Sensitivity of red blood cell antibody screening test systems for anti-D: A multicentre study using external quality assessment programmes
A
ChristophBuchta1✉Emailchristoph.buchta@oequasta.at
WimCoucke2
CécileToly-Ndour3
AgnèsMailloux3
KarinaHellbert4
GianfrancoAvveduto5
IsabelleBertin-Jung6
AnnaBode7
LobnaBouacida2
StefanieGemein7
SabineGoseberg8
AndreaGriesmacher1,9
RichardHaggas10
JunhoKim11
TruschaNiekerk12
Jean-PascalSiest6
NannaSkeie13
HeidiStøen13
RenateBecker14
ThomasWagner15
GüntherF.Körmöczi1✉,16Emailguenther.koermoeczi@meduniwien.ac.at
F.Günther16
Körmöczi16
1Austrian Association for Quality Assurance and Standardization of Medical and Diagnostic Tests (ÖQUASTA)Hörlgasse 18/51090ViennaAustria
2Sciensano, BrusselsBelgium
3Centre National de Référence en Hémobiologie Périnatale (CNRHP)ParisFrance
4Maybach Bechter Hellbert Rechtsanwälte GesbRViennaAustria
5CRRVEQ - SOD Sicurezza e Qualità, AOU CareggiFirenzeItaly
6Biologie ProspectiveVillers-les-NancyFrance
7Reference Institute for Bioanalytics (RfB)BonnGermany
8INSTAND e.V. Society for Promoting Quality Assurance in Medical LaboratoriesDüsseldorfGermany
9Central Institute for Medical and Chemical Laboratory DiagnosisInnsbruck University HospitalInnsbruckAustria
10NEQAS (BTLP), West Herts Hospitals NHS TrustWatfordUK, UK
11Royal College of Pathologists of Australasia Quality Assurance ProgramSt Leonards, SydneyAustralia
12South African National Blood ServiceJohannesburgSouth Africa
13Norwegian EQA Immunohaematology (Nasjonalkontrollen)Oslo University HospitalOsloNorway
14ANTITOXIN GmbHBammentalGermany
15Department of Blood Group Serology and Transfusion MedicineMedical University of GrazGrazAustria
16Department of Transfusion Medicine and Cell TherapyMedical University of ViennaWähringer Gürtel 18-201090ViennaAustria
Christoph Buchta1,*, Wim Coucke2, Cécile Toly-Ndour3, Agnès Mailloux3, Karina Hellbert4, Gianfranco Avveduto5, Isabelle Bertin-Jung6, Anna Bode7, Lobna Bouacida2, Stefanie Gemein7, Sabine Goseberg8, Andrea Griesmacher1,9, Richard Haggas10, Junho Kim11, Truscha Niekerk12, Jean-Pascal Siest6, Nanna Skeie13, Heidi Støen13, Renate Becker14, Thomas Wagner15, Günther F. Körmöczi1,16,*
1) Austrian Association for Quality Assurance and Standardization of Medical and Diagnostic Tests (ÖQUASTA), Vienna, Austria
2) Sciensano, Brussels, Belgium
3) Centre National de Référence en Hémobiologie Périnatale (CNRHP), Paris, France
4) Maybach Bechter Hellbert Rechtsanwälte GesbR, Vienna, Austria
5) CRRVEQ - SOD Sicurezza e Qualità, AOU Careggi, Firenze, Italy
6) Biologie Prospective, Villers-les-Nancy, France
7) Reference Institute for Bioanalytics (RfB), Bonn, Germany
8) INSTAND e.V. Society for Promoting Quality Assurance in Medical Laboratories, Düsseldorf, Germany
9) Central Institute for Medical and Chemical Laboratory Diagnosis, Innsbruck University Hospital, Innsbruck, Austria
10) UK NEQAS (BTLP), West Herts Hospitals NHS Trust, Watford, UK
11) Royal College of Pathologists of Australasia Quality Assurance Program, St Leonards, Sydney, Australia
12) South African National Blood Service, Johannesburg, South Africa
13) Norwegian EQA Immunohaematology (Nasjonalkontrollen), Oslo University Hospital, Oslo, Norway
14) ANTITOXIN GmbH, Bammental, Germany
15) Department of Blood Group Serology and Transfusion Medicine, Medical University of Graz, Graz, Austria
16) Department of Transfusion Medicine and Cell Therapy, Medical University of Vienna, Vienna, Austria
*Corresponding authors:
Regarding EQA: Christoph Buchta, Austrian Association for Quality Assurance and Standardization of Medical and Diagnostic Tests (ÖQUASTA), Hörlgasse 18/5, 1090 Vienna, Austria; christoph.buchta@oequasta.at
Regarding immunohaematology: Günther F. Körmöczi, Department of Transfusion Medicine and Cell Therapy, Medical University of Vienna, Währinger Gürtel 18–20, 1090 Vienna, Austria; guenther.koermoeczi@meduniwien.ac.at
Keywords:
Sensitivity
limit of detection
LoD
anti-D
red blood cell antibody screening
Abstract
A
A
A
A
Background
Red blood cell (RBC) antibodies can cause serious and even fatal haemolytic transfusion reactions or haemolytic disease of the foetus and newborn. Therefore, the application of reliable testing procedures for their detection is of utmost importance. However, only little is known about the sensitivity of common RBC antibody screening tests (AST).
Methods
Detection rates of quantified but undisclosed anti-D antibodies in low concentrations, traceable to an International Standard, were investigated in four serum samples.
A
In a concerted effort by ten external quality assessment (EQA) providers, these samples were tested by 2500 participating laboratories using RBC ASTs based on four test principles and test cells and devices from more than 24 and five manufacturers, respectively.Resulzs
Detection rates decreased with decreasing anti-D antibody concentrations. The vast majority of test systems detected samples with 0.1 IU/mL anti-D antibodies, some were still reliable at 0.01 IU/mL, and a few appeared to easily detect concentrations of 0.005 IU/mL. The sensitivity of ASTs depended on the test principle they applied. Erythrocyte magnetized technique (EMT) and solid-phase RBC adhesion technique (SPRCA) consistently demonstrated above-average detection rates, and automated result reading was superior to manual reading. Test tube technology (TTT) performed below average. Column agglutination technology (CAT) represented the average, and, surprisingly, no differences in detection rates were found between automated and manual reading of the reaction results.
Interpretation
The results indicate significant differences between RBC ASTs in routine use. The clinical relevance of these differences has yet to be assessed.
A
A
A
A
Funding
None.
Introduction
A
Individuals are sensitized by allogeneic red blood cell (RBC) antigens mainly through transfusion of mismatched blood components or foeto-maternal haemorrhage. The probability of evoking irregular RBC antibodies depends on the immunogenicity of the antigen and the functionality of the immune system of the individual [1]. A recent prospective study confirmed the data from earlier retrospective studies that 2–5% of transfused individuals are sensitized and develop detectable RBC antibodies [2, 3]. The immune response to initial sensitization is characterized by slow kinetics and low or moderate antibody concentrations. Without further antigen exposure, the concentration of alloantibodies decreases at different rates and at some point, they may no longer be detectable, or phases of detectability and non-detectability alternate cyclically [1, 3–5]. Individuals are at risk of developing a haemolytic transfusion reaction (HTR) on re-exposure to the RBC antigen against which they are sensitised. Such a secondary immune response is characterized by rapid kinetics and high peak antibody concentrations. While such pronounced immune responses upon contact with pathogens are the purpose of preventive vaccinations, they are feared in the context of blood transfusions and pregnancies. To date, no limit has been defined for antibody concentrations below which HTR is unlikely to occur. Upon foeto-maternal haemorrhage, the alloimmunization rate is reported to vary between 0.4% and 8.74% [6]. As maternal RBC IgG antibodies pass the placenta and react with antigens on foetal erythrocytes or hematopoietic precursors, they may cause haemolytic disease of the foetus and newborn (HDFN) and in the most severe cases intrauterine foetal demise [7]. Serum concentrations of maternal anti-D antibodies exceeding 3.5 or 5 IU/ml have been reported to be relevant for the development of HDFN [8, 9].Sensitization by RBC antigens is demonstrated by the detection of specific antibodies in patient serum. To detect the presence of RBC antibodies, suspensions of various test RBCs expressing clinically significant erythrocyte antigens are added to patient serum, and a positive result is indicated by agglutination of the test RBC. The probability of detecting RBC antibodies depends on the antibody concentration in relation to the sensitivity of the test system used. Each diagnostic test is characterized by its proprietary sensitivity and does not detect measurands in concentrations below the limit of detection (LOD). It is a major and potentially disastrous mistake to take the non-detection of a measurand as evidence for its non-presence [10]. In recent decades, various techniques have been developed to enhance agglutination reactions, like the use of proteins, enzymes, low ionic strength solution (LISS), or polyspecific anti-human globulin (Coombs) reagent [11]. RBC antibodies detected in serum are named after the identified RBC antigen against which they are directed, so that, for example, an antibody against antigen D is referred to as an anti-D antibody. RBC antibodies are commonly quantified by titration, and the result of this functional determination is reported as the last non-negative in a series of increasing dilutions, e.g. 1:1024.
The sparse literature on RBC antibody screening reports almost exclusively on specificity, while data on sensitivity of test systems is still largely lacking [1–5, 12]. According to European Union (EU) regulations, blood grouping test systems are in vitro diagnostic medical devices (IVD-MD) of the highest risk class [13]. The fact that their sensitivity is still unknown is highly concerning. To address the apparent lack of knowledge about the sensitivity of RBC antibody screening tests (AST), a research collaboration between several external quality assessment (EQA) providers was initiated. EQA programs are ideal tools for evaluating the performance of test systems in routine laboratory operations. They include large numbers of participant laboratories in which numerous applications of the same and different test systems, devices, reagents, and batches of reagents to analyse samples with identical known but undisclosed properties. An EQA “super-challenge” is the use of identical sample materials in approximately simultaneous challenges of several EQA providers [14]. This unique approach considerably expands the range of laboratories and IVD MDs whose performance can be evaluated.
Materials and Methods
Pilot tests showed that common RBC ASTs detected anti-D antibodies at concentrations of 1 IU/mL to 0.05 IU/mL as positive, 0.025 IU/mL as weakly positive, and at 0.01 IU/mL, first test systems presented false negative results. It was decided to use one clearly positive, two different weakly positive and one sample presumably below the detection limit of most test systems for this study. The sample panel therefore consisted of four samples with Sample 1 (S1) yielding 0.1 IU/mL, S2 0.25 IU/mL, S3 0.01 IU/mL, and S4 0.005 IU/mL anti-D.
Preparation, testing and shipment of samples
Sample materials were produced by ANTITOXIN GmbH (Bammental, Germany). Sample matrix was a defibrated donor plasma blood group O, Rh(D)+, spiked with the WHO International Standard Anti-D Immunoglobulin (NIBSC, Potters Bar, Hertfordshire, United Kingdom, NIBSC code 16/332) to achieve the target concentration of anti-D [15]. Samples were stored at + 4°C and shipped to the EQA providers under the same conditions.
A
Shipping to participant laboratories was carried out under ambient conditions.A
Anti-D antibodies in S1 to S4 were quantified at the Centre National de Référence en Hémobiologie Périnatale (CNRHP) in Paris, France, by using the anti-D microtitration test [16]. This semi-quantitative method classifies as a titration test on column gel filtration using papain-treated D positive RBC (D + C-E-c + e+ (RH:1,-2,-3,4,5 phenotype) and serial 2-fold dilutions of the samples or of a 0.12 IU/ml polyclonal anti-D standard (connected to the WHO International Standard Anti-D Immunoglobulin) [15]. Concentrations of anti-D were determined by multiplying the inverse of the last reactive dilution of the sample by the concentration of the standard with the same hemagglutination strength. Twelve aliquots of each sample (S1, S2, S3 and S4) were double measured and the mean of the 24 values per sample was calculated.The measured concentration of anti-D as compared to the concentration calculated using the dilution factor was 0.11 vs. 0.1 IU/mL for S1, 0.026 vs. 0.025 IU/mL for S2, and 0.011vs. 0.01 IU/mL for S3. In S4, anti-D was detected, but the concentration was below the limit of quantification (0.0075 IU/mL). It can be assumed that the dilution procedure to prepare S4 was as precise as that of the other samples and S4 was therefore released without verification due to the lack of an available method. For reasons of completeness of stability testing, the concentrations of the anti-D antibodies were determined for a second time after the last EQA samples were sent to the participants and the results of these determinations confirmed the stability of the measurand until the end of the study period. Vial-to-vial homogeneity in samples was determined according to applicable international standards and were found appropriate [17].
The “EQA super-challenge”
The professional network of the European Organisation of External Quality Assurance Providers in Laboratory Medicine (EQALM) was used to invite EQA providers to join the study and to use the study samples in their EQA programs in 2024. A total of ten EQA providers agreed to participate in the study. (Supplement 1). It was left to the EQA providers to decide about the order and designation of the samples, and whether all samples were sent to the participating laboratories in one or several consecutive cycles. A definition of “EQA super challenge” was provided elsewhere [18]. The EQA provider UK NEQAS (BTLP) used only S2 and S3 in their program. The participants were informed that samples were ready to use and should be analysed in the same way as patient samples.
Data collection and evaluation
EQA providers collected data in their routine way. They were asked to report the results of their participants in Excel files in a predefined format to the study database. The information requested included pseudonyms (numbers) of the participant laboratories, the methods (manual or automated), test cells and disposables (e.g. gel cards) used for determination, and for automated test systems, the name of the device and its manufacturer and the test principle applied. Results reported for RBC antibody screening were to be submitted as “positive” or “negative”. The EQALM central database was used to compile the data. Routines exist to semi-automatically transfer the data into the database. Once the data was centralized, procedures were executed to uniform the data and make them comparable between EQA providers. These procedures include definition of common measurement methods and expression of results. Uniformed data were analysed statistically.
Statistics
Only descriptive statistics were used.
Preparation of this manuscript
This study was planned, and the manuscript was written considering the EQALM guidelines for publishing about interlaboratory comparison studies (PubILC) [19].
Results
A total of 2500 laboratories were enrolled in this study via ten national and international EQA providers. Test systems with instruments from five manufacturers and test cells from > 24 manufacturers were used, eight of which were represented at least 15 times, the minimum number required for an evaluation. This vague information is because a total of nine laboratories reported their test systems as “in house preparation”.
Total detection rates of anti-D antibody in Samples 1–4
S1 was detected as positive for irregular RBC antibodies by 1867/1955 (95.5%) of participants; S2 by 2184/2460 (88.8%), S3 by 841/2374 (35.4%), and S4 by 130/1931 (6.7%). Detection rates of automated vs. manual methods were 507/509 (99.6%) vs. 277/353 (78.4%) for S1, 922/961 (95.9%) vs. 224/413 (54.2%) for S2, 429/947 (45.3%) vs. 77/386 (19.9%) for S3, and 44/507 (8.7%) vs. 5/348 (1.4%) for S4. (Fig. 1, Supplement 3)
A
Fig. 1
Total detection rates of anti-D antibody in Samples 1–4
Legend
The difference between “all“ and the sum of “automated” and “manual” reading of results is explained by a high number of results where the way they were read was not reported.
Anti-D antibody detection rates of different test principles and test systems
A
Test systems used by participants were based on different methodological test principles, namely test tube technology (TTT), column agglutination technology (CAT), solid phase RBC adherence (SPRCA) technology, and erythrocyte-magnetized technique (EMT) [20]. While EMT and SPRCA were used exclusively by Diagast (Loos, France) and Immucor (Norcross, GA, USA) systems respectively, CAT was used by Bio-Rad (Hercules, CA, USA), Grifols (Barcelona, Spain) and Ortho Clinical Diagnostics (OCD, QuidelOrtho, Athens, OH, USA) test systems, with Bio-Rad and Grifols using a dextran-acrylamide matrix and OCD using a glass beads matrix. Detection rates of EMT were 38/39 (97.4%) for S1, 39/40 (97.5%) for S2, 37/39 (94.9%) for S3, and 38/39 (97.4%) for S4; detection rates of automated vs. manual reading were 22/23 (95.7%) vs. 19/19 (100%) for S1, 24/24 (100%) vs. 2/2 (100%) for S2, 22/23 (95.7%) vs. 2/2 (100%) for S3, and 23/23 (100%) vs 2/2 (100%) for S4. Detection rates of SPRCA were 75/75 (100%) for S1, 113/115 (98.3%) for S2, 99/110 (90.0%) for S3, and 30/66 (45.5%) for S4, and for automated vs. manual reading 14/14 (100%) vs. 3/3 (100%) (S1), 49/51 (96.1%) vs. 1/3 (33.3%) (S2), 46/47 (97.9%) vs. 0/3 (0%) (S3), and 9/14 (64.3%) vs. 0/2 (0%) (S4). Detection rates of CAT systems were 1527/1539 (99.2%) for S1, 1850/1935 (95.6%) for S2, 640/1886 (33.9%) for S3, and 60/1531 (3.9%) for S4, and for automated vs. manual reading 442/443 (99.8%) vs. 141/144 (97.9%) (S1), 799/829 (96.4%) vs. 154/164 (93.9%) (S2), 322/807 (39.9%) vs. 61/160 (38.1%) (S3), and 12/441 (2.7%) vs. 2/143 (1.4%) (S4). Among test systems using CAT, detection rates of automated vs. manual reading were for Bio-Rad 274/274 (100%) vs. 112/112 (100%) (S1), 398/422 (94.3%) vs. 119/123 (96.7%) (S2), 182/414 (44.0%) vs. 46/115 (40.0%) (S3), and 9/274 (3.3%) vs. 2/112 (1.8%) (S4). Grifols test systems had detection rates of 40/40 (100%) vs. 16/16 (100%) (S1), 151/152 (99.3%) vs. 17/19 (89.5%) (S2), 64/142 (45.1%) vs. 5/21 (23.8%) (S3), and 1/38 (2.6%) vs. 0/15 (0%) (S4). OCD test systems had detection rates of 128/129 (99.2%) vs. 13/16 (81.3%) (S1), 250/255 (98.0%) vs. 18/22 (81.8%) (S2), 76/251 (30.3%) vs. 10/24 (41.7%) (S3), and 2/129 (1.6%) vs. 0/16 (0%) (S4). Detection rates of TTT were 107/159 (67.3%) for S1, 35/181 (19.3%) for S2, 1/156 (0.6%) for S3, and 0/151 (0%) for S4. (Fig. 2, Supplement 4)A
Fig. 2
Anti-D antibody detection rates of different test principles, test systems, and ways of reading reaction results
Legend
Differences between the result counts of all CAT systems vs. the sums of automated and manual read results is caused by a high number of results for which the way of reading was not reported. Results from groups of 10 or fewer are displayed transparently.
Anti-D antibody detection rates of automated and manual read test systems with different EQA providers
Automated and manual reading outcome could not be compared for EMT and SPRCA test systems due to low numbers of manually read results. Median detection rate of automatedly vs. manually read Bio-Rad test systems was 100% for both methods for S1; 97.1% (median; range 86.3%-100%) vs. 100% (90.9%-100%) for S2; 55.6% (0%-85.4%) vs. 31.6% (0%-57.1%) for S3, and 0% (0%-23.7%) vs. 0% (0%-3.0%) for S4. Median detection rate of automatedly vs. manually read Grifols test systems was 100% for both methods for S1, 100% (87.5%-100%) vs. 100% (85.7%-100%) for S2, 3.7% (0%-56.1%) vs. 15.4% (0%-75.0%) for S3, and 0% (0%-3.7%) vs. 0% for S4. Median detection rate of automatedly vs. manually read OCD test systems was 100% (median; range 98.2%-100%) vs. 100% (0%-100%) for S1, 100% (77.8%-100%) vs. 100% (0%-100%) for S2, 28.5% (12.7%-51.7%) vs. 20.0% (0%-75.0%) for S3, and 0% (0%-3.6%) vs. 0% for S4. TTT reaction results were only read manually. (Fig. 3, Supplement 5)
A
Fig. 3
Anti-D antibody detection rates of test systems with different EQA providers
Legend
The number of results per EQA provider is given as the average value of the participants in the individual rounds. Only results accompanied by information on the methodology used were included; results missing this information are not shown in this Figure. Results from groups of 10 or fewer are displayed transparently.
Consideration of the legal position
Reagents and test cells for immunohaematology tests have been classified as the highest risk class of IVD-MDs in European legislation for decades. Both the currently applicable IVDR and, previously, the Directive on In Vitro Diagnostic Medical Devices (IVDD), classify reagents and reagent products for determining ABO, Rh and Kell blood group systems (Kidd and Duffy are also mentioned in the IVDR) to the highest risk class, formerly “List A” and currently Class D [13, 21]. The IVDR uses the terms “marker”, “biomarker” and “target marker”, but does not define the term “marker” itself. According to constant case law as rendered by the Court of the European Union (CJEU) “in interpreting a provision of EU law, it is necessary to consider not only its wording, but also the context in which it occurs and the objectives pursued by the rules of which it is part” [22]. A “marker” can be defined as “a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention” [23]. Taking into account the aim of the IVDR (strengthening patient safety and the performance of the IVDs), the term “marker” can therefore be interpreted as including the detection of blood group antigens and antibodies against them.
Discussion
To our knowledge, this is the first study on the sensitivity of RBC-ASTs used in routine laboratory practice, conducted through an interlaboratory comparison using samples with quantified concentrations of anti-D antibodies traceable to an international standard. The sensitivity of the test systems varied according to their analytical principle, with EMT- and SPRCA-based test system consistently presenting above-average detection rates for all samples. The former even gives the impression that its LOD is still far from being met by the 0.005 IU/mL samples. No differences were found between the detection rates of the CAT systems using dextran acrylamide matrix (Bio-Rad and Grifols) or glass bead matrix (OCD), nor between manual and automatic reading of the reaction results. TTT appears to lag behind other test principles in terms of sensitivity, to such an extent that a comparison of automated and manual evaluation of reaction results without considering the underlying test principles leads to manually read results from CAT systems appearing less sensitive than they actually are. The same patterns of detection rates achieved by the same test systems at different EQA providers underscore their robust performance. A direct comparison of detection rated of automated and manual methods is not acceptable, because the effect of “automated” vs. “manual” reading is much smaller than the effect of the determination method used.
All test systems appeared to be capable of detecting anti-D antibodies at concentrations considered relevant for the development of HDFN (3.5-5 IU/mL). However, this threshold was only established for immune responses during pregnancy. For situations in which antigen re-exposure may lead to a secondary immune response in already sensitised individuals, the thresholds below which booster-induced HTR is unlikely, are not yet known, if they exist at all. Until then, one approach to reducing the risk of missing previous sensitization due to evanescence of RBC antibodies could be the implementation of registers for individuals who have ever been diagnosed with RBC antibodies, as has been proposed [24]. To make such a register complete, ASTs would have to be performed on all individuals after transfusions or pregnancies. If these options are not available, ASTs with the highest possible sensitivity are the option of choice for risk reduction.
Sensitivity is a characteristic of diagnostic tests that must be declared by manufacturers under the IVDR, and the International Standard ISO 15189:2022 requires laboratories to have this information available [25]. The fact that blood group testing is classified in the highest risk categories of IVD-MDs makes it all the more incomprehensible that AST has never gained the dubious reputation it actually deserves due to a lack of knowledge about this essential performance criterion. With the full entry into force of the IVDR, the CE-certification of certain CE-IVDs (Class B or higher) is based on the performance assessment by notified bodies as competent and independent third parties, and the evaluation of Class D products also requires the involvement of a European Union Reference Laboratory (EURL) [26]. It is expected that more importance will be attached to the sensitivity of immunohaematological tests in the future, especially once individual batches of reagents need to be assessed and approved by them. For the time being, it will be helpful to raise awareness among clinicians that negative screening results should not be taken lightly as evidence that the patient has not been sensitized before. To this end, results should be reported as “(irregular RBC antibodies) not detected” instead of “negative”, “none” or “-”.
The lack of specifications for the LOD of ASTs may also be reflected in reports of “disappearing” and “returning” antibodies. Further studies are needed to investigate whether this apparent on-off phenomenon actually exists or whether it is merely due to differences in the sensitivity of the ASTs used in the patient's medical history.
A
Beyond that, the type and screen practices in transfusion medicine require clear guidelines on the LOD of AST. In this concept, the decision on whether to perform a serological crossmatch prior to transfusion is based on the results of the AST. Since the resource requirements for a non-selective, complete serological cross-matching prior to transfusion are under discussion, studies are needed on the booster potential of previous sensitisation that can no longer be identified by existing and detectable RBC antibodies [27]. These must be followed by studies on the required detection limit for ASTs in order to justify the type and screen concept as a resource-saving method without additional risks for patients. However, given the long-standing discussions about the existence and possible definition of correlates of protection (CoP) in the closely related field of vaccinology, the path to establishing a threshold value for the improbability of a booster response in individuals after re-exposure to RBC antigens still seems to be a long way off [28]. To top it all off, it is not unusual that polytransfused individuals or pregnant women develop several different RBC antibodies simultaneously or in succession [29, 30]. The possible significance of this fact also needs to be investigated.Immunohaematology is a laboratory discipline that is not exactly known for its innovative spirit. Although some of its laboratory methods have been refined since their introduction at the beginning of the 20th century, it took about fifty years to recognise that immunohematology must deal not only with qualitative but also with quantitative properties of patient samples, and another fifty years before the first attempts were made to actually measure the concentration of RBC antibodies. However, the potency of irregular RBC antibodies is currently still generally expressed as a dimensionless functional measure, i.e. as a titre value, rather than as a concentration, e.g. IU/mL. The correlation between anti-D titres and anti-D concentrations reported by Wikman et al. shows that the threshold value of 5 IU/mL established for the prevention of HDFN corresponds to titres in a range of 128 to 1024, and, conversely, a titre of 128 corresponds to concentrations between ∼1 and ∼11 IU/mL [31]. Such inaccuracy is just as risky as the reporting of titres is outdated. Examples from both medical laboratory diagnostics and pharmaceutical analysis show that quantitative methods can nowadays be routine [31, 32].
As with all reports on EQA data, a limitation of our study is that results could only be included as they were reported by participants and passed on by their EQA providers. It must be trusted that they were generated properly and with the reported test system. Further limitations include the imbalance in the number of results obtained with individual test principles and the frequent lack of information on the use of automated or manual test systems. The effects of different batches of reagents and test cells, as well as any agglutination enhancers used, were not investigated.
Conclusion
This study shows a clear difference in the analytical sensitivity regarding detection of anti-D antibodies of different test principles used in RBC ASTs. The clinical significance of these findings remains to be assessed but is deemed to be high due to their impact on patient welfare and safety. Furthermore, the sensitivity of ASTs to RBC antibodies other than anti-D and also in the case of multiple antibodies being present simultaneously still needs to be evaluated.
Autor contributions
CB conceptualized this study. CB and WC analysed the data and verified its reproduction in the manuscript. CTN, AM provided methodology and resources for laboratory determinations. CB, GA, IBJ, AB, LB, SG, SG, RH, JK, TN, JPS, NS, and HS acquired data from their EQA programs. CB and WC visualized data. CB, CTN, AM, KH, RB, and GFK wrote the original draft of the manuscript. All authors (CB, WC, CTN, AM, KH, GA, IBJ, AB, LB, SG, SG, AG, RH, JK, TN, JPS, NS, HS, RB, TW, and GFK) had full access to all data in the study, critically reviewed and edited the manuscript, and accept responsibility for publication.
References
1.
Evers D, Middelburg RA, de Haas M et al (2016) Red-blood-cell alloimmunisation in relation to antigens' exposure and their immunogenicity: a cohort study. Lancet Haematol 3:e284–e292
2.
Yamada C, Ono T, Ino K et al (2024) The development and evanescence of red blood cell antibodies after transfusion: A multi-institutional prospective study in Japan. Transfusion 64:1980–1992
3.
Tormey CA, Hendrickson JE (2019) Transfusion-related red blood cell alloantibodies: induction and consequences. Blood 133:1821–1830
4.
Körmöczi GF, Mayr WR (2014) Responder individuality in red blood cell alloimmunization. Transfus Med Hemother 41:446–451
5.
Hauser RG, Esserman D, Karafin MS et al (2020) (for the NHLBI Recipient Epidemiology and Donor Evaluation Study-III (REDS-III)). The evanescence and persistence of RBC alloantibodies in blood donors. Transfusion 60:831–839
6.
Adani SN, Mohd Ashari NS, Johan MF, Edinur HA, Mohd Noor NH, Hassan MN (2024) Red Blood Cell Alloimmunization in Pregnancy: A Review of the Pathophysiology, Prevalence, and Risk Factors. Cureus 16:e60158
7.
van 't Oever RM, Zwiers C, de Winter D et al (2022) Identification and management of fetal anemia due to hemolytic disease. Expert Rev Hematol 15:987–998
8.
Ghesquière L, Leroy J, Deken V et al (2023) Anti-RH1 alloimmunization: At what maternal antibody threshold is there a risk of severe fetal. anemia? Transfus 63:629–637
9.
Toly-Ndour C, Mourtada H, Huguet-Jacquot S et al (2018) Clinical input of anti-D quantitation by continuous-flow analysis on autoanalyzer in the management of high-titer anti-D maternal alloimmunization. Transfusion 58:294–305
10.
Buchta C, Zeichhardt H, Osterman A, Perrone LA, Griesmacher A (2024) Do not blindly trust negative diagnostic test results! Lancet Microbe 5:e102–e3
11.
Armstrong B (2020) Antigen-antibody reactions. Vox Sang 15:68–80
12.
Guglielmino J, Morris FJ, Grattidge CM, Jackson DE (2024) Evaluation of the analytical performance of four different manufacturer's reagent red blood cells in antibody detection and identification. BMC Res Notes 17:293
13.
Regulation (EU) 2017/746 of the European Parliament and of the Council of 5 April 2017 on in vitro diagnostic medical devices and repealing Directive 98/79/EC and Commission Decision 2010/227/EU (Text with EEA relevance). Available at http://data.europa.eu/eli/reg/2017/746/oj
14.
Buchta C, De la Salle B, Marrington R et al (2025) Behind the scenes of EQA - Characteristics, Capabilities, Benefits and Assets of External Quality Assessment (EQA) Part V - Benefits for Stakeholders Other than Participants. Clin Chem Lab Med 63:898–915
15.
Available at https://nibsc.org/documents/ifu/16-332.pdf, accessed on 08 June 2025
16.
Dupont M, Gouvitsos J, Dettori I, Chiaroni J, Ferrera V (2007) Intérêt de la technique de microtitrage des anticorps anti-RH1 dans le suivi immunohématologique des femmes enceintes [Microtitration of anti-RH1 antibodies: interest in the follow-up of pregnant women]. Transfus Clin Biol 14:381–385
17.
Buchta C, Marrington R, De la Salle B et al (2025) Behind the scenes of EQA - characteristics, capabilities, benefits and assets of External Quality Assessment (EQA), Part III - EQA samples. Clin Chem Lab Med 63:868–878
18.
Buchta C, Aberle SW, Albarède S, Albe X, Badrick T, Bietenbeck A et al (2025) The concept of external quality assessment (EQA) super challenges with special consideration of their importance during pandemics. Lancet Microbe (in press).
19.
Buchta C, Gidske G, Henriksen GM, Badrick T (2024) The European Organisation of External Quality Assurance Providers in Laboratory Medicine (EQALM) Statement: Guidelines for publishing about interlaboratory comparison studies (PubILC). Crit Rev Clin Lab Sci 61:588–598
20.
Li HY, Guo K (2022) Blood group testing. Front Med 9:827619
21.
European Parliament and Council Directive 98/79/EC of the European Parliament and of the Council of 27 October 1998 on in vitro diagnostic medical devices. Available at http://data.europa.eu/eli/dir/1998/79/oj
22.
Judgment of the Court (Fifth Chamber) of 14 March 2024 Case C-46/23, ECLI identifier: ECLI:EU:C:2024:239. Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:62023CJ0046
23.
Biomarkers Definitions Working Group (2001) Biomarkers and surrogate endpoints: preferred definitions and conceptual framework. Clin Pharmacol Ther 69:89–95
24.
Mathur G, Wilkinson MB, Island ER, Menitove JE, Tilzer L (2022) A case for a national registry of red blood cell antibodies. Vox Sang 117:738–740
25.
ISO 15189:2022 (2022) Medical laboratories - requirements for quality and competence. International Organization for Standardization (ISO), Geneva
26.
Buchta C, Benka B, Delatour V et al (2024) Reference, calibration and referral laboratories - a look at current European provisions and beyond. Clin Chem Lab Med 63:656–669
27.
Arthi R, Soundharya V, Suresh Kumar I, Hari Haran A, Sahayaraj J (2024) Impact of Type and Screen Method on Turnaround Time and Man-Hour Utilization Compared to Conventional Coomb's Crossmatch: A Cross-Sectional Analytical Study. Cureus 16:e69564
28.
King DF, Groves H, Weller C et al (2024) Realising the potential of correlates of protection for vaccine development, licensure and use: short summary. NPJ Vaccines 9:82
29.
Cruz RO, Mota MA, Conti F, Pereira RA, Kutner JM, Aravechia MG, Castilho L (2011) Prevalence of erythrocyte alloimmunization in polytransfused patients. Einstein (Sao Paulo) 9:173–178
30.
Singh B, Chaudhary R, Katharia R (2021) Effect of multiple maternal red cell alloantibodies on the occurrence and severity of Hemolytic Disease of the Fetus and Newborn. Transfus Apher Sci 60:102958
31.
Wikman A, Jalkesten E, Ajne G, Höglund P, Mörtberg A, Tiblad E (2018) Anti-D quantification in relation to anti-D titre, middle cerebral artery Doppler measurement and clinical outcome in RhD-immunized pregnancies. Vox Sang 113:779–786
32.
Oviedo SA (2020) Flow Cytometry Assay for Quantitation of Therapeutical Anti-D IgG during Process Control in the Pharmaceutical Production. Innovations in Cell Research and Therapy. IntechOpen, London (UK)