Tissue Antigens ISSN 0001-2815 LETTER TO THE EDITOR Association with HLA-DRB1 in Egyptian and German pemphigus vulgaris patients O. Haase 1,† , R. Alneebari 1,† , M.A. Eldarouti 2 , M. Abd El Hady 2 , D. Dorgham 2 , E. El-Nabarawy 2 , S. B. El Din Mahmoud 2 , H. Mosaad El Sayed 3 , M. Darwish 3 , F. Abbas 3 , S. Salah 3 , Y. Mosaad 3 , F. El-Chennawi 3 , S. Al Mongy 4 , A. M. Abdelaziz 4 , S. Abd El Gaber 4 , M. Hertl 5 , R. Eming 5 , A. Recke 1 , S. Möller 1 , E. Schmidt 1 , D. Zillikens 1 & S. Ibrahim 1 1 Department of Dermatology, University of Luebeck, Luebeck, Germany 2 Department of Dermatology, Cairo University, Cairo, Egypt 3 Clinical Immunology Unit (Clinical Pathology Department), Mansoura University, Mansoura, Egypt 4 Department of Dermatology, Mansoura University, Mansoura, Egypt 5 Department of Dermatology, University of Marburg, Marburg, Germany To the Editor, Evidence collected over the last decades brought up the hypothesis that a strong genetic background is present in pem- phigus vulgaris (PV) (1–4). In this study, we examined for the frst time the HLA-DRB1 locus in Egyptian PV patients and controls for association by low and high resolution sequencing and compared these results to a cohort of German patients and controls. Egyptian PV patients (n = 47; 49% female, mean age of 38.6 ± 13.7 years) were treated at the Universities of Cairo and El Mansoura. Egyptian Blood donors (n = 73; 53% female, mean age 38.7 ± 15.5 years) came from the same region. Simi- larly, 46 German PV patients were treated at the Universities of Lübeck and Marburg (52% females, mean age 56.4 ± 14.7 years and 74 blood donors were collected from the same region (50% female; mean age 43.2 ± 13.1 years). DNA of ethylene- diaminetetraacetic acid (EDTA) blood samples was extracted using the DNeasy ® exraction kit (Qiagen, Hilden, Germany) and stored at -80 ∘ C until use. First, low resolution genotyping (two digits, allelic level) with the polymerase chain reaction sequence specifc primer (SSP) method was used to reveal associations of the DRB1 gene (PROTRANS HLA-DRB1* Cyclerplate). Then, the resolu- tion of the associated subtypes was increased (high resolution: four digits corresponding to exonic level, six digits correspond- ing to intronic level) by a human leukocyte antigen (HLA) sequencing-based typing strategy (PROTRANS Domino Sys- tem, Medipro, Hockenheim, Germany) (Tables 1 and 2). Statistical analysis was performed with the Fisher’s exact test and Bonferroni correction were used. Signifcance level: p = 0.05. The low resolution genotyping showed an association with HLA-DRB1*04 and HLA-DRB1*14 in both the Egyptian and in German PV patients. The frequency of HLA-DRB1*04 was 38.0% in German PV patients [controls: 17.6%; corrected p-value: 1.8E-4, odds ratio (OR): 2.88, confdence interval (CI): 1.59 – 5.23] and 44.7% in Egyptian PV patients (controls: 18.5%; corrected p-value: 5.86E-5, OR: 3.56, CI: 1.99–6.38). † These authors contributed equally to this work. HLA-DRB1*14 was found in 21.7% of German PV patients (controls: 2.0%; corrected p-value: 1.85E-6, OR: 13.46, CI: 3.86 – 46.87) and in 14.9% of Egyptian patients (controls: 1.4%; corrected p-value: 6.86E-3, OR: 12.6, CI: 2.79 – 56.84). In addi- tion, a signifcant association of HLA-DRB*08 was seen in Egyptian patients (13.8%; controls, 1.4%; corrected p-value: 1.7E-3, OR: 11.56), but not in Germans. On the other hand HLA-DRB1*13 (patients: 4.35%, controls: 16.89%, corrected p-value: 1.7E-3, OR: 0.22, CI 0.075–0.665) seems to have a protective effect in Germans, but not in Egyptians (Tables 1 and 2). In the second step, we further increased the resolution of the associated DRB1 subtypes by a HLA sequencing-based typing strategy. The high resolution analysis of HLA-DRB1*04 revealed a signifcant association of HLA-DRB1*04:02:01 in both Germans and Egyptians (corrected p-values 3.11E-11 and 1.23E-8, respectively). In Germans, a negative correlation was observed with HLA-DRB1*04:01:01 (corrected p-value 0.49) (Table 3). Furthermore, HLA-DRB1*14:01:01/14:54 was associated with disease in Germans and Egyptians alike (corrected p-value 1.37E-3 and 3.15E-3, respectively). Egyptian patients were also signifcantly associated with HLA-DRB1*08:04:01 (corrected p-value 4.6E-4) (Table 4). In this study, for the frst time, low and high resolution geno- typing was performed in Egyptian PV patients in comparison to a German cohort. In both populations, HLA-DRB1*04:02:01 and HLA-DRB1*14:01:01/14:54 showed a strong association with PV (Tables 3 and 4). The alleles HLA-DRB1*14:54 and HLA-DRB1*14:01:01 differ only outside of their antigen recognition site for a nucleotide polymorphism on exon 3, changing codon 112 from tyrosine to histidine (5). They were not distinguish- able by common genotyping methods leading to conficting reports in the literature. The meta-analysis of Yan et al. (6) does not mention HLA-DRB1*14:54, but describes a strong association to HLA-DRB1*14:01, whereas Saha et al. reported that HLA-DRB1*14:54, but not HLA-DRB1*14:01:01, is asso- ciated with PV in a British cohort (7). Similarly Parnicka © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd 283 Tissue Antigens, 2015, 85, 283–286