SEARCH

SEARCH BY CITATION

Keywords:

  • Antibody-mediated rejection;
  • donor-specific antibodies;
  • HLA-sensitized patients;
  • Long-term outcomes;
  • positive crossmatch kidney transplant

Abstract

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

Renal transplant candidates with high levels of donor-specific anti-HLA antibodies have low transplantation rates and high mortality rates on dialysis. Using desensitization protocols, good short-term outcomes are possible in “positive crossmatch kidney transplants (+XMKTx)”, but long-term outcome data are lacking. The aim of the current study was to determine actual 5-year graft outcomes of +XMKTx. We compared graft survival and the functional and histologic status of 102 +XMKTx to 204 −XMKTx matched for age and sex. Actual 5-year death-censored graft survival was lower in the +XMKTx group (70.7% vs. 88.0%, p < 0.01) and chronic injury (glomerulopathy) was present in 54.5% of surviving grafts. Graft survival was higher in recipients with antibody against donor class I only compared with antibody against class II (either alone or in combination with class I) (85.3% vs. 62.6%, p = 0.05) and was similar to −XMKTx (85.3 vs. 88.0%, p = 0.64). Renal function and proteinuria ranged across a wide spectrum in all groups reflecting the different histological findings at 5 years. We conclude that when compared to −XMKTx, +XMKTx have inferior outcomes at 5 years, however, almost half of the surviving grafts do not have glomerulopathy and avoiding antibodies against donor class II may improve outcomes.


Abbreviations
AMR

antibody-mediated rejection

BFXM

B cell flow cytometric crossmatch

CDC

complement-dependent cytotoxicity

eGFR

estimated glomerular filtration rate

+XMKTx

positive crossmatch kidney transplants

−XMKTx

negative crossmatch kidney transplants

TFXM

T cell flow cytometric crossmatch

Introduction

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

Antibodies against donor HLA, termed donor-specific alloantibodies (DSA), pose a significant barrier to successful kidney transplantation because of the increased risk of both early and late graft loss (2001, 2009). However, since many candidates have antibodies which react against a broad range of HLA, finding a donor against whom they have no antibody can be difficult. Over the past decade, novel “desensitization” protocols have been developed to enable patients with high levels antidonor antibodies (termed here positive crossmatch kidney transplants, +XMKTx) to receive a transplant with acceptable short term outcomes (2003, 2000, 2003, 2002). Recently, paired donation and “acceptable mismatch” programs have provided other means for providing sensitized patients a donor against whom they have little or no antibody (2011, 2011, 2011, 2009).

+XMKTx appears to have higher patient survival compared to either dialysis or waiting for a HLA compatible transplant (2011). However, with regard to graft survival, few studies have provided data beyond 1 or 2 years after transplantation and those that have suggest inferior graft survival in +XMKTx compared to negative crossmatch kidney transplantation (−XMKTx) (2006). In addition, the late outcomes of transplants in patients with low levels of antibody and with antibody against class II HLA remain unclear. The goal of the current study was to determine the actual 5-year graft outcomes in patients with antibodies against donor HLA.

Materials and Methods

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

Antibody detection

The presence of antidonor alloantibody was determined using conventional crossmatch assays as previously described (2003). During the study period, the primary methods for determining the presence and level of DSA pretransplant were crossmatch assays including (1) complement dependent cytotoxic (CDC) crossmatch using the T cell antihuman globulin enhanced technique and; (2) T and B flow cytometric crossmatch (FXM) for CDC− patients. All serum was tested against donor cells for cytotoxicity using the T cell antihuman globulin enhanced complement-dependent cytotoxicity assay. However, since the T cells only express class I (and not class II), the only cytotoxic donor-specific alloantibody (DSA) identified was against class I. CDC negative patients were those who had a positive T cell flow cytometric crossmatch (TFXM) and/or a positive B cell flow cytometric crossmatch (BFXM). Solid-phase assays were utilized in our program beginning in 2006. Therefore, in 2006 using predesensitization stored serum, confirmation of DSA and its HLA specificity generally was performed retrospectively using a solid phase, single antigen bead assay (LABScreen, One Lambda, Canoga Park, CA, USA) with the amount reported as the mean fluorescence intensity of the highest single level against donor class I and/or class II. A mean fluorescence intensity of 1000 or greater was considered positive.

Study populations

In this IRB approved study, we analyzed the outcomes of 102 patients who underwent living donor +XMKTx at the Mayo Clinic in Rochester, Minnesota between January 2000 and December 2006 and met the following criteria: (1) positive crossmatch against their live donor prior to transplantation (i.e. prior to any pretransplant therapy); (2) retrospective verification of the presence of antibody with specificity against their live donor's HLA using a solid-phase assay; (3) achieved a negative complement-dependent cytotoxic (CDC) crossmatch assay on the day of transplant and (4) minimum of 5 years of follow-up after transplantation. A comparison group included 204 −XMKTx recipients matched 2:1 for era, age (±10 years) and sex for +XMKTx recipients.

We chose the cutoff of positivity of an MFI >1000 because during the 5-year follow-up we found that no graft losses occurred in any +XMKTx in which class I and/or class II DSA was <1000 at baseline (i.e. 100% 5 year death-censored graft survival). Of 154 +XMKTx, 102 recipients met inclusion criteria while 52 were excluded. These included 8 lost to follow-up; 16 had no DSA at baseline despite a positive crossmatch; 5 had DSA in which the highest MFI was <1000; 11 had no sera available for testing DSA retrospectively and 12 CDC+ patients who did not achieve a negative CDC crossmatch despite desensitization.

We also studied the outcomes of 204 living donor kidney transplants from the same time period who had a negative crossmatch (−XMKTx) at baseline against their donor. This group was developed from a pool of 831 −XMKTx recipients from the same time period by identifying two −XM living donors recipients of similar age (±10 years) and sex for + XMKTx recipient. Thirty-four −XMKTx patients who met these criteria were lost to follow-up and were excluded from matching. This group was not retrospectively tested for HLA antibodies.

Desensitization and immunosuppression

The goal of “desensitization” was to achieve a negative CDC crossmatch on the day of transplant or a channel shift less than 300 in the B cell flow cytometric assay. Three general approaches to desensitization were utilized including (1) plasma exchange with low-dose intravenous immune globulin plus splenectomy at the time of transplant (n = 16), (2) plasma exchange plus immune globulin (2 g/kg) without splenectomy (n = 48), (3) high-dose immune globulin alone (n = 21) and finally no pretransplant desensitization (n = 17). All +XMKTx received induction with rabbit polyclonal antilymphocyte antibodies and received tacrolimus, mycophenolate mofetil and prednisolone as maintenance immunosuppression. −XMKTx generally also received the same triple therapy. Thirty-three (16.2%) −XMKTx received other regimens including cyclosporine and sirolimus-based immunosuppression.

Biopsy scoring, renal function and proteinuria

Patients underwent surveillance allograft biopsy at 1- and 5-year time-points following transplantation using a percutaneous ultrasound-guided 18-gauge biopsy gun (Bard, Murray Hill, NJ, USA) as previously described. Each biopsy was deemed adequate for interpretation by a Mayo Clinic renal pathologist and scored using the Banff 97 classification (1999). In addition, the presence of C4d was assessed by immunoflourescence in the +XMKTx. Paired biopsies were examined retrospectively for peritubular capillaritis according to Banff 2007 criteria (2008).

Graft loss was defined as return to dialysis or retransplantation. Renal function was calculated from creatinine readings using the Modification of Diet in Renal Disease (MDRD) equation. Proteinuria was assessed using 24-hour urine collection.

Statistical analyses

Analysis was made on JMPv9. (SAS, Cary, NC, USA) using the Kruskal–Wallis/Wilcoxon, Pearson and paired t-tests according to the variable data comparison. Survival curves differences were calculated by the log-rank test. Matching was done using the GREEDY algorithm (1989).

Results

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

Patients and matched control subjects

Of the 102 +XMKTx patients, the mean age was 46.0 ±12 years, 69.6% were women and 96.1% were Caucasian (Table 1). Comparing this group to 204 −XMKTx matched controls, differences existed in factors primarily related to being sensitized to HLA including: a longer duration of dialysis pretransplant (44.1 ± 46.8 vs. 17.8 ± 19.6 months, p < 0.0001); a lower rate of preemptive transplantation (18.6% vs. 51.0 %, p < 0.0001); a higher rate of patients who had a prior kidney transplant (39.2% vs. 22.1 %, p = 0.002) and a higher mean HLA ABDR mismatch (3.7 ± 1.3 vs. 3 ± 1.7, p = 0.005). All patients had at least 5 years of follow-up (+XMKTx mean = 1781.4 ± 962.3 days vs. −XMKTx mean 2215.6 ± 859.2 days).

Table 1. Comparison of HLA-sensitized kidney transplant recipients and nonsensitized match control subjects*
 Nonsensitized N = 204HLA sensitized N = 102Class I only N = 36Combined class II/I N = 66p-Value
  • Data are expressed as mean and standard deviation, unless specified.

  • 1

    The results are pairwise comparisons between HLA-sensitized and nonsensitized matched controls

  • 2

    p value is comparison between sensitized (+XMKTXx) with antibody against donor class I only or donor class II/I (class II only or class I and II).

Age (years)46.9 ± 10246.3 ± 1247.6 ± 11.745.6 ± 12.20.691
Gender (female) % (n)69.6% (142)69.6% (71)83.3% (30)62.1% (41)1.01
Ethnicity (Caucasian) % (n)90.7% (185)96.1% (98)100% (36)93.9% (62)0.821
Retransplant % (n)22.1% (45)39.2% (40)27.8% (10)45.5% (30)0.0021
ABDR mismatch3 ± 1.73.7 ± 1.33.4 ± 1.43.8 ± 1.30.0051
Preemptive % (n)51% (104)18.6% (19)16.7% (6)19.7% (13)<0.0011
Dialysis time (months)17.8 ± 19.644.1 ± 46.851.5 ± 58.740.1 ± 38.8<0.0011
Living donor % (n)100% (204)100% (102)100% (36)100% (66)1.01
Living-related donor % (n)38.7% (79)44.1% (45)36.1% (13)48.5% (32)0.371
Primary disease % (n)
Diabetes26.5% (54)10.8% (11)13.9% (5)9.1% (6)0.0021
Polycystic13.7% (28)9.8% (10)11.1% (4)9.1% (6)0.331
AHGCDC (dilution)3.7 ± 9.95 ± 11.33 ± 9.10.182
CDC negative %(n)59.8% (61)44.4% (16)68.2% (45)0.022
CDC positive %(n)40.2% (41)55.6% (20)31.8% (21) 
TFXM channel shift187 ± 124.4267.6 ± 86.8150.6 ± 122.30.0012
BFXM channel shift329.7 ± 92.5310.2 ± 80.1338.5 ± 97.20.2392
Class I35.3% (36)100% (36)0% (0)n/a
Class II19.6% (20)0% (0)30.3% (20) 
Class I and II45.1% (46)0% (0)69.7% (46) 
Baseline class I MFI7388.2 ± 5515.89545.1 ± 4399.66211.8 ± 5733.20.0022
Baseline class II MFI5584.8 ± 6210.7138.8 ± 320.18555.4 ± 5873.1<0.0012
Follow-up (days)2215.6 ± 859.21781.4 ± 962.31904.3 ± 916.81714.3 ± 986.5<0.0011
Donor gender (Female)56.4% (115)59.8% (61)66.7% (24)56.1% (37)0.331
Donor age (years)42.6 ± 1241 ± 11.939.7 ± 11.341.7 ± 12.10.261

At baseline, 40% (n = 41) of the +XMKTx had a crossmatch that was CDC+ (median dilution 1:4, range 1:1–1:64). In the +XMKTx group, 36 had antibody against donor class I only, 20 had antibody against donor class II alone and 46 had antibody against both donor class I and II. The level of antibody against class I specificities was similar in the class I only group and in the class I and II combined group with similar mean highest MFI against donor class I specificities (MFI = 9545.1 ± 4400.0 vs. 8849.1 ± 4896.2, p = 0.51) and a similar proportion of CDC+ recipients (55.6% vs. 45.7%, p = 0.37). The mean highest MFI against class I was higher in these CDC+ patients compared to the FXM+ group (11398.7 ± 4196.7 vs. 4692.7 ± 4592.8, p < 0.0001); however, there was considerable overlap of MFI between these CDC+ and FXM+ patients. The level of DSA against class II specificities was similar in the class II only and both class I and II groups (7711.8 ± 6741.1 vs. 8922.2 ± 5493.9, p = 0.45).

There was poor correlation between MFI and the crossmatch technique in both class I and class II antibodies. The best fit line of class I MFI to TFXM was r2 = 0.28 and this was similarly poor in the class II MFI to BFXM (r2 = 0.24).

Five-year patient and graft survival

Patient survival at 5 years was lower in +XMKTx compared to −XMKTx (83.5% vs. 92.5% p = 0.01, Figure 1 panel A). Death-censored graft survival at 5 years also was lower in the +XMKTx (70.7% vs. 88.0%, p < 0.01, Figure 1 panel B). The only factor associated with increased patient death was prior dialysis chi-squared (6.8; p = 0.009).

image

Figure 1. Five-year outcomes after positive crossmatch live donor kidney transplant (+XMKTx). Actual 5-year outcomes are shown for +XMKTx including patient survival (panel A); overall death-censored graft survival (panel B); grafts survival by baseline crossmatch assay type (+XMKTx recipients who were CDC+ vs. CDC−, panel C); and graft survival by donor-specific HLA specificity (+XMKTx recipients with antibody against donor class I only, class II only and both class I and II, panel D). (A) 5-year patient survival. (B) 5-year overall graft survival. (C) Graft survival by crossmatch assay (CDC+ vs. CDC−/FXM+). (D) Graft survival by donor-specific HLA antibody specificity.

Download figure to PowerPoint

At 1 year, CDC+ transplants had lower graft survival than the CDC−/FXM+ recipients (82.4% vs. 96.7%, p = 0.02), but the difference was not statistically different at 5 years (64.9% vs. 72.7%, p = 0.26, Figure 1 panel C) demonstrating a similar rate of graft loss in CDC− grafts between 1 and 5 years.

The anti-HLA specificity had an important impact on 5-year death-censored graft survival (Figure 1 panel D). Graft survival was higher in recipients with antibody against donor class I only compared with those with class II (either alone or with class I) (85.3 vs. 62.6%, p = 0.05, Figure 1 panel D) and was similar to that of the −XMKTx (85.3 vs. 88.0%, p = 0.64). In the class I only group, 60% of graft losses occurred in the first year and the subsequent rate of graft loss averaged only 1.6% per year. This compares favorably with the rate of graft loss in the −XMKTx group (2.9% per year). In the class I only group, patients who had a CDC− crossmatch had graft survival similar to that of −XMKTx recipients (93.3% vs. 88.0%, p = 0.51). In contrast, antibody against donor class II was associated with a higher rate of graft loss over time (7.0% per year in the class II/I combined group).

The improved 5-year graft survival of class I versus class II/I recipients did not appear to be due to differences in anticlass I DSA levels (Figure 2). Also, there was no statistical difference in graft survival between CDC+ and CDC− patients with class II antibodies (54.3% vs. 66.2%, p = 0.23).

image

Figure 2. Antidonor antibody levels and outcome by specificity for donor HLA. Antibody levels against donor class I HLA were similar in the class I only group and the class II/I group (panel A). For class I, antibody levels were similar in surviving and failed allografts. Antibodies against donor class II were absent in class I group (panel B). The mean antibody levels were higher in class II/I CDC−/FXM+ grafts that failed compared to those that survived (11557.6 ± 6093.9 vs.7015.0 ± 5368.4,*p = 0.003). Scatterplots of antidonor antibody levels as determined by solid phase assay and expressed as mean fluorescence intensity demonstrate distribution across surviving and failed grafts at 5 years. (A) Antibody against donor class I. (B) Antibody against donor class II. The different antibody groups are represented by different colors. Green circles are class I only; orange circles are class II only and red circles are class I + II.

Download figure to PowerPoint

In general, baseline antibody levels by solid phase were similar in grafts that survived compared with grafts that failed (Figure 2). However, class II antibody at lower levels (e.g. a mean fluorescence intensity of <3000) had better survival compared to those with higher levels (5-year DCGS 86.7% vs. 54.6%, p = 0.05, data not shown). There was no significant difference in 5-year outcomes comparing DSA against HLA-DR or HLA-DQ (data not shown).

Causes of graft loss in class I only and class II/I groups were similar, with 60% related to transplant glomerulopathy overall (class I 60.0% (3/5) vs. class II/I 60.9% (14/23), p = 0.97).

Antibody-mediated rejection

The incidence of antibody-mediated rejection (AMR) was 37.2% (n = 38) in the +XMKTx group and 2.5% (n = 5) in the −XMKTx group (data not shown). Early AMR was less common in the class II only group (15%) compared to either the class I only (38.9%, p = 0.06) or recipients with both class I and II (45.7%, p = 0.02). +XMKTx patients with an episode of AMR had lower graft survival compared to those without early AMR (53.2% vs. 81.2%, p < 0.01) (Supporting Figure S1).

Renal function and proteinuria at 5 years

Renal function in the +XMKTx at 5 years spanned a wide spectrum and was not significantly different from −XMKTx (Figure 3 panel A) [+XMKTx mean estimated glomerular filtration rate (eGFR) = 44.4 ± 19.6 mL/min, (range 10–101) vs. −XMKTx mean eGFR = 48.5 ± 17.9, (range 11–106), p = 0.22)]. At 5 years, the percentage of surviving grafts with good function (GFR >40 mL/min) also was similar in +XMKTx vs. −XMKTx. (52.4% vs. 64.6%, p = 0.09).

image

Figure 3. Renal function and proteinuria after +XMKTx. Estimated glomerular function rate (eGFR) by MDRD at 1 (panel A) and 5 years (panel B) after transplantation. Twenty-four hour urine protein excretion at 1 (panel C) and 5 years (panel D) after transplantation.

Download figure to PowerPoint

Proteinuria (24-hour urinary protein excretion), has been associated with chronic antibody-mediated injury (2007). At 5 years, the mean 24-h urine protein excretion was higher in the +XMKTx vs. −XMKTx patients (1161.8 ± 2089.8 vs. 282.7 ± 783.4, p < 0.01; Figure 3 panel B). In the +XMKTx, 44.4% (n = 20) of patients had 24-h urinary protein excretion greater than 500 mg/24 h compared to only 13.8% (n = 18, p < 0.01) in −XMKTx at 5 years. Proteinuria was significantly associated with the presence of transplant glomerulopathy at both 1 year (“CG = 0” 108 [range 26–7788] vs. “CG > 0” 496 [range 75–9997], p < 0.01) and 5-year (“CG = 0” 72 [range 23–522] vs. “CG > 0” 694 [range 49–10826], p < 0.01). The level of proteinuria at 1 year was significantly higher in patients with no CG at 1 year biopsy who developed CG by the 5 year biopsy (“5-year CG = 0” 76 [range 26–229] vs. “5-year CG > 0” 134 [range 26–2405], p < 0.01).

Allograft histology at 1 and 5 years

Similar to functional data, allograft histology provides insight into the extent of injury in functioning allografts. At 1 year after transplantation, surveillance biopsies were obtained in 75.1% (142/189) of the −XMKTX; 77.4% (24/31) of the class I only group; and 80.4% (45/56) of the class II/I group. The entire dataset is presented in Supporting Table S1.

At 1 year, chronic glomerulopathy (cg, score >0), a histologic finding associated with chronic antibody mediated injury, was more common in +XMKTX compared to −XMKTX (27.5% vs. 3.5%, p < 0.0001, Table 2). CG was more common in the class II/I group compared to class I only (37.8% vs. 8.4%, (p = 0.009) at 1 year, and was only seen in class II/I patients with DSA MFI>3000. By 5 years, there was equivalent prevalence of CG between the class II group regardless of baseline MFI (<3000 50% vs. 62.5%, p = 0.34).

Table 2. Summary of histologic lesions commonly associated with anti-HLA antibody. The overall incidence of histologic changes commonly associated with anti-HLA antibody found on paired surveillance biopsies obtained at 1 and 5 years after transplantation are shown
 Chronic glomerulopathyAcute glomerulitisPeritubular capillaritis
 1 year5 years1 years5 years1 year5 years
−XMKTx4.2%7.4%7.4%6%NotNot
     donedone
+XMKTx21.2%54.5%30.3%63.6%63.3%50%
Class I10%50%20%50%50%55.7%
Class II/I26.1%56.5%34.7%65.2%66.6%90.9%

Acute glomerulitis at 1 year was common in +XMKTx occurring in 44.4% of class II/I patients, significantly higher than 20.9% of class I patients (p = 0.05) and 7.7% of −XMKTX patients. In contrast to acute glomerulitis, acute peritubular capillaritis (ptc) was not significantly different between class I or class II/I patients at 1 year (53.3% vs. 75%, p = 0.14) but at 5 years this was significantly different (45.5% vs. 91.3%, p = 0.003). Interstitial fibrosis (ci) was similar in all groups at 1 year (range 52.1–68.9%). By 5 years, the prevalence of cg had increased to 58.3% in the class II/I group, 50.0% in the class I group and was only 7.7% in the −XMKTX.

The presence of TG at 1 year resulted in graft loss in 30.4% (n = 7) class II patients at 5 years and 20% (n = 1) class I patient. In contrast, the absence of TG at 1 year was associated with graft survival rates of >80% by 5 years.

Paired allograft histology at 1 and 5 years

The data from the subset of patients who were biopsied at both 1 and 5 years (paired biopsies) also might yield data regarding the progression of antibody-mediated injury. Of the grafts that survived 5 years, paired surveillance biopsies at both 1 and 5 years were available for analysis in 56% (n = 95) of the −XMKTx and 52% (n = 33) +XMKTx. At 1 year, the prevalence of chronic glomerulopathy, the histologic lesions most closely associated with chronic antibody-mediated injury was 21.2% in +XMKTx and only 4.2% in −XMKTx (Table 2). This increased to 54.5% and 7.4% respectively by 5 years. At 5 years, chronic glomerulopathy was equally common in grafts in the class I only group and in the class II/I group. Importantly, slightly more than half of all +XMKTx patients with no cg at 1 year also had no cg at 5 years (15/26) including: 50% (5/10) of patients in the class I group and 43.5% (10/23) of patients in the class II/I group. Figure 4 shows the actual Banff scores in paired biopsies obtained at 1 and 5 years in +XMKTx. These data show that in addition to an increase in the prevalence of glomerulopathy between 1 and 5 years, the severity also increased in many patients. However, at 5 years, 50% of the class I only group and 43.5% of the class II/I group had no evidence of chronic glomerulopathy.

image

Figure 4. Detailed histologic score at 1 and 5 years in paired biopsies in ±XMKTx. Actual Banff scores at 1 and 5 years in paired biopsies in +XMKTx including chronic glomerulopathy (panel A); glomerulitis (panel B), peritubular capillaritis (panel C), interstitial fibrosis (panel D) and tubular atrophy (panel E).

Download figure to PowerPoint

The prevalence and severity of inflammatory changes associated with antibody-mediated injury (glomerulitis and peritubular capillaritis) increased from 1 to 5 years in +XMKTx (Figure 4 panels B and C). Interstitial fibrosis and tubular atrophy also was commonly seen, but was mild in many cases (Figure 4 panels D and E). See Supporting Tables S1 and S2 for all biopsy data.

Outcomes of +XMKTx excluded

In addition to the 102 +XMKTx studied above, we excluded 52 +XMKTx including: 8 lost to follow-up prior to 5 years; 12 CDC+ patients who did not achieve a negative CDC− crossmatch by the day of transplantation (5 DCGS was 41.7% and all surviving grafts developed CG); 16 (1 CDC+, 15 FXM+) had no DSA detected by solid phase assay (the 5 year DCGS was 93.8% and none of the grafts developed CG); 5 had an MFI <1000 (the 5 year DCGS was 100%) and 11 had no pretransplant sera stored for DSA testing (5 year DCGS was 62.3%).

Discussion

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

In this study, we analyzed the actual 5-year outcomes of +XMKTx. Our results showing a 5-year patient survival of 83.5% % were similar to those of HLA-incompatible kidney transplant recipients recently reported by the Johns Hopkins group (2011). The overall 5-year graft survival of 70.7% was similar to that of FXM+ living donors at the University of Maryland (70% in 19 patients followed to 5 years) (2009) and in deceased donors with donor-specific antibodies at Hôpital Saint-Louis de Paris (70% in 30 patients followed 5 years) (2010).

The current study supports prior studies showing that CDC+ allografts with high levels of anticlass I antibody have a high rate of early graft loss (2011). However, our study suggests that the anti-HLA specificity may have an even greater impact on long-term graft survival. Most of the graft losses in the class I only group occurred in the first year and the rate of graft loss between 1 and 5 years was similar to −XMKTx. In contrast, class II DSA appears associated with a low rate of early AMR (15% in those with class II only), yet a very high rate of chronic injury with more than 40% of allografts failing within the first 5 years after transplantation. Patients with antibody against both donor class I and II (almost half of the +XMKTx cohort) had similar poor long-term outcomes as patients with class II only. The rate of allograft loss is associated with the presence of CG at an earlier stage in the class II/I group, than class I only. The whole group histology at 1 year demonstrates 37.8% in class II/I group, compared to only 8.3% in the class I only cohort (p < 0.01). Overall, if 1 year CG is present it is highly correlated with increased graft failure at 5 years (42.1% vs. 12%, p < 0.01). The development of proteinuria in allografts at 1 year without histology evidence of CG was a good predictor of the development of CG at 5 years.

The cohort in this study was from the early “era” of our positive crossmatch transplant program. More recent data have shown a trend toward fewer graft losses in the first year after transplantation. This is likely due to many different protocol changes including improved DSA monitoring. However, we believe that the actual graft survival data presented here reflects the true incidence of chronic injury that occurs over time in +XMKTx—especially in patients with anticlass II DSA.

The functional and histologic data provide important additional insight into the outcomes of +XMKTx. Chronic glomerulopathy was common at 1 year and increased over time in patients with antidonor HLA antibody. Similar to the findings of Haas et al. (2007) and Loupy et al. (2009), chronic inflammation (peritubular capillaritis and glomerulitis) also was extremely common at 1 and 5 years after +XMKTx (2007). An encouraging finding from the histology data was that approximately half of the biopsies at 5 years showed no evidence of transplant glomerulopathy suggesting that some grafts may avoid chronic antibody-mediated injury. Conversely, a worrisome finding is the increasing prevalence of inflammation and glomerulopathy from 1 to 5 years in the class I only group suggesting that despite having survived 5 years, these grafts may still be destined to fail from chronic injury. It should also be emphasized that more grafts with class II DSA were lost in the first 5 years after transplantation. Thus, the negative impact of class II DSA on graft outcomes is underestimated by the paired histologic data presented.

A limitation of this study is that posttransplant antidonor antibody data were not available for sufficient number of patients to be analyzed in this study. Longitudinal studies correlating antibody levels and specificity with outcomes after +XMKTx are needed. In addition, the histologic data at 5 years should be interpreted with some caution because only 52% of eligible patients underwent paired surveillance biopsies.

Sensitization to HLA is a major problem affecting approximately 30% of wait-listed kidney transplant candidates. The results of this study have important implications in both informed consent for patients and for the development of protocols for the management of renal transplant candidates with anti-HLA antibody. The data suggest that protocols designed to enhance the transplantation rate and outcomes of sensitized patients should consider the presence of antibody against donor class II HLA in their algorithms and the sole reliance on a negative CDC crossmatch is insufficient. When available, we contend that both paired donation and acceptable mismatch programs are important and viable options for some sensitized patients. Most, if not all, sensitized candidates should be entered into a paired donor program for at least a short period of time. However, despite these efforts, many sensitized candidates will never be matched with a donor against whom they have no antibody (2011). Thus, a +XMKTx may be the only realistic option for transplantation. While many of the grafts in +XMKTx may fail within 5 years necessitating another kidney transplant, others will show no signs of chronic injury. Given the expected good outcome, +XMKTx patients with antibody against donor class I only that are CDC− might be considered relatively low risk and thus transplanted without a prolonged wait in the paired or acceptable mismatch programs. The same consideration also applies to +XMKTx patients with antibodies against class II with MFI < 3000 who also comprise a lower risk group within the +XMKTx population.

Finally, the study highlights the fact that chronic injury affects the majority of +XMKTx in the first 5 years leading to a very high rate of graft loss. Thus, these data clearly suggest that the prevention and treatment of chronic antibody-mediated injury should be a major focus of future research.

Acknowledgments

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

The authors thank the Mayo Clinic transplant coordinators for their tireless efforts in patient follow-up and data collection.

AB received a grant from The University of Birmingham, UK: Clinical Fellowship Program.

An abstract of this project was accepted for presentation at the 2012 American Transplant Congress and also at the 2012 – XXIV TTS□Congress – Berlin, Germany. Thus, the abstract was published in the American Journal of Transplantation Special Issue: 2012 American Transplant Congress Volume 12, Issue Supplement s3, May 2012.

Disclosure

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

The authors of this manuscript have conflicts of interest to disclose as described by the American Journal of Transplantation. MDS has research contracts with Alexion Pharmaceuticals and Millennium Pharmaceuticals. LDC has research contracts with Alexion Pharmaceuticals.

References

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

Supporting Information

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

Disclaimer: Supplementary materials have been peer-reviewed but not copyedited.

FilenameFormatSizeDescription
ajt4291-sup-0001-suppmat.doc528K

Table S1: Demographics describing the class II only and the class I + II groups

Table S2: One and 5-year biopsies

Table S3: One and 5-year paired biopsies

Figure S1: Allograft survival and antibody mediated rejection.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.