Secondary Logo

Journal Logo

DSA IN ORGAN TRANSPLANTATION: CONUNDRUMS AND SOLUTIONS: Edited by Didier A. Mandelbrot

Measuring human leukocyte antigen alloantibodies: beyond a binary decision

Maguire, Chelsea H.a; Schinstock, Carrie A.b; Tambur, Anat R.a

Author Information
Current Opinion in Organ Transplantation: December 2020 - Volume 25 - Issue 6 - p 529-535
doi: 10.1097/MOT.0000000000000822
  • Open

Abstract

INTRODUCTION

The humoral theory

Historically, the damage caused by human leukocyte antigen (HLA) antibodies leading to hyperacute allograft rejection [1,2] mandated the performance of donor lymphocyte and patient serum crossmatch testing prior to transplantation. A binary answer to the question of whether or not the patient has HLA antibodies against the prospective donor was provided; leading to a binary decision of whether to proceed with transplantation (yes/no). Percentage panel reactive antibody (%PRA) and HLA antibody specificity testing followed, to provide proactive information on patients’ transplantability. The notion was that once hyper-acute/accelerated rejection were avoided, posttransplant immune responses would be driven mostly by T cells. In fact, the vast majority of immunosuppressive medications at the time were directed towards cellular immunity (T cells).

The ‘cellular theory’ was boldly challenged by Terasaki [3] when he elegantly demonstrated the significance of HLA antibodies in graft loss before and after transplantation. The ‘Humoral Theory of Transplantation[4] summarizes evidence to link HLA antibodies with both acute early rejection and chronic rejection. HLA antibodies were detected among many patients with graft loss. Often the detection of HLA antibodies preceded rejection, which suggested antibodies were the potential cause of rejection. Terasaki [4] argued that an important implication of adopting the Humoral Theory was that treatment to remove those antibodies might lead to improved allograft survival. That article marked a pivotal change in how HLA antibodies were viewed and monitored. Approaches to remove HLA antibodies or suppress the cells producing them were sought, leading to the use of plasmapheresis/IVIg, anti-CD20 monoclonal antibodies, and so forth. Importantly, with this appreciation came the need for more quantitative approaches to specifically measure HLA antibodies, particularly donor-specific antibodies (DSA). 

Box 1
Box 1:
no caption available

The tools to measure HLA antibodies

Historic assays to detect HLA antibodies relied on dye-exclusion assays such that cells not recognized by the antibody remained alive, and thus excluded the dye, whereas cells that were killed by the antibody/complement activation internalized the dye. This Complement Dependent Cytotoxicity (CDC) assay was shown to lack in sensitivity and specificity. The assay was further augmented by the addition of antihuman-globulin (AHG). The microscope was later replaced with flow cytometry and cell fluorescence rather than cell death is measured when performing flow cytometric crossmatch (FCXM) assays. These modifications to cell-based assays increased their sensitivity but did not improve their specificity.

A barrier to achieving specificity in cell-based assays is the expression of molecules other than HLA on the cell membrane that can serve as a target for antibodies. With the appreciation of the Humoral Theory and the need for tools to specifically identify HLA antibodies, Terasaki developed solid-phase-based assays with recombinant HLA antigens coating microparticle beads leading to increased specificity of antibody detection assays. Currently, the most commonly used test is the single antigen bead (SAB) assay. The output is mean fluorescence intensity (MFI) value on a scale of zero to roughly 25 000 leading the community to consider is as semi-quantitative. For an overview on the utility and limitations of these assays, the reader is referred to the literature [5–7,8▪▪,9].

In search of clinically relevant tools

In an attempt to provide more clinically meaningful tools, the transplant community explored different approaches. The first utilized multiple available assays demonstrating that progressive positive results in the different tests (SAB, FCXM, CDC) can help risk stratify patients pretransplantation [10]. Additionally, as complement activation was shown to correlate with transplant outcome, permutations of the SAB assay were sought to identify involvement of initial steps in this cascade, that is, identification of C1q or C3d-binding assays. Indeed, multiple reports followed, demonstrating a correlation between complement binding in a solid-phase in-vitro assay, especially when serum samples were collected during an episode of acute rejection, and increased risk to develop rejection or present with graft loss [11,12]. It is important to highlight that solid phase complement assays are vulnerable to similar limitations as the basic HLA-SAB assay [13▪,14].

With the understanding that the 4 IgG subclass antibodies can initiate different effector functions [15▪], an additional permutation of the SAB assay was proposed, using IgG subclass-specific reporters. The evolution of IgG subclass switches follows a one-way direction IgG3 IgG1 IgG2 IgG4 [16], suggesting that patients with acute kidney injury will likely exhibit the early subclasses. Indeed, Lefaucheur et al.[17] analyzed serum samples from patients obtained during an episode of acute rejection or within the first year post transplant, demonstrating that 91% of the IgG3-positive DSA recipients had acute ABMR whereas the majority of the IgG4-containing DSA were present among patients with chronic features of rejection include transplant glomerulopathy. Enthusiasm was dampened [18▪▪] though by the fact that about 20% of patients with DSAs detected in the basic SAB assay had no demonstrable DSA using the subclass assay, an observation supported by multiple studies [17,19], questioning the ability to make clinical decisions based on these results. Earlier on, Lowe et al.[20] demonstrated that patients usually exhibit a mixture of all subclasses, complicating assignment of patients to a specific subclass category. Navas et al.[19] challenged the paradigm that IgG2/IgG4 do not bind C1q as these subclasses were commonly found among the C1q-binding antibodies in their cohort (78.9 vs. 38.6%; P < 0.001 compared with non-C1q-binding antibodies). At this point, it is difficult to draw robust conclusions regarding the impact of the individual subclasses of DSA and the clinical value of IgG subclass assays [18▪▪].

A model for understanding antibody-mediated complement activation was presented by Diebolder et al.[21], who elegantly documented the formation of antibody hexamers after binding to their target, leading to the recruitment and activation of C1. These hexamers can be formed by all four human IgG subclasses. In other words, activation of C1 requires some critical mass of antibodies, regardless of the subclass. At around the same time, we demonstrated a strong correlation between increasing titers of HLA antibodies and complement binding in the solid-phase assay [22]. Further, we demonstrated a linear correlation between the basic SAB assay and titer (up to a certain titer) compared with a logarithmic correlation between the C1q SAB assay and titer [6] (Fig. 1). The need for six antibody molecules to bind to the HLA target on the SAB, in close proximity to form the hexamer, likely explains the slow increase in C1q MFI values in lower titers, until the necessary amount of antibody clusters has formed.

FIGURE 1
FIGURE 1:
Correlation between antibody titer (provided as dilution of 2n), basic SAB MFI values (IgG MFI) and C1q MFI. Data from 596 antibodies is presented. The data demonstrate a linear correlation between the IgG MFI and the titer of the antibody (that is lost in higher titers) but a logarithmic correlation between the C1q SAB assay and the antibody titer. Note that the C1q MFI begins to curve only at a titer of 25–26. This is in line with the need to have at least five or six HLA molecules in close proximity to allow activation of C1q and the complement cascade. MFI, mean fluorescence intensity; SAB, single antigen bead.

Use of serum dilutions to overcome the limitations of single antigen bead assays

In order to evaluate the clinical utility of an assay, one must have a thorough understanding of all variables associated with its performance [18▪▪]. Specific for SAB assays, it is important to appreciate that beyond issues related to the reagents, some patients may exhibit serum-specific factors that can affect results and interpretation of the test. For example, especially among the highly sensitized patients, a common occurrence is the presence of inhibitory factors that affect the binding, and therefore, the detection of some of the HLA antibodies present in a serum sample. As a result, antibodies can falsely appear to be at a low level because of their low MFI values (if serum is not treated to mitigate this so called ‘prozone’ effect). This observation was reported by multiple investigators and can lead to under-appreciation of antibody strength [23,24]. Many believe that complement interference leads to the observed inhibition. Regardless of the cause, several approaches to minimize the impact of the inhibition have been documented [9]. Importantly, inhibition does not affect complement binding assays to the same extent as the basic IgG assay [22].

Another factor that may lead to under appreciation of antibody strength is the presence of a shared-epitope phenomenon, namely, some patients exhibit an antibody that recognizes multiple HLA antigens (hence, ‘spread’ over multiple beads) leading to low MFI value readout per bead but no readout that correlated with the antibody strength. The phenomenon, described and explained in a recent review by Garcia-Sanchez [8▪▪], is found only in some patients because of their specific/inherent antibody makeup. In order to obtain accurate data from the SAB assay, the Sensitization in Transplantation: Assessment of Risk (STAR) workgroup highlighted the above limitations and strongly recommended utilizing approaches that will unmask antibody strength in these situations [9].

Both inhibition and the shared-epitope phenomenon often lead to presentation of low MFI values that can mistakenly be interpreted as negative. However, under-appreciation of antibody strength can happen also when the SAB assay report high MFI values. Although not a limitation if transplantation is considered only in the absence of any DSA, some centers are willing to consider transplantation across DSA barriers and estimation of antibody strength in these cases is critical. Appreciation of antibody strength can also be helpful when treating ABMR or contemplating desensitization pretransplantation.

In this regard, it is also important to recognize that each bead in the assay is coated with a finite amount of HLA antigens and some patients have more antibodies than HLA targets in the assay. Once those antigens are saturated, additional antibodies in the serum will have no target to bind to, and therefore, will go undetected. Saturation is reached at an MFI value of about 25 000. Thus, antibodies with titers of 1 : 512 and beyond show similar MFI values (Fig. 2). Without performing some dilution, the unbound antibodies will remain masked, and antibody strength underestimated.

FIGURE 2
FIGURE 2:
Demonstration of HLA single antigen bead saturation. Violin plots of neat IgG-MFI correlation with antibody titer value are presented separately for HLA class I and class II assays. The linear correlation continues roughly up to a titer of 1 : 512--1 : 1024 after which the median IgG-MFI value remains pretty unchanged despite increase in titer. This observation highlights the fact that there is a finite amount of HLA antigens attached to each bead and once the target-antigen is saturated, the remaining HLA antibodies go undetected. However, when the serum is diluted and this information is factored in, it is possible to identify the presence of those additional antibodies. MFI, mean fluorescence intensity.

Clinical applications of antibody titers

The need to reduce levels of HLA antibodies arises in two main clinical scenarios: pretransplantation desensitization: among patients with an incompatible living donor (known DSA) or patients awaiting a deceased donor organ (unknown DSA), and posttransplantation among patients with ABMR and known DSA. Unfortunately, no therapy that has been proven effective to remove HLA antibodies, and there is a critical need to develop new agents. Deliberation by the Food and Drug Administration (FDA) [25] deemed SAB MFI results as unsatisfactory to support clinical trials, which is confounded by issues affecting assay reproducibility. Some patients appear to respond favourably to antibody depletion therapies, whereas other patients do not support the need to develop tools to help predict a favorable therapeutic response [26,27]. The use of complement fixing antibodies was also shown to be insufficient in this regard [12].

Recently, Pinelli et al.[28▪] reported that the amount of antibody removed using PP/IVIg (plasmapheresis/intravenous immunoglobulin) can be quantified and monitored using antibody titers. Moreover, it was shown that PP/IVIg removes HLA antibodies in a predictable manner -- linearly for the first five cycles (approximately), after which the quantity of antibodies removed declined with each cycle. As a consequence, by determining the DSA titer, it was possible to predict the number of PP/IVIg cycles required to proceed with transplant with confidence.

Titration studies have revealed the spectrum of antibody strength among patients. In our experience, the strongest antibodies are usually detected in sera of the very highly sensitized patients (although they can be found also in patients with lower %PRA). As shown in Fig. 2, bead saturation is reached at around a titer of 1 : 512 but some patients have antibodies at titers greater than 1 : 16 384. This is at least 26 more antibody than can be detected with MFI alone. Therefore, PP/IVIg can potentially remove an enormous amount of antibody (roughly 25 of antibodies) without the SAB showing a change in MFI. This could lead a clinician or investigator to erroneously conclude that PP/IVIg or other treatment is ineffective at antibody removal [22].

Titration studies also reveal the heterogeneity among the highly sensitized patients. Currently, the United Network for Organ Sharing (UNOS) utilizes calculated PRA (cPRA) as a metric to define levels of sensitization. Patients with cPRA of at least 98% are considered highly sensitized, recognized to have exponentially lower likelihood of finding a compatible donor, and thus receive significant priority in organ allocation [29]. Importantly, even a fraction of a percentage change in the cPRA translates into a significant reduction in compatible donors for these patients [30▪]. Here we demonstrate further heterogeneity among patients. Even among patients with 100% cPRA, vast differences in antibody strength (titer) are apparent. Specifically, Fig. 3 is a heat map of cPRAs, computed individually for each dilution after performing titration studies (N = 20 highly sensitized patients). All patients demonstrated cPRA greater than 99.9% with undiluted serum. Important observations include the rapid decline of cPRA for patients I and II, reaching cPRA below 98%; with only 1 : 8 and 1 : 16 dilutions, and the very slow decline of cPRA for patients XIX and XX who reached a cPRA less than 98% only when their serum was diluted 1 : 16 384. Given these observations, it is reasonable to assume that patients I and II will potentially respond well to desensitization and more likely to receive a transplant. Additional information is shown for patients IV and V who both have a cPRA of 100% with undiluted serum but drop to less than 98% cPRA when serum is diluted 1 : 64 (essentially removing 25 of antibodies). However, patient IV drops to a cPRA of 60% and patient V will drop to cPRA of 94%. Thus, although both patients will benefit from reduced cPRA below 98%, patient IV is theoretically more likely to find a donor. Figure 3 illustrates that the ability to determine cPRA by titers can be used as an approach to identify those patients that are more likely to respond favorably to desensitization therapy.

FIGURE 3
FIGURE 3:
Heat Map of cPRAs (calculated panel reactive antibodies). Titration studies were performed on 20 highly sensitized patients and cPRA was computed individually for each dilution. All patients demonstrated cPRA above 99.9% at their undiluted serum. Several important observations are highlighted in the body of the manuscript (specifically comparisons between patients I and II; patients XIX and XX, and IV and V). This data illustrates the ability of titration studies to provide an additional layer of information helping to distinguish, among the very highly sensitized patients, those that have stronger versus not as strong antibodies. We submit that this information can be beneficial in determining which patients are more likely to benefit from desensitization on the waitlist.

Unanswered questions

Published data explain, at least in part, why some patients seem to favorably respond to desensitization using PP/IVIg whereas others, with similar MFI values, do not respond [28▪]. Similar studies evaluating responses to desensitization using other immunosuppressive agents need to be performed in order to confirm that titration studies can, indeed, serve as a prognostic tool for such treatments. It is important to appreciate that responses to desensitization pretransplantation are not affected by strong continuous immune activation by a graft, which can significantly impact the kinetics and dynamics of observed responses. Thus, studies to evaluate responses to antibody removal therapies in the posttransplant period must be conducted. Moreover, variability between patients in the posttransplantation period is more complex given the differences between the immunologic/immunogenic barrier that are being crossed; and the involvement of a memory component that cannot be predicted.

Importantly, all antibody assays provide a snapshot in time, which does not necessarily correlate with potential changes in antibodies over time. Therefore, high-level antibodies, whether complement binding or high titer, are likely linked to a poor clinical state at the present time but low-level antibodies (low titer or no binding with the complement assays) cannot predict the rigor of an immune memory response that can be awakened. Moreover, damage to the graft can be associated with pathways other than complement activation [31].

CONCLUSION

Herein we provide evidence supporting the utility of titration studies, highlighting why and when this information can support clinical decision-making. Specifically, although titration studies can identify presence of inhibition, approaches, such as the use of EDTA or complement binding assay can eliminate prozone in many cases. The shared epitope phenomenon, which is another confounder of the SAB assay, cannot be identified or resolved by either of the approaches mentioned above and require high-level consultation with the histocompatibility team.

Titration studies provide clinically relevant information for patients considered for desensitization with a live donor where they can add prognostic value. Determining DSA titer can forecast the likelihood of lowering DSA levels to levels acceptable by the center to cross as barrier to transplantation. It can also project the number of PP/IVIg cycles that will be required to reach that point, thus allowing to schedule date of transplant with high accuracy. Additional studies are required to test whether titration studies can be used to monitor efficacy of other desensitization tools or approaches to treat AMR.

Titration studies provide clinically relevant information for highly sensitized patients considered for desensitization on the waitlist by providing more granular view of antibody strength. This in turn translates into information regarding the likelihood of lowering antibody levels to have impact on availability of potential donors.

Performing serial dilutions of serum followed by testing with SAB is an expensive proposition. Initial reports highlighting the added value that can be obtained by using this approach were met with significant opposition because of the added cost and confusion of how to interpret such results. Moreover, the term ‘titer’ was misused in many publications referring to high MFI value antibodies as high-titer antibodies, without performing any dilutions [32,33]. In fact, most of the clinically relevant information can be obtained by using only two to three dilutions. For example, centers considering removal of DSA by PP/IVIg desensitization could probably obtain all required data by running a single dilution at a titer of 1 : 64. This should discriminate between antibodies that are likely to be eliminated or significantly reduced by PP/IVIg desensitization from those that are not likely to favorably respond to treatment. Adding one more dilution, at approximately titer of 1 : 256, will likely show the upper limit of antibodies that can be aggressively treated versus those that will fail to respond completely. These recommendations are given based on our experience and clearly more data is required to make strong recommendations. It is also clear, given data presented here and elsewhere that determining HLA antibody strength using titer metrics has the potential to serve as a prognostic biomarker.

Acknowledgements

A.R.T. is a Paul I. Terasaki Research Fund Scholar.

Financial support and sponsorship

Part of the work presented here was supported by a research grant from the Paul I. Terasaki Research Fund; part of the reagents were provided on One Lambda, a brand of ThermoFisher Scientific Brand.

Conflicts of interest

A.R.T. received honoraria and free research reagents from One Lambda/A band of ThermoFisher Scientific Brand; and free research reagents from Immucor.

REFERENCES AND RECOMMENDED READING

Papers of particular interest, published within the annual period of review, have been highlighted as:

▪ of special interest

▪▪ of outstanding interest

REFERENCES

1. Terasaki PM, Marchioro TL, Starzl TE. Sero-typing of human lymphocyte antigens: Preliminary trials on long-term kidney homograft survival. First Histocompatibility Workshop. 1964. pp. 83–84.
2. Starzl TE, Lerner RA, Dixon FJ, et al. Shwartzman reaction after human renal homotransplantation. N Engl J Med 1968; 278:642–648.
3. Terasaki PI. Humoral theory of transplantation. Am J Transplant 2003; 3:665–673.
4. Terasaki PI. A personal perspective: 100-year history of the humoral theory of transplantation. Transplantation 2012; 93:751–756.
5. Konvalinka A, Tinckam K. Utility of HLA antibody testing in kidney transplantation. J Am Soc Nephrol 2015; 26:1489–1502.
6. Tambur AR, Wiebe C. HLA diagnostics: evaluating DSA strength by titration. Transplantation 2018; 102: (1s Suppl 1): S23–S30.
7. Wehmeier C, Honger G, Schaub S. Caveats of HLA antibody detection by solid-phase assays. Transpl Int 2020; 33:18–29.
8▪▪. Garcia-Sanchez CU, Usenko CY, YC Herrera ND, Tambur AR. The shared epitope phenomenon—a potential impediment to virtual crossmatch accuracy. Clin Transplant 2020; 34:e13906.
9. Tambur AR, Campbell P, Claas FH, et al. Sensitization in Transplantation: Assessment of Risk (STAR) 2017 Working Group Meeting Report. Am J Transplant 2018; 18:1604–1614.
10. Schinstock CA, Gandhi M, Cheungpasitporn W, et al. Kidney transplant with low levels of DSA or low positive B-flow crossmatch: an underappreciated option for highly sensitized transplant candidates. Transplantation 2017; 101:2429–2439.
11. Loupy A, Lefaucheur C, Vernerey D, et al. Complement-binding anti-HLA antibodies and kidney-allograft survival. N Engl J Med 2013; 369:1215–1226.
12. Viglietti D, Bouatou Y, Kheav VD, et al. Complement-binding anti-HLA antibodies are independent predictors of response to treatment in kidney recipients with antibody-mediated rejection. Kidney Int 2018; 94:773–787.
13▪. Lan JH, Tinckam K. Clinical utility of complement dependent assays in kidney transplantation. Transplantation 2018; 102: (1S Suppl 1): S14–S22.
14. Karahan GE, Claas FHJ, Heidt S. Technical challenges and clinical relevance of single antigen bead C1q/C3d testing and IgG subclass analysis of human leukocyte antigen antibodies. Transpl Int 2018; 31:1189–1197.
15▪. Valenzuela NM, Schaub S. The biology of IgG subclasses and their clinical relevance to transplantation. Transplantation 2018; 102: (1S Suppl 1): S7–S13.
16. Stavnezer J, Guikema JE, Schrader CE. Mechanism and regulation of class switch recombination. Annu Rev Immunol 2008; 26:261–292.
17. Lefaucheur C, Viglietti D, Bentlejewski C, et al. IgG donor-specific anti-Human HLA antibody subclasses and kidney allograft antibody-mediated injury. J Am Soc Nephrol 2016; 27:293–304.
18▪▪. Tambur AR, Campbell P, Chong AS, et al. Sensitization in transplantation: assessment of risk (STAR) 2019 Working Group Meeting Report. Am J Transplant 2020; 18:1604–1614. in press.
19. Navas A, Molina J, Aguera ML, et al. Characterization of the C1q-binding ability and the IgG1-4 subclass profile of preformed anti-HLA antibodies by solid-phase assays. Front Immunol 2019; 10:1712.
20. Lowe D, Higgins R, Zehnder D, Briggs DC. Significant IgG subclass heterogeneity in HLA-specific antibodies: implications for pathogenicity, prognosis, and the rejection response. Hum Immunol 2013; 74:666–672.
21. Diebolder CA, Beurskens FJ, de Jong RN, et al. Complement is activated by IgG hexamers assembled at the cell surface. Science 2014; 343:1260–1263.
22. Tambur AR, Herrera ND, Haarberg KM, et al. Assessing antibody strength: comparison of MFI, C1q, and titer information. Am J Transplant 2015; 15:2421–2430.
23. Schnaidt M, Weinstock C, Jurisic M, et al. HLA antibody specification using single-antigen beads--a technical solution for the prozone effect. Transplantation 2011; 92:510–515.
24. Wahrmann M, Hlavin G, Fischer G, et al. Modified solid-phase alloantibody detection for improved crossmatch prediction. Hum Immunol 2013; 74:32–40.
25. Archdeacon P, Chan M, Neuland C, et al. Summary of FDA antibody-mediated rejection workshop. Am J Transplant 2011; 11:896–906.
26. Patel J, Everly M, Chang D, et al. Reduction of alloantibodies via proteasome inhibition in cardiac transplantation. J Heart Lung Transplant 2011; 30:1320–1326.
27. Lefaucheur C, Antoine C, Suberbielle C, Glotz D. Mastering the risk of HLA antibodies in kidney transplantation: an algorithm based on pretransplant single-antigen flow bead techniques. Am J Transplant 2011; 11:1592–1598.
28▪. Pinelli DF, Zachary AA, Friedewald JJ, et al. Prognostic tools to assess candidacy for and efficacy of antibody-removal therapy. Am J Transplant 2019; 19:381–390.
29. Israni AK, Salkowski N, Gustafson S, et al. New national allocation policy for deceased donor kidneys in the United States and possible effect on patient outcomes. J Am Soc Nephrol 2014; 25:1842–1848.
30▪. Schinstock CA, Smith BH, Montgomery RA, et al. Managing highly sensitized renal transplant candidates in the era of kidney paired donation and the new kidney allocation system: is there still a role for desensitization? Clin Transplant 2019; 33:e13751.
31. Hickey MJ, Valenzuela NM, Reed EF. Alloantibody generation and effector function following sensitization to human leukocyte antigen. Front Immunol 2016; 7:30.
32. Morrow WR, Frazier EA, Mahle WT, et al. Rapid reduction in donor-specific antihuman leukocyte antigen antibodies and reversal of antibody-mediated rejection with bortezomib in pediatric heart transplant patients. Transplantation 2012; 93:319–324.
33. O’Leary JG, Demetris AJ, Friedman LS, et al. The role of donor-specific HLA alloantibodies in liver transplantation. Am J Transplant 2014; 14:779–787.
Keywords:

HLA antibody; titer; transplantation

Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc.