Oral Abstracts

Alison Cleaton, Emma Burrows, Kimberley Robinson, Michael Richardson, Deborah Pritchard, Tracey Rees

Welsh Blood Service, Ely Valley Road, Talbot Green, UK

Regular HLA antibody testing is undertaken for patients awaiting renal transplantation, using LABScreen™ HLA antibody assays. During the COVID-19 pandemic, we observed unexplained changes to some HLA antibody profiles. Investigation revealed that several patients had COVID-19 prior to the changes, therefore, a review of all patients on the transplant waiting list with known COVID-19 infection was undertaken. Sixty-six out of two hundred thirty-seven patients on the transplant waiting list had COVID-19 (March 2020–July 2022). The HLA antibody results from samples prior to and following COVID-19 infection were analysed for changes in existing HLA antibody levels (increased Luminex Median Fluorescent Intensity (MFI) values), or expanded antibody profiles (increased cRF). Fifty-two (78.8%) patients had no detectable change in cRF or MFI; five (7.6%) had changes in MFI (but no change in cRF); nine (13.6%) had changes in MFI and cRF. Two out of nine patients had no recorded prior sensitisation event; four had a previous transplant, four blood transfusions, four pregnancy; and three multiple sensitising events. All nine patients had sustained cRF changes in subsequent samples (follow up to December 2022). Three out of nine patients were consequently identified as having altered immunosuppression due to the COVID-19 infection; these patients had a 20%–76% rise in cRF and now all have a cRF 98%–100%. While the majority of patients awaiting kidney, transplantation had no change to their HLA antibody profile following COVID-19 infection, nine patients had an increase in cRF, which has not been transient. Reduction or withdrawal of immunosuppression to aid recovery from COVID-19 was identified as the cause for three patients.

Adrienne Seitz, Clive Carter, Brendan Clark, Richard Baker

Leeds Teaching Hospitals NHS Trust, Leeds, UK

The level of pre-transplant immune risk is assessed through measuring serum IgG HLA antibodies which can be produced by long lived plasma cells and memory B-cells. Memory B-cells can circulate without producing antibodies, therefore their contribution to the antibody pool may not be fully appreciated. We describe an in vitro method for improving the assessment of pretransplant risk through the non-specific stimulation of peripheral memory B-cells. Peripheral blood mononuclear cells from three unsensitised volunteers and six sensitised patients were cultured for 9 days with the toll-like receptor agonist R848 and interleukin-2. Cell culture supernatant was tested for IgG HLA antibodies using single antigen beads. This was compared with a matched serum sample. Resting Day-0 and stimulated Day-9 B-cell phenotypes were assessed using flow cytometry, confirming the switch to antibody secreting (CD24-CD38hi), class-switched memory (CD27+IgD-) and plasma (CD38+CD138+) cells. HLA Class I and II antibodies were found in the cell supernatant, and 65% were present in the matched serum sample. When the supernatant demonstrated additional HLA antibodies, these could be attributed to a previous transplant, or had been present in the patient's historic serum profile. We demonstrate a method that can uncover peripheral memory using technology accessible to most H&I laboratories. This assay could be useful when assessing live donor pairs where the donor may repeat mismatches associated with pregnancy, and in re-grafts, prior to removal of ‘other unacceptable antigens’. Finally, it could be considered alongside delisting strategies in the context of novel peri-transplant agents.

Sophie Chambers, Robert Whittle, John Goodwin, Tim Key

NHS Blood and Transplant, Barnsley, UK

Differences in amino acids (aa) at positions 76–83 of exon II of HLA-B and a subset of HLA-A primarily account for the highly immunogenic public epitopes Bw4 and Bw6. Bw4-specific antibodies are commonplace in alloimmunised Bw6 homozygotes, whilst Bw6-specific antibodies are encountered in Bw4 homozygotes. As variation exists within the Bw4 complex it has been reported previously that in a proportion of Bw6 homozygous individuals with Bw4 specific antibodies, reactivity can be restricted to HLA Bw4 subtype epitopes. We report three individuals with a Bw4 phenotype who demonstrate Luminex Single Antigen Bead determined alloantibodies to Bw4 epitopes distinct from their own. For renal patient L bearing B13, reactivity was present to all Bw4-positive beads other than B13. Reactivity was consistent with Epitope 249 defined by aa at positions 82L + 145R/ 83R + 145R. Platelet refractory patient G bearing A*24:02 demonstrated reactivity to all Bw4 beads other than A24, consistent with Epitope 423 defined by 144Q. For allogeneic-HSCT patient F bearing B*27:05, reactivity was present to all Bw4 beads except B*27:05 with reactivity defined by aa 77N + 81A + 82L. Our observations highlight that Bw4 subtype epitope profiles should not be overlooked in Bw4-positive individuals, that single amino acid differences in the Bw4 complex appear sufficient to generate alloantibodies and differences outside aa positions 76–83 may contribute to antibody binding. Bw4 subtype antibodies can impact organ allocation in the renal transplant setting, restrict appropriate platelet support for immunological refractoriness and complicate donor selection in HLA-mismatched allogeneic-HSCT.

Ryan Stevens, Felicity May

Welsh Transplantation and Immunogenetics Laboratory, Pontyclun, UK

The Welsh Transplantation and Immunogenetics Laboratory maintain a register of local patients active on the national renal/pancreas waiting list. Previously, data pulled from local IT systems was distributed monthly to service users in PDF format and a printed copy was held locally. This system had multiple drawbacks, including lack of ability to easily interrogate/update the data. In collaboration with the NHS Wales Microsoft 365 Centre of Excellence, a digital solution was created using Power BI. This allows up-to-date data to be instantly available, which gives users a more accurate overview to support decision making, and eliminates the requirement for PDF/paper distribution. The Dashboard can only be accessed by approved users, and access can be restricted to specific datasets (e.g., dialysis unit staff can only view patients at that unit). The Dashboard consists of two pages. A ‘Statistics’ page graphically displays the Register by organ type, blood group, dialysis unit, and so forth. The ‘demographics’ page tabulates data within the Register, which can be ordered or filtered in a variety of ways to aid in patient selection for donor offers. Each patient also has a colour coded ‘Sample Status’ to highlight when routine antibody screening samples are due. Comments and/or attachments can be added to each patient entry. Since successful go live December 2022, the dashboard has been praised by staff and service users for ease of access of data to assist delivery of safe and efficient patient care, as well as reduced incidence of overdue samples.

Kirsty Clark, Jane Matthews, Claire Romaines, Ruth Chisman, Arash Akbarzad-Yousefi

NHSBT Newcastle, Newcastle upon Tyne, UK

Despite advancements in organ allocation and immunosuppression, cardiothoracic transplantation continues to have the lowest 5-year survival when compared to all other forms of solid organ transplantation. A key factor in allograft loss is the formation of de novo Donor Specific HLA Antibodies (dnDSA). Recently, research has been driven away from the conventional method of defining mismatch at the antigenic level and has instead focused on HLA eplet mismatches. This study aimed to evaluate the role of eplet mismatching within our local cardiothoracic patient cohort. A retrospective analysis of 2 years of cardiothoracic transplant data was performed. The antibody data from all eligible patients were reanalysed to investigate the role of eplet mismatch load and to identify possible high risk eplets. This study found no evidence to directly support the theory of eplet load mismatches. However, four previously identified high risk eplets (McCaughan et al., American Journal of Transplantation 2018) namely, 55PP, 52LR, 55R and 75S were confirmed within our patient cohort. Findings from this single centre study provide potential evidence of high-risk HLA eplet mismatches; all of which are present within the HLA-DQ locus, indicating possible high immunogenicity for mismatches at this locus. Particular attention to minimise HLA-DQ mismatches may reduce the incidence of dnDSA and subsequently, allograft loss. Whilst avoiding mismatches in cardiothoracic transplantation is not always possible, further understanding of this area could lead to refinements in post-transplant monitoring and immunosuppression regimens.

Charlotte A. Cambridge¹, Jonathan A.M. Lucas¹, Xenia Georgiou¹, Gabriel J. Benitez¹, Neema P. Mayor^1,2, Steven G.E. Marsh^1,2

¹Anthony Nolan Research Institute, London, UK; ²UCL Cancer Institute, Royal Free Hospital, UK

Submission of novel sequences to the IPD-IMGT/HLA Database from patients with haematological malignancies is not permitted, unless confirmed in the germline. To investigate if sequencing blood and buccal DNA yields different results, we typed 47 patients in remission from malignant disease and 58 healthy donors for HLA-A, -B, -C and -E using PacBio SMRT sequencing. Higher Phred scores (33.2 vs. 32.7, p < 0.05) and lower cluster diversity (0.06 vs. 0.11, p < 0.0001) were observed in sequences from patient buccal DNA versus blood, indicating sequences of higher quality with fewer background errors. No differences were observed in donor blood or buccal material. Blood DNA generated better HLA typing results that were automatically accepted (80.7% vs. 79.8%) with less allele dropout (2.0% vs. 7.2%). HLA typing results were concordant between blood and buccal derived DNA, including four novel sequences observed in patient samples, confirming these as germline mutations. Next, we analysed sequences from blood DNA for patients in remission from malignant (n = 406) versus non-malignant (n = 46) disease across nine libraries containing ≥3 of HLA-A, -B, -C, -E, -F and -G. No significant differences in Phred score (33.60 vs. 33.56) or cluster diversity (0.05 vs. 0.06) were observed. Overall, there were no differences in final HLA typing results for blood and buccal DNA samples for the same individual, and no difference in sequence quality between malignant and non-malignant patient samples. All sample types, if taken at the point of remission, are reliable sources for HLA typing and identification of novel sequence variation.

Ravneet Kaur Bola, Madalina Pinzaru, Marlowe Macadangdang, Ufot Udoffia, Sandra Frater, Franco Tavarozzi

Anthony Nolan Histocompatibility Laboratories, London, UK

Buccal epithelial cells are used for germline HLA testing, particularly useful in patients where loss of heterozygosity (LOH) is suspected. Anthony Nolan's (AN) patient buccal collection method utilises CytoSoft® brushes, however, we are always looking for ways to optimise our processes and realise that the design of the brushes may lead to an increased chance of blood contamination with some patients, potentially confounding HLA typing in LOH cases.

AN uses a gentler collection tool for our registry donor recruitment, FLOQSwabs®, therefore a trial was devised to identify the best buccal collection method for our patients, collaborating with the Royal Marsden Hospital. Each consented patient was swabbed using three scenarios: (1) FLOQSwabs®, cheek; (2) FLOQSwabs®, left gutter; (3) CytoSoft® brushes, right gutter. Samples from 54 patients were processed. DNA was extracted from each swab and one extraction per patient was selected for Next Generation Sequencing (NGS) in a pre-determined sequence, producing equal amounts of NGS data per scenario. Assessment looked at DNA extraction and NGS failure rates, together with DNA quantity produced.

DNA extraction failure rates were 9.3%, 17.6%, 29.9% for the three scenarios respectively.

On average, the DNA concentration from FLOQSwabs® were at least two times higher than with CytoSoft® brushes. Due to NGS robustness, all extractions selected for NGS were successfully typed. In summary, FLOQSwabs® swabbed from the cheek are the preferred collection method. These produced the lowest number of failed DNA extractions, while providing a softer and gentler swabbing experience for the patient.

Claire Lenehan, James Kelleher, David Keegan, Mary Keogan, Khairin Khalib

NHISSOT Beaumont Hospital, Dublin, Ireland

Antibodies to human leukocyte antigens (HLA) are a complication for transplantation. The introduction of Luminex® technology has allowed for precise characterization of these antibodies with high sensitivity. The clinical significance of these antibodies remains controversial due to the detection of biologically irrelevant antibodies directed against denatured HLA molecules (dHLA). We aimed to determine the correlation of donor specific antibodies (DSA) detected by Luminex single antigen bead assay with flow cross match (FXCM) results, using a previously published protocol, with clinically validated cut-offs.

The capability of HLA antibodies in 170 non-classically sensitised patients to result in a positive crossmatch was investigated. Forty-nine FCXM against incompatible donor cells were completed. Fifty-six percent of T cell and 61% of B cell FCXM were positive. The rate of positive FCXM results was significantly higher for HLA-A and -B antibodies (p = 0.013). MFI value was a poor predictor of FCXM results. Conversely, the rate of negative FCXM results for DSA with MFI values of 5000 to 10,000 was not significantly different to that of MFI values <5000. The rate of positive FCXM was higher for patients with DSA MFI > 10,000. A significant proportion of the FCXM results were positive regardless of MFI strength indicating that DSA in non-classically sensitised patients are not limited to reactivity to dHLA. FXCM facilitates risk assessment. Based on the data in this study, we now offer a FCXM in living donors, or highly sensitised patients when DSA MFI is less than 10,000.

Kay Poulton ,^1,2 Madeleine Harris¹, Andrew Canterbury², Marie Hampson¹, Judith Worthington¹, Marcus Russell-Lowe¹

¹Manchester Royal Infirmary, Oxford Road, UK, ²MC Diagnostics Limited, St Asaph, Wales, ³University of Manchester, Manchester, UK

Single Antigen Bead assays have revolutionised the identification and definition of HLA-specific antibodies. They have enabled the widespread use of virtual crossmatching and almost eliminated hyperacute rejection due to HLA incompatibility during the 20 years since their introduction. But for highly sensitised patients, there is a low likelihood of finding a compatible donor. The only option for some patients is strategic de-listing of specificities which may impact the lowest immunological risk to a new graft. In this study, 257 serum samples from 82 potential recipients were crossmatched against cells from 84 potential donors. Only results where serum samples have been tested by LABScreen, HISTO SPOT® HLA AB, Complement Dependent Cytotoxicity (CDC) and Flow Cytometry were included in the analysis. The results were analysed to assess the ability of HISTO SPOT® HLA AB to predict the crossmatch result. Of the 136 samples analysed, 17 (12.5%) were CDC positive, and 82 (60.3%) were positive by flow cytometry. Twenty-eight sera tested negative using HISTO SPOT. All of these (100%) were also negative by CDC and 25 (89.3%) were also negative by Flow Cytometry. Three sera which tested negative by HISTO SPOT with a positive flow cytometry crossmatch were CDC negative, each had cumulative MFIs of <3000 against Donor Specific Antigens. In this early study, HISTO SPOT® HLA AB has 100% negative predictive value for CDC crossmatch negativity and 89.3% by Flow Cytometry. It may therefore prove a useful additional tool to inform de-listing strategies used to facilitate transplantation in highly sensitised patients.

David Wimbury

Transplant Laboratory, University Hospitals of Leicester NHS Trust, Leicester, UK

Previous work has shown that variations in temperature, incubation time and operator methodology can have a drastic effect on results in solid phase assays. In HLA antibody screening, differences in overall MFI levels could impact the number of antibodies detected as positive and/or their risk level, thereby potentially affecting transplantation opportunities. Normalisation of MFI in each sample to a baseline could create more consistent results between samples at different time points, reducing the measurement uncertainty. This method normalises overall sample MFIs to a baseline. Positive control samples and three patients with extensive sample histories were tested with this methodology. CVs of MFIs before and after normalisation were analysed with paired t-tests.

Overall, MFI data became more comparable between samples post-normalisation. MFI ranges on a bead-by-bead basis initially varied between runs from around 500 to as much as 20,000. This was decreased after normalisation; the average MFI range between runs decreased by 1337 for Class I and by 2402 for Class II. All samples showed a statistically significant decrease in coefficient variation (CV) of MFIs between runs after normalisation, both Class I and Class II (p < 0.0001). Inconsistencies in HLA antibody screening results between samples from the same patient can make for troublesome clinical interpretations, especially when there is no sensitising event accounting for reactivity changes. The procedure shown here provides a novel method for reducing the variability caused by differences in assay conditions and has the potential to give more consistent results therefore providing a clearer clinical picture.

Deborah Pritchard

Welsh Transplantation Laboratory, Pontyclun, UK

Immunological compatibility testing for potential live kidney donors and patients requires multiple tests on patient and donor samples (HLA typing, HLA antibody testing, crossmatch assessment). Only on completion of all tests can a report be produced. The local KPI is 90% of cases reported in a turnaround time (TAT) of 10 working days, but performance since Q1 2021/22 has been <75%. This quality improvement project (May–August 2022) identified delays from process maps and selected two areas as targets for improvement; antibody testing and report production. Two plan-do-study-act (PDSA) cycles were performed testing six change ideas generated from staff workshops. Changes included identification of case samples within the testing process to allow more efficient workflow; a dedicated individual with responsibility for reporting rather than shared responsibility; E-mail notification when tests have been completed; coordination of HLA antibody batch testing with sample arrival to reduce wait time, and introduction of visualisation boards to track cases more efficiently. One ‘just do it’ change was also introduced; an interface to import antibody results from HLA Fusion software to the laboratory information management system (LIMS). The mean TAT was reduced from 11.9 to 8.8 days. Improvements were seen in the antibody testing mean TAT (9.1 days→6 days) and report production TAT (3.6 days→2.2 days). This resulted in compliance with the KPI: 95% of cases were reported in 10 days (October–December 2022). The improvement was due to removing manual data entry processes, reducing wait times between procedures and eliminating duplication of work.

Evelien Little, Nicola Brosnan, Jade Kally, Adrian Silk, Lisa Walsh, Franco Tavarozzi

Anthony Nolan Histocompatibility Laboratories, London, UK

With the onset of Next Generation Sequencing (NGS), HLA typing has become more streamlined with the ability to use either singleplex or multiplex primer strategies. The Anthony Nolan Histocompatibility Laboratory utilises GenDx NGSgo® singleplex primers for NGS to achieve high-allelic resolution typing for HLA-A, -B, -C, -DRB1, -DRB3,4,5, -DQB1, -DQA1, -DPA1, and -DPB1. Prior to this work, all samples received for HLA typing were tested for 11 loci, but the number of loci requested may vary. The use of singleplex, as opposed to multiplex, primers has allowed us to implement a flexible approach to provide a tailored and bespoke service. The aim of this work was to develop a process to type samples that need various combinations of testing on the same run. Modifications were needed to our Laboratory Information Management System (LIMS) to allow significant changes throughout the NGS testing process. As a result of this project and as our ongoing flexibility improvement process, we also implemented the ability to repeat PCR failures within the same run, allowing the process to proceed without delay, and merge multiple NGS libraries.

Since we no longer perform unnecessary testing, this generated cost savings by increased capacity on our Illumina® flow cells, allowing more samples per run, potentially reducing our turnaround times, as well as meeting our customer needs. To conclude, singleplex primers have proved extremely useful in allowing flexibility with no discernible increase in run time, using LIMS to simplify and track the process, end to end.

Victoria Wood^1,2, Brendan Clark¹, Eric Hewitt², Sunil Daga¹

¹Leeds Teaching Hospitals NHS Trust, Leeds, UK; ²University of Leeds, Leeds, UK

A core aspect of a renal Transplant Immunology service is the detection of HLA antibodies in the context of prospective donors. The presence of HLA-antibodies is detrimental to chances of transplant and, as donor-specific antibodies (DSA), to graft outcomes. With growing numbers of highly sensitised patients requiring renal transplant across HLA-antibody barriers, an increased understanding of antibody functional characteristics could lead to more informed donor choices. The avidity of an antibody-antigen interaction provides insight into the antibody's capability to induce antibody-mediated changes, such as intracellular signalling, leading to tissue remodelling and graft damage. Chaotropic agents reduce protein stability and have been previously used within ELISA protocols to estimate avidity of antibody-antigen interactions through a process called chaotropic disruption.

Modification of our standard One Lambda LABScreen Single Antigen bead (SAB) protocol to include a chaotropic agent has demonstrated technical viability of applying chaotropic disruption to a solid-phase assay and that this manipulation can be confidently interpreted. A maximum molarity of chaotropic agent was established which is not detrimental to bead surface antigen integrity - 1 molar. This was verified using flow cytometric analysis with HCA2 monoclonal, which binds to both native and denatured HLA-Class I, in comparison with HLA-A, -B, -C monoclonal which binds to native HLA-Class I only. Initial testing using sera with known HLA antibody profiles showed antibody-specific patterns of binding and disassociation. This novel method represents a potential accessible method of testing HLA antibody avidity in an NHS laboratory setting.

Mr Steven Jervis^1,2, Dr Antony Payton³, Dr Marcus Lowe^1,2, Dr Altug Didikoglu⁴, Professor Arpana Verma², Professor Kay Poulton^1,2

¹Manchester Transplantation Laboratory, Oxford Road, United Kingdom; ²Faculty of Biology, Medicine and Health, Division of Population Health, Health Services Research and Primary Care, School of Health Sciences, University of Manchester, Manchester, United Kingdom; ³Division of Informatics, Imaging & Data Sciences, School of Health Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester, United Kingdom; ⁴The Centre for Biological Timing, Division of Neuroscience, School of Biological Sciences, University of Manchester, Manchester, United Kingdom

Studies have hypothesised that the combination of pre-existing genetic traits and specific environmental triggers determine the onset of narcolepsy. The most impactful genetic risk factor is the presence of Human Leukocyte Antigen (HLA) DQB1*06:02 encoded on the Major Histocompatibility Complex (MHC), however, the presence of HLA-DQB1*06:02 is not ubiquitous in all narcolepsy cases. The most poignant genetic risk factors outside the MHC are predominantly located in genes associated with the immune system. In addition to the traditional symptoms of narcolepsy, the co-morbidities can vary with a cohort of sufferers complaining of cognitive dysfunction, particularly memory and attention. These self-reports are not substantiated by consistent scientific evidence whereas there is significant evidence outlining the genetic contribution underpinning variation in cognitive abilities in the general population. In this study we impute targeted non-MHC narcolepsy associated single nucleotide polymorphisms (SNPs) from 1,558 non-pathological elderly volunteers who have been followed for change in cognitive function for up to a 24-year period. Specifically, we investigate 13 previously documented narcolepsy associated SNPs with a odds ratio greater than or equal to 1.00 combined with a minor allele frequency of greater than 0.05. We observed an association between rs306336, rs4290173 and rs2834168 and a faster decline in long term memory. Similarly, we observed a protective effect of rs10995245 against the decline of long-term memory loss. This investigation suggests that the cognitive problems reported by cohorts of narcoleptic patients may be due to genetic predispositions and supports the variation seen in the co-morbidities associated with narcolepsy.

Raji Patel, Aliyye Karasu, Liezelle Pagala, Jyoti Bhatt, Edgar Correa, Gloria Adeyemo, Maame-Esi Yeboah, Abigail Sarkodie, Andrew Joahill, Suzette Cavanna, Carla Rosser, Colin Brown

NHS Blood and Transplant-Colindale, London, UK

The British Bone Marrow Registry (BBMR) utilises Oragene® saliva collection kits to facilitate extended HLA typing of potential donors by Next Generation Sequencing (NGS). Saliva is a good alternative source of DNA as its collection is non-invasive, allows for self-collection and is stable at ambient temperature. In January 2023, the Qiagen EZ2 robot was introduced into our laboratory to provide DNA extraction for clinical samples. However, Qiagen were unable to provide a validated protocol for the extraction of DNA from saliva samples using this instrument. In this study, we assessed the capability of Qiagen EZ2 tissue kits to obtain DNA from saliva samples using the Qiagen EZ2 Connect robot. Using the protocol, we developed, DNA was successfully extracted from 10 saliva samples, with an average DNA concentration of 33 ng/μl and a 260/280 ratio of 1.72. The quality of DNA was comparable to DNA extracted from saliva samples using the validated Roche Magnapure method. The suitability of saliva DNA for high resolution HLA typing was assessed using the One Lambda FASTplex kit. Using TypeStream Visual (TSV) NGS analysis software, we achieved average mapped reads of 334,691 and the results met our documented acceptance criteria for all quality metrics. In conclusion, we have demonstrated that the Qiagen EZ2 Connect robot can be used for successful extraction of DNA from saliva samples to a comparable quality of blood. Validation of this process has allowed our department to avoid unnecessary delays in BBMR extended HLA typing requests and thereby supports patient care.

Sebastian Fernando, Jennifer Lord, Nicola Martin, Alison Logan, Kay Poulton

University of Manchester NHS Foundation Trust, Manchester, UK

Our current routine HLA typing methodology to support disease association testing is LABType™ SSO (One Lambda). An alternative HLA typing methodology EUROArray (EUROIMMUN), was evaluated for concordance and efficiency. This system combines polymerase chain reaction and microarray technologies. Amplified target DNA labelled with a fluorescent dye hybridises to complimentary DNA probes using BIOCHIP technology. Fluorescence signals are evaluated automatically using the EUROIMMUN Microarray Scanner and EUROArrayScan software. Routine and external proficiency scheme peripheral blood and DNA samples previously HLA typed by LABType™ SSO were tested using the appropriate EUROArray assay. Forty-six samples were tested with the HLA-B27 Direct assay, 41 samples with the HLA-B*57:01 Direct assay and 42 samples with the HLA-DQ2/DQ8-h Direct assay. A range of relevant HLA-B and HLA-DQ alleles were selected to ensure the EUROArray system could differentiate between alleles of interest. We found 100% concordance when compared to the LABType™ SSO results for all tests. The workflow was simple and straightforward with a time saving when compared to LABType™ SSO, however the cost of the EUROArray assay was higher. This study has shown the EUROArray Direct assays for HLA-B27, -B57 and -DQ2/8 detection are valid alternative methodologies to support disease association testing. Advantages of the EuroArray system include no additional DNA isolation, the inclusion of numerous integrated controls for high reliability of results and fully automated standardised evaluation and result generation. Limitations of this assay include the use of blood samples within 14 days of venepuncture, although samples can be frozen to mitigate this.

Amy De'Ath, Deborah Pritchard, Tracey Rees

UK NEQAS for H&I, Talbot Green, UK

Scheme 2A and 2B assesses participants’ ability to correctly determine the cytotoxic and flow cytometry crossmatch status across 40 cell/serum combinations per year, respectively. An analysis of performance by UK and Ireland laboratories from 2018–2022 was performed.

There were between 15–22 participants in 2A and 19–22 in 2B. Unsatisfactory performance (UP) in 2A ranged between 0% (2020/21)–38.9% (2018), latest 20%. UP in 2B ranged from 0% (2020/21)–10.5% (2022). Over the 5 years, four laboratories had UP in 2B (one lab in two consecutive years) and 11 in 2A. 2 labs had UP in both schemes. Two out of four labs with UP in 2B were due to performance in the T-cell crossmatch, 1/4 in B-cell and 1/4 in T and B cell crossmatch performance. Six out of eleven labs with UP in 2A were in the B-cell without DTT category, 1/11 in B-cell with DTT, 3/11 in both B-cell with and without DTT and 1/11 in T-cell with and without DTT and B-cell with DTT. In 2B, an average of 36/40 crossmatch combinations per year were assessed. An average of 3% assignments per year were incorrect with a 50:50 split of false negatives and false positives. The use of equivocal reporting was low, average 0.5%. Four percent of samples were reported as not tested. Performance in flow cytometry crossmatching is better than cytotoxic crossmatching during the 5 years. These schemes offer a technical assessment of crossmatching, NEQAS encourage laboratories to participate in our educational schemes which more closely mimic clinical practice.

Amy Bedford, Kathryn Howson, Graham Knighton, Jacqueline Pires, Sarah Maxfield

Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK

HLA-B*27 testing is supportive in the diagnosis of autoimmune diseases including Ankylosing Spondylitis due to their strong association with the HLA-B*27 positive genotype. Testing an average of 383 samples per month, Cambridge Tissue Typing laboratory aimed to introduce a direct-from-blood HLA-B*27 detection assay to eradicate DNA extraction required by current polymerase chain reaction (PCR) sequence specific primer (SSP) based methods, and the associated workload. Utilising the BAG Diagnostics FastQ® B*27 Direct assay, DNA is amplified directly from EDTA whole blood by PCR using SSP. Fluorescent-labelled TaqMan® hydrolysis probes are utilised to enable detection of the amplified product by real-time PCR on the QuantStudioTM 3. Implementation of the FastQ® B*27 Direct assay reduced DNA extraction workload by an average of 64%. Increased batch size capacity (93, vs. 20 using the previous technique) streamlined workflows. Risks associated with exposure to carcinogens ethidium bromide and ultraviolet light and handling boiling agarose gel were eradicated. The absence of DNA extraction and adjustment leads to variance in the volume of nucleated cells utilised, which can result in delayed amplification/false negative results. This highlights the importance of sufficient sample mixing prior to aspiration and stringent result acceptance criteria during analysis using PlexTyper® software, with an average repeat rate of 4%. In summary, implementation of robust mixing procedures and stringent quality metrics have facilitated transition to BAG Diagnostics FastQ® B*27 Direct assay leading to a decrease in workload and demands on staff time, and improvement in the health and safety within the laboratory.

Kelly Spence, Sylvia McConnell, David Turner

H&I Department, SNBTS, Edinburgh, UK

H&I supports the Clinical Apheresis Unit (CAU) and Tissues, Cells and Advanced Therapeutics (TCAT) departments by performing CD34+ stem cell enumeration in acute myeloid leukaemia and multiple myeloma patients receiving autologous stem cell transplantation. The enumeration of CD34+ cells using flow cytometry and the International Society of Hematotherapy and Graft Engineering (ISHAGE) protocol is an established method for the evaluation of stem cell numbers in peripheral blood and apheresis products. Whilst a single (flow cytometry) platform is the recommended approach, locally a dual platform method is in operation, using both flow cytometry and Haematology analyser data. The BD Stem Cell Enumeration assay has been validated on the BD FACSLyric flow cytometer. A comparison between the single and dual platform methodologies using the ISHAGE protocol was undertaken. Analysis was performed on 30 stem cell harvest samples. Strong correlation between the single and dual platform methods was observed for CD34+ absolute count (cells/μl) for stem cell harvest samples (r = 0.99, p < 0.05). There was also a strong correlation between the White Blood Cell (WBC) count from the Haematology analyser and the CD45+ count from the flow cytometer (r = 0.94, p < 0.05). In this validation both the single and dual platform methods for calculating CD34+ cell counts were similar, as were the WBC counts and CD45+ counts from the Haematology analyser and the flow cytometer. This data supports using the recommended single platform method as part of the CD34+ enumeration testing.

Rebecca McGuire^1,2, Paul Wright³, Steven Jervis¹, Stephine Whiteside¹, Malcolm Guiver⁴, Kay Poulton^1,2

¹Transplantation Laboratory, Manchester Royal Infirmary, Manchester University NHS Foundation Trust, Manchester, UK; ²Faculty of Biology, Medicine and Health, Division of Medical Education, School of Medical Sciences, University of Manchester, Manchester, UK; ³H&I Laboratory, Liverpool Clinical Laboratories, Liverpool, UK; ⁴Department of Clinical Virology, Manchester University NHS Foundation Trust, Manchester, UK

Acute myeloid leukaemia is an aggressive haematological malignancy with a poor prognosis. Allogeneic haematopoietic progenitor cell transplantation is the only available curative treatment, but it does not eliminate the risk of relapse. Hypotheses for improving relapse rates include utilising donors with advantageous KIR2DL1 allele groups. However, an inexpensive, rapid, and reliable method to discern KIR2DL1 groups does not currently exist. Here, we show that homology between several Killer-cell Immunoglobulin-like Receptor (KIR) loci restricts accurate genotyping of KIR2DL1 allele groups by traditional TaqMan™- based real-time PCR methods. Using multiple sequence analysis, we found that the single nucleotide polymorphisms within codons 114 (rs11673144) and 245 (rs34721508), used to discern KIR2DL1 allele groups, are shared between other KIR loci. In a traditional TaqMan™ real-time PCR assay, this results in non-specific binding and incorrect allele group assignment. Contrary to existing literature, our analysis demonstrated that specificity could not be achieved through the introduction of 3´ terminal mismatches in the forward primers. The results described illustrate the challenges in designing an efficient real-time PCR assay for functional group typing of KIR2DL1. We anticipate our assay to provide the foundation for a more sophisticated real-time PCR assay. Developing a test capable of defining KIR2DL1 allele groups remains of interest to research groups focused on reducing the incidence of acute myeloid leukaemia relapse post-transplant. With further improvements, the described assay could fulfil this goal.

Abigail Levy, Arthi Anand, John Wintour-Pittom

Hammersmith H&I NWLP, London, UK

The Histocompatibility & Immunogenetics (H&I) department at NWLP support renal transplantation at West London Renal Transplant centres. A 24/7 on-call service operates for crossmatching of local patients for transplantation. Timely and effective communication between the renal team receiving deceased donor offers and H&I on-call team is critical for turnaround of compatibility assessment and testing. Historically H&I on-call team have been contactable via ICHT switchboard with provision of a weekly on-call rota. A number of challenges were experienced using this method including calls being directed to the wrong team member interrupting rest and the potential for GDPR breach with patient identifiable information recorded in home environment. ICHT implementation of ALERTIVE app in June 2022 provided alternative on-call communication. H&I were the first pathology service to adopt the app and collaborated with Trust Telecoms & ThamesNet Services to set up the App to meet H&I on-call needs, with Go-Live in February 2023. ALERTIVE improves the speed and quality of communication, between clinical staff and H&I on-call team, in time critical deceased donor transplantation pathway. The ALERTIVE app has been in routine use for three months and has had a very favourable response from H&I on-call team. The app has simplified the process of communicating between Renal and H&I on-call teams, enabling more efficient decision-making. The clinical messaging app is giving us a wealth of new data that was either very hard to get or was unavailable including number of offer related call outs. There are exciting new uses for this data including workforce planning.

Dayna Badaro^1,2, Sarah Maxfield¹

¹Cambridge University Hospital NHS Foundation Trust, Cambridge, UK; ²University of Manchester, Manchester, UK; ³Organ and Tissue Donation and Transplantation, NHS Blood and Transplant, Stoke Gifford, UK

In the United Kingdom, prior to September 2019, deceased donor kidneys were allocated following the National Health Service Blood and Transplant-Organ Donation and Transplantation (NHSBT-ODT) 2006 Kidney Allocation Scheme (KAS). A review by the Kidney Advisory Group prompted significant changes to national allocation with the aim of improving fairness in kidney offering, reducing transplant waiting times and improving longevity of matched transplants. A retrospective clinical audit aimed to review the impact for patients on the Cambridge deceased donor kidney transplant waiting list (TWL), and determine whether the objectives of the 2019 KAS have been met. Transplant data was obtained from NHSBT-ODT spanning 24 months pre-(n = 298) and post-(n = 303) implementation. Data was reviewed taking into consideration the impact of the COVID-19 pandemic. Results show reduced average wait time (days) for highly sensitised patients with a calculated reaction frequency (cRF) ≥85%. In accordance with simulation data, donor/recipient index matching has been achieved, with the majority of ‘low risk’ donors allocated to ‘low risk’ recipients, and the same being true for ‘high risk’ donors/recipients. Divergent donor/recipient age matching was accounted for by difficult to match patients prioritised according to Tier A (cRF 100%/wait time >7 years/matchability score of 10). Contrary to simulation data, Cambridge did not observe a reduction in donation after cardiac death (DCD) transplants following changes to national allocation (n = 174 pre, n = 192 post). In summary data indicates a reduction in average wait time for highly sensitised patients on the Cambridge TWL and suggests objectives of the 2019 KAS have been met.

Michelle Carr¹, Shelley Harris¹, Judith Worthington¹, Alex Woywodt², Kay Poulton¹

¹Transplant Laboratory MFT, Manchester, UK; ²Lancashire Teaching Hospitals NHS Foundation Trust, Preston, UK

In January 2020 a 24-year-old male with chronic renal failure secondary to reflux nephropathy received his third kidney transplant from a fully HLA matched DBD donor. In March 2020 his kidney function declined with a concurrent rise in creatinine and proteinuria. Tacrolimus levels were variable throughout the course of the patient's transplant history and non-adherence was suspected. A retrospective analysis of non-HLA antibodies using LABScreen™ Autoantibody kits identified that the patient developed antibodies to Glutathione S-Transferase Theta 1 (GSTT1). After his second transplant tacrolimus levels were below target range (<2.5 μg/L), which was followed by the first appearance of the GSTT1 antibody (6000 MFI = 95% percentile). This coincided with a biopsy showing chronic transplant glomerulopathy. Following his third transplant tacrolimus levels remained variable with some levels above and below target range. His transplant function deteriorated further in Spring 2023. A biopsy at this time showed borderline T cell mediated rejection with moderate interstitial fibrosis and tubular atrophy. There was a rapid increase in GSTT1 antibody levels with a peak of 11,000 MFI (>95% percentile). We propose that immune-mediated inflammatory processes triggered by non-adherence episodes caused cellular damage. In response to such damage, intracellular components such as GSTT1 would be released, enabling recognition and the subsequent immune response leading to GSTT1 antibody production. We propose that the detection of antibodies to GSTT1 should be regarded a marker of nephrotoxicity and ongoing trauma to the transplanted kidney.

Adrian Handley, David Briggs, Clare Collins

NHS Blood and Transplant, Birmingham, UK

All measurements come with an element of uncertainty and are only truly useful when that degree of uncertainty is understood. The use of semi-quantitative data from Luminex HLA antibody testing is essential in enabling cardiothoracic transplant compatibility assessment; however, the level of uncertainty was not understood. BSHI/BTS CTAG guidelines set risk levels based on MFI values produced in Luminex Single Antigen Bead HLA antibody assays. Additionally, ISO15189:2012 states that laboratories should calculate measurement uncertainty (MU) for all accredited tests. In this study we used simple statistical tools to identify degrees of MU and some of the contributing factors. These allowed us to quickly assess the effectiveness of changes subsequently made. We identified significant difference between results obtained by individual testing personnel and were able to reduce this through our interventions. Reed et al. (Am J Transplant. 2013;13(7):1859–1870) suggested MU as %CV was in the region of 20%–62%. Our study found this to range from 8% at 20,000 MFI up to 18% at 1000 MFI, the threshold for positivity.

The tools we developed allow ongoing monitoring of assay performance and provide the basis for future quality initiatives, contributing to sustainable, evidence-based quality improvement. MU calculations are an effective quality assurance and improvement tool and are easily implemented. By combining the tools used in this study, levels and sources of variation can be identified, and the effect of any changes to process can be easily assessed for effectiveness.

Saima Azhar Salim, Louise Walsh, Geraldine Donnelly, David Keegan, Joseph Kelly, Mary Keogan

H&I Lab, Beaumont Hospital, Dublin, Ireland

Flow cytometric crossmatch (FXM) is performed during pre-transplant histocompatibility workup. False positive results may prevent a potential recipient receiving a suitable transplant. Many labs use a three-colour FXM assay established using a dual-laser flow cytometer, which includes phycoerythrin (PE), fluorescein isothiocyanate (FITC) and peridinin-chlorophyll proteins (PerCP). There is significant spectral overlap between PE and FITC potentially leading to high background fluorescence, and possible false positive B cell FXM. This study was undertaken to determine optimum fluorochrome combinations to minimise spectral overlap and maximise consistency with the current method for which clinically validated cut-offs are available. Fluorochromes Brilliant Violet (BV421) and Allophycocyanin (APC) were selected for labelling CD3+T cells and CD19+B cells respectively. Thirty-four FXM (total 63 samples for T cell and 61 samples for B cell FXM) were performed with fluorochromes CD3BV421, CD19APC and anti-human IgG FITC. Fourteen FXM (total 35 samples for T and B cell FXM) were performed with fluorochromes CD3PerCP, CD19APC and anti-human IgG FITC. FXM assays were performed on Becton Dickenson (BD) FACSLyric™ and results compared to the standard FXM assay. In comparison to the standard FXM, a statistically significant difference in sample T cell ratio and B cell ratios (p < 0.005) were observed for CD3BV421-CD19APC FXM. Our pilot data suggest no significant difference in T and B cell ratios for CD3PerCP-CD19APC (p > 0.05). Data suggest that substitution of CD19PE with CD19APC minimises spectral overlap, reducing the risk of false positive FXM results, without significantly altering T and B cell ratios.

Ana Bultitude¹, Anthony Poles², Sue Jordan¹, Anthony Calvert², Deborah Sage¹

¹NHS Blood and Transplant, Tooting, UK; ²NHS Blood and Transplant, Filton, UK

HNA-3 is a bi-allelic antigen, -3a/-3b, with 95% of the UK population encoding at least one HNA-3a allele. HNA-3 is expressed on a variety of cell types including lymphocytes and renal endothelial cells. Accordingly, HNA-3 specific antibodies, which can develop against the non-self-variant in homozygous individuals, have been implicated in rejection episodes following renal transplantation. Here, we report a case of a renal transplant recipient displaying anti-HNA-3a antibodies that prevented transplantation. In August 2022, the patient was offered an altruistic kidney with a 2,1,1 mismatch grade, yielding a negative virtual crossmatch result. However, final wet laboratory flow cytometric crossmatch (FCXM) results presented an unexplained strong positive T and B cell result with a negative auto FCXM. The patient had no previous transplant history and no detectable HLA antibodies since their initial referral. HNA genotyping was performed on both patient and donor, determining HNA-3b3b and HNA-3a3a genotypes respectively. Patient sera was screened for anti-HNA antibodies, confirming the presence of HNA-3a-specific antibodies. The patient was removed from the virtual crossmatching programme as they no longer met the eligibility criteria but remained on the deceased donor waiting list. A further nine deceased donor kidneys were accepted and crossmatched, all of which were T and B cell positive. In March 2023, the patient received a deceased donor kidney which produced a T and B cell negative FCXM. The donor was retrospectively genotyped as HNA-3b3b. The patient continues to do well with no antibody-mediated rejection detected thus far.

Hawzhin Jabar, Luke Foster

NHS Blood and Transplant, Birmingham, UK

Accurate HLA typing is essential to facilitate safe solid organ transplantation. Within the UK, H&I laboratories are required to meet the minimum typing requirements set by NHSBT-OTDT, which includes reporting results that allow for the assessment of any donor specific antibodies (DSA) in a particular donor-recipient combination. Failure to do so can lead to inappropriate allocation or an increased risk of transplant rejection. Here we report the identification of the DRB1*14:15 allele in a deceased solid organ donor who was HLA typed on-call using LinkSeq real-time PCR (One Lambda). Ordinarily, it would not be a requirement to report HLA-DRB1*14 to the second field under the minimum typing requirements, however, interestingly, HLA-DRB1*14:15 does not encode the DR14 antigen, but codes for the DR8 antigen. Therefore, in this instance there was a requirement to report HLA-DRB1*14 to the second field to allow accurate allocation and assessment of any potential DSA. Despite being locally rare, HLA-DRB1*14:15 is listed as well-documented in European populations within the CIWD 3.0.0 catalogue, and common in Asian/Pacific Islands and Native American populations. In this case, donation proceeded with both kidneys being transplanted in two patients, one locally, and out of region. Although the patient transplanted locally was cRF 0% and therefore had no DSA, elucidation of the DR8 antigen may be important for any future post-transplant DSA monitoring.

Richard Battle, Emma Ross, Sylvia McConnell, David Turner

H&I Department, SNBTS, Edinburgh, UK

The SNBTS H&I laboratory provides ∼2300 apheresis donor platelet units annually to HLA sensitised refractory patients across Scotland from a panel of ∼800 typed donors. Data is recorded for each transfusion, including match grade (A = matched for HLA-A and B antigens, B1 = one HLA antigen mismatch, B2 = two antigen mismatch etc.) and cumulative HLA antibody MFI (cMFI) against mismatches. Platelet transfusions were grouped by match and cMFI and assessed against post counts relative to pre-counts that is, a measure of transfusion increment. Pre- and post-platelet counts were available on 1486 cases between 2015–2022. Recipient and donor HLA-A and B types at first field defined A, B1–B4 matching. cMFI was calculated following One Lambda SAB I testing. Analyses used ANOVA or t-test between groups. In 1166 HLA compatible transfusions (cMFI < 2000) no differences were seen in mean increment between A, B1, B2, B3 and B4 matches; 22.5, 22.9, 22.1, 24.1, 30.2 respectively (ANOVA p = 0.14). In 1486 patients grouped according to cMFI, cases with < 2000 had a mean increment = 23.0, with > 2000 < 10,000 mean = 18.0 (p < 0.001) and > 10,000 mean = 8.0 (ANOVA p = 1.55 × 10–15). This analysis of platelet counts after HLA selected platelet provision shows that, as expected, the match grade of the platelets does not impact on the immediate increment. The level of HLA antibody, as defined by cMFI, affects the post transfusion count, especially when cMFI > 10,000. This data will help locally in selection of optimal platelet units for patients.

Patrick Flynn¹, Sebastian Fernando², Judith Worthington¹, Kay Poulton¹

¹Transplantation Laboratory, Manchester Royal Infirmary, Manchester, UK; ²School of Health Education and Public Health Sciences, University of Manchester, Manchester, UK

The aim of this study was to devise an algorithm that would predict Flow Cytometry crossmatch results using SAB Median Fluorescent Intensity (MFI) levels and to test this correlation using samples tested from a NEQAS Scheme 2B cohort.159 NEQAS 2B serum samples were screened using LABScreen™ SAB and 40 NEQAS 2B peripheral blood samples were HLA typed with LABType™ SSO. Donor-Specific Antibodies (DSA) were identified for each cell-serum combination tested and cumulative MFI values calculated for each test. HLA Class I MFIs were combined to predict the T cell crossmatch. For the B cell crossmatch prediction, two options were considered: (i) HLA Class II MFI values alone and (ii) HLA Class I + Class II MFIs. Receiver Operating Characteristic analysis was carried out to identify the combined MFI cut off that predicted NEQAS consensus results with the greatest sensitivity and specificity value. HLA Class I combined MFI > 5000 predicted T cell crossmatch results with 96% sensitivity, 100% specificity, 100% Positive Predictive Value (PPV) and 92% Negative Predictive Value (NPV). For B cell results, HLA Class I + Class II combined MFIs > 11,000 gave the best model showing 97% sensitivity, 79% specificity, 95% PPV and 85% NPV. However, for samples with only HLA Class II sensitisation, combined MFIs > 13,000 improved the B cell crossmatch predictions: 92% Sensitivity, 91% specificity, 92% PPV and 91% NPV. Using this model, combined MFI values can be used to predict the immunological risk posed by DSA when it is not possible to carry out a crossmatch test.

Jonathan A.M. Lucas¹, Richard M. Szydlo^1,2, Shelley Hewerdine¹, Steven G.E. Marsh^1,3, Neema P. Mayor^1,3

¹Anthony Nolan Research Institute, Royal Free Hospital, UK; ²Department of Medicine, Imperial College, UK; ³UCL Cancer Institute, Royal Free Hospital, UK

The effect of matching for HLA-E on the outcome of hematopoietic cell transplantation (HCT) has thus far been inconsistent and has not been studied in a UK cohort where there is an abundant use of alemtuzumab for T-cell depletion. We analysed HLA-E genotypes in 1513 UK HCT patients with a haematological malignancy and their unrelated donors using full-length PacBio Single Molecule Real-Time DNA sequencing. After adjusting for clinical factors that affected outcome prognoses including classical HLA matching out of 12, the presence of mismatches at both HLA-E loci (HLA-Emm; n = 30) was significantly associated with a reduced risk of relapse (HR 0.44; 95% CI 0.20–0.98, p = 0.04) in comparison to being HLA-E matched (HLA-Em; n = 788). A non-significant but beneficial effect on Overall Survival (OS) and Event-Free Survival (EFS) was also observed for two HLA-Emm loci compared to HLA-Em (OS: HR 0.73, p = 0.30; EFS: HR 0.77, p = 0.38). There were no significant differences in HCT outcomes correlated with a single HLA-Emm (n = 450). Assessing directionality of mismatches showed that a bi-directional HLA-Emm (n = 45) was significantly associated with reduced risks of relapse (HR 0.40; 95% CI 0.20–0.83, p = 0.01), increase in EFS (HR 0.55; 95% CI 0.32–0.93, p = 0.02) and a non-significant increase in OS (HR 0.72, p = 0.16) compared to HLA-Em. We hypothesise that mismatching alleles at the HLA-E locus results in sufficient genetic disparity to provide a stronger Graft-versus-Leukemia effect, without eliciting detrimental Graft-versus-Host responses, hence no significant differences in acute Graft-versus-Host Disease or Transplant Related Mortality, as observed in this study.

Sajadhossein Bazrafshani¹, Mohammadreza Bazrafshani²

¹ENT Department, Worcester, UK; ²Kerman University of Medical Sciences, Kerman, Iran

Cyclosporine, a calcineurin inhibitor, has a narrow therapeutic index and shows considerable inter-individual variability in pharmacokinetics. Cyclosporine is a P-glycoprotein (P-gp) substrate, a multidrug resistance gene (MDR1) product. Some of the single nucleotide polymorphisms (SNPs) of MDR-1 correlate with the variable activity of P-gp in vivo, and it is thought that these polymorphisms are associated with pharmacokinetic variations in cyclosporine therapy. Genotyping assays (PCR-SSP and PCR-RFLP) were performed for detection of frequency within two functional MDR-1 SNPs (C1236T in exon 12 and C3435T in exon 26) in 60 patients and the correlation between genotyping and concentration/dose ratio of cyclosporine was investigated. Data analysis revealed that C3435T polymorphism correlated with the concentration/dose ratio significantly. The concentration/dose ratios were 59.36 μg/L/kg/12 h (p < 0.001) and 30 μg/L/kg/12 h (p = 0.049) lower in homozygous wild-type patients (CC) rather than homozygous mutant-type (TT) and heterozygous patients (CT) respectively. Moreover, heterozygous patients (CT) had 29.36 μg/L/kg/12 h (p = 0.001) which was less than homozygous mutant-type patients (TT). These findings suggest that, for the given dose, the blood concentration is lower in homozygous wild-type individuals (CC type in 3435 position). Our results revealed that MDR-1 genotype appears to influence cyclosporine drug levels and MDR-1 genotyping may provide a useful clinical guide in predicting the required dose after renal transplantation.

Mazen Mabrok¹, Renuka Palanicawander², Betia Nouri¹, Rachel Smith¹, Arthi Anand¹, Natalia Brodaczewska³

¹Histocompatibility & Immunogenetics Laboratory, North West London Pathology, Imperial College Healthcare NHS Trust, London, UK; ²Centre for Haematology, Imperial College Healthcare NHS trust, London, UK; ³Specialist Integrated Haematological Malignancy Diagnostic service (SIHMDS), Imperial College Healthcare NHS Trust, London, UK

Monitoring of donor chimerism after haematopoietic stem cells transplantation (HSCT) is vital for early effective therapeutic interventions. Short tandem repeat (STR) assays are the current gold standard for chimerism monitoring after allogeneic HSCT. The emergence of Next Generation Sequencing (NGS) CE-IVD approved assays with improved limit of detection of around 0.05% offer a promising alternative. In this evaluation exercise, we explored the technical capacity of NGS in post-transplant chimerism monitoring, by retrospectively analysing samples from 13 patients from our centre using the Dvysr® NGS chimerism assay (Sweden, Stockholm). All patients enrolled in this evaluation were tested at three time points (whole blood & T- Cell), selected in collaboration with the clinical team based on clinical signs of relapse, increase in MRD markers and the profile obtained using the PowerPlex 16 Multiplex STR system (Promega). Our results showed very strong correlation between the NGS and the STR assays (Pearson score 0.998 & 0.999) with a shift of 1%–1.5% at the pre-relapse time point observed in 55% of the patients enrolled in this evaluation. The NGS assay needed 50% less time to analyse and in 70% of the patients tested, the NGS assay yielded more informative markers than the STR assay. NGS chimerism assay promises improved diagnostic performance and usability in our evaluation. Increased sample size and inclusion of lineage specific cell separation will further strengthen validity of the improved diagnostic performance and usability of NGS chimerism assay as alternative to STR assays including correlation with clinical presentation.

Emma Holmes¹, Jasmaine Lee², Winnie Chong², Deborah Sage¹, Martin Howell²

¹NHSBT Tooting Centre, Tooting, UK; ²NHSBT Service Development, Colindale, UK

Nanopore sequencing presents a new technology for high resolution HLA typing that is, considered faster and potentially cheaper than existing methods used for next generation sequencing (NGS). NanoTYPE (Omixon) can be used for batches of up to 24 samples or for single samples, which has the potential to be used for deceased donor HLA typing. The aim of this study was to evaluate the NanoTYPE assay in two NHSBT H&I laboratories; Colindale Service Development and Tooting, using the same batch of 96 samples that had previously been typed by current rapid HLA typing or NGS methods. All NanoTYPE reagents, R9.4 flow cells and a MinION mk1B device were provided by Omixon and Oxford Nanopore Technologies for this study. Eight samples were set-up using the single-sample protocol. The remaining samples were set-up in multi-batches of varying sizes. There was 100% concordance at second field resolution between both sites, and with previous typing data, for HLA-A, -B, -C, DRB4, DRB5 and DPA1. Non-concordance with previous typing data was seen for either HLA-DRB1, DRB3, DQA1, DQB1 or DPB1 loci in 10/96 (10%) samples at Tooting and 9/96 (9%) samples at Colindale, which was attributed to allele imbalance, known low amplification of certain genotypes such as DQB1*03, allele dropout and detection of potential novel alleles. After manual review of the sequencing data, 95/96 samples at Colindale and 93/96 samples at Tooting were concordant with previous results. Our study has proven that the NanoTYPE assay is simple to use and enables high resolution HLA typing.

Emma Ross¹, Sylvia McConnell¹, Richard Battle¹, Nicole Priddee², David Turner¹

¹H&I Department, SNBTS, Edinburgh, UK; ²Donor Medicine, SNBTS, Edinburgh, UK

The SNBTS H&I laboratory is responsible for the provision of HLA selected platelets in Scotland. In 2022, 96 patients were supported with a total of 2317 platelet units. A key performance indicator (KPI) for the service is >60% of allocated donor platelets should be an A grade or B1 grade match with the patient (A grade = no HLA-A or HLA-B mismatches at first field; B1 = only one HLA-A or HLA-B mismatch). This target has averaged 57.7% (range 46%−72%) between January 2016 and September 2022, with an overall downward trend observed. In contrast, the number of individual donor platelets issued as HLA selected increased from 1288 units in 2016 to 2317 in 2022 (+55.5%). To make the best use of a limited donor panel, H&I have worked with SNBTS Donor Services (DS) to increase availability of best matched donor platelets. Utilising a Quality Improvement approach, business analytic tools and specialist donor communications, we concentrated on the national identification and (re)engagement of HLA-A and -B homozygous donors which constitute 3.5% of the panel but have the potential to provide A grade matches for ∼68% of patients (Jan–June 2023). Since the project initiation we have recorded an upward trend in the percentage of A and B1 platelet allocations over the period October 2022 (49.5%)–May 2023 (72.4%). The target of >60% has been achieved consistently from Jan 2023. Collaborative working with DS has enhanced the donor panel, improving ‘off the shelf’ access to best matched HLA selected platelets.

Felicity May¹, Sian Griffin², Madhvi Menon³, Tracy Hussell³, Tracey Rees¹

¹Welsh Transplantation and Immunogenetics Laboratory, Welsh Blood Service, Pontyclun, UK; ²Department of Nephrology and Transplantation, University Hospital of Wales, Cardiff, UK; ³Division of Immunology, Immunity to Infection and Respiratory Medicine, University of Manchester, Manchester, UK

Human leukocyte antigen incompatible (HLAi) transplantation remains an important option for very highly sensitised patients. Crossing the HLA barrier is associated with increased risk of antibody-mediated rejection (AMR) and graft failure. Desensitisation, induction and maintenance immunosuppression aim to minimise this risk. There is poor consensus on optimal treatment protocols, and patient response varies. In this single centre, retrospective study, we assessed incidence of rejection and graft survival in a clinical cohort of 27 HLAi transplant recipients desensitised with rituximab and cycles of double filtration plasmapheresis. There was poor association between established risk factors and incidence of AMR, suggesting the involvement of other, currently unknown, factors. HLA antibody response and AMR are influenced by a complex cytokine network supporting the generation and survival of antibody producing cells. We conducted a literature review and graded 459 serum protein analytes (cytokines, chemokines, hormones etc.) based on the quality of evidence and proximity of involvement in humoral memory response. We evaluated serum concentrations of the 40 highest graded analytes in longitudinal samples derived from our clinical cohort using a bespoke multiplexed Luminex assay. Due to low samples numbers and heterogeneity of the analyte profiles, we were unable to establish clear association with patient clinical outcomes. However, we observed significant and sustained changes to key signalling molecules known to influence germinal centres, B-cells and plasma cells. We noted a reduction in levels post-desensitisation/immunosuppression for the majority of our panel, but increased or unchanged levels of several key supportive signals and suppression of several immune regulators.

Agnieszka Ojrzynska¹, Kylara Hassall¹, Katie Butler¹, Graham Shirling¹, Sharon Vivers², Raymond Fernando¹

¹Solid Organ Group, Royal Free Hospital, London, UK; ²Anthony Nolan Histocompatibility Laboratories, London, UK

Increasing numbers of multiple myeloma patients, treated with Daratumumab (Dara), are being listed for kidney transplantation. Dara is a human monoclonal anti-CD38 IgG antibody that interferes with both pre-transplant allogeneic lymphocyte crossmatches and pre-transfusion compatibility testing, as it binds to CD38 molecules expressed on many blood cells including lymphocytes. This can simulate the presence of donor-specific antibodies (DSA) and lead to false positive crossmatch results. The aim of this study was to investigate methods to mitigate Dara interference with flow cytometry crossmatches (FCXM). Our centre has transplanted four patients undergoing Dara treatment. Three of these patients were HLA antibody negative and one patient was HLA antibody positive but HLA DSA negative. Six third party peripheral blood FCXMs, and one retrospective transplant FCXM with spleen cells, were performed using our standard procedure, as well as an amended protocol using donor cells pre-incubated with dithiothreitol (DTT) (0.05–0.1 M) which can cleave the stabilising disulphide bonds of the CD38 molecule. All third-party crossmatches using our standard procedure were T cell positive, two were also B cell positive. The retrospective crossmatch performed with spleen cells was T and B cell positive. All crossmatches performed with DTT were T and B cell negative. Our results show that DTT eliminates the false positivity observed when sera from Dara treated patients are used in FCXM. HLA positive control results were unaffected suggesting that HLA molecules are not affected by DTT treatment. Therefore, DTT can be used to mitigate DARA mediated false positive results in FCXM.

Daniel Eggleston, Helena Lee

Manchester Royal Infirmary, Manchester, UK

Haematopoietic progenitor cell transplantation (HPCT) is well-established as a curative treatment for malignant and non-malignant haematological disorders. However, patient monitoring is still essential following transplantation. Post-transplant donor chimerism testing can identify complications including relapse or graft failure. Relapse is a particular area of concern for patients diagnosed with Myelodysplastic syndromes (MDS) or high-risk leukaemias who have undergone HPCT. Most laboratories perform chimerism monitoring by PCR-Short Tandem Repeats (PCR-STR). This technique uses differences in short repeated sequences of DNA and capillary electrophoresis to determine percentage donor chimerism. However, other methods are emerging that offer improved sensitivity and more informative markers. One such technique is digital PCR (dPCR). During dPCR, a PCR reaction mixture is partitioned into individual droplets. Each function as an individual assay, with the presence of a reaction indicating donor or recipient DNA, increasing technique sensitivity. Using dPCR, we investigated eleven patients diagnosed with MDS or Myeloid Leukaemias whose percentage donor chimerism reached 100% by PCR-STR but dropped in subsequent samples. We tested our cohort using JETA DigitalTRACE™ dPCR technology with the Qiagen QIAcuity machine at two or more timepoints preceding this reduction in donor chimerism using three markers. In 10 of our 11 patients, a reduction in donor chimerism was identifiable by dPCR whereas analysis by PCR-STR had suggested samples were 100% donor chimerism. This raises the possibility of an earlier detection of changes in donor chimerism and an improved follow up process. Chimerism by dPCR is therefore a viable replacement for PCR-STR and may identify events earlier post-HPCT.