Behçet disease (BD) is a systemic inflammatory disease of unknown etiology. It is characterized by recurrent oral aphthous and genital ulcers, uveitis, and skin lesions. Vascular injury, hyper-coagulability, hyper-function of neutrophils, and autoimmune response are pathological findings of the disease.1,2

BD is a complex clinical entity where genetic and environmental factors influence disease’ susceptibility. Several previous studies showed that immunogenetics factors (mainly HLA class A and B) are present in the development of BD.3

HLA-B*51 allele is the most strongly associated genetic risk factor linked with BD.1,2,4–6 This association is present along different geographical regions and ethnic groups.7 So far, two studies from Mexico are the only evidence on the association between the HLA-B51 allele and BD in Latin American countries; however, the results are contradictory.8,9

The aim of this study was to assess the association between BD and the HLA-B*51 allele in Argentinean patients.

The most significant finding in our study is a strong association between HLA-B*51 allele and Behçet disease in Argentinean patients. This association is well known in patients belonging to Asian and Eurasian populations from Japan to the Middle East.7 However, there is less data on Caucasian patients with BD from other regions, such as North Western Europe and North America.12–16 Furthermore, there are two Latin American studies focusing on the association between HLA-B*5 allele and BD. Lavalle, et al. found that the HLA-B*5 allele was more frequent in Mexican Mestizo patients with BD than ethnically matched controls (70% vs. 31%, p<0.05).8 On the other hand, Soto Vega et al., found no association between HLA-B*51 allele and BD among Mexican Mestizo patients and controls [9% vs. 5%, OR=1.88 (95%CI=0.58–5.93, p=0.24)].9Table 3 shows the HLA-B*51 allele frequencies in our BD patients and the Mexican Mestizo patients from Soto Vega study.

In our study the frequency of the HLA-B*51 allele in BD patients was 38.2% as compared to 14.2% in controls. Strikingly, our data are quite similar to the pooled prevalence estimated for HLA-B*51 allele in the Northern/Eastern regions of Europe (BD=39%; 95%CI=28.2–51.1 vs. controls=11.2%; 95%CI=8.1–15.3) and North America (BD=34.2%; 95%CI=6.0–80.8 vs. controls=18%; 95%CI=7.6–37.1), but lower than the pooled prevalence estimated in the Southern Europe (BD=60.6%; 95% CI=51.9–68.7 vs. controls=16.8%; 95% CI=13.3–21.0) where the biggest proportion of Argentine immigration comes from.7 This finding could be partially explained by the fact that our studied population was predominantly Mestizo, a mixture of European (Caucasian) and native ancestors (Amerindians). Two studies showed low prevalence of HLA-B*51 in Amerindian tribes of South America (mainly Argentine natives). Fernandez-Viña et al.,17 analyzed the prevalence of HLA antigens in a Toba tribe, finding an HLA-B*51 allele frequency of 3%. Parham et al.,18 studied the prevalence of HLA antigen in a Guarani tribe, finding an HLA-B*51 allele frequency of 6.6%. Therefore, we assume that the admixture between European and South American Amerindian could explain that mixed populations have more HLA class I allele diversity than each ethnic group separately.

Our study has limitations. Classical case-control studies designed to evaluate gene association are exposed to bias due to the existence of population substructures that cannot be certainly determined. It is possible that our cases are not representative of all BD patients from an Argentinean population as a consequence of a selection bias. Finally, as we did not have the sex and ethnicity data of the controls, we could not match both groups for these variables.

The HLA-B*51 allele is strongly associated with BD in Argentinean patients. Individuals carrying HLA B*51 allele have a 4-fold susceptibility risk increase of BD compared to non-carrier individuals of the allele. Our finding is consistent with previous studies indicating that the HLA-B*51 allele is an important susceptibility gene in BD regardless the geographical region and ethnicity.

This research was funded by an Argentinean Rheumatology Society grant.

The authors have no conflicts of interest to declare.