Editorial: Endomysial Antibodies for a No-Biopsy Diagnosis of Coeliac Disease—The Jury Is Still Out

Abstract

Studies in children and adults have shown that immunoglobulin (Ig)A anti-tissue transglutaminase (tTG) antibody levels ≥ 10× upper limit of normal are highly specific for a diagnosis of coeliac disease and may obviate the need for histological confirmation in approximately 30% of patients [1, 2]. Given that many patients prefer to avoid endoscopy [3], the ‘no-biopsy’ approach to diagnosis is an area of interest with the potential to improve patient experience and reduce associated healthcare costs.

In this issue of Alimentary Pharmacology & Therapeutics, Maimaris and colleagues present the findings of a retrospective/prospective study evaluating the accuracy of endomysial antibodies (EMA) in combination with clinical risk stratification to support a no-biopsy diagnosis of adult coeliac disease [4]. A positive EMA test had a specificity and a positive predictive value (PPV) of 100% for a diagnosis of coeliac disease, suggesting that this may be a suitable alternative to IgA-tTG testing in this no-biopsy approach. However, EMAs are typically identified in serum using indirect immunofluorescence, which is more costly and labour-intensive than conventional tTG-based immunoassays; this methodology is also subject to inter-observer variability. Importantly, the EMA test also lacks a quantitative readout, which is a central tenet of the no-biopsy approach, as low titre IgA-tTG levels (i.e.,1-5× upper limit of normal) confer less specificity for the duodenal inflammation typical of coeliac disease than high titre levels (i.e., ≥ 10× upper limit of normal) [5], despite both being identified as a positive result.

Notably, all patients with a positive EMA result were diagnosed with coeliac disease in this study. However, patients with potential coeliac disease—defined by positive serology without duodenal villous atrophy—were included within this group. If these individuals were re-classified as false positives, the specificity would fall from 100% to 93.7%, and the PPV would fall from 100% to 87.4%, based on a disease prevalence/pre-test probability of 31%. On the one hand, does this matter if the end goal is a gluten-free diet? On the other hand, serology will normalise in up to a third of patients with potential coeliac disease despite ongoing gluten exposure [6]. Therefore, it is important not to blur the lines between the two conditions. Indeed, this is one of the limitations of the no-biopsy approach, which should be balanced against the shortcomings of the serology-biopsy pathway before either is considered for diagnosis.

Beyond this, the study provides reassuring data on the safety of avoiding upper gastrointestinal endoscopy in patients with positive coeliac serology, as no major concomitant pathology was identified among those with a positive EMA regardless of their age or the presence of alarm symptoms. Conversely, most non-coeliac enteropathies were diagnosed in those aged ≥ 45 years with alarm symptoms and negative coeliac serology, underscoring the importance of thoroughly investigating patients in this setting.

In conclusion, the accuracy and applicability of EMA testing to safely replace diagnostic biopsies remain uncertain and future work evaluating EMAs within combined serological strategies is ongoing. Ultimately, clinicians should consider test performance, clinical context, and patient preferences when investigating for coeliac disease to ensure informed and appropriate patient-centred care.