostic certainty. Those with incomplete evaluations, conflicting test results, or self-reported gluten sensitivity would need further assessment before a diagnosis could be confirmed.
When researchers applied this framework to patients at Mount Sinai who had been diagnosed with celiac disease, only about one-third were classified as having high diagnostic certainty. Many more fell into the moderate-certainty category because, although they reported a previous diagnosis, supporting medical records were no longer available.
New diagnostic tools may eventually help address these challenges. Dr. Cao highlighted HLA testing, gluten challenge protocols, and emerging biomarkers such as interleukin-2 (IL-2) assays as potential ways to clarify uncertain cases without relying solely on traditional diagnostic approaches. Although IL-2 testing is still under investigation, he said it could one day help confirm or support a diagnosis in patients who are unable or unwilling to undergo a gluten challenge.
Dr. Cao reported having no relevant disclosures.