Human Immunology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
www.ashi-hla.org
journal homepage: www.elsevier.com/locate/humimm
The dilemma of DQ HLA-antibodies Rabab Al Attas a,⇑, Dalal Al Abduladheem a, Adel A. Shawhatti a, Ricardo Lopez a, Abdelhamid Liacini a, Saber AlZahrani a, Khalid Akkari b, Nasreen Hasan c, Abdulnaser Alabadi b a
Histocompatibility & Immunogenetics Laboratory (HIL), King Fahad Specialist Hospital-Dammam, Dammam, Saudi Arabia Multi Organ Transplant Center, King Fahad Specialist Hospital-Dammam, Dammam, Saudi Arabia c Pathology and Laboratory Medicine, King Fahad Specialist Hospital-Dammam, Dammam, Saudi Arabia b
a r t i c l e
i n f o
Article history: Received 17 July 2014 Revised 3 March 2015 Accepted 11 March 2015 Available online xxxx
a b s t r a c t Accurate identification of antibody reactivity against HLA-DQ antigens was difficult by using the old serological assays because of the strong linkage disequilibrium between HLA-DR and HLA-DQ (the usual inheritance of a certain HLA-DR molecule that ties together with the same DQ molecule within a racial group). The accurate and precise identifications of anti-HLA-antibodies of DQ specificities were made possible with the introduction of multiplex-bead arrays (Luminex), using single antigen bead (SAB) assay. The SAB assay is also considered today to be the most sensitive and specific method for alloimmunization assessment even for the low titer anti-HLA-antibodies. However, it is becoming clear that the detection of the HLA antibodies by SAB is not absolutely perfect due to the variation in densities, conformations and orientations of the antigen coated beads. Unlike HLA-DR, the HLA-DQ antigens are made of two polymorphic chains, both (alpha and beta chains) can contribute to the process of immunization individually or jointly. Routine SAB testing approach, which assigns the specificities based on beta chains and ignores the contribution of the DQa chains, can lead to erroneous DQ-antibody assignments. Therefore, it is important to recognize both the peculiarity of the HLA-DQ antigens as well as the nature of the assay format used in order to reach the correct antibody assignments. Erroneous donor specific antibodies (DSA) assignment may lead to denial of an otherwise immunologically compatible organ transplant, or exposing transplant recipients to unnecessary investigations or immunosuppression. The following two patients presented with HLA-antibodies against DQ antigens (anti-DQ-Abs) highlight these two scenarios. Ó 2015 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.
1. Introduction The detrimental effect of HLA Class II-antibodies directed against allograft antigens on transplantation outcome is well established [1–3]. Until recently, the most commonly reported antibodies to HLA-Class II antigens were those targeting the HLADR molecules. The significance of the DQ antibodies is increasingly recognized following the introduction of the SAB assay. HLA-DQ antigen is an ab heterodimer of the HLA Class II type [4]. Unlike DR, the HLA-DQA1 and HLA-DQB1 genes that encode the a and b chains of DQ antigens respectively are both polymorphic although a chain exhibits less polymorphism [5]; consequently, sensitization is usually and dominantly directed to the b chain, but both chains contribute to the complete structure and can induce immunologic response [6]. It is a common practice to identify sensitization to Class II antigens based on its b chains; however,
⇑ Corresponding author.
this approach may not be accurate for DQ due to its chains polymorphism. Therefore, it is important to understand the nature and the complex data generated by SAB based testing in order to accurately interpret the results. The two patients that are presented in this study highlight how misinterpretations of DQ antibodies can potentially prevent an otherwise HLA-compatible transplant or unnecessarily alter patient management.
2. Case 1 A 23-year-old male with end-stage renal disease (ESRD) underwent evaluation for living donor kidney transplantation. Table 1 summarizes the HLA-typing for the patient and his four living related potential donors. Screening of HLA-antibodies using LABScreen on Luminex platform was negative for Class I, but positive for HLA-Class II antibodies. Antibody identification by SAB confirmed specificities against HLA-DQ7, DQ8, and DQ9 (Figs. 1A and B). The right column of Fig. 1B provides the median
http://dx.doi.org/10.1016/j.humimm.2015.03.008 0198-8859/Ó 2015 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
2
R. Al Attas et al. / Human Immunology xxx (2015) xxx–xxx
Table 1 HLA typing result (Case 1). A
B
C
DRB1
DQA1
DQB1
Patient
02, 33
49, 51
07, 15
03(17), 16
01:02 05:01
05:02 02:01
D (1) – Father
68, 33
35, 51
04, 15
04, 16
01:02 03:02/03
05:02 03:02(8)
D (2) – Sister
11, 68
35, 51
04, 15
04, 11
03:02/03 05:05
03:02(8) 03:01(7)
D (3) – Cousin 1
01, 03
51, –
06, 16
04, –
03:02/03 03:01
03:02(8) 03:02(8)
D (4) – Cousin 2
01, 31
35, 51
06, 07
03(17), 04
03:01 05:01
03:02(8) 02:01
fluorescence intensity (MFI) values for different HLA-DQ-coated single antigen beads while the middle column illustrates the different combinations of DQA1 and DQB1alleles. No anti-HLA Class I-antibodies were detected by SAB. The patient was flow cross matched (FXM), with three out of the four potential donors. The results of the FXM with the second (sister) and the third donor (cousin 1) were positive for B-cells (MCS = 170, 223, respectively, (cutoff point 106 using 1024-channel scale)), but B-FXM was negative with the fourth donor. T-FXM results were negative with all the donors. The positive B crossmatches were consistent with the DSA that were demonstrated in the patient sera against donor HLAClass II antigens (DQ7 and DQ8, (second donor), and DQ8, (third donor)). However, the negative result with the fourth donor (cousin 2) was unexpected because he also carried DQ8 antigens. Both cousins were full siblings. The crossmatch results were reproducible. A closer analysis of the data revealed that there was a positive reactivity of all DQ8 coated beads except for one bead (bead number 82). Then, it was thought that the DQA1⁄ component was responsible for the positivity and not the b component (DQB1⁄03:02). However, a comprehensive bead reactivity analysis baffled us further as the positive reactions of bead number 42 and 43 could only be attributed to DQ8 b (B⁄03:02) component, in view of the clear negative reactions with the corresponding a components for both beads. Low or unexpression of DQ8 antigen in the fourth donor was considered initially, but this possibility was not supported by the serological and high resolution (HR) HLA typings that were performed for DQ8. Results of HR typing showed that both the second and third donors who crossmatched positive carry DQA1⁄03:02, DQB1⁄03:02 combination while this combination was not found in the fourth donor in whom DQ8 antigens were composed of DQA1⁄03:01, DQB1⁄03:02. As the difference between DQA1⁄03:02 and DQA1⁄03:01 alleles is in the 50 -UTR and exon 3, (outside the antigen recognition site (ARS)); in addition, studies have shown sharing of all epitopes between these two alleles [7,8]; therefore, the individual DQA1⁄03:01/03:02 molecules could not be responsible for the different immune response. We then thought that it was possibly the DQA1⁄03:02, DQB1⁄02:02 combinations with the created antigenic epitopes accounted for the positive crossmatch while the DQA1⁄03:01, DQB1⁄03:02 conformation did not expose these epitopes efficiently, leading to the negative reaction in the fourth donor. Further bead analysis still did not support the latter speculation since DQA1⁄03:01, DQB1⁄03:02 coated beads (bead# 42) reacted strongly (8996 MFI). It has been known that HLA antigen purification and beads coating can lead to improper conformation of antigens which may give rise to detection of clinically irrelevant antibodies in SAB assays [9,10]; thus, it is recommended that for those sera with questionable reactions should be subjected to different source of HLA protein like the phenotype beads and additional testing with assays such as LABScreen PRA from a second vendor may eliminate
false positive reactions [11,12]. To resolve this dilemma, flow PRA Class II bead on BD cell analyzer was performed. Interestingly, one combination that included homozygous DQA1⁄03:02 coated beads in one group was the only positive response found (Fig. 2). The homozygous DQA1⁄03:02 was only present in donor 3; and a single copy of this allele was present in donor 2 (who also carried DQ7), while donor 4 did not carry this allele completely. Thus, we speculated that there may be another antigenic epitope located in DQA1⁄03:02, but it might be expressed more efficiently in the homozygous conformation of this allele. This might explain the higher channel shift of the B-cell-FXM with the third donor as compared to the second donor who apparently demonstrated two incompatible antigens (DQ7 and DQ8). To verify further, the assessment of the degree of contribution to the sensitization of DQA1⁄03:02, DQ7, and the DQ8 (DQA1⁄03:02, DQB1⁄03:02 combination) individually was tested by performing crossmatch tests with three surrogate cells and adsorption studies. The crossmatches with the DQA1⁄03:02 homozygous and heterozygous surrogate cells were positive for B-FXM (Table 2), but the surrogate cells that typed DQ7/DQ6 (with same DQ7-allele as present in donor 2) reacted surprisingly negative with patient serum. Due to the absence of the family members, we were unable to crossmatch the adsorbed/eluted serum (obtained after treatment of patient serum with the third surrogate cells) with second donor to prove that the positive B-FXM was attributed to DQ8 only. The inconclusive results in assessing the immunological risk with the fourth donor, had been denying patient transplantation for 6 months with extensive workup before he underwent uneventful transplant with graft offered by the fourth donor. The patient is currently enjoying good kidney function 8 months post transplant. 3. Case 2 A 26-year-old female with ESRD had her first kidney transplant from her mother in 2011. Her graft failed over the next 3 years due to chronic rejection, with a history of medication non-compliance. Her brother offered the second graft. HLA-typing for the patient and her two donors (mother and brother) is represented in Table 3. During the second transplant work up, her investigation showed negative HLA-Class I, but positive HLA-Class II antibody screening. Antibody identification by SAB revealed weak DQ2 specificity of 600 MFI (Figs. 3A and B), which was interpreted as DSA to the first donor, due to DQ2 antigen mismatch. Typing showed the new donor (brother) to share DQ2 antigen with the first graft donor (mother) but FXM was negative for both T and B cells with the second donor. Patient proceeded to a Thymoglobulin induced second transplant in February 2014 from her brother. The first routine post transplant SAB results showed remarkable increase in DQ2, MFI of 11380 (Figs. 4A and B). There
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
3
R. Al Attas et al. / Human Immunology xxx (2015) xxx–xxx
Fig. 2. Flow Histogram (Case1): The figure illustrates examples of FL1 vs. FL2 dot plots of FlowPRAÒ Class II Specific beads (FL2SP). Histogram A illustrates group 1 reactions with negative control serum (FL-NC) for all the beads, the vertical line indicates the boundary of the negative control. In histogram B the beads that have shifted to the right due to binding of the secondary antibody indicate positivity against DQA1⁄03:02 alloserum.
Table 2 Crossmatch results with surrogate cells (Case1). DRB1⁄
Surrogate cells 1 2 3
DRB1⁄
04 11
DRB3/4/5⁄ B4⁄01(53) B3⁄02(52)
16 15
DRB3/4/5⁄
DQB1⁄
DQB1⁄
DQA1⁄
DQA1⁄
T-FCXM
B-FCXM
B5⁄01(51) B5⁄01(51)
03(8) 03:02 03(8) 03:02 03 (7) 03:01
03(8) 03:02 05:02 06:02/47
03:01 03:01 05:05/09
03:01/02/03 01:02 01:02
Negative Negative Negative
Positive (MCS: 249) Positive (MCS: 119) Negative
Table 3 HLA-typing – Case 2.
Patient D1 (mother) D2 (brother)
A
B
C
DRB1
DQA1
DQB1
23, 33 23, – 03, 23
42, 14 42, 50 27, 50
08, 17 06, 17 02, 06
03, 01 03, 07 07, 11
01, 04 02, 04 02, 05
05, 04 02, 04 02, 03(7)
were no other HLA Class I or II alloantibodies. The treating nephrologist was alerted, and patient was put under close surveillance, including repeating DSA tests despite a normal serum creatinine level. A kidney biopsy showed mild tubular injury in some tubules, with negative C4d staining. Because of the persistence elevation of DSA and in view of normal kidney function, the treating team kept questioning the SAB results. While having a stable normal serum creatinine, a second biopsy was obtained which showed the same previous histological findings, but with C4d focally positive. As a result, the patient received pulse steroids, five sessions of plasma exchange, intravenous immunoglobulin, and a single dose of Rituximab. Follow up DSA levels remained unchanged over the subsequent 3 weeks. As a result, patient and donors’ HLA data were reanalyzed, including performing DQ high resolution typing. HR typing confirmed HLA-DQB1⁄02:02 associated with DQA1⁄02:01 for both donors. Careful analysis of DQ2 coated bead reaction showed that the relevant antibodies (directed against
donor’s antigens) were always on the low side while what we reported as DSA was actually non-donor specific. It was directed against a subgroup of DQ2 (HLA-DQB1⁄02:01 associated with DQA⁄05:01). As DQB1⁄02:01 and DQB1⁄02:02 differ only in one nucleotide at position 135, outside the ARS, the difference in bead reactions between DQB1⁄02:02 and DQB1⁄02:01 coated beads is unlikely to be due to the b component alone. Instead, this was likely to be caused by different target that was generated by pairing of b with different a chains. Alternatively, the antibodies might have been directed solely towards the a component (DQ1⁄05:01). The latter possibility could not be unequivocally excluded with the current panel of beads. The possibility that the individual DQB1⁄ molecules that were not responsible for the reaction was also supported by the negative reactions of the DQB1⁄02:01 in beads number 40, 41 and 42, where these coated beads have different a chain components. Repeated Post-transplantation crossmatches were persistently negative. An amended report was issued with the correct assignment. The patient is now 12 months post-transplant with very good renal function.
4. Methods Medium resolution typing was performed by PCR-based sequence-specific oligonucleotide (SSO) on Luminex platform.
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
4
R. Al Attas et al. / Human Immunology xxx (2015) xxx–xxx
The primary panel to assess the immunization status of our patients includes HLA-antibody detection (screening) and specificity identification on Luminex platform using pooled antigens (LabScreen Mixed), and single antigen bead assay (LabScreen Single Antigen) respectively, (One Lambda Inc., Canoga Park, USA). The threshold for a positive response was set at mean fluorescence intensity (MFI) of 1000 in SAB assay. Only DSA between 500 and 1000 MFI will be reported as weak reactivity. Phenotype panels’ coated beads (PRA) test is not performed routinely in our laboratory unless indicated to resolve ambiguities that are encountered during data analysis of SAB. PRA was detected using one Lambda FlowPRA Specific HLA Class II Beads on BD FACSCanto II flow cytometer. High resolution typing was performed by SBT for DQB1⁄ using allelsSEQR (Celera, Alameda, CA), distributed by Abbott (Germany) on ABI 3130, and assignment of DQA1 was extrapolated from the medium resolution typing, family segregation, and from the common linkage association between DQB1 serologic equivalents and DQA1allele tables [6]. Serology typing was performed using Terasaki tissue typing tray, LMT First HLA Class II MDR172 (One Lambda Inc.). Flow crossmatch was performed by incubating donor cells with patient serum followed by the addition of a fluoresceinated (FITC) goat anti-human polyclonal antibody, Phycoerythrin (PE) labeled monoclonal anti-CD19 (B cells) and peridinin chlorophyll (PerCP) conjugated anti-CD3 (T cells). With a three color combination, T and B cell alloantibodies could be detected simultaneously. Results were analyzed and called positive or negative based on a shift in the median channel fluorescence intensity value from the negative control serum. 5. Discussion Although, the introduction of SAB assays using Luminex technology has recognizably allowed precise characterization of HLAantibodies, it is becoming clearer that the detection of antibodies by SAB has not been absolutely perfect due to several reasons [13]. The negative-B-cell Flow-XM result, in the first case with the fourth donor, was puzzling for us. Given the current panel of beads, the results of SAB did not conclusively determine whether the reactivity was because of the sole response against the DQA1⁄03:02 or DQB1⁄03:02, or it was directed against epitopes that were made of the combination of DQA1⁄03:02 and DQB1⁄03:02. The degree of mismatching at the structural level could be assessed by using the HLA-matchmaker [14] or by extensive adsorption/elusion studies [15]. However, the applications of such tools have not been widely available in most HLA-laboratories. The extensive testing using different antibody assays and the repeated crossmatches with surrogate cells, showed us that the different combinations in the three donors created different epitopes, and likely made the DQ8 antigen more immunogenic in donor three than donor two, but not immunogenic in donor four. It seems that the antibodies recognized DQB1⁄03:02 preferentially when it is coupled with DQA1⁄03:02. DQA1⁄03:02 molecules individually may also constitute immunogenic targets that are enhanced in homozygous conformations as evident by the stronger FXM results with donor 3 and the first surrogate cells (both were homozygous for DQA1⁄03:02) compared to the lower channel shift with donor 2 who apparently had two DSAs (DQ7 and DQ8). Since donor 2 carried both DQ7 and DQ8, it was difficult initially to assess the degree of contribution to the immunogenicity for each antigen individually. Given the information of the cross match with the third surrogate cell, we were able to deduce at least theoretically that the reactivity observed with donor 2 was solely due to DQ8 antigens. Unfortunately, assays relying on cellular absorption techniques are severely limited in their ability to reliably differentiate HLA from non-HLA antibodies. Despite of the absence of definitive
studies, our experiments highlighted the strong immunogenicity created by DQA1⁄03:02 compared to other DQ antigens and perhaps stimulate others to look into similar cases and investigate them further. Mikkelsen et al. [16] reported that the possibility of a/b chains rearrangement of HLA-DQ leading to expression of different combinations on the surface of antigen presenting cells (APC) giving rise to ‘semi’ – semi-direct allorecognition pathway. Whether this or similar events occurred in donor 3 (homozygous for DQ8), creating accessible epitopes or hiding epitopes in donor 4 need further studies. The discordant results between the positive reaction response of the DQA1⁄03:01, DQB1⁄03:02 (bead# 42) in the DQ8 coated beads and the in vivo response with donor four who carries the same combination as demonstrated by both the negative cross match and the excellent transplant outcome might indicate false positive reaction due to denaturation of antigen during coating process which may uncover cryptic epitopes that comprised of amino acids buried within HLA-binding groove, making them accessible only on the dissociated antigens [17,18]. It is our policy to report the highest MFI for DSA. In the second patient, the highest bead was coated with DQB1⁄02:01, in both pre-transplant and post-transplant sera. Our report of strong DSA may have disadvantaged this patient when, in reality, she only had weak level of DSA. In view of the known linkage disequilibrium between DQA1 and DQB1 chains (HLA-DQB1⁄02:02 usually associated with DQA1⁄02:01 (HLA-DQ2.2) and DQB1⁄02:01 is usually associated with DQA1⁄05:01(2.5)), a more careful interpretation of the results could have prevented the erroneous reporting of DSA, and the distress that had been created. Theoretically, this patient could have developed memory response against the repeated mismatch antigens, and the aggressive treatment may have prevented serious outcome. However, the persistence of normal kidney function in the absence of significant histological findings at least during the initial post-transplant period, and the minimal increase in the actual DSA, with the negative post-transplant crossmatch results did not support this possibility. We believe that this case may represent a status of tolerance to noninherited maternal antigen (NEMA) shared by both donors. Therefore, the correct antibody assignment could have alleviated unjustifiable worry and repeated laboratory testing. Consequently, the aggressive treatment could have been potentially avoided. Although allele-specific antibodies can occur at any locus, the complexity of HLA-DQ molecules and the increased frequency of DQ immunization [19–22], make the chance of the misinterpretation of DQ allele-specific antibodies greater than other HLAantigens and justify public attention. In conclusion, analysis of the DQ HLA antibodies can be misleading the current serological definition of the beads in the SAB assay may be in some cases insufficient. Additional antibody detection techniques, and extended high resolution DQA1/DQB1 typing, for both donor and recipient, may be required to resolve ambiguities for the correct DQ-antibody assignment. The issue is further complicated by the ramification of the immunologic response of the antigen coated beads in the SAB assay, and the natural behavior of antigens in vivo. Until an assay is developed to provide a practical algorithm for definition of clinically relevant DQ antibodies, and tools defining epitopes specificity become available for routine use, the transplant communities may continue to be misled by the DQ HLA antibody results. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.humimm.2015. 03.008.
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
R. Al Attas et al. / Human Immunology xxx (2015) xxx–xxx
References [1] Campos EF, Tedesco-Silva H, Machado PG, et al. Post transplant antibodies class II as risk factor for late kidney allograft failure. Am. J. Transplant. 2008;6:2316. [2] Dequesnoy RJ. Human leukocyte antigen class II antibodies and transplant outcome. Transplantation 2008;86:638. [3] Terasaki P, Mixuani K. Antibody mediated rejection: update 2006. Clin. J. Am. Soc. Nephrol. 2006;1:400–3. [4] March B. Genetics of histocompatibility. Curr. Opin. Hematol. 1994;1:4. [5] Anthony Nolan Research Institute. IMGT/HLA sequence database. Available at: (accessed 01.31.10). [6] Tambur AR, Leventhal JR, Friedewald JJ, Ramon D. The complexity of HLA-DQ antibodies and its effect on virtual crossmatching. Transplantation 2010;90: 1117. [7] Deng CT, El-Awar N, Ozawa M, Cai J, Lachmann N, Terasaki. Human leukocyte antigen class II DQ alpha and beta epitopes identified from sera of kidney allograft recipients. Transplantation 2008;86:452–9. [8] El-awar N, Nguyen A, Almeshari K, Alawami M, Alzayer F, Alharbi M, et al. HLA classII DQA and DQB epitopes: recognition of the likely binding sites of HLADQ alloantibodies eluted from recombinant HLA-DQ single antigen cell lines. Hum. Immunol. 2013;74:1141–52. [9] Cai J, Terasaki PI, Anderson N, Lachmann N, Schönemann C. Intact HLA not beta2m-free heavy chain-specific HLA class I antibodies are predictive of graft failure. Transplantation 2009;88:226. [10] Pereira S, Perkins S, Lee JH, Shumway W, LeFor W, Lopez-Cepero M, et al. Donor-specific antibody against denatured HLA-A1: clinically nonsignificant ? Hum. Immunol. 2011;72:492. [11] Tait BD, Süsal C, Gebel HM, Nickerson PW, Zachary AA, Claas FH, et al. Consensus guidelines on the testing and clinical management issues associated with HLA and non-HLA antibodies in transplantation. Transplantation 2013; 95:19–47. [12] Gombos P, Opelz G, Scherer S, Morath C, Zeier M, Schemmer P, et al. Influence of test technique on sensitization status of patients on the kidney transplant waiting list. Am. J. Transplant. 2013;13:2075–82.
5
[13] Middelton D, Jones J, Lowe D. Nothing’s perfect. Transpl. Immunol. 2014;30: 115–21. [14] Dequesnoy RJ, Marrari M. HLAMatchmaker-based definition of structural human leukocyte antigens epitopes detected by alloantibodies. Curr. Opin. Organ Transplant. 2009;14:403–9. [15] El-awar N, Terasaki PI, Nguyen A, Sasaki N, Morales-Buenrostro LE, Saji H, et al. Epitopes of human leukocyte antigen class I antibodies found in sera of normal healthy males and cord blood. Hum. Immunol. 2009;70:844–5. [16] Mikkelsen S, Korsholm T, Burg I, Petersen MS, MØller BK. Case report: binding of a clinically relevant HLA-DQa-specific antibody in a kidney graft recipient is inhibited by donor type HLA-DQb chain. Transplant. Proc. 2013;45: 1209–12. [17] Zachary AA, Vega RM, Lucas DP, Leffel MS. HLA antibody detection and characterization by solid phase immunoassay: methods and pitfalls. Methods Mol. Biol. 2012;882:289–308. [18] Lachmann N, Todorova K, Schulze H, Schönemann C. LuminexÒ and its applications for solid organ transplantation, hematopoietic stem cell transplantation, and transfusion. Transfus. Med. Hemother. 2013;40: 182–9. [19] Almeshari K, Pall A, Elgamal H, Alzayer F, Alawwami M. HLA-DQ antibodies are the most frequent antibodies encountered in antibody mediated rejection (AMR) of renal allograft. Hum. Immunol. Suppl. 2011;72:187. [20] Willicombe M, Brookes P, Sergeant R, Santos-Nunez E, Steggar C, Galliford J, et al. De novo DQ donor-specific antibodies are associated with a significant risk of antibody-mediated rejection and transplant glomerulopathy. Transplantation 2012;94:172. [21] Worthington JE, Martin S, Al-Husseini DM, Dyer PA, Johnson RW. Post transplantation production of donor HLA-specific antibodies as a predictor of renal transplant outcome. Transplantation 2003;75:1034. [22] DeVoes JM, Gaber AO, Knight RJ, Land GA, Suki WN, Gaber LW, et al. Donorspecific HLA-DQ antibodies may contribute to poor graft out come after renal transplantation. Kidney Int. 2012;82:598.
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
Contents lists available at ScienceDirect
www.ashi-hla.org
journal homepage: www.elsevier.com/locate/humimm
The dilemma of DQ HLA-antibodies Rabab Al Attas a,⇑, Dalal Al Abduladheem a, Adel A. Shawhatti a, Ricardo Lopez a, Abdelhamid Liacini a, Saber AlZahrani a, Khalid Akkari b, Nasreen Hasan c, Abdulnaser Alabadi b a
Histocompatibility & Immunogenetics Laboratory (HIL), King Fahad Specialist Hospital-Dammam, Dammam, Saudi Arabia Multi Organ Transplant Center, King Fahad Specialist Hospital-Dammam, Dammam, Saudi Arabia c Pathology and Laboratory Medicine, King Fahad Specialist Hospital-Dammam, Dammam, Saudi Arabia b
a r t i c l e
i n f o
Article history: Received 17 July 2014 Revised 3 March 2015 Accepted 11 March 2015 Available online xxxx
a b s t r a c t Accurate identification of antibody reactivity against HLA-DQ antigens was difficult by using the old serological assays because of the strong linkage disequilibrium between HLA-DR and HLA-DQ (the usual inheritance of a certain HLA-DR molecule that ties together with the same DQ molecule within a racial group). The accurate and precise identifications of anti-HLA-antibodies of DQ specificities were made possible with the introduction of multiplex-bead arrays (Luminex), using single antigen bead (SAB) assay. The SAB assay is also considered today to be the most sensitive and specific method for alloimmunization assessment even for the low titer anti-HLA-antibodies. However, it is becoming clear that the detection of the HLA antibodies by SAB is not absolutely perfect due to the variation in densities, conformations and orientations of the antigen coated beads. Unlike HLA-DR, the HLA-DQ antigens are made of two polymorphic chains, both (alpha and beta chains) can contribute to the process of immunization individually or jointly. Routine SAB testing approach, which assigns the specificities based on beta chains and ignores the contribution of the DQa chains, can lead to erroneous DQ-antibody assignments. Therefore, it is important to recognize both the peculiarity of the HLA-DQ antigens as well as the nature of the assay format used in order to reach the correct antibody assignments. Erroneous donor specific antibodies (DSA) assignment may lead to denial of an otherwise immunologically compatible organ transplant, or exposing transplant recipients to unnecessary investigations or immunosuppression. The following two patients presented with HLA-antibodies against DQ antigens (anti-DQ-Abs) highlight these two scenarios. Ó 2015 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.
1. Introduction The detrimental effect of HLA Class II-antibodies directed against allograft antigens on transplantation outcome is well established [1–3]. Until recently, the most commonly reported antibodies to HLA-Class II antigens were those targeting the HLADR molecules. The significance of the DQ antibodies is increasingly recognized following the introduction of the SAB assay. HLA-DQ antigen is an ab heterodimer of the HLA Class II type [4]. Unlike DR, the HLA-DQA1 and HLA-DQB1 genes that encode the a and b chains of DQ antigens respectively are both polymorphic although a chain exhibits less polymorphism [5]; consequently, sensitization is usually and dominantly directed to the b chain, but both chains contribute to the complete structure and can induce immunologic response [6]. It is a common practice to identify sensitization to Class II antigens based on its b chains; however,
⇑ Corresponding author.
this approach may not be accurate for DQ due to its chains polymorphism. Therefore, it is important to understand the nature and the complex data generated by SAB based testing in order to accurately interpret the results. The two patients that are presented in this study highlight how misinterpretations of DQ antibodies can potentially prevent an otherwise HLA-compatible transplant or unnecessarily alter patient management.
2. Case 1 A 23-year-old male with end-stage renal disease (ESRD) underwent evaluation for living donor kidney transplantation. Table 1 summarizes the HLA-typing for the patient and his four living related potential donors. Screening of HLA-antibodies using LABScreen on Luminex platform was negative for Class I, but positive for HLA-Class II antibodies. Antibody identification by SAB confirmed specificities against HLA-DQ7, DQ8, and DQ9 (Figs. 1A and B). The right column of Fig. 1B provides the median
http://dx.doi.org/10.1016/j.humimm.2015.03.008 0198-8859/Ó 2015 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
2
R. Al Attas et al. / Human Immunology xxx (2015) xxx–xxx
Table 1 HLA typing result (Case 1). A
B
C
DRB1
DQA1
DQB1
Patient
02, 33
49, 51
07, 15
03(17), 16
01:02 05:01
05:02 02:01
D (1) – Father
68, 33
35, 51
04, 15
04, 16
01:02 03:02/03
05:02 03:02(8)
D (2) – Sister
11, 68
35, 51
04, 15
04, 11
03:02/03 05:05
03:02(8) 03:01(7)
D (3) – Cousin 1
01, 03
51, –
06, 16
04, –
03:02/03 03:01
03:02(8) 03:02(8)
D (4) – Cousin 2
01, 31
35, 51
06, 07
03(17), 04
03:01 05:01
03:02(8) 02:01
fluorescence intensity (MFI) values for different HLA-DQ-coated single antigen beads while the middle column illustrates the different combinations of DQA1 and DQB1alleles. No anti-HLA Class I-antibodies were detected by SAB. The patient was flow cross matched (FXM), with three out of the four potential donors. The results of the FXM with the second (sister) and the third donor (cousin 1) were positive for B-cells (MCS = 170, 223, respectively, (cutoff point 106 using 1024-channel scale)), but B-FXM was negative with the fourth donor. T-FXM results were negative with all the donors. The positive B crossmatches were consistent with the DSA that were demonstrated in the patient sera against donor HLAClass II antigens (DQ7 and DQ8, (second donor), and DQ8, (third donor)). However, the negative result with the fourth donor (cousin 2) was unexpected because he also carried DQ8 antigens. Both cousins were full siblings. The crossmatch results were reproducible. A closer analysis of the data revealed that there was a positive reactivity of all DQ8 coated beads except for one bead (bead number 82). Then, it was thought that the DQA1⁄ component was responsible for the positivity and not the b component (DQB1⁄03:02). However, a comprehensive bead reactivity analysis baffled us further as the positive reactions of bead number 42 and 43 could only be attributed to DQ8 b (B⁄03:02) component, in view of the clear negative reactions with the corresponding a components for both beads. Low or unexpression of DQ8 antigen in the fourth donor was considered initially, but this possibility was not supported by the serological and high resolution (HR) HLA typings that were performed for DQ8. Results of HR typing showed that both the second and third donors who crossmatched positive carry DQA1⁄03:02, DQB1⁄03:02 combination while this combination was not found in the fourth donor in whom DQ8 antigens were composed of DQA1⁄03:01, DQB1⁄03:02. As the difference between DQA1⁄03:02 and DQA1⁄03:01 alleles is in the 50 -UTR and exon 3, (outside the antigen recognition site (ARS)); in addition, studies have shown sharing of all epitopes between these two alleles [7,8]; therefore, the individual DQA1⁄03:01/03:02 molecules could not be responsible for the different immune response. We then thought that it was possibly the DQA1⁄03:02, DQB1⁄02:02 combinations with the created antigenic epitopes accounted for the positive crossmatch while the DQA1⁄03:01, DQB1⁄03:02 conformation did not expose these epitopes efficiently, leading to the negative reaction in the fourth donor. Further bead analysis still did not support the latter speculation since DQA1⁄03:01, DQB1⁄03:02 coated beads (bead# 42) reacted strongly (8996 MFI). It has been known that HLA antigen purification and beads coating can lead to improper conformation of antigens which may give rise to detection of clinically irrelevant antibodies in SAB assays [9,10]; thus, it is recommended that for those sera with questionable reactions should be subjected to different source of HLA protein like the phenotype beads and additional testing with assays such as LABScreen PRA from a second vendor may eliminate
false positive reactions [11,12]. To resolve this dilemma, flow PRA Class II bead on BD cell analyzer was performed. Interestingly, one combination that included homozygous DQA1⁄03:02 coated beads in one group was the only positive response found (Fig. 2). The homozygous DQA1⁄03:02 was only present in donor 3; and a single copy of this allele was present in donor 2 (who also carried DQ7), while donor 4 did not carry this allele completely. Thus, we speculated that there may be another antigenic epitope located in DQA1⁄03:02, but it might be expressed more efficiently in the homozygous conformation of this allele. This might explain the higher channel shift of the B-cell-FXM with the third donor as compared to the second donor who apparently demonstrated two incompatible antigens (DQ7 and DQ8). To verify further, the assessment of the degree of contribution to the sensitization of DQA1⁄03:02, DQ7, and the DQ8 (DQA1⁄03:02, DQB1⁄03:02 combination) individually was tested by performing crossmatch tests with three surrogate cells and adsorption studies. The crossmatches with the DQA1⁄03:02 homozygous and heterozygous surrogate cells were positive for B-FXM (Table 2), but the surrogate cells that typed DQ7/DQ6 (with same DQ7-allele as present in donor 2) reacted surprisingly negative with patient serum. Due to the absence of the family members, we were unable to crossmatch the adsorbed/eluted serum (obtained after treatment of patient serum with the third surrogate cells) with second donor to prove that the positive B-FXM was attributed to DQ8 only. The inconclusive results in assessing the immunological risk with the fourth donor, had been denying patient transplantation for 6 months with extensive workup before he underwent uneventful transplant with graft offered by the fourth donor. The patient is currently enjoying good kidney function 8 months post transplant. 3. Case 2 A 26-year-old female with ESRD had her first kidney transplant from her mother in 2011. Her graft failed over the next 3 years due to chronic rejection, with a history of medication non-compliance. Her brother offered the second graft. HLA-typing for the patient and her two donors (mother and brother) is represented in Table 3. During the second transplant work up, her investigation showed negative HLA-Class I, but positive HLA-Class II antibody screening. Antibody identification by SAB revealed weak DQ2 specificity of 600 MFI (Figs. 3A and B), which was interpreted as DSA to the first donor, due to DQ2 antigen mismatch. Typing showed the new donor (brother) to share DQ2 antigen with the first graft donor (mother) but FXM was negative for both T and B cells with the second donor. Patient proceeded to a Thymoglobulin induced second transplant in February 2014 from her brother. The first routine post transplant SAB results showed remarkable increase in DQ2, MFI of 11380 (Figs. 4A and B). There
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
3
R. Al Attas et al. / Human Immunology xxx (2015) xxx–xxx
Fig. 2. Flow Histogram (Case1): The figure illustrates examples of FL1 vs. FL2 dot plots of FlowPRAÒ Class II Specific beads (FL2SP). Histogram A illustrates group 1 reactions with negative control serum (FL-NC) for all the beads, the vertical line indicates the boundary of the negative control. In histogram B the beads that have shifted to the right due to binding of the secondary antibody indicate positivity against DQA1⁄03:02 alloserum.
Table 2 Crossmatch results with surrogate cells (Case1). DRB1⁄
Surrogate cells 1 2 3
DRB1⁄
04 11
DRB3/4/5⁄ B4⁄01(53) B3⁄02(52)
16 15
DRB3/4/5⁄
DQB1⁄
DQB1⁄
DQA1⁄
DQA1⁄
T-FCXM
B-FCXM
B5⁄01(51) B5⁄01(51)
03(8) 03:02 03(8) 03:02 03 (7) 03:01
03(8) 03:02 05:02 06:02/47
03:01 03:01 05:05/09
03:01/02/03 01:02 01:02
Negative Negative Negative
Positive (MCS: 249) Positive (MCS: 119) Negative
Table 3 HLA-typing – Case 2.
Patient D1 (mother) D2 (brother)
A
B
C
DRB1
DQA1
DQB1
23, 33 23, – 03, 23
42, 14 42, 50 27, 50
08, 17 06, 17 02, 06
03, 01 03, 07 07, 11
01, 04 02, 04 02, 05
05, 04 02, 04 02, 03(7)
were no other HLA Class I or II alloantibodies. The treating nephrologist was alerted, and patient was put under close surveillance, including repeating DSA tests despite a normal serum creatinine level. A kidney biopsy showed mild tubular injury in some tubules, with negative C4d staining. Because of the persistence elevation of DSA and in view of normal kidney function, the treating team kept questioning the SAB results. While having a stable normal serum creatinine, a second biopsy was obtained which showed the same previous histological findings, but with C4d focally positive. As a result, the patient received pulse steroids, five sessions of plasma exchange, intravenous immunoglobulin, and a single dose of Rituximab. Follow up DSA levels remained unchanged over the subsequent 3 weeks. As a result, patient and donors’ HLA data were reanalyzed, including performing DQ high resolution typing. HR typing confirmed HLA-DQB1⁄02:02 associated with DQA1⁄02:01 for both donors. Careful analysis of DQ2 coated bead reaction showed that the relevant antibodies (directed against
donor’s antigens) were always on the low side while what we reported as DSA was actually non-donor specific. It was directed against a subgroup of DQ2 (HLA-DQB1⁄02:01 associated with DQA⁄05:01). As DQB1⁄02:01 and DQB1⁄02:02 differ only in one nucleotide at position 135, outside the ARS, the difference in bead reactions between DQB1⁄02:02 and DQB1⁄02:01 coated beads is unlikely to be due to the b component alone. Instead, this was likely to be caused by different target that was generated by pairing of b with different a chains. Alternatively, the antibodies might have been directed solely towards the a component (DQ1⁄05:01). The latter possibility could not be unequivocally excluded with the current panel of beads. The possibility that the individual DQB1⁄ molecules that were not responsible for the reaction was also supported by the negative reactions of the DQB1⁄02:01 in beads number 40, 41 and 42, where these coated beads have different a chain components. Repeated Post-transplantation crossmatches were persistently negative. An amended report was issued with the correct assignment. The patient is now 12 months post-transplant with very good renal function.
4. Methods Medium resolution typing was performed by PCR-based sequence-specific oligonucleotide (SSO) on Luminex platform.
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
4
R. Al Attas et al. / Human Immunology xxx (2015) xxx–xxx
The primary panel to assess the immunization status of our patients includes HLA-antibody detection (screening) and specificity identification on Luminex platform using pooled antigens (LabScreen Mixed), and single antigen bead assay (LabScreen Single Antigen) respectively, (One Lambda Inc., Canoga Park, USA). The threshold for a positive response was set at mean fluorescence intensity (MFI) of 1000 in SAB assay. Only DSA between 500 and 1000 MFI will be reported as weak reactivity. Phenotype panels’ coated beads (PRA) test is not performed routinely in our laboratory unless indicated to resolve ambiguities that are encountered during data analysis of SAB. PRA was detected using one Lambda FlowPRA Specific HLA Class II Beads on BD FACSCanto II flow cytometer. High resolution typing was performed by SBT for DQB1⁄ using allelsSEQR (Celera, Alameda, CA), distributed by Abbott (Germany) on ABI 3130, and assignment of DQA1 was extrapolated from the medium resolution typing, family segregation, and from the common linkage association between DQB1 serologic equivalents and DQA1allele tables [6]. Serology typing was performed using Terasaki tissue typing tray, LMT First HLA Class II MDR172 (One Lambda Inc.). Flow crossmatch was performed by incubating donor cells with patient serum followed by the addition of a fluoresceinated (FITC) goat anti-human polyclonal antibody, Phycoerythrin (PE) labeled monoclonal anti-CD19 (B cells) and peridinin chlorophyll (PerCP) conjugated anti-CD3 (T cells). With a three color combination, T and B cell alloantibodies could be detected simultaneously. Results were analyzed and called positive or negative based on a shift in the median channel fluorescence intensity value from the negative control serum. 5. Discussion Although, the introduction of SAB assays using Luminex technology has recognizably allowed precise characterization of HLAantibodies, it is becoming clearer that the detection of antibodies by SAB has not been absolutely perfect due to several reasons [13]. The negative-B-cell Flow-XM result, in the first case with the fourth donor, was puzzling for us. Given the current panel of beads, the results of SAB did not conclusively determine whether the reactivity was because of the sole response against the DQA1⁄03:02 or DQB1⁄03:02, or it was directed against epitopes that were made of the combination of DQA1⁄03:02 and DQB1⁄03:02. The degree of mismatching at the structural level could be assessed by using the HLA-matchmaker [14] or by extensive adsorption/elusion studies [15]. However, the applications of such tools have not been widely available in most HLA-laboratories. The extensive testing using different antibody assays and the repeated crossmatches with surrogate cells, showed us that the different combinations in the three donors created different epitopes, and likely made the DQ8 antigen more immunogenic in donor three than donor two, but not immunogenic in donor four. It seems that the antibodies recognized DQB1⁄03:02 preferentially when it is coupled with DQA1⁄03:02. DQA1⁄03:02 molecules individually may also constitute immunogenic targets that are enhanced in homozygous conformations as evident by the stronger FXM results with donor 3 and the first surrogate cells (both were homozygous for DQA1⁄03:02) compared to the lower channel shift with donor 2 who apparently had two DSAs (DQ7 and DQ8). Since donor 2 carried both DQ7 and DQ8, it was difficult initially to assess the degree of contribution to the immunogenicity for each antigen individually. Given the information of the cross match with the third surrogate cell, we were able to deduce at least theoretically that the reactivity observed with donor 2 was solely due to DQ8 antigens. Unfortunately, assays relying on cellular absorption techniques are severely limited in their ability to reliably differentiate HLA from non-HLA antibodies. Despite of the absence of definitive
studies, our experiments highlighted the strong immunogenicity created by DQA1⁄03:02 compared to other DQ antigens and perhaps stimulate others to look into similar cases and investigate them further. Mikkelsen et al. [16] reported that the possibility of a/b chains rearrangement of HLA-DQ leading to expression of different combinations on the surface of antigen presenting cells (APC) giving rise to ‘semi’ – semi-direct allorecognition pathway. Whether this or similar events occurred in donor 3 (homozygous for DQ8), creating accessible epitopes or hiding epitopes in donor 4 need further studies. The discordant results between the positive reaction response of the DQA1⁄03:01, DQB1⁄03:02 (bead# 42) in the DQ8 coated beads and the in vivo response with donor four who carries the same combination as demonstrated by both the negative cross match and the excellent transplant outcome might indicate false positive reaction due to denaturation of antigen during coating process which may uncover cryptic epitopes that comprised of amino acids buried within HLA-binding groove, making them accessible only on the dissociated antigens [17,18]. It is our policy to report the highest MFI for DSA. In the second patient, the highest bead was coated with DQB1⁄02:01, in both pre-transplant and post-transplant sera. Our report of strong DSA may have disadvantaged this patient when, in reality, she only had weak level of DSA. In view of the known linkage disequilibrium between DQA1 and DQB1 chains (HLA-DQB1⁄02:02 usually associated with DQA1⁄02:01 (HLA-DQ2.2) and DQB1⁄02:01 is usually associated with DQA1⁄05:01(2.5)), a more careful interpretation of the results could have prevented the erroneous reporting of DSA, and the distress that had been created. Theoretically, this patient could have developed memory response against the repeated mismatch antigens, and the aggressive treatment may have prevented serious outcome. However, the persistence of normal kidney function in the absence of significant histological findings at least during the initial post-transplant period, and the minimal increase in the actual DSA, with the negative post-transplant crossmatch results did not support this possibility. We believe that this case may represent a status of tolerance to noninherited maternal antigen (NEMA) shared by both donors. Therefore, the correct antibody assignment could have alleviated unjustifiable worry and repeated laboratory testing. Consequently, the aggressive treatment could have been potentially avoided. Although allele-specific antibodies can occur at any locus, the complexity of HLA-DQ molecules and the increased frequency of DQ immunization [19–22], make the chance of the misinterpretation of DQ allele-specific antibodies greater than other HLAantigens and justify public attention. In conclusion, analysis of the DQ HLA antibodies can be misleading the current serological definition of the beads in the SAB assay may be in some cases insufficient. Additional antibody detection techniques, and extended high resolution DQA1/DQB1 typing, for both donor and recipient, may be required to resolve ambiguities for the correct DQ-antibody assignment. The issue is further complicated by the ramification of the immunologic response of the antigen coated beads in the SAB assay, and the natural behavior of antigens in vivo. Until an assay is developed to provide a practical algorithm for definition of clinically relevant DQ antibodies, and tools defining epitopes specificity become available for routine use, the transplant communities may continue to be misled by the DQ HLA antibody results. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.humimm.2015. 03.008.
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
R. Al Attas et al. / Human Immunology xxx (2015) xxx–xxx
References [1] Campos EF, Tedesco-Silva H, Machado PG, et al. Post transplant antibodies class II as risk factor for late kidney allograft failure. Am. J. Transplant. 2008;6:2316. [2] Dequesnoy RJ. Human leukocyte antigen class II antibodies and transplant outcome. Transplantation 2008;86:638. [3] Terasaki P, Mixuani K. Antibody mediated rejection: update 2006. Clin. J. Am. Soc. Nephrol. 2006;1:400–3. [4] March B. Genetics of histocompatibility. Curr. Opin. Hematol. 1994;1:4. [5] Anthony Nolan Research Institute. IMGT/HLA sequence database. Available at: (accessed 01.31.10). [6] Tambur AR, Leventhal JR, Friedewald JJ, Ramon D. The complexity of HLA-DQ antibodies and its effect on virtual crossmatching. Transplantation 2010;90: 1117. [7] Deng CT, El-Awar N, Ozawa M, Cai J, Lachmann N, Terasaki. Human leukocyte antigen class II DQ alpha and beta epitopes identified from sera of kidney allograft recipients. Transplantation 2008;86:452–9. [8] El-awar N, Nguyen A, Almeshari K, Alawami M, Alzayer F, Alharbi M, et al. HLA classII DQA and DQB epitopes: recognition of the likely binding sites of HLADQ alloantibodies eluted from recombinant HLA-DQ single antigen cell lines. Hum. Immunol. 2013;74:1141–52. [9] Cai J, Terasaki PI, Anderson N, Lachmann N, Schönemann C. Intact HLA not beta2m-free heavy chain-specific HLA class I antibodies are predictive of graft failure. Transplantation 2009;88:226. [10] Pereira S, Perkins S, Lee JH, Shumway W, LeFor W, Lopez-Cepero M, et al. Donor-specific antibody against denatured HLA-A1: clinically nonsignificant ? Hum. Immunol. 2011;72:492. [11] Tait BD, Süsal C, Gebel HM, Nickerson PW, Zachary AA, Claas FH, et al. Consensus guidelines on the testing and clinical management issues associated with HLA and non-HLA antibodies in transplantation. Transplantation 2013; 95:19–47. [12] Gombos P, Opelz G, Scherer S, Morath C, Zeier M, Schemmer P, et al. Influence of test technique on sensitization status of patients on the kidney transplant waiting list. Am. J. Transplant. 2013;13:2075–82.
5
[13] Middelton D, Jones J, Lowe D. Nothing’s perfect. Transpl. Immunol. 2014;30: 115–21. [14] Dequesnoy RJ, Marrari M. HLAMatchmaker-based definition of structural human leukocyte antigens epitopes detected by alloantibodies. Curr. Opin. Organ Transplant. 2009;14:403–9. [15] El-awar N, Terasaki PI, Nguyen A, Sasaki N, Morales-Buenrostro LE, Saji H, et al. Epitopes of human leukocyte antigen class I antibodies found in sera of normal healthy males and cord blood. Hum. Immunol. 2009;70:844–5. [16] Mikkelsen S, Korsholm T, Burg I, Petersen MS, MØller BK. Case report: binding of a clinically relevant HLA-DQa-specific antibody in a kidney graft recipient is inhibited by donor type HLA-DQb chain. Transplant. Proc. 2013;45: 1209–12. [17] Zachary AA, Vega RM, Lucas DP, Leffel MS. HLA antibody detection and characterization by solid phase immunoassay: methods and pitfalls. Methods Mol. Biol. 2012;882:289–308. [18] Lachmann N, Todorova K, Schulze H, Schönemann C. LuminexÒ and its applications for solid organ transplantation, hematopoietic stem cell transplantation, and transfusion. Transfus. Med. Hemother. 2013;40: 182–9. [19] Almeshari K, Pall A, Elgamal H, Alzayer F, Alawwami M. HLA-DQ antibodies are the most frequent antibodies encountered in antibody mediated rejection (AMR) of renal allograft. Hum. Immunol. Suppl. 2011;72:187. [20] Willicombe M, Brookes P, Sergeant R, Santos-Nunez E, Steggar C, Galliford J, et al. De novo DQ donor-specific antibodies are associated with a significant risk of antibody-mediated rejection and transplant glomerulopathy. Transplantation 2012;94:172. [21] Worthington JE, Martin S, Al-Husseini DM, Dyer PA, Johnson RW. Post transplantation production of donor HLA-specific antibodies as a predictor of renal transplant outcome. Transplantation 2003;75:1034. [22] DeVoes JM, Gaber AO, Knight RJ, Land GA, Suki WN, Gaber LW, et al. Donorspecific HLA-DQ antibodies may contribute to poor graft out come after renal transplantation. Kidney Int. 2012;82:598.
Please cite this article in press as: Al Attas R et al. The dilemma of DQ HLA-antibodies. Hum Immunol (2015), http://dx.doi.org/10.1016/ j.humimm.2015.03.008
