SEARCH

SEARCH BY CITATION

Keywords:

  • Antibody-mediated rejection;
  • flow-beads;
  • HLA-antibodies;
  • renal allograft rejection virtual crossmatch

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Preformed donor-specific HLA-antibodies antibodies (DSA) are a major risk for early antibody-mediated rejection (AMR). This prospective study evaluated the accuracy of pretransplant risk assessment using virtual crossmatching (virtualXM) (i.e. comparing HLA-typing of the donor with the recipient's HLA-antibody specificities determined by flow-beads). Sixty-five consecutive patients were stratified according to virtualXM results: patients without DSA (n= 56) were considered low risk and received standard immunosuppression; patients with DSA (n= 9) were considered high risk and received additional induction with anti-T-lymphocyte-globulin (ATG) and intravenous immunoglobulins. Despite induction therapy 4 of 9 patients with DSA (44%) had clinical/subclinical AMR, whereas only 2 of 56 patients without DSA (4%) (p = 0.002). Notably, one of these two patients had early AMR likely induced by non-HLA-antibodies; the other had subclinical AMR at month 6 consistent with de novo DSA. The results of virtualXM and retrospectively obtained flow-cytometric crossmatches (FCXM) (n= 59) were concordant in 51 patients (86%), four patients (7%) were virtualXM−/FCXM+ and none had AMR, four patients (7%) were virtualXM+/FCXM− and one had AMR. VirtualXM can accurately define absence or presence of DSA and may become an invaluable tool for organ allocation and pretransplant risk assessment. However, further studies need to address whether all HLA-antibodies detected by flow-beads are clinically relevant.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The presence of preformed donor-specific HLA-antibodies (DSA) is a major risk factor for early renal allograft rejection and allograft loss (1). For the last 35 years, the standard complement-dependent cytotoxicity crossmatch (CDCXM) and the antihuman globulin augmented assay (AHG-CDCXM) were used the detect DSA (2, 3). More recently, the flow cytometric crossmatch (FCXM) proved to be more sensitive than the CDC-assays to detect clinically relevant DSA and is therefore widely accepted today as the best possible test to assess the presence of DSA (1). However, the FCXM has some limitations. First, nonspecific binding of irrelevant antibodies to the cell surface may lead to false positive results (4). Second, the assay is not standardized and the definition of a cut-off for a positive FCXM is difficult, which may cause both false positive and false negative results. Furthermore, for deceased donor transplantation, the FCXM is often performed in a few patients as a ‘final XM’ after screening eligible patients on the waiting list by (AHG) CDCXM. This procedure requires additional resources in the laboratory and prolongs cold ischemia time.

Flow cytometry-based solid-phase assays (flow-beads) allow to detect and specify HLA-antibodies with at least similar sensitivity than the FCXM (5). Therefore, it is possible to determine the presence of DSA without performing an FCXM by comparing the HLA-typing of the donor with the HLA-antibody specificities of the recipient (i.e. virtualXM). Up to date, this strategy has been evaluated only in lung transplantation. In this study, organs were allocated to 16 sensitized recipients without DSA defined by flow-beads and all had a negative retrospective FCXM (6).

While the best solution is to perform transplantation only when no DSA are present, this is usually not achievable for highly sensitized patients without access to a large donor pool. For such patients, transplantation across DSA with an intensified immunosuppressive regimen can be a reasonable solution (7, 8). A recent study demonstrated that DSA levels at the time of transplantation are of critical importance and should be below the threshold of detection of an AHG-CDCXM for better outcomes (i.e. low-level DSA) (7). Therefore, a strategy using virtualXM for risk stratification together with a CDC-based XM to determine the level of potentially present DSA at the time of transplantation may be feasible. The aim of this prospective single center study was to evaluate this strategy in unselected consecutive renal transplant recipients.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Patients

All therapeutic protocols were approved by the local institutional review board and all patients gave informed consent. Sixty-eight consecutive patients who received a kidney transplant at the University Hospital Basel from November 2004 to February 2006 were evaluated. Three patients were excluded (AB0-incompatible transplantation (n= 1), transplant from an identical twin (n= 1), allograft loss due to renal vein thrombosis at day 6 posttransplant (n= 1)) leaving 65 for analysis. Patients were divided into three groups based on the results of the virtualXM: (i) patients with no HLA-antibodies (no HLA-Ab; n= 46), (ii) patients with HLA-antibodies, but no DSA (HLA-Ab/no DSA; n= 10), (iii) patients with DSA (DSA; n= 9). All patients had negative current and remote T- and B-cell CDCXM.

Patients with DSA (n= 9) were considered as high risk for rejection and received an induction therapy consisting of a polyclonal anti-T-lymphocyte globulin (ATG-Fresenius, Fresenius Medical Care, Switzerland) 9 mg/kg body weight prior to reperfusion of the allograft and 3 mg/kg body weight on day 1–4 as well as intravenous immunoglobulins (IvIg) (Octagam, Octapharm, Hurdal, Norway) 0.4 g/kg body weight per day on day 0–4. Maintenance immunosuppression consisted of tacrolimus (Tac) (Prograf, Astellas), mycophenolate-mofetil (MMF) (CellCept, Roche, Basel, Switzerland) and steroids. Target tacrolimus trough levels were 12–15 ng/mL for the first month, 10–12 ng/mL for months 2–3, 8–10 ng/mL for months 4–6 and 6–8 ng/mL thereafter. Steroids were tapered to 0.1 mg/kg body weight by month 3 posttransplant.

Patients in the no HLA-Ab (n= 46) and HLA-Ab/no DSA-group (n= 10) were considered as low risk for rejection. They received 20 mg basiliximab (Simulect, Novartis, Basel, Switzerland) on day 0 and 4 and maintenance triple immunosuppression consisting of either Tac, MMF and steroids or Tac, mycophenolate-sodium (Myf) (Myfortic, Novartis) and sirolimus (Sir) (Rapamune, Wyeth, Grafenauweg, Switzerland) according to ongoing protocols. In the Tac-MMF-steroids regimen target tacrolimus trough levels were 10–12 ng/mL for the first month, 8–10 ng/mL for months 2–3, 6–8 ng/mL for months 4–6 and 4–6 ng/mL thereafter. Steroids were tapered to 0.1 mg/kg body weight by month 3 posttransplant and further tapered to 0 within six weeks when the protocol biopsy at month 3 posttransplant showed no rejection. In the Tac-Myf-Sir regimen, target tacrolimus trough levels were the same as in the Tac-MMF-steroids regimen; target sirolimus trough levels for month 1–3 were 4–8 ng/mL. If the protocol biopsy at month 3 posttransplant showed no rejection, tacrolimus trough levels were further decreased to 4–8 ng/mL.

Diagnosis and treatment of rejection

All rejection episodes were biopsy-proven. Protocol biopsies were performed at month 3 and month 6 posttransplant. Biopsy specimens (two cores obtained with a 16-gauge needle) were evaluated by light microscopy, immunofluorescence (C4d, HLA-DR and immunoglobulins), and immunohistochemistry (SV40 antigen). Findings were graded according to the updated Banff 97 classification (9). Positive C4d-staining was defined as either focal or diffuse detection of C4d in peritubular capillaries (PTC) by immunofluorescence. Antibody-mediated rejection (AMR) was defined as diffuse C4d positivity in PTC with or without transplant-glomerulitis or severe capillaritis in PTC. AMR was also assumed in the presence of transplant-glomerulitis without C4d positivity in PTC. Clinical and subclinical rejection (Banff borderline tubulitis, IA and IB) were treated with steroid pulses (3*500 mg iv.). Subclinical AMR was treated with IvIg (5*0.4 g/kg body weight) and steroid pulses (3*500 mg iv.). In case of clinical AMR plasmapheresis±rituximab (Mabthera, Roche) was added.

Detection and specification of HLA-antibodies with flow-beads

FlowPRA screening beads class I and II and FlowPRA single antigen beads class I and II (OneLambda, Canoga Park, CA) were used according to the instructions of the manufacturer. Staining was performed with fluorescein isothiocyanate (FITC)-conjugated goat-anti human IgG and beads were analyzed on a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA). FlowPRA screening beads class I and II were considered positive when >3% of beads exhibited fluorescence above the negative control. FlowPRA single antigen beads were considered positive when a bead shifted to the right of the negative control bead and the corresponding bead in the negative control serum.

Screening algorithm for HLA-antibodies

Patients who had presensitizing events were evaluated for the presence of HLA-antibodies with FlowPRA screening beads I/II in the current serum and historic sera. HLA-antibodies found only in historic sera were regarded as risk for transplantation and were included for virtualXM analysis. Patients without presensitizing events were evaluated for the presence of HLA-antibodies with FlowPRA screening beads I/II in the current serum only. In case of a positive result as defined above, HLA-antibody specificities were determined with FlowPRA single antigen beads I/II. After subsequent sensitizing events, analysis with FlowPRA screening beads I/II was repeated.

Complement-dependent cytotoxicity crossmatch (CDCXM) assay

Immunomagnetic beads (Dynabeads, Dynal Biotech, Oslo, Norway) were used to separate T- and B-cells according to the instruction of the manufacturer. One μL of donor T- and B-cells were incubated with 1 μL of recipient sera for 30 and 40 min, respectively. A 5 μL rabbit complement and staining solution was added and incubated for 45 min. T- and B-cell CDCXM were considered positive when the observed cell death exceeded 10% above background.

Flow cytometric crossmatch (FCXM) assay

All FCXM were done retrospectively with thawed PBMC's or spleen cells. Cells were washed and treated with pronase (Sigma-Aldrich, St. Louis, MO) as previously reported (10). A 20 μL of 1:4 diluted recipient serum and four control sera (three negative controls and one positive control) were incubated with 0.25*106 donor PMBC's for 30 min on ice. After three washes with PBS containing 4% goat serum, cells were incubated for 20 min on ice with FITC-conjugated goat-anti human IgG (Jackson Immunoresearch, West Grove, PA), phycoerythrin (PE)-conjugated anti-CD19 (BD Biosciences) and allophycocyanin (APC)-conjugated anti-CD3 (BD Biosciences). After three washes, cells were fixed in 1% paraformaldehyde. Data were collected on an FACSCalibur (BD Biosciences). Molecules of equivalent soluble fluorochrome (MESF)-FITC reference beads (Bangs Laboratories, Fischers, IN) were measured with the same instrument settings. The T- and B-cell FCXM was considered positive when MESF-value of the patient serum exceeded the mean of the three negative control sera by 20% and 36%, respectively. These cut-offs were defined by evaluation of 140 negative sera-donor cell combinations (PBMC's or spleen cells) as follows: percent MESF-value increase = 100 / mean of MESF-values * 3 standard deviations of delta MESF-shift. In case of a positive allo-FCXM, an auto-FCXM was performed to determine the presence of auto-antibodies. A positive allo-FCXM together with a negative auto-FCXM was considered to indicate the presence of DSA.

Investigated outcomes

Primary outcomes were biopsy-proven rejection and graft loss due to rejection within the first three weeks posttransplant. Secondary outcomes were biopsy-proven rejection episodes beyond the first three weeks posttransplant and allograft function at month 6.

Statistical analysis

We used JMP software version 6.0 (SAS Institute Inc., Cary, NC) for statistical analysis. For categorical data, Fisher's exact test or Pearson's chi-square test was used. Parametric continuous data were analyzed by Student t-tests or one-way analysis of variance. For nonparametric continuous data, Wilcoxon or Kruskal-Wallis rank sum tests were used. A p-value <0.05 (two-sided test) was considered to indicate statistical significance.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Patient characteristics

Patient characteristics are summarized in Table 1. Presensitizing events were noted in 20 of 46 patients (43%) in the no HLA-Ab group, in 8 of 10 patients (80%) in the HLA-Ab/no DSA group and in all patients in the DSA group (p = 0.002). This difference was due to more patients with prior transplants and pregnancies (p = 0.0004 and p = 0.0001, respectively). CDC-PRA were significantly different between the no HLA-Ab group, the HLA-Ab/no DSA group and the DSA group in the peak serum (p < 0.0001), but not in the current one (p = 0.08). In contrast, Flow-PRA class I and II were different in the peak and the current serum (p < 0.0001).

Table 1.  Baseline characteristics
VariableNo HLA-antibodies (n= 46)HLA-antibodies, but no DSA (n= 10)DSA (n= 9)P-level
  1. *A patient can contribute to more than one group.

Recipient age, median (range)53 (18–70)51 (36–67)48 (34–68)0.92
Recipient sex, no. females (%)13 (28)7 (70)6 (67)0.01
Donor, no. living donor (%)27 (59)5 (50)2 (22)0.13
Donor age, median (range)52 (1–83)57 (37–73)47 (7–67)0.48
Patients with presensitizing events, no. (%)20 (43)8 (80)9 (100)0.002
Presensitizing events*
 - Patients with prior transplants, no. (%)2 (4)3 (30)5 (56)0.0004
 - Patients with blood transfusions, no. (%)15 (33)7 (70)6 (75)0.02
 - Patients with pregnancies, no. (%)8 (17)6 (60)6 (67)0.0001
CDC-PRA [%]
 - Peak serum, median (range)00 (0–52)52 (0–78)<0.0001
 - Current serum, median (range)00 (0–14)0 (0–44)0.08
Flow-PRA class I [%]
 - Peak serum, median (range)010 (0–69)71 (6–97)<0.0001
 - Current serum, median (range)09 (0–69)60 (0–97)<0.0001
Flow-PRA class II [%]
 - Peak serum, median (range)00 (0–22)57 (0–98)<0.0001
 - Current serum, median (range)00 (0–20)57 (0–99)<0.0001
HLA-mismatches
 - A/B, median (range)3 (0–4)2 (1–4)2 (1–2)0.04
 - DR, median (range)1 (0–2)2 (1–2)1 (1–2)0.03

Clinical rejection and graft loss rate within the first 3 weeks posttransplant

Clinical rejection within the first 3 weeks posttransplant was observed in 3 of 65 patients (4.6%). All were AMR, which occurred on posttransplant day 1 (patient in the no HLA-Ab group), 11 and 21 (both patients in the DSA group). The patient in the no HLA-Ab group lost the graft due to severe AMR on the third day posttransplant. In this particular patient, we could not detect any HLA-antibodies in all pre- or posttransplant sera by FlowPRA screening and single antigen beads, and both CDCXM and FCXM were negative. Furthermore, antiphospholipid antibodies were tested negative.

Clinical rejection rate beyond the first 3 weeks after transplantation and development of allograft function

One allograft was lost beyond the first 3 weeks after transplantation in the HLA-Ab/no DSA group due to death of the patient with a well-functioning allograft (cardiac arrest at home). Six months' graft and patient survival were 97% and 98.5%, respectively. Two of 54 patients (4%) without DSA defined by virtualXM (i.e. 45 in the no HLA-Ab group and nine in the HLA-Ab/no DSA) had a biopsy-proven acute clinical rejection, which were both Banff grade IA/IB. No clinical rejection episode occurred in the DSA group. Glomerular filtration rate (GFR) was similar in all three groups at month 6 posttransplant (p = 0.12) (Table 2).

Table 2.  Allograft function and protocol biopsy results
 No HLA-antibodies (n= 45)*HLA-antibodies, but no DSA (n= 9)*DSA (n= 9)P-level
  1. * One patient in each group lost their allograft within the first month. These two patients were not included in the analysis.

  2. $ GFR calculated with the MDRD study equation: 186*serum creatinine-1.154*age-0.203*0.742 (if subject is female).

  3. AMR = antibody-mediated rejection; PVAN = polyoma BK-virus associated nephropathy.

Allograft function
 - Creatinine at month 1, median (range)1.6 (0.7–5.9)1.4 (0.8–4.0)1.4 (1.0–2.3)0.21
 - Creatinine at month 3, median (range)1.7 (0.8–4.9)1.3 (0.7–1.9)1.4 (0.9–2.1)0.02
 - Creatinine at month 6, median (range)1.5 (0.8–3.5)1.0 (0.7–1.9)1.1 (0.9–2.1)0.04
 - GFR$ at month 6, median (range)50 (19–100)61 (42–90)55 (28–81)0.12
Protocol biopsy at month 3 
 - No rejection, no.38950.08
 - Borderline tubulitis / Tubulitis IA, no.2 / 10 / 01 / 0 
 - AMR, no.002 
 - PVAN, no201 
 - Not performed, no.200 
Protocol biopsy at month 6
 - No rejection, no.34740.08
 - Borderline tubulitis / Tubulitis IA, no.6 / 10 / 01 / 0 
 - AMR, no.012 
 - PVAN, no.202 
 - Not performed, no.210 

Histological findings in protocol biopsies at 3 and 6 months posttransplant

Fifty-two of 54 patients (96%) without DSA (i.e. 45 in the no HLA-Ab group and nine in the HLA-Ab/no DSA) had a 3-month protocol biopsy. No AMR was observed and C4d-staining was negative in 51 of 52 biopsies (98%). One patient showed focal positive C4d-staining, which was subsequently negative in the 6-month protocol biopsy without any treatment. Fifty-one of 54 patients (94%) had a 6-month protocol biopsy. C4d-staining was negative in all 51 biopsy samples. Transplant-glomerulitis was observed in 1 of 51 patients (2%). This patient had stable allograft function (creatinine 0.8–1.1 mg/dL) and a normal three-month protocol biopsy. Before transplantation (FlowPRA class I, 69%; class II, 14%) and at the time of biopsy showing transplant-glomerulitis (FlowPRA class I, 77%; class II, 4%) the same 16 HLA-antibody specificities were detectable by single antigen beads without any DSA. Polyoma BK-virus associated nephropathy (PVAN) was diagnosed in 2 of 52 biopsies (4%) at month 3 and 2 of 51 biopsies (4%) at month 6 (Table 2).

All nine patients with DSA had a 3- and 6-month protocol biopsy. AMR was observed in 2 of 9 biopsies at month 3 and 2 of 9 biopsies at month 6. PVAN was diagnosed in 1 of 9 biopsies at three months and 2 of 9 biopsies at month 6 (Table 2). Details of patients with DSA are summarized in Table 3.

Table 3.  Details of patients with donor-specific antibodies
Pat No.DonorPresensitizing eventsFlow-PRA I/II [%]*HLA-antibody specificities§FCXM T/BClinical rejectionProtocol BX month 3Protocol BX month 6
  1. *Flow-PRA values in current sera. Remote values are denoted in brackets.

  2. §DSA are bold and underlined. Strength of DSA is indicated semiquantitatively based on the extent of the shift of the beads: += 2/8 shift; ++= 4/8 shift; +++= 6/8 shift; ++++= 8/8 shift. Remote HLA-antibodies are denoted in brackets.

  3. $This patient received a haplo-identical first transplant from his mother with an HLA-A31 mismatch. The second transplant was from his haplo-identical sister, again with an HLA-A31 mismatch.

  4. LD = living donor; DD = deceased donor; LM = light microscopy; IF = immunofluorescence; AMR = antibody-mediated rejection; PVAN = polyoma BK-virus associated nephropathy; sCr = serum creatinine [mg/dL].

#1LDPrior transplants: 16/0A31+++ (repeated mismatch)$−/−NoLM: NormalLM: Normal
Blood transfusions: 2B – IF: C4d –IF: C4d –
DR – sCr: 1.2sCr: 1.1
DQ – 
 
#2LDBlood transfusions: >1060/57A1, 2, 25, 26, 29, 30, 31, 32+, 33, 34, 68−/−NoLM: AMR, borderlineLM: Borderline tubulitis
B7, 8, 18, 35, 38, 44, 45, 51, 55, 56++, 57, 58, 60, 64, 65 IF: C4d –IF: C4d –
DR7, 10 sCr: 2.1sCr: 2.1
DQ5, 6, 9 
 
#3DDPrior transplants: 168/73A30, 31–/+NoLM: NormalLM: PVAN
Blood transfusions: unknownB7++, 45 IF: C4d –IF: C4d –
DR1, 4, 7, 8, 9, 12, 15++, 16, 17, 103 sCr: 1.7sCr: 2.0
DQ4, 7++++, 8, 9 
 
#4DDPregnancies: 271/20A2, 3++, 11, 23, 24, 25, 26, 29, 32, 68+/+NoLM: NormalLM: Normal
B7++, 13, 18, 27, 35, 44, 45, 51, 52, 55, 57, 62, 65 IF: C4d –IF: C4d +
 sCr: 1.1
DR4 sCr: 1.4 
DQ7, 8, 9 
 
#5DDPrior transplants: 197/88A1, 2, 11, 32n.d.Day 11LM: AMRLM: AMR
Pregnancies: 2B7, 8, 27, 57, 60(no cells)LM: AMRIF: C4d –IF: C4d +
DR1, 4+++, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 IF: C4d +sCr: 1.9sCr: 1.9
DQ7, 8++++, 9 sCr: 2.7 
 
#6DDPrior transplants: 181/69A1, 3, 23, 24, 25, 29+++, 32, 33−/−NoLM: NormalLM: Normal
Blood transfusions: 1B13, 27, 38, 44, 49, 51, 52, 57 IF: C4d –IF: C4d –
Pregnancies: 2DR4, 7, 9, 10 sCr: 0.9sCr: 0.9
DQ2, 7, 8, 9 
 
#7DDPrior transplants: 269/98A11, 23, 24, 25, 26, 32, 34, 68–/+NoLM: Borderline tubulitisLM: Normal
Blood transfusions: >10B13, 27, 38, 44, 49, 51, 52, 57 IF: C4d –IF: C4d –
Pregnancies: 1DR1, 4, 8+++, 9, 13, 14, 15, 16, 17, 18, 103 sCr: 1.0sCr: 1.1
DQ4++, 5, 6, 8, 9 
 
#8DDBlood transfusions: 212/0A30, 31, 32, (2++, 11, 25, 68)current −/−Day 21LM: PVANLM: PVAN
Pregnancies: 2(93/0)B8, 18, 35, 51, 52, 57, 62, 65, (13, 38+++, 44, 45)remote +/+LM: AMRIF: C4d –IF: Cd4 –
DR – IF: C4d +sCr: 1.7sCr: 1.6
DQ – sCr: 3.5 
 
#9DDBlood transfusions: 20/0A (2++++, 68)current −/−NoLM: NormalLM: Normal
Pregnancies: 3(79/0)B (7, 8, 18, 27, 35, 38, 45, 49, 55, 60, 62, 65)remote −/− IF: C4d –IF: C4d –
DR – sCr: 1.0sCr: 0.9
DQ – 

Taken together, despite induction therapy with ATG/IvIg, 4 of 9 patients (44%) with DSA defined by virtualXM (Table 3; #2, 4, 5 and 8) had biopsy-proven AMR, whereas only 2 of 56 patients (4%) without DSA defined by virtualXM (p = 0.002).

Comparison of FCXM and virtualXM to predict AMR

We had sufficient donor cells to perform retrospective FCXM in 59 of 65 patients (91%). The results of virtualXM and FCXM were concordant in 51 of 59 patients (86%). Four patients (7%) were virtualXM−/FCXM+ and all had good allograft function and normal 3- and 6-month protocol allograft biopsies including negative C4d-staining. Four patients (7%) were virtualXM+/FCXM−. One of these four patients had subclinical AMR, the other three had good allograft function with normal 3- and 6-month protocol allograft biopsies including negative C4d-staining. Notably, these last four patients received an induction therapy with ATG/IvIg.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

To our knowledge this is the first prospective study that used virtualXM for pretransplant risk assessment in unselected consecutive renal allograft recipients and correlated these results with clinical and histopathological outcomes as the ‘gold standard’. Despite induction therapy with ATG/IvIg 4 of 9 patients (44%) with DSA defined by virtualXM demonstrated clinical/subclinical AMR, whereas only 2 of 56 patients (4%) without DSA. Notably, one of these two patients had early AMR most likely induced by preformed non-HLA-antibodies, the other had subclinical AMR at month 6, consistent with de novo DSA. This suggests that (i) the absence of DSA defined by virtualXM is associated with a very low risk for early AMR due to HLA-antibodies and (ii) that DSA defined by virtualXM are clinically relevant.

The concept of virtualXM relies on an accurate HLA-typing and a thorough evaluation of HLA-antibodies by flow-beads. It bears the inherent risk that only antibodies against HLA-antigens present on the beads can be detected, while non-HLA-antibodies and HLA-antibodies against those HLA-antigen alleles not represented on the flow-beads will be missed (11–15). Although FlowPRA screening and single antigen beads already cover a large proportion of the most prevalent alleles of HLA-antigens, there are still gaps in the whole repertoire of existing HLA-antigens. These gaps are likely to be filled by the manufacturers in the near future. However, preformed non-HLA-antibodies will remain an unpredictable risk for early AMR as long as no easy-to-use and reliable assays exist to measure them. As illustrated by the only allograft loss in our study cohort, non-HLA-antibodies were not only missed by flow-beads but also by all cell-based assays (i.e. CDC-PRA, CDCXM and FCXM).

While a negative virtualXM proved to be very reliable to rule out the presence of donor-specific HLA-antibodies, it becomes more a concern whether all HLA-antibodies detected by flow-beads are in fact clinically relevant. The number of detected HLA-antibody specificities by single-antigen flow-beads can be enormous. This becomes even more astonishing when a patient had contact to only few foreign HLA-antigens and the amount of detected HLA-antibodies seems to be out of proportion. Two factors can explain this discrepancy. First, HLA-antibodies are often directed against public epitopes that are shared by more than one HLA-antigen and therefore can lead to positive results with many single-antigen flow-bead (e.g. antibody against Bw4- or Bw6-epitope) (16). Second, infections with viruses or bacteria may induce production of antibodies that cross-react with foreign HLA-antigens (i.e. heterologous immunity) (17). Another important aspect regarding pretransplant risk assessment is the relevance of HLA-antibodies detected only in remote sera. Several studies and our data indicate that the number and titer of HLA-antibodies change over time and may become undetectable even by the most sensitive current assays (18, 19). Nevertheless, remote donor-specific HLA-antibodies are widely accepted as a risk factor for early allograft rejection and graft loss (1). For these reasons we regarded all HLA-antibodies detected by flow-beads in the current and remote sera as a risk and included them for virtualXM analysis.

Despite induction therapy, 4 of 9 patients with DSA defined by virtualXM showed biopsy-proven AMR indicating that the detected HLA-antibodies were indeed donor-specific and relevant. Lack of even subtle signs for AMR in protocol biopsies in the remaining four patients may be due to the successful induction therapy with ATG/IvIg. Interestingly, only one of these four patients had a positive FCXM (Table 3, #7), whereas they were negative in the other three patients (Table 3, #1, 6, 9). The combination of DSA defined by virtualXM with a negative FCXM is a diagnostic challenge and allows for two different interpretations. First, flow-beads can be false positive due to antibodies directed against an epitope only present on the flow-beads but not on the donor-cells (e.g. allele-specific HLA-antibodies, antibodies against denatured epitopes, antibodies against the peptide presented by the HLA-molecule) (4, 20). Second, flow-beads may be true positive and the negative FCXM can be explained with its lower sensitivity to detect HLA-antibodies. Indeed, our data suggest that the cumulative semiquantitative strength of DSA assessed by flow-beads has to reach a certain value to induce a positive FCXM (Table 3). Further support for this interpretation comes from a study by Gloor et al., who reported that flow-beads had superior sensitivity to monitor DSA posttransplant compared to the FCXM (21). Therefore, the favorable outcome observed in those patients with DSA detected only by flow-beads could be due to very low HLA-antibody levels that can be successfully controlled by ATG/IvIg. Clearly, the clinical relevance of DSA only detectable by flow-beads has to be further evaluated in patients not having received an induction therapy.

Due to the accuracy of the virtualXM to predict clinical outcome, is may become an invaluable tool for organ allocation. Indeed, organs can be given with priority to sensitized patients with a negative virtualXM. Furthermore, in patients without DSA defined by virtualXM a pretransplant CDCXM or FCXM could be omitted, which will reduce cold ischemic time and the incidence of delayed graft function (22). These potential benefits will make the virtualXM approach of particular interest in lung and heart transplantation (6, 23). However, responsible use of virtualXM implies thorough assessment of presensitizing events and careful determination of HLA-antibody specificities in current and remote sera. Although this will require additional resources in the laboratory, a recent study found that screening for HLA-antibodies by flow-beads is cost-effective in renal allograft recipients (24).

This study raises the question whether there is still a need for cell-based XM. Clearly, the CDCXM is not sensitive enough to detect clinically relevant low level DSA. Based on the CDC-assay, none of the 65 patients in this study had DSA as both T- and B-cell XM were negative in current and remote sera. The FCXM has a higher sensitivity than the CDCXM and still offers the advantage of cell-based XM-assays to detect antibodies against the ‘real’ donor HLA-antigens. Unfortunately, this theoretical high specificity is hampered by binding of irrelevant antibodies to the cell surface leading to false positive FCXM results. Indeed, four patients with an uneventful clinical and histopathological course had a positive FCXM, but no DSA defined by virtualXM. Assuming that these patients had in fact no DSA, the rate of false positive FCXM in our cohort was 7% (4 of 59 patients) despite optimization of the FCXM assay using pronase pretreatment and standardization with MESF-beads. On the other hand, a current limitation of flow-beads is the difficulty to assess the total load and strength of DSA, particularly when several DSA are present. This load can be determined by cell-based XM assays (e.g. FCXM+/CDCXM−) and these semiquantitative levels at the time of transplant are predictive for subsequent outcomes (7). Therefore, a strategy using virtualXM for risk stratification and a cell-based XM assay to determine the load of potentially present DSA seems currently to be a good solution for pretransplant risk assessment.

Our study has several advantages to evaluate the diagnostic value of virtualXM. First, the patients represent ‘real-life’ as we included all consecutive patients transplanted within 15 months. Second, risk assessment with consecutive adaptation of the immunosuppressive regimen was only based on virtualXM results. Third, serial protocol biopsies were used to determine the histopathological outcome. And fourth, the results of virtualXM were compared with retrospectively obtained FCXM. Since our study included only 65 patients and was conducted in a single center the results have to be confirmed in a larger cohort and different transplant centers.

In conclusion, virtualXM can accurately define absence or presence of DSA and may become an invaluable tool for organ allocation and pretransplant risk assessment. However, it does not include non-HLA-antibodies, which can cause severe AMR including graft loss. In addition, further studies need to address whether all HLA-antibodies detected by flow-beads are in fact clinically relevant.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

The authors thank Ineke Oehri and the staff of the renal transplant unit for their help with clinical data collection. SS is supported by grants from the University of Basel (VFWAWF, AstraZeneca) and the Swiss National Foundation (3200B0-109302).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Gebel HM, Bray RA, Nickerson P. Pre-transplant assessment of donor-reactive, HLA-specific antibodies in renal transplantation: Contraindication vs. risk. Am J Transplant 2003; 3: 14881500.
  • 2
    Patel R and Terasaki PI. Significance of the positive crossmatch test in kidney transplantation. N Engl J Med 1969; 280: 735739.
  • 3
    Johnson AH, Rossen RD, Butler WT. Detection of alloantibodies using a sensitive antiglobulin microcytotoxicity test: Identification of low levels of pre-formed antibodies in accelerated allograft rejection. Tissue Antigens 1972; 2: 215226.
  • 4
    Bray RA, Nickerson PW, Kerman RH, Gebel HM. Evolution of HLA antibody detection: Technology emulating biology. Immunol Res 2004; 29: 4154.
  • 5
    Pei R, Lee JH, Shih NJ, Chen M, Terasaki PI. Single human leukocyte antigen flow cytometry beads for accurate identification of human leukocyte antigen antibody specificities. Transplantation 2003; 75: 4349.
  • 6
    Appel JZ III, Hartwig MG, Cantu E III, Palmer SM, Reinsmoen NL, Davis RD. Role of flow cytometry to define unacceptable HLA antigens in lung transplant recipients with HLA-specific antibodies. Transplantation 2006; 81: 10491057.
  • 7
    Stegall MD, Gloor J, Winters JL, Moore SB, Degoey S. A comparison of plasmapheresis versus high-dose IVIG desensitization in renal allograft recipients with high levels of donor specific alloantibody. Am J Transplant 2006; 6: 346351.
  • 8
    Akalin E, Ames S, Sehgal V et al. Intravenous immunoglobulin and thymoglobulin induction treatment in immunologically high-risk kidney transplant recipients. Transplantation 2005; 79: 742.
  • 9
    Racusen LC, Colvin RB, Solez K et al. Antibody-mediated rejection criteria – an addition to the Banff 97 classification of renal allograft rejection. Am J Transplant 2003; 3: 708714.
  • 10
    Vaidya S, Cooper TY, Avandsalehi J et al. Improved flow cytometric detection of HLA alloantibodies using pronase: Potential implications in renal transplantation. Transplantation 2001; 71: 422428.
  • 11
    Sumitran-Holgersson S, Wilczek HE, Holgersson J, Soderstrom K. Identification of the nonclassical HLA molecules, mica, as targets for humoral immunity associated with irreversible rejection of kidney allografts. Transplantation 2002; 74: 268277.
  • 12
    Sumitran-Karuppan S, Tyden G, Reinholt F, Berg U, Moller E. Hyperacute rejections of two consecutive renal allografts and early loss of the third transplant caused by non-HLA antibodies specific for endothelial cells. Transpl Immunol 1997; 5: 321327.
  • 13
    Dragun D, Muller DN, Brasen JH et al. Angiotensin II type 1-receptor activating antibodies in renal-allograft rejection. N Engl J Med 2005; 352: 558569.
  • 14
    Zou Y, Heinemann FM, Grosse-Wilde H et al. Detection of anti-MICA antibodies in patients awaiting kidney transplantation, during the post-transplant course, and in eluates from rejected kidney allografts by Luminex flow cytometry. Hum Immunol 2006; 67: 230237.
  • 15
    Mizutani K, Terasaki P, Rosen A, Esquenazi V, Miller J, Shih RN et al. Serial ten-year follow-up of HLA and MICA antibody production prior to kidney graft failure. Am J Transplant 2005; 5: 22652272.
  • 16
    Piazza A, Poggi E, Ozzella G et al. Public epitope specificity of HLA class I antibodies induced by a failed kidney transplant: Alloantibody characterization by flow cytometric techniques. Transplantation 2006; 81: 12981305.
  • 17
    Adams AB, Pearson TC, Larsen CP. Heterologous immunity: An overlooked barrier to tolerance. Immunol Rev 2003; 196: 147160.
  • 18
    Karpinski M, Rush D, Jeffery J et al. Flow cytometric crossmatching in primary renal transplant recipients with a negative anti-human globulin enhanced cytotoxicity crossmatch. J Am Soc Nephrol 2001; 12: 28072814.
  • 19
    Noreen HJ, McKinley DM, Gillingham KJ, Matas AJ, Segall M. Positive remote crossmatch: Impact on short-term and long-term outcome in cadaver renal transplantation. Transplantation 2003; 75: 501505.
  • 20
    Gebel HM, Harris SB, Zibari G, Bray RA. Conundrums with flowPRA beads. Clin Transplant 2002; 16(Suppl 7): 2429.
  • 21
    Gloor JM, Degoey S, Ploeger N et al. Persistence of low levels of alloantibody after desensitization in crossmatch-positive living-donor kidney transplantation. Transplantation 2004; 78: 221227.
  • 22
    Taylor CJ, Smith SI, Morgan CH et al. Selective omission of the donor cross-match before renal transplantation: Efficacy, safety and effects on cold storage time. Transplantation 2000; 69: 719723.
  • 23
    Zangwill SD, Ellis TM, Zlotocha J et al. The virtual crossmatch—A screening tool for sensitized pediatric heart transplant recipients. Pediatr Transplant 2006; 10: 3841.
  • 24
    McLaughlin K, Manns B, Nickerson P. The routine use of high-resolution immunological screening of recipients of primary deceased donor kidney allografts is cost-effective. Transplantation 2006; 81: 12781284.