A set of serum proteomic biomarkers differentiates celiac children from age and human leukocyte antigen‐matched healthy controls
Journal of Pediatric Gastroenterology and Nutrition, 2025
Mandile R., di Tolla M., Carpinelli M., Samsom J., De Cegli R., Barone M., Troncone R., Formisano P., Greco L., Auricchio R.
| Disease area | Application area | Sample type | Products |
|---|---|---|---|
Immunological & Inflammatory Diseases | Patient Stratification | Serum |  Olink Target 96 |
Abstract
Objective
The latest European Society of Gastroenterology Hepatology and Nutrition (ESPGHAN) criteria for celiac disease (CD) diagnosis reduced the requirement for a small intestinal biopsy but still, for most of the cases a small intestinal biopsy is required for a safe diagnosis: hence the attempt to identify serum biomarkers that could replace, in most of these latter cases, the requirement of the biopsy (what is called a “liquid biopsy”). The aim of this study is to identify a set of serum biomarkers able to differentiate celiac patients from age and human leukocyte antigen (HLA)‐matched healthy controls.
Methods
Relative concentration of 92 inflammation‐linked proteins was examined in the sera of 50 children with CD compared with 50 HLA DQ2/8 and age‐matched healthy controls born in genetically at‐risk families, using proximity extension immunoassay technology (Olink Proteomics®) with the ProSeek Multiplex Inflammation panel.
Results
Three different multivariate analysis (Univariate and multivariate distribution analysis, random forest classification and linear discriminant analysis) localized a cluster of seven molecules (CASP8, CXCL9, NT‐3, SIRT2, STAMBP, ST1A1, and TNFSF14) with a remarkable diagnostic potential, able to differentiate around 90% (95% confidence interval [CI]; 0.7–0.99) of CD patients from controls.
Conclusion
Patients with CD, compared to age‐ and HLA‐matched healthy controls, show in their sera an increased expression of inflammatory molecules, involved in NF‐κB cytokine signaling, cell apoptosis, and crypt proliferation pathways. A set of seven of these proteins can differentiate cases from controls with an accuracy higher than 90%. The implementation of this approach in clinical setting could in future facilitate a noninvasive and individualized approach for CD diagnosis.