Dermatitis herpetiformis in two American blacks: HLA type and clinical characteristics

Dermatitis herpetiformis is a rare blistering skin disease characterized in part by granular IgA deposits at the dermoepidermal junction, an associated gluten-sensitive enteropathy, and a strong association with the human histocompatibility leukocyte antigen (HLA)-A1 (74% of patients with dermatitis herpetiformis), -B8 (88%), -DR3 (95%), and -DQw2 (100%). Dermatitis herpetiformis is rarely seen in American blacks and some investigators have postulated that this finding may be due to the decreased frequency of HLA-A1 and -B8 in American blacks compared with Caucasians (American blacks: HLA-A1 = 15.3%, HLA-B8 = 10.7%; Caucasian: HLA-A1 = 26.4%, HLA-B8 = 18.3%). This report describes two American blacks with dermatitis herpetiformis and reports the results of HLA typing of these subjects for HLA-A, -B, -C, -DR, and -DQ antigens. HLA typing revealed that neither patient expressed HLA-A1 or -B8; however, both patients did express the class II antigens most frequently seen in dermatitis herpetiformis, HLA-DR3 and -DQw2. Comparison of HLA class II antigen frequency in normal American blacks and Caucasians reveals a similar frequency of HLA-DR3 and -DQw2 (American blacks: HLA-DR3 = 27.6%, HLA-DQw2 = 40.9%; Caucasian: HLA-DR3 = 22.6%, HLA-DQw2 = 32.9%). These data confirm the importance of the HLA class II region in the pathogenesis of dermatitis herpetiformis. In addition, these data suggest that the rare occurrence of dermatitis herpetiformis in American blacks is not due to the decreased frequency of the HLA class I antigens -A1 and/or -B8 but rather is related to differences in the HLA class II region not detected by routine HLA typing.     

A comparative serological and molecular study of linear IgA disease and dermatitis herpetiformis

The class I and class II HLA serologically defined antigens and DQ alpha and DX alpha restriction fragment length polymorphism (RFLP) in 23 patients with linear IgA disease (LAD) were determined and their frequencies compared with those in a group of patients with dermatitis herpetiformis (DH) and healthy controls. In LAD there was a significant increase in HLA-B8 and DR3 and a larger increase in the DQw1-DR2/DRw6 related DQ alpha 6.2 kb and 6.8 kb RFLP. In DH there was a significantly increased frequency of HLA-A1, B8, DR3, and DQw2 with a concomitant increase in the DR3-DQw2 related DQ alpha 4.6 kb RFLP. The difference in DR3 frequencies and the increased frequency of DQw1 rather than DQw2 in LAD indicates that different susceptibility genes operate in the two diseases.     

Dermatitis herpetiformis

The state of our understanding of the pathogenesis of DH relies on the integration of several key characteristics: (1) a high frequency of the HLA antigens HLA-B8, HLA-DR3, and HLA-DQw2, (2) an associated GSE, (3) the resolution of both the skin lesions and gut abnormalities in response to a gluten-free diet, and (4) the presence of granular deposits of IgA in normal and perilesional skin. The role of the HLA class II antigens expressed in patients with DH most likely relates to the afferent or initiating arm of the immune system. The association of the HLA-A1, -B8, -DR3, -DQw2 haplotype with Sjogren's syndrome, chronic hepatitis, Graves' disease, and other presumably immunologically mediated diseases, as well as the evidence that some normal HLA-B8, -DR3 individuals have an abnormal in vitro lymphocyte response to wheat protein and mitogens and have abnormal Fc-IgG receptor-mediated functions, suggests that this HLA haplotype or genes linked closely to it may confer a generalized state of immune susceptibility on its carrier, the exact phenotypic expression of which depends on other genetic or environmental determinants. It also is clear, from the association of DH with GSE and the ability to control the cutaneous manifestations of DH with a gluten-free diet, that the gut disease is a critical factor in the pathogenesis of DH. Several pathogenetic theories about the origin of the cutaneous IgA deposits in DH have been proposed, one of which states that the IgA is produced in the gut mucosa as a response to a dietary antigen or gut epithelial antigen and then cross-reacts with the skin of patients with DH. A second hypothesis is that the IgA produced in the gut binds to an antigen and is deposited in skin as an antigen-antibody complex. Finally, it could be that the gut mucosal abnormality simply allows an unknown antigen access to the central immune system where an IgA antibody is produced that binds to skin. The failure to detect circulating IgA anti-basement membrane zone antibodies in patients with DH suggests that either the structures to which the IgA binds are not present in normal skin without DH, that IgA cannot bind to these structures in vitro, or that the circulating IgA is too scant for detection with conventional methods. Finally, it must be considered that the IgA deposited in DH skin may bind as a result of non-antigen-antibody interactions that cannot be duplicated in vitro.(ABSTRACT TRUNCATED AT 400 WORDS)     

Alterations in HLA-DP and HLA-DQ antigen frequency in patients with dermatitis herpetiformis

Dermatitis herpetiformis (DH) is associated with a markedly increased frequency of the HLA class II antigens DR3 and DQw2. To investigate a possible role of HLA-DP (or closely associated genes) in the pathogenesis of DH as well as to confirm the previously described alterations of HLA-DR3 and HLA-DQw2 antigen frequency, we have typed 43 patients with DH for HLA-DP, HLA-DQ, and HLA-DR antigens. All patients with DH had typical clinical and histologic features, as well as granular deposits of IgA at the dermal-epidermal junction by direct immunofluorescence. HLA-DR3 was expressed in 41 of 43 (95%) DH patients, whereas HLA-DQw2 was expressed in all 43 (100%). The overall distribution of HLA-DP antigens in patients with DH was significantly different from that seen in all controls and in HLA-DR3 and HLA-DQw2 controls (p less than 0.02). Examination of the frequency of individual DP antigens revealed that HLA-DPw1 was increased (42% of patients with DH vs 11% of all controls and 26% of DR3 positive controls), but this increase was not statistically greater than that expected due to the disequilibrium linkage of DPw1 with DR3/DQw2. Patients with DH, however, did have a statistically significant decreased frequency of DPw2 (14% of patients vs 31% of all controls and 41% of DR3 positive controls) (pc less than 0.05). Studies of three informative families demonstrated that the DPw2 genes of the DH patients were not present on the haplotype thought to carry a DH susceptibility gene (HLA-A1, HLA-B8, HLA-DR3, HLA-DQw2). A role of HLA-DP region genes in the pathogenesis of DH is further suggested by the observation that HLA-DPw1 was expressed in 82% (9 of 11) of DH patients with IgA antibodies against dietary antigens as compared with only 33% (4 of 12) of patients without IgA antibodies. HLA-DP genes or genes closely linked to them may be important in DH either as markers of the disease haplotype or by direct involvement in its pathogenesis.     

HLA Antigens and Linkage Disequilibrium Patterns in Turkish Behçet's Patients

Behçet's disease (BD) is a multisystem disorder featuring mucocutaneous, ocular, articular, vascular, intestinal, pulmonary, and neurologic involvement. Although the pathogenesis of the disease is still unknown, most studies have proposed that immunologic factors may play a major role in its development in genetically predisposed individuals. Seventy-one Turkish patients with BD, diagnosed according to the International Study Group for Behçet's Disease criteria, were studied and compared with 600 healthy controls to determine not only frequencies of HLA-A, B, and DR antigens but also whether BD shows any distinct linkage disequilibrium (LD) patterns. In addition, three-point linkage disequilibrium and relative risk (RR) values were determined. Of the HLA-A, B, and DR antigens examined, only B5 (51) was significantly increased in the patient group (X²=55.4; p<0.05; RR=6.44). DR7 was significantly decreased in the patient group (X²=6.9; p>0.05; RR=0.31). HLA haplotype B5-DR5 was found to be in negative LD in the control group, but Behçet's patients showed a strong positive LD between these two antigens. HLA haplotypes A2-B12, B5-DR2, and B12-DR4 showed negative LD in the patients; A1-B5 and B5-DR5 had positive LD in the patients. HLA haplotype A2-B5-DR5 was found to be more frequent than expected in both patients and control gorups. A2-B12-DR4 showed a negative LD in the patients. The strong LD patterns between HLA-B and DR antigens in BD suggest that the susceptibility gene to BD could reside between the these two antigens.     

Analysis of HLA antigens in Croatian patients with psoriasis

In common with most autoimmune diseases, psoriasis is associated with some HLA antigens. We studied the distribution of HLA antigens in Croatian patients with psoriasis: 108 patients were divided into groups according to family history and age of disease onset. HLA antigens were analyzed serologically and HLA-C alleles were analyzed using polymerase chain reaction. We found significant increases in HLA-A2, -B17, -B37 and -B13 antigens and highly significant increases in HLA-Cw*0602 and DR7 antigens in psoriatic patients compared with controls. Patients with type I psoriasis (early onset, positive family history) showed highly significant associations with Cw*0602 [p < 0.00001; relative risk (RR) = 14.45] and DR7 (p < 0.00001; RR = 15.09) antigens. Patients with type II psoriasis (late onset, no family history) had a significant association with Cw*03 antigen (p = 0.008; RR = 0.17). In conclusion, HLA-B13, -B17, Cw*0602 and -DR7 antigens are associated with a significant risk of psoriasis in the Croatian population and the Cw*0602 allele has the strongest association, especially for type I psoriasis.     

HLA-class 1 and class 2 antigens in Turkish patients with pemphigus

BACKGROUND: Pemphigus is an autoimmune disease which is more frequently seen in certain ethnic groups such as Jews. It is thought that exogenous factors may induce pemphigus in genetically predisposed individuals. Recent reports on HLA antigens indicate an increased frequency of HLA-class II antigens particularly HLA-DR4 among Jewish patients. Herein we investigated the antigen frequencies of HLA-A, B, C, HLA-DR and DQ in Turkish patients with pemphigus. METHODS: HLA class I and II antigens were typed by microdroplet lymphocyte cytotoxicity test in 33 patients with pemphigus and 100 healthy individuals. RESULTS: HLA-B35, B44, CW4, DR4, DR14, DQ8 and DQ4 antigens were significantly high in the study group whereas HLA-DR11, DQ7 and DQ2 antigens were high among the controls. The most striking differences were observed in HLA class II antigens. HLA DR14-DQ8 and HLA B35-DR14 haplotypes were the most frequently observed ones in the study group. CONCLUSIONS: We postulate that HLA-B35, B44, CW4, DR4, DR14, DQ4 and DQ8 antigens may be responsible for susceptibility to pemphigus while HLA-DR11, DQ7 and DQ2 antigens may have a protective role in the Turkish population.     

Association of herpes simplex virus-induced erythema multiforme with the human leukocyte antigen DQw3

We studied the genetic markers in human leukocyte antigen (HLA) region HLA-A, -B, -C, -DR, -DQ, C2, Bf, C4A, and C4B of 31 patients with erythema multiforme (EM) and their families. In contrast to the complement allotypes, which showed no deviation from the distribution in the normal population, two HLA class II antigens occurred in much higher frequency in patients with EM. The frequency of HLA-DQw3 (77.4%) increased with high significance compared with normal Caucasian control individuals (41.2%). An even stronger DQw3 association was found in the patient group with postherpetic EM (88.8%; relative risk, 9.41). Interestingly, all patients suffering from frequently recurrent EM were found to have the DQw3 allele (relative risk 44.2). The previously reported association to HLA-B15 was also seen and may be due to a linkage disequilibrium with the HLA-DR4 allele (66.6% in recurrent EM). Our data provide further evidence that classic recurrent EM is related to herpes simplex virus infection and should be regarded as a distinct entity within the enigmatic EM syndrome. DQw3 may serve as a helpful marker for distinguishing this entity from other diseases with EM-like lesions.     

HLA-DR3 in dermatitis herpetiformis

Twenty-one patients with dermatitis herpetiformis were typed for HLA-ABC and -DR determinants. The incidence of HLA-AI, -B8 and -DR3 antigens was found to be significantly higher (P: = 10^?3, <10^?6 and <10^?6, respectively) among patients with dermatitis herpetiformis than among the normal population. HLA-DR3 was found in 85.7% of patients, HLA-B8 in 66.7% and HLA-Ai in 61.9% only. These results indicate that HLA-DR3 is the antigen primarily associated in dermatitis herpetiformis and the latter antigens (HLA-Ai and -B8) are present in increased incidence, probably due to the known linkage disequilibrium of these antigens with HLA-DR3.     

Association of HLA loci alleles and antigens in Saudi patients with vitiligo

HLA complex is composed of several closely linked loci, each containing several alleles, yielding a high expression of polymorphism. Vitiligo, a commonly acquired dermatological disorder, has been associated with different HLA antigens in different ethnic groups. In this study, HLA classes I (HLA-A, B, and C) and II (HLA-DR, DQ) antigens/alleles were analyzed in a group of 80 Saudi subjects consisting of vitiligo patients (40) and matched controls (40). The frequency of antigens of various HLA loci was tested using two-stage microcytotoxicity assays, while the frequency of alleles of HLA-DR was screened by polymerase chain reaction/sequence specific primers (PCR/SSP) method. The frequencies of HLA-B7, B15, Bw6, Cw6, Cw7, and DRB4*010101 were found to be significantly higher in vitiligo patients compared to controls [P = 0.029, 0.015, 0.033, 0.009, 0.043, and 0.015, respectively, with relative risk (RR) > or = 3, etiologic fraction (EF) > or = 0.4]. On the other hand, HLA-A9, B5, DQ1, and DRB3*010101 were significantly decreased in vitiligo patients compared to healthy Saudis [P = 0.008, 0.004, 0.028, and 0.04, respectively, with RR < 1 and preventive fraction (PF) < 0.5]. Among the patients, the highest allele frequency was noted for DRB4*010101(70%), while in controls it was for DRB3*010101 (72.5%). These results for antigens and allele frequency of various HLA Loci in vitiligo patients and control subjects suggested that HLA-B7, Bw6, Cw6, Cw7, and DRB4*010101 could be susceptible to vitiligo, while HLA-A9, B5, DQ1, and DRB3*010101 might be negatively associated with the development of vitiligo in Saudis.     

Recurrent erythema multiforme: tissue typing in a large series of patients

Previous reports have shown an increased frequency of certain HLA antigens in association with erythema multiforme, including HLA-B15(B62), HLA-B35, HLA-A33, HLA-DR53 and, more recently, HLA-DQB1*0301. A strong association with HLA-DQ3 has been documented in patients with recurrent erythema multiforme. We have performed HLA typing in 39 patients with recurrent erythema multiforme, of whom 33 were associated with herpes simplex virus infection. The results were compared with 309 controls. In the recurrent erythema multiforme patients there was a statistically significant increase in HLA-B62 and HLA-B35. An increase in HLA-DR53 was also found, although this did not reach statistical significance. There was no increase in HLA-A33. The presence of HLA-DQ3 in the study population approached that in the controls. Finally, the study population demonstrated a trend towards a reduction in the HLA antigens A1, B8, and DR3. The study confirms the previously reported associations with HLA-B62 (B15), HLA-B35 and HLA-DR53. We have been unable to confirm an association of HLA-A33 or HLA-DQ3 with erythema multiforme. The HLA antigens A1, B8, and DR3 are associated with autoimmune disease, reflecting an increased host response to tissue self antigens. Their absence in patients with recurrent erythema multiforme (REM) may be an indicator of a poor host response to an antigen, which in the case of REM is the herpes simplex virus.     

Association of HLA loci alleles and antigens in Saudi patients with vitiligo

HLA complex is composed of several closely linked loci, each containing several alleles, yielding a high expression of polymorphism. Vitiligo, a commonly acquired dermatological disorder, has been associated with different HLA antigens in different ethnic groups. In this study, HLA classes I (HLA-A, B, and C) and II (HLA-DR, DQ) antigens/alleles were analyzed in a group of 80 Saudi subjects consisting of vitiligo patients (40) and matched controls (40). The frequency of antigens of various HLA loci was tested using two-stage microcytotoxicity assays, while the frequency of alleles of HLA-DR was screened by polymerase chain reaction/sequence specific primers (PCR/SSP) method. The frequencies of HLA-B7, B15, Bw6, Cw6, Cw7, and DRB4*010101 were found to be significantly higher in vitiligo patients compared to controls [P = 0.029, 0.015, 0.033, 0.009, 0.043, and 0.015, respectively, with relative risk (RR) ≥ 3, etiologic fraction (EF) ≥ 0.4]. On the other hand, HLA-A9, B5, DQ1, and DRB3*010101 were significantly decreased in vitiligo patients compared to healthy Saudis [P = 0.008, 0.004, 0.028, and 0.04, respectively, with RR < 1 and preventive fraction (PF) < 0.5]. Among the patients, the highest allele frequency was noted for DRB4*010101(70%), while in controls it was for DRB3*010101 (72.5%). These results for antigens and allele frequency of various HLA Loci in vitiligo patients and control subjects suggested that HLA-B7, Bw6, Cw6, Cw7, and DRB4*010101 could be susceptible to vitiligo, while HLA-A9, B5, DQ1, and DRB3*010101 might be negatively associated with the development of vitiligo in Saudis.     

Class II major histocompatibility complex typing across the ethnic barrier in pemphigoid gestationis. A study in Mexicans

BACKGROUND: Pemphigoid gestationis (PG), also called herpes gestationis, is a rare autoimmune disease of pregnancy or puerperium (estimated 1 out of 50,000 pregnancies among Caucasians). A previous series has demonstrated an association of PG with human leukocyte antigen (HLA)-DR3 or HLA-DR4 haplotypes. While these haplotypes are most commonly found in individuals of European ancestry, they have also been found in African-American patients affected with PG. PG has rarely been reported in other ethnic groups, and the HLA association in non-Europeans has not been examined. METHODS: We have characterized eight patients of Mexican ancestry who have PG by clinical, histologic, and immunofluorescence criteria. Class I and class II major histocompatibility complex (MHC) antigens were studied by standard microlymphocytotoxicity assays. Class II MHC antigens were further studied by polymerase chain reaction (PCR) amplification of HLA-DRB1, DQA, and DQB genes and allele-specific oligonucleotide hybridization. For comparison purposes, we used results obtained from a group of 100 ethnically matched healthy individuals. RESULTS: We found that all eight patients had the HLA-DR3/DR4 phenotype; all HLA-DR3 haplotypes were HLA-DRB1*0301, DQA1*0501, and DQB1*0201, whereas half of the HLA-DR4 haplotypes were from the DRB1*0401 subtype and the other half were DRB1 *0407. CONCLUSIONS: These results suggest that, in Mexicans, the genetic susceptibility for the development of PG is strongly influenced by the genetic admixture of Caucasian origin, and the role of class II MHC antigens in the pathophysiology of this disease is confirmed.     

No human leukocyte antigen-A, -B or -DR association in Swedish patients with hidradenitis suppurativa.

Hidradenitis suppurativa (HS) is a cicatrizing, inflammatory and recurrent disease restricted to inverse skin, such as that of the axilla and groin of younger adults. In a previous study, using serological tissue-typing techniques, no significant increases in the human leukocyte antigen (HLA)-A and -B specificities were found in patients with HS. The aim of this study was to determine the frequencies of HLA-A, -B and, for the first time, HLA-DR alleles, using genomic tissue-typing methods in patients with HS. Forty-two unrelated Swedish patients with HS were included and compared with 250 controls. According to clinical staging adopted from Hurley all of the patients had stage II HS, i.e. recurrent abscesses with tract formation and cicatrization and single or multiple widely separated lesions. No association with HLA-A, -B or -DRB1 alleles was found in patients with HS. Genetic factors associated with the HLA class I or II regions do not appear to contribute significantly to the possible genetic susceptibility of HS.     

HLA -A, -B, -C and -DR antigens in individuals with sensitivity to cobalt

In a skin investigation of 853 individuals working with hard metal manufacturing 39 cases of cobalt allergy were found. Thirty-five of the individuals with cobalt sensitivity and 102 matched controls were HLA-A, -B, -C and -DR typed. No significantly deviating HLA antigen frequencies were observed when the two groups were compared. Thus, there are no signs that a certain HLA antigen would dispose to cobalt allergy. In the cobalt sensitive group the B7 positive individuals showed particularly often simultaneous reactions to other contact allergens (p = less than 0.025). The B12 positive individuals had low reactivity (p = less than 0.0001) while the A28 positive showed high reactivity (p = less than 0.015).     

A study of HLA antigen association in localized and generalized granuloma annulare

HLA-A and -B antigen frequencies were studied in 78 patients, 35 with localized granuloma annulare and 43 with generalized granuloma annulare (GA). Twenty-eight patients in each group were also typed for HLA-DR antigens. A group of 200 healthy age-matched subjects served as controls. HLA-A31 and B35 were increased significantly in patients with generalized GA, but not in the localized form. HLA-DR antigen distribution showed no significant variation.     

Human leukocyte antigen associations in basal cell carcinoma

Basal cell carcinoma is the most common form of skin cancer and is one in which both host and environmental factors are thought to play a role in its pathogenesis. For an investigation of the role of human leukocyte antigen (HLA)-associated variations in genetic susceptibility, thirty-one patients with multiple basal cell carcinomas were typed for HLA-A, B, C, and DR antigens. Patients were compared with both local and appropriate ethnic group controls. No statistically significant association with HLA-A, B, or C antigens was noted in any group. However, a significant increase in HLA-DR1 was noted in non-Irish, non-Ashkenazi patients. A tendency toward a decrease in HLA-DR3 was also noted among patients of Irish or Ashkenazi Jewish descent. The role of HLA-associated genetic factors in this form of skin cancer needs further investigation.     

MHC haplotypic association in Chinese Han patients with vitiligo

BACKGROUND: Serological typing of the human leucocyte antigen (HLA) has shown discrepancies in HLA associations with vitiligo in different ethnic populations. OBJECTIVE: To evaluate the distributions of HLA at class I and II loci that may contribute to the genetic susceptibility of vitiligo patients in Chinese Hans population. METHODS: We analysed the allelic frequencies of HLA class I and II by using the polymerase chain reaction sequence-specific primer (PCR-SSP) method in 187 patients with vitiligo and 273 controls in Chinese Hans. The linkage disequilibrium was calculated from a 2 x 2 table. RESULTS: Two-locus haplotypes including HLA-A25-B13, HLA-A25-B27, HLA-A25-Cw*0602, HLA-A25-DQA1*0302, HLA-A25-DQA1*0601, HLA-A25-DQB1*0303, HLA-B13-Cw*0602, HLA-B13-DQA1*0302, HLA-B13-DQA1*0601, HLA-B27-Cw*0602, HLA-B27-DQA1*0302, HLA-B27-DQA1*0601, HLA-B27-DQB1*0303, HLA-B27-DQB1*0503, HLA-Cw*0602-DQA1*0302, HLA-Cw*0602-DQA1*0601, HLA-Cw*0602-DQB1*0303, HLA-Cw*0602-DQB1*0503 and HLA-DQA1*0302-DQB1*0503 were associated with all types of vitiligo in Chinese Hans. The extended haplotypes HLA-A25-B13-Cw*0602, HLA-A25-B27-Cw*0602, HLA-DQA1*0302-DQB1*0303-Cw*0602 and HLA-B13-DQB1*0303-Cw*0602 were found to be associated with all types of vitiligo in Chinese Hans, whereas the frequency of HLA-A25-Cw*0602-DQA1*0302 was significantly increased in generalized vitiligo but not in localized vitiligo. The frequencies of HLA-A25-DQA1*0302-DQB1*0503 and HLA-A30-DQA1*0302-DQB1*0303 were higher in childhood vitiligo than in adult vitiligo, and the frequency of HLA-A25-B13-DQB1*0303-Cw*0602 was shown to be associated with adult vitiligo but not childhood vitiligo. CONCLUSION: This study demonstrates not only the differential association between HLA markers and types of vitiligo according to distribution or age at onset but also newly found high-risk haplotypes in Chinese vitiligo patients.     

Subacute cutaneous lupus erythematosus in multiple members of a family with C2 deficiency

Deficiency of the second component of complement (C2d) has been associated with systemic lupus erythematosus (LE)-like syndromes as well as recurrent infections. In particular, C2d has been associated with the LE subset of subacute cutaneous LE (SCLE), the presence of anti-Ro antibodies (anti-Ro or SS-A), and the human leukocyte antigen (HLA) types A25, B18, and DR2. A family with C2d in which three members have developed SCLE was observed and studied clinically, serologically, and immunogenetically. Deficiency of the second component of complement was present in all six family members, while anti-Ro was present in only two. There was a strong but incomplete association of C2d and SCLE with HLA-DR2, but the association was not complete with positivity of anti-Ro or antinuclear antibodies. Study of this family reconfirmed the close association of HLA-A25, -B18 and -DR2 with the C2 gene, but indicated a less close association of these loci with serologic markers.     

Association of HLA class I alleles with psoriasis vulgaris in southeastern Chinese Hans

BACKGROUND: Predisposing genetic factors in psoriasis include associations with human leukocyte antigen (HLA). Accumulative evidence has shown that certain HLA at class I locus, especially HLA-Cw6, are associated closely with psoriasis. OBJECTIVE: To evaluate the association of HLA class I alleles with susceptibility to psoriasis in the southeastern Chinese Han population. METHODS: We performed genotype for HLA-A, -B and -C loci in 166 patients with psoriasis vulgaris by means of polymerase chain reaction with sequence-specific primers technique. The distribution of HLA allelic frequencies was further analyzed according to age of onset, i.e. under 35-y and beyond 35-y groups. These data were compared with the healthy controls of 204 unrelated Hans. RESULTS: The frequencies of HLA-A*26 (24.7% vs. 13.1%, OR=2.36, Pc<0.01), -B*13 (27.2% vs. 14.8%, OR=2.34, Pc<0.01), -B*27 (12.2% vs. 4.0%, OR=3.49, Pc<0.01) and -Cw*0602 (17.9% vs. 5.3%, OR=4.20, Pc<0.001) were significantly increased in psoriasis patients, whereas HLA-Cw*0304 frequency (4.9% vs. 13.4%, OR=0.32, Pc<0.01) was highly decreased, when compared to the controls. HLA-A*26-B*27-Cw*0602 was identified as a high-risk haplotype of HLA class I in developing psoriasis in the test. HLA-Cw*0602 was found to be strongly associated with the early-onset psoriasis (age of onset <35 y). CONCLUSION: This study demonstrated the positive associations of HLA class I markers with psoriasis vulgaris, of which HLA-Cw*0602 was the strongest susceptibility determinant for development of early-onset psoriasis, in the southeastern Chinese Han population.