A new strategy for systematically classifying HLA alleles into serological specificities

Hiding in plain sight: Misinterpretation of immunogenic DPB epitopes within G/P groups

The clinical impact of HLA DP antibodies is poorly understood, resulting in variable clinical strategies for transplant candidates and recipients with donor-directed HLA-DP antibodies. Complicating matters further, the DPB naming convention is not based on allelic homology and requires sequence alignments to identify potential immunogenic epitopes. Historically, G and P codes, which consolidated alleles that were identical over Exon 2, were used to simplify the reporting of HLA Class II typing as differences outside of Exon 2 have not been considered immunogenic (i.e., able to induce an antibody response). Herein, we present four cases demonstrating that polymorphisms at codons 96R/K and 170I/T, in Exon 3 of DPB, are targets for alloantibody recognition. These regions "hide in plain sight" due to the current use of G/P code-level typing, potentially leading to incorrect compatibility assessments (i.e., virtual crossmatches) and misinterpreted antibody responses. The unintentional crossing of an HLA-DPB donor-specific antibody (DSA) in a solid organ or hematopoietic stem cell transplant may lead to unforeseen deleterious clinical outcomes. Our data underscore the complexities of DPB histocompatibility assessments and highlight the need for adaptable systems that align with evolving research and clinical outcomes.

A new strategy for systematically classifying HLA alleles into serological specificities: Update and refinement

HLA antigens were historically defined according to the unique reactivity pattern of cells expressing HLA molecules with distinctive clusters of allo-antisera and/or monoclonal antibodies. Subsequently, amino acid residues determining epitopes (DEP) in the HLA molecule were correlated with reactivity patterns. In current clinical practice, the presence of allo-antibodies is assessed using Luminex-based solid phase single antigen bead (SAB) assays for transplantation. Recently, novel antigens were proposed for HLA molecules with DEP patterns that do not match any serologically defined antigens recognised by the WHO Nomenclature Committee. To validate the antigens, mean fluorescence intensity values of SABs tested on >13,000 patients' sera were extracted from clinical databases and analysed by scatter plots using a linear regression model. We found that when two proteins were considered as the same antigen in the original study, for example, HLA-A*02:01 and -A*02:06, their correlation ranked among the highest values at each locus. In contrast, discrete asymmetric outliers were observed when there were different antigens, for example, HLA-A*30:01 and -A*30:02, allowing validation and confirmation of 20 novel antigens for HLA-A, -B, -C and -DR. The outliers were confirmed to be true or false by flow cytometric crossmatches. In addition to the previously defined residues for antigen assignments, findings suggest that further distinction should be made for common antigens by including the substitutions at residue 67 of HLA-B, 67 and 74 of -DR. These serologic analyses can be applied systematically to identify and confirm novel antigens. These developments will lead to designing optimal SAB panels and further improving virtual donor-specific antibodies assessment.

Quantifying uncertainty of molecular mismatch introduced by mislabeled ancestry using haplotype-based HLA genotype imputation

Introduction

Modern histocompatibility algorithms depend on the comparison and analysis of high-resolution HLA protein sequences and structures, especially when considering epitope-based algorithms, which aim to model the interactions involved in antibody or T cell binding. HLA genotype imputation can be performed in the cases where only low/intermediate-resolution HLA genotype is available or if specific loci are missing, and by providing an individuals' race/ethnicity/ancestry information, imputation results can be more accurate. This study assesses the effect of imputing high-resolution genotypes on molecular mismatch scores under a variety of ancestry assumptions.

Methods

We compared molecular matching scores from "ground-truth" high-resolution genotypes against scores from genotypes which are imputed from low-resolution genotypes. Analysis was focused on a simulated patient-donor dataset and confirmed using two real-world datasets, and deviations were aggregated based on various ancestry assumptions.

Results

We observed that using multiple imputation generally results in lower error in molecular matching scores compared to single imputation, and that using the correct ancestry assumptions can reduce error introduced during imputation.

Discussion

We conclude that for epitope analysis, imputation is a valuable and low-risk strategy, as long as care is taken regarding epitope analysis context, ancestry assumptions, and (multiple) imputation strategy.

Proficiency testing within Eurotransplant

Eurotransplant is responsible for the international allocation of organs between eight countries in Europe. All HLA laboratories affiliated to Eurotransplant must be EFI or ASHI-accredited and must participate in the Eurotransplant external proficiency testing (EPT) program, organized by the Eurotransplant Reference Laboratory (ETRL). EPT within Eurotransplant has a long tradition, starting in 1978. The current EPT program consists of the following schemes: HLA typing including serology, CDC crossmatching, HLA-specific antibody detection, and identification. Participants enter the results of laboratory tests using a web-based application. Assessed results are visible on the website. An additional component called "patient-based cases" runs since 2016. Results are summarized and published on the EPT website. Furthermore, these results are discussed during the annual extramural tissue typers meeting, which is organized by the ETRL. Thanks to this EPT program, the performance of all HLA laboratories affiliated to Eurotransplant can be monitored and corrected, if necessary. Because all affiliated laboratories are assessed in the same EPT program, where these laboratories show to be consistent in most of their results, Eurotransplant EPT has proven to be an efficient tool to create a more uniform level of quality of histocompatibility testing within Eurotransplant.

Identification of the novel HLA-B*56:100 allele by next-generation sequencing

HLA-B*56:100 differs from HLA-B*56:20:02 by one nucleotide substitution at codon 147.2 in exon 3.

Introduction of the donor centre virtual crossmatch in Eurotransplant

On 24 January 2023, Eurotransplant has introduced the virtual crossmatch for kidney and pancreas allocation as a better alternative for the physical Complement Dependent Cytotoxicity (CDC) crossmatches at the donor centre, which were associated with a longer cold ischaemia time and false positive reactions. For the time being, the physical CDC crossmatch at the recipient centre will remain in place as the final histocompatibility check. While Eurotransplant is certainly not the first organ allocation organisation to introduce virtual crossmatching, several novel aspects have been introduced, such as calculation of the virtual panel reactive antibody (vPRA) on 11 loci at the second-field level in addition to the serological broad and split level, electronic HLA typing data transmission using Histoimmunogenetics Markup Language (HML) file format, and the actual virtual crossmatch based on ambiguous, second-field HLA typing of the donor on all 11 loci. This short communication will focus on these novel aspects of the virtual crossmatch in Eurotransplant.

Resolving unknown nucleotides in the IPD-IMGT/HLA database by extended and full-length sequencing of HLA class I and II alleles

In the past, identification of HLA alleles was limited to sequencing the region of the gene coding for the peptide binding groove, resulting in a lack of sequence information in the HLA database, challenging HLA allele assignment software programs. We investigated full-length sequences of 19 HLA class I and 7 HLA class II alleles, and we extended another 47 HLA class I alleles with sequences of 5' and 3' UTR regions that were all not yet available in the IPD-IMGT/HLA database. We resolved 8638 unknown nucleotides in the coding sequence of HLA class I and 2139 of HLA class II. Furthermore, with full-length sequencing of the 26 alleles, more than 90 kb of sequence information was added to the non-coding sequences, whereas extension of the 47 alleles resulted in the addition of 5.5 kb unknown nucleotides to the 5' UTR and > 31.7 kb to the 3' UTR region. With this information, some interesting features were observed, like possible recombination events and lineage evolutionary origins. The continuing increase in the availability of full-length sequences in the HLA database will enable the identification of the evolutionary origin and will help the community to improve the alignment and assignment accuracy of HLA alleles.

Impact of allele-specific anti-human leukocyte antigen class I antibodies on organ allocation

Technological advances in the field of histocompatibility have allowed us to define anti-human leukocyte antigen (HLA) antibody specificity at the allelic level. However, how allele-specific antibodies affect organ allocation is poorly studied. We examined allelic specificities of class I HLA antibodies in 6726 consecutive serum samples from 2953 transplant candidates and evaluated their impact on the corresponding crossmatch and organ allocation. Out of 17 class I HLA antigens represented by >1 allele in the LABScreen single antigen bead assay, 12 had potential allele-specific reactivity. Taking advantage of our unbiased cohort of deceased donor-candidate testing (123,135 complement-dependent cytotoxicity crossmatches between 2014 and 2017), we estimated that the presence of allele-specific antibody detected using a single antigen bead assay (median fluorescence intensity, >3000) against only the rare allele was a poor predictor of a positive complement-dependent cytotoxicity crossmatch, with a positive predictive value of 0% to 7%, compared with 52.5% in allele-concordant class I HLA antibodies against A or B locus antigens. Further, we confirmed allele-specific reactivity using flow crossmatch in 3 scenarios: A11:01/A11:02, A68:01/A68:02, and B44:02/B44:03. Our results suggest that allele-specific antibodies may unnecessarily exclude transplant candidates (up to 10%) from organ offers by overcalling unacceptable antigens; incorporation of selective reactivity pattern in allocation may promote precision matching and more equitable allocation.