Evaluation of HLA-G Expression in Multipotent Mesenchymal Stromal Cells Derived from Vitrified Wharton's Jelly Tissue

BACKGROUND

Mesenchymal Stromal Cells (MSCs) from Wharton's Jelly (WJ) tissue express HLA-G, a molecule which exerts several immunological properties. This study aimed at the evaluation of HLA-G expression in MSCs derived from vitrified WJ tissue.

METHODS

WJ tissue samples were isolated from human umbilical cords, vitrified with the use of VS55 solution and stored for 1 year at -196 °C. After 1 year of storage, the WJ tissue was thawed and MSCs were isolated. Then, MSCs were expanded until reaching passage 8, followed by estimation of cell number, cell doubling time (CDT), population doubling (PD) and cell viability. In addition, multilineage differentiation, Colony-Forming Units (CFUs) assay and immunophenotypic analyses were performed. HLA-G expression in MSCs derived from vitrified samples was evaluated by immunohistochemistry, RT-PCR/PCR, mixed lymphocyte reaction (MLR) and immunofluorescence. MSCs derived from non-vitrified WJ tissue were used in order to validate the results obtained from the above methods.

RESULTS

MSCs were successfully obtained from vitrified WJ tissues retaining their morphological and multilineage differentiation properties. Furthermore, MSCs from vitrified WJ tissues successfully expressed HLA-G.

CONCLUSION

The above results indicated the successful expression of HLA-G by MSCs from vitrified WJ tissues, thus making them ideal candidates for immunomodulation.

Non-Inherited Maternal Antigens Identify Acceptable HLA Mismatches: A New Policy for the Hellenic Cord Blood Bank

BACKGROUND

During pregnancy, the maternal-fetal contact may lead to the development of tolerance against the maternal human leukocyte antigen (HLA) that is not inherited by the fetus. These non-inherited maternal antigens (NIMAs) define acceptable HLA mismatches; therefore, the number of HLA phenotypes that are suitable matches for patients who need a hematopoietic stem cell transplant could be increased. Cord blood unit (CBU) transplantations to patients mismatched for a HLA loci, but similar to the ΝΙΜAs of the CBU, have a prognosis similar to 6/6-matched ones.

METHODS

The Hellenic Cord Blood Bank (HCBB) identified the maternal HLA of 380 cord blood donors, specifying the NIMA haplotypes of the related cryostored CBUs.

RESULTS

The HCBB extended the pool of HLA phenotypes through the generation of unique virtual phenotypes (VPs). A "VP database" was set up, using Microsoft Office-Access™, in order to provide NIMA-matched CBUs for potential recipients. The effectiveness of VPs' matching was tested in 80 Greek patients.

CONCLUSION

This methodology may contribute to the increase of the number of available CBUs for patients, in the case where there is no available CBU, or in case an additional one is needed. Through this method, the CBUs could be used faster and more effectively, rather than being cryostored for long periods of time.

A multi-factorial genetic model for prognostic assessment of high risk melanoma patients receiving adjuvant interferon

Purpose

IFNa was the first cytokine to demonstrate anti-tumor activity in advanced melanoma. Despite the ability of high-dose IFNa reducing relapse and mortality by up to 33%, large majority of patients experience side effects and toxicity which outweigh the benefits. The current study attempts to identify genetic markers likely to be associated with benefit from IFN-a2b treatment and predictive for survival.

Experimental design

We tested the association of variants in FOXP3 microsatellites, CTLA4 SNPs and HLA genotype in 284 melanoma patients and their association with prognosis and survival of melanoma patients who received IFNa adjuvant therapy.

Results

Univariate survival analysis suggested that patients bearing either the DRB1*15 or HLA-Cw7 allele suffered worse OS while patients bearing either HLA-Cw6 or HLA-B44 enjoyed better OS. DRB1*15 positive patients suffered also worse RFS and conversely HLA-Cw6 positive patients had better RFS. Multivariate analysis revealed that a five-marker genotyping signature was prognostic of OS independent of disease stage. In the multivariate Cox regression model, HLA-B38 (p = 0.021), HLA-C15 (p = 0.025), HLA-C3 (p = 0.014), DRB1*15 (p = 0.005) and CT60*G/G (0.081) were significantly associated with OS with risk ratio of 0.097 (95% CI, 0.013–0.709), 0.387 (95% CI, 0.169–0.889), 0.449 (95% CI, 0.237–0.851), 1.948 (95% CI, 1.221–3.109) and 1.484 (95% IC, 0.953–2.312) respectively.

Conclusion

These results suggest that gene polymorphisms relevant to a biological occurrence are more likely to be informative when studied in concert to address potential redundant or conflicting functions that may limit each gene individual contribution. The five markers identified here exemplify this concept though prospective validation in independent cohorts is needed.

Evaluation of six CTLA-4 polymorphisms in high-risk melanoma patients receiving adjuvant interferon therapy in the He13A/98 multicenter trial

Purpose

Interferon is approved for adjuvant treatment of patients with stage IIb/III melanoma. The toxicity and uncertainty regarding survival benefits of interferon have qualified its acceptance, despite significant durable relapse prevention in a fraction of patients. Predictive biomarkers that would enable selection of patients for therapy would have a large impact upon clinical practice. Specific CTLA-4 polymorphisms have previously shown an association with response to CTLA-4 blockade in patients with metastatic melanoma and the development of autoimmunity.

Experimental design

286 melanoma patients and 288 healthy controls were genotyped for six CTLA-4 polymorphisms previously suggested to be important (AG 49, CT 318, CT 60, JO 27, JO30 and JO 31). Specific allele frequencies were compared between the healthy and patient populations, as well as presence or absence of these in relation to recurrence. Alleles related to autoimmune disease were also investigated.

Results

No significant differences were found between the distributions of CTLA-4 polymorphisms in the melanoma population compared with healthy controls. Relapse free survival (RFS) and overall survival (OS) did not differ significantly between patients with the alleles represented by these polymorphisms. No correlation between autoimmunity and specific alleles was shown. The six polymorphisms evaluated where strongly associated (Fisher's exact p-values < 0.001 for all associations) and significant linkage disequilibrium among these was indicated.

Conclusion

No polymorphisms of CTLA-4 defined by the SNPs studied were correlated with improved RFS, OS, or autoimmunity in this high-risk group of melanoma patients.

Correlation of molecular human leukocyte antigen typing and outcome in high-risk melanoma patients receiving adjuvant interferon

BACKGROUND:

Interferon is approved for adjuvant treatment of patients with stage IIB/III melanoma. The identification of predictive markers that would permit selection of patients would be beneficial. Specific human leukocyte antigen (HLA) class I and II antigens have previously shown an association with response to therapy or overall survival of patients with metastatic melanoma.

METHODS:

A total of 284 high-risk melanoma patients participating in a randomized trial and 246 healthy controls were molecularly typed for HLA class I and II. Specific allele frequencies were compared between the healthy and patient populations, as well as presence or absence of these in relation to recurrence. Alleles related to autoimmune disease were also investigated.

RESULTS:

No significant differences were found between the distribution of HLA genotype in the melanoma population compared with healthy controls. Correlations between nonrecurrence and the presence of HLA-Cw*06 allele were noted present in 19.3% of melanoma patients. The median relapse-free survival of the Cw*06-positive cohort was 100.2 months versus 37.3 months in the Cw*06-negative cohort (P = .013). The median overall survival for the Cw*06-positive cohort has not yet been reached, versus 78.9 months in the Cw*06-negative cohort (P = .025). HLA-Cw*06 was present in 29.79% of patients in the autoimmunity group and 15.38% of patients in the nonautoimmunity group (P = .049).

CONCLUSIONS:

No allele was associated with absence of recurrence in patients receiving adjuvant interferon with the exception of HLA-Cw*06, an allele correlated with psoriasis. HLA-Cw*06-positive patients have better relapse-free and overall survival. Cancer 2010. © 2010 American Cancer Society.

HLA-DQA1*0505 sharing and killer immunoglobulin-like receptors in sub fertile couples: report from the 15th International Histocompatibility Workshop

This aim of the study was to investigate whether human leukocyte antigen (HLA)-DQA1*0505 sharing or the maternal killer immunoglobulin-like receptor (KIR) repertoire is associated with recurrent spontaneous abortion (RSA) or repeated implantation failure (RIF). The study included 224 couples with RSA, 61 couples with RIF, 182 fertile couples, and 10 couples with successful in vitro fertilization and embryo transfer (IVF)/ET at first cycle. HLA-DQA1*0505 typing using polymerase chain reaction-sequence-specific oligonucleotide (PCR-SSO) was performed in 185 RSA (117 with alloimmune abnormalities and 68 of autoimmune etiology), 61 RIF and 182 control couples, and KIR genotyping using polymerase chain reaction-sequence-specific primer (PCR-SSP) in 167 RSA and 55 RIF cases as well as 46 RSA and 10 IVF controls. No differences in DQA1*0505 sharing were found between patients and controls. In RSA and RIF women, the ratio of inhibitory to activating KIRs was slightly lower (1.53 and 1.85 vs 2.03 in controls). The analysis of maternal inhKIR and fetal HLA-C molecule pairs showed that the 'less inhibiting' combination KIR2DL3-C1 was found in higher percentage in subfertile (mainly RIF) than in fertile couples. In contrast, the percentage of cases possessing the 'strong inhibiting' combination KIR2DL1-C2 was lower in the RSA and RIF groups in comparison with that in the control groups (17.36% vs 23.91 and 16.36% vs 40%, respectively). In women with >or= 6 implantation failures, the KIR2DL1-C2 combination was not found in any of them (P = 0.0014), and the KIR2DL3-C1 combination was not found in the control IVF group. The results oppose the suggestion that increased HLA-DQA1*0505 sharing predispose to RSA or RIF. The KIR2DL3-C1 combination (or lack of the KIR2DL1-C2 one) is associated with implantation failure.

Evaluation of IL-10 polymorphisms in high-risk melanoma patients receiving adjuvant interferon

9049

Background: Attempts to identify patients who benefit from adjuvant treatment of interferon alfa-2b (IFN) have been disappointing. Interleukin-10 polymorphisms have been implicated with the prognosis of patients with advanced melanoma and associated with response to biochemotherapy. Several polymorphisms have been found within the IL-10 gene. We evaluated three IL-10 Single Nucleotide Polymorphisms (SNPs) in high-risk melanoma patients enrolled in a study of two regimens of high-dose IFN and compared the distribution of SNPs found in healthy controls. Methods: We genotyped DNA from peripheral blood of 280 stage IIb, IIc and III melanoma patients and 288 healthy controls for 592 C/A, 819 C/T and 1082 G/A with PCR and pyrosequencing technology (Biotage, Uppsala, Sweden). Results: At a median follow up of 56.3 months (95% CI 47.4–63.7), 147 patients have recurred and 94 have died. The median DFS was 53 months and the median OS 86 months. There were no statistically significant differences in the incidence of IL-10 polymorphisms between the melanoma patients and healthy controls. The incidence of these polymorphisms is presented in table . RFS and OS did not differ significantly between the alleles of these polymorphisms (p=0.88 and p=0.55 for 592 C/A, p=0.84 and p=0.68 for 819 C/T and p=0.26 and p=0.30 for 1082 G/A respectively). Conclusions: No SNP studied was correlated with improved RFS and OS in this high-risk group of melanoma patients.

[Table: see text]

[Table: see text]

Takayasu arteritis: epidemiological, clinical, and immunogenetic features in Greece

OBJECTIVE

Takayasu arteritis (TA) is an uncommon disease with clinical heterogeneity across different ethnic groups. We aimed to evaluate the epidemiological, clinical, and immuno-genetic features of TA in Greece.

METHODS

Demographic, clinical, laboratory, angiographic, and therapeutic data of 42 patients from 4 large referral centers were retrieved. Serology and Human Lymphocyte Antigen (HLA) typing was performed in 22 patients.

RESULTS

We studied 37 women and 5 men with a median age of 31 years at disease onset. Median delay in diagnosis was 24 months and median follow-up was 47 months (range 0-178). Constitutional or musculoskeletal symptoms were present in 86%, especially early in the disease course. Vascular findings were universal with reduced or absent pulse being the most common manifestation (98%). Hypertension was frequent (78%). Extensive disease prevailed and stenotic lesions were more common than aneurysms (95% vs. 40%). Erythrocyte sedimentation rate and C-reactive protein showed modest correlation with disease activity. HLA-B52 was expressed by 37% of the patients vs. 2.4% of the controls (p<0.001). Glucocorticoids and cytotoxic agents were used in most patients with remission rates of 83%. A total of 42 surgical procedures were performed with success rates of 87%.

CONCLUSION

TA in Greece clinically and epidemiologically resembles the pattern of disease in Japan and the Western hemisphere. There is considerable delay in diagnosis, which may partially reflect failure to recognize a rare disease. New surrogate markers are needed to assess disease activity. Glucocorticoids are the cornerstone of treatment and cytotoxic drugs are frequently used as steroid sparing agents.

14th International HLA and Immunogenetics Workshop: Report from the reproductive immunology component

The aim of the study was to investigate whether human leukocyte antigen (HLA) allele sharing between partners or the maternal killer immunoglobulin-like receptor (KIR) repertoire is associated with recurrent spontaneous abortion (RSA) and repeated implantation failure after in vitro fertilization (IVF)/embryo transfer. From a total population of 158 RSA couples, 40 couples with repeated implantation failures (IVF) and 81 control couples, reported by five different laboratories, analysis was performed for (a) HLA sharing in 50 RSA, 31 IVF and 31 control couples, (b) DQA1*0505 sharing/homozygosity among partners in 108 RSA, 40 IVF and 36 control couples, and (c) the women's KIR repertoire in 46 RSA, 26 IVF and 36 control wives. RSA couples were divided into alloimmune aborter (RSAallo) and autoimmune aborter (RSAauto). The results oppose to the suggestion that increased HLA sharing per se or a limited maternal KIR repertoire predisposes to RSA or IVF failure. However, the observation of a slightly higher percentage of DQA1*0505 sharing in the RSAauto and the IVF group needs further investigation. The ratio of inhibitory to activating KIR (actKIR) was slightly lower in RSAallo and IVF women (1.9 vs 2.6 in controls), while in a high percentage of these women, the standard receptors of the KIR A haplotype were combined with actKIR/s of the haplotype B (66.6% and 45.4% vs 20% and 15.3% in RSAauto and control groups). This may suggest a possible involvement of actKIRs in embryo implantation and the maintenance of pregnancy and also requires further investigation.

Correlation of molecular HLA typing and outcome in high-risk melanoma patients receiving adjuvant interferon

8029

Background: Serological typing for both HLA class I and class II antigen expression, has previously shown association of specific HLA antigen expression with clinical response and survival in patients with metastatic melanoma treated with IL-2 (e.g. HLA-DQ1). Purpose: To evaluate the impact of HLA class I (low-resolution) and class II (high-resolution) expression, on the outcome of high-risk melanoma patients receiving adjuvant high-dose interferon. Methods: 181 stage IIB, IIC and III melanoma patients (88 female and 93 male), median age 52.1 years and 246 healthy controls were included in this study. DNA was used for the determination of HLA-A, HLA-B, HLA-Cw, HLA-DRB1 and HLA-DQB1 genotypes. Results: With a median follow-up of 37 months, 59 (group 1) patients have remained with no evidence of recurrence and 122 have recurred (group 2). Statistical significant differences between the two groups, were found in the following genotypes: HLA-A*02 (42% vs. 57.3%, p=0.08), HLA-A*33 (15.2% vs. 6.5%, p=0.05), HLA-B*51 (15.2% vs. 34.4%, p=0.01), HLA-B*57 (11.8% vs. 2.4%, p=0.02). Statistical significant differences between group 1 and healthy controls, were found in the following genotypes: HLA-A*33 (15.2% vs. 6.5%, p=0.05), HLA-B*51 (15.2% vs. 28.5%, p=0.05), HLA-B*57 (11.8% vs. 4.5%, p=0.05), HLA-Cw*03 (23.7% vs. 11%, p=0.01), HLA-Cw*06 (27.1% vs. 16.1%, p=0.06), HLA-DRB1*0701 (27.1% vs. 13.4%, p=0.01), HLA-DRB1*1601 (35.6% vs. 22.3%, p=0.01), HLA-DQB1*0202 (23.8% vs. 10.1%, p=0.09). Conclusions: Statistical significant differences were seen in HLA-A and HLA-B alleles between the patients with high-risk melanoma free of recurrence and those who recurred after treatment with adjuvant interferon. Additionally, differences were seen between healthy controls and melanoma patients free of recurrence.

No significant financial relationships to disclose.

Lack of the appropriate natural killer cell inhibitory receptors in women with spontaneous abortion

Previous studies have revealed that women with unexplained recurrent spontaneous abortions have a limited repertoire of inhibitory KI receptors (inhKIRs) and that the inhKIRs they possess do not have specificity for the human leukocyte antigen (HLA)-Cw molecules that would be expressed on trophoblast. We sought to confirm these findings by direct definition of maternal inhKIR and trophoblastic HLA-Cw allotypes on the placental material of spontaneously missed pregnancies. The study included 30 women undergoing vacuum uterine curettage for first-trimester missed pregnancy (group A; n = 15) or for elective termination of normal pregnancy (group C, n = 15). DNA extracted from isolated decidual and trophoblastic cells was used for molecular detection of maternal inhKIRs (2DL1, 2DL2, 2DL3) and fetal HLA-Cw alleles, respectively. The results revealed that in the group of women who experienced abortion, 60% did not have the full repertoire of three inhKIRs (group A vs group C; p = 0.006); that in five of 15 patients (none in the controls), no epitope matching existed between maternal inhKIRs and trophoblastic HLA-Cw alleles (group A vs group C; p = 0.01); and that more cases were found with limited epitope matching (less than three inhKIRs with specificity for fetal HLA-Cw alleles). The results provide additional evidence that in some cases of spontaneous abortions, the women lack the appropriate inhKIRs to interact with the HLA-Cw molecules on trophoblasts and to deliver signals to inhibit natural killer cell activation and protect the embryo.

Distribution of killer cell immunoglobulin-like receptors in the Greek population

The interaction between killer cell immunoglobulin-like receptors (KIR) expressed on natural killer (NK) cells, and human leukocyt antigen (HLA) molecules expressed on target cells is known to regulate the cytolytic activity. A wide range of KIR genotypes is observed in the population, as the number of KIR loci can vary. In the present study we have determined the frequencies and combinations of 13 KIR genes and two CD94:NKG2 receptor genes and their distribution in the two haplotype groups in a panel of 233 unrelated healthy Greek Caucasians. We have typed genomic DNA for the presence of the putative KIR loci KIR2DL1, KIR2DL2, KIR2DL3, KIR2DL4, KIR3DL1, KIR3DL2, KIR3DL3, KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4, KIR2DS5, and KIR3DS1 using modified polymerase chain reaction sequence-specific primers. The frequency of KIR loci combined with the linkage disequilibrium values suggest that the Greek population shares several general features with other Caucasoid populations studied before, but still distinguishes itself by the increased or decreased frequency of several alleles. The majority of the 45 different KIR genotypes seen in Greeks have been observed in Caucasoid populations studied before. Nevertheless, two of these genotypes, although met in other populations, have not been found in Caucasians before. One individual possesses a novel profile with no KIR inhibitory gene. The A haplotypes remain the most prevalent, with AA1 being the most common genotype, and the number of inhibitory KIRs being more variable than the number of noninhibitory KIRs in most haplotypes.

Natural killer (NK) cell receptors' repertoire in couples with recurrent spontaneous abortions

PROBLEM

Natural killer (NK) cell receptors (NKRs) have been suggested to protect trophoblast, but their function at the fetomaternal interface remains unknown. To investigate if the outcome of pregnancy depends on women's NKRs, we studied the NKR repertoire in couples with recurrent spontaneous abortions (RSA).

METHODS

Twenty-six childless couples with > or = 2 abortions, characterized by alloimmune abnormalities, and 26 control couples were genotyped for five killer immunoglobulin-like receptors (KIR) and two CD94/NKG receptors, known to have as ligands human leukocyte antigen (HLA) class I molecules with trophoblastic expression: inhibitory 2DL1,2,3 and activating 2DS1,4 KIRs, inhibitory NKG2A and activating NKG2C. Detected repertoires of women and partners were compared between the two groups.

RESULTS

Less aborters than controls were found to have all three inhibitory KIRs (30.77% versus 69.23%, P = 0.01), some of them had only one inhibitory KIR (19.23% versus 3.85%, P = 0.08) and most of them were lacking inhibitory KIRs possessed by their husbands (57.69% versus 15.38%, P = 0.001).

CONCLUSIONS

Women with alloimmune abortions have a limited inhibiting KIR repertoire and such miscarriages may occur because trophoblastic HLA class I molecules are recognized by decidual NK cells lacking the appropriate inhibitory KIRs.

MHC class I and class II phenotype, gene, and haplotype frequencies in Greeks using molecular typing data

In the present study, DNA typing for HLA-A, C, B, DRB1, DRB3, DRB4, DRB5, DQA1, DQB1, and DPB1 was performed for 246 healthy, unrelated Greek volunteers of 20-59 years of age. Phenotype, genotype frequencies, Hardy-Weinberg equilibrium fit, and 3-locus haplotype frequencies for HLA-A, C, B, HLA-A, B, DRB1, HLA-DRB1, DQA1, DQB1, and HLA-DRB1, DQB1, DPB1 were calculated. Furthermore, linkage disequilibrium, deltas, relative deltas and p-values for significance of the deltas were defined. The population studied is in Hardy-Weinberg equilibrium, and many MHC haplotypes are in linkage disequilibrium. The most frequent specificities were HLA-A*02 (phenotype frequency = 44.3%) followed by HLA-A*24 (27.2%), HLA-B*51 (28.5%), HLA-B*18 (26.8%) and HLA-B*35 (26.4%) and HLA-Cw*04 (30.1%) and HLA-Cw*12 (26.8%). The most frequent MHC class II alleles were HLA-DRB1*1104 (34.1%), HLA-DQB1*0301 (54.5%) and HLA-DPB1*0401 with a phenotype frequency of 59.8%. The most prominent HLA-A, C, B haplotypes were HLA-A*24, Cw*04, B*35, and HLA-A*02, Cw*04, B*35, each of them observed in 21/246 individuals. The most frequent HLA-A, B, DRB1 haplotype was HLA-A*02, B*18, DRB1*1104 seen in 20/246 individuals, while the haplotype HLA-DRB1*1104, DQB1*0301, DPB1*0401 was found in 49/246 individuals. Finally, the haplotype DRB1*1104, DQA1*0501, DQB1*0301 was observed in 83/246 individuals. These results can be used for the estimation of the probability of finding a suitable haplotypically identical related or unrelated stem cell donor for patients of Greek ancestry. In addition, they can be used for HLA and disease association studies, genetic distance studies in the Balkan and Mediterranean area, paternity cases, and matching probability calculations for the optimal allocation of kidneys in Greece.

Systemic scleroderma in Greece: low mortality and strong linkage with HLA-DRB1*1104 allele

OBJECTIVE

Description of Greek patients with scleroderma with reference to (a) major organ disease, (b) autoantibodies, (c) survival rate, and (d) HLA associations.

METHODS

The clinical files of 254 patients were analysed retrospectively and a standardised clinical chart was completed with age at disease onset, sex, date of first and last visit, clinical and serological findings, organs affected, reasons for death, and HLA class II alleles. HLA class II alleles (DRB1, DQA1, DQB1, DPB1) were determined by polymerase chain reaction amplification using oligopeptide probes. DNA was extracted from 98 patients and 130 Greek controls.

RESULTS

124 patients (49%) had limited systemic sclerosis (lSSc), 114 (45%) had diffuse systemic sclerosis (dSSc), and 16 (6%) had overlap syndromes. Patients with dSSc, compared with lSSc, were characterised by a higher prevalence of lung disease (p=0.0011), oesophageal, heart, and peripheral vessel disease (p=0.027, p=0.0025, and p=0.012, respectively). Anticentromere antibodies (ACA) occurred exclusively in lSSc (34%), whereas antibodies to topoisomerase I (anti-topo I) were associated with dSSc (p<0.0001). Anti-topo I were associated with interstitial pulmonary fibrosis, oesophageal and peripheral vessel disease (p=0.028, p=0.012, and p=0.01, respectively). The HLA-DRB1*1104 allele was associated with the disease (p<0.0001) and anti-topo I (p<0.001), whereas it was not associated with ACA serum reactivity (p<0.001). Renal disease occurred in 4% of patients with SSc. The estimated survival probability for this cohort of patients with SSc, four years after the first visit, is 94.8%.

CONCLUSION

SSc among Greek subjects has the same pattern of organ disease as in other white populations. However, the prevalence of kidney disease is low. The HLA class II DRB1*1104 allele is associated with the disease, with anti-topo I, and not associated with ACA serum reactivity.

HLA associations of anti-beta2 glycoprotein I response in a Greek cohort with antiphospholipid syndrome and meta-analysis of four ethnic groups

Using molecular typing, we evaluated the strength of class II HLA associations in 67 Greek patients with antiphospholipid syndrome (APS), 54 of whom had antibodies against beta2-glycoprotein I (beta2GPI), as compared to 246 controls. To further clarify and delineate HLA associations of the beta2GPI response, we combined these data with individual patient data from three other ethnic groups including an additional 74 patients with beta2GPI response and 403 ethnically matched controls of white, African-American, and Mexican-American origin in a formal meta-analysis. The major alleles associated with anti-beta2GPI response are HLA-DQA1*03 (in particular *0301) and the HLA-DRB1*1302-DQB1*0604 haplotype, while protection against developing an anti-beta2GPI response is related primarily to the HLA-DRB1*0101-DQA1*0101 haplotype and the HLA-DRB1*1101 allele. These effects are not significantly heterogeneous across ethnic groups. The previously observed association with HLA-DQB1*0302 may simply reflect linkage disequilibrium with HLA-DQA1*0301 and the previously reported HLA-DQB1*06 effect is limited to HLA-DQB1*0604/0605, while HLA-DQB1*0602 is unlikely to be important. The meta-analysis clearly documents that the anti-beta2GPI response is determined by a few specific class II alleles and haplotypes.