Journal Logo

Original Clinical Science—General: Antibodies

Technical Limitations of the C1q Single-Antigen Bead Assay to Detect Complement Binding HLA-Specific Antibodies

Taylor, Craig J. PhD1; Kosmoliaptsis, Vasilis PhD2; Martin, Jessie MSc1; Knighton, Graham BSc1; Mallon, Dermot PhD2; Bradley, J. Andrew FMedSci2; Peacock, Sarah MSc1

Author Information
doi: 10.1097/TP.0000000000001270

Over the last decade there have been major technological advances in the detection and characterisation of HLA-specific antibodies which in turn have enabled a better understanding of the role of alloantibodies in the rejection of kidney allografts.1-3 Luminex single-antigen HLA-antibody detection bead (HLA-single antigen beads [SAB]) technology allows semi-quantitative detection of donor HLA-specific antibodies (DSA) with a high level of sensitivity and specificity.4 Screening for the presence of DSA in the sera of potential recipients of a kidney allograft allows selection of suitable antibody-compatible donors and the development of de novo DSA together with the deposition of C4d on graft biopsy are diagnostic of antibody-mediated allograft rejection and associated with inferior long-term graft survival.

Although the availability of HLA-SAB has revolutionized the detection of HLA-specific alloantibodies, it has become clear that not all patients with DSA identified by HLA-SAB experience allograft rejection and the period between development of de novo DSA and deterioration in graft function can vary from months to many years.5 The conventional HLA-SAB detects all IgG subclasses, irrespective of their ability to initiate complement activation. It has been suggested that those DSA that fix complement most effectively (IgG1 and IgG3) are those most likely to cause allograft injury, and in an attempt to detect such DSA, the solid phase Luminex HLA-SAB assay has been modified to identify only those DSA that bind C1q, the first component of complement activation (C1q-SAB).6-9 In the C1q-SAB assay, the presence and level of complement fixing IgG isotypes bound to HLA-SAB are identified by the addition of exogenous C1q that bind the Fc region of complement fixing IgG and is detected using a fluorescent-conjugated antihuman C1q antibody. Clinical application of the C1q-SAB assay has produced intriguing results that suggest C1q binding DSA are associated with graft damage, whereas IgG-DSA that do not bind C1q are clinically benign.10 It is not clear, however, whether this is due to the ability of the C1q-SAB assay to distinguish between complement fixing and noncomplement fixing IgG subclasses or whether positivity may also be affected by HLA-specific antibody level and by technological artefacts that interfere with IgG and/or C1q binding.11,12

The C1q-SAB assay represents a substantial additional cost for routine clinical use, and we and others have questioned the justification for this given the potential uncertainty about its interpretation.13,14 In the present study, we show that, in addition to the influence of antibody levels and interfering factors (commonly referred to as the prozone phenomenon), the presence of denatured HLA class I protein on SAB may interfere with the C1q-SAB assay.

MATERIALS AND METHODS

Study Group

Sera were obtained from 25 highly sensitised patients awaiting deceased donor kidney transplantation at the Cambridge Transplant Unit (12 men and 13 women; median age, 41 years; age range, 20-61 years). All patients had calculated reaction frequency greater than 85% determined against a standardised panel of 10 000 consecutive UK organ donors15 and had become sensitized by previous transplant alone (N = 1), previous transplant plus blood transfusions and/or pregnancy (N = 21), or blood transfusion plus pregnancy (N = 3). Patient sera used in this study had been obtained and stored for the purposes of antibody screening and characterization to guide selection of future antibody compatible donor and recipient pairs. This study was conducted in accordance with the UK Human Tissue Act that allows patients samples obtained for scheduled purposes (testing for HLA-specific antibodies) to be anonymized and used for pathology quality control and assay performance assessment.

Detection of IgG HLA Specific Antibodies and C1q Binding Using Luminex SAB

Patient sera were tested undiluted (neat), after ethylene diamine tetraacetic acid (EDTA) treatment and at 1 in 20 dilution using LabScreen single-antigen HLA class I–specific antibody detection beads (One Lambda, Canoga Park, CA). The addition of EDTA to sera used in the SAB assay acts as a chelating agent that sequesters calcium ions and abrogates C1q formation and obviates complement interference of FITC-labeled IgG detection reagent. Addition of EDTA to sera results in only 5% dilution (5 μL 6% EDTA solution to 95 μL test serum) and in validation tests by ourselves and others, does not alter antibody binding to SAB.16 In parallel, undiluted sera were tested using the C1QScreen (One Lambda) according to standard procedures. IgG-SAB and C1q-SAB binding levels were expressed as normalized mean fluorescence intensity (MFI) using HLA Fusion software (v3.2.0; One Lambda). All tests were undertaken at the same time using the same kit batches to minimise technical and operator variability.

Detection of Conformationally Folded and Denatured HLA on SAB

The level of conformationally folded (native) HLA class I protein expressed on the HLA-SAB was determined using W6/32 mouse monoclonal antibody (mAb) that recognizes a monomorphic HLA class I epitope expressed upon association of heavy chain and β2-microglobulin.17,18 Denatured HLA protein expressed on SAB was determined using HC-10 mAb that was raised against free HLA class I heavy chain and loses reactivity upon association of heavy chain with β2-microglobulin.19 HC-10 shows heterogeneity in the level of binding to different class I free heavy chain, and to account for this, denatured HLA was expressed as HC-10 MFI value obtained using untreated SAB as a percentage of maximal HC-10 MFI value obtained using denatured HLA class I after acid treatment of SAB. The purpose of the acid treatment of the HLA-SAB was to denature all the HLA protein and determine the maximal binding of HC-10 as shown in Figure S1 (SDC,http://links.lww.com/TP/B289).20

Data Analysis

The correlation between IgG-SAB MFI (using undiluted, EDTA-treated, and diluted sera) and C1q-SAB MFI data was assessed using scatter plots in Microsoft Excel (Seattle, WA). Pearson product moment correlation coefficients (r) were calculated to describe the proportion of variance between IgG-SAB MFI and C1q-SAB MFI. We predicted that the correlation would be a sigmoid curve, as turned out to be the case, because C1q binding will remain negative until a threshold of IgG binding is reached which allows IgG bound to adjacent HLA molecules to be colinked by C1q as described by Peacock et al.13 When this threshold is reached one would expect a linear correlation, until there are saturating amounts of IgG present, and therefore, no additional binding of C1q. A polynomial trendline allowing 3 orders of data fluctuation (negative, increasing slope, and saturation) was fitted by using the following equation to calculate the least squares fit through points: y = b + c1x + c2x2 + c3x3 + …+ c6x6 where b and C1…C6 are constants (Microsoft Excel).

RESULTS

Effect of EDTA Treatment and Dilution of Patient Test Sera on IgG-SAB MFI and C1q-SAB MFI

Luminex HLA class I SAB assays were undertaken using undiluted sera, EDTA treated sera, and sera diluted 1 in 20. The results were compared with those obtained for unmodified sera tested using the C1q-SAB assay (Figure 1).

FIGURE 1
FIGURE 1:
Effect of serum treatment on IgG-SAB and C1q-SAB binding. Undiluted sera (panel A), EDTA treated sera (panel B) and 1 in 20 diluted sera (panel C) obtained from 25 highly sensitised patients were tested using Luminex HLA class I IgG-SAB. The results (IgG-SAB MFI, x-axis) were compared with that obtained for undiluted sera tested using C1QScreen (C1q-SAB MFI, y-axis). The results show improved correlation coefficients (r 2) between IgG-SAB MFI and C1q-SAB MFI after correction for the prozone effect (EDTA-treated sera) and taking account of high titre IgG HLA specific antibody (diluted sera) compared with the conventional assay performed using untreated sera.

The correlation between IgG-SAB MFI using undiluted sera and C1q-SAB MFI (Figure 1, panel A) was low (r2 = 0.418) with many SAB populations showing high MFI values for IgG-SAB but low MFI for C1q-SAB and vice versa. This suggests that many sera contain high-level IgG HLA class I–specific antibodies that do not bind C1q and low-level IgG with strong C1q binding. The addition of EDTA to obviate complement interference in the IgG-SAB assay (Figure 1, panel B) improved the correlation between the MFI values for IgG-SAB and C1q-SAB (r2 = 0.568). Sera that previously displayed low-level IgG binding but strong C1q-SAB binding were revealed by EDTA treatment to contain high-level IgG-SAB binding. After EDTA treatment sera still displayed high level IgG-SAB MFI against some antigen specificities that did not bind C1q-SAB. Dilution of test sera to identify only high titre HLA-specific antibodies (Figure 1 panel C) further improved the correlation between MFI values for IgG-SAB and C1q-SAB (r2 = 0.769).

Analysis of Conformationally Folded and Denatured HLA Class I Protein Expression on SABs

Figure S1 (SDC,http://links.lww.com/TP/B289) shows the levels of conformationally folded (native) and of denatured HLA class I antigen (W6/32 and HC-10 mAb binding respectively) bound to the surface of HLA class I SABs. The levels of native HLA class I antigen (W6/32 mAb binding) bound to the different bead populations was remarkably similar for all of the HLA-A and -B specificities and for most (13 of 16, 81%) HLA-C specificities. In contrast, the levels of denatured HLA class I antigen detected (HC-10 mAb binding) varied markedly between different bead populations and ranged between 19% and 91% (mean, 69%; SD, 21%) of maximal HC-10 binding on the beads (Figure 2). Nine of the 31 HLA-A specificities (29%) expressed low level (≤30%) denatured HLA antigen, whereas all 49 HLA-B and all 16 HLA-C bead specificities expressed greater than 30% denatured HLA.

FIGURE 2
FIGURE 2:
Assessment of levels of denatured HLA class I protein on single antigen bead populations. The level of denatured HLA class I protein expressed on SAB was determined by comparing HC-10 MFI value for each SAB population tested using untreated SAB as a percentage of maximum HC-10 MFI value tested using acid treated (denatured) SAB.

Effect of Denatured HLA Protein on IgG-SAB MFI and C1q-SAB MFI

HLA-SAB populations were stratified according to the level of bound denatured HLA, and the relationship with IgG-SAB MFI and C1q-SAB MFI was analyzed (Figure 3). For HLA-SAB with greater than 30% denatured HLA, the correlation coefficient between IgG-SAB MFI and C1q-SAB MFI was lower than that observed for SAB populations with 30% or less denatured HLA. This was the case for undiluted test sera (r2 = 0.401 vs 0.647), EDTA-treated sera (r2 = 0.555 vs 0.721), and diluted sera (r2 = 0.760 vs 0.861). These results indicate that denatured HLA on SAB can interfere with the C1q assay and hence lead to a poorer correlation between IgG-SAB MFI and C1q-SAB MFI: the difference was most marked using untreated and EDTA-treated sera, and less notable using diluted sera. Of note, SAB specificities that displayed high-level IgG-SAB MFI and low-level C1q-SAB MFI were predominantly found in SAB populations with greater than 30% denatured HLA, suggesting that denatured HLA protein interferes with the ability of IgG to bind C1q in this solid phase assay.

FIGURE 3
FIGURE 3:
Effect of denatured HLA class I protein expression on IgG-SAB and C1q-SAB binding. Undiluted sera (panels A and B), EDTA treated sera (panels C and D) and 1 in 20 diluted sera (panels E and F) were tested using Luminex HLA class I IgG-SAB. The results (IgG-SAB MFI, x-axis) were compared to that obtained for undiluted sera tested using C1QScreen (C1q-SAB MFI, y-axis). SAB populations were stratified into 2 groups according to 30% or less denatured HLA protein expression (panels A, C, and E) and greater than 30% denatured protein (panels B, D, and F). The results show that for each serum treatment, the correlation coefficient between IgG-SAB MFI and C1q-SAB MFI was higher for SAB populations that express low levels (≤30%) of denatured HLA compared to high levels (>30%) of denatured HLA.

We next considered the effect of higher cutoff levels of denatured HLA class I protein expressed on SAB on the correlation between IgG-SAB MFI and C1q-SAB MFI (Table 1). As the cutoff level of denatured HLA was increased, the correlation between IgG-SAB MFI and C1q-SAB MFI decreased progressively. This effect was more marked when undiluted test sera were used, than when diluted sera or sera treated with EDTA were used in the assays.

TABLE 1
TABLE 1:
Effect of denatured HLA protein level expressed on HLA class I SAB and the relationship between IgG-SAB MFI and C1q-SAB MFI

Relationship Between IgG-SAB MFI and C1q-SAB MFI for Individual Patient Sera

Analysis and display of sera from individual patients provided additional insights into the relationship between IgG-SAB MFI and C1q-SAB MFI. Individual patient sera showed 2 main patterns of HLA-SAB binding, as illustrated by the results for the 2 selected patient sera shown in Figure 4. Using undiluted test serum, IgG-SAB MFI and C1q-SAB MFI for both patients correlated poorly (r2 = 0.497 and 0.241, respectively). After correction for the complement interference using EDTA-treated sera and detection of high titre IgG antibodies using diluted sera, sera from patient 1 displayed a very good correlation between IgG-SAB MFI and C1q-SAB MFI (r2 = 0.959 and 0.983, respectively). In contrast, use of EDTA and serum dilution of serum from patient 2 produced only a modest improvement in the correlation between IgG-SAB MFI and C1q-SAB MFI (r2 = 0.720 and 0.819, respectively). When, for serum from patient 2, the effect of denatured HLA on C1q-SAB MFI was considered (Figure 5), the overall poor correlation observed for IgG-SAB MFI and C1q-SAB MFI was shown to be almost entirely restricted to SAB with greater than 30% denatured HLA (r2 = undiluted sera 0.170; EDTA treated sera 0.593; diluted sera 0.719) and a very close correlation was seen for EDTA treated and diluted sera with 30% or less denatured HLA (r2 = EDTA treated 0.949; diluted sera 0.975).

FIGURE 4
FIGURE 4:
Examples of IgG-SAB and C1q-SAB MFI binding profiles for 2 selected patients. Undiluted sera (panels A and B), EDTA treated sera (panels C and D) and sera diluted 1 in 20 (panels E and F) obtained from 2 selected patients (patient 1 and patient 2) were tested using Luminex HLA class I IgG-SAB. MFI values for IgG-SAB (x-axis) were compared to that obtained for undiluted sera tested using C1QScreen (C1q-SAB, y-axis). For patient 1 untreated serum, the correlation coefficient (r 2) between MFI values obtained using IgG-SAB MFI and C1q-SAB was poor, but after correction for the prozone effect (EDTA treated serum) and taking account of high level IgG (diluted serum), the correlation was near perfect. For patient 2, the correlation between IgG-SAB MFI and C1q-SAB MFI using untreated serum was poor, and EDTA treatment and serum dilution gave only a relatively modest improvement.
FIGURE 5
FIGURE 5:
Effect of denatured HLA class I protein expression on IgG-SAB and C1q-SAB binding profiles for patient 2. For patient 2 serum, MFI values obtained using IgG-SAB (x-axis) were compared with those tested using C1QScreen (C1q-SAB, y-axis). The results were stratified according to the level of denatured HLA (≤30% versus >30%) expressed on the different SAB populations. For SAB that expressed 30% or less denatured HLA there was a high correlation between MFI values obtained using IgG-SAB and C1q-SAB assays, particularly when using EDTA treated and diluted serum. In contrast, for SAB that expressed greater than 30% denatured HLA, for undiluted serum the correlation was very poor, and showed only a modest improvement using EDTA-treated and diluted serum.

DISCUSSION

The findings from the present study highlight limitations in the use of solid phase bead assays to differentiate between complement binding and noncomplement binding HLA-specific antibodies in the serum of highly sensitized patients. The standard IgG-SAB assay has revolutionized HLA alloantibody screening, but it detects all IgG subclasses regardless of their ability to fix complement. It has been proposed that IgG antibodies that are complement binding are of greater clinical significance for predicting kidney transplant outcome.6-10,21-25 The introduction of the C1q-SAB assay as a tool to differentiate complement binding from noncomplement binding IgG is therefore a very promising development.

Although early studies using C1q-SAB suggested that detection of C1q binding DSA was independent of alloantibody level,10 no correction was made for interfering factors that cause a misleadingly low assessment of IgG-SAB binding.11-14,26-29 The strong association of C1q and C3d donor-specific antibodies detected using Luminex SAB assays with antibody-mediated rejection and graft loss observed by Loupy et al10 and Sicard et al30 provides an important basis for prognostic monitoring and to guide treatment options. It remains to be shown, however, that in vitro detection of complement products in solid phase assays is independent of antibody strength.31 The present analysis confirms that undiluted sera tested in the standard IgG-SAB assay gives results that correlate poorly with those obtained using the C1q-SAB assay, in keeping with the hypothesis that the 2 assays detect functionally distinct alloantibody populations. However, when antibody level and the contribution of complement interference are taken into account, there is a clear correlation between the presence of high level IgG detected by IgG-SAB and the ability to bind C1q in the C1q-SAB assay. This finding is in keeping with other recent studies questioning whether the C1q-SAB assay provides useful additional information.32 There has been growing awareness that interfering factors29,33,34 and antibody level31,32 are important and the correlation between C1q-SAB binding and worse graft outcome may be an indication of antibody strength. In addition to C1q complement interference in HLA solid phase assays, Schwaiger et al33 also demonstrated a role for complement split product, in particular C3d and C4d, as a cause of complement interference. Studies have shown that de novo formation of posttransplant donor-HLA–specific antibodies may be associated with C4d deposition in the capillaries of the graft and graft failure (reviewed in Böhmig et al35). It is, however, likely that denatured HLA class I might pose similar limitations on other solid phase assay-based complement detection techniques for C4d and C3d.14,30,35

A limitation of the present study is that we did not investigate potential complement fixing and noncomplement fixing IgG isotypes, although most sera contain a mixture of IgG isotypes and very few contain DSA with noncomplement fixing IgG2 and IgG4 isotypes alone.11 Otten et al20 reported that donor HLA-specific antibodies can corecognize intact and denatured HLA. They are present in the serum of patients in the absence of an allosensitization event and may have high MFI, but are not associated with renal transplant rejection.20,36,37 The presence of antibodies to both denatured and native HLA on SAB could, therefore, give an artificially high MFI to bead specificities that carry large amounts of denatured HLA protein on their surface. Further, such antibodies might also bind C1q in the C1q-SAB assay, but have no clinical relevance after kidney transplantation.

Correction for the level of alloantibody alone in the IgG-SAB assay does not fully account for the differences obtained with IgG-SAB and C1q-SAB and there remain examples of high level IgG binding that do not bind C1q. The novel aspect of the present study is the observation that the presence and level of denatured HLA protein on SAB may interfere with the ability of IgG to bind C1q in the solid phase assay. After correction for complement interference (using EDTA-treated sera) and HLA-specific antibody titre (using diluted sera), and taking account of denatured HLA protein expressed on HLA-SAB, the correlation between results obtained using IgG-SAB and C1q-SAB was shown to be very good. This suggests that a positive IgG-SAB but negative C1q-SAB result does not necessarily indicate the presence of noncomplement fixing antibodies, but instead, may indicate interference by denatured HLA in the C1q solid phase assay.

Antibody-mediated activation of the classic pathway of complement is initiated when C1q binds the Fc region of antigen bound IgG. The relatively low affinity of the interaction between IgG-Fc and C1q is strengthened through the high avidity provided by the hexameric C1q molecule.38 In physiological conditions, antigen present within a cell membrane are crosslinked by IgG monomers to form antigen/antibody clusters that enable multivalent C1q to bind exposed Fc regions of multiple IgG molecules, giving a higher C1q binding constant. In contrast, in a solid phase assay, immobilized antigen bound to polystyrene beads is not able to form areas of high IgG/antigen clusters. In this situation, the ability of C1q to bind IgG-Fc is dependent on correct antigen/antibody spatial orientation that requires permissive spacing of antigen bound IgG-Fc to enable high avidity multivalent C1q binding.13 Therefore, the combination of high SAB antigen density and high IgG levels are critical for C1q binding. In addition, the presence of denatured antigen on SAB interferes with native antigen spacing and may interfere with the ability of complement binding IgG isotypes to bind C1q.

In the present study, expression on SAB of high levels of conformationally folded HLA class I heavy chain/β2 microglobulin protein was, with the exception of 3 HLA-C locus antigens, remarkably consistent. This contrasts with early experience of Luminex HLA-SAB kits that showed overall lower and highly variable antigen expression on SAB populations.39 It seems, however, that consistently high levels of native HLA class I protein expressed on most SAB populations has been achieved, in some cases, at the expense of high levels of denatured (free class I heavy chain) HLA protein, and this is most notable for HLA-B and HLA-C.20 It may be that the manufacturing process to increase the levels of HLA protein has contributed to the variable levels of denatured protein or that some HLA specificities are more stable than others. The absolute percentages of HC-10 binding (before and after acid treatment) shown in Figure 2 are likely to be influenced by (a) the extent to which HC-10 binds the given HLA class I specificity (because there is published evidence that HC-10 binds different specificities to a variable extent), and (b) the “nonspecific” effects of acid treatment on antigen, which means the percentage of binding before acid treatment and as a percentage of total binding after acid treatment are likely to be proportionate.

It has been reported by Tran et al40 that W6/32 can bind both complete HLA molecules and free HLA-B heavy chains, although the absence of W6/32 binding free heavy chain expressed on acid-treated SAB clearly demonstrates that W6/32 does not bind to a significant extent to denatured HLA class I protein (all MFI <150).20 Although expression of denatured HLA will have little or no effect on IgG binding to native HLA proteins in the conventional IgG-SAB assay, our data suggest that high levels of denatured HLA causes technical interference in the C1q-SAB assay that can produce misleading results. Our data indicate that the effect of increasing levels of denatured HLA on SAB is most notable using untreated sera, but there is little effect using diluted sera. The likely explanation for the less pronounced correlation is that when sera are diluted, only high titre HLA-specific antibodies are detected by binding to SAB, and the ability of these to bind C1q is less influenced by the presence of denatured HLA on SAB.

It should also be remembered that the patient cohort in our present study were particularly highly sensitized, and none were sensitized by blood transfusion alone. It is not, therefore, clear if our findings are also applicable to patients with lower levels of sensitisation (eg, by blood transfusion alone), but the ability of HLA antibodies to bind C1q is likely to relate to antibody titre and affinity and not the nature of the alloantibody priming event per se. Further, the thrust of our study was to highlight the technical limitations of the C1q-SAB, and we did not therefore seek to correlate such data with biopsy findings and clinical outcomes which was beyond the scope of this study.

Given our findings with respect to HLA class I, analysis of denatured HLA class II on C1q binding would now be of interest, but again was beyond the scope of the present study. Although there are reports of suspected antibody binding to denatured HLA class II molecules in Luminex SAB assays, there is no direct evidence that antibody binding is to denatured protein.41,42 A potential problem of undertaking a similar analysis for HLA class II is that we are unaware of comparable mAbs to HC-10 that detect denatured HLA class II. To undertake such a study would ideally require analysis of free alpha and β chains of HLA-DR, -DQ, and -DP, and we are not aware that appropriate reagents are available to perform such a study.

In conclusion, the present study shows that antibody level, interfering factors, and the presence of denatured HLA protein on class I SAB may all affect the clinical interpretation of the C1q assay. Interest in the C1q binding assay to detect clinically relevant HLA antibodies continues to increase, and our study highlights the importance of technical factors that may affect its clinical interpretation. Given these uncertainties and the considerable costs involved, the addition of the C1q-SAB assay into routine clinical practice is difficult to justify.

REFERENCES

1. Lefaucheur C, Loupy A, Hill GS, et al. Preexisting donor-specific HLA antibodies predict outcome in kidney transplantation. J Am Soc Nephrol. 2010;21:1398–1406.
2. Gloor JM, Winters JL, Cornell LD, et al. Baseline donor-specific antibody levels and outcomes in positive crossmatch kidney transplantation. Am J Transplant. 2010;10:582–589.
3. Otten HG, Verhaarb MC, Borsta HPE, et al. Pretransplant donor-specific HLA class-I and -II antibodies are associated with an increased risk for kidney graft failure. Am J Transplant. 2012;12:1618–1623.
4. Pei R, Lee JH, Shih NJ, et al. Single human leukocyte antigen flow cytometry beads for accurate identification of human leukocyte antigen antibody specificities. Transplantation. 2003;75:43–49.
5. Vlad G, Ho EK, Vasilescu ER, et al. Relevance of different antibody detection methods for the prediction of antibody-mediated rejection and deceased-donor kidney allograft survival. Hum Immunol. 2009;70:589–594.
6. Chen G, Sequeira F, Tyan DB. Novel C1q assay reveals a clinically relevant subset of human leukocyte antigen antibodies independent of immunoglobulin G strength on single antigen beads. Hum Immunol. 2011;72:849–858.
7. Sutherland SM, Chen G, Sequeira FA, et al. Complement-fixing donor-specific antibodies identified by a novel C1q assay are associated with allograft loss. Pediatr Transplant. 2012;16:12–17.
8. Yabu JM, Higgins JP, Chen G, et al. C1q-fixing human leukocyte antigen antibodies are specific for predicting transplant glomerulopathy and late graft failure after kidney transplantation. Transplantation. 2011;91:342–347.
9. Lachmann N, Todorova K, Schulze H, et al. Systematic comparison of four cell- and Luminex-based methods for assessment of complement-activating HLA antibodies. Transplantation. 2013;95:694–700.
10. Loupy A, Lefaucheur C, Vernerey D, et al. Complement-binding anti-HLA antibodies and kidney-allograft survival. N Engl J Med. 2013;369:1215.
11. Schaub S, Hönger G, Koller M, et al. Determinants of C1q-binding in the single antigen bead assay. Transplantation. 2014;98:387–393.
12. Tinckam KJ, Heeger PS. Transplantation: Complementing donor-specific antibody testing. Nat Rev Nephrol. 2013;9:713–714.
13. Peacock S, Kosmoliaptsis V, Bradley AJ, et al. Questioning the added value of Luminex single antigen beads to detect C1q binding donor HLA-specific antibodies. Transplantation. 2014;98:384–386.
14. Visentin J, Vigata M, Daburon S, et al. Deciphering complement interference in anti-human leukocyte antigen antibody detection with flow beads assays. Transplantation. 2014;98:625–631.
15. Fuggle SV, Martin S. Tools for human leukocyte antigen antibody detection and their application to transplanting sensitized patients. Transplantation. 2008;86:384–390.
16. Schnaidt M, Weinstock C, Jurisic M, et al. HLA antibody specification using single-antigen beads—a technical solution for the prozone effect. Transplantation. 2011;92:510–515.
17. Barnstable CJ, Bodmer WF, Brown G, et al. Production of monoclonal antibodies to group A erythrocytes, HLA and other human cell surface antigens—new tools for genetic analysis. Cell. 1978;14:9–20.
18. Martayan A, Sibilio L, Tremante E, et al. Class I HLA folding and antigen presentation in beta 2-microglobulin-defective Daudi cells. J Immunol. 2009;182:3609–3617.
19. Stam NJ, Vroom TM, Peters PJ, et al. HLA-A- and HLA-B-specific monoclonal antibodies reactive with free heavy chains in western blots, in formalin-fixed, paraffin-embedded tissue sections and in cryo-immuno-electron microscopy. Int Immunol. 1990;2:113–125.
20. Otten HG, Verhaar MC, Borst HP, et al. The significance of pretransplant donor-specific antibodies reactive with intact or denatured human leucocyte antigen in kidney transplantation. Clin Exp Immunol. 2013;173:536–543.
21. Lawrence C, Willicombe M, Brookes PA, et al. Preformed complement-activating low-level donor-specific antibody predicts early antibody-mediated rejection in renal allografts. Transplantation. 2013;95:341–346.
22. Piazza A, Poggi E, Ozzella G, et al. Post-transplant development of C1q-positive HLA antibodies and kidney graft survival. Clin Transpl. 2013:367–375.
23. Freitas MC, Rebellato LM, Ozawa M, et al. The role of immunoglobulin-G subclasses and C1q in de novo HLA-DQ donor-specific antibody kidney transplantation outcomes. Transplantation. 2013;95:1113–1119.
24. Süsal C, Wettstein D, Döhler B, et al. Collaborative Transplant Study Report. Association of kidney graft loss with de novo produced donor-specific and non-donor-specific HLA antibodies detected by single antigen testing. Transplantation. 2015;99:1976–1980.
25. Calp-Inal S, Ajaimy M, Melamed ML, et al. The prevalence and clinical significance of C1q-binding donor-specific anti-HLA antibodies early and late after kidney transplantation. Kidney Int. 2016;27:293–304.
26. Kosmoliaptsis V, O'Rourke C, Bradley JA, et al. Improved Luminex-based human leukocyte antigen-specific antibody screening using dithiothreitol-treated sera. Hum Immunol. 2010;71:45–49.
27. Lefaucheur C, Viglietti D, Bentlejewski C, et al. IgG donor-specific anti-human HLA antibody subclasses and kidney allograft antibodymediated injury. J Am Soc Nephrol. 2016;27:293–304.
28. Weinstock C, Schnaidt M. The complement-mediated prozone effect in the Luminex single-antigen bead assay and its impact on HLA antibody determination in patient sera. Int J Immunogenet. 2013;40:171–177.
29. Zachary AA, Lucas DP, Detrick B, et al. Naturally occurring interference in Luminex assays for HLA-specific antibodies: characteristics and resolution. Hum Immunol. 2009;70:496–501.
30. Sicard A, Ducreux S, Rabeyrin M, et al. Detection of C3d-binding donor-specific anti-HLA antibodies at diagnosis of humoral rejection predicts renal graft loss. J Am Soc Nephrol. 2015;26:457–467.
31. Tambur AR, Herrera ND, Haarberg KM, et al. Assessing antibody strength: comparison of MFI, C1q, and titer information. Am J Transplant. 2015;15:2421–2430.
32. Yell M, Muth BL, Kaufman DB, et al. C1q binding activity of de novo donor-specific HLA antibodies in renal transplant recipients with and without antibody-mediated rejection. Transplantation. 2015;99:1151–1155.
33. Schwaiger E, Wahrmann M, Bond G, et al. Complement component C3 activation: the leading cause of the prozone phenomenon affecting HLA antibody detection on single-antigen beads. Transplantation. 2014;97:1279–1285.
34. Guidicelli G, Anies G, Bachelet T, et al. The complement interference phenomenon as a cause for sharp fluctuations of serum anti-HLA antibody strength in kidney transplant patients. Transpl Immunol. 2013;29:17–21.
35. Böhmig GA, Kikic Z, Wahrmann M, et al. Detection of alloantibody-mediated complement activation: a diagnostic advance in monitoring kidney transplant rejection? Clin Biochem. 2016;49(4–5):394–403.
36. Gombos P, Opelz G, Scherer S, et al. Influence of test technique on sensitization status of patients on the kidney transplant waiting list. Am J Transplant. 2013;13:2075–82.
37. Visentin J, Guidicelli G, Bachelet T, et al. Denatured class I human leukocyte antigen antibodies in sensitized kidney recipients: prevalence, relevance, and impact on organ allocation. Transplantation. 2014;98:738–744.
38. Diebolder CA, Beurskens FJ, de Jong RN, et al. Complement is activated by IgG hexamers assembled at the cell surface. Science. 2014;343:1260–1263.
39. Kosmoliaptsis V, Bradley JA, Peacock S, et al. Detection of immunoglobulin G human leukocyte antigen-specific alloantibodies in renal transplant patients using single-antigen-beads is compromised by the presence of immunoglobulin M human leukocyte antigen-specific alloantibodies. Transplantation. 2009;87:813–820.
40. Tran TM, Ivanyi P, Hilgert I, et al. The epitope recognized by pan-HLA class I-reactive monoclonal antibody W6/32 and its relationship to unusual stability of the HLA-B27/beta2-microglobulin complex. Immunogenetics. 2001;53:440–446.
41. In JW, Rho EY, Shin S, et al. False-positive reactions against HLA class II molecules detected in Luminex single-antigen bead assays. Ann Lab Med. 2014;34:408–410.
42. Grenzi PC, de Marco R, Silva RZ, et al. Antibodies against denatured HLA class II molecules detected in luminex-single antigen assay. Human Immunol. 2013;74:1300–1303.

Supplemental Digital Content

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.