Methodologies for Anti-HLA Antibody Screening in Patients Awaiting Kidney Transplant: A Comparative Study

Transcription

1 ARTiCle Methodologies for Anti-HLA Antibody Screening in Patients Awaiting Kidney Transplant: A Comparative Study Pellegrino B. Minucci, 1 Vincenzo Grimaldi, 1 Amelia Casamassimi, 1 Francesco Cacciatore, 2 Linda Sommese, 1 Antonietta Picascia, 1 Marianna Resse, 1 Chiara Sabia, 1 Adolfo Russo, 1 Claudio Napoli 1 Abstract Objectives: The relevance of anti-hla antibodies in patients awaiting kidney transplants is well recognized. During the past 40 years, kidney transplant candidates have been tested for these antibodies, and the choice of the detection assay has become essential. Recently, the pioneer method, the complement-dependent cytotoxicity, has been integrated but has not been replaced by more-sensitive solid-phase assays, such as the enzyme-linked immunosorbent assay and the beadbased technology (ie, flow cytometry: FlowPRA, and FlowAnalyzer: Luminex). Materials and Methods: We compared the sensitivity and antibody specificity of these 4 techniques for detecting panel-reactive antibodies in a population of 101 consecutive patients awaiting a renal transplant (which had already resulted positive in a prescreening analysis). Results: Sera positive for class I and class II antibodies were 62 and 90 as assessed by the complement-dependent cytotoxicity method, 76 and 58 by using enzyme-linked immunosorbent assay, 83 and 65 with Flow-panel-reactive antibodies, and 90 and 79 by Luminex. Luminex From the 1 U.O.C. of Immunohematology and Immunology of Transplantation, CRT-AORN Cardarelli, and Department of General Pathology, Second University of Naples, Naples, IT; and the 2 Cardiovascular Rehabilitation, Salvatore Maugeri Foundation, IRCCS, Institute of Telese, 82037, Benevento, IT Acknowledgements: This study was presented at 25th European Immunogenetics and Histocompatibility Conference, Prague, May 4-7, We thank Francesco Paolo De Luca for technical assistance. We are grateful to the President of Regione Campania and the Subhead of Health for the support of our Institution. Moreover, we acknowledge the General Manager of the Azienda Ospedaliera Universitaria (AOU) of the Second University of Naples for his support. Address reprint requests to: Dr. Pellegrino Biagio Minucci, Department of General Pathology, U.O.C. Division of Immunohematology, Transfusion Medicine and Transplant Immunology (SIMT), Regional Reference Laboratory of Transplant Immunology (LIT), Second University of Naples, Via Costantinopoli, 16, 80138, Naples (ITALY) Phone: Fax: pellegrinob.minucci@unina2.it Experimental and Clinical Transplantation (2011) 6: gave more positive scores than the others for class I HLA antibodies, whereas complement-dependent cytotoxicity revealed more positives for those of class II. Conclusions: Although Luminex appears more efficient among these assays, our results indicate that use of multiple methods is still the best approach for characterizing the immunologic status of these patients. Key words: CDC, ELISA, Flow-PRA, Luminex, HLA antibody. Class I and class II antibodies Introduction A key challenge in renal transplant is identifying acceptable kidney donors for individuals with renal failure. The importance of anti-hla antibodies in the outcome of these transplants has been recognized for over 40 years, 1, 2 especially to avoid hyperacute rejection. 3 Patel and Terasaki established that presence of recipient antibodies to antigens expressed on donor white blood cells was a major risk factor for immediate graft loss. 4 Since then, transplant candidates have been tested for these antibodies, and the assay used for detection has become an essential point. 5 Cell-based crossmatch assays were developed in which recipient sera were incubated with donor lymphocytes in the presence of the complement, and cell lysis was used to measure donor-specific antibodies. Different histocompatibility laboratories have modified this procedure to the preferred (higher/lower) sensitivity, so that there is no standard protocol that is universally adopted. Despite this variability in testing, complement-dependent cytotoxicity crossmatching has reduced the incidence of hyperacute rejection. 3-5 Copyright Başkent University 2011 Printed in Turkey. All Rights Reserved.

2 382 Pellegrino B. Minucci et al /Experimental and Clinical Transplantation (2011) 6: Exp Clin Transplant Early rejection episodes still occur, thereby stimulating the search of novel assays with greater sensitivity. 5 The transplant literature up to the year 2000 reveals many reports with important differences in protocols and conclusions regarding the use of different crossmatch assays. The complementdependent cytotoxicity method (CDC) has been used for over 40 years since its introduction 6, 7 has been recently integrated by flow cytometric techniques or more-sensitive solid-phase assays, 8, 9 such as enzymelinked immunosorbent assay (ELISA) and bead-based technology 10 (including Flow Cytometry [Flow-PRA - FL] and Flow Analyzer [Luminex LX]). Introduction of these techniques into clinical practice has overcome the limited sensitivity and specificity of the CDC assay. 11 These technologic advances, combined with a better understanding of all the epitopes of HLA antigens, have provided a more-efficient approach to determine HLA compatibility. These promising approaches may be used to predict crossmatches among highly sensitized patients and to monitor their development of clinically relevant anti-hla antibodies after transplant. 4, 12 In patients on a waiting list for an organ transplant, the sensitivity of a screening test for detecting class I and class II HLA antibodies plays a crucial role, especially at the initial stage of the patient enrollment. This study sought to define a better assay for detecting HLA antibodies by comparing 4 techniques: CDC, ELISA, FlowPRA, and Luminex. To do this, we analyzed 101 positive samples from a prescreening performed on a population of patients awaiting a renal transplant. Materials and Methods Patients Samples were derived from candidates for renal transplant at the Division of Immunohematology and Immunology of Transplantation, at the Second University of Naples between 2007 and Patients were routinely screened at least every 3 months for HLA antibodies with ELISA and Luminex methods. This study has been done on sera obtained from a single blood collection of a group of 632 consecutive patients awaiting a kidney transplant. About 80% of analyzed sera (506/632) were negative to HLA antibody screening with the above-mentioned procedures. The described techniques were tested on 101/126 positive sera where there was enough sample amount to analyze them, and on the same serum stock, to reduce at minimum all the possible variables (antibody title, cryoconservation). This examined group included 50 men and 51 women (age range, 18 to 75) (64/101 had previously received a transplant, 8 had pregnancies, and 8 patients had been transfused). Three out of 101 positive patients displayed Lupus, and 2 of them had an IgA nephropathy. All the sera were collected and were stored at -80 C until analysis. Methods Complement-dependent cytotoxicity method Complement-dependent cytotoxicity was performed through a whole lymphocyte population consisting of a panel of 50 cells from HLA class I and class II phenotyped blood and/or bone marrow donors. The CDC protocol used was the standard method with long incubation. 6 Whole cells were isolated with Dynabeads HLA class I and II. The trays were read using a fluorescent microscope connected to Lambda Scan Software. By using this CDC technique, we have screened for classes I (CD8 + lymphocytes) and II (CD19 + lymphocytes) and determined antibody specificity and calculated the percentage positive panel reactive antibody (PRA). Enzyme-linked immunoabsorbent assay method Enzyme-linked immunosorbent assay was performed using automatic and commercial enzyme-linked immunosorbent assay for detecting or indentifying HLA antibodies after the manufacturer s instructions (AbScreen HLA class I and II and AbIdent class I and II QuickStep, Biorad). We used microtiter plates, coated with HLA class I and II antigens, highly purified from a large pool of human platelets. The antibody-antigen complex was detected using a specific enzyme-labeled antibody directed against human IgG (conjugate) and bound antibodies were revealed by adding a chromogenic substrate. The Biorad software analyzed the results. Each test run included positive and negative controls. A sample was considered as positive for the presence of HLA antibodies when the optical density average value was at least twice that of negative controls (cutoff). FlowPRA method The FlowPRA screening test provides precalibrated reagents for rapid flow cytometry detection of PRA in human sera. Serum samples were tested for IgG anti-hla PRA using FlowPRA class I and class II

3 Pellegrino B. Minucci et al /Experimental and Clinical Transplantation (2011) 6: screening beads (One Lambda, Canoga Park, CA, USA), which consists of a pool of 30 different bead populations, each coated with purified HLA class I antigen derived from a single cell line. Negative (FL- NC, One Lambda) and positive (FL1- and FL2-PC, One Lambda) control sera were included in each assay. Flow-cytometry analysis was performed using FACSCanto and Diva software (Becton Dickinson, Mountain View, CA, USA). At least beads were collected from each tube. Two gates were set on the dot plot of SSC (side scatter) versus FL2 (fluorescence 2) to analyze class I (FL2-negative particles) and class II (FL2 high-fluorescence particles) beads separately. The percentage of beads that shifted to the right cutoff set on the FL1 histogram (anti-human IgG) of the negative control serum represented the amount of class I and/or class II antibodies (% PRA) of the sample test. Luminex method Luminex uses microbeads coated with purified class I or class II antigens. LABSscreen products (LSM12, LS1PRA, LS2PRA - One Lambda) uses the lambda Array Beads Multi-Analyte System, with the LABScan 100/200 flow analyzer, for data acquisition and analysis. This flow analyzer has 2 lasers: the green 532 nm laser, and the red 633 nm laser. The combination of the 2 signals indicates both the binding of HLA-antibodies and specific HLA molecules. Luminex was performed according to the manufacturers instructions. The One Lambda negative control serum was used in parallel; plates were incubated with phycoerythrin-conjugated goat anti-human IgG. In the screening test for class I and II (LSM12), a cutoff value of 1.2 was considered, as proposed by the supplier of the negative control. Specifically, a ratio 1.5 defined the positive sample (corresponding to score 8 of cytotoxicity), a ratio between 1.2 and 1.5 sample was undetermined (corresponding to score 4 of cytotoxicity), and a ratio below 1.2 labeled the analyzed sample as negative (corresponding to score 1 of cytotoxicity). In the analysis of test identifier (LS1PRA and LS2PRA), the positivity or negativity of a serum was determined according to the following values: i bead iii beads Result < 500 < 300 Negative > 500 > 300 Positive > 500 < 300 < 500 > 300 For attribution of HLA antibody specificity (LS1A04 and LS2A01), those with a mean of FL > 100 were considered specific. Data were analyzed with HLA Visual Software (One Lambda). Results We tested a group of sera obtained from a collection of 632 consecutive patients awaiting a kidney transplant. Patients were routinely screened for HLA antibodies every 3 months at least with ELISA and Luminex assays (as described in the Methods section), and about 80% of sera (506/632) were negative to HLA antibody screening with these procedures. The complete study was performed on 101/126 positive sera (see the Methods section). As shown in Table 1, sera positives for class I and class II antibodies were 62 and 90 with the CDC method, 76 and 58 using the ELISA method, 83 and 65 with Flow-PRA, and 90 and 79 by Luminex. Thus, if considering these absolute values, the Luminex method gave more positive scores than did the others for class I HLA antibodies, whereas CDC revealed more positives for class II. Table 1. Comparison of class I and II. Results for the 101 sera using the 4 techniques. Class I HLA antibodies Class II HLA antibodies Positive Negative Positive Negative Flow-PRA ELISA Luminex CDC Abbreviations: CDC, complement-dependent cytotoxicity; ELISA, Enzymelinked immunosorbent assay; Flow-PRA, Flow cytometric method for panel reactive antibody; HLA, human leucocyte antigen; Luminex, Flow Analyzer bead-based technology Data comparison of all the techniques is reported in Table 2. When considering class I antibodies, Luminex and Flow-PRA revealed the higher number of positive sera with both methods (82 positive sera). Taking into account discordant values, the Luminex assay is the technique giving the more positive results (8 Luminex vs 1 FlowPRA [89%]; 16 Luminex vs 2 ELISA [89.9%]; 34 Luminex vs 6 CDC [85%]). When analyzing the number of negative data, Luminex and CDC together showed the lower values (5 negative sera). Conversely, for class II antibodies, Luminex and CDC gave 72 positive sera and 4 negative sera with both methods. When observing discordant values, we found 14 Luminex versus 0 FlowPRA (100%), 22 Luminex versus 1 ELISA (95.7%), and 7 Luminex versus 18 CDC (28%)

4 384 Pellegrino B. Minucci et al /Experimental and Clinical Transplantation (2011) 6: Exp Clin Transplant (Table 2). Concordance among data obtained with the 4 methods was calculated as the sample number giving the same result (both positives and negatives). Concordance between positive results obtained with all the used procedures was 50 for class I antibodies and 54 for class II antibodies (Table 3). Concordance between negative results in all techniques was found for 4 sera both classes (Table 3). Discrepancy data were found for 47 sera (46.5%) for class I antibodies and 43 for class II antibodies (42.6%) (Table 3). Indeed, 12, 9, and 26 sera were concordant for 1, 2, or 3 assays for class I antibodies. Similarly 19, 11, and 13 sera showed concordance for class II antibodies. To recognize donor specific antibodies, we analyzed only the sera from 38 individuals already transplanted and with known donor HLA typing. Particularly, 37 sera displayed mismatches for class I antibodies, and 25 for class II (Table 4). The antibody specificity was determined through the use of 3 methods (ELISA, Luminex, and CDC). Flow-PRA was not included in this analysis because of the high costs of reagents that are also based on the same principle of Luminex. By this analysis, we could identify class I antigen Table 2. Comparison of results among Flow, Luminex, ELISA, and CDC techniques. -/- +/- -/+ +/+ Class I antibodies FL/ELISA FL/LX FL/CDC ELISA/LX ELISA/CDC LX/CDC Class II antibodies FL/ELISA FL/LX FL/CDC ELISA/LX ELISA/CDC LX/CDC immunosorbent assay; FL, flow cytometric method for panel reactive antibody; LX, bead-based technology Table 3. CDC, FL, ELISA, LX, Concordance. HLA Class I HLA Class II immunosorbent assay; FL, flow cytometric method for panel reactive antibody; HLA, Human leucocyte antigen; LX, bead-based technology Expressed as score levels: (+) concordance / (-) no concordance. mismatches in 35/37 samples with Luminex (94.6%), 15/37 with ELISA (40.5%), and 13/37 with CDC (35.0%). Similarly, for class II, we observed specificity correlations in 17/25 with Luminex (68%), 6/25 with ELISA (24%), and 4/25 with CDC (16%) (Table 4). Table 5 illustrates how many times the single antibody specificity could be revealed by the 3 techniques studied. As shown, the antibody specificities found with ELISA, Luminex, and CDC were 13, 30, and 11 for HLA-A; 3, 13, and 3 for HLA- B; and 8, 17, and 6 for HLA-DR. By this analysis, the Luminex method could detect antibody specificities more frequently than the others. Table 4. Specificity correlations. CDC ELISA Luminex HLA I 13/37 (35%) 15/37 (40.5%) 35/37 (94.6%) HLA II 4/25 (16%) 6/25 (24%) 17/25 (68%) immunosorbent assay; HLA I, human leucocyte antigen class I; HLA II, human leucocyte antigen class II; Luminex, flow analyzer bead-based technology Table 5. Analysis of antibodies specificities detected in sera by CDC, ELISA, LX. HLA antibodies CDC ELISA LX A A A A9 (23, 24) A A A A Total B5 (51, 52) B B12 (44, 45) B14 (64, 65) B15 (62, 63, 74-76) 1-1 B17 (57, 58) B B22 (54-56) B B B Total DR DR2 (15, 16) DR3 (17, 18) DR5 (11, 12) DR6 (13, 14) DR7-1 1 DR9-1 - Total immunosorbent assay; HLA, human leucocyte antigen; LX, bead-based technology.

5 Pellegrino B. Minucci et al /Experimental and Clinical Transplantation (2011) 6: Discussion All the standard techniques analyzed have some limitations, and none of these methods can determine with certainty a positive or a negative result, with the exception perhaps, of Luminex. Although our sample is small, our data are substantially in line with those reported in the literature, thus permitting us to draw some considerations. First, our analysis revealed that Luminex method is more sensitive and specific than the others. 11, 16 Indeed, for class I antibodies, this assay could detect most of the positive samples (90/101). When we considered class II antibodies, CDC would seem more sensitive than Luminex (90/101 vs 79/101) (Table 1). This discrepancy can be caused by several methods. The CDC method cannot, without proper methodologic strategies (such as absorption of HLA class I antibodies with pooled platelets), to discriminate between class I and II antibodies, thus providing false-positive reactions. Moreover, we did not use dithiothreitol to differentiate between IgM and IgG antibodies as reported in other studies. 11 Clinical relevance of these antibodies is still controversial, 17, 18 and in our laboratory, we always perform pretransplant crossmatch analysis with Ts (CD8 + ) and B (CD19 + ) lymphocytes. A further problem can be the quality of B cells, which often show a high background of cell death, even though we did not find this technical problem. Despite CDC being widely accepted and feasible, it showed more discrepancies than the other systems in the detection of class I antibodies. Further limitations are represented by the panel specificity wholeness and the lack of response by antibodies that do not fix the complement, which, as shown recently, 14 may be the cause of both acute and chronic rejection. Interestingly, a study investigating the clinical relevance of Luminex technology reported sensitivity percentages similar to our data; the authors also suggested that Luminex could be valid for class I antibody detection but uncertain for class II. 16 Differently, the ELISA method offers the important advantage of making all the tests automatically, especially for those laboratories working on large sample numbers. Enzyme-linked immunosorbent assay screening appeared more sensitive for class I than for CDC, but less sensitive if compared to Luminex and FlowPRA (Table 2). This difference probably depends on several factors. First, the ELISA panel composition consists of a platelet antigen pool from donors of different ethnicities, whereas FlowPRA and Luminex both consist of recombinant HLA antigens. A second point may be the serum dilution (1:4 dilution in ELISA vs undiluted serum in FlowPRA and Luminex). Indeed, in our experience, excessive serum dilution can lead to the failure of antibody detection in some samples. Another aspect to be considered with ELISA, and even more with CDC, is the difficulty to efficiently identify anti-dq and anti-dp antibodies, owing to the linkage disequilibrium between these alleles, because their presence in the serum is of everincreasing interest. 2, 19 Thus, based on the obtained data, the flow cytometry test (FlowPRA) and the flow analyzer test (Luminex) appear to be more sensitive and therefore able to reveal specificity at low antibody titers. Moreover, they allow recognition of specific epitopes directed to public and private antibodies more easily and with high reproducibility degree (Tables 4 and 5). We conclude that all these techniques have some limitations and there is no standard method for determining with certainty a positive or a negative result. Indeed, each screening test can give false negative results for several reasons: (1) the reduced sensitivity of the commercial kits; (2) the low antibody titer in the serum; (3) the under-represented specificities in the commercially kits or in the antigenic cell panel; or (4) the excessive serum dilution. Nevertheless, from our results, Luminex seems to be more reliable than the other assays. It also would be interesting to evaluate the clinical relevance of positive antibody reactions detected in 1 assay, but the nonreactive reactions in the other assays. Thus, future studies should explore this important and difficult task. A recent report from the Collaborative Transplant Study found no association of kidney graft loss with HLA antibodies detected exclusively by sensitive Luminex single-antigen testing. 20 We should consider that higher sensitivity is not necessarily an improvement in terms of clinical relevance. Indeed, despite the increased sensitivity of the newly developed antibody assays may account for the increased number of sensitized patients awaiting transplant, it also has been associated with decreased specificity, and some non-hla antigens with no clinical relevance have been able to give a positive crossmatch. 5 These false-positive antibody

6 386 Pellegrino B. Minucci et al /Experimental and Clinical Transplantation (2011) 6: Exp Clin Transplant results have as a consequence a decreased chance of the patient to receive an organ by way of exchange organizations, thus decreasing chances for the patient. We believe that each laboratory should perform a strategy for identifying HLA antibodies. A working hypothesis could be the use of a basic technique, which is considered enough sensitive, like Luminex or FlowPRA, together with another technique such as ELISA or CDC. Indeed, our results indicate that to date, the use of multiple methods in this diagnostic field is still an indispensable approach for a better characterization of the immunologic status of the patient awaiting organ transplant. References 1. Porter KA. The effects of antibodies on human renal allografts. Transplant Proc. 1976;8(2): Stastny P, Salvador IM, Lavingia B. Evaluation of the highly sensitized transplant recipient. Pediatr Nephrol [Epub ahead of print]. 3. Kissmeyer-Nielsen F, Olse S, Petersen VP, Fjeldborg O. Hyperacute rejection of kidney allografts, associated with pre-existing humoral antibodies against donor cells. Lancet. 1966;2(7465): Patel R, Terasaki PI. Significance of the positive crossmatch test in kidney transplantation. New Engl J Med. 1969;280(14): Gloor J, Stegall MD. Sensitized renal transplant recipients: current protocols and future directions. Nat Rev Nephrol. 2010;6(5): Terasaki PI, McClelland JD. Microdroplet assay of human serum cytotoxins. Nature. 1964;204: Zeevi A, Girnita A, Duquesnoy R. HLA antibody analysis: sensitivity, specificity, and clinical significance in solid organ transplantation. Immunol Res. 2006;36(1-3): Aubert V, Venetz JP, Pantaleo G, Pascual M. Low levels of human leukocyte antigen donor-specific antibodies detected by solid phase assay before transplantation are frequently clinically irrelevant. Hum Immunol. 2009;70(8): Batal I, Zeevi A, Lunz JG 3rd, et al. Antihuman leukocyte antigen specific antibody strength determined by complement-dependent or solid-phase assays can predict positive donor-specific crossmatches. Arch Pathol Lab Med. 2010;34(10): Gebel HM, Bray RA, Ruth JA, et al. Flow PRA to detect clinically relevant HLA antibodies. Transplant Proc. 2001;33(1-2): Colombo MB, Haworth SE, Poli F, et al. Luminex technology for anti- HLA antibody screening: evaluation of performance and of impact on laboratory routine. Cytometry B Clin Cytom. 2007;72(6): Lee PC, Ozawa M, Hung CJ, Lin YJ, Chang SS, Chou TC. Reappraisal of HLA antibody analysis and crossmatching in kidney transplantation. Transplant Proc. 2009;41(1): Mansour I, Messaed C, Azoury M, Klayme S, Naaman R. Panelreactive antibodies using complement-dependent cytotoxicity, flow cytometry, and ELISA in patients awaiting renal transplantation or transplanted patients: a comparative study. Transplant Proc. 2001;33(5): Poli F, Cardillo M, Scalamogna M. Clinical relevance of human leukocyte antigen antibodies in kidney transplantation from deceased donors: the North Italy Transplant program approach. Hum Immunol. 2009;70(8): Muro M, Llorente S, Marín L, et al. Acute vascular rejection mediated by HLA antibodies in a cadaveric kidney recipient: discrepancies between FlowPRA, ELISA and CDC vs luminex screening. Nephrol Dial Transplant. 2005;20(1): Billen EV, Christiaans MH, van den Berg-Loonen EM. Clinical relevance of Luminex donor-specific crossmatches: data from 165 renal transplants. Tissue Antigens. 2009;74(3): Chapman JR, Taylor CJ, Ting A, Miach PJ, Chapman JR, Morris PJ. Immunoglobulin Class and specificity of antibodies causing positive T cell crossmatches. Relationship to renal transplant outcome. Transplantation. 1986;42(6): Roelen DL, van Bree J, Witvliet MD, et al. IgG antibodies against an HLA antigen are associated with activated cytotoxic T cells against this antigen, IgM are not. Transplantation. 1994;57(9): Tait BD, Hudson F, Cantwell L, et al. Luminex technology for HLA antibody detection in organ transplantation. Nephrology (Carlton). 2009;14(2): Süsal C, Ovens J, Mahmoud K, et al. No association of kidney graft loss with human leukocyte antigen antibodies detected exclusively by sensitive Luminex single-antigen testing: a Collaborative Transplant Study report. Transplantation. 2011;91(8):