Oral Abstract

Abstract

Christopher Byrnes¹, Andrew Hastings², Ira Lacej², Renuka Palanicawandar², Eduardo Olavarria², Arthi Anand¹

¹Histocompatbility and Immunogenetics, North West London Pathology, London, UK; ²Department of Haematology, Imperial College Healthcare NHS Trust, London, UK

Relapse is a major cause of treatment failure in haploidentical haematopoietic progenitor cell transplant (HPCT) with PTCy. Natural killer cells are the first to reconstitute post HSCT, suppressing graft versus host disease and mediating the graft versus leukaemia effect, driven by killer cell immunoglobulin-like receptors (KIRs). Emerging research suggests that donor KIR genotype may influence outcomes of haploidentical HPCT. Haploidentical donors are readily available, and donor selection could hinge on predicted KIR NK cell alloreactivity. This study investigates whether donors with greater KIR B motifs associate with greater relapse free survival (RFS), overall survival (OS), non-relapse mortality (NRM), acute graft versus host disease (GvHD) and infection.

Following KIR genotyping, seventy-seven haploidentical donor recipient pairs (myeloablative n = 30, RIC n = 47) with various haematological malignancies are categorised into neutral (n = 49) or better and best (n = 28), using KIR B motif content. Kaplan-Meier and Cox Regression survival functions are performed to investigate associations with potential outcomes.

Our results show that the better and best category has significantly reduced RFS (p = .004) (HR 4.13, 95% CI 1.45–11.74: p = .008) and trend towards greater infections (p = .080) (HR 2.09, 95% CI 0.90–4.84: p < .1), decreasing OS (p = .008) (HR2.44, 95% confidence interval [CI] 1.24–4.81: p = .01), without impacting GvHD or NRM.

In our study, neutral donor outcomes are favourable in T cell depleted haplo-HPCT, potentially due to alloresponsive donor NK cells being targeted by immunosuppressive PTCy treatment delaying reconstitution. Further studies focusing on a homogenous pathology and treatment modality would determine the utility of KIR B content calculator in haploidentical donor selection.

Miss Rebecca Cope^1,2, Rhea McArdle^1,2, Veena Surendrakumar³, Afzal Chaudhry^1,4, Vasilis Kosmoliaptsis³, Gavin Pettigrew³, Stephen Pettit⁵, Peter Riddle⁵, Sarah Peacock¹

¹Addenbrooke's Hospital, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK; ²Faculty of Biology, Medicine and Health, Division of Medical Education, School of Medical Sciences, The University of Manchester, Manchester, UK; ³Department of Surgery, University of Cambridge, Cambridge, UK; ⁴Department of Medicine, University of Cambridge, Cambridge, UK; ⁵Royal Papworth Hospital, Cambridge, UK

Development of de novo donor-specific antibody (dnDSA) is linked to poor outcomes after cardiac transplantation, including antibody-mediated rejection and cardiac allograft vasculopathy (CAV). This study aimed to assess the role of dnDSA and pre-transplant immunological risk stratification, as defined by the cardiothoracic advisory group (CTAG) guidelines (2013), in outcome.

We retrospectively analysed 187 transplants performed between 2009 and 2015. A new research tool, the mismatch data aggregator (MDA), was validated and used with human leukocyte antigen (HLA) antibody screening and typing data to analyse dnDSA production and its effect on outcome. We also compared immunological risk stratification pre- and post-2013 and assessed the impact of risk on outcome.

Twenty-eight percent of patients developed dnDSA, the most common being to HLA-DQ (29.7%). CAV was more common in patients who developed dnDSA (33% vs. 13%). Thirteen transplants above standard immunological risk were performed. These had a higher risk of death (38.5% vs. 24.3%) with similar causes of death to standard risk transplants. Deaths attributed to factors including infection and malignancy were the largest cause of death across both analyses. No change in type of transplants performed with CTAG guideline introduction was identified.

This study supports the detrimental impact of dnDSA on cardiac transplant outcome, particularly dnDSA to HLA-DQ. However, there is a need for larger studies to identify the role of pre-transplant immunological risk and whether this is linked to dnDSA development. We have also demonstrated the utility of the MDA in transplantation research, offering this tool to aid future DSA development studies.

Rachael Cole^1,2, Shahram Hemmatpour¹, Mariam Amer³, Kim Orchard³, Deborah Sage¹

¹NHS Blood and Transplant (H&I), Tooting, UK; ²University of Manchester, Manchester, UK; ³University Hospital Southampton, Southampton, UK

HLA-DPB1 may play an integral role in clinical outcomes following haematopoietic stem cell transplant (HSCT), although the significance of this remains uncertain.

We retrospectively analysed the impact of HLA-DPB1 on the outcome of 200 patients that underwent primary allogeneic HSCT within our single centre, all of whom were otherwise 10/10 matched at HLA-A, -B, -C, -DRB1 and -DQB1. The number of HLA-DPB1 mismatches were assessed for each recipient-donor pair and subsequently any that were mismatched classified as either permissive or non-permissive according to the revised T-cell epitope matching algorithm (TCE-FD). For any non-permissive mismatched pairs, the vector of mismatch was also considered.

The incidence of grade I-II acute graft-versus-host disease (aGvHD) was significantly influenced by HLA-DPB1 mismatching, although this did not reach significance in grade III-IV aGvHD. Those pairs with non-permissive mismatches in the GvH direction correlated with the highest risk of grades I-II aGvHD compared to those that were HLA-DPB1 matched (OR 8.93, 95% CI 2.52–31.62, p = .001). This significance was also confirmed in multivariate analysis (OR 9.52, 95% CI 2.62–34.63, p = .001). Although no impact was observed in chronic GvHD, relapse or transplant-related mortality, those that were HLA-DPB1 matched showed a trend towards better overall survival.

Our data helps to understand the role of HLA-DPB1 matching within our single centre and supports the importance of incorporating this locus as a determinant in unrelated HSCT. These findings will help to guide our local donor selection strategy in order to optimise transplant outcomes.

Sophie Chambers¹, Sean Druce¹, John Goodwin¹, Tim Key¹, Matthew Hopkins², Jayne Johnson², Kirti Mepani³, Marina Karakantza^4,5, Anthony Poles², Colin Brown³, Angela Kanny⁵, Lishel Horn^5,6, Robert Whittle¹

¹H&I Barnsley, NHS Blood and Transplant, Barnsley, UK; ²H&I Filton, NHS Blood and Transplant, Bristol, UK; ³H&I Colindale, NHS Blood and Transplant, Colindale, UK; ⁴Therapeutic Apheresis Unit, NHS Blood and Transplant, Leeds, UK; ⁵Haematology Department, Leeds Teaching Hospitals, Leeds, UK; ⁶Leeds Haemophilia Centre, Leeds Teaching Hospitals, Leeds, UK

Glanzmann's thrombasthenia (GT) is an inherited bleeding disorder in which ITGA2B /IGTB3 gene mutations produce quantitative and qualitative defects in platelet glycoprotein GPIIb/IIIa. Fifty-one-year-old female patient with GT presented with intracranial haemorrhage (ICH), staphylococcal endocarditis, partial mitral valve destruction and subsequent cardiac failure. This necessitated valve replacement surgery which is life-threatening in GT with high-risk of haemorrhage. Patient was HLA Class I antibody positive (cRF > 95%), received multiple HLA-selected platelet (HLA-SP) units daily with challenging platelet donor selection. Initial favourable platelet increment response deteriorated and post-transfusion < 5% circulating platelets were donor origin determined by functional GPIIb/IIIa expression. Anti-GPIIb/IIIa antibodies were detected rendering further platelet transfusions ineffective and patient managed with recombinant coagulation factor VIIa. With functional platelet count of <1 × 10^9/L, surgery was not feasible. Plasma exchange (PE) was used aiming to reduce antibody levels enabling effective platelet transfusion. Six rounds of PE were performed by NHSBT Therapeutic Apheresis Services with samples taken pre- and post for HLA and GPIIb/IIIa antibody detection. Platelet donors matched for HLA-A and –B or compatible single HLA-antigen mismatch were coordinated to provide 12 HLA-SP. Post 2nd round PE anti-GPIIb/IIIa antibodies were undetectable, remaining undetectable before surgery. HLA-specific antibodies also diminished. Four HLA-SP transfused prior to surgery raised functional platelet count to >100 × 10^9/L, permitting successful surgery using further 6× HLA-SP and minimal bleeding. Patient stable, extubated, conscious with ICH stabilisation d+1. Patient discharged clinically well d+29. This case exemplifies NHS inter-speciality collaboration facilitating GPIIb/IIIa and HLA-specific antibody removal enabling surgery with appropriate HLA-SP support and favourable outcome.

Katie Whittle¹

¹Welsh Transplantation and Immunogenetics Laboratory, Pontyclun, UK

In solid organ transplantation, the clinical significance of human leukocyte antigen (HLA) IgM antibodies remains uncertain though may indicate an elevated risk of rejection due to prior sensitisation.

The main aims of this project were to validate Luminex® technology for anti-HLA IgM detection and to screen a patient cohort for anti-HLA IgM. Antibody screening and characterisation was performed using LABScreen™ mixed and single antigen bead kits (One Lambda), respectively. Twenty-two CDC-defined IgM positive sera were tested to determine cut-offs and assess Luminex® concordance with CDC. Next, 96 sera were screened from patients including those with a potential sensitisation history and no known sensitisation history.

Overall concordance of HLA class I mixed and single bead kits with CDC were both 82% (p = .008), while class II mixed bead was 64% (p = .204) and single bead 77% (p = .012). Several sera were class I or II Luminex® positive but CDC negative, unsurprising given the enhanced sensitivity of Luminex® compared to CDC. However, one CDC and single antigen bead positive class I antibody was undetected by mixed bead indicating poor sensitivity. The poor concordance of class II mixed bead was mainly due to the resolution capability of CDC. Of the patient cohort, 36% of ‘sensitised’ and 16% of ‘non-sensitised’ patients had Luminex®-defined IgM anti-HLA antibodies, the majority of which co-existed as IgG. The data gathered suggested an IgM testing strategy based on single antigen bead alone would be most accurate, though the cost implications would need to be weighed against clinical benefit.

Deeya Balgobin¹, Emma Lougee¹, Corinna Freeman¹, Olivia Shaw¹

¹Viapath Guy's Hospital, London, UK

Differential rejection, of the pancreas and not the kidney, following simultaneous pancreas and kidney (SPK) transplantation, may be explained by the greater expression of MICA antigens on pancreas epithelium, creating more target for MICA donor specific antibody (DSA). There is some data to support this, however, previous studies lacked MICA typing to determine presence of MICA DSA (Rangel et al., European Society for Organ Transplantation 2009, 23, 602–610). This study evaluated the potential significance of MICA DSA in this group of patients.

Twenty SPK donor and recipient pairs, where differential rejection of the pancreas was suspected, in the absence of HLA antibody, were MICA genotyped using LABType SSO (One Lambda). Pre-transplant and time of rejection sera were screened using LABScreen MICA single antigen beads (One Lambda). Five rejection-free SPK transplant recipients were included as controls.

The presence of MICA DSA was compared.

There was no significant difference in the presence of MICA DSA in recipients with differential rejection (5%), compared to the controls (0%) (p = .99).

MICA DSA was not shown to be associated with differential rejection of pancreas following SPK.

This study had several limitations beyond its size. Notably, the MICA Ab screening kit employed did not detect Ab to MICA 008, the most common antigen, meaning DSA may have been missed.

A larger scale study, using NGS typing and an alternative screening panel, including MICA 008, would be warranted before the significance of MICA DSA in the differential rejection of pancreas following SPK transplantation can be disregarded.

Richard Issitt¹, Matthew Fenton¹, Eamonn Cudworth², Arun Gupta, Delordson M Kallon

¹Great Ormond Street Hospital, London, UK; ²Clinical Transplantation Laboratory, Barts Health NHS Trust, London, UK

Human leukocyte antigen (HLA) antibody sensitisation represents a major barrier to cardiac transplantation. Patients supported by Ventricular Assist Devices (VAD), or those who have undergone cardiac surgery are at higher risk of developing HLA-antibodies. Currently there is no consensus as to how, or when desensitisation should take place, meaning that sensitised patient waits much longer for a compatible donor. We aimed to determine if it were possible to use immunoadsorption in an ex-vivo setting to provide a potential, intraoperative desensitisation methodology.

Three HLA antibody-containing whole blood units were obtained from NHSBT and quantified using a Luminex Single Antigen Bead assay. A Cardiopulmonary bypass (CPB) circuit was set up to mimic a 20 kg patient undergoing cardiac transplantation, into which a plasma separator was placed. Plasma was diverted to a standalone, secondary immunoadsorption system, with antibody-depleted plasma return to the CPB circuit. A total treatment volume of 3-fold the pseudo-patient's plasma volume (4500 ml at 75 ml/kg) was treated. Samples for HLA antibody quantification were taking at baseline, and every 20 min for the duration of treatment (total 3 h).

All three experiments demonstrated a reduction in individual bead mean fluorescence intensity (MFI) to below clinically relevant levels (<1000 MFI), even in HLA specificities with a baseline MFI > 4000. Flowcytometric crossmatching of suitable pseudo-patients demonstrated a flipping from T- and B-cell positive to negative.

Intraoperative immunoadsorption in an ex-vivo¬ setting demonstrates a clinically relevant reduction in HLA-antibodies. This method represents a potential desensitisation technique that could enable sensitised children to accept a donor organ earlier.

Alexandra Kreins¹, Sejal Morjaria², Evey Howley¹, Graham Davies¹, Delordson Kallon

¹Great Ormond Street Hospital, London, UK; ²Clinical Transplantation Laboratory, Barts Health NHS Trust, London, UK

Congenital athymia is most commonly associated with DiGeorge syndrome due to microdeletions of chromosome 22q11.2, but also with other genetic/non-genetic causes. T-cell progenitors complete their development in the thymus and therefore athymic patients have a severe primary immunodeficiency due to the absence of functional T-cells. Allogeneic thymus transplantation is the most appropriate treatment and worldwide, only two centres offer this pioneering treatment including Great Ormond Street Hospital (GOSH). Donor thymus tissue is obtained from infants undergoing cardiac surgery when it is normally discarded. It is then cultured for 2–3 weeks to deplete donor thymocytes prior to implantation into the athymic patient. Donor selection based on tissue type is not feasible, but chance matching at some loci with the recipient may occur and is monitored. So far, 60 patients have been transplanted at GOSH of which 35 have been followed up for >18 months. Preliminary data suggest a trend towards high resolution (HR) HLA chance matching at two or more Class II loci having a beneficial effect on outcomes including fewer new autoimmune complications after transplantation (incidence of autoimmunity 22% vs. 62% in those with and without two or more high resolution DR/DQ matches, respectively). Thymic output as measured by T-cell receptor excision circles also tends to be higher when patient and donor have some HLA-CII matching. Continued monitoring of HLA matching and long-term follow-up of the increasing numbers of treated patients will be required to confirm a possible role for partial HLA matching in improving outcomes after thymus transplantation.

Jessica Brookes¹, Tom Browne¹, Jane Horler¹, Elizabeth Wroe¹, Anthony Poles¹

¹NHS Blood & Transplant, Bristol, UK

HNA-1 (FCGR3B) genotyping at NHS Blood & Transplant, Filton, is performed by PCR-SBT. Using this method, we are able to detect polymorphisms at known mutation sites as well as other regions along the gene sequence; however, FCGR3A and FCGR3B genes are co-amplified, making it difficult to assign definitive types to either gene. A potentially novel polymorphism (NM_000570.4: c.197T > G p.Leu66Arg) was identified in a donor sample. Retrospective analysis of 229 clinical samples between January and May 2022 identified 50 samples (22%) with up to three polymorphisms in this position (c.197T > AGT p.Leu66His/Arg/Leu). In FCGR3A, polymorphisms at this triallelic position leads to functional change in the presence of p.176Val (Nagelkerke et al., Frontiers in Immunology 2019, 10, 7); however, the clinical significance of c.197 polymorphisms in FCGR3B is yet to be elucidated for granulocytes. Long-range amplification with FCGR3B-specific primers was performed on the 50 samples identified above. In ninety-four percent of samples, no polymorphism was observed in FCGR3B, suggesting that the polymorphisms observed by PCR-SBT (e.g. c.197T > A) were in FCGR3A. In FCGR3B only, we identified two samples with c.197T > TG (p.Leu66Leu/Arg) polymorphisms, and one sample with a c.197T > G (p.Leu66Arg) polymorphism. Further work exploring molecular interactions with FCγRIIIB in the presence of these amino acid substitutions is indicated to ascertain the clinical significance of the c.197 polymorphisms identified in FCGR3B. Furthermore, one sample was confirmed as FCGR3B*Null, demonstrating the clinical utility of this additional FCGR3B-specific method to remove ambiguity, analyse the gene of interest, and confirm Null genotypes without phenotyping, which has its constraints.

Matthew Hopkins¹, Anthony Calvert¹, Lorraine Flores¹, Mary Assall¹, Kate Giddings¹, Anthony Poles¹

¹NHSBT, Filton, UK

In March 2021, a new clinical condition was identified which leads to thrombocytopenia and thrombosis in a proportion of people vaccinated with the ChAdOx1 CoV-19 vaccine (AstraZeneca). Presentation is similar to heparin-induced thrombocytopenia (HIT) but without previous exposure to heparin. By April 2021, the association with the AZ vaccine was confirmed and designated vaccine induced/associated thrombotic thrombocytopenia (VITT/VATT). The antigen binding site was identified as platelet factor 4 (PF4) (Huynh et al., Nature. 2021;596(7873):565-569).

H&I, NHSBT Filton became a reference laboratory for VITT/VATT referrals. Our testing protocol includes an ELISA (Immucor HAT45G) and a functional platelet assay, HITAlert (IQ Products, Groningen, Netherlands), which identifies activated platelets by flow cytometry using platelet specific and activation markers. The patient sample can be prepared without heparin, important as patients diagnosed with VITT/VATT display high-titre antibodies to PF4 that activate platelets in the absence of heparin. Heparin is shown to inhibit this reaction, contrary to activation by PF4/heparin complexes in HIT. Table one shows the results of suspected VITT/VATT cases referred to H&I Filton between March 2021 and May 2022.

We have implemented HITAlert for the detection of PF4/heparin dependent antibodies seen in HIT and PF4 antibodies in VITT/VATT. The kit offers a rapid, easy to use assay, which is safer than the radio-labelled serotonin release assay (SRA) which was considered the gold-standard for HIT. This has enabled us to have a definitive functional test available, increasing our scope and improving the service provision (Table 1).

Betia Nouri¹, Loretta Brown¹, Fiona Powell¹, Mazen Mabrok¹, Ambika Camille¹, Julian Cano-Flanagan¹, Rachel Smith¹, Arthi Anand¹

¹Imperial College Healthcare NHS Trust, London, UK

Allogeneic hematopoietic stem cell transplantation (HSCT) is an effective therapy for patients with Acute Myeloid Leukaemia (AML). Approximately 30% of such patients experience relapse. While the vast majority of these post-HSCT AMLs are of host origin, donor cell leukaemia (DCL) in which malignancy develops in the donor's transplanted cells has been reported to occur rarely, with DCL accounting for 2–5% of relapsed cases.

Here we describe a patient with a past medical history of Ulcerative Colitis for which he received Mercaptopurine and was subsequently diagnosed with therapy related AML with monsomy 7 detected by cytogenetics. Post paternal haploidentical HSCT, he achieved MRD negativity by flow, cytogenetics and aspirate as well as a high chimerism of >97% indicating a complete response. Within months he showed continued morphologic remission and high chimerism however there was evidence of cytogenetic relapse.

We demonstrate, a multi-disciplinary team approach between H&I and SIHMDS in reaching a diagnosis of de novo development of DCL MDS in the engrafted donor cells. This was established using STR Chimerism analysis, NGS Chimerism monitoring, FISH, karyotyping and Flow cytometry; alongside HLA typing by NGS to identify whether there was relapse with or without HLA loss. In light of this diagnosis a second HSCT was considered to be the best treatment option.

Furthermore, we discuss our donor selection strategy for the second HSCT, in which we selected donors based on the patient's initial HLA type and not that of his post-transplant donor type in order to exploit the graft-versus-leukaemia effect.

Dario Merlo¹, Zdenka Edwards¹, Sandra Frater¹, Elizabeth De Mendonca¹, Lisa Walsh¹, Sharon Vivers¹

¹Anthony Nolan Research Institute, London, UK

Somatic mutation in HLA genes is a potential mechanism through which cancer cells can evade immune surveillance. Loss of heterozygosity for all or part of one HLA haplotype and Single Nucleotide Polymorphisms (SNPs) have been observed in haematologic malignancies. These anomalies can be increasingly identified due to the introduction of Next Generation Sequencing (NGS) techniques. Here we describe the case of a 27-year-old patient with T-ALL, referred for allogeneic stem cell transplant. Peripheral blood samples were sent to our laboratory during blast crisis (91% lymphoblast in bone marrow). HLA typing was performed at HLA-A, -B, -C, -DRB1, -DRB3/4/5, -DQA1, -DQB1, -DPA1 and -DPB1 loci using GenDx NGSgo®-AmpX v2 kits utilising Illumina Sequencers. Analysis of the base variation plot for the HLA-DRB1 locus showed a sub-population of reads (17%) for the HLA-DRB1*13 allele, displaying a C > T SNP in position gDNA 5421 (cDNA 149), distinguishable from the background and potentially responsible for an amino acid change in codon 21 (Thr > Met). A buccal swab sample was requested to determine if this was a germline mutation or cancer derived. DNA extracted from the buccal swab underwent NGS typing, and the analysis did not detect any mutation in the HLA-DRB1 locus, confirming the SNP observed in a subset of reads as cancer-derived. This study demonstrates that NGS can detect novel single-nucleotide mutations associated with malignant diseases, corroborating the use of germline DNA from buccal swabs for confirmation. Failure to detect these mutations as cancer-derived may impact the HLA type assigned to the patient.

Sandra Lloyd¹, Deborah Pritchard¹, Tracey Rees¹

¹Welsh Blood Service, Talbot Green, UK

AQ, a 55-year-old male (cRF-99%) returned to the deceased donor transplant list in 2014 following the failure of his first kidney transplant. Although AQ and his first donor were both DRB1*15-DQB1*06, DQB1*06:02, 06:03 and 06:09 allele-specific antibodies were detected. Sequence-based typing determined that AQ was DRB1*15-DQB1*06:01 rather than the more common DRB1*15-DQB1*06:02. HLA-DQ6 was not listed as unacceptable and AQ remained on the deceased donor list with no offers until early 2020 when two offers were received in quick succession. Both donor 1 (DR13-DQ6) and donor 2 (DR15-DQ6) were declined due to positive T- and B-cell flow cytometry crossmatches. Due to the complexity of his antibody profile listing of DQ6 would severely restrict AQ's already limited pool of donors. Listing of DRB1*15 could however be used as a surrogate for DQB1*06:02 due to strong linkage disequilibrium (LD). This would reduce but not eliminate the chance of a positive crossmatch as AQ could still receive incompatible offers of DRB1*13:01-DQB1*06:03 or DRB1*13:02-DQB1*06:09 donors. It would however leave the possibility of compatible DRB1*13:02-DQB1*06:04 donors. The proposal was discussed at the local multidisciplinary team meeting, and it was agreed to list DRB1*15 as unacceptable. Following a pause in transplantation due to the COVID-19 pandemic, AQ received a further two incompatible donor offers, both DRB1*13-DQ6, before finally being transplanted with a crossmatch negative donor in July 2021. Whilst unacceptable listing at the allelic level is not currently possible, strategic listing of surrogate antigens can be used to reduce the risk of positive donor crossmatch offers.

Sarah Peacock¹, David Turner²

¹Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK; ²SNBTS, Edinburgh, Scotland, UK

Despite accumulating evidence that preformed IgG HLA donor specific antibodies (HLA-DSA) increase the risk of deleterious outcomes in liver transplant recipients, the current BSHI/BTS HLA antibody guidelines do not consider HLA-DSA. This abstract provides an update to the liver guidelines for H&I illustrated through a patient case study. The patient was listed for his 3rd liver transplant in April 2021 and, due to the high level of sensitisation to predominantly HLA class II, it was agreed to avoid previous HLA class II donor mismatches at MFI > 20,000 by providing virtual crossmatching (vXM) on call. The patient waited 7 months with 11 potential donor offers before receiving an offer deemed appropriate immunological risk (HLA mismatch grade 1.1.1, no repeat transplant mismatches and negative for class II HLA-DSA). He was discharged 1-month post-transplant with good liver function and no DSA detected at last follow up 3 months later. The UK liver guideline update proposes to include a recommendation that the presence of IgG HLA specific antibodies that are circulating at levels likely to cause a positive crossmatch and are directed against known previous donor HLA mismatches associated with adverse events (e.g. rejection), warrant consideration for prospective avoidance in the liver re-transplantation setting. This decision should be taken with the clinical team and the risk of avoidance balanced against the risk of not transplanting. This will require the H&I laboratory to provide an on-call service to allow for an individualised immunological risk assessment for a given donor and recipient pair.

Sarinder Day¹, Ella Brewer¹, Jack Galliford¹

¹North Bristol Nhs Trust, Bristol, UK

We report here our second HNA-3a antibody patient case identified for renal transplantation. A 49-year-old female patient with polycystic kidney disease was assessed for kidney transplantation with a potential living kidney donor. The flow-cytometry crossmatch was unexpectedly positive for T and B cells with repeat and third party crossmatching, but negative by CDC. No HLA or autoantibodies were detectable. HNA-3a antibodies were identified in the patient sera using Luminex based screening and recombinant cell lines. The patient genotyped as HNA-3b3b and donor HNA-3a3b by PCR-SBT.

HNA-3a antibodies have been reported in several cases in the UK, for both deceased and living donor kidney transplants, most of these are reported to be associated with early and chronic antibody rejection with some resulting in graft loss. Options for such patients are limited as approximately 95% of the population express HNA-3a and donor genotyping is not routinely performed to identify HNA-3b compatible donors. Desensitisation protocols have proved to enable HLA and ABO incompatible direct living donor transplants for patients with limited alternative options. Here we report our first planned HNA-3a antibody incompatible case for desensitisation using typical HLA/ABO antibody removal protocols including Rituximab and plasma exchange. As there are no established quantitative or semi-quantitative methods currently available to monitor HNA-3a antibody removal, T cell flow-cytometry is used where serum titrations have shown to reduce both the T and B cell RMF.

Katherine Mounsey¹, Pallavi Yadav², Sophie Holland², Mark Lobb¹, Victoria Wood1, Brendan Clark¹

¹Transplant Immunology Laboratory, Leeds NHS Trust, Leeds, UK; ²Department of Children's Nephrology, Leeds NHS Trust, Leeds, UK

Samples from a 17-month old patient, with renal dysplasia, were received for virtual crossmatching versus his father. HLA antibody screening revealed the presence of HLA-A2 and A28 antibodies, despite an absence of sensitising events (transplantation, blood products). The HLA-A2 antibody resulted in an incompatible virtual crossmatch versus the patient's father who was a 111 HLA mismatch, but HLA-A2 positive. Further investigations into the patient's clinical history revealed he had received a live vaccine against Varicella Zoster (Varilrix), prior to his samples being sent to the H&I laboratory. Varilrix is propagated in a human cell line which is HLA-A2 positive.

To investigate if the HLA-A2 antibody was functional, a flow cytometric crossmatch was performed for the pair, which was T and B cell positive. Following this the patient was entered into the national kidney sharing scheme and activated on the deceased donor list with a calculated reaction frequency of 56%.

This case prompted us to investigate a further nine paediatric patients who had received a Varicella vaccine. The analysis was complicated by several patients having received blood products and by the cell line HLA type coinciding with ‘self’ HLA. Two patients showed some HLA-A2 positivity but had a transfusion history. Two further patients, with no sensitising events, showed low level sensitisation against antigens present on the vaccine cell line.

This case describes investigations in a paediatric patient who has potentially developed HLA antibodies following a live vaccination, and highlights the need for vigilance in using such vaccines in patients awaiting transplantation.

Maria Irvine¹, Alison Logan¹, Anna Barker¹, Karen Wood¹, Michelle Carr¹, Natalia Diaz Burlinson¹

¹Manchester Transplantation Laboratory, Manchester, UK

A 47-year-old male with high-risk AML was referred for HPCT. Simultaneous searches of the unrelated donor and cord registries were initiated in addition to typing two family members.

The patient and potential donors were HLA typed by NGS (AlloSeq™Tx17, CareDx). The related donors both shared an HLA haplotype with the patient. HLA-specific antibody testing by LABScreen® single antigen beads identified high-level antibodies against multiple HLA class II specificities, including mismatched antigens in the related donors.

The unrelated donor search did not yield any viable options. Similarly, for the cord blood search only mismatched units were available and the patient was sensitised to the mismatched antigens in all cord units with an adequate cell dose.

Clinical urgency and lack of alternative options expedited selection of a related donor with antibody removal pre-HPCT to reduce the chance of graft rejection. CDC and flow cytometry risk assessment crossmatches were performed against the two related donors. Both crossmatches were historic CDC positive and flow cytometry positive (current and historic). The relative to whom the patient had the lowest level of donor-directed antibodies was selected as the donor of choice; the antibodies were HLA-DR15 (MFI: 18,976), DR51 (MFI: 10,097) and DQ6 (MFI: 16,617).

The patient underwent three cycles of plasma exchange prior to a RIC haploidentical HPCT; donor-directed antibody levels were monitored pre- and post-plasma exchange. The patient engrafted successfully, and post-transplant chimerism monitoring of PBL and CD15+ cell lineages showed 100% donor engraftment. The patient is currently well post-transplant with no indication of rejection.

Anna Barker¹, Alison Logan¹, Karen Wood¹, Stephine Whiteside¹, Madeleine Harris¹, Judith Worthington¹, Natalia Diaz Burlinson¹

¹Manchester University NHS Foundation Trust, Manchester, UK

A 62-year-old female with myelodysplasia was referred for an urgent unrelated donor stem cell transplant.

Three donors were high resolution matched at HLA-A, B, C, DRB1 and DQB1 (10/10). The patient was HLA-DPB1*01:01, 04:01. Two donors were HLA-DPB1 mismatched: -DPB1*02:01 and -DPB1*04:02, respectively. The third donor was HLA-DPB1*04:01 homozygous but was found to be discrepant (HLA-A*01, 02) with the registry type (HLA-A*02 homozygous).

Subsequently, HLA antibody testing was performed. HLA class I and II specific antibodies were defined (LABScreen™ Single Antigen, One Lambda), including HLA-DPB1*02:01 (MFI 27,381), -DPB1*04:02 (MFI 26,672) and HLA-A1 (MFI 25,545) antibodies.

Since the need for transplant was urgent, crossmatch blood samples were requested via the registries from the HLA-DPB1*02:01 and -DPB1*04:02 mismatched donors. The flow cytometry crossmatches were B-cell positive. Two further 10/10 donors were ordered; both were HLA-DPB1*04:01 homozygous.

Donor options and test results were discussed at MDT; ideally waiting for HLA typing of the expected donors, or, if immediate transplant was required, plasma exchange prior to transplant with one of the HLA-DPB1 mismatched donors. The aim was for transplant in three weeks, allowing for urgent HLA typing of the expected donor samples. Unfortunately, the patient relapsed; she is undergoing further chemotherapy, with the aim of transplant at the end of June 2022.

The importance of receiving samples for HLA antibody testing early enough in transplant work-up for the selection of the most suitable donors is highlighted, enabling short time to transplant for urgent patients, whilst saving time and money for laboratory and clinical teams.