Review of genetic risk haplotypes in Celiac disease across diverse ancestries

Photo by Google DeepMind / Unsplash

medRxiv Published April 24, 2026 Medically reviewed April 24, 2026 Study authors: Long, X.; Karnati, H.; Ying, W.; Touma, M.-J.; Smith, I.; Lewis, S. K.; Xing, C.; Burstein, E.; Gree… DOI ↗ By Dr. Amelia Tan, PhD · Internal Medicine & Chronic Disease

Key Takeaway

Note higher genetic risk allele enrichment in Celiac disease across diverse ancestries.

This observational review examines genetic risk haplotypes, specifically HLA-B8, HLA-DQ2.5, and DQB1*02:01, within the All of Us Research Program. The study population includes 3,481 individuals with Celiac disease and 5 sex-, age-, and ancestry-matched controls per case. The analysis focuses on enrichment patterns and linkage disequilibrium rather than establishing causal mechanisms.

The authors report that enrichment in the DQB1*02:01 allele is strongest in Europeans (32.3% vs. 11.6%), followed by admixed Americans (18.5% vs. 8.1%) and Africans (15.7% vs. 8.1%). Additionally, 72.3% of Europeans and 42.3% of admixed Americans carry HLA-B8 alongside DQ2.5, with lower prevalence observed in African participants.

Polygenic risk scores demonstrated 86% accuracy in distinguishing seropositive Celiac disease from controls. This predictive accuracy increased to 92% when clinical risk factors were incorporated. Identification of patients with tTG-IgA levels greater than 10 IU/mL yielded 93% detection rates.

The authors note limited data availability for Hispanic and Black populations as a key constraint. Because this is an observational study, associations should not be interpreted as causation. Practice relevance regarding these genetic markers remains to be defined by further research.

Study Details

EvidenceLevel 5

PublishedApr 2026

View Original Abstract ↓

Abstract Background: Most genetic studies on celiac disease (CeD) have focused on individuals of European descent. Limited data are available for the Hispanic and black populations. Methods: We analyzed whole-genome sequencing data, electronic health records (EHR), and laboratory results from the All of Us Research Program. We identified 3,481 individuals with CeD through EHR, self-reporting, or both. Of these, 2,899 carried one of the four well-established risk haplotypes, including 262 of admixed American (89% Hispanic) and 108 of African (70% black) ancestry. Five sex-, age-, and ancestry-matched controls per case were selected for the assessment of genetic and clinical risk factors. Results: An enrichment in the DQB1*02:01 allele was observed in CeD patients across all ancestries, with the strongest association in Europeans (32.3% vs. 11.6%), followed by Americans (18.5% vs. 8.1%) and Africans (15.7% vs. 8.1%). Among individuals carrying the DQ2.5 (DQA1*05:01-DQB1*02:01 haplotype), HLA-B8 was present in 72.3% of Europeans, 42.3% of Admixed Americans, and lower in Africans. This linkage disequilibrium was higher in CeD patients than in controls across all three ancestries. A polygenic risk score distinguished seropositive CeD from controls with 86% accuracy. Incorporating clinical risk factors, including family history, hypothyroidism, diarrhea, vitamin D deficiency, and anemia, increased predictive accuracy to 92%. The model identified 93% of CeD patients with tTG-IgA levels greater than 10 IU/mL. Conclusion: Linkage between HLA-B8 and DQ2.5 differs significantly among individuals of European, admixed American, and African ancestry, contributing to ancestry-dependent genetic risk for CeD.