Search

As part of the 7th Annual francophone workshop series on the harmonization of clinical practices in allogeneic stem cell transplantation held in Lille in September 2016, our workgroup discussed how transplant centers might follow a collective approach to coding data. This was done mainly by analyzing the study results found in the literature that do not provide clear answers. In addition, we discuss practical ways of coding for both donor and recipient HLA typing in the European bone marrow transplantation database called ProMISe which is managed by the European Society for Blood and Marrow Transplantation (EBMT).

In an attempt to harmonize clinical practices among French hematopoietic stem cell transplantation centers, the Francophone Society of Bone Marrow Transplantation and Cellular Therapy (SFGM-TC) held its sixth annual workshop series in September 2015 in Lille. This event brought together practitioners from across the country with the purpose of offering careful analysis of published studies on clinical practice issues that remain to be disputed. This article addresses the impact of HLA and KIR gene polymorphism on the outcome of the transplantation in order to optimize unrelated donor selection.

Epidermolysis bullosa acquisita (EBA) is a rare autoimmune bullous disease (AIBD). However, higher EBA incidence and predisposing genetic factor(s) involving an HLA haplotype have been suspected in some populations. This retrospective study assessed the overrepresentation of black patients with EBA, its link with HLA- DRB1*15:03 , and their clinical and immunological characteristics. Between 2005 and 2009, 7/13 (54%) EBA and 6/183 (3%) other-AIBD patients seen consecutively in our department were black ( P =10 −6 ); moreover 7/13 (54%) black patients and 6/183 (3%) white patients had EBA ( P =10 −6 ). In addition, between 1983 and 2005, 12 black patients had EBA. Finally, among the 19 black EBA patients, most of them had very atypical clinical presentations, 9 were natives of sub-Saharan Africa, 1 from Reunion Island, 7 from the West Indies, and 2 were of mixed ancestry. HLA- DRB1*15:03 allelic frequencies were 50% for African patients, significantly higher than the control population ( P −3 ), and 21% for the West Indians (nonsignificant). High EBA frequencies have already been reported in American blacks significantly associated with the HLA- DR2 . In conclusion, black-skinned patients developing EBA seem to have a genetic predisposition, and EBA should be suspected systematically for every AIBD seen in this population.

HLA matching between the donor and recipient improves the success of unrelated hematopoietic stem cell transplantation (HSCT). Because many patients in need of an unrelated transplant have only donors with mismatch, information is needed to evaluate the limits of HLA mismatching. We examined the association of survival, acute graft-versus-host disease (aGVHD) and relapse with HLA-A, -B, -C, -DRB, -DQB1, and -DPB1 mismatching in 334 patients coming from 12 French transplant centers and who received a non-T cell-depleted bone marrow graft from an unrelated donor. All patients were prepared with the use of myeloablative conditioning regimens. Our analyses demonstrate negative effects of HLA mismatching for either HLA-A, -B, -C, -DRB1, or -DQB1 loci on survival. Multivariate Cox analyses showed that a single mismatch was associated with a significant decrement in survival (P = .046, hazard ratio [HR] = 1.41, confidence interval [CI] 95% 1.1-1.98). The presence of multiple mismatches was worse for survival (P = .003, HR = 1.91, CI 95% 1.26-2.91) and severe aGVHD (grade III-IV) (P = .002, HR = 2.51, CI95% 1.41-4.46). The cumulative incidences of aGVHD and relapse in those HLA-A, -B, -C, -DRB1, and -DQB1 identical pairs with 2, 1, or 0 DPB1 incompatibilities were 63%, 50%, and 51%, and 12%, 27%, and 20%, respectively, but these differences were not statistically significant. Similar differences of aGVHD and relapse, but not statistically significant, were observed in those HLA-A, -B, -C, -DRB1, and -DQB1 identical pairs with DPB1 disparities classified into permissive or nonpermissive mismatches according to Zino’s classification based on a hierarchy of the immunogenicity of the HLA-DP molecules. “Missing killer cell immunoglobulin-like receptor (KIR) ligand” evaluated on the presence of HLA-C1, -C2, and Bw4 groups in the recipients was not associated with aGVHD, survival, and relapse in this cohort of non-T cell-depleted HSCT.

Epidemiological studies performed in different ethnic populations and family studies, notably based on a partial phenotype of the autoimmune process, indicate that genetic factors are involved in the occurrence of pemphigus. However, the precise heritability remains uncertain in the absence of twin concordance rate studies. Among the different strategies available to identify genetic factors participating in autoimmune disease susceptibility, only population studies based on case-control design have been performed in pemphigus. These studies consistently showed that MHC locus, in particular HLA class II alleles, are associated with pemphigus vulgaris and pemphigus foliaceus. Other genes of the MHC locus may also participate in disease susceptibility as shown by studies using microsatellite markers across different regions of the MHC. It is likely that other non-MHC genes are involved in the pathogenesis of pemphigus. In particular, involvement of a polymorphic variant of desmoglein 1 gene was shown to be associated with pemphigus foliaceus and to interact in an epistatic manner with MHC class II genes to contribute to the autoimmune process. Other candidate genes to which a role can be assigned in the disease pathogenesis should be considered to design case-control or family-based association studies. Genome scan studies which require a large number of multiplex families to reach statistical power, should also be considered in the endemic form of pemphigus foliaceus because of the high number of familial cases.

Pemphigus is a group of autoimmune blistering diseases caused by autoantibodies directed against keratinocyte adhesion molecules. Pemphigus vulgaris (PV) and pemphigus foliaceus (PF), in which autoantibodies bind, respectively, to desmoglein 3 and desmoglein 1, are strongly associated with HLA-class II DR4 and DR14 alleles. In paraneoplastic pemphigus (PNP), a rare variant associated with neoplasia, autoantibodies target proteins of the plakin family in addition to desmogleins 1 and 3. The presence of anti-desmoglein antibodies in all types of pemphigus raises the question of common molecular mechanisms of susceptibility, particularly similar MHC-class II allele associations, in the different forms of the disease. HLA-DRB1 typing was performed in 13 PNP patients and results were compared to those obtained from 84 healthy controls, 37 PV and 31 PF patients. Our data demonstrate a significant association of PNP with HLA-DRB1*03 allele which was found in 61.5% of the patients, whereas DRB1*04 and DRB1*14 appear not to be involved in PNP susceptibility. Therefore, the HLA-genetic background of PNP differs from that of other types of pemphigus, which suggests that distinct mechanism(s) initiate(s) the immunological response in this form of pemphigus.

Susceptibility to Pemphigus, an autoimmune disease of the skin, has been previously linked to DRB1*0402, 1401/04 and DQB1*0503 in pemphigus vulgaris (PV), to DRB1*0102, 0404, 1402/06 in endemic pemphigus foliaceus in Brazil and to DRB1*04 in Italian patients suffering from pemphigus foliaceus (PF). The disease is caused by autoantibodies against desmoglein (Dsg1 in PF, Dsg3 in PV). Molecular typing of 57 French patients suffering from PV (37) and from PF (20) confirmed previous results concerning PV and showed that DRB1*0102 and 0404 are susceptible molecules to PF in France. We have analysed the characteristics of the 'pockets' of the susceptibility-associated molecules to PV and PF and we showed that (i) in PV, two kinds of Dsg3 derived peptides may be presented by HLA-DR according to HLA polymorphism (DRB1*0402 or DRB1*14/0406), (ii) the same Dsg1 peptides may be presented by DRB1*0102, DQB1*0404 or DRB1*14 in PF, (iii) the DRB1*14/0406 PV-related molecules may be able to present Dsg1 and Dsg3 peptides thereby providing an explanation for the cases of PV with combined responses to Dsg1 and to Dsg3 which are typified by a muco-cutaneous clinical phenotype.

The genetic polymorphism of the HLA-DPB1 and DPA1 loci was studied in 60 unrelated caucasoid individuals by PCR-RFLP. The polymorphic second exon of DPB1, the third exon of DPA1, and the transmembrane DPA1 exon were specifically amplified in vitro by polymerase chain reaction (PCR). Amplified DNAs were digested with selected enzymes. Twenty patterns were obtained with DPB1 defining 20 DPB1 alleles. Thirty-nine homozygous cell lines were used as HLA-DP reference cells. The results obtained with these cell lines were compared to those obtained by PLT, RFLP, and SSO. Although three subdivisions of the allele DPA1*01 were reported, DPA1*0103 was the only represented one in the caucasoid population. In the studied population, it was the most frequent DPA1 allele (76.6%), whereas DPA1*0201 frequency is 23.3%. DPB1*0401 and DPB1*0402 are the most frequent among the DPB1 alleles (40.0% and 13.3%, respectively). This may lead to a lower HLA-DPB1 diversity among caucasoids. Certain HLA-DPB1 alleles associate exclusively with one DPA1 allele (DPB1*0401, 0402, and 0301 with DPA1*01 and DPB1*0101, 0501, and 1701 with DPA1*0201) whereas the others can associate with both DPA1 alleles. This by itself can create another kind of polymorphism, indicating the importance of HLA-DPA1 typing. Thus, PCR-RFLP seems to be one of the best DNA typing methods: it represents direct, accurate, fast, and nonradioactive typing for both HLA-DPA1 and HLA-DPB1 alleles.