Challenges of anti–human leukocyte antigen antibody measurement in blood donors

Introduction

Human leukocyte antigen system

The human leukocyte antigen (HLA) system is a complex system of genes encoding proteins that play an important role in immune system regulation. The diversity of HLA genes is crucial for the effectiveness of the immune response. Each individual has a unique HLA profile, which contributes to the specificity of immune responses and is a key factor in organ transplantation, as compatibility in HLA types is essential to minimize the risk of graft rejection.¹

HLA antigens have a strong sensitization potential. Alloimmunization can occur as a result of a transplant, but also after blood transfusions and pregnancies.^2,3

Anti-HLA antibodies can also be detected in individuals who have not been exposed to the abovementioned immunization / sensitization factors. The clinical significance of these “naturally” occurring HLA antibodies in transfusion or transplantation medicine is not clear.^4,5

Anti–human leukocyte antigen antibodies in blood components

Presence of anti-HLA antibodies in blood components increases the risk of transfusion-related acute lung injury (TRALI) in blood recipients.² TRALI is a rare complication, but is characterized by a high mortality rate. It manifests as noncardiogenic pulmonary edema with sudden dyspnea, hypoxia, and respiratory failure.⁶

The pathogenesis of TRALI is complex and not fully characterized. Antibody-dependent mechanisms are supposed to be responsible for 50%–80% of TRALI cases. In addition to anti-HLA antibodies, TRALI can also be triggered by anti–human neutrophil antigen antibodies.^7,8

The incidence of TRALI is reported within a wide range, varying from 0.04% to 8% per transfused patient or—when reported per blood component—from approximately 1 in 500 to 1 in 100 000 components.^9-12

Passive transfer of anti-HLA antibodies through blood components may also contribute to graft rejection in transplant patients.^13,14

Despite the risk of post-transfusion complications, anti-HLA antibodies are not routinely screened in blood donors. Such tests are conducted in a specific donor when investigating the cause of a post-transfusion reaction in a patient who received blood collected from that particular donor. To decrease the risk of TRALI, dedicated risk-reduction strategies have been applied, such as inclusion of male-only plasma and platelet apheresis donors or exclusion of all-exposure donors.⁷

Some blood donation centers are implementing strategies that involve screening platelet apheresis donors for the presence of antileukocyte antibodies. However, there is no industry standard regarding the choice of an assay platform for such a screening. Additionally, the cutoff for a positive test interpretation for healthy blood donors is not defined.

Anti–human leukocyte antigen antibodies in blood recipients

Presence of anti-HLA antibodies in blood recipients may be associated with the occurrence of nonhemolytic febrile reactions following transfusion of blood components containing leukocytes, platelet transfusion refractoriness, TRALI, and transplant rejection.² Due to the latter complication, recipients are tested for the presence of anti-HLA antibodies prior to transplantation. Determination of the antibody status in potential organ recipients is essential, as the presence of antibodies in the recipient that are specific to HLA incompatibilities present in the donor can cause graft rejection.¹⁵

Convalescent plasma

The necessity for anti-HLA antibody detection in blood donors arose in connection with the COVID-19 pandemic. As the pandemic developed, there emerged an idea to treat COVID-19 patients with plasma obtained from individuals who had recovered from the disease and developed anti–SARS-CoV-2 antibodies (so-called convalescents). Convalescent plasma therapy has shown efficacy in the past for other infections, for example, the Spanish flu, measles, and influenza viruses.^16,17 As convalescent plasma was intended for transfusion to patients with respiratory system failure, it was necessary to either collect it exclusively from donors with no history of transfusion / pregnancy or to ensure that the transfused plasma did not contain anti-HLA antibodies that could potentially trigger TRALI.

In this study, we analyzed test results obtained as part of a routine procedure adopted at blood donation centers in Poland for the qualification of convalescent plasma donors. The aim was to assess the prevalence of anti-HLA antibodies among blood donors with a positive medical history of pregnancy and / or transfusion of blood or its components. We also evaluated the utility of the LABScreen Mixed immunofluorescence test (One Lambda, Inc., Canoga, California, United States) on the Luminex platform (Luminex 100/200TM; Luminex Corp., Austin, Texas, United States) for detecting anti-HLA antibodies in blood donors.

Methods

In this study, we analyzed the results of anti-HLA antibody testing in donors who were registered at the Blood Donation Center in Kraków (between August 31, 2020 and September 27, 2021) to donate convalescent plasma and had a positive medical history of pregnancy and / or transfusion (risk group; 617 donors). As a general rule, the donors were in good health, with no history of recent infection (within the last 2 weeks) or therapy, including antibiotics. Individuals with autoimmune diseases were not eligible to donate convalescent plasma. Blood transfusion was defined as a medical procedure during which blood or blood components (such as red blood cells, plasma, or platelets) were transferred from the bloodstream of one person (the donor) into that of another person (the recipient).

Additionally, we analyzed the results of 3 donors tested for the presence of anti-HLA antibodies due to reported adverse post-transfusion reactions in the recipients (patients) who had received their blood components (1 case of TRALI, 1 case of transfusion-associated circulatory overload, and 1 adverse post-transfusion reaction presenting as skin lesions and shortness of breath). All of these donors were men with no medical history of transfusion.

Serum samples were screened for anti-HLA antibodies using the LABScreen Mixed Class I & II immunofluorescent test on the Luminex platform. The measurements were classified based on a normalized background (NBG) ratio cutoff of 2.2, as recommended by the manufacturer for transplant patients.

Statistical analysis

Categorical variables were expressed as numbers and percentages and compared using the χ² test. P values below 0.05 were considered significant. All analyses were performed using STATISTICA, version 13.3 (TIBCO Software Inc., Palo Alto, California, United States).

Ethics

This study was performed in accordance with the Declaration of Helsinki and approved by the Research Ethics Committee of Jagiellonian University Medical College (118.0043.1.423.2024), Kraków, Poland. As it was a retrospective analysis of test results obtained as part of standard procedures during blood donor qualification, informed consent statement was not required.

Results

A total of 620 tests were performed, of which 617 were performed in high-risk donors. Detailed breakdown of subgroups according donor types is presented in Figure 1. Based on the NBG ratio cutoff value of 2.2, anti-HLA antibodies were detected in 66.13% (408 of 617) of the tested donors in the risk group.

The percentage of anti-HLA–positive and anti-HLA–negative serum samples among donors with different blood groups is illustrated in Figure 2. We found no association between the presence of anti-HLA antibodies and the donor blood group (P = 0.98).

Data on medical history of pregnancy and / or transfusion, as well as the test results for the presence of anti-HLA antibodies are presented in Table 1.

Table 1. Human leukocyte antigen antibody screening test results in donors by pregnancy and transfusion history

Variable	Positive		Negative
	Men	Women	Men	Women
Pregnancy	0	377	0	190
Transfusion	7	5	7	3
Pregnancy + transfusion	0	19	0	9

The prevalence of anti-HLA antibodies (expressed as percentage; Figure 3) was similar in all 3 subgroups of women—transfused women, parous women, and transfused and parous women. There were no differences between the groups (P = 0.96), which indicated that both pregnancy and transfusion history were equally immunogenic. We also found no significant difference in the anti-HLA antibody prevalence between the analyzed groups of transfused men (n = 14) and transfused women (n = 8).

Among the 595 female donors with a positive pregnancy history, data on the number of pregnancies were available in 472 cases. Based on this information, these women were classified into 4 groups: 1, 2, 3, and 4 or more pregnancies (Figure 4). We did not observe an association between the number of pregnancies and the presence of anti-HLA antibodies (P = 0.54).

Discussion

Presence of anti-HLA antibodies in blood components may cause transfusion complications in blood recipients. Testing for these antibodies is not routinely performed in blood donors. Some blood donation centers do not allow individuals exposed to immunization with HLA antigens in connection with pregnancy or transfusion to become donors of plasma and blood components collected by apheresis.

The need for anti-HLA antibody screening in blood donors emerged in relation to the collection of plasma from convalescents during the COVID-19 pandemic. The specific eligibility criteria for donating convalescent plasma were, among others, negative results of screening tests for antibodies against HLA class I, class II, and human neutrophil antigens, which were performed in donors with a history of pregnancy or transfusion (risk group). The anti-HLA antibody testing was conducted using a sensitive immunofluorescence test.

Over the years, several methods have been developed to detect and assess the specificity of anti-HLA antibodies. The classic method is the microcytotoxicity test, which detects complement-dependent anti-HLA antibodies. Recently, new techniques with improved sensitivity have been developed. These methods involve HLA antibody screening using solid-phase immunoassays, either by employing HLA molecules bound to microtiter plates in an enzyme-linked immunosorbent assay (ELISA) or, more commonly, using polystyrene beads on the Luminex platform. Another specific method for detecting HLA-specific complement-fixing antibodies is a flow cytometry–based approach. Over the past decade, the bead-based assay utilizing the Luminex fluorocytometer has become the gold standard for HLA antibody testing.^5,15,18

In the literature, there are discrepancies regarding both the main cause of alloimmunization with HLA antigens and the prevalence of anti-HLA antibodies in blood donors with a history of transfusion and / or pregnancy. The prevalence of anti-HLA antibodies in our risk group was 66.13%, according to the NBG ratio cutoff of 2.2. Statistically, both scenarios (pregnancy and transfusion) were shown to have a similar immunogenic potential.

The available data on the immunization rate after pregnancy are inconsistent, ranging from 4.5% to as high as 72.5%.^19-25 Many studies have also shown that the prevalence of anti-HLA antibodies is related to the number of previous pregnancies.^22,24,26,27

The percentage of anti-HLA–positive results obtained in our group of parous women was 66.55%, which is relatively high. Although we did not observe a significant relationship between the number of pregnancies and the presence of anti-HLA antibodies, a higher percentage of antibodies in the groups with 4 or more pregnancies may suggest a tendency toward an increased risk of immunization with a higher number of pregnancies.

Regarding transfusion as an immunizing factor, some reports showed that transfusions posed a low risk of immunization (1.3%–1.7%).^19,21,26 In other studies, the frequency of immunization after transfusion was 12%²² or even as high as 62.5%.²⁰ In our report, the prevalence of anti-HLA antibodies was 50% in the transfused men and 62.5% in the transfused women. These values are significantly higher than those reported elsewhere; however, in another Polish study, anti-HLA antibodies were present in 60% of women with a history of pregnancy and in 62.5% of blood donors with a history of transfusion (using the Luminex method with a cutoff of 2.2—the same as in our study).²⁰

Of note, in other studies using the Luminex method, the incidence of anti-HLA antibodies in the control group (no medical history of transfusion or pregnancy) was surprisingly high: Morales-Buenrostro et al⁴ reported an anti-HLA antibody rate of 63% in men without a history of transfusion, whereas Priyadarsini et al,²⁷ who investigated the prevalence of anti-HLA antibodies in Indian blood donors, reported positive results in 23.9% of donors from the control group. In our study, among the 3 male donors without a history of transfusion, 2 obtained a positive test result.

As previously mentioned, anti-HLA antibodies can be detected in individuals who have not been exposed to immunization factors (transfusion, pregnancy, or transplant).^4,5 The presence of anti-HLA antibodies has been reported in 1%–2% of unexposed individuals.²⁴ Some studies suggest even higher percentages, ranging from 34%²⁰ up to 63%,⁴ depending on the method used for antibody detection and the chosen cutoff. These cases have been attributed to nonspecific cross-reactions during certain infections (eg, with HIV or hepatitis C virus),^28,29 or other stimuli, such as ingested proteins and allergens.⁴ Technical aspects of the testing method used could also be responsible for such results. Denatured, recombinant HLA molecules used in the fluorescence bead method can express cryptic epitopes. These epitopes are detected by antibodies with a low clinical relevance or naturally occurring antibodies.^15,30

Generally, there are few publications regarding the frequency of immunization with HLA antigens in healthy blood donors, and the results obtained in various studies differ significantly. They are undoubtedly related to the method used for testing and the cutoff point adopted for a given method.

Many research centers and diagnostic laboratories worldwide conduct anti-HLA antibody testing. New methods are emerging, and guidelines regarding the criteria for defining the cutoff for a positive result are usually provided by the manufacturer. However, there is no cutoff defined for healthy blood donors—the cutoff value has only been established for transplant recipient groups.

The impact of using different laboratory assays to detect anti-HLA antibodies in female blood donors was analyzed by Lopes et al.³¹ First, they collected serum samples from 300 previously pregnant female blood donors and tested them for the presence of anti-HLA antibodies using a solid-phase mixed-antigen assay (ELISA). Anti-HLA class I and / or class II antibodies were detected by ELISA in 26.7% of all women and in 37% of those with a history of 3 or more pregnancies. Then, samples from 60 women with a history of 3 or more pregnancies and a negative ELISA result were further tested using 3 different microbead flow assays (One Lambda LABScreen mixed, panel-reactive antibodies [PRA], and single-antigen assay). When a more sensitive testing method was used, HLA antibodies were detected in 8.3%, 18.3%, and 21.7% of the ELISA-negative women (using the LABScreen mixed, PRA or single-antigen methods, respectively).

The issue of comparing HLA antibody results obtained using different methods was also addressed by Carrick et al.³² They analyzed 525 samples using 5 different HLA antibody assays: ELISA tests from 2 different manufacturers, multi-analyte bead-based assays on the Luminex platform (2 different test versions), and a flow cytometry assay. In each method, they used cutoffs suggested by the respective manufacturers (Table 2).

Table 2. Methods of anti–human leukocyte antigen antibody testing and the adopted cutoffs for positive results, based on the study by Carrick et al³²

Method	Cutoff
GTI Diagnostics QuickStep (ELISA)	OD value greater than or equal to the cutoff, where the cutoff was calculated as 2 × the mean of the OD values of the negative controls
One Lambda LATM20 (ELISA)	Greater than the calculated 20% cutoff (calculated according to the product insert)
One Lambda LABScreen Mixed (Luminex)	NBG >2.2
Tepnel Lifescreen (Luminex)	Calculation of 3 adjusted MFI ratios for each bead as described in the product insert, followed by determination of whether at least one of the class I or class II HLA beads was positive
3Ti Aegis (flow cytometry)	Percentage of beads above the calculated cutoff (2 × negative control)
Abbreviations: ELISA, enzyme-linked immunosorbent assay; HLA, human leukocyte antigen; MFI, mean fluorescence intensity; NBG, normalized background; OD, optical density

As the quantitative results from 5 different assays on the same group of samples differed significantly (QuickStep [GTI Diagnostics, Waukesha, Wisconsin, United States; ELISA], 19.1%; LATM20 [One Lambda; ELISA], 19.5%; 3Ti Aegis [3Ti, Atlanta, Georgia, United States; flow cytometry], 29.3%; Lifescreen [Tepnel Lifecodes, Stamford, Connecticut, United States; Luminex] 37.9%; LABScreen Mixed [One Lambda; Luminex], 52.5%), the authors employed a latent variable analysis to establish a new set of cutoffs (consensus cutoffs). They established consensus cutoffs at the 99th percentile of the never-pregnant donors. The consensus cutoffs resulted in comparable sensitivity across various test platforms (QuickStep [ELISA], 24.6%; LATM20 [ELISA], 25.7%; 3Ti Aegis (flow cytometry), 25.3%; Lifescreen [Luminex], 22.6%; LABScreen Mixed [Luminex], 21.8%), consequently enhancing the agreement and consistency among the assays. The method proposed by the authors for standardizing the results yielded comparable values and represents a potential solution to the problem of different assays and different cutoffs; however, it still does not address the issue of clinical significance of the obtained results.

Limitations

This study is a retrospective analysis of the anti-HLA screening results of individuals who volunteered to donate convalescent plasma; consequently, the obvious limitation is a small number of donors after transfusions, as well as a lack of a control group. The obtained results prompt further planned research involving a control group.

Conclusions

The prevalence of anti-HLA antibodies in individuals exposed to immunization with HLA antigens, determined with the use of a sensitive diagnostic test, was high and reached 66.13%.

The anti-HLA antibody prevalence in blood donors is clearly dependent on the testing methodology used to detect the antibodies, as well as the assay cutoff chosen for the particular method. If a very sensitive test is used for screening, the percentage of anti-HLA–positive donors may be high. Consequently, this would significantly reduce the number of donors, which, in the context of blood shortages and an aging society, may be a problem.

Also, considering that positive test results may occur in blood donors without a history of transfusion or pregnancy, further extensive research on the clinical significance of anti-HLA antibodies present in blood components is required. The obtained results highlight the importance of establishing cutoff points for the applied research methods in relation to the clinical data related to post-transfusion complications.