T cell neoepitope discovery in colorectal cancer by high throughput profiling of somatic mutations in expressed genes

OBJECTIVE

Patient-specific (unique) tumour antigens, encoded by somatically mutated cancer genes, generate neoepitopes that are implicated in the induction of tumour-controlling T cell responses. Recent advancements in massive DNA sequencing combined with robust T cell epitope predictions have allowed their systematic identification in several malignancies.

DESIGN

We undertook the identification of unique neoepitopes in colorectal cancers (CRCs) by using high-throughput sequencing of cDNAs expressed by standard cancer cell cultures, and by related cancer stem/initiating cells (CSCs) cultures, coupled with a reverse immunology approach not requiring human leukocyte antigen (HLA) allele-specific epitope predictions.

RESULTS

Several unique mutated antigens of CRC, shared by standard cancer and related CSC cultures, were identified by this strategy. CD8⁺ and CD4⁺ T cells, either autologous to the patient or derived from HLA-matched healthy donors, were readily expanded in vitro by peptides spanning different cancer mutations and specifically recognised differentiated cancer cells and CSC cultures, expressing the mutations. Neoepitope-specific CD8⁺ T cell frequency was also increased in a patient, compared with healthy donors, supporting the occurrence of clonal expansion in vivo.

CONCLUSIONS

These results provide a proof-of-concept approach for the identification of unique neoepitopes that are immunogenic in patients with CRC and can also target T cells against the most aggressive CSC component.

Effects of transmembrane region variability on cell surface expression and allorecognition of HLA-DP3

The functional relevance of polymorphisms outside the peptide binding groove of HLA molecules is poorly understood. Here we have addressed this issue by studying HLA-DP3, a common antigen relevant for functional matching algorithms of unrelated hematopoietic stem cell transplantation (HSCT) encoded by two transmembrane (TM) region variants, DPB1(*)03:01 and DPB1(*)104:01. The two HLA-DP3 variants were found at a overall allelic frequency of 10.4% in 201 volunteer stem cell donors, at a ratio of 4.2:1. No significant differences were observed in cell surface expression levels of the two variants on B lymphoblastoid cell lines (BLCL), primary B cells or monocytes. Three different alloreactive T cell lines or clones showed similar levels of activation marker CD107a and/or CD137 upregulation in response to HLA-DP3 encoded by DPB1(*)03:01 and DPB1(*)104:01, either endogenously on BLCL or after lentiveral-vector mediated transfer into the same cellular background. These data provide, for the first time, direct evidence for a limited functional role of a TM region polymorphism on expression and allorecognition of HLA-DP3 and are compatible with the notion that the two variants can be considered as a single functional entity for unrelated stem cell donor selection.

Effect of T-cell-epitope matching at HLA-DPB1 in recipients of unrelated-donor haemopoietic-cell transplantation: a retrospective study

BACKGROUND

The risks after unrelated-donor haemopoietic-cell transplantation with matched HLA-A, HLA-B, HLA-C, HLA-DRB1, HLA-DQB1 alleles between donor and recipient (10/10 matched) can be decreased by selection of unrelated donors who also match for HLA-DPB1; however, such donors are difficult to find. Classification of HLA-DPB1 mismatches based on T-cell-epitope groups could identify mismatches that might be tolerated (permissive) and those that would increase risks (non-permissive) after transplantation. We did a retrospective study to compare outcomes between permissive and non-permissive HLA-DPB1 mismatches in unrelated-donor haemopoietic-cell transplantation.

METHODS

HLA and clinical data for unrelated-donor [corrected] transplantations submitted to the International Histocompatibility Working Group in haemopoietic-cell transplantation were analysed retrospectively. HLA-DPB1 T-cell-epitope groups were assigned according to a functional algorithm based on alloreactive T-cell crossreactivity patterns. Recipients and unrelated donors matching status were classified as HLA-DPB1 match, non-permissive HLA-DPB1 mismatch (those with mismatched T-cell-epitope groups), or permissive HLA-DPB1 mismatch (those with matched T-cell-epitope groups). The clinical outcomes assessed were overall mortality, non-relapse mortality, relapse, and severe (grade 3-4) acute graft-versus-host disease (aGvHD).

FINDINGS

Of 8539 transplantations, 5428 (64%) were matched for ten of ten HLA alleles (HLA 10/10 matched) and 3111 (36%) for nine of ten alleles (HLA 9/10 matched). Of the group overall, 1719 (20%) were HLA-DPB1 matches, 2670 (31%) non-permissive HLA-DPB1 mismatches, and 4150 (49%) permissive HLA-DPB1 mismatches. In HLA 10/10-matched transplantations, non-permissive mismatches were associated with a significantly increased risk of overall mortality (hazard ratio [HR] 1·15, 95% CI 1·05-1·25; p=0·002), non-relapse mortality (1·28, 1·14-1·42; p<0·0001), and severe aGvHD (odds ratio [OR] 1·31, 95% CI 1·11-1·54; p=0·001), but not relapse (HR 0·89, 95% CI 0·77-1·02; p=0·10), compared with permissive mismatches. There were significant differences between permissive HLA-DPB1 mismatches and HLA-DPB1 matches in terms of non-relapse mortality (0·86, 0·75-0·98; p=0·03) and relapse (1·34, 1·17-1·54; p<0·0001), but not for overall mortality (0·96, 0·87-1·06; p=0·40) or aGvHD (OR 0·84, 95% CI 0·69-1·03; p=0·09). In the HLA 9/10 matched population, non-permissive HLA-DPB1 mismatches also increased the risk of overall mortality (HR 1·10, 95% CI 1·00-1·22; p=0·06), non-relapse mortality (1·19, 1·05-1·36; p=0·007), and severe aGvHD (OR 1·37, 95% CI 1·13-1·66; p=0·002) compared with permissive mismatches, but the risk of relapse was the same in both groups (HR 0·93, 95% CI 0·78-1·11; p=0·44). Outcomes for HLA 10/10-matched transplantations with non-permissive HLA-DPB1 mismatches did not differ substantially from those for HLA 9/10-matched transplantations with permissive HLA-DPB1 mismatches or HLA-DPB1 matches.

INTERPRETATION

T-cell-epitope matching defines permissive and non-permissive HLA-DPB1 mismatches. Avoidance of an unrelated donor with a non-permissive T-cell-epitope mismatch at HLA-DPB1 might provide a practical clinical strategy for lowering the risks of mortality after unrelated-donor haemopoietic-cell transplantation.

FUNDING

National Institutes of Health; Associazione Italiana per la Ricerca sul Cancro; Telethon Foundation; Italian Ministry of Health; Cariplo Foundation; National Cancer Institute; National Heart, Lung and Blood Institute; National Institute of Allergy and Infectious Diseases; Office of Naval Research; IRGHET Paris; Swedish Cancer Society; Children's Cancer Foundation; Swedish Research Council; Cancer Society in Stockholm; Karolinska Institutet; and Leukemia and Lymphoma Society.

Molecular and functional characterization of allogantigen-specific anergic T cells suitable for cell therapy

BACKGROUND

CD4(+) regulatory T cells are a specialized subset of T cells that actively control immune responses. Several experimental protocols have been used to expand natural regulatory T cells and to generate adaptive type 1 regulatory T cells for regulatory T-cell-based therapies.

DESIGN AND METHODS

The ability of exogenous recombinant human interleukin-10 to induce alloantigen-specific anergy in T cells was investigated and compared to that of interleukin-10 derived from tolerogenic dendritic cells, in mixed lymphocyte cultures. A detailed characterization of the effector functions of the resulting anergized T cells is reported.

RESULTS

Interleukin-10, whether exogenous or derived from tolerogenic dendritic cells, induces a population of alloantigen-specific T cells (interleukin-10-anergized T cells) containing type 1 regulatory T cells, which are anergic and actively suppress alloantigen-specific effector T cells present within the mixed population. Interleukin-10-induced anergy is transforming growth factor-β independent, and is associated with a decreased frequency of alloantigen-specific cytotoxic T lymphocyte precursors, but interleukin-10-anergized T cells are still responsive to third-party, bacterial, and viral antigens. Tolerogenic dendritic cells are more powerful than exogenous interleukin-10 in generating type 1 regulatory T-cell precursors, and are also effective in the context of HLA-matched donors.

CONCLUSIONS

Based on these studies, we have developed an efficient and reproducible in vitro method to generate antigen-specific type 1 regulatory T-cell precursors starting from total peripheral blood cells with minimal cell manipulation and suitable for generating type 1 regulatory T cells for regulatory T-cell-based therapies.

Platelet transfusion refractoriness in highly immunized beta thalassemia children undergoing stem cell transplantation

Immune-mediated refractoriness to platelet transfusion is a major problem in patients undergoing HSCT. In a cohort of 50 pediatric patients affected by beta thalassemia coming from Middle East countries, we experienced a high incidence of refractoriness because of anti-HLA antibodies during post-HSCT aplasia. In a risk factors analysis, factors predicting a negative transfusion outcome were presence of spleen and the number of anti-HLA antibodies. We adopted a policy to select platelet donors by avoiding HLA antigens against which the patient had specific antibodies. Transfusion of dedicated units resulted in 26% refractoriness compared to 74% to random units (p < 0.0001). When dedicated transfusions were used, the presence of spleen did not influence transfusion outcome. Analyzing transfusion outcome depending on the degree of HLA match and ABO compatibility, 76% successful transfusions were obtained with HLA-matched- ABO compatible followed by 67% in HLA-1mismatch- ABO compatible or HLA-matched- ABO incompatible and by 46% in HLA-1mismatch- ABO incompatible. In conclusion, we provide evidence that the selection of platelet donors according to patient characteristics, anti-HLA antibodies and ABO matching, is successful in reducing platelet refractoriness in heavily alloimmunized thalassemia patients undergoing transplantation.

Frequency and targeted detection of HLA-DPB1 T cell epitope disparities relevant in unrelated hematopoietic stem cell transplantation

The majority of unrelated donor (UD) hematopoietic stem cell (HSC) transplants are performed across HLA-DP mismatches, which, if involving disparity in a host-versus-graft (HVG) direction for an alloreactive T cell epitope (TCE), have been shown by our group to be associated with poor clinical outcome in 2 cohorts of patients transplanted for hematopoietic malignancies and beta-thalassemia, respectively. Using site-directed mutagenesis of DPB1*0901, we show here that the TCE is abrogated by the presence of amino acids LFQG in positions 8-11 of the DP beta-chain. Based on this and on alloreactive T cell responsiveness, we have determined the presence or absence of the TCE for 72 DPB1 alleles reported in the ethnic groups representative of the worldwide UD registries, and predict that 67%-87% (mean 77%) of UD recipient pairs will not present a DPB1 TCE disparity in the HVG direction. We developed and validated in 112 healthy Italian blood donors an innovative approach of DPB1 epitope-specific typing (EST), based on 2 PCR reactions. Our data show that DPB1 TCE disparities may hamper the clinical success of a considerable number of transplants when DPB1 matching is not included into the donor selection criteria, and that a donor without DPB1 TCE disparities in the HVG direction can be found for the majority of patients. Moreover, the study describes the first protocol of targeted epitope-specific DPB1 donor-recipient matching for unrelated HSC transplantation. This protocol will facilitate large-scale retrospective clinical studies warranted to more precisely determine the clinical relevance of DPB1 TCE disparities in different transplant conditions.

Therapeutic and Diagnostic Applications of Minor Histocompatibility Antigen HA-1 and HA-2 Disparities in Allogeneic Hematopoietic Stem Cell Transplantation: A Survey of Different Populations

Minor histocompatibility antigens (mHags) HA-1 and HA-2 are encoded by biallelic loci, with immunogenic variants, HA-1H and HA-2V, which induce strong HLA-A2-restricted alloreactive T-cell responses, and nonimmunogenic counterparts, HA-1R and HA-2M, which represent functional null alleles that are poorly presented by HLA class I molecules. HA-1 and HA-2 are potential targets of selective graft-versus-leukemia and graft-versus-tumor reactivity after allogeneic hematopoietic stem cell transplantation (HSCT); however, these applications are restricted to a limited number of patients. Here, we show that a far more frequent application of HA-1 and HA-2 disparity relies on their use as markers for the state of host chimerism after allogeneic HSCT. We have determined allelic frequencies of 29.3% and 70.7% for HA-1H and HA-1R, respectively, and of 83.7% and 16.3% for HA-2V and HA-2M, respectively, in >200 healthy individuals from northern Italy. Similar frequencies were observed in nearly 100 patients affected by hematologic malignancies or solid tumors, thus showing that HA-1 and HA-2 variability are not associated with the presence of cancer. On the basis of these data, we predict that HA-1 and HA-2 can be used in 32.8% and 23.5% of Italian transplant patients, respectively, as markers for the state of host chimerism, whereas exploitation of disparity for these mHags for targeted immunotherapy will be possible in 10.7% and 1.1% of Italian patients, respectively. Retrospective HA-2 typing of bone marrow aspirates obtained from a patient during complete remission or recurrence of acute myeloid leukemia after haploidentical HSCT showed the feasibility of using HA-2 as a surrogate marker for disease monitoring. Because of an apparent north-south gradient for HA-1 allelic frequencies, with higher frequencies for the HA-1H variant reported in white populations from Southern Europe as compared with Northern Europe and North America, the diagnostic applicability of HA-1 disparity will be slightly more frequent in transplant patients from the north. Taken together, our data show that determination of HA-1 and HA-2 variability can be an important parameter for the selection of allogeneic stem cell donors, in particular for patients affected by hematologic malignancies without a tumor-specific molecular marker.

Graft rejection after unrelated donor hematopoietic stem cell transplantation for thalassemia is associated with nonpermissive HLA-DPB1 disparity in host-versus-graft direction

The success of allogeneic hematopoietic stem cell transplantation (HSCT) from matched unrelated donors (UDs) for beta-thalassemia may be hampered by the occurrence of graft rejection. Here, we show that the rate of this complication can be reduced by selecting 5-loci HLA-matched donors without nonpermissive mismatches at HLA-DPB1, defined according to an algorithm previously described and based on principles of central T-cell tolerance. Seventy-two consecutive patients and their UDs, prospectively selected for matching at the allelic level for HLA-A, -B, -C, -DRB, and -DQB1 loci, were enrolled in the analysis. These pairs were either DPB1 matched/permissively mismatched (n = 45, control group) or had at least one nonpermissive DPB1 mismatch in the host-versus-graft (HvG; n = 17) or in the graft-versus-host (GvH; n = 10) direction. In multivariate analysis, the risk of rejection was significantly increased in the group with HvG disparity (RR = 7.42; 95% CI = 1.29-42.68; P = .02) as compared to the control group. A lower, statistically significant, probability of thalassemia-free survival was found in patients belonging to the HvG group as compared to controls (RR = 5.15; 95% CI = 1.58-16.82; P = .01). These data suggest that in patients with thalassemia, the incidence of graft failure after HSCT may be reduced by appropriate selection of UDs, with such selection taking into account the functional rules of immunogenetics.

Rapid detection of all HLA-B*27 alleles (B*2701-B*2725) by group-specific polymerase chain reaction

Human leucocyte antigen-B*27 is strongly associated with a number of rheumatic diseases, including ankylosing spondylitis and reactive arthritis. Targeted detection of the B*27 group by molecular methods is hampered by the extreme heterogeneity of the serological B*27 group. Here, we describe a simple, rapid sequence-specific primer-based method for detection of all 28 B*27 alleles defined to date. The method involves an initial screening with two sequence-specific polymerase chain reactions (PCRs), which has to be followed by two additional PCR amplifications in samples carrying a few rare subtypes of B*27, B*4202 or B*7301. The described protocol should be useful for laboratories involved in diagnostics and research of rheumatoid diseases.

A T-cell epitope encoded by a subset of HLA-DPB1 alleles determines nonpermissive mismatches for hematologic stem cell transplantation

The importance of HLA-DPB1 matching for the outcome of allogeneic hematologic stem cell (HSC) transplantation is controversial. We have previously identified HLA-DPB1*0901 as a target of cytotoxic T cells mediating in vivo rejection of an HSC allograft. Here we show that HLA-DPB1*0901 encodes a T-cell epitope shared by a subset of DPB1 alleles that determines nonpermissive mismatches for HSC transplantation. Several T-cell clones obtained from the patient at the time of rejection showed HLA-DP restricted recognition of allogeneic targets expressing HLA-DPB1*0901, *1001, *1701, *0301, *1401, and *4501, but not other alleles. Based on these findings, we developed an algorithm for prediction of nonpermissive HLA-DPB1 mismatches. Retrospective evaluation of 118 transplantations showed that the presence of nonpermissive HLA-DPB1 mismatches was correlated with significantly increased hazards of acute grade II to IV graft-versus-host disease (HR = 1.87, P =.046) and transplantation-related mortality (HR = 2.69, P =.027) but not relapse (HR = 0.98, P =.939), as compared with the permissive group. There was also a marked but statistically not significant increase in the hazards of overall mortality (HR = 1.64, P =.1). These data suggest that biologic characterization of in vivo alloreactivity can be a tool for definition of clinically relevant nonpermissive HLA mismatches for unrelated HSC transplantation.

Peripheral blood stem cell allograft rejection mediated by CD4(+) T lymphocytes recognizing a single mismatch at HLA-DP beta 1*0901

Little is known about the molecular characteristics of alloantigens recognized by alloreactive T cells mediating hematologic stem cell graft rejection. In particular, it has never been shown that such alloantigens can be encoded by HLA-DP beta alleles. Indeed, matching for HLA-DP antigens is generally not considered to be of functional importance for the outcome of allogeneic bone marrow or peripheral blood stem cell transplantation. In this study, a case of peripheral blood stem cell allograft rejection was investigated in which the patient and donor differed for a single mismatch at HLA-DP in the rejection direction. Patient-derived T lymphocytes circulating at the time of rejection showed direct ex vivo cytotoxic activity against donor-derived B-lymphoblastoid cells as well as other HLA-DP beta 1*0901--expressing targets. The presence of HLA-DP beta 1*0901--specific effectors in vivo was further confirmed by in vitro stimulation experiments. CD4(+) T-cell lines and clones with specific cytotoxic activity against HLA-DP beta 1*0901--expressing targets including donor B-lymphoblastoid cells were generated both by nonspecific and by donor-specific in vitro stimulation. Taken together, these data demonstrate that HLA-DP can be the target antigen of cytotoxic CD4(+) T lymphocytes involved in peripheral blood stem cell allograft rejection. (Blood. 2001;98:1122-1126)

Identification of tumor antigen-specific cytotoxic T lymphocytes cross-recognizing allogeneic major histocompatibility class I molecules

Adoptive immunotherapy of cancer utilizes tumor antigen-specific cytotoxic T lymphocytes (CTL) as mediators of a targeted anti-tumor effect. In this study, we show that such CTL can be able to cross-recognize allogeneic major histocompatibility complex (MHC) molecules in a phenomenon of molecular mimicry. A self histo-leukocyte antigen (HLA) A*0201-restricted CTL specific for peptide MT27-35 from the human differentiation antigen Melan-A/MART-1 was shown to cross-recognize allogeneic A*0220 molecules which differ from syngeneic A*0201 for a single amino acid substitution at position 66 of the antigen-binding groove. A*0220 molecules were recognized on a variety of human cells of different histological origin but not on COS-7 cells. A second self-A*0201-restricted CTL, specific for peptide D10/6-271 encoded by the tumor-specific DAM-gene family, was shown to cross-recognize allogeneic B*3701 molecules which differ from syngeneic A*0201 by 32 amino acids in the peptide antigen-binding cleft. B*3701 molecules were recognized on a variety of cell types including COS-7 cells. These data raise new safety issues for clinical trials of cancer immunotherapy using adoptive transfer of in vitro generated, allogeneic CTL with specific anti-tumor activity.

Molecular characterization of HLA class I in Colombians carrying HLA-A2: high allelic diversity and frequency of heterozygotes at the HLA-B locus

Polymerase chain reaction using sequence-specific oligonucleotide probes (PCR-SSOP) typing was used to analyze HLA class I A, B and C alleles in three different Colombian populations. Fifty-nine samples were from Hispano-American Mestizos living in the urban areas of Cali (referred to here as Aso population). Forty-four and thirty samples were from the African Black populations of Zacarias (Zac) and Punta Soldado (PS), respectively. Samples were selected for expression of HLA-A2 by monoclonal antibody staining and allele-specific hybridization, and their HLA-A2 subtype distribution has been reported previously. Although only a limited number of samples was analyzed, the data suggest the existence of a remarkable degree of HLA class I polymorphism in the populations studied, with representatives of most serological classes. Despite their common African origin, the populations Zac and PS, both resident in malaria endemic regions, showed some striking differences in allelic distribution for all three class I loci. Furthermore, the samples from Aso and PS, but not Zac, showed a low percentage of blank alleles at the HLA-B locus (0 and 0.4%, respectively), suggesting the possibility of a heterozygote advantage for HLA-B alleles in Colombian populations.

Accurate spatial and temporal transgene expression driven by a 3.8-kilobase promoter of the bovine beta-casein gene in the lactating mouse mammary gland

The spatial, temporal, and hormonal pattern of expression of the beta-casein gene is highly regulated and confined to the epithelial cells of the lactating mammary gland. Previous studies have shown that 1.7 kb of the bovine beta-casein promoter were able to drive cell-specific and hormone-dependent expression to a mouse mammary cell line but failed to induce accurate expression to the mammary gland of transgenic mice. We investigated here the ability of 3.8 kb of the bovine beta-casein gene promoter to drive the expression of the human growth hormone (hGH) gene in transgenic mice. A Northern blot analysis using total RNA obtained from different tissues of lactating and nonlactating females revealed the presence of hGH mRNA only in the mammary gland of lactating females. hGH mRNA was not detectable in the mammary gland of virgin females or males. A developmental analysis showed that hGH mRNA only peaked on parturition, resembling more closely the bovine beta-casein temporal expression pattern rather than the murine. In situ hibridization studies performed on mammary gland sections showed that the cellular pattern of hGH expression was homogeneous in all lobules from heterozygous and homozygous transgenic mice. Silver grain counts on the tissue sections highly correlated with the hGH contents in the milk determined by radioimmunoassay (r = 0.996). Thus 3.8 kb of the bovine beta-casein promoter direct a high-level expression of a reporter gene to the lactating mammary gland of transgenic mice in a tissue-specific and developmentally regulated manner.

HLA-A*02 subtype distribution in Caucasians from northern Italy: identification of A*0220

This study describes a comprehensive easy to perform PCR-SSOP typing approach suitable for complete genomic subtyping of HLA-A*02. A single 1.6 kb PCR-amplificate spanning exons 2, 3 and 4 of the HLA-A*02 gene was used for hybridization with a panel of twenty-four SSOPs. This allowed unequivocal assignment of all so far known HLA-A2 subtypes, including A*0209 and A*0215N which differ for nucleotide substitutions in exon 4, without the need for two separate amplifications. Using this approach, HLA-A*02 subtype distribution was analyzed in 218 samples from unrelated, healthy individuals from northern Italy enrolled in the Italian Bone Marrow Registry and typed as HLA-A2 by serology or generic molecular analysis. As expected, A*0201 was found in the majority (92.6%) of samples. However, a significant number (6.8%) of individuals carried A*0205. Furthermore, A*0202, A*0208, A*0209 and A*0217, so far not described in Caucasians, were detected in a low number of samples (frequency ranging from 0.45% to 1.8%). Finally, a novel HLA-A*02 subtype, A*0220, was detected in 0.9% of the samples. As confirmed by DNA sequencing of exons 2 and 3, this allele is identical to A*0201 except for a single nucleotide substitution in codon 66 which changes the predicted amino acid sequence form Lys to Asn. The findings of this study have implications for the selection of HLA-A*02+ donors in unrelated bone marrow transplantation and of patients for specific immuno-therapy with HLA-A*02 restricted peptide vaccines.

Complete generic and extensive fine-specificity typing of the HLA-B locus by the PCR-SSOP method

This study describes sequence specific oligonucleotide probe (SSOP) typing of hypervariable regions in exons 2 and 3 of HLA-B locus genes. A single HLA-B specific PCR-product spanning from bp 84 in exon 2 to bp 241 in exon 3 was used for dot blot hybridization to forty-seven chemiluminescent labeled oligonucleotide probes. Thirty-one of these probes were derived from four hypervariability zones in exon 3 of HLA-B genes and covered most known sequence polymorphisms within these regions. In addition, sixteen probes derived from polymorphic regions in exon 2 were used to discriminate alleles not unequivocally characterized by the exon 3 based probes. This SSOP panel gave rise to eighty-six distinct hybridization patterns that could be used to unequivocally define all WHO-designated serological HLA-B specificities except for HLA-B54 in all homo- and heterozygous combinations. Furthermore, sixty-six out of ninety-seven molecularly defined HLA-B subtypes were characterized by unique hybridization patterns in all homozygous and most (possibly all) heterozygous combinations. The reproducibility of these results was confirmed by analysis of forty-four Workshop reference cell lines and of seventy-eight randomly chosen samples (one-hundred forty-seven alleles) from unrelated individuals serologically typed in the laboratory. For sixty-five samples (one-hundred-thirty-three alleles), molecular typing confirmed the results obtained by serology and allowed molecular subtype assignment for ninety-one alleles tested. A serologically blank allele could be defined by molecular analysis in three cases. The method presented here for molecular typing of the HLA-B locus can be used as an alternative to biochemical methods such as one-dimensional isoelectric focusing for assignment of serologically cross-reacting HLA-B molecules as well as for subtype characterization of a large variety of HLA-B alleles.