Review article: Luminex technology for HLA antibody detection in organ transplantation

Authors


Dr Brian D Tait, Victorian Transplantation and Immunogenetics Service, 2nd Floor Rotary Bone Marrow Research Building, c/o Royal Melbourne Hospital, Grattan St, Parkville, Vic. 3050, Australia. Email bdtait@arcbs.redcross.org.au

SUMMARY

Since its inception in the early 1960s, the serologically based complement-dependent cytotoxicity (CDC) assay has been the cornerstone technique for the detection of human leucocyte antigen (HLA) antibodies, not only in pre-transplant renal patients, but also in other forms of organ transplantation. Recently, solid phase assays have been developed and introduced for this purpose, and in particular the Flow-based bead assays such as the Luminex system. This latter assay has proved to be far more sensitive than the CDC assay and has revealed pre-sensitization in potential transplant recipients not detected by other methods of HLA antibody detection. However, the clinical implications of this increased sensitivity have not been convincingly demonstrated until recently. This technology for HLA antibody detection permits the evaluation of the clinical importance of antibodies directed at, for example, HLA-DPB1 and HLA-DQA1, which has not been possible to date. There are Luminex issues, however, requiring resolution such as the ability to distinguish between complement fixing and non-complement fixing antibodies and determination of their relative clinical significance. Luminex technology will permit a re-evaluation of the role of HLA antibodies in both early and late antibody-mediated rejection.

Since the first demonstration by Patel and Terasaki1 that pre-formed donor-specific human leucocyte antigen (HLA) class 1 antibodies (DSA) in renal transplant recipients caused hyperacute rejection, there has been an imperative to test all potential recipients prospectively for HLA antibodies to avoid transplanting incompatible grafts. Until recently, the method universally used for antibody detection was the complement-dependent cytotoxicity (CDC) assay,2 which involves incubating test sera with cells of known HLA type followed by the addition of rabbit serum as a source of complement, and the use of cell dyes to assess the degree of cell death. Variations to this technique have been added over the years to improve sensitivity such as the use of anti-human globulin as a second antibody3 and the use of extended incubation times.4 By testing sera against a panel of cells of known HLA type covering the majority of known HLA antigens, antibody specificities can be determined and a per cent reactive antibody (PRA), which is the percentage of cells in a random panel giving positive results with an antiserum, can be calculated. The same technique has been used for crossmatching recipients against potential donors to determine their suitability. The PRA therefore is particularly useful in estimating the probability of a patient receiving a crossmatch-negative kidney. In more recent years, the same technique has been used in matching heart, lung, liver and pancreas transplants.

Given this background, the consensus view until the early 1990s was that patients with no detectable DSA and a negative CDC crossmatch would be considered suitable candidates for transplantation. The introduction of Flow Cytometry crossmatching changed perceptions with the discovery that this technique could detect donor-specific HLA antibodies, which in some cases was associated with rejection,5–7 in the presence of a negative donor CDC crossmatch. In recent years, solid phase assays have been introduced as methods for HLA antibody screening which have again redefined the definition of pre-sensitization. These assays which basically fall into two categories, namely ELISA-based methods8–10 and HLA antigen-coated beads used in either a Flow Cytometry system11 or a Luminex platform,12 are more sensitive than CDC for detecting both HLA class 1 and class 2 antibodies.13–15

The ELISA method essentially consists of either groups or individual HLA molecules bound to the surface of plastic wells. The test serum is added to the well which binds to the HLA molecule of appropriate specificity, a second enzyme-linked anti-human IgG antibody is added followed by a suitable substrate, the readout being a colour change. The Luminex platform-based assay has demonstrated even greater sensitivity than ELISA16,17 and has become the most popular method of HLA antibody detection, and is the subject of the remainder of this review.

LUMINEX HLA ANTIBODY SCREENING TECHNOLOGY

The Luminex antibody screening technology consists of a series of polystyrene microspheres (beads), containing fluorochromes of differing intensity embedded within the bead giving each group of beads with an HLA molecule or molecules derived from lymphoblastoid cell lines attached a unique signal . . . There are three levels of bead attachment. The first level consists of beads bound with a large number of class 1 or class 2 molecules which essentially provides a positive or negative result. At the second level, the bead is equivalent to a cell with each bead containing two molecules derived from two alleles at each loci; HLA-A, -B and -C in the case of class 1 and HLA-DR, -DQ in the case of class 2. The third level consists of beads with one HLA molecule attached (either class 1 or 2) referred to as a single antigen bead (SAG). This third level is particularly useful for characterizing complex sera with a high PRA, accurately defining the antibodies present.

Test sera are added to the bead mix and the HLA antibodies bind to the bead with the appropriate HLA molecule attached. A second phycoerythrin-labelled anti-human IgG antibody is then added which binds to the primary HLA antibody. When the sample is passed through the detector, one laser excites the fluorochrome in the bead which exhibits a unique signal and the other laser excites phycoerythrin bound to the second antibody. The combination of these signals defines the specificity of the antibody in the test serum. The test method is depicted in Figure 1.

Figure 1.

The top panel depicts the basic principles of the Luminex bead assay (LABScreen, One Lambda Inc.). Shown is a bead with a human leucocyte antigen (HLA) molecule (antigen) attached. If the appropriate HLA antibody is present in the test serum, it binds to the HLA molecule. Each HLA-specific bead gives a unique signal when excited by one of the lasers built into the Luminex instrument due to a specific intensity of fluorochrome embedded into the bead. The excitation and detection of the signal from phycoerythrin (PE) bound to the second anti-human IgG antibody by the second laser indicates the presence of specific HLA antibody. The second panel shows the steps involved in preparing assay for measurement by the Luminex machine. The test serum and beads are incubated together at room temperature (RT) for 30 minutes, washed three times with buffer and a second antibody is then added and incubated at RT for 30 minutes. After a further two washes, the mixture is re-suspended in phosphate-buffered saline (PBS) and analysed in the Luminex instrument. The figure shown was supplied by One Lambda Inc.

The remainder of the review will focus on the advantages of this technology over the conventional CDC technique which has been in regular use for the last 40 years.

SENSITIVITY AND RELEVANCE TO CLINICAL OUTCOME

Published data indicate the Luminex platform is the most sensitive of the solid phase antibody detection techniques. Gibney et al.18 reported on 136 renal transplant patients tested by Luminex of whom 16 had pre-transplant antibodies defined as a CDC PRA >30%. In comparison, 55 patients had antibodies for either HLA class 1 or class 2 detectable by the solid phase assay and defined as a Luminex PRA >15%. Twenty out of these 55 patients were shown to have DSA. These 20 patients had a significantly higher rate of primary or delayed graft function, biopsy proven acute rejection (BPAR), and lower rates of graft survival at 6 months. In particular, the BPAR incidence was 25% in this group compared with 3% in the non-DSA group (P < 0.01). These results indicate that the Luminex technique was detecting additional clinically relevant HLA antibodies.

Smith et al.19 reported similar results in a cohort of 565 cardiac transplant recipients whose pre-transplant sera were retrospectively tested for the presence of HLA antibodies using CDC and Luminex technology. Fourteen patients developed class 1 or class 2 antibodies antibodies pre-transplant as defined by a CDC PRA of >5% with a strength of reactivity >10% above background, of whom five were shown to have DSA resulting in a positive CDC crossmatch with the donor. Luminex testing at level 1 revealed a further 53 patients with HLA antibodies, 14 of whom had DSA identified using SAG. The 1-year actuarial graft survival for patients with CDC-positive DSA who were also Luminex-positive was 40% compared with 42% for those patients with Luminex-positive CDC-negative detected DSA. For patients with non-DSA, the graft survival was 75% for CDC and 77% for Luminex. The graft survival for patients with no detectable antibodies was 75%. The results obtained for DSA-positive compared with DSA-negative and antibody-negative patients were highly statistically significant in both the CDC-positive, Luminex-positive and CDC-negative, Luminex-positive groups. These results indicate that a further subset of patients with DSA detectable only by Luminex are at high risk of graft rejection.

There are reports which have addressed the issue of predicting CDC crossmatch results by Luminex screening, the so-called virtual crossmatch. For example, Cherikh et al.20 reported on a multi-centre study of 6620 crossmatches performed in 225 patients The predicted result of the CDC crossmatch was based on various techniques for antibody screening. Solid phase assays including Luminex and Flow cytometry had a higher predictive value for a positive CDC crossmatch than CDC antibody screening, 84.7–84.8% for T-cells and 91–97.9% for B-cells. The higher rate for B-cells is due to the fact that they detect both HLA class 1 and class 2 antibodies in a CDC assay. The prediction of a negative CDC crossmatch was lower for solid phase assays and Flow methods (50.9%) than CDC PRA (66.7%). These results presumably reflect additional DSA detected by solid phase assays which were predicted to give a positive CDC crossmatch but in fact were negative. The low rate of concordance between CDC antibody identification and crossmatch prediction may reflect the failure of the CDC method, which is based on the probability of an antibody being present, to determine specificity accurately particularly in sensitized individuals with a high PRA.

Vaidya et al.21,22 assessed HLA antibodies detected by Luminex as a predictor of both CDC and Flow crossmatching. The Luminex antbody detection assay correctly predicted Flow B- and T-cell crossmatch results in 93% of cases with a 1% false positive and 6% false negative rate. As Flow crossmatches are performed with cells and not purified HLA molecules as used in the Luminex assay, the latter's false negative rate may be partly due to non-HLA antibodies. The predictive rates for Flow crossmatch results compares with prediction rates of 79% and 68% for T- and B-cell CDC crossmatch results, respectively. In this study, the authors quantitated HLA antibodies detected by Luminex using MESF (Molecules of Equivalent soluble Fluorochromes).22 They demonstrated that CDC crossmatches had a threshold of 250 000, below which they were negative. By contrast, Flow crossmatches were able to detect DSA below 20–25 000 MESF indicating Flow crossmatching is at least 10-fold more sensitive than CDC crossmatching. The results also indicate a similar level of sensitivity of the Luminex antibody detection assay with Flow crossmatching given the level of agreement between the two techniques. With this high level of agreement, an argument can be made for using Luminex antibody results as a virtual crossmatch.

The use of MSEF also permits the study of the strength of HLA antibodies on clinical outcome. Mizutani et al.23 demonstrated that there was a direct correlation between MSEF and the titre of antibody, and by converting fluorochrome values to MSEF they further showed that patients who had subsequent graft failure had higher titre antibodies than those patients whose grafts continued to function. This approach has potential for the ongoing monitoring of HLA antibodies post transplant and for identification of patients with higher risks of graft failure.

OTHER FACTORS THAT REFLECT DIFFERENCES IN RESULTS BETWEEN SOLID PHASE ASSAYS AND CDC FOR HLA ANTIBODY SCREENING

Potentially the CDC method can detect any antibody directed at the cell surface of a B or T lymphocyte. In addition to HLA antibodies, these can include autoantibodies, antibodies directed at polymorphic non-HLA molecules and it is feasible that immune complexes or aggregated IgG combine with the Fc receptor on the B lymphocyte and under the conditions of the CDC test give the appearance of antibody. Non-HLA antibodies and immune complexes are not detected by the Luminex system. IgM HLA antibodies are readily detected by the CDC method and can be detected by the Luminex method if a second IgM antibody is utilized. However, there is no clear consensus on the role these antibodies may play in renal graft rejection.24,25 As a result of the above differences, there is some difficulty in precisely determining the true difference in sensitivity of the two assays as no study has taken all these factors into account. However, there is overwhelming evidence from individual case studies and reports presented at international meetings12,13,18,19 that the Luminex assay has far greater sensitivity than that of complement-dependent assays in detecting HLA pre-sensitization.

Conversely, as the CDC assay is complement-dependent, non-complement fixing antibodies of the IgG2 and IgG4 class will not be detected by this method and these can account for a substantial percentage (approximately 28%) of antibodies eluted from explanted renal allografts.26

NON-COMPLEMENT FIXING ANTIBODIES

Because Luminex technology detects all classes of IgG, it does not discriminate between complement fixing and non-complement fixing HLA antibodies. The critical issue is whether the non-complement fixing antibodies are detrimental to the graft and, if not, are we excluding patients from receiving compatible grafts using Luminex technology for HLA antibody identification?

Wahrmann et al.27,28 adapted a Flow-based bead assay (Flow PRA, One Lambda) to measure complement fixing antibodies. Pre-transplant sera from 338 out of 352 consecutive transplant patients were tested in parallel using the standard Flow-based bead and a modified assay using normal human serum as a source of C4d which covalently linked to the beads in the presence of complement fixing antibody. A second FITC-labelled anti-C4d antibody was used for detection. Surprisingly 35% of Flow detected HLA class 1 antibodies and 64% of class 2 antibodies were shown to be non-complement fixing by this method. As the sera were collected pre-transplant, both donor-specific and non-specific antibodies were tested. Wahrmann et al.28 further demonstrated the clinical relevance of complement fixing antibodies in the cohort of 352 renal patients. Patients with C4d+ Flow+ HLA class 1 antibodies had a significantly inferior graft survival (75% at 3 years) compared with C4d− Flow+ and Flow-negative patients (91% and 89% respectively). However, complement fixing HLA class 2 antibodies were not associated with inferior graft survival compared with the other groups. Of further interest was the fact that in the class 1 antibody-positive group the C4d− Flow+ group had a higher rate of C4d associated graft dysfunction than the Flow-negative group as demonstrated by renal biopsies. This may reflect on the sensitivity of the C4d assay in that some of the complement fixing antibodies were recorded as negative or a reflection of conversion to complement fixing antibodies in vivo.28

Smith et al.19 demonstrated similar results to Wahrmann et al. in 565 cardiac transplant recipients. In this case, they modified the Luminex-based assay in a similar fashion to the method described by Wahrmann et al.27 utilizing fresh human serum as a source of complement. Using a mouse anti-human C4d monoclonal reagent and a phycoerythrin-labelled donkey anti-mouse IgG, they were able to demonstrate a profound effect of complement fixing donor-specific HLA antibodies on graft survival. The 1-year graft survival for patients with C4d+ DSA was 20%, 54% for C4d− DSA patients and 91% for C4d+ non-DSA patients. The interesting observations are the profound negative effect of the complement fixing antibodies on graft survival and also the reduction in graft survival of the C4d− DSA+ group.

Heinemann et al.26 demonstrated by using antibodies specific for the four isotypes of IgG (IgG1-4) that non-complement fixing HLA antibodies could be demonstrated in the eluates of 28% of 58 rejected kidneys. The issue is whether these antibodies play a role in graft destruction or whether they are simply bystanders. Nevertheless, if the results obtained by Wahrmann et al.28 and Smith et al.19 can be confirmed, it would indicate a need to discriminate between complement fixing and non-complement fixing antibodies either by using antibodies to the IgG isotypes or by the more functional C4d assay as described. The delineation of complement fixing and non-complement fixing antibodies using Luminex technology would clearly be of value to clinicians in determining whether to proceed with a transplant and will also be of assistance in planning post transplant pre-emptive rejection treatment.

OTHER SPECIFIC ADVANTAGES OF THE LUMINEX PLATFORM

Single bead antibody identification

The outstanding advantage of the Luminex platform is the introduction of SAG. SAG allows complex sera reacting with many HLA molecules to be dissected and accurate antibody specificities assigned as described fully by El-Awar et al.29 Often these multiple reactions are directed at one or two epitopes which are shared across several HLA class1 or class 2 molecules. With the CDC method, the specificities of sera with complex reactivity patterns are assigned based on mathematical probability. A more definitive answer is therefore provided by the Luminex assay representing a distinct advantage.

While both HLA class 1 and class 2 antibodies are associated with early graft rejection,30 there is evidence that HLA class 2 antibodies are more strongly associated with long-term chronic rejection.31 Luminex SAG technology permits evaluation of HLA class 2 antibody in pre- and post-transplant patients with a high level of sensitivity. An additional advantage of the Luminex platform utilizing SAG is that antibodies directed at loci such as HLA-DRB3,-DRB4,-DRB5, HLA-DQB1 and HLA-DPB1 can be clearly delineated from HLA-DRB1 activity. Antibodies with specificity for allelic products of HLA-DRB3, -DRB4, -DRB5 and HLA-DQB1 can be particularly problematical when tested by CDC because of the linkage disequilibrium between alleles at these loci.

Since the introduction of beads containing HLA-DPB1 molecules, there has been further understanding of the role that HLA-DP antibodies may play in renal rejection processes. For example, we have reported on a 32-year-old woman transplant recipient (BW)32 who received her first transplant in November 2001 from her 54-year-old father with donor mismatches HLA-A11,B35,Cw4,DRB1*0103,DQ1 and HLA-DPB1*0901. DSA A11, Cw4 and DRB1*0103 developed post transplant. The graft deteriorated and nephrectomy was performed in February 2006. BW's 60-year-old mother was considered a second potential donor. On HLA typing, it was established that BW's mother was HLA-A-,B,-C,-DR,-DQ identical. The only donor mismatch between BW and her mother was HLA-DPB1*0101. Luminex SAG testing at the time of transplant revealed antibodies to HLA-DRB1*0103,03,08,09,10,12, HLA-DQB01,04,05,06 and HLA-DPB1*0101,-0301,0501,1001,1101,1301 and 1701. A bead for HLA-DPB1*0901 was not available at that time. A Flow crossmatch was performed pre transplant which was T-cell-negative but B-cell-positive presumably because of the presence of HLA-DPB1*0101 in the donor and a decision was made to proceed with the transplant. Post transplant, the creatinine rose to 287 µmol/L on day 6 and a biopsy confirmed early humoral rejection but C4d testing was negative. Several rounds of plasma exchange with low-dose IVIG and an additional dose of 1 g/kg of IVIG on day 16 reduced the creatinine to 82 µmol/L which then remained low. A repeat biopsy 4 months post transplant revealed mild cellular rejection but the C4d remained negative. It appears that humoral rejection was caused by pre-existing immunization to the HLA-DPB1*0901 of the first donor, as examination of the allele sequences corresponding to the DPB1 antibodies detected revealed a shared epitope at positions 84–87 (DEAV) including the two donor DP mismatches, HLA-DPB1*0901 and HLA-DPB1*0101. This epitope seems to be particularly immunogenic as Piazza et al.33 reported on a series of patients who had undergone renal graft failure. DP-specific antibodies were found in 47% of 81 patients. Six of these patients were immunized only to DP, four of whom had been grafted with zero HLA-DR/-DQ mismatched kidneys. The remaining patients had DP antibodies in combination with DR and/or DQ antibodies. When DP sequences were correlated with DP antibody specificities detected in individual patients, the DEAV (84–89) motif was implicated in 51% of cases indicating the importance of this sequence for matching purposes.

Qiu et al.34implicated DP humoral rejection by identifying DP antibodies in 5.1% of 138 patients with a functioning graft and 19.5% of 185 patients with a rejected graft. In total, 13% of those who rejected a graft had a DP antibody in the absence of class 1 or DR/DQ antibodies. Arnold et al.35 showed that 24 of 27 patients on the transplant waiting list who had been previously grafted and had developed class 2 antibodies had DP included in their antibody specificities. Again in the majority of cases, the antibodies could be defined by shared epitopes. These results clearly implicate DP in some cases of rejection. With the advent of beads containing DP molecules, it is now possible to routinely screen patients for DP antibodies, which historically was difficult by CDC because of the lack of DP allele typed screening cells, and Luminex technology provides the opportunity to monitor patients for the presence of DP antibodies and to gather more data on their relevance in transplant rejection. These results also demonstrate clearly the detection of antibodies directed at shared epitopes which appear on first analysis as multispecific antibodies.

Once Luminex technology is established in the laboratory, it becomes possible to add to the repertoire of antibodies being screened. For example, beads containing HLA-C, HLA-DQA, -DQB and MICA molecules are now available which allows screening of antibodies to these specificities. The clinical usefulness of this is highlighted by convincing evidence that MICA antibodies occur relatively frequently in renal transplant patients and can be responsible for renal allograft rejection.36–38

Allele-specific antibodies

A further advantage of the Luminex assay is the ability to detect antibodies directed at allelic specific differences within a patient's own antigen group. Pancoska et al.39 described a male caucasian patient who had received multiple transfusions and rejected a kidney from a zero antigen-matched deceased donor. The pre-transplant ELISA PRA on the patient was negative and the AHG-enhanced CDC crossmatch was negative. The graft was slowly rejected because of recipient immmunization to the donor and the patient was shown to have antibodies to HLA-A1, A23 and A24. The HLA-A24 antigen was present in both recipient and donor. On retyping, the patient was shown to be HLA-A*2403 while the antibody was specific for the donor allele HLA-A*2402. The amino acid differences between HLA-A*2403 and HLA-A*2402 are aspartic acid and glycine substitutions replacing glutamic acid and tryptophan at amino acid positions 166 and 167, respectively, creating a serological epitope in HLA-A*2402 which can be recognized by an HLA-A*2403 individual. There are now numerous reports of this level of antibody discrimination using Luminex bead technology which appear to have clinical relevance. For example, an HLA-A*0302 individual was shown to produce an antibody to HLA-A*0301.40

These antibodies would not have been readily recognized in the past using CDC techniques as generically typed cells were used for antibody screening in most cases. Even with allele typing of screening cells, it is problematic whether precise allele specificities could be assigned with confidence using this technique.

Repeat HLA mismatches

Repeat HLA mismatches are HLA incompatibilities to which a patient has previously been exposed and are present on the current graft. There are two instances where repeat HLA mismatches can compromise the fate of a second graft.

First, multiparous females who are prospective solid organ transplant recipients who have not had a recent immunizing event are in many cases negative when screened by CDC but have covert sensitization. In many of these patients, the last pregnancy may be in excess of 20 years ago and HLA antibody is no longer detectable by CDC, but memory cells and low levels of antibodies are still present. In negative donor crossmatch heart transplant recipients, multiparous females have a high rate of early rejection when a spousal HLA mismatch is repeated on the transplant.41 This early rejection is presumably due to a failure to detect HLA pre-sensitization caused by a prior pregnancy. Luminex technology now provides the opportunity to detect these low level antibodies and avoid the corresponding antigen on the transplant. With routine use of Luminex technology, the higher rejection rates experienced in multiparous females may be overcome.

Second, mismatches present on a first graft have historically been considered a contraindication in subsequent grafts on the basis of likely pre-sensitization. However, in recent years several centres have reported no detrimental effect of repeating HLA mismatches.42–44 Conversely, Cecka and Terasaki45 demonstrated that class 2 repeat mismatches resulted in a poorer outcome in subsequent renal grafts than repeat class 1 mismatches. This observation could be explained by the fact that HLA class 1 CDC antibody screening can exclude many of the previous mismatches to which the patient is immunized. These antibodies will manifest as a positive crossmatch and hence be avoided. Historically, laboratories have not screened renal patients routinely for HLA class 2 antibodies by CDC and as a result a greater percentage of patients will be transplanted with class 2 antibodies directed at donor-specific repeat mismatches. With the advent of Luminex testing, the detection of sensitization to previous mismatches prior to subsequent transplantation is more likely to be detected and time will show whether the repeat mismatch concern of the last 30 years simply reflects a failure to adequately detect pre-sensitization, and that in the absence of donor-specific pre-sensitization second graft success rates will equal that of first grafts. Post transplant, HLA antibody monitoring would be recommended in such cases.

Virtual panel reactive antibody and virtual crossmatch

The use of Luminex, particularly SAG, permits the calculation of a virtual PRA based on the frequencies of the various antigens in the local population.46 This can replace the conventional calculation of PRA which is the per cent of a random panel positive by CDC. The use of such a virtual PRA can assist the clinicians in predicting the overall likelihood of positive crossmatches against deceased donors. Likewise, the use of SAG can be used as a predictor of a positive crossmatch with a prospective living or deceased donor.

Can Luminex HLA antibody screening replace Flow crossmatching?

Flow crossmatching is currently the most sensitive crossmatch used in transplant matching. There is abundant evidence that demonstrates this technique is far more sensitive than CDC and at least in many cases detects additional clinically relevant antibodies.47–49 However, some patients with CDC-negative and Flow-positive crossmatches have uneventful post transplant courses with no evidence of rejection. This raises the question as to whether the Flow crossmatch is too sensitive and is detecting antibodies which are not clinically relevant and thereby denying some patients a suitable graft. The advantage of the Luminex technique is that by definition all antibodies detected are HLA-directed. A comparison of donor-directed antibodies detectable by Luminex with donor Flow crossmatches will determine whether Luminex antibody testing can act as a virtual Flow crossmatch and therefore render the Flow crossmatch redundant remains to be seen. The relevance of donor-specific HLA antibodies detected by Luminex which are negative by Flow crossmatching and conversely the relevance of positive Flow crossmatches in the absence of detectable donor-specific HLA antibodies also remain to be seen.

Algorthim for the use of Luminex

With the minimal use of blood transfusion in renal patients, the need for continual screening of patients' sera for HLA antibodies in the absence of any known immunizing event seems excessive if one is able to accurately define the antibodies present. With the use of CDC screening, the rationale for repeated testing is to gain a consensus on the specificity of antibodies present. However, with the increased sensitivity and more accurate definition of antibody specificity, the need for repeated testing with Luminex is unnecessary provided accurate information is received from the clinical units concerning potential immunizing events. If that information is provided, screening on a three to six monthly basis seems reasonable.

The role of Luminex antibody testing in de-sensitization protocols

Many centres are now performing living-related donor renal transplants with an initial positive crossmatch by either CDC or Flow cytometry. Treatment with plasma exchange, IVIG and in some cases the anti-CD20 reagent Rituximab can diminish DSA present with the aim of obtaining a negative CDC crossmatch. The use of Luminex screening in these cases can be extremely useful because although the CDC crossmatch may convert to negative, the Luminex assay by virtue of its greater sensitivity can still detect the presence of DSA. Pre-transplant results obtained may guide post transplant treatment and in the long-term post-transplant monitoring by Luminex may provide information regarding chronic rejection in these patients. Because Rituximab is a monoclonal antibody directed at the CD20 molecule, it binds to B-cells and results in false positive reactions in the CDC assay. An additional benefit of screening by Luminex technology is that Rituximab does not interfere with the results obtained and allows a virtual crossmatch to be obtained in these circumstances. It is now possible to use cell lysates from the donor, immunoprecipitate the HLA molecules, attach them to the beads and perform a direct crossmatch using Luminex technology.50 This approach may eventually replace conventional CDC crossmatching.

Cost benefit of Luminex testing

Important questions with the introduction of any new technology include ‘what does it cost’, ‘do the benefits justify the cost’ and ‘what is the cost, in clinical and monetary terms of not doing the test?’ No studies have specifically examined the cost–benefit of Luminex technology in transplantation but recent assessment of Flow Cytometry microbead HLA antibody screening and de-sensitization protocols are available. In general, kidney transplantation allows greater survival for less cost compared with remaining on dialysis and in most instances immunological assessment and manipulation around the time of transplant is cost-neutral within 2 years of surgery. In Australia although the estimated first year cost of kidney transplantation ($62 375) is greater than the annual cost of either satellite or home haemodialysis ($44 739–$56 828), the low ongoing annual cost of transplantation ($10 749) makes it the economically preferable treatment after only 2 years of graft survival.51 Considering the current longevity of allografts, this amounts to a considerable saving. It is on this basis that de-sensitization protocols for highly sensitized patients are economically justified, despite the extra $30 000 initial cost. Previous economic analysis of these protocols confirms significant economic benefits and the additional cost of Luminex is small in comparison.52 Performing Flow Cytometry antibody screening rather than serological screening for wait listed patients has been shown to be cost-effective with a gain in life years, transplant life years and quality-adjusted life years per patient, despite the higher initial cost.53 In this context, not identifying DSA and relying entirely on older technologies may put the recipient at risk of early humoral rejection with possible graft loss, increased immunosuppression and longer-term consequences.

Ancillary