Donor-reactive HLA antibodies in renal allograft recipients: Considerations, complications, and conundrums

Human Immunology 70 (2009) 610 – 617

Contents lists available at ScienceDirect

Donor-reactive HLA antibodies in renal allograft recipients: Considerations, complications, and conundrums Howard M. Gebel a,*, Omar Moussa b, David D. Eckels c, Robert A. Bray a a

Department of Pathology, Emory University, Atlanta, Georgia, USA Department of Pathology, Medical University of South Carolina, Charleston, South Carolina, USA c Department of Pathology, University of Utah, Salt Lake City, Utah, USA b

A R T I C L E

I N F O

Article history: Received 8 March 2009 Accepted 9 April 2009 Available online 16 April 2009

Keywords: HLA antibodies HLA antigens HLA alleles Donor-directed antibodies Solid phase assays

A B S T R A C T

Whether sensitized patients wait for a compatible crossmatch with a deceased donor, enter a paired exchange program with the hope of finding a compatible living donor, or go through a desensitization protocol depends on a number of factors, not the least of which is the overall philosophy of the transplant center. Centers such as ours take the position that donor-directed antibodies detected by solid phase assays (even those that are “weak”) present an unacceptable risk factor to the patient. This philosophy is predicated on the biologic role of the immune system, specifically that antibodies were generated in response to a non-self (allo) antigen and that a successful immune response eliminates that which caused its stimulation. Although obviously an oversimplification, this philosophy mandates a comprehensive evaluation of HLA antibodies in sensitized recipients. This article addresses the challenges and conundrums associated with human leukocyte antigen antibody identification. 䉷 2009 Published by Elsevier Inc. on behalf of American Society for Histocompatibility and Immunogenetics.

1. Introduction Following the landmark report of Patel and Terasaki [1], a positive cytotoxic crossmatch between donor cells and recipient serum was considered a contraindication to renal transplantation. The high rate of immediate graft loss among patients undergoing transplantation across this barrier was unacceptable. It soon became apparent that the clinically relevant antibodies in a lymphocyte crossmatch were those directed against antigens encoded by the human major histocompatibility complex (MHC), henceforth referred to as the human leukocyte antigen (HLA) complex. The paradigm to not cross the positive crossmatch barrier was modified when it was recognized that cytotoxic crossmatches resulting from non-HLA antibodies (e.g., autoantibodies) had no impact on allograft survival and could be safely ignored (reviewed in [2]). However, there was a growing appreciation that complement-fixing, donor-directed HLA antibodies undetectable by standard cytotoxicity assays were clinically relevant (reviewed in [3]). Such observations led to the development of more sensitive antibody detection tests including the antiglobulin-enhanced cytotoxicity (AHG-CDC) and flow-cytometric crossmatch assays [4 – 8]. Although these tests were more sensitive, their shortcoming was their relative lack of specificity [9,10]. Subsequently, the development and implementation of solid phase antibody detection systems (SPADS) that specifically identified HLA antibodies represented a major advancement

* Corresponding author. E-mail address: [email protected] (H.M. Gebel).

[3,11,12]. These sensitive and specific assays provided the transplant community with analytical tools not previously available and, in many situations, actually changed how crossmatch data were interpreted. For example, in a single-center study by Kerman et al., no differences were initially reported in the number of rejection episodes and/or graft losses among recipients undergoing transplantation with renal allografts from donors whose flow crossmatches were positive or negative as long as the AHG-CDC crossmatches were negative [13]. Subsequently, retesting of sera from these patients by solid phase technology revealed that several of the “positive” flow-cytometric crossmatches were actually caused by non-HLA antibodies [14]. Importantly, the above patients experienced no rejection episodes or graft loss during the study period. Thus, it should not be surprising that SPADS have supplanted cell-based assays and have become the goldstandard for identifying HLA antibodies. Nonetheless, how information from solid phase assays is translated into clinical practice is quite controversial. For example, in terms of risk, should “weak” (low titered, low fluorescence intensity) donor-directed antibodies be given the same clinical significance as “strong” (high titered, strong fluorescence intensity) donor-directed antibodies [15–18]? Can non– complement-fixing donor-directed HLA antibodies be considered less problematic than complement-fixing antibodies, as suggested by Bohmig et al. [19], or should they be considered to confer the same degree of risk to longterm graft survival as proposed by Cai and Terasaki [20]? Such disparate data beg the question of how antibodies are classified as “strong” or “weak” or as complement fixing/non– complement fixing. Indeed,

0198-8859/09/$32.00 - see front matter 䉷 2009 Published by Elsevier Inc. on behalf of American Society for Histocompatibility and Immunogenetics. doi:10.1016/j.humimm.2009.04.012

H.M. Gebel et al. / Human Immunology 70 (2009) 610 – 617

we recently demonstrated that HLA antibodies considered to be non– complement fixing by AHG-CDC assays were in fact complement fixing when assayed by a more sensitive flow-cytometry– based assay [21]. Thus, the fact that an antibody does not fix complement in vitro should not automatically be taken as evidence that the same antibody will not fix complement in vivo. There are numerous other controversies involving antibodies detected by SPADS, with literature supporting either side of the argument. Can donor-directed antibodies detected in a solid phase assay that do not result in a positive lymphocyte crossmatch, even by the most sensitive of assays, be ignored [22,23]? Are antibodies directed against certain HLA antigen specificities less problematic than others (reviewed in [24])? Are donor-directed class II antibodies more tolerable than donor-directed class I antibodies [25,26]? Over the past 3 years, the concept of virtual crossmatching (i.e., predicting the crossmatch outcome without actually performing a physical crossmatch) has taken center stage as an approach to more efficient allocation and distribution of deceased-donor

611

organs [27–29]. Although virtual crossmatching has proved successful for predicting actual negative crossmatches between specific donor:recipient combinations, it can be argued that, if the threshold for calling an antigen “unacceptable” is set too low, those patients will never be considered for donors that might be suitable. Without question, not all patients with low, moderate, or even high levels of donor-directed antibody have poor outcomes if they receive donor organs that express the corresponding antigens. Unfortunately, other patients with characteristics virtually identical to those just described will experience severe episodes of rejection and may lose their grafts if they undergo transplantation. There are currently no assays that unequivocally categorize patients into one group or the other. Over the last decade, desensitization strategies have been advocated as an approach to performing transplantations in highly sensitized patients, especially those who present with crossmatch-incompatible donors [30,31]. Albeit an excellent strategy for some patients, there is clear evidence that patients who are

Fig. 1. (A) At least three different types of glycoproteins are expressed on the surface of B cells: namely, MHC class I antigens, MHC class II antigens (which can bind to MHC class II antibodies), and Fc receptors. Low-titer class I antibodies can bind to B cells and not T cells based on B cells expressing more class I than T cells. Antibodies to class II will bind to B cells and not T cells (which, under normal circumstances, do not express class II). Fc receptors (which have a significantly higher expression on B cells than on T cells) can bind the Fc portion of antibodies independent of their antigen specificity. Thus, in a flow-cytometric crossmatch, B cells will have a higher background than T cells. (B) Reduction/elimination of Fc receptors from lymphocytes before performing a crossmatch. Cells are treated with pronase, a proteolytic enzyme that will digest Fc receptors without affecting the expression of class I or class II antigens [44]. Pronase also digests CD20 (a target of Rituxan) from the surface of B cells [58].

612

H.M. Gebel et al. / Human Immunology 70 (2009) 610 – 617

desensitized before undergoing transplantation will not all have good long-term prognoses, with up to 40% of these patients experiencing antibody-mediated rejection, transplant glomerulopathy, and/or graft loss [32–34]. Why some desensitized patients are more likely to experience antibody-mediated injury than others is not known. One explanation is that susceptible patients have a higher number of memory B cells and/or plasma cells dedicated to the production of donor-directed HLA compared with patients with good prognoses [35]. New tests recently developed that identify HLA antibody–producing cells in vitro may eventually help to discern sensitized patients into high-risk and low-risk categories [36,37]. Should sensitized patients wait for a compatible crossmatch with a deceased donor, enter a paired exchange program [38] with the hope of finding a compatible living donor, or go through a desensitization protocol? This is not always a simple decision. How sensitized is the patient? How long has she or he been waiting? What is the patient’s overall health status? Another major factor is the philosophy of the transplant center. Our center takes the position that donor-directed HLA antibodies (even those that are “weak”) are an unacceptable risk to the patient. This philosophy is predicated on the biologic role of the immune system, namely, that these alloantibodies were generated in response to a non– self-antigen and that a successful immune response will eliminate that which caused its stimulation. Albeit an oversimplification, this philosophy requires a thorough evaluation of HLA antibodies in sensitized recipients. 2. HLA antibody detection and identification Methods to detect and categorize HLA antibodies have steadily evolved during the past 40 years. Initially, a complement-dependent cytotoxicity (CDC) assay was the only assay available to identify antibody activity. Although easy to perform and relatively inexpensive, allograft recipients with pretransplantation HLA antibodies undetectable in CDC assays experienced early (accelerated) antibody-mediated episodes of rejection and graft loss [1]. These clinical ramifications accelerated the development and implementation of more sensitive assays to identify (HLA) antibodies. Early variations included additional wash steps and/or longer incubation times (to eliminate anticomplementary factors and/or to increase the binding of low avidity antibodies) [4]. Subsequently, the use of a secondary reagent, namely anti-human immunoglobulin (AHG), permitted the detection of non– complement-fixing antibodies [5–7]. Even so, it was soon recognized that these approaches were limited. For example, when assaying for HLA antibodies using a panel of frozen cells, panel composition could skew the results. A given panel may indicate that the patient’s serum reacts with every cell (100% PRA), indicating that the patient would need a “perfect” match to have a negative crossmatch. But what if the patient had only one antibody (e.g., A*02) and every target cell expressed HLAA*02? Obviously, this would be an extreme example, but it indicates how panel composition could provide misinformation. Similarly, analysis of patient sera with a panel of cells might reveal that the patient has antibodies to only HLA-A*01. However, if several of the HLA*01 target cells also expressed HLA-B*08 and no cells expressed HLA-B*08 without also expressing HLA-A*01, antibodies to HLA-B*08 would be masked. Under these circumstances, a positive crossmatch with a cell expressing HLA-B*08 without HLA-A*01 would be interpreted as being caused by a non-HLA antibody. Another limitation was the requirement for viable cells as targets. Thus, if a patient possessed a lymphocyte-reactive, non-HLA antibody, an incorrect assumption would be made about that individual’s chances of finding a compatible donor. Finally, most of the antibody identification assays using lymphocyte panels were limited to T cells and therefore could only detect antibodies detected to class I HLA antigens.

In 1983, Garavoy et al. took advantage of a relatively new technology (flow cytometry) in clinical laboratory practice, and observed that a flow cytometry– based crossmatch (FCXM) could detect levels of HLA antibodies on lymphocytes undetectable by other techniques. Although clearly more sensitive than other crossmatch assays, as the FCXM was a cell-based assay, positive T-cell and/or (especially) B-cell FCXMs lacked antigen specificity. That some positive crossmatches were due to HLA and others to non-HLA antibodies contributed to a conflicted and confused literature. For example, when donor:recipient pairs transplanted after a negative CDC or AHG-CDC crossmatch were re-evaluated based on whether their flow-cytometric crossmatches were positive or negative, some centers reported significant differences in episodes of rejection/graft loss, whereas other centers reported no differences between the two groups [13,39 – 42]. The underlying assumptions were that a positive T-lymphocyte FCXM was caused by HLA class I antibodies, whereas a positive B-cell FCXM (with a negative T cell FCXM) was considered the result of class II– directed HLA antibodies. It is important to recognize that numerous other antigens are also expressed on T and B lymphocytes. Thus, positive crossmatches could result from the following: 1) antibodies to selfantigens on recipient lymphocytes that bind to the same epitope on donor lymphocytes; 2) Fc receptor binding of immunoglobulin independent of antigen specificity; or 3) binding of antibodies to lymphocyte antigens other than those encoded by the major histocompatibility complex [43– 45]. In any of the above circumstances, a positive crossmatch is not considered a contraindication to transplantation [3]. In the current era, new tools have become available that provide a rationale to help interpret these exquisitely sensitive cellular assays. In today’s HLA laboratory, methods to identify, reduce, and eliminate non-HLA antibodies in FCXM assays are routinely applied. Such approaches include ultracentrifugation of patient serum to eliminate immune complexes that nonspecifically bind to donor lymphocytes and treatment of donor cells with pronase, a proteolytic enzyme that cleaves Fc receptors from the cell surface [44]. Elimination (or at least reduction) of surface Fc receptors significantly decreases nonspecific binding of immunoglobulin and reduces the rate of false-positive B-cell crossmatches (Figs. 1A, 1B). However, the most significant breakthrough was the development of assays using solid phase matrices coated with purified HLA class I or class II antigens. These new technologies have completely transformed the histocompatibility laboratory’s approach and ability to detect HLA antibodies. In these assays, HLA class I or class II antigens are isolated from transformed or transfected cell lines (either as an entire cluster or single allele, respectively) and attached to solid phase matrices such as a plastic plate or inert microparticles [11,12]. 3. Confounding issues: Pretransplantation conundrums and challenges Recent evidence revealed that patients awaiting allograft transplantation can produce antibodies not only to HLA antigens but to specific alleles of HLA antigens [14,46]. Our first experience with allele-specific antibodies occurred in a patient who displayed antibodies to HLA-A*68. Importantly, this patient expressed the HLAA*68 antigen. Was this an example of HLA autoantibodies? Further investigation using high-resolution histocompatibility typing revealed that the patient possessed HLA-A*6802, an allele of HLAA*68, whereas her husband and son (with whom she was crossmatch positive) both possessed the HLA-A*6801 allele. Assessment of the protein sequences of A*6801 and A*6802 revealed that these two alleles differed by five amino acids. More recently, we identified a patient who was homozygous for HLA-A*02 who was found to have antibodies only to HLA-A*25,26,34,43 and 66 (the so-called 10 CREG antigens) using a flow-based detection assay. Interest-

H.M. Gebel et al. / Human Immunology 70 (2009) 610 – 617

613

Fig. 2. Highlighted portion reveals a shared amino acid sequence (epitope) between HLA-A* 0203, 2501, 2601, 3401, 4301, and 6601. It should be noted that HLA-A*0201 and HLA-A*0206 do not possess this sequence.

ingly, this patient also had an antibody to HLA-A*0203 when the serum was tested on an extended bead panel using Luminex technology. High resolution typing of the patient documented the presence of HLA-A*0201. Assessment of protein sequences revealed that HLA-A*0203 shares an identical sequence common to HLAA*25, 26, 34, 43, and 66, which is not expressed on A*0201 or A*0205 (Fig. 2). Moreover, this potential epitope occurs on an antibody accessible part of the HLA molecule (␣-helix). This particular case clearly indicates that epitopes can be shared among HLA antigens previously not considered to be in the same CREG. More importantly, the antibody to HLA-A*0203 we detected was “strong”

Table 1 HLA class I and class II allele–specific antibodies Class I HLA-A*

HLA-B*

*0201 *0203 *0206 *1102 *3007 *6801

*2708 *4402 *4403 *5002 *5704

Class II HLA-DQA

HLA-DRB1

HLA-DQB1

HLA-DRB3

*0201 *0301

*0101 *0103 *0401 *0404 *0405 *0801 *0803 *1301 *1303 *1401 *1404

*0301 *0302 *0303

*0101 *0201

and would be considered an unacceptable antigen. However, because this antibody was not identified with what was then our routine panel for antibody identification, we would certainly have accepted any donor that did not express the A10 CREG antigens, especially donors that were HLA-A*02 (since our patient was homozygous for that antigen), including donors who expressed HLA-*0203. A positive crossmatch with such a donor would most likely be attributed to non-HLA antibodies. These data underscore the need to be vigilant for allele-specific antibodies (at the very least alleles that are common in the donor population) and support the concept that high-resolution typing of donors (again, for the common alleles) is warranted in solid organ transplantation. In addition to antibodies against HLA-A*6801 and A*0203, several other allele specific antibodies have been identified by our laboratory (Table 1). The implications of allele-specific antibodies for organ allocation are significant. Currently, in the United States, patients with antibodies to HLA-B*4402 but not HLA-B*4403 are not distinguished from patients who have antibodies to both alleles. This is because data entry of “unacceptable/avoid” antigens into the United Network for Organ Sharing (UNOS) database is typically limited to antigens, not alleles. As a result, patients listed as having antibodies to HLA-B44 but who have antibodies only to HLA-B*4403 will not be offered kidneys from any HLA-B44 positive donor, some of whom would be predicted to have a negative crossmatch. The only option available for these patients is not to consider HLA-B44 as “unacceptable” and to crossmatch with them

Table 2 FCXM before and after intravenous immunoglobulin (IVIG)

Patient serum ⫹ carrier Patient serum ⫹ IVIG

⌬T CELL

⌬ B CELL

⫹196 ⫹43

⫹153 ⫹30

Interpretation: IVIG converted FCXM from positive to negative.

614

H.M. Gebel et al. / Human Immunology 70 (2009) 610 – 617

with all HLA-B44 positive donors. Based on the frequencies of HLA-B*4402 and 4403, positive crossmatches would be expected approximately half of the time. Neither of the above approaches is acceptable. A better solution would be to allow for the entry of allele-specific antibodies in the UNOS matching system. This would necessitate high-resolution typing of all donors (at least for those antigens to which allele specific antibodies have been identified), not for the purpose of HLA matching but, rather, for identification of acceptable mismatches. Instead of waiting (perhaps too long) for a crossmatch-compatible donor, many centers will apply desensitization strategies to remove/ inhibit HLA antibodies [30,31]. The most successful of these approaches treats patients with either plasmapheresis and a low dose (100 mg/kg) of intravenous immunoglobulin (IVIG) or a high dose (2 g/kg) without plasmapheresis. For either approach, the goal is to

convert donor-reactive crossmatches from positive to negative or, at the very least, from strongly positive to weakly positive. There are two underlying premises here: 1) by reducing the strength of the antibody the transplant can be performed without fear of hyperacute rejection; and 2) desensitized patients who develop antibody-mediated rejection post-transplantation can be rescued with appropriate therapy. In monitoring for effective desensitization, it is important to understand the limitations of in vitro assays. For example, although low-dose IVIG does not interfere with the detection of HLA antibodies on lymphocytes or microparticles [47], the same cannot be said for high-dose IVIG [48]. An example of how high-dose IVIG can interfere with the detection of HLA antibodies begins in Table 2. Here, it appears as though the addition of high-dose IVIG to a patient’s serum converted a flow-cytometric T-cell–positive and

Fig. 3. (A) Flow-cytometric T-cell (top) and B-cell (bottom) crossmatches in the presence of normal human serum (NHS) or patient serum diluted 1:1 with carrier. Numbers in upper right-hand quadrant of each histogram represent the median channel value (MCF) of the cells on a 1024 scale. The difference (⌬) between the NHS and patient serum is 196 channels for T cells and 128 channels for B cells. A crossmatch is considered positive when the ⌬ MCF is ⬎60 for T cells and ⬎100 for B cells. (B) Flow-cytometric T-cell (top) and B-cell (bottom) crossmatches in the presence of NHS or patient serum diluted 1:1 with IVIG. The ⌬ MCF between the NHS and patient serum is 43 for T cells and 55 for B cells, suggesting that the crossmatch became negative. Of note, however, the MCF of T cells and B cells in NHS diluted in carrier vs NHS diluted in IVIG are significantly different (T cells:258 vs 441; B cells: 279 vs 525). With such a high background, the crossmatches (T and B) cannot be interpreted.

H.M. Gebel et al. / Human Immunology 70 (2009) 610 – 617

B-cell–positive crossmatch with his donor to negative. These data suggest that it would be reasonable for this donor:recipient combination to proceed to transplantation. However, upon closer examination (Figs. 3A, 3B), additional information changes the interpretation. Specifically, when comparing flow-cytometric crossmatch results performed with normal human serum (which contains no HLA antibodies) “spiked” with IVIG, it became clear that the background binding had increased substantially above that of normal human serum without IVIG. Thus, quantifiable differences including channel values and molecules of equivalent solution fluorescence (MESF) values between the control and patient specimens will not be reliable. Obviously, for the example shown here, it is inappropriate to interpret the crossmatch between donor lymphocytes and recipient serum containing IVIG as negative. A more accurate approach is a recently described assay in which donor cells are treated with recipient serum, with or without IVIG, followed by monitoring C3b deposition on the donor cells as a means to determine whether IVIG treatment will be beneficial to a particular donor:recipient pair [49]. 4. Confounding issues: Post-transplantation conundrums and challenges Although a large number of patients have been successfully desensitized and undergone transplantation at centers around the world, many of these patients experience acute antibody-mediated rejection and/or transplant glomerulopathy [32,50 –52]. As a result, the use of Thymoglobulin as an induction and/or rescue agent has steadily increased [53]. Thymoglobulin is a polyclonal rabbit antihuman reagent that possesses at least 23 different antibody specificities, including antibodies to human HLA class I and class II antigens as well as antibodies against ␤2-microglobulin [54]. Hence, when monitoring post-transplantation for the presence of HLA antibodies, it is critical to know that Thymoglobulin has been administered. Because Thymoglobulin contains antibodies against both class I and class II antigens, it can directly interfere with the detection of human HLA antibodies (Fig. 4). We recently demonstrated that the

615

presence of a therapeutic dose (100 ng/ml) of Thymoglobulin in a patient’s serum with known HLA alloantibodies masked the detection of HLA antibody even when using appropriate secondary reagents [55]. To be certain that Thymoglobulin will not interfere with HLA alloantibody detection, it must first be removed from the sample before testing. One method involves the use of goat– anti-rabbit immunoglobulin– coated beads [56]. For the sake of completeness, the adsorbed serum should then be tested to determine whether any rabbit immunoglobulin (i.e., Thymoglobulin) remains postadsorption. Only by completing this rather tedious process is it feasible to reliably determine whether low levels of HLA antibody are present in sera obtained from patients receiving Thymoglobulin. It is important to follow these protocols because of the clinical implications. Specifically, consider patients already receiving Thymoglobulin who present with symptoms of acute rejection: If they are monitored for HLA antibodies without Thymoglobulin removal, they would be considered to be antibody negative, and the therapeutic strategies to treat these patients may not include antibody depletion. Another immunosuppressive agent, Rituxan, a chimeric monoclonal antibody directed against CD20, also presents a challenge to the detection of clinically relevant HLA antibodies [57], specifically when performing B-cell crossmatches. In the presence of Rituxan, B-cell crossmatches will always appear positive, as Rituxan present in recipient serum will bind to donor cells expressing CD20. Recent studies have demonstrated that, rather than removing the antibody, CD20 can be removed from the B-cell surface via proteolytic cleavage with pronase [58]. Thus, under the appropriate conditions (and with the appropriate controls), positive B-cell crossmatches can be interpreted even in the presence of Rituxan. Other therapeutic monoclonal antibodies, such as Campath (anti-CD52), cannot be adsorbed from patient sera. In these circumstances, a crossmatch will always appear positive, although positive SPADS, including a recently developed immunosorbent crossmatch assay [59] will not be affected. It is critical that the laboratory be informed when patients have received any monoclonal or polyclonal antibody

Fig. 4. The presence of Thymoglobulin can interfere with the detection of low levels of HLA antibodies. As shown, a high concentration of Thymoglobulin can bind to surface HLA antigens and effectively block human HLA alloantibodies from binding to the same targets. The FITC-conjugated rabbit–anti-human Ig would have nothing to bind, as all human alloantibodies would be in solution, not cell bound. The result would be interpreted as negative.

616

H.M. Gebel et al. / Human Immunology 70 (2009) 610 – 617

treatments so that the appropriate procedures can be implemented and crossmatches are not misinterpreted. 5. Conclusions Over the past 40 years, many advances have taken place in the detection and identification of HLA antibodies. The most notable of these has been the introduction of sensitive and specific SPADS. The more exacting identification of HLA alloantibodies has significantly altered the manner in which the histocompatibility community interprets positive crossmatches between potential allograft recipients and their donors. In a similar manner, significant advances in immunosuppression and antibody modulation have created opportunities to perform transplantations in highly sensitized patients. For many transplant programs, the presence of pretransplantation, donor-directed alloantibody does not necessarily mean the same contraindication to transplantation as it once did; rather, the presence of donor-directed alloantibody represents one risk factor among many that must be considered. At present, how to assess the degree of risk is incompletely defined. For example, although it is certainly conceivable that some donor-directed antibodies are more detrimental to patient outcomes than other antibodies, this distinction is not clearly understood. Similarly, although target antigens such as MHC class I and/or class II may be differentially expressed on an allograft, how antigen expression affects outcomes in a sensitized patient is unknown. The relative strength (titer) of class I/class II HLA antibodies is also likely to play a role in graft outcome; but how, when, and for who must still be determined. One fact remains unambiguous. The human immune response is protective and its major function is to generate effectors (humoral and cellular) to eliminate the target(s) that provoked its response. With that in mind, the detection of donor-directed HLA antibody in an allograft recipient, regardless of the quantitative level, should be viewed as an indication that the immune system has been primed to allogeneic MHC antigens. At the very least, recipients who present with such antibodies should be considered to be at a different risk level from subjects devoid of HLA antibodies. References [1] Patel R, Terasaki PI. Significance of the positive crossmatch test in kidney transplantation. N Engl J Med 1969;280:735–9. [2] Braun WE. Laboratory and clinical management of the highly sensitized organ transplant recipient. Hum Immunol 1989;26:245– 60. [3] Gebel HM, Bray RA, Nickerson P. Pre-transplant assessment of donor-reactive, HLA-specific antibodies in renal transplantation: Contraindication vs. risk. Am J Transplant 2003;3:1488 –500. [4] Amos DB, Bashir H, Boyle W, MacQueen M, Tiilikainen A. A simple micro cytotoxicity test. Transplantation 1969;7:220 –3. [5] Cross DE, Whittier FC, Weaver P, Foxworth J. A comparison of the antiglobulin versus extended incubation time crossmatch: Results in 223 renal transplants. Transplant Proc 1977;9:1803– 6. [6] Fuller TC, Cosimi AB, Russell PS. Use of an antiglobulin-ATG reagent for detection of low levels of alloantibody-improvement of allograft survival in presensitized recipients. Transplant Proc 1978;10:463– 6. [7] Johnson AH, Rossen RD, Butler WT. Detection of alloantibodies using a sensitive antiglobulin microcytotoxicity test: Identification of low levels of preformed antibodies in accelerated allograft rejection. Tissue Antigens 1972;2: 215–26. [8] Iwaki Y, Lau M, Cook DJ, Takemoto S, Terasaki PI. Crossmatching with B and T cells and flow cytometry. Clin Transpl 1986:277– 84. [9] Christiaans MH, Overhof R, ten Haaft A, Nieman F, van Hooff JP, van den Berg-Loonen EM. No advantage of flow cytometry crossmatch over complement-dependent cytotoxicity in immunologically well-documented renal allograft recipients. Transplantation 1996;62:1341. [10] Gebel HM, Bray RA. Laboratory assessment of HLA antibodies circa 2006: Making sense of sensitivity. Transplantation Reviews 2006;20:189 –94. [11] Kao KJ, Scornik JC, Small SJ. Enzyme-linked immunoassay for anti-HLA antibodies–an alternative to panel studies by lymphocytotoxicity. Transplantation 1993;55:192– 6. [12] Pei R, Wang G, Tarsitani C, Rojo S, Chen T, Takemura S, et al. Simultaneous HLA class I and class II antibodies screening with flow cytometry. Hum Immunol 1998;59:313–22. [13] Kerman RH, Susskind B, Buyse I, Pryzbylowski P, Ruth J, Warnell S, et al. Flow cytometry-detected IgG is not a contraindication to renal transplantation: IgM may be beneficial to outcome. Transplantation 1999;68:1855– 8.

[14] Bray RA, Nickerson PW, Kerman RH, Gebel HM. Evolution of HLA antibody detection: Technology emulating biology. Immunol Res 2004;29:41–54. [15] Lefaucheur C, Suberbielle-Boissel C, Hill GS, Nochy D, Andrade J, Antoine C, et al. Clinical relevance of preformed HLA donor-specific antibodies in kidney transplantation. Am J Transplant 2008;8:324 –31. [16] Reinsmoen NL, Lai CH, Vo A, Cao K, Ong G, Naim M, Wang Q, Jordan SC. Acceptable donor-specific antibody levels allowing for successful deceased and living donor kidney transplantation after desensitization therapy. Transplantation 2008;86:820 –5. [17] Vaidya S. Clinical importance of anti-human leukocyte antigen-specific antibody concentration in performing calculated panel reactive antibody and virtual crossmatches. Transplantation 2008;85:1046 –50. [18] van den Berg-Loonen EM, Billen EV, Voorter CE, van Heurn LW, Claas FH, van Hooff JP, Christiaans MH. Clinical relevance of pretransplant donor-directed antibodies detected by single antigen beads in highly sensitized renal transplant patients. Transplantation 2008;85:1086 –90. [19] Bohmig GA, Bartel G, Wahrmann M. Antibodies, isotypes and complement in allograft rejection. Curr Opin Organ Transplant 2008;13:411– 8. [20] Cai J, Terasaki PI. Humoral theory of transplantation: Mechanism, prevention, and treatment. Hum Immunol 2005;66:334 – 42. [21] Saw CL, Bray RA, Gebel HM. Cytotoxicity and antibody binding by flow cytometry: A single assay to simultaneously assess two parameters. Cytometry B Clin Cytom 2008;74:287–94. [22] Bielmann D, Honger G, Lutz D, Mihatsch MJ, Steiger J, Schaub S. Pretransplant risk assessment in renal allograft recipients using virtual crossmatching. Am J Transplant 2007;7:626 –32. [23] Patel AM, Pancoska C, Mulgaonkar S, Weng FL. Renal transplantation in patients with pre-transplant donor-specific antibodies and negative flow cytometry crossmatches. Am J Transplant 2007;7(10):2371–7. [24] Zachary AA, Montgomery RA, Leffell MS. Defining unacceptable HLA antigens. Curr Opin Organ Transplant 2008;13:405–10. [25] Eng HS, Bennett G, Tsiopelas E, Lake M, Humphreys I, Chang SH, Coates PT, Russ GR. Anti-HLA donor-specific antibodies detected in positive B-cell crossmatches by Luminex predict late graft loss. Am J Transplant 2008;8:2335– 42. [26] Issa N, Cosio FG, Gloor JM, Sethi S, Dean PG, Moore SB, DeGoey S, Stegall MD. Transplant glomerulopathy: Risk and prognosis related to anti-human leukocyte antigen class II antibody levels. Transplantation 2008;86:681–5. [27] Appel JZ, 3rd, Hartwig MG, Cantu E, 3rd, Palmer SM, Reinsmoen NL, Davis RD. Role of flow cytometry to define unacceptable HLA antigens in lung transplant recipients with HLA-specific antibodies. Transplantation 2006;81:1049 –57. [28] Bray RA, Nolen JD, Larsen C, Pearson T, Newell KA, Kokko K, Guasch A, et al. Transplanting the highly sensitized patient: The Emory algorithm. Am J Transplant 2006;6:2307–15. [29] Zangwill S, Ellis T, Stendahl G, Zahn A, Berger S, Tweddell J. Practical application of the virtual crossmatch. Pediatr Transplant 2007;11:650 – 4. [30] Jordan SC, Vo A, Bunnapradist S, Toyoda M, Peng A, Puliyanda D, et al. Intravenous immune globulin treatment inhibits crossmatch positivity and allows for successful transplantation of incompatible organs in living-donor and cadaver recipients. Transplantation 2003;76:631– 6. [31] Montgomery RA, Zachary AA, Racusen LC, Leffell MS, King KE, Burdick J, et al. Plasmapheresis and intravenous immune globulin provides effective rescue therapy for refractory humoral rejection and allows kidneys to be successfully transplanted into cross-match-positive recipients. Transplantation 2000;70:887–95. [32] Gloor J, Cosio F, Lager DJ, Stegall MD. The spectrum of antibody-mediated renal allograft injury: Implications for treatment. Am J Transplant 2008;8:1367–73. [33] Haririan A, Nogueira J, Kukuruga D, Schweitzer E. Hess J, Gurk-Turner C, et al. Positive cross-match living donor kidney transplantation: Longer-term outcomes. Am J Transplant 2009;3:536 – 42. [34] West-Thielke P, Herren H, Thielke J, Oberholzer J, Sankary H, Raofi V, et al. Results of positive cross-match transplantation in African American renal transplant recipients. Am J Transplant 2008;8:348 –54. [35] Stegall M, Dean PG, Gloor J. Mechanisms of alloantibody production in sensitized renal allograft recipients. Am J Transplant (in press). [36] Mulder A, Eijsink C, Kardol MJ, Franke-van Dijk ME, van der Burg SH, Kester M, et al. Identification, isolation, and culture of HLA-A2-specific B lymphocytes using MHC class I tetramers. J Immunol 2003;171:6599 – 603. [37] Perry DK, Pollinger HS, Burns JM, Rea D, Ramos E, Platt JL, Gloor JM, Stegall MD. Two novel assays of alloantibody-secreting cells demonstrating resistance to desensitization with IVIG and rATG. Am J Transplant 2008;8:133– 43. [38] Hanto RL, Reitsma W, Delmonico FL. The development of a successful multiregional kidney paired donation program. Transplantation 2008;86:1744 – 8. [39] Iwaki Y, Cook DJ, Terasaki PI, Lau M, Terashita GY, Danovitch G, et al. Flow cytometry crossmatching in human cadaver kidney transplantation. Transplant Proc 1987;19:764 – 6. [40] Mahoney RJ, Ault KA, Given SR, Adams RJ, Breggia AC, Paris PA, et al. The flow cytometric crossmatch and early renal transplant loss. Transplantation 1990; 49:527–35. [41] Ogura K, Terasaki PI, Johnson C, Mendez R, Rosenthal JT, Ettenger R, et al. The significance of a positive flow cytometry crossmatch test in primary kidney transplantation. Transplantation 1993;56:294 – 8. [42] Pelletier RP, Orosz CG, Adams PW, Bumgardner GL, Davies EA, Elkhammas EA, et al. Clinical and economic impact of flow cytometry crossmatching in primary cadaveric kidney and simultaneous pancreas-kidney transplant recipients. Transplantation 1997;63:1639 – 45. [43] Ettenger RB, Jordan SC, Fine RN. Cadaver renal transplant outcome in recipients with autolymphocytotoxic antibodies. Transplantation 1982;35:429 –31.

H.M. Gebel et al. / Human Immunology 70 (2009) 610 – 617

[44] Vaidya S, Cooper TY, Avandsalehi J, Barnes T, Brooks K, Hymel P, et al. Improved flow cytometric detection of HLA alloantibodies using pronase: Potential implications in renal transplantation. Transplantation 2001;71:422– 8. [45] Karuppan SS, Ohlman S, Moller E. The occurrence of cytotoxic and noncomplement-fixing antibodies in the crossmatch serum of patients with early acute rejection episodes. Transplantation 1992;54:839 – 44. [46] Bray RA, Grebel HM. Allele specific HLA alloantibodies: Implication for organ allocation. Am J Transplant 2005;5(s11):p 488. [47] Gloor JM, DeGoey S, Ploeger N, Gebel H, Bray R, Moore SB, et al. Persistence of low levels of alloantibody after desensitization in crossmatch-positive livingdonor kidney transplantation. Transplantation 2004;78:221–7. [48] Jordan SC, Vo AA, Peng A, Toyoda M, Tyan D. Intravenous gammaglobulin (IVIG): A novel approach to improve transplant rates and outcomes in highly HLA-sensitized patients. Am J Transplant 2006;6:459 – 66. [49] Watanabe J, Scornik JC. IVIG and HLA antibodies. Evidence for inhibition of complement activation but not for anti-idiotypic activity. Am J Transplant 2005;5:2786 –90. [50] Gloor JM, Cosio FG, Rea DJ, Wadei HM, Winters JL, Moore SB, et al. Histologic findings one year after positive crossmatch or ABO blood group incompatible living donor kidney transplantation. Am J Transplant 2006;6:1841–7. [51] Haas M, Rahman MH, Racusen LC, Kraus ES, Bagnasco SM, Segev DL, et al. C4d and C3d staining in biopsies of ABO- and HLA-incompatible renal allografts: Correlation with histologic findings. Am J Transplant 2006;6:1829 – 40. [52] Stegall MD, Gloor J, Winters JL, Moore SB, Degoey S. A comparison of plasmapheresis versus high-dose IVIG desensitization in renal allograft

[53]

[54]

[55] [56]

[57]

[58]

[59]

617

recipients with high levels of donor specific alloantibody. Am J Transplant 2006;6:346 –51. Akalin E, Ames S, Sehgal V, Fotino M, Daly L, Murphy B, Bromberg JS. Intravenous immunoglobulin and Thymoglobulin facilitate kidney transplantation in complement-dependent cytotoxicity B-cell and flow cytometry T- or B-cell crossmatch-positive patients. Transplantation 2003;76:1444 –7. Rebellato LM, Gross U, Verbanac KM, Thomas JM. A comprehensive definition of the major antibody specificities in polyclonal rabbit antithymocyte globulin. Transplantation 1994;57:685–94. Gebel HM, Bunce M, Crosby I, Nolan JDL, Bray RA. Masking of HLA antibodies by Thymoglobulin. Hum Immunol 2005;66(Suppl 1):S12. Bearden CM, Book BK, Sidner RA, Pescovitz MD. Removal of therapeutic antilymphocyte antibodies from human sera prior to anti-human leukocyte antibody testing. J Immunol Methods 2005;300:192–9. Vieira CA, Agarwal A, Book BK, Sidner RA, Bearden CM, Gebel HM, et al. Rituximab for reduction of anti-HLA antibodies in patients awaiting renal transplantation: 1. Safety, pharmacodynamics, and pharmacokinetics. Transplantation 2004;77:542– 8. Bearden CM, Agarwal A, Book BK, Sidner RA, Gebel HM, Bray RA, Pescovitz MD. Pronase treatment facilitates alloantibody flow cytometric and cytotoxic crossmatching in the presence of rituximab. Hum Immunol 2004;65: 803–9. Book BK, Agarwal A, Milgrom AB, Bearden CM, Sidner RA, Higgins NG, Pescovitz MD. New crossmatch technique eliminates interference by humanized and chimeric monoclonal antibodies. Transplant Proc 2005;37:640 –2.