hla antibody analysis · efficient, structurally based strategy to determine hla compatibil-ity. in...

Introduction

Humoral sensitization to human leukocyteantigens (HLA) is an important barrier forsolid organ transplantation. The exposure toHLA antigens can occur during pregnancy,blood transfusion, or previous transplants.The HLA system has multiple antigen-encod-ing highly polymorphic loci. There are anti-

genic epitopes expressed on individual HLAmolecules while many others are sharedamong various HLA antigens (1,2). This com-plexity of the HLA system constitutes a sig-nificant challenge for the histocompatibilitylaboratories in order to develop sensitive andspecific methods to analyze the repertoire ofanti-HLA antibodies in transplant candidates.Furthermore, anti-HLA antibodies can appear

AbstractThe clinical relevance of humoral allosensitization has gained a lotof attention in the last few years. An increasing number of studieshave demonstrated adverse graft survival in patients who have eitherpreformed or post-transplant-developed anti-HLA antibodies. Thedetection of HLA antibodies and the specificity analysis haveevolved over time from primarily cell-based to solid-phase methods,including the availability of single-HLA antigen preparations.These technological advances combined with a better understand-ing of the epitope structure of HLA antigens have provided a moreefficient, structurally based strategy to determine HLA compatibil-ity. In conclusion, these emerging approaches can be reliably usedto predict crossmatch results in highly sensitized patients and alsoto monitor the development of clinically relevant anti-HLA anti-body after transplantation.

Key WordsHLAAlloantibodyCrossmatchTransplantationAntibody-mediated rejectionHLAMatchmaker

Adriana Zeevi, PhD200 Lothrop St, w1551 Biomedical Science Tower,Pittsburgh, PA 15261, USAE-mail: [email protected]

255© 2006Humana Press Inc.0257–277X/(Online)1559-0755/06/36/1–3:255–264/$30.00

HLA Antibody AnalysisSensitivity, Specificity, and Clinical Significance in Solid Organ Transplantation

Immunologic Research 2006;36/1–3:255–264

Adriana Zeevi1,2

Alin Girnita1

Rene Duquesnoy1,2

1Departments of Pathology and2Immunology, Thomas E Starzl,Transplantation Institute,University of Pittsburgh,Pittsburgh, PA

following transplantation in previously un-sensitized individuals and the efficiency todetect, characterize, and remove these anti-bodies may influence the long term outcomeof solid organ transplantation (3).

The detection of circulating anti-HLA anti-bodies and the specificity analysis have evolvedover time, from primarily cell-based to solid-phase immunoassays, using solubilized andrecombinant HLA molecules (4,5). Further-more, the application of HLAMatchmaker, acomputer-based algorithm to determinedonor–recipient HLA compatibility at the struc-tural level, has enhanced our ability to interpretthe patterns of anti-HLA antibodies in sensi-tized transplant recipients (6–9). In this publi-cation, we will discuss the various methods forantibody detection and specificity analysis. Wewill also address the clinical impact of anti-HLA antibodies before and after transplantationin various solid organs, a topic that has beenextensively reviewed in recent publications.

Assays to Detect Anti-HLA Antibodies

The screening for anti-HLA antibodies hasevolved from complement-dependent methods[complement dependent cytotoxicity (CDC)and CDC with anti-human globulin (AHG)] tothe more sensitive flow-based assays (10,11).

These methods are referred to as “cell-based”or “membrane-dependent” assays, as thetarget for the antibody in patient’s serum is theHLA antigen expressed on intact cell mem-brane of lymphocytes. Advances in the purifi-cation technology of HLA antigens havefacilitated the development of solid-phaseassays, whereby the HLA antigens can bebound to a solid matrix.

The “membrane-independent” assays includeELISA and flow technology using beads coatedwith HLA antigens (12,13). In Table 1 wesummarize the pros and cons of these twoapproaches. Cell-based assays are consideredclinically relevant having the HLA moleculesdisplayed in their natural configuration. How-ever, these assays may lack specificity becausepositive reactions (false positive) may occur inthe presence of non-HLA antibodies and auto-antibodies. Different cytolytic therapies appliedpre- and post-transplantation, such as anti-CD3antibody, polyclonal anti-thymoglobulin, anti-CD20 antibody, may interfere with the cell-based techniques for the detection of anti-HLAantibodies. Solid-phase methods are more spe-cific than cell-based assays because the HLAmolecules can be purified and bound to theplates (ELISA) or beads (flow) and these assaysare not influenced by lymphocytotoxicimmunosuppressive drugs. However, there are

Zeevi, Girnita, and Duquesnoy256

Table 1. Screening for Anti-HLA Antibodies

Cell based Solid phase

Sensitivity of the method CDC < CDC+AHG<Flow ELISA < Flow beadsHLA molecules Natural configuration on cell surface Isolated proteins bound on artificial

surfaceHLA antigens HLA phenotypes Pooled HLA antigens, phenotypes

and single antigensFalse-positive reactions Non HLA-specific antibody Reactions with cryptic epitopes on

denatured HLA molecules.False negative reactions Antibody level below detection Loss of epitope expression on isolated

molecules

CDC = Complement-dependent cytotoxicity; AHG = Anti-human immunoglobulin.

potential problems with the isolation of HLAantigens, whereby new cryptic molecules fromthe denatured HLA antigen can be exposed andbind to patient’s sera (false-positive reactions),or antigenic epitopes may be lost as a conse-quence of changes in configuration of the iso-lated HLA molecules (false-negative reactions).The analysis of antibody specificity hasimproved with the availability of single anti-gen–coated plates/beads. For the highly sensi-tized patients awaiting transplant, it was verydifficult to identify their anti-HLA antibodyspecificity and to find a suitable donor. Cur-rently, with the new reagents, it is possible todefine the HLA antigens that are consideredunacceptable and/or acceptable mismatches.Furthermore, knowledge of patient’s history ofsensitizing events such as previous grafts, bloodtransfusions, or pregnancy is required for thelaboratory assessment of HLA sensitization.The characterization of anti-HLA antibodyshould be based on a combination of cell-basedand solid-phase assays with single antigens.

Interpretation of Serum Screening Results

The so-called PRA (Panel-Reactive Anti-body) represents a semiquantitative estimateof the degree of HLA sensitization. It is cal-culated as the percentage of an HLA panelthat reacts with a serum. Patients with >80%PRA are considered highly sensitized and forthem it is difficult to find crossmatch-negativedonors. The accumulation of highly sensitizedpatients on transplant waiting lists representsa growing problem.

The analysis of serum reactivity of transplantcandidates has two goals. Most commonly usedis the identification of unacceptable HLA anti-gens that should be avoided on donor organs.This system is designed to identify donors whomust be excluded, but it does not necessarilymean that all other HLA antigens would becompatible for a patient. The other goal is to

determine HLA antigens that are acceptablemismatches. This strategy represents a directapproach of finding a compatible donor for asensitized patient (14,15).

The analysis of serum reactivity patternswith HLA phenotyped panels in cell-basedand solid-phase assays is primarily done with2×2 table statistical methods such as chisquare to identify antigens and epitopes withsignificant correlations. Unfortunately, thismethod is of limited value for >80% PRAsera. The use of single HLA antigens inELISA and Luminex assays permits a betterinterpretation of antibody reactivity patterns.An important consideration is that each HLAantigen carries multiple epitopes that can bestructurally defined by amino acid residues inpolymorphic positions of the HLA molecule.Stereo chemical modeling of crystallizedHLA antigens has visualized these ratherextreme structural polymorphisms. Figure 1shows examples of three class I molecules:HLA-A2, HLA-B27, and HLA-Cw3.

The molecular surface around the boundpeptide (see top view) has similar numbers ofexposed polymorphic positions on the α1helices of HLA-A and HLA-B antigens butmore polymorphic positions are visible on theα2 helices of HLA-A antigens. The α helicesof HLA-C antigens have much fewer poly-morphic positions.

In contrast, HLA-C antigens have morepolymorphic positions in the membrane-prox-imal region, which becomes visible upon sideviewing. HLA-A antigens have also moresurface-exposed polymorphic positions in thatregion than HLA-B antigens. It should benoted that the sequence positions in the mem-brane-proximal domain of HLA-B are allmonomorphic.

Class II HLA antigens have similarly com-plex structural polymorphisms (not shown).This applies to all DRB, DQB, and DPBchains. DQA chains have more structural

Anti-HLA Alloantibody in Transplantation

257

polymorphisms than DPA chains whereasDRA chains are primarily monomorphic.

Considering the high number of HLA anti-gens (and alleles) and their extensive poly-

morphisms, one can expect that HLA anti-body formation in transplant patients is exten-sive and complex. A better understanding ofthe epitope structure of HLA antigens is


Fig. 1. Polymorphic residues on class I molecules controlled by HLA-A, B, and C loci (β2M = β2microglobulin).

important not only for the characterization ofHLA-specific antibodies but also will permita more efficient, structurally based strategy todetermine HLA compatibility.

HLAMatchmaker is a matching programthat considers the structural basis of epitopeson class I HLA antigens (7). Each HLA anti-gen can be viewed as a string of shortsequences (triplets) involving polymorphicamino acid residues in antibody-accessiblepositions; they are considered key elements ofepitopes that can induce the formation of spe-cific antibodies. The patient’s HLA phenotyperepresents the repertoire of self-triplets towhich no antibodies can be made and HLA-Matchmaker determines for each mismatchedHLA antigen, which triplets in correspondingsequence positions are different. HLAMatch-maker-based matching improves transplantoutcome (9,16,17), and is useful in serumanalysis and the identification of acceptablemismatches for alloimmunized kidney trans-plant candidates (6,15,18–23) and refractorythrombocytopenic patients requiring matchedplatelet transfusions (24).

The number of amino acid triplet differencesbetween patient and donor correlates with thedevelopment of anti-HLA antibody followingpregnancy and kidney transplantation (25).Certain patients become highly sensitized fol-lowing exposure to a single mismatched HLAantigen. The following two cases illustratehow HLAMatchmaker can explain this.

The first case was a patient who afterremoval of a rejected kidney graft 7 mo post-transplant developed a serum PRA of about90% due to class I antibodies (22). This kidneywas a one antigen mismatch, namely, HLA-B13, but patient’s serum reacted not only withHLA-B13 but also with a large number ofother HLA-A and HLA-B antigens. An HLA-Matchmaker-based analysis showed antibodyspecificity to the 144tQl triplet unique toHLA-B13 and the 76En, 80rTa, 82aLr triplets

shared between HLA-B13 and other HLAantigens such as HLA-A9, HLA-B17, HLA-B27, and many more. (The triplet notationsystem uses the amino acid letter code and thenumber represents the sequence position of theresidue in capital letters.) The76En, 80rTa,82aLr carrying antigens must be consideredunacceptable mismatches although the patientmight have never been exposed to them. Thereason why they become unacceptable wasthat they share one or more epitopes with theimmunizing HLA-B13.

The second example is a high anti-class IIantibody activity following sensitization to aone HLA-DR antigen mismatch. This patienttyped homozygous for HLA-DR7 and hadrejected a kidney transplant with a HLA-DR11 mismatch. Patient’s serum reacted withall DRB1 antigens except HLA-DR7. HLA-Matchmaker identified a mismatched triplet14ER on HLA-DR11 that is shared with allDRB1 antigens except HLA-DR7, which has14QK. Thus, the high reactivity of thispatient’s serum might be due to antibodies toa single epitope defined in this case by 14ER.This conclusion is consistent with descrip-tions of monoclonal antibodies reacting withall DRB1 antigens except HLA-DR7 (26).Exposure to single HLA-DQ mismatches mayalso lead to antibody reactivity to allDQB1antigens except self-DQB1 and corre-sponding structurally defined DQB1 epitopescan readily be identified.

It should be noted that HLA antibodyresponses are generally restricted to a limitednumber of epitopes. High PRA sera reflect thepresence of antibodies against high-frequencyepitopes. The HLAMatchmaker-based inter-pretation of serum reactivity incorporatespatient’s HLA type determined preferably byDNA methods at the 4-digit allele level. HLAinformation of the immunizer (i.e., a previoustransplant) will identify structurally definedepitopes the patient has been exposed to. This


259

facilitates the interpretation of serum screen-ing results and the determination of mismatchacceptability for prospective donors.

Clinical Impact of Anti-HLA Antibodies

Humoral rejection is emerging as a leadingcause of graft failure and is associated with allforms of allograft rejection: hyperacute,acute, and chronic (4, 27–29). Although cir-culating antibodies were found in patientsrejecting their allografts, the lack of histolog-ical evidence of antibody-mediated processhampered the diagnosis of antibody-mediatedrejection (AMR). Detection of complementactivation in tissue by staining for C4d and therecognition of other clinical features, such asgraft dysfunction that does not respond tostandard therapies, greatly facilitated thediagnosis of AMR (28,30,31).

HLA antibodies are associated with acuteand chronic rejection in kidney, kidney–pan-creas, heart, and lung transplant recipients andare present after almost all kidney failures(32,33). Donor-specific HLA antibodies notfound prior to graft removal becamedetectable after transplant nephrectomy inmost patients (22). This study supports theconcept that the allograft may remove thehigh affinity circulating donor-specific anti-bodies and upon re-transplantation those HLAantigens recognized by the antibodies shouldbe avoided. Furthermore, the serum analysisperformed with the HLAMatchmaker pro-gram can identify additional HLA antigensthat may share antibody-reactive epitopeswith the immunizing donor and thereforeshould also be avoided (22).

The deleterious effect of anti-HLA anti-bodies developed post-transplantation wasaddressed in a large collaborative study of2231 kidney recipients from 23 centers fol-lowed for 2 yr (34). Among those patients whowere pre-Tx antibody negative and developed

de novo HLA antibodies, 16.7% (n=233 anti-body positive) failed within the 2-yr follow-up, while only 6.5% of antibody negativepatients (n=1331) lost their grafts. When thepatients were further divided by the serum cre-atinine levels, at the time of antibody testing,a progressive decline was observed in patientswho had antibody and increased serum creati-nine (34).

The incidence of allosensitized patients oncardiac transplant waiting lists is on the rise,owing to the use of left ventricular assistdevices, blood transfusions, and the increas-ing number of re-transplants. Cardiac recip-ients with a history of sensitization have anincreased incidence of antibody-mediatedrejection. AMR has been shown to predisposeheart transplant recipients to coronary vas-culopathy (29,35,36). Furthermore, patientswho develop AMR after cardiac transplanta-tion progress to transplant-associated coro-nary artery disease earlier and at increasedfrequency compared with controls (37,38).

AMR in the heart, like the kidney, mayoccur in combination with cellular rejection.The incidence of AMR in biopsies with cel-lular rejection has been reported to be 23%,while the prevalence of AMR without cellularrejection is about 15% (35).

We studied the role of pre-formed and denovo developed anti-HLA antibodies in lungtransplantation (39). HLA antibodies detectedprimarily by solid-phase ELISA were associ-ated with severe forms of acute allograftrejection that required multiple treatments(persistent and recurrent acute perivascularrejection, ACR-PR, Table 2). As depicted inTable 2, the relative risk for ACR-PR inpatients with circulating HLA antibodies, wasfivefold higher than in patients who did notdevelop anti-HLA antibodies. Moreover,there was no influence of the immunosup-pressive protocol on the frequency of de novoanti-HLA antibody production.


We, and others, have observed a significantassociation between anti-HLA antibodies andchronic lung allograft dysfunction referred toas bronchiolitis obliterans syndrome (BOS)(40,41). A multivariate risk factor analysis forthe development of BOS in 51 lung transplantrecipients followed for 4.2 + 1.6 yr, showedsignificant associations between the develop-ment of de novo antibodies and lymphocyticbronchiolitis, which is considered the airwayrejection form (Fig 2), as well as betweenantibodies and BOS (40). The HLA antibod-ies preceded the development of BOS bymore than 1 yr, and both anti-class I and anti-class II HLA antibodies were associated withworse outcomes (40).

The hallmark of complement activationdue to antibody-mediated processes in allo-graft tissue is C4d deposition (30). Specificimmunostaining patterns are considered whencontinuous, linear, subendothelial depositionis detected in microcirculation (capillaries,arterioles, and venules) (42). Lung transplantrecipients who rejected their allografts andalso had circulating antibodies, had higherfrequency of C4d deposition than histologicalmatched patients without anti-HLA antibod-ies (42). Circulating HLA antibodies in lungtransplant recipients during ACR episodeswere also associated with increased levels ofsoluble C4d in bronchoalveolar lavage fluids(43). These results support the notion thathumoral immunity may contribute to allograftdysfunction, and improved methods of anti-

body detection pre- and post-transplantation,combined with complement deposition stain-ing and novel treatment protocols, should beincorporated in the clinical management oftransplant recipients.

Virtual Crossmatch

The impact of sensitization on transplantoutcome has been recognized since the firstreports of antibody-mediated hyperacuterejection in renal transplant recipients. Sensi-tized patients, those with PRAs > 10%, cur-rently comprise 33% of the patients on thewaiting list, and the proportion of sensitizedwomen (PRA ≥ 10) is twice that of men (3).There is also disparity in rates of sensitization


261

Table 2. Increased Risk of Persistent Lung Allograft Rejection in Patients with Circulating Anti-HLA Antibodies

HLA antibody No HLA antibody

Therapeutic Protocol N ACR-PR N ACR-PR p< Relative risk

Triple drug (CsA, Ster, Aza) 12 11/12 (90%) 35 10/35 (34%) 0.005 5Pre-transplant thymoglobulin 14 10/14 (71%) 23 3/23 (13%) 0.001 5.5

ACR-PR = Persistent / recurrent acute cellular rejection; CsA = Cyclosporine A; Ster = steroids; Aza = Azathyoprine.

Fig. 2. Higher prevalence of lymphocytic bron-chiolitis (LBB) in lung transplanted patients withanti-HLA antibodies (HLA-Ab).

among different racial groups with the highestpercentage of highly sensitized patients (PRA≥80) observed among African Americans (3).

The level of sensitization significantly pro-longs waiting time and prospective CXM haveto be performed pre-transplant to avoid thedeleterious effects of antibody-mediated rejec-tion. Owing to the need to minimize organischemia time, especially for thoracic organs,pre-transplant CXMs are performed withdonor samples prior to organ recovery. Thisapproach limits the pool of donors for highlysensitized thoracic transplant candidatesbecause it precludes the use of donors obtainedat a distant site.

With the application of newer, specific, andsensitive techniques for detection and charac-terization of anti-HLA antibodies, the clinicallaboratory can determine an appropriate donorwithout the actual CXM test. By using a virtualCXM, we exclude donors that express thoseHLA antigens, or epitopes that are recognizedby the patient’s antibody (unacceptable anti-gens), while donors that carry the acceptableHLA antigens can be considered. The use ofthe virtual CXM has improved the ability totransplant sensitized heart and lung transplantrecipients, and it resulted in a significantdecrease in their waiting list time (44,45).

We also implemented the virtual CXM atour institution for sensitized thoracic trans-plant candidates. The pre-transplant antibodyscreening and specificity analysis is performedwith solid-phase assays including single HLAantigen preparations. Following transplanta-tion, the next working day, the actual CXM isdone to confirm the negative virtual CXM.

Based on this approach, two sensitized lungtransplant recipients were successfully trans-planted. Both patients had multiple sera samplesfor antibody analysis collected over a period of6 mo pre-transplantation (6, 4, 2 mo), includingsera obtained within 2 wk of the allograft.Based on the antibody reactivity pattern we

could identify the unacceptable antigens. Fur-thermore, the application of the HLAMatch-maker program provided additional informationregarding the acceptable HLA mismatches, andincreased our predictability of a negative virtualCXM. Although both patients were sensitized,we could find a compatible donor in a timelyfashion so they could proceed to transplantation.Both patients had an uneventful post-transplantcourse and did not develop de novo anti-DSAwithin the 2–3 mo of follow-up.

This virtual CXM relies on complete infor-mation of sensitizing events and patient history.In any case whereby the transplant candidatehad a sensitizing event, such as blood transfu-sion or exposure to left ventricular device(heart transplant candidates), the antibodyanalysis needs to be reassessed. The predictivevalue of the virtual CXM is highly dependenton the accuracy of the latest antibody analysis.

Summary

There is good evidence that pre-formed andde novo production of anti-HLA class I andclass II antibodies contribute to graft deteriora-tion at all times after transplantation. Further-more, recent studies suggest that monitoring foranti-HLA antibodies post-transplantation isprognostic of allograft outcome and can providea useful measure of therapeutic efficiency.Combination of solid-phase and cell-basedmethods should be used to identify low levelsand clinically significant anti-HLA antibodies.The analysis of antibody reactivity patternsincorporating the HLAMatchmaker program isa valuable tool to determine the acceptable mis-matches and provide important information ofdonor compatibility for highly sensitized trans-plant candidates.

Acknowledgments

This work was partialy supported by grantsAI 55933 and 1P50 HL074732.


References1. Rodey GE, Fuller TC: Public epitopes and the antigenic

structure of the HLA molecules. Crit Rev Immunol1987; 7: 229–267.

2. Konoeda Y, Terasaki PI, Wakisaka A, Park MS, MickeyMR: Public determinants of HLA indicated by preg-nancy antibodies. Transplantation 1986; 41: 253–259.

3. Zachary AA, Montgomery RA, Leffell MS: Desensiti-zation protocols improving access and outcome in trans-plantation. Clin Appl Immunol Rev 2005; 5: 373–395.

4. Terasaki PI: Humoral theory of transplantation. Am JTransplant 2003; 3: 665–673.

5. Bray RA, Nickerson PW, Kerman RH, Gebel HM: Evo-lution of HLA antibody detection: technology emulatingbiology. Immunol Res 2004; 29 41–54.

6. Duquesnoy RJ, Marrari M: HLAMatchmaker: a molec-ularly based algorithm for histocompatibility determina-tion. II. Verification of the algorithm and determinationof the relative immunogenicity of amino acid triplet-defined epitopes. Hum Immunol 2002; 63: 353–363.

7. Duquesnoy RJ: HLAMatchmaker: a molecularly basedalgorithm for histocompatibility determination. I.Description of the algorithm. Hum Immunol 2002; 63:339–352.

8. Duquesnoy RJ, Howe J,Takemoto S: HLAmatchmaker:a molecularly based algorithm for histocompatibilitydetermination. IV. An alternative strategy to increase thenumber of compatible donors for highly sensitizedpatients. Transplantation 2003; 75: 889–897.

9. Duquesnoy RJ, Takemoto S, de Lange P, et al: HLA-matchmaker: a molecularly based algorithm for histo-compatibility determination. III. Effect of matching atthe HLA-A,B amino acid triplet level on kidney trans-plant survival. Transplantation 2003; 75: 884–889.

10. Amos DB, Bashir H, Boyle W, MacQueen M,Tiilikainen A: A simple micro cytotoxicity test. Trans-plantation 1969; 7: 220–223.

11. Zachary AA, Klingman L, Thorne N, Smerglia AR,Teresi GA: Variations of the lymphocytotoxicity test. Anevaluation of sensitivity and specificity. Transplantation1995; 60: 498–503.

12. Kao KJ, Scornik JC, Small SJ: Enzyme-linkedimmunoassay for anti-HLA antibodies—an alternativeto panel studies by lymphocytotoxicity. Transplantation1993; 55: 192–196.

13. Pei R, Wang G, Tarsitani C, et al: Simultaneous HLAclass I and class II antibodies screening with flowcytometry. Hum Immunol 1998; 59: 313–322.

14. Claas FH, Gijbels Y, von Veen A, et al: Selection ofcross-match negative HLA-A and/or -B mismatcheddonors for highly sensitized patients. Transplant Proc1989; 21(1 Pt 1): 665–666.

15. Claas FH, Witvliet MD, Duquesnoy RJ, Persijn GG,Doxiadis II: The acceptable mismatch program as a fasttool for highly sensitized patients awaiting a cadaverickidney transplantation: short waiting time and excellentgraft outcome. Transplantation 2004; 78: 190–193.

16. Bohringer D, Reinhard T, Duquesnoy RJ, et al: Benefi-cial effect of matching at the HLA-A and -B amino-acid

triplet level on rejection-free clear graft survival in pen-etrating keratoplasty. Transplantation 2004; 77: 417–421.

17. Laux G, Mytilineos J, Opelz G: Critical evaluation of theamino acid triplet-epitope matching concept in cadaverkidney transplantation. Transplantation 2004; 77:902–907.

18. Lobashevsky AL, Senkbeil RW, Shoaf JL, et al: Thenumber of amino acid residues mismatches correlateswith flow cytometry crossmatching results in high PRArenal patients. Hum Immunol 2002; 63: 364–374.

19. Duquesnoy RJ, Witvliet M, Doxiadis II, de Fijter H,Claas FH: HLAMatchmaker-based strategy to identifyacceptable HLA class I mismatches for highly sensitizedkidney transplant candidates. Transpl Int 2004; 17:22–30.

20. Iniotaki-Theodoraki A, Kalogeropoulou E, ApostolakiM, Doxiadis IN, Stavropoulos-Giokas C: Humoral sen-sitization against rejected grafts: specific antibodies tograft immunogenic amino acid triplets. Transplant Proc2004; 36: 1728–1731.

21. Varnavidou-Nicolaidou A, Doxiadis II, Iniotaki-Theodor-aki A, Patargias T, Stavropoulos-Giokas C, KyriakidesGK: HLA class I donor-specific triplet antibodies detectedafter renal transplantation. Transplant Proc 2004; 36:1732–1734.

22. Adeyi OA, Girnita AL, Howe J, et al: Serum analysisafter transplant nephrectomy reveals restricted antibodyspecificity patterns against structurally defined HLAclass I mismatches. Transpl Immunol 2005; 14: 53–62.

23. Doxiadis II, Duquesnoy RJ, Claas FH: Extendingoptions for highly sensitized patients to receive a suitablekidney graft. Curr Opin Immunol 2005; 17: 536–540.

24. Nambiar A, Duquesnoy RJ, Adams S, et al: HLA-Matchmaker-driven analysis of responses to HLA-typedplatelet transfusions in alloimmunized thrombocy-topenic patients. Blood 2006; 107: 1680–1687.

25. Dankers MK, Witvliet MD, Roelen DL, et al: Thenumber of amino acid triplet differences between patientand donor is predictive for the antibody reactivity againstmismatched human leukocyte antigens. Transplantation2004; 77: 1236–1239.

26. Bodmer JEA: HLA-D region monoclonal antibodies, inAlbert MME, Mayr W (eds.). Histocompatibility Test-ing. Berlin, Springer-Verlag, 1984, pp. 217–236.

27. McKenna RM, Takemoto SK, Terasaki PI: Anti-HLAantibodies after solid organ transplantation. Transplan-tation 2000; 69: 319–326.

28. Takemoto SK, Zeevi A, Feng S, et al: National conferenceto assess antibody-mediated rejection in solid organtransplantation. Am J Transplant 2004; 4: 1033–1041.

29. Girnita AL, Webber SA, Zeevi A: Anti-HLA alloanti-bodies in pediatric solid organ transplantation. PediatrTransplant 2006; 10: 146–153.

30. Feucht HE, Schneeberger H, Hillebrand G, et al: Capil-lary deposition of C4d complement fragment and earlyrenal graft loss. Kidney Int 1993; 43: 1333–1338.

31. Nickeleit V, Zeiler M, Gudat F, Thiel G, Mihatsch MJ:Detection of the complement degradation product C4din renal allografts: diagnostic and therapeutic implica-tions. J Am Soc Nephrol 2002; 13: 242–251.


263

32. Terasaki PI, Ozawa M: Predicting kidney graft failure byHLA antibodies: a prospective trial. Am J Transplant2004; 4: 438–443.

33. Lee PC, Terasaki P(I), Takemoto SK, et al: All chronicrejection failures of kidney transplants were preceded bythe development of HLA antibodies. Transplantation2002; 74: 1192–1194.

34. Terasaki PI, Ozawa M: Predictive value of HLA anti-bodies and serum creatinine in chronic rejection: resultsof a 2-year prospective trial. Transplantation 2005; 80:1194–1197.

35. Reed EF, Demetris AJ, Hammond E, et al: Acute anti-body-mediated rejection of cardiac transplants. J HeartLung Transplant 2006; 25: 153–159.

36. Reinsmoen NL, Nelson K, Zeevi A: Anti-HLA antibodyanalysis and crossmatching in heart and lung transplan-tation. Transpl Immunol 2004; 13: 63–71.

37. Michaels PJ, Espejo ML, Kobashigawa J, et al: Humoralrejection in cardiac transplantation: risk factors, hemo-dynamic consequences and relationship to transplantcoronary artery disease. J Heart Lung Transplant 2003;22: 58–69.

38. Itescu S, Tung TC, Burke EM, et al: An immunologicalalgorithm to predict risk of high-grade rejection in car-diac transplant recipients. Lancet 1998; 352: 263–270.

39. Girnita AL, McCurry KR, Iacono AT, et al: HLA-spe-cific antibodies are associated with high-grade and per-

sistent-recurrent lung allograft acute rejection. J HeartLung Transplant 2004; 23: 1135–1141.

40. Girnita AL, Duquesnoy R, Yousem SA, et al: HLA-spe-cific antibodies are risk factors for lymphocytic bron-chiolitis and chronic lung allograft dysfunction. Am JTransplant 2005; 5: 131–138.

41. Jaramillo A, Smith MA, Phelan D, et al: Development ofELISA-detected anti-HLA antibodies precedes thedevelopment of bronchiolitis obliterans syndrome andcorrelates with progressive decline in pulmonary func-tion after lung transplantation. Transplantation 1999; 67:1155–1161.

42. Ionescu DN, Girnita AL, Zeevi A, et al: C4d depositionin lung allografts is associated with circulating anti-HLAalloantibody. Transpl Immunol 2005; 15: 63–68.

43. Miller GG, Destarac L, Zeevi A, et al: Acute humoralrejection of human lung allografts and elevation of C4din bronchoalveolar lavage fluid. Am J Transplant 2004;4: 1323–1330.

44. Zangwill SD, Ellis TM, Zlotocha J, et al: The virtualcrossmatch—a screening tool for sensitized pediatricheart transplant recipients. Pediatr Transplant 2006; 10:38–41.

45. Appel JZ 3rd, Hartwig MG, Cantu E 3rd, et al: Role offlow cytometry to define unacceptable HLA antigens inlung transplant recipients with HLA-specific antibodies.Transplantation 2006; 81: 1049–1057.


hla antibody analysis · efficient, structurally based strategy to determine hla compatibil-ity. in...

Documents