Secondary Logo

Journal Logo

Analysis of Luminex-based Algorithms to Define Unacceptable HLA Antibodies in CDC-crossmatch Negative Kidney Transplant Recipients

Zecher, Daniel MD1; Bach, Christian PhD2; Preiss, Adrian MD1; Staudner, Christoph1; Utpatel, Kirsten MD3; Evert, Matthias MD3; Jung, Bettina MD1; Bergler, Tobias MD1; Böger, Carsten A. MD1; Spriewald, Bernd M. MD2; Banas, Bernhard MD1

doi: 10.1097/TP.0000000000002129
Original Clinical Science—General
Free
SDC

Background HLA-specific antibodies detected by solid phase assays are increasingly used to define unacceptable HLA antigen mismatches (UAM) before renal transplantation. The accuracy of this approach is unclear.

Methods Day of transplant sera from 211 complement-dependent cytotoxicity crossmatch-negative patients were retrospectively analyzed for donor-specific anti-HLA antibodies (DSA) using Luminex technology. HLA were defined as UAM if DSA had mean fluorescence intensity above (I) 3000 (patients retransplanted and those with DSA against HLA class I and II) or 5000 (all other patients), (II) 5000 for HLA-A, -B, and -DR and 10 000 for HLA DQ or (III) 10 000 (all HLA). We then studied the accuracy of these algorithms to identify patients with antibody-mediated rejection (AMR) and graft loss. UAM were also determined in 256 transplant candidates and vPRA levels calculated.

Results At transplantation, 67 of 211 patients had DSA. Of these, 31 (algorithm I), 24 (II) and 17 (III) had UAM. Nine (I and II) and 8 (III) of 11 early AMR episodes and 7 (I), 6 (II) and 5 (III) of 9 graft losses occurred in UAM-positive patients during 4.9 years of follow-up. Algorithms I and II identified patients with persistently lower glomerular filtration rate even in the absence of overt AMR. Of the waiting list patients, 22–33% had UAM with median virtual panel reactive antibody of 69.2% to 79.1%.

Conclusions Algorithms I and II had comparable efficacy but were superior to Algorithm III in identifying at-risk patients at an acceptable false-positive rate. However, Luminex-defined UAM significantly restrict the donor pool of affected patients, which might prolong waiting time.

In a retrospective series of 211 kidney transplant recipients, the authors suggest 3 algorithms based on DSA-MFI in order to more accurately define unacceptable mismatches.

1 Department of Nephrology, University Hospital Regensburg, Regensburg, Germany.

2 Department of Internal Medicine 5-Hematology and Oncology, University Hospital Erlangen, Friedrich-Alexander University Erlangen-Nürnberg, Germany.

3 Institute of Pathology, Regensburg University, Regensburg, Germany.

Received 21 August 2017. Revision received 29 December 2017.

Accepted 3 January 2018.

The authors declare no funding or conflicts of interest.

D.Z. designed the study, performed data analysis, and wrote the article. C.B. performed Luminex analyses and data interpretation, and contributed to the writing of the article. C.S. and A.P. collected data. K.U. and M.E. performed histological analyses of kidney biopsies. T.B., B.J., C.A.B., B.S., and B.B. gave conceptual advice and contributed to the writing of the article.

Correspondence: Daniel Zecher, MD, Department of Nephrology Regensburg University Hospital Franz Josef Strauss-Allee 11 93042 Regensburg, Germany. (daniel.zecher@ukr.de).

Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.transplantjournal.com).

Ever since recipient immunization against foreign HLAs has been recognized as one of the major barriers to successful solid organ transplantation,1 the periodical determination of anti-HLA antibodies in the serum of patients on the waiting list and the subsequent definition of unacceptable HLA antigen mismatches (UAM) has become a routine procedure for HLA laboratories and transplant physicians. The determination of UAM is a critical decision step because it can prevent a considerable number of positive crossmatches and thus prevent futile organ shipments between transplant centers and loss of high-quality organs. However, the stricter the definition of UAM, the more donors are excluded for an individual patient, potentially leading to a significant prolongation of waiting time and an increase of patient death on dialysis.

It is commonly accepted to define all those HLA antibody specificities as UAM that are cytotoxic in cell-based complement-dependent cytotoxicity (CDC) assays as they have been associated with an increased risk of early antibody-mediated rejection (AMR) and graft loss.1 The introduction of more sensitive solid phase assays such as the single-antigen bead (SAB) assay using Luminex technology have greatly improved risk stratification before transplantation especially for the identification of low risk patients without detectable donor-specific anti-HLA antibodies (DSA).2 The relevance of a positive SAB test result, however, is less clear as the positive predictive value (PPV) of a DSA detected by Luminex testing for the occurrence of early AMR and graft loss in an individual patient is low.3,4 The strength of an antibody found in the Luminex assay is commonly reported as mean fluorescence intensity (MFI). Besides technical aspects that have to be considered when interpreting MFI levels as an approximate of biological strength of an antibody,5,6 neither the prognostic value of different MFI levels nor specific MFI cutoffs for the segregation of patients at risk from those without have clearly been established so far.6

As a consequence, UAM algorithms vary greatly between transplant centers.7 Whereas some centers only consider HLA specificities determined in cell-based CDC assays for the definition of UAM, others integrate Luminex test results into their algorithms but use varying MFI cutoffs with or without a plausibility check taking into account the patient’s immunization history.7 Until now, there have been only a few reports on the outcome of patients after the introduction of center-specific Luminex-based UAM (SAB-UAM) algorithms.7,8 Given their prospective design, these studies were unable to investigate the accuracy of these algorithms, as the clinical course of SAB-UAM-positive patients could not be evaluated. In an attempt to homogenize UAM algorithms between transplant centers, the German Society for Immunogenetics (DGI) recently published a consensus statement on the definition of SAB-UAM before renal transplantation.9 These UAM criteria, however, have not yet been clinically validated.

In our center, until the end of 2013, the assignment of UAM was exclusively based on CDC assays and information on pretransplant Luminex-testing was not routinely available and hence had no influence on risk stratification and subsequent choice of induction therapy at the time of transplantation. This allowed us to retrospectively compare the ability of 3 different SAB-UAM algorithms to segregate patients at risk for early AMR and graft loss from those without in a cohort of CDC crossmatch (CM)-negative renal transplant patients. Moreover, the HLA antibody profiles of an independent cohort of patients on the waiting list for renal transplantation were used to estimate the potential impact of these algorithms on organ allocation.

Back to Top | Article Outline

MATERIALS AND METHODS

Study Population

From all patients that received a kidney transplant at our institution between January 2005 and December 2013, we retrospectively selected all those treated with an anti–IL2-receptor-based induction therapy (basiliximab, Simulect; Roche; Basel, Switzerland) followed by a maintenance regimen consisting of a calcineurin-inhibitor, mycophenolate-mofetil and prednisolone (n = 211, Table 1). Patients treated without any induction therapy, depleting-antibody induction, ie, antithymocyte globulins, or an mTOR inhibitor–based maintenance regimen, were excluded as well as patients for whom no serum sample was available before transplantation. All patients were transplanted with a negative CDC-CM using current sera. During the study period, only those HLA against which antibodies were detected in recipient sera before transplantation using cell-based CDC assays were defined as UAM. Donor and recipient characteristics as well as clinical data were obtained by careful chart review or were extracted from the Eurotransplant Network Information System. All retrospective analyses were performed with approval of the local institutional review board.

TABLE 1

TABLE 1

Back to Top | Article Outline

Diagnosis of Rejection and Definition of AMR

All rejection episodes were biopsy-proven. Biopsies were obtained either as protocol biopsy on days 14 and 90 posttransplantation or when clinically indicated. Specimens were evaluated by 2 independent pathologists (K.U. and M.E.) on light microscopy and immunohistochemistry for C4d staining and were graded according to the BANFF 2013 classification.10 If AMR was diagnosed together with T cell–mediated rejection, it was classified as AMR.

Back to Top | Article Outline

Detection and Definition of DSA and Donor HLA Typing

Sera taken at the time of kidney transplantation were retrospectively screened for the presence of anti-HLA class I and class II IgG antibodies. All sera were stored at −80°C and heat inactivated at 52°C for 20 minutes before analysis. Screening was done using a commercial solid-phase microsphere-based assay (LSM12; One Lambda Inc., Los Angeles, CA). Sera were analyzed on a LABScan 100 Luminex (Luminex Corp., Austin, TX) flow analyzer, applying a threshold ratio for positive results of 2.5. In positive sera, HLA specificity was determined by a single-antigen assay for HLA class I and/or HLA class II antigens (LABScreen Single Antigen, Class I or II, respectively, both One Lambda Inc.). The tests were performed according to the manufacturers’ instructions and analyzed on a LABScan 100 Luminex flow analyzer, applying a baseline-adjusted MFI cutoff for positive reactions of 500. Donor-specificity of anti-HLA antibodies was defined based on the available donor HLA typing data. Donor HLA-typing was performed according to standard Eurotransplant (ET) protocols. Typing for HLA-A, -B, and -DR was done for all donors. HLA-DQ and Cw typing data were available for 196 (92.9%) and 49 (23.2%) donors, respectively. DP typing was not routinely done and therefore anti-DP HLA antibodies were not evaluated for donor-specificity. If donor-specificity of anti-HLA antibodies could not be determined due to lack of high-resolution typing of a donor, they were classified as non-DSA. This occurred in 3 recipients for HLA class I and in 19 patients for HLA class II antibodies, respectively. Lack of high-resolution typing in the corresponding donors resulted in potential misclassification of 3 recipients with respect to pretransplant DSA status (yes/no) with all 3 having an uneventful follow up. In case Luminex analysis revealed the presence of antibodies for all different splits of an HLA antigen, the bead with the highest MFI was used for MFI categorization.

Back to Top | Article Outline

Definitions of SAB-UAM

If Luminex analysis of pretransplant sera revealed the presence of anti-HLA IgG antibodies, the corresponding HLA were classified as unacceptable HLA-antigen mismatches (SAB-UAM) based on the following algorithms (Table 2): (I) DGI consensus: antibodies against class I and II HLA found in the LSM screening assay that had individual MFI values above 3000 in the SAB assay in recipients of a first kidney transplant or antibodies against class I or II HLA upon screening with MFI above 3000 in the SAB assay in patients awaiting retransplantation. Alternatively, antibodies against class I or II HLA upon screening with MFI above 5000 using SAB in recipients of a first kidney transplant.9 (II) ME algorithm: all antibodies against HLA-A, -B, or -DR with MFI above 5000 or antibodies against HLA DQ with MFI above 10 000.8 (III) MFI_10 000: all anti-HLA antibodies with MFI above 10 000.

TABLE 2

TABLE 2

Back to Top | Article Outline

Virtual Panel-Reactive Antibodies in Patients on the Waiting List

Sera from all 256 patients on the waiting list for kidney transplantation at our institution in October 2016 were screened for the presence of anti-HLA antibodies as detailed above. For calculation of virtual panel reactive antibodies (vPRA), all antibodies meeting the above SAB-UAM criteria were entered in the vPRA calculator at https://www.etrl.org/Virtual%20PRA/Default.aspx (last accessed on 12/03/2017). Anti-HLA C antibodies were included in all algorithms with an MFI threshold of 5000 in the ME algorithm (Table 2), the latter therefore designated “modified ME algorithm” (ME_mod). Anti-DP antibodies were not included, as they are not considered in the vPRA calculator until now.

Back to Top | Article Outline

Statistical Analysis

Statistical analysis was performed using IBM SPSS version 23.0 (SPSS Inc., Chicago, IL). Survival analysis was performed by the Kaplan-Meier method and differences between groups were compared using the log-rank test. Differences in kidney function (glomerular filtration rate [GFR]) between groups were evaluated using the Kruskal-Wallis test including pairwise post hoc testing. A P value less than 0.05 was considered statistically significant.

Back to Top | Article Outline

RESULTS

Patient and Treatment Characteristics

Among the 211 patients studied, 115 (54.5%) had anti-HLA-specific IgG antibodies retrospectively detected by the Luminex assay using day of transplant sera. Antibody-positive patients had a median of 15 HLA-specificities with 56/115 (48.7%) having antibodies against both class I and class II HLA. Median MFI of the highest individual anti-HLA antibody (MFImax) was 6218 (range, 525-24 432). All patients received nondepleting induction therapy with basiliximab. Initial maintenance immunosuppression consisted of tacrolimus, mycophenolate mofetil and prednisolone in 99% of patients. Median follow up was 5.0 years (Table 1).

Back to Top | Article Outline

Characteristics of SAB-UAM-positive Patients at the Time of Transplantation

At the time of transplantation, 67 (31.8%) of 211 patients had donor-specific anti-HLA-antibodies (DSA). In 31 (46.3%) of 67 patients, the DSA characteristics retrospectively met the SAB-UAM criteria proposed by the DGI, whereas 24 (35.8%) of 67 had SAB-UAM according to the ME algorithm. Defining UAM as all DSA with an MFI above 10 000, 17 (25.4%) patients were positive. All SAB-UAM algorithms consistently identified recipients considered to be at an increased immunological risk defined by conventional risk factors such as CDC-PRA or a history of previous kidney transplantation. The high immunological risk was also reflected by the DSA patterns of SAB-UAM–positive patients with 46.9% to 58.8 % having DSA against HLA class I and II and median MFImax between 11.011 and 15.766, respectively (Table 3).

TABLE 3

TABLE 3

Back to Top | Article Outline

Outcome of SAB-UAM-Positive Patients

AMR

Among the 67 DSA-positive patients, 16 (23.9%) experienced AMR during a median follow up of 4.9 years. Four (25%) of 16 were classified as C4d-negative. Thirteen (81.3%) of 16 AMR episodes occurred during the first year posttransplantation. Eleven (68.8%) of 16 AMR episodes occurred in patients that would have been identified as SAB-UAM-positive before transplantation using the DGI algorithm (Figure 1 A). All 11 patients met the “high-risk” criteria proposed by the DGI. None of the 3 patients classified as “intermediate risk” experienced AMR (Figure 1 B). The ME algorithm retrospectively identified 10 of 16 AMR-positive patients (Figure 1C). Both the DGI and the ME algorithm identified 9/13 patients with an AMR during the first year posttransplantation. Lastly, 8 of 16 patients experiencing AMR had at least 1 DSA with an MFI above 10 000, all occurring during the first year posttransplantation (Figure 1 D).

FIGURE 1

FIGURE 1

Back to Top | Article Outline

Graft Loss

After exclusion of graft losses secondary to nonimmunological reasons (n = 3, 1 primary nonfunction, 1 vascular problem and 1 BK-nephropathy), 9 graft losses were identified in DSA-positive patients. Seven of 9 patients were identified as SAB-UAM-positive by the DGI algorithm, all fulfilling the "high risk" criteria (Figure 2 A-B). 6/9 were identified by the ME algorithm (Figure 2 C). Five of 9 patients with early graft loss had a least 1 DSA with MFI above 10 000 (Figure 2 D).

FIGURE 2

FIGURE 2

Back to Top | Article Outline

Kidney Function

Patients with DSA classified as SAB-UAM-positive by the DGI and ME algorithms had a lower GFR compared to DSA-positive but SAB-UAM-negative patients at 1, 3, and 5 years posttransplantation (not shown). A lower GFR in SAB-UAM-positive compared with SAB-UAM-negative patients was still found at all time points studied when patients who had experienced AMR were excluded from both groups (Figures 3A-C). Tacrolimus trough levels at 1 and 3 years posttransplantation were not different between the groups (Table S2, SDC, http://links.lww.com/TP/B535). Moreover, there was no correlation between tacrolimus trough level and GFR (not shown). Proteinuria over time showed high variability but tended to be higher in SAB-UAM-positive compared to SAB-UAM-negative patients (not shown).

FIGURE 3

FIGURE 3

Back to Top | Article Outline

Accuracy of the SAB-UAM Algorithms in Identifying Patients at Risk

Depending on the SAB-UAM algorithm used, 71% (22/31, DGI consensus), 62.5% (15/24, ME algorithm), and 52.9% (9/17, MFI ≥ 10 000) of SAB-UAM-positive patients at the time of transplantation were free from AMR during the first year posttransplantation. This was reflected by a low PPV for the occurrence of early AMR for all algorithms tested. However, the negative predictive value (NPV) was excellent ranging between 88.9% and 90.7%. No graft loss during follow-up was observed in 77.4% (24/31), 75% (18/24), and 70.6% (11/17) of SAB-UAM–positive patients, respectively, again resulting in low PPV (Table 4 and Figure 4).

TABLE 4

TABLE 4

FIGURE 4

FIGURE 4

Back to Top | Article Outline

SAB-UAM in Patients on the Waiting List for Kidney Transplantation

We next assessed the prevalence of SAB-UAM in an independent cohort of 256 patients on the waiting list for kidney transplantation at our institution at the end of 2016 (Table S1, SDC,http://links.lww.com/TP/B535). One hundred thirty-four (52.3%) of 256 patients had anti-HLA antibodies in the Luminex screening assay that was then complemented by an SAB test. Eighty-five (33.2% of the total population) of 256 patients had antibodies that fulfilled the DGI criteria with 56 classifying as “high risk” and 29 as “intermediate risk.” Seventy-nine (30.9%) of 256 patients had SAB-UAM based on the ME_mod algorithm (including HLA C), whereas 59 (23%) of 256 had at least 1 DSA with MFI above 10 000. If only HLA antibodies from cell-based CDC assays and those Luminex-derived antibodies against HLA from previously failed grafts were considered (the algorithm previously in place at our institution), 51 (19.9%) of 256 patients had UAM (Table 5).

TABLE 5

TABLE 5

Back to Top | Article Outline

Potential Impact on Waiting Time

To calculate the extent of the ET donor pool excluded during organ allocation due to the presence of SAB-UAM, the vPRA was calculated.11 Depending on the algorithm applied, median vPRA was between 69.2% and 79.1% (Table 5).

It has recently been reported that each 1% increase in vPRA would prolong a standard ET kidney allocation system (ETKAS) patient's waiting time by 1.3 weeks. Patients older than 65 years listed in the ET Senior Program (ESP), however, were reported to wait an additional 5 weeks for every 1% increase in vPRA.12 The majority of SAB-UAM-positive patients on our waiting list were listed for kidney only, did not participate in the ET acceptable mismatch program and were younger than 65 years, therefore, representing standard ETKAS patients (Table 5). Depending on the algorithm, these patients had a vPRA between 60% and 71.5%, translating into a potential prolongation of waiting time between 1.5 and 1.8 years, respectively. Those patients qualifying for the ESP program had a vPRA between 56.2% and 67.7%, translating in a potential longer time on dialysis between 5.4 and 6.5 years, respectively (Table 5).

Back to Top | Article Outline

DISCUSSION

In this study, we retrospectively compared the ability of 3 Luminex-based UAM algorithms to predict early AMR and graft loss in a cohort of CDC-CM–negative renal transplant patients. We made 3 major observations: First, the sensitivity of the algorithms to identify patients at risk decreased with increasing the MFI threshold, but absolute differences between the 3 algorithms were small. Second, most patients retrospectively identified to have SAB-UAM did not experience AMR or graft loss. Compared with the total number of patients transplanted, however, the absolute number of patients, that would have been wrongly excluded, was small. Third, even in the absence of overt AMR, kidney function during follow-up was significantly lower in SAB-UAM-positive compared with SAB-UAM-negative patients. In addition, Luminex-based assignment of UAM to an independent cohort of waiting list patients identified a relevant proportion of patients as SAB-UAM–positive with high vPRA levels.

Although all patients in our study had a negative CDC-CM at the time of transplantation, additional flow cytometry crossmatch (FCM) testing as a means to further improve risk stratification was not performed.8,13 Used in combination by many transplant centers, FCM is an objective method to add plausibility to SAB results.6 However, as with Luminex testing, FCM has technical limitations, and there is no established cutoff to segregate patients at risk from those without.6,14 Further studies are needed to clarify whether a combined approach is superior to using SAB testing alone for the identification of at-risk patients in the setting of a negative CDC-CM.

Given the absence of SAB test results at the time of transplantation, all patients in our study received nondepleting induction therapy and no desensitization procedures were performed before transplantation. It is therefore possible that an intensified immunosuppressive regimen would have resulted in comparable outcomes even with a less stringent SAB-UAM definition. However, recent data indicate that highly immunized patients still experience a high rate of AMR despite intensive desensitization regimens.15-17 Moreover, antibody-induction protocols might not significantly reduce the risk for early AMR, especially in DSA-positive patients with high MFI.18 Of note, 5-year allograft survival in DSA-positive but SAB-UAM–negative patients was above 90% for all algorithms tested in our study and comparable to DSA-negative patients treated with the same immunosuppressive regimen at our center.4 These results simultaneously justify the treatment approach in SAB-UAM–negative patients and validate the UAM algorithms tested.

It has been recommended to define DSA that are positive in solid phase but negative in cell-based assays only as UAM if they can be explained by a patient's immunization history.6,9 Moreover, DSA against HLA from previously rejected grafts are commonly defined as UAM irrespective of their method of detection.19 In our study, 16 of 33 previously transplanted patients had DSA against repeat HLA mismatches. Of note, only 5 of 16 experienced AMR, and only 2 lost their graft during follow up. Patients with AMR had similar MFImax compared with those without AMR (not shown). Therefore, integration of additional DSA properties such as C1q- or C3d-binding into SAB-UAM algorithms might help improve their predictive power.

Given the lower MFI threshold of the first algorithm, it is not surprising that it had the highest sensitivity in identifying patients at risk. However, the numerical differences between the 3 approaches, most notably between the DGI and the ME algorithm, were small. A larger study population will have to be analyzed to see whether the differences between the 3 algorithms become more pronounced. As the second algorithm had a slightly higher accuracy (less patients wrongly excluded) compared to the first, one might favor the easy-to-use ME algorithm over the more complex DGI consensus. Of note, DSA against HLA C and DP are not part of the ME algorithm, potentially questioning the clinical relevance of DSA against these epitopes. However, evidence is accumulating that DSA against these epitopes are equally immunogenic even if they occur in isolation.20-22 The lack of systematic HLA C and DP donor typing in our cohort does not allow any conclusion on this question.

One of the major aspects of our study was to assess how many patients with an uneventful follow up were identified as “high risk” by any of the 3 algorithms and therefore would not have been transplanted in case SAB-UAM algorithms had been in place at the time of organ allocation. Extrapolation of the rate of 62.5% to 71% of patients without early AMR to the roughly 30% SAB-UAM-positive patients identified by the DGI and the ME algorithm on the waiting list suggests that about 20% of patients on the waiting list would have to wait longer for no reason. In the ET region, it remains speculative to what degree high vPRA levels affect waiting time.12,23 This is primarily because the restriction of the donor pool as a consequence of increased vPRA levels is not integrated in any of the ET allocation algorithms until today. The acceptable mismatch program has been set up to enable transplantation for the 2% to 3% highest immunized patients requiring greater than 85% mainly cytotoxic PRA as an entry criterion. In ETKAS, only the most recent PRA from quarterly screening is part of the so-called mismatch probability score. As this has historically been a CDC-PRA, the true restriction of the donor pool as determined by Luminex-based vPRA is not reflected in the mismatch probability.12,24 Finally, in the ESP, the degree of immunization is not considered at all. Given the regional donor pool with poor HLA variability in ESP, high vPRA levels might result in an inacceptable increase of waiting time and patient death on dialysis.

In summary, we have found that all tested SAB-UAM algorithms allow for the identification of most patients with poor outcomes due to early AMR and graft loss. However, they affect many patients on the waiting list with, until now, unknown effects on waiting time. In face of these data, it seems imperative to homogenize UAM algorithms between transplant centers and to integrate the vPRA into the ET organ allocation algorithms to compensate for the restricted donor pool of patients with a large number of SAB-UAM. Until then, the decision which algorithm to use will remain a judgement call for HLA laboratories and transplant physicians and has to be balanced between an unknown prolongation of waiting time for a considerable number of potentially low risk patients and maximization of success rates for those patients transplanted.

Back to Top | Article Outline

REFERENCES

1. Patel R, Terasaki PI. Significance of the positive crossmatch test in kidney transplantation. N Engl J Med. 1969;280:735–739.
2. Amico P, Hirt-Minkowski P, Honger G, et al. Risk stratification by the virtual crossmatch: a prospective study in 233 renal transplantations. Transpl Int. 2011;24:560–569.
3. 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.
4. Zecher D, Bach C, Staudner C, et al. Characteristics of donor-specific anti-HLA antibodies and outcome in renal transplant patients treated with a standardized induction regimen. Nephrol Dial Transplant. 2017;32:730–737.
5. Konvalinka A, Tinckam K. Utility of HLA antibody testing in kidney transplantation. J Am Soc Nephrol. 2015;26:1489–1502.
6. Tait BD, Süsal C, Gebel HM, 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.
7. Süsal C, Roelen DL, Fischer G, et al. Algorithms for the determination of unacceptable HLA antigen mismatches in kidney transplant recipients. Tissue Antigens. 2013;82:83–92.
8. Marfo K, Ajaimy M, Colovai A, et al. Pretransplant immunologic risk assessment of kidney transplant recipients with donor-specific anti-human leukocyte antigen antibodies. Transplantation. 2014;98:1082–1088.
9. Susal C, Seidl C, Schonemann C, et al. Determination of unacceptable HLA antigen mismatches in kidney transplant recipients: recommendations of the German Society for Immunogenetics. Tissue Antigens. 2015;86:317–323.
10. Haas M, Sis B, Racusen LC, et al. Banff 2013 meeting report: inclusion of c4d-negative antibody-mediated rejection and antibody-associated arterial lesions. Am J Transplant. 2014;14:272–283.
11. Eurotransplant Reference Laboratory. Eurotransplant. https://www.etrl.org/VirtualPRA/Default.aspx. vPRA Calculator. Accessed December 3, 2017.
12. Ziemann M, Hessler N, Konig IR, et al. Unacceptable human leucocyte antigens for organ offers in the era of organ shortage: influence on waiting time before kidney transplantation. Nephrol Dial Transplant. 2017;32:880–889.
13. Schinstock CA, Gandhi MJ, Stegall MD. Interpreting anti-HLA antibody testing data: a practical guide for physicians. Transplantation. 2016;100:1619–1628.
14. Burns JM, Cornell LD, Perry DK, et al. Alloantibody levels and acute humoral rejection early after positive crossmatch kidney transplantation. Am J Transplant. 2008;8:2684–2694.
15. Schwaiger E, Eskandary F, Kozakowski N, et al. Deceased donor kidney transplantation across donor-specific antibody barriers: predictors of antibody-mediated rejection. Nephrol Dial Transplant. 2016;31:1342–1351.
16. Marfo K, Lu A, Ling M, et al. Desensitization protocols and their outcome. Clin J Am Soc Nephrol. 2011;6:922–936.
17. Marfo K, Ling M, Bao Y, et al. Lack of effect in desensitization with intravenous immunoglobulin and rituximab in highly sensitized patients. Transplantation. 2012;94:345–351.
18. Kannabhiran D, Lee J, Schwartz JE, et al. Characteristics of circulating donor human leukocyte antigen-specific immunoglobulin G antibodies predictive of acute antibody-mediated rejection and kidney allograft failure. Transplantation. 2015;99:1156–1164.
19. Abramowicz D, Cochat P, Claas FH, et al. European Renal Best Practice Guideline on kidney donor and recipient evaluation and perioperative care. Nephrol Dial Transplant. 2015;30:1790–1797.
20. Aubert O, Bories MC, Suberbielle C, et al. Risk of antibody-mediated rejection in kidney transplant recipients with anti-HLA-C donor-specific antibodies. Am J Transplant. 2014;14:1439–1445.
21. Bettinotti MP, Zachary AA, Leffell MS. Clinically relevant interpretation of solid phase assays for HLA antibody. Curr Opin Organ Transplant. 2016;21:453–458.
22. Jolly EC, Key T, Rasheed H, et al. Preformed donor HLA-DP-specific antibodies mediate acute and chronic antibody-mediated rejection following renal transplantation. Am J Transplant. 2012;12:2845–2848.
23. Bostock IC, Alberu J, Arvizu A, et al. Probability of deceased donor kidney transplantation based on % PRA. Transpl Immunol. 2013;28:154–158.
24. Wissing KM, Abramowicz D. Unacceptable human leucocyte antigens: how to navigate between increased immunological risk and waiting time? Nephrol Dial Transplant. 2017;32:745–747.

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.