Allele frequency and the associations of HLA-DRB1 and HLA-DQB1 polymorphisms with pemphigus subtypes and disease severity.

ABSTRACT: Pemphigus is a rare, devastating, bullous autoimmune disease that damages the skin and mucous membranes, and has high morbidity and mortality. Studies have shown associations of pemphigus vulgaris (PV) and pemphigus foliaceus (PF) with human leukocyte antigen (HLA) class II polymorphisms.This study examined the frequency of Major Histocompatibility Complex, Class II, DR Beta 1, a Protein Coding gene (HLA-DRB1) and Major Histocompatibility Complex, Class II, DQ Beta 1 (HLA-DQB1) alleles in Vietnamese PV and PF patients, and the association of these polymorphisms with pemphigus subtypes and disease severity.The study enrolled 31 unrelated Vietnamese who underwent HLA typing using Sanger sequencing.HLA-DRB1∗14:54 was the most frequent allele in both PV (20.5%) and PF (33.3%) patients. The percentage of HLA-DQB1∗03:02 was significantly higher in PF than PV patients, while the percentage of HLA-DQB1∗05:03 was approximately 10 times higher in PV patients. Pemphigus patients who have the HLA-DRB1∗04 alleles are more likely to have mild or moderate disease.The HLA-DRB1 and DQB1 alleles may influence susceptibility to pemphigus subtypes, with DQB1∗05:03 being specific for PV and DQB1∗03:02 for PF. Our findings suggest that the DRB1∗04 alleles are likely to be associated with mild and moderate disease.

Introduction

Pemphigus is a rare, potentially devastating autoimmune disease targeting the skin and mucous membranes that has high morbidity and mortality. Its pathological mechanism is attributed to the direct effect of autoantibodies on desmoglein antigens on the surface of keratinocytes, which disrupts cell–cell adhesion and causes the formation of intraepithelial blisters on the skin and mucosa. Pemphigus vulgaris (PV) and pemphigus foliaceus (PF) are the 2 most common major clinical forms. The main target of antibodies in PV patients is desmoglein 3 , while PF patients mostly possess autoantibodies against desmoglein 1 . It has been suggested that the diseases result from genotype–environment interactions that regulate the autoimmune response to desmoglein antigens. Therefore, genetic susceptibility to pemphigus has been investigated worldwide.

Most genetic research on pemphigus has focused on the association between the disease and human leukocyte antigens (HLAs), especially HLA class II genes. Studies have shown that HLA class II polymorphisms make a significant contribution to the identification of desmoglein-derived peptides in PV and PF patients. Approximately 95% of pemphigus patients possess specific HLA-DRB1 or DQB1 alleles, which highlights the important role of these gene polymorphisms in the pathogenesis and invasion of the disease. While some HLA class II polymorphisms are population-specific, others have been reported to be universally associated with the pemphigus subtype. The 2 alleles most commonly associated with PV are DRB1∗04:02 and DQB1∗05:03. Associations of PV with DRB1∗04:02 and DQB1∗03:02 have been recorded in Jewish populations, whereas DQB1∗05:03 was found to be more common in non-Jewish PV patients.[2,6] Because of the heterogeneity in disease manifestations and progression among individual patients, HLA alleles are considered promising genetic markers of susceptibility to pemphigus.

Due to the rarity of the disease, little is known about the genotypic characteristics of Vietnamese pemphigus patients. This study examined the frequency of HLA class II DRB1 and DQB1 alleles in Vietnamese PV and PF patients, and investigated the associations of these polymorphisms with pemphigus subtypes and disease severity.

Methods

Study subjects

The study was performed at Ho Chi Minh City Hospital of Dermato-Venereology from November 2019 to June 2020. The study was designed to find any associations between HLA-DRB1 or DQB1 alelles and the pemphigus subtypes (PV and PF) and the disease severity in both subtypes. Thirty-one unrelated Vietnamese aged 18 years or older were confirmed diagnosis of PV or PF based on clinical manifestations, histopathological results with superficial epidermal blistering formation within the stratum granulosum for PF or intraepidermal acantholysis within the suprabasal epidermis for PV, and a direct immunofluorescence test proving the presence of immunoglobulin G (IgG) at the intercellular borders of the epidermis within the superficial layers of the epidermis for PF and within the lower suprabasal epidermis for PV before enrolled in the study. Patients were excluded if they had previously undergone allogeneic stem cell transplantation or had other coexisting autoimmune diseases. All patients did not receive specific treatment for pemphigus before enrollment in the study.

Eligible patients were invited to participate in the study and underwent clinical evaluation by a research doctor. Their baseline characteristics, including age, age of onset, and sex, were recorded during an interview. Disease activity was assessed using the Pemphigus Disease Area Index (PDAI) score which representing for disease activity and disease damage. The PDAI score is ranging from 0 to 263 points (120, 10, 120 points for skin, scalp, and mucosal activity respectively and 13 points for postinflammatory hyperpigmentation). The disease severity was categorized to 3 levels: mild as PDAI scores were less than 15, moderate as PDAI scores were range from 15 to 44, and severe as PDAI scores were from 45 to 263.

DNA extraction

Two milliliters of peripheral blood was collected from each patient in an EDTA-K2 anticoagulant tube. Genomic DNA was extracted by using a GeneJET Whole Blood Genomic DNA Purification Mini Kit (Thermo Scientific, Waltham, MA, USA).

Sequencing

PCR amplification and target gene sequencing were performed using primers designed based on the DNA sequences of HLA-DRB1 (acc. no. NG_002392) and HLA-DQB1 (acc. no. NG_02922), according to the National Center for Biotechnology Information and Lazaro et al.

Sanger sequencing was performed using POP-7 Polymer and a 50-cm capillary array on an ABI 3500 Genetic Analyser (Applied Biosystems, USA). The sequences were analyzed using CLC Main Workbench software, and Basic Local Alignment Search Tool software (European Bioinformatics Institute ) was used to identify HLA-DRB1 and HLA-DQB1 alleles.

Statistical analysis

Continuous data are presented as the mean ± standard deviation or median with interquartile range depending on the normality of the distribution. Categorical data are described as the frequency and percentage. Quantitative variables were analyzed with Student t test. Fisher exact test was used to compare allele frequencies between pemphigus subtypes. Findings were considered significant at a 2-sided P value < .05. All statistical analyses were performed using R software (ver. 4.0.2; R Development Core Team, Austria).

Ethical issues

This study was approved by the Biomedical Research Ethics Review Committee of the University of Medicine and Pharmacy at Ho Chi Minh City, Ho Chi Minh City, Vietnam (Approval no. 593/DHYD-HDDD, November 4, 2019). Informed consent was obtained from all participants.

Results

Participant epidemiological characteristics

Thirty-one pemphigus patients, 22 PV and 9 PF, were enrolled in the study. Table 1 showed the patient’ epidemiological data, including age, gender, age of onset, PDAI score, and disease severity classification. The average age of the patients was 51.55 ± 15.40 years (range: 23–83 years). Most participants had been diagnosed around 1 year before this study, with a mean age of onset of 50.19 ± 15.67 years. Most patients were female (74.2%). Participants with PV had more severe disease than PF patients (91.0% vs. 55.6% classified as severe).

Table 1

Participants’ epidemiological characteristics.

Characteristics	PV (N = 22)	PF (N = 9)	Total (N = 31)
Age	49.73 ± 13.78 (30.00–83.00)	56.00 ± 18.97 (23.00–80.00)	51.55 ± 15.40 (23.00–83.00)
Age of onset	48.27 ± 13.63 (28.00–81.00)	54.89 ± 19.95 (22.00–79.00)	50.19 ± 15.67 (22.00–81.00)
Sex
Male	6 (27.3)	2 (22.2)	8 (25.8)
Female	16 (72.7)	7 (77.8)	23 (74.2)
PDAI score	76.14 ± 25.91 (14.00–121.00)	54.11 ± 35.37 (12.00–135.00)	69.74 ± 30.11 (12.00–135.00)
Disease severity classification
Mild	1 (4.5)	1 (11.1)	2 (6.5)
Moderate	1 (4.5)	3 (33.3)	4 (12.9)
Severe	20 (91.0)	5 (55.6)	25 (80.6)

N = number of patients, PDAI score = Pemphigus Disease Area Index score, PF = pemphigus foliaceus, PV = pemphigus vulgaris.

Allele frequencies

The frequencies of HLA-DRB1 and HLA-DQB1 alleles were summarized in Table 2. HLA-DRB1∗14:54 was the most frequent allele in both PV (20.5%) and PF patients (33.3%). Regarding HLA-DQB1, DQB1∗05:03 (31.8%) was the most frequent allele in PV patients, while DQB1∗03:01 and DQB1∗03:02 (27.8% each) were the most prevalent among PF patients. The second-most frequent HLA-DQB1 allele was DQB1∗03:01 (29.5%) among PV patients and DQB1∗05:02 (22.2%) among PF patients.

Table 2

Frequencies of HLA-DRB1 and HLA-DQB1.

	PV (n = 44)	PF (n = 18)	Total (n = 62)
HLA-DRB1
14:54	9 (20.5)	6 (33.3)	15 (24.2)
03:02	3 (6.8)	1 (5.6)	4 (6.5)
14:04	2 (4.5)	2 (11.1)	4 (6.5)
03:01	2 (4.5)	2 (11.1)	4 (6.5)
04:04	3 (6.8)	0 (0.0)	3 (4.8)
08:03	1 (2.3)	2 (11.1)	3 (4.8)
13:07	2 (4.5)	0 (0.0)	2 (3.2)
14:01	2 (4.5)	0 (0.0)	2 (3.2)
14:03	2 (4.5)	0 (0.0)	2 (3.2)
14:12	2 (4.5)	0 (0.0)	2 (3.2)
15:02	2 (4.5)	0 (0.0)	2 (3.2)
16:02	2 (4.5)	0 (0.0)	2 (3.2)
04:03	2 (4.5)	0 (0.0)	2 (3.2)
07:01	1 (2.3)	1 (5.6)	2 (3.2)
08:02	1 (2.3)	1 (5.6)	2 (3.2)
08:08	1 (2.3)	0 (0.0)	1 (1.6)
11:01	1 (2.3)	0 (0.0)	1 (1.6)
12:02	1 (2.3)	0 (0.0)	1 (1.6)
13:03	1 (2.3)	0 (0.0)	1 (1.6)
14:05	1 (2.3)	0 (0.0)	1 (1.6)
14:06	1 (2.3)	0 (0.0)	1 (1.6)
15:03	1 (2.3)	0 (0.0)	1 (1.6)
16:01	1 (2.3)	0 (0.0)	1 (1.6)
04:05	0 (0.0)	1 (5.6)	1 (1.6)
04:06	0 (0.0)	1 (5.6)	1 (1.6)
04:07	0 (0.0)	1 (5.6)	1 (1.6)
HLA-DQB1
03:01	13 (29.5)	5 (27.8)	18 (29.0)
05:03	14 (31.8)	1 (5.6)	15 (24.2)
05:02	8 (18.2)	4 (22.2)	12 (19.4)
03:02	2 (4.5)	5 (27.8)	7 (11.3)
02:02	5 (11.4)	0 (0.0)	5 (8.1)
02:01	1 (2.3)	1 (5.6)	2 (3.2)
03:03	1 (2.3)	0 (0.0)	1 (1.6)
03:09	0 (0.0)	1 (5.6)	1 (1.6)
05:01	0 (0.0)	1 (5.6)	1 (1.6)

HLA = human leukocyte antigen, n = number of alleles, PF = pemphigus foliaceus, PV = pemphigus vulgaris.

HLA-DRB1 and HLA-DQB1 polymorphisms and pemphigus subtypes

Table 3 compared the HLA-DRB1 and HLA-DQB1 polymorphisms between the 2 groups of pemphigus subtypes. Our study indicated no significant difference in the percentage of HLA-DRB1 alleles between the 2 pemphigus subtypes. The percentage of HLA-DQB1∗03:02 was significantly (P = .01) higher in PF patients (55.6%) than PV patients (9.1%). Patients with DQB1∗03:02 were less likely to develop PV than PF [odds ratio (OR) = 0.09; 95% confidence interval (CI): 0.01–0.61]. Conversely, the percentage of HLA-DQB1∗05:03 in PV patients (59.1%) was approximately 10 times higher than in PF patients (11.1%) (OR = 9.73; 95% CI: 1.38–276.20; P = .02).

Table 3

Associations of HLA-DRB1 and HLA-DQB1 polymorphisms with pemphigus subtypes.

	PV (N = 22)	PF (N = 9)	OR (CI 95%)	P
HLA-DRB1
03:01	2 (9.1)	2 (22.2)	0.36 (0.03–4.05)	.56
03:02	3 (13.6)	1 (11.1)	1.17 (0.11–37.20)	1.0
04:03	1 (4.5)	2 (22.2)	0.19 (0.01–2.65)	.19
04:04	2 (9.1)	0 (0.0)		1.0
04:05	0 (0.0)	1 (11.1)		.29
04:06	0 (0.0)	1 (11.1)		.29
04:07	0 (0.0)	1 (11.1)		.29
07:01	1 (4.5)	1 (11.1)	0.39 (0.01–16.65)	.50
08:02	1 (4.5)	0 (0.0)		1.0
08:03	2 (9.1)	0 (0.0)		1.0
08:08	1 (4.5)	0 (0.0)		1.0
11:01	1 (4.5)	0 (0.0)		1.0
12:02	1 (4.5)	1 (11.1)	0.39 (0.01–16.65)	.50
13:03	1 (4.5)	0 (0.0)		1.0
13:07	2 (9.1)	1 (11.1)	0.77 (0.05–26.41)	1.0
14:01	2 (9.1)	0 (0.0)		1.0
14:03	2 (9.1)	0 (0.0)		1.0
14:04	3 (13.6)	0 (0.0)		.53
14:05	1 (4.5)	0 (0.0)		1.0
14:06	1 (4.5)	0 (0.0)		1.0
14:12	2 (9.1)	0 (0.0)		1.0
14:54	8 (36.4)	5 (55.6)	0.47 (0.09–2.38)	.32
15:02	2 (9.1)	0 (0.0)		1.0
15:03	1 (4.5)	0 (0.0)		1.0
16:01	1 (4.5)	0 (0.0)		1.0
16:02	2 (9.1)	0 (0.0)	0.36 (0.03–4.05)	1.0
HLA-DQB1
02:01	1 (4.5)	1 (11.1)	0.39 (0.01–16.65)	.50
02:02	4 (18.2)	0 (0.0)		.29
03:01	9 (40.9)	4 (44.4)	0.87 (0.17–4.58)	1.0
03:02	2 (9.1)	5 (55.6)	0.09 (0.01–0.61)	.01
03:03	1 (4.5)	0 (0.0)		1.0
03:09	0 (0.0)	1 (11.1)		.29
05:01	0 (0.0)	1 (11.1)		.29
05:02	6 (27.3)	4 (44.4)	0.48 (0.09–2.64)	.41
05:03	13 (59.1)	1 (11.1)	9.73 (1.38–276.20)	.02

CI = confidence interval, HLA = human leukocyte antigen, N = number of patients, OR = odd ratio, P = P value, PF = pemphigus foliaceus, PV = pemphigus vulgaris. Bold values are statistically significant.

Association between HLA-DRB1 and HLA-DQB1 polymorphisms and disease severity

The relationship of the PDAI score and the presently studied HLA alleles was demonstrated in Table 4. For convenience of identification, we called the group of patients having at least 1 specific allele of DRB1 or DQB1 the “with allele” group, and group of patients without that specific DRB1 or DQB1 allele the “without allele” group. In both DRB1 and DQB1 phenotypes, there was no difference in the PDAI score between 2 groups with allele and without allele.

Table 4

Associations of HLA-DRB1 and HLA-DQB1 polymorphisms with PDAI score.

	PDAI score
Allele	With allele	Without allele	P
HLA-DRB1
03	65.00 ± 38.34	71.39 ± 27.53	.67
04	49.00 ± 20.89	73.73 ± 30.26	.06
13	71.25 ± 16.54	69.52 ± 31.85	.87
14	70.68 ± 30.92	67.44 ± 29.70	.78
HLA-DQB1
02	75.17 ± 24.60	68.44 ± 31.59	.58
03	63.10 ± 31.30	81.82 ± 24.70	.07
05	69.05 ± 32.53	71.44 ± 24.87	.82

HLA = human leukocyte antigen, P = P value, PDAI score = Pemphigus Disease Area Index score.

However, when disease severity was categorized according to the PDAI score, the percentage of the HLA-DRB1∗04 allele was higher in mild/moderate patients (50.0%) than in the severe group (8.0%). Pemphigus patients with the HLA-DRB1∗04 allele were more likely to have mild or moderate disease (OR = 9.91; 95% CI: 1.11–116.78; P = .04; Table 5).

Table 5

Associations of HLA-DRB1 and HLA-DQB1 polymorphisms with disease severity.

Allele	Mild/moderate (N = 6)	Severe (N = 25)	OR (CI 95%)	P
HLA-DRB1
03	2 (33.3)	6 (24.0)	1.59 (0.16–11.28)	.63
04	3 (50.0)	2 (8.0)	9.91 (1.11–116.78)	.04
07	0 (0.0)	2 (8.0)		1.0
08	1 (16.7)	2 (8.0)	2.34 (0.07–34.51)	.49
11	0 (0.0)	1 (4.0)		1.0
12	0 (0.0)	2 (8.0)		1.0
13	0 (0.0)	4 (16.0)		.56
14	4 (66.7)	18 (72.0)	0.77 (0.11–7.31)	.79
15	0 (0.0)	2 (8.0)		1.0
16	0 (0.0)	2 (8.0)		1.0
HLA-DQB1
02	0 (0.0)	6 (24.0)		.31
03	6 (100.0)	14 (56.0)		.06
05	5 (83.3)	17 (68.0)	2.11 (0.26–62.57)	.64

CI = confidence interval, HLA = human leukocyte antigen, N = number of patients, OR = odd ratio, P = P value. Bold values are statistically significant.

Discussion

This is the first study to compare HLA-DRB1 and DQB1 polymorphisms between the 2 main pemphigus subtypes (PV and PF) in Vietnamese patients. The mean age of our patients (51.55 ± 15.40 years) was similar to other cohort studies of genetic polymorphisms in HLA associated with pemphigus subtypes, such as Zhang et al (49.3 ± 13.5 years) in China and Aboobaker et al in South Africa (43 years; range: 12–93 years). A female predominance of pemphigus, reflected in our study by a 3:1 female-to-male ratio, has been reported worldwide.[13,14]

The most frequent HLA-DRB1 allele in both pemphigus subtypes was DRB1∗14:54. DRB1∗14 alleles were found in 43.5% of the participants. This was similar to Dere et al, who reported that the 2 most prevalent alleles in Turkish pemphigus patients were DRB1∗04 (41.7%) and DRB1∗14 (25.0%). Studies in other Asian countries have provided inconsistent results. A Korean study reported that the most common HLA-DRB1 allele was DRB1∗01 in PV patients (60.0%) and DRB1∗04 (73.3%) in PF patients. Niizeki et al examined HLA polymorphisms in Japanese patients and found that the most common alleles were DRB1∗14:01, DRB1∗14:05, DRB1∗04:06, DRB1∗14:06, and DRB1∗14:08. In Europe, the DRB1∗14:54 allele was reported to be prevalent. Svecova et al found that the most common HLA-DRB1 alleles in Slovakians were DRB1∗04:02, DRB1∗04:04, DRB1∗14:54, DRB1∗14:04, and DRB1∗14:05. A study of White British and Indo-Asian pemphigus patients reported that the most prevalent HLA-DRB1 alleles were DRB1∗04:02 and DRB1∗14:54. Note that DRB1∗14:54 was included in DRB1∗14:01 until it was acknowledged as a distinct allele in 2005 by the World Health Organization Nomenclature Committee, due to a difference in exon 3 of gene DRB1. Concerns have been raised that the DRB1∗14:01 characterized in studies conducted before 2005 may actually be DRB1∗14:54.[7,18]

The most frequent HLA-DQB1 allele in the PV patients in our study was DQB1∗05:03, while in PF patients the most common allele was DQB1∗03:02. This is in line with a Japanese study in which the DQB1∗05:03 allele also accounted for the largest proportion (27.5%) among PV patients. Loiseau et al reported a high percentage of the DQB1∗05:03 allele (25.7%), which was the second-most common allele in French PV patients, and found that DQB1∗03:02 was the most frequent allele in their PF patients. Abida et al concluded that DQB1∗03:02 might be a common susceptibility allele for pemphigus worldwide.

In our study, the percentage of the DQB1∗03:02 allele in the PF group was significantly higher than in the PV group; this is consistent with Zhang et al, who also concluded that DQB1∗03:02 is specific for the PF subtype among Chinese. This characteristic is stable across ethnicities. In a study of white British patients, Saha et al found that DQB1∗03:02 was a susceptibility allele for PF (OR = 5.35; 95% CI: 2.38–12.1). In Tunisia, Abida et al found a specific association between DQB1∗03:02 and PF, while DQB1∗05:03 was possibly specific for PV, as the proportion of this allele in the PV group (31.8%) was significantly higher than in the PF group (5.6; OR = 6.91; 95% CI: 1.20–178.61). However, Zhang et al showed that allele DQB1∗05:03 is shared by both pemphigus subtypes in Chinese patients.

Since the natural history of pemphigus is frequently unpredictable, it is desirable to identify biomarkers that can predict the clinical outcome and thus improve treatment. Our study showed that the HLA phenotype was not relevant to the PDAI score. However, a study was conducted on 54 Indian PV patients by Priyadarshini, demonstrated that HLA-DQB1∗ 03 was significantly associated with higher PDAI scores (P = .02). In Slovakia, Svecova et al examined 44 PV patients and proposed that the DRB1∗04:02 and DQB1∗03:02 alleles were common in severe patients (P = .001), contrary to our finding that DRB1∗04 alleles were relative to mild/moderate level. These results suggest that HLA class II polymorphisms are promising predictors of prognosis. However, the findings are not fully consistent, probably due to differences in research populations, since ethnicity influences susceptibility or resistance to PV via direct involvement of HLA molecules, which act as antigen presenters or neighbour a linked gene.[22,23]

The main limitation of this study was the relatively small number of participants, due to the rarity of the disease. As the majority of participants had severe disease, the association between alleles and disease severity may be biased. Multicentre studies with more subjects are needed to confirm our findings and obtain more comprehensive data on pemphigus patients.

Conclusions

The HLA-DRB1 and DQB1 alleles may influence the susceptibility to pemphigus subtypes. We found that DRB1∗14:54, DQB1∗05:03, DQB1∗03:01, and DQB1∗03:02 were the most frequent alleles in Vietnamese PV and PF patients, of which DQB1∗05:03 was specific for PV and DQB1∗03:02 was specific for PF. The findings suggest that the DRB1∗04 alleles might be associated with mild and moderate disease.

Acknowledgment

The authors thank Ms Hoang Le Gia Linh (Center for Molecular Biomedicine, University of Medicine and Pharmacy at Ho Chi Minh City, Ho Chi Minh City) for technical support with the HLA typing.

Author contributions

Conceptualization: Thanh Thai Van Le.

Data curation: Thinh Phuc Ong.

Formal analysis: Thinh Phuc Ong, Minh Duc Do.

Funding acquisition: Thanh The Bich Vuong.

Investigation: Thanh The Bich Vuong.

Methodology: Thanh The Bich Vuong.

Project administration: Thanh Thai Van Le.

Resources: Thanh The Bich Vuong.

Software: Thinh Phuc Ong, Minh Duc Do.

Supervision: Thanh Thai Van Le.

Validation: Thanh Thai Van Le.

Visualization: Thanh The Bich Vuong.

Writing – original draft: Thanh The Bich Vuong.

Writing – review & editing: Thanh Thai Van Le, Thanh The Bich Vuong.