Effect of persistent versus transient donor-specific HLA antibodies on graft outcomes in pediatric cardiac transplantation

http://www.jhltonline.org

ORIGINAL CLINICAL SCIENCE

Effect of persistent versus transient donor-specific HLA antibodies on graft outcomes in pediatric cardiac transplantation Claire A. Irving, MRCPCH, MD,a Vaughan Carter, PhD, FRCPath,b Andrew R. Gennery, MD, FRCPCH,c Gareth Parry,d Massimo Griselli, FRCS,a Asif Hasan, FRCS,a and C. Richard Kirk, MBBChir, FRCPCHa From the aInstitute of Transplantation, Department of Paediatric Cardiothoracic Transplantation, Freeman Hospital, Newcastle upon Tyne, UK; bNHS Blood and Transplant, Newcastle upon Tyne, UK; cInstitute of Cellular Medicine, Newcastle University, Newcastle upon Tyne, UK; and the dDepartment of Cardiology, Freeman Hospital, Newcastle upon Tyne, UK.

KEYWORDS: donor-specific; HLA antibodies; pediatric cardiac transplantation; graft outcomes

BACKGROUND: De novo donor-specific HLA antibodies (DSA) are a risk for poor graft outcomes, but there is little evidence of their long-term effect in pediatric cardiac transplantation or of the effect of transient versus persistent DSA found using newer antibody testing methods. METHODS: Archived serum samples were obtained from patients o18 years of age who underwent primary cardiac transplantation during the period from 1996 to 2009. Luminex antibody testing was performed at 3 months, 6 months and 1 year post-transplant, and then annually. Outcomes including cardiac allograft vasculopathy (CAV), rejection and graft loss were correlated with the presence or absence of DSA or non-donor-specific HLA (non-DSA) antibodies. RESULTS: Six hundred ninety-one samples from 108 patients, with mean age at transplant of 7.4 (0.1 to 15.9) years and mean follow-up 8.2 (1.9 to 15.7) years, were studied. Forty-three (40%) patients had DSA (which were persistent in 58%), 41 (38%) had non-DSA (persistent in 46%) and 24 (22%) had no antibodies. In those with DSA, 30% had Class I antibodies, 47% Class II and 23% both Class I and II, whereas, in the subgroup with persistent DSA, 88% had Class II antibodies. There were 14 cases of graft loss, 9 of these in patients with persistent DSA. All had Class II antibodies. There was an increased incidence of CAV, rejection and graft loss in those with persistent DSA. Outcomes were similar between the group with non-DSA antibodies and the group with no antibodies. CONCLUSIONS: De novo HLA antibodies are detectable post-transplant in the majority of patients, but non-DSA and transient DSA do not appear to be associated with poor outcomes. Patients with persistent DSA, especially those with Class II DQ antibodies, have worse survival. J Heart Lung Transplant ]]]];]:]]]–]]] r 2015 International Society for Heart and Lung Transplantation. All rights reserved.

The presence of HLA antibodies in the serum of solidorgan transplant recipients, directed against donor antigens, Reprint requests: Claire A. Irving, MRCPCH, MD, Department of Paediatric Cardiothoracic Transplantation, Freeman Hospital, Freeman Road, Newcastle upon Tyne NE7 7DN, UK. Telephone: þ44 797 6877920. Fax: þ44 191 2231314. E-mail address: [email protected]

correlates with an increased risk of rejection and reduced graft survival.1–3 There is evidence of adverse outcomes and increased antibody-mediated rejection (AMR) in adult heart transplant recipients with donor-specific antibodies (DSA).2,4,5 The presence of Class II DSA has also been linked to an increased incidence of cardiac allograft vasculopathy (CAV).6

1053-2498/$ - see front matter r 2015 International Society for Heart and Lung Transplantation. All rights reserved. http://dx.doi.org/10.1016/j.healun.2015.05.001

2

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

One study examining kidney graft function and appearance of HLA antibodies post-transplantation over a 5-year period demonstrated that de novo donor-specific HLA antibody formation was detrimental to graft function. In that study, patients who developed HLA antibodies formed these 41 year before the graft failed, indicative of a long-term, ongoing rejection process.7 This study also showed an increased risk of rejection in patients with non-DSA with these antibodies detected well in advance of the graft failure. There has been an increase in the number of reports of DSA in pediatric renal8 and cardiac transplantation,1,9–11 but there remains little data on the incidence of de novo donor and non-donor-specific HLA antibodies in this group and possible effects on graft outcomes. There have been no longitudinal studies examining the appearance of DSA over the medium and long term in children who have undergone cardiac transplantation. There is evidence in renal transplantation that persistent DSA cause ongoing immunologic damage; however, patients with non-persistent DSA have similar outcomes to DSA-negative patients,8 suggesting that patients with non-persistent DSA may develop a degree of immunologic accommodation to the donor antigens. In view of these findings we hypothesized that: (1) de novo HLA antibodies are identifiable in a significant proportion of post-transplant patients, but, in some, they may be transient, with less of an effect on medium- and long-term graft outcomes; and (2) patients in whom DSA are persistent have worse outcomes when compared with patients who do not develop antibodies. Developments in HLA antibody screening assays have led to a change in methodology from the use of complement-dependent cytotoxicity (CDC) and enzymelinked immunosorbent assays9,12 to more sensitive and specific methods of antibody detection using flow beads and Luminex methodology.10 These new techniques allow for identification of HLA antibodies that were previously undetectable, and have provided new areas for research. We used these sensitive techniques to investigate the appearance and significance of HLA antibodies over time in patients who underwent cardiac transplantation in childhood to determine the potential effect of both transient and persistent DSA and non-DSA on medium- and long-term graft outcomes.

Methods Patients o18 years of age who underwent primary cardiac transplantation between 1996 and 2009 at a single center were included in this study. Archived stored serum samples from these patients were analyzed. Patients were excluded if they underwent concomitant lung transplantation (n ¼ 4), died within 1 year of transplant (n ¼ 8), or if their stored samples were not available for study (n ¼ 8). One patient with DSA detected on retrospective testing of pre-transplant samples was also excluded. Clinical data and transplant outcomes were obtained from the departmental cardiothoracic transplant database and from hospital paper-based medical records. Due to the retrospective nature of the study, patients in whom donor-specific HLA antibodies were detected did not receive specific treatment for antibody-mediated rejection. Only 1 patient

treated for rejection was known to have DSA and received treatment with rituximab in addition to standard treatment with anti-thymocyte globulin (ATG) and methylprednisolone.

HLA antibody testing and patient groups Luminex antibody testing was performed on archived serum samples taken at 3 months, 6 months and 1 year post-transplant and then annually until either time of death, re-transplantation or the end of the study period. Luminex screens for HLA Class I and Class II antibodies were performed using LABScreen mixed HLA antibody screening kits (One Lambda, Inc., Canoga Park, CA) with the flow analysis (LABScan 100) for data acquisition and analysis. Samples positive on screening were further tested by single-antigen beads using LABScreen single-antigen kits (One Lambda). For HLA antibody screening, a positive reaction was determined by the manufacturer’s recommended guidelines (1.5 times the negative control bead as positive, 1.2 to 1.5 times as reactive). For all positive and reactive results, HLA antibody specificities were determined using LABScreen single-antigen kits with a mean fluorescence intensity of 4500 considered positive. Patients with antibodies were classified as having DSA or nonDSA based on an analysis of donor HLA typing. Patients were further sub-divided as having persistent HLA antibodies (presence of antibodies on Z3 consecutive samples) or transient HLA antibodies. A correlation between outcomes of CAV, rejection and graft loss (re-transplantation or death) was made between groups.

Outcomes Rejection was defined according to the International Society for Heart and Lung Transplantation (ISHLT) grading system for cardiac allograft rejection.13,14 Episodes of acute severe rejection were defined as either endomyocardial biopsy samples with cellular rejection Grade Z3A (2R), antibody-mediated rejection Grade 3, or hemodynamic compromise and poor ventricular function on echocardiography requiring augmentation with corticosteroids. CAV was diagnosed on the basis of coronary angiographic imaging or on post-mortem examination.

Immunosuppression and post-transplant management Immunosuppression consisted of cyclosporine and azathioprine, with methylprednisolone given intra-operatively at the time of cross-clamp release. All patients received induction therapy with anti-thymocyte globulin (ATG) or basilixumab. In patients o5 years of age at transplant, 3 doses of methylprednisolone were given on Day 1 post-operatively and patients were subsequently maintained on a steroid-free regimen with a calcineurin inhibitor. In the majority of cases, an anti-metabolite drug (azathioprine or mycophenolate mofetil) was added. In patients 45 years of age at transplant, steroids were weaned over the first 6 to 12 months posttransplant. Routine endomyocardial biopsies and coronary angiography were not performed in children o5 years of age, but urgent biopsies were performed at any age if rejection was suspected. In children 45 years of age, biopsies were performed at 1 week posttransplant in most cases, but thereafter only if clinically indicated until 2 years post-transplant, after which they were performed annually during routine surveillance. Older teenagers followed the

Irving et al.

Persistent and Transient DSA

3

adult biopsy protocol of weekly biopsies for the first month, every 2 weeks until 8 weeks, then monthly until 3 months posttransplant, followed by annual biopsies and coronary angiography (via cardiac catheterization with selective injection of the coronary arteries).

Statistical analysis Unpaired t-tests and 1-way analysis of variance were used to compare normally distributed groups and the Mann-Whitney U-test was used for non-parametric tests. Kaplan-Meier curves were used to assess survival with the log-rank test assessing for differences between groups, acknowledging the fact that development of HLA antibodies is a time-dependent variable. Univariate analysis was performed using Cox proportional hazards regression analysis with number of HLA mismatches at transplant considered significant if 2 mismatches were present and ischemic time prolonged if 4240 minutes. Multivariable Cox proportional hazards models were also performed (MEDCALC for Windows version 9; MedCalc Software, Mariakerke, Belgium). p o 0.05 was considered statistically significant for all tests.

Results

patients had a CDC-negative cross-match at time of transplant. Figure 1 depicts the results of the antibody analysis and number of patients in each group. Patient demographics for each group are shown in Table 1.

Antibody classes and specifications in DSA group

One hundred and eight patients (56% female) were included in the study. Mean age at transplant was 7.4 years (range 0.1 to 15.9 years), with mean follow-up 8.2 years (range 1.9 to 15.7 years). One patient underwent an additional kidney transplant during the study period. Patients who underwent concomitant lung transplantation (n ¼ 4), those who died within 1 year of transplant (n ¼ 8) and those for whom stored samples were not available for study (n ¼ 8) were excluded. In the 8 excluded patients who died at o1 year post-transplant, causes of death included acute graft failure (4), infection (1), intracranial hemorrhage (1), lymphoma (1) and unknown (1). One patient with DSA detected on retrospective testing of pre-transplant samples was also excluded. Seventy-eight of 108 (72%) patients had cardiomyopathy (CM), and 30 of 108 patients (28%) had congenital heart disease (CHD). All patients with CHD had undergone previous surgical (n ¼ 29) or transcatheter (n ¼ 1) interventions. Twenty-eight patients (26%) were on mechanical support before transplant (18 ventricular assist device, 10 extracorporeal membrane oxygenation). Twelve patients (11%) received an ABO-incompatible graft. All Table 1

Figure 1 Flow diagram of the study groups with the DSA and non-DSA patients, divided into transient and persistent groups.

Mean age at time of first appearance of DSA was 10.9 (range 2.5 to 26.6) years, with no difference in age at transplant and time to appearance of DSA (p ¼ 0.76). Median time from transplant to DSA development was 2.0 (range 0.25 to 11) years. Patients who developed persistent DSA produced them later post-transplant than those with transient DSA (median 5.0 years vs 0.5 years posttransplant, p ¼ 0.006). The majority of patients with DSA had Class II antibodies (Table 2). Antibodies to HLA Class II were the most common, and the majority of DSA in all groups were against HLA-DQ antigens, with antibodies specific for HLA-DQ2 being the most prevalent. In the persistent DSA group, 17 patients (68%) had HLA-DQ2 antibodies, as compared with 7 patients (39%) in the transient DSA group. Overall, 90.7% of patients had not developed any DSA at 3 months, 86.1% at 6 months, 84.3% at 1 year, 76.7% at 2 years, 68.6% at 5 years, 58% at 10 years and 47.9% at 15 years post-transplant. The mean fluorescence intensity (MFI) at the time of antibody detection in the persistent DSA group was 2,369

Patients’ Demographics by Antibody Status

Number Age at transplant (years) Follow-up (years) Gender Pre-transplant diagnosis Pre-transplant mechanical support ABO-incompatible graft

DSA group

Non-DSA group

No antibodies group

43 7.6 (range 0.4–15.9, SD 5.22) 8.4 (range 2.6–15.5, SD 4.11) Female 26 (60.4%), male 16 (40.6%) CM 30 (70%), CHD 13 (30%) 12 (28%)

41 6.5 (range 0.1–15.9, SD 4.97) 8.7 (range 1.9–15.7, SD 3.98) Female 18 (44%), male 23 (56%) CM 31 (76%), CHD 10 (24%) 11 (27%)

24 8.6 (range 0.3–15.1, SD 5.22) 6.8 (range 2.1–15.5, SD 4.22) Female 16 (67%), male 8 (33%) CM 17 (71%) CHD 7 (29%) 5 (21%)

4 (9.3%)

4 (9.7%)

4 (16.7%)

CHD, congenital heart disease; CM, cardiomyopathy; DSA, donor-specific antibodies.

p 0.29 0.17 0.20 0.82 0.70 0.62

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

4 Table 2

Breakdown of Antibody Class in Patients With DSA

n Class I alone Class II alone Both Class I and Class II

Overall DSA group

Persistent DSA

Transient DSA

43 12 (28%) 23 (53%) 8 (19%)

25 5 12 8

18 7 11 0

(58%) (20%) (48%) (32%)

(42%) (39%) (61%) (0%)

(range 500 to 10,520) and, although there was a trend toward a higher MFI in the persistent compared with transient DSA group, it did not reach statistical significance (p ¼ 0.08).

differences [p ¼ 0.31], age at transplant [p ¼ 0.2 to 0.6], size mismatch [p ¼ 0.63], ischemic time [p ¼ 0.59], indication for transplant [p ¼ 0.49] and use of pre-transplant mechanical support [p ¼ 0.28]), the only significant factor was presence of post-transplant DSA, particularly persistent DSA (p ¼ 0.007). In multivariate analysis, only the development of persistent DSA was found to be a significant independent predictor of poor graft survival (p ¼ 0.0009). The MFI for patients with DSA and graft loss was 3,148 (range 600 to 9,000, standard deviation 2,012) when DSA were last detected. The strength of the antibody assessed by MFI did not correlate with graft loss when mean MFI values for those who lost their grafts were compared with those with persistent DSA who did not lose their grafts.

Cardiac allograft vasculopathy

Overall outcomes Overall there was an increased incidence of CAV, acute rejection and graft loss in patients with persistent DSA when compared with those with transient DSA or non-DSA (Table 3), but numbers were too small to determine statistical significance.

Graft outcomes Graft survival Patients with DSA had worse survival when compared with the non-DSA group or those with no HLA antibodies (p ¼ 0.004; Figure 2). Median time from development of DSA to graft loss was 1.7 (0.1 to 4.8) years. Patients with persistent DSA had worse graft survival when compared with those with transient DSA (p ¼ 0.049; Figure 3). Graft survival by antibody group (Figure 2) and DSA status (Figure 3) showed that development of antibodies is a time-dependent variable and therefore not consistent for each patient in these groups. Fourteen of 108 (13%) patients had graft loss (Table 3). Ten of these (71%) had DSA, which were persistent in 9. Nine patients died of acute rejection (all had evidence of acute cellular rejection with 3 patients having additional features of antibody-mediated rejection), 2 of CAV and 3 of non-cardiac causes (Table 4). One patient with DSA died from a non-cardiac cause. On univariate analysis of other factors that could influence post-transplant graft survival (including number of HLA mismatches [p ¼ 0.79], gender

Ten patients developed CAV. Nine of these (90%) had DSA, 6 in the persistent and 3 in the transient DSA group. Seven of the 9 (78%) patients with DSA and CAV had Class II DSA (all including DQ antibodies). Two patients had Class I DSA alone (both had HLA-B antibodies). Angiographic findings, intervention and outcome for each patient with CAV are outlined in Table 5. Mean time from transplant to CAV was 5.1 (range 1.8 to 9.9) years. DSA were detected at a median of 2.2 (1 to 4.9) years before CAV diagnosis. Freedom from CAV was significantly less in the group with DSA (68% at 10 years and 88% in the non-DSA or no antibodies group; p ¼ 0.048). On univariate analysis of factors potentially implicated in the development of CAV, the development of posttransplant DSA (p ¼ 0.0005) and persistent DSA (p ¼ 0.006) had a strong association with the development of CAV. The only other factors of significance were a previous episode of rejection (p ¼ 0.019) and older age at the time of transplant (p ¼ 0.02 for the 1- to 10-year age group, and p ¼ 0.016 for the 410-year age group). Eight of the 10 patients with CAV were transplanted at 410 years of age (4 of whom had persistent DSA, 3 transient DSA and 1 nonDSA). One patient was transplanted in the 1- to 5-year age range (persistent DSA), and only 1 patient with persistent DSA and CAV was transplanted in infancy. On subsequent multivariate analysis, the presence of DSA (p ¼ 0.038) and age at transplant between 1 and 10 years (p ¼ 0.0096) were found to be independent predictors of development of CAV.

Graft rejection Table 3

Outcomes According to HLA Antibody Status No Persistent Transient DSA Non-DSA antibodies DSA (n ¼ 25) (n ¼ 18) (n ¼ 41) (n ¼ 24)

Death/ 9 (36%) re-transplant Rejection 11 (44%) CAV 6 (24%)

1 (5%)

1 (2%)

3 (13%)

3 (17%) 3 (17%)

5 (12%) 1 (2%)

1 (4%) 0 (0%)

CAV, cardiac allograft vasculopathy.

Twenty patients (19%) had an episode of rejection. In 14 patients there was documented cellular rejection, although features of antibody-mediated rejection were also present in 3 of these patients. In 6 patients, the type of rejection on histology was not documented or biopsy not performed and slides were not available for retrospective review. Rejection was more frequent in the DSA group (33%) when compared with the non-DSA (12%) or no antibodies (5%) groups (p ¼ 0.007). There was no difference in time from transplant

Irving et al.

Persistent and Transient DSA

5

Figure 2 Survival by antibody grouping compared with the group with no antibodies and patients overall (noting the fact that development of DSA or non-DSA is a time-dependent variable).

to first rejection or number of rejection episodes between groups. Of the 14 patients with DSA who rejected, 9 had DSA detected before the rejection episode with a median time of 2.8 (0.2 to 7.2) years from DSA detection to rejection. In the remaining 5 patients DSA were detected at the time of rejection. Four of the 14 patients (29%) had both Class I and Class II antibodies, 3 of 14 (21%) had Class I antibodies alone and 7 of 14 (50%) had Class II antibodies alone. The most common DSA were to HLA-DQ (64%), followed by HLA-DR (57%), HLA-A (43%) and HLAB (36%).

Figure 3

On univariate analysis, there was an association between the development of post-transplant HLA antibodies (p ¼ 0.0431) and rejection, but this association was only significant for the overall DSA group (p ¼ 0.0017) and the persistent DSA group (p ¼ 0.0075). The only other factor of significance on univariate analysis was an association between an increased number of HLA-B mismatches (p ¼ 0.0282) and an increase in acute rejection (Table 6). Multivariate analysis showed that only the development of DSA was an independent predictor of rejection (p ¼ 0.011).

Comparison of survival by DSA group (overall, transient or persistent).

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

6 Table 4

Cases of Graft Loss

Pt

Antibody group

Antibody specificity (for DSA group)

Cause of graft loss

Age at transplant (years) 10.9 0.9

2.0 2.2

Death

Acute rejection Acute rejection and CAV Acute rejection

2.5

2.5

Re-Tx

CAV

11.3

4.7

Re-Tx

Chronic graft failure Acute rejection Non-cardiac (post-partum hemorrhage) CAV Acute rejection Acute rejection Chronic graft failure PTLD Acute rejection Pneumonia related to underlying desmin myopathy

12.3

5.8

1.1 14.1

6.6 7.2

11.3 11.5 3.4 3.7

7.2 9.8 2.7 9.9

14.2 1.6 7.6

5.0 1.1 14.2

Outcome

1 2

DSA (persistent) DSA (persistent)

3

DSA (persistent)

4

DSA (persistent)

5

DSA (persistent)

DQ5, DQ6 A25, A32, B27, Cw2, DR51, DQ7 A26, A25, A66, A34, DQ3, DR53 DR103, DQ4, DQ5, DQ6 DQ8

6 7

DSA (persistent) DSA (persistent)

DR4 DQ6, DQ7

Re-Tx Death

8 9 10 11

DSA (persistent) DSA (persistent) DSA (transient) Non-DSA (persistent) No antibodies No antibodies No antibodies

DQ5 A2, B44, DR12, DQ7 DR52

Death Death Death Re-Tx

12 13 14

Re-Tx Death

Death Death Death

Time to graft loss (years)

CAV, cardiac allograft vasculopathy; PTLD, post-transplant lymphoproliferative disease; Re-Tx, re-transplant.

Discussion In this study we have shown that 40% of our patient cohort had DSA detected at some point during the study period. These antibodies were transient in almost half of the patients. The transient presence of DSA did not appear to be related to poor graft outcomes during our follow-up period. In contrast, persistent DSA were associated with an increased incidence of CAV, rejection and graft loss. Although there are increasing data on the deleterious effect of DSA on cardiac graft function, this is the first study to address long-term longitudinal effects of de novo HLA antibody production, either transient or persistent, on outcomes in childhood cardiac transplant recipients. Freitas et al presented data demonstrating the effects of HLA-DQ DSA in kidney transplantation and showed a significantly higher prevalence of acute rejection and worse allograft survival in patients with HLA-DQ antibodies alone.15 Adult patients with persistent DSA have been found to be at increased risk of CAV, acute rejection and graft loss.5,16–19 Early reports of post-transplant HLA antibody development relied on less sensitive methods of antibody detection compared with the current Luminex methodology.4,9,16,20 Using Luminex, Stastny et al21 found a significant decrease in graft survival and an increase in rejection episodes and CAV in the presence of DSA. However, they only tested for DSA on one occasion, not serially. They did show, however, that non-donor-specific HLA antibodies appear to have no effect on graft outcome, a finding that we have

corroborated in this work. In a more recent study, serial retrospective HLA antibody testing using Luminex technology was performed5 and showed that the majority of DSA were directed against HLA-DR and -DQ antigens, with persistent DSA more likely to be directed at HLA-DQ, as also shown in our cohort. This is in keeping with recent renal transplant data, which showed a negative impact on renal allograft outcomes of persistent de novo HLA-DQ antibodies.15,22 In contrast, the majority of non-donorspecific HLA antibodies in our study were directed against Class I A or B antigens. Earlier studies showed an association between both HLA Class I and Class II antibodies and early graft rejection.3,23,24 More recent studies have addressed the use of MFI levels for HLA antibodies and found some utility in using these measurements in acute allograft vasculopathy.25 There is increasing evidence, however, that Class II DSA are more strongly associated with long-term chronic rejection, particularly in the form of CAV.6,26 Interestingly, Smith et al5 showed no association between HLA antibodies (DSA or non-DSA) and the development of CAV. In our study, 60% of the patients who developed CAV had persistent DSA. We also found an association with older age at transplant, a factor that has been described previously and is in keeping with data from the ISHLT registry.27 When rejection was analyzed, post-transplant HLA antibody development (particularly DSA and persistent DSA) were significant variables. Limitations to our study, taking into account the retrospective study design, include the fact that, although

Irving et al. Table 5

Persistent and Transient DSA

7

Patients With Cardiac Allograft Vasculopathy

Antibody group (DSA specificities)

Age at Tx (years)

Time from Tx to CAV (years)

1

DSA persistent (DQ4, DQ5, DQ6, DR103)

11.3

4.2

2 3

DSA persistent (DQ8) DSA persistent (DQ5)

12.3 11.3

3.0 4.9

4

DSA persistent (DQ2)

2.7

9.9

5

12.3

9.3

0.9

2.2

7 8

DSA persistent (A29, DQ2, DR17) DSA persistent (A25, A32, B27, Cw2, DQ7, DR51) DSA transient (DQ2) DSA transient (B44)

15.1 9.3

6.9 3.2

9

DSA transient (B15)

14.2

1.8

10

Non-DSA persistent

11.1

6.0

Pt

6

Angiographic findings Moderate narrowing to proximal Cx, mild narrowing to proximal obtuse marginal Diffuse LCA stenosis Tight stenosis of first diagonal and right ventricular branch RCA, diffuse disease of proximal LAD and distal RCA RCA stenosis with 80% narrowing, 30% narrowing of main LCA, 90% narrowing of proximal LAD Tight stenosis Cx, 100% occlusion LAD Found on post-mortem—complete occlusion RCA, ectatic LCA 70–90% stenosis LAD Narrowed and calcified RCA, Calcification Cx and LAD with tight stenosis origin Cx Narrowing of first diagonal, moderate narrowing of midportion of LAD, diffusely small RCA, 50% stenosis obtuse marginal Narrowed LAD

Intervention for CAV

Status

None

Re-Tx

None Stented, LIMA

Re-Tx Died

Stented

Alive

Angioplasty

Alive

None

Died

Stented None

Alive Alive

None

Alive

None

Alive

Cx, circumflex; DSA, donor-specific antibodies; LAD, left anterior descending coronary artery; LCA, left coronary artery; LIMA, left internal mammary artery graft; Pt, patient; RCA, right coronary artery; Tx, transplant.

the maximum follow-up period was almost 16 years, the mean follow-up period was just under 10 years posttransplant, even though this is longer than in previous renal transplantation reports.16 We did exclude patients who died within 1 year of transplant as we aimed to focus our Table 6

analysis on those patients with medium- and long-term follow-up periods available; therefore, it is possible that patients who died early on may have had DSA that were not detected. As this was a retrospective study, it did not address the use of interventions (e.g., in the form of increased

Univariate Analysis of DSA and Other Variables on Development of Acute Rejection

Variable

Estimated hazard ratio

95% CI interval for hazard ratio

p

Development of post-transplant HLA antibodies Development of post-transplant DSA Development of post-transplant persistent DSA Development of post-transplant transient DSA Development of post-transplant non-DSA Number of HLA-A mismatches Number of HLA-B mismatches Number of HLA-Cw mismatches Number of HLA-DR mismatches Number of HLA-DQ mismatches Male donor Gender mismatch Ischemic time 44 hours ABO-incompatible transplant Age o1 year at transplant Age 1–10 years at transplant Age 410 years at transplant Size mismatch 42:1 Pre-transplant mechanical support

5.0263 4.4296 3.4300 1.2169 0.5084 1.1086 3.3843 2.2358 0.6460 0.7905 0.8827 0.9686 1.1380 1.1508 0.7451 0.7291 1.5700 1.0208 1.3499

0.6789–37.2138 1.6144–12.1542 1.4325–8.2127 0.4077–3.6321 0.1851–1.3961 0.4543–2.7054 0.9914–11.5533 0.8756–5.7089 0.2579–1.6180 0.2307–2.7079 0.3645–2.1378 0.4012–2.3381 0.4151–3.1201 0.2655–4.9880 0.1731–3.2068 0.3010–1.7660 0.6516–3.7830 0.3648–2.8560 0.4801–3.7959

0.0431 0.0017 0.0075 0.7314 0.1702 0.8223 0.0282 0.1090 0.3449 0.7023 0.7829 0.9437 0.8048 0.8544 0.6828 0.4850 0.3228 0.9690 0.5801

CI, confidence interval; DSA, donor-specific antibodies.

8

The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]

immunosuppression, or immune modulation using agents such as rituximab or eculizumab or antibody removal with plasmapheresis or immunoadsorption) in asymptomatic patients who were found to have persistent DSA. Very few routine endomyocardial biopsies were performed and therefore some early rejection episodes that were clinically insignificant may have been missed. We did not look at complement fixation in our study. In addition, it is not yet known whether patients with no antibodies or only non-DSA at presentation will go on to develop DSA in time and if development of DSA is an inevitable finding with increasing graft age. The non-DSA group did include patients who may or may not have had HLA antibodies before transplant; however, as the aim of our study was to examine the importance of DSA, this information was not analyzed. Even so, no patient in the non-DSA group ever had a response to donor antigens, and therefore these patients were characterized as negative for DSA. This study does, however, further support the fact that if Class II DSA, particularly DQ antibodies, are persistent, then additional management should be considered, as these patients are at higher risk of rejection, CAV and eventual graft loss. One of the strengths of our study is its longitudinal rather than cross-sectional design. The finding that transient DSA do not appear to have associations with poor graft outcomes in the time frame studied is important and highlights the need for repeated monitoring of antibody status in all patients after transplantation. Nevertheless, it would be useful to carry out prospective data collection to evaluate longer term outcomes and to consider a review or change of immunosuppressive regimens in patients who have persistent DSA. Information on antibody status may aid in tailoring of individual immunosuppression. Further work in patients who have undergone desensitization protocols to remove HLA antibodies pre-transplant would also be of interest.

Disclosure statement The authors have no conflicts of interest to disclose.

References 1. Xydas S, Yang JK, Burke EM, et al. Utility of post-transplant antiHLA antibody measurements in pediatric cardiac transplant recipients. J Heart Lung Transplant 2005;1296:1289-96. 2. Tambur AR, Pamboukian SV, Costanzo MR, et al. The presence of HLA-directed antibodies after heart transplantation is associated with poor allograft outcome. Transplantation 2005;1025:1019-25. 3. McKenna RM, Takemoto SK, Terasaki PI. Anti-HLA antibodies after solid organ transplantation. Transplantation 2000;326:319-26. 4. Tambur AR, Bray RA, Takemoto SK, et al. Flow cytometric detection of HLA-specific antibodies as a predictor of heart allograft rejection. Transplantation 2000;1059:1055-9. 5. Smith JD, Banner NR, Hamour IM, et al. De novo donor HLA-specific antibodies after heart transplantation are an independent predictor of poor patient survival. Am J Transplant 2011;319:312-9. 6. Vasilescu ER, Ho EK, de la Torre L, et al. Anti-HLA antibodies in heart transplantation. Transpl Immunol 2004;183:177-83.

7. Mao Q, Terasaki PI, Cai J, et al. Extremely high association between appearance of HLA antibodies and failure of kidney grafts in a five year longitudinal study. Am J Transplant 2007;871:864-71. 8. Kim JJ, Balasubramanian R, Michaelides G, et al. The clinical spectrum of de novo donor-specific antibodies in pediatric renal transplant recipients. Am J Transplant 2014;2358:2350-80. 9. Di Filippo S, Girnita A, Webber SA, et al. Impact of ELISA-detected anti-HLA antibodies on pediatric cardiac allograft outcome. Hum Immunol 2005;518:513-8. 10. Tait BD, Hudson F, Cantwell L, et al. Luminex technology for HLA antibody detection in organ transplantation. Nephrology 2009;254: 247-254. 11. Irving C, Carter V, Parry G, et al. Donor-specific HLA antibodies in pediatric cardiac transplant recipients are associated with poor graft survival. Pediatr Transplant 2011;197:193-7. 12. Patel R, Terasaki P. Significance of the positive crossmatch test in kidney transplantation. N Engl J Med 1969;739:735-9. 13. Billingham ME, Cary NR, Hammond ME, et al. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: heart rejection study group. J Heart Transplant 1990;593:587-93. 14. Stewart S, Winters GL, Fishbein MC, et al. Revision of the 1990 working formulation for the standardization of nomenclature in the diagnosis of heart rejection. J Heart Lung Transplant 2005;1720:1710-20. 15. Freitas MCS, Rebellato LM, Ozawa M, et al. The role of immunoglobulin-G subclasses and C1q in de novo HLA-DQ donorspecific antibody kidney transplantation outcomes. Transplantation 2013;1119:1113-9. 16. Christiaans MH, Nieman F, van Hooff JP, et al. Detection of HLA class I and II antibodies by ELISA and complement-dependent cytotoxicity before and after transplantation. Transplantation 2000;927: 917-27. 17. Sutherland SM, Chen G, Sequeira FA, et al. Complement-fixing donorspecific antibodies identified by a novel C1q assay are associated with allograft loss. Pediatr Transplant 2012;17:12-7. 18. Jordan SC. Complement fixing donor-specific antibodies and allograft loss. Pediatr Transplant 2012;16:1-3. 19. Kaczmarek I, Deutsch MA, Kauke T, et al. Donor-specific HLA alloantibodies: long-term impact on cardiac allograft vasculopathy and mortality after heart transplant. Exp Clin Transplant 2008;235: 229-35. 20. El-Awar N, Lee J, Terasaki PI. HLA antibody identification with single antigen beads compared to conventional methods. Hum Immunol 2005;997:989-97. 21. Stastny P, Lavingia B, Fixler DE, et al. Antibodies against donor human leukocyte antigens and the outcome of cardiac allografts in adults and children. Transplantation 2007;745:738-45. 22. DeVos JM, Gaber AO, Knight RJ, et al. Donor-specific HLA-DQ antibodies may contribute to poor graft outcome after renal transplantation. Kidney Int 2012;82:598-604. 23. Raess M, Frohlich G, Roos M, et al. Donor-specific anti-HLA antibodies detected by Luminex: predictive for short-term but not long-term survival after heart transplantation. Transpl Int 2013;1107: 1097-1107. 24. Zeevi A, Lunz J, Feingold B, et al. Persistent strong anti-HLA antibody at high titer is complement binding and associated with increased risk of antibody-mediated rejection in heart transplant recipients. J Heart Lung Transplant 2013;32:98-105. 25. Fine NM, Daly RC, Shankar N, et al. The role of donor-specific antibodies in acute cardiac allograft dysfunction in the absence of cellular rejection. Transplantation 2014;238:229-38. 26. Frank R, Molina MR, Wald JW, et al. Correlation of circulating donorspecific anti-HLA antibodies and presence of C4d in endomyocardial biopsy with heart allograft outcomes: a single-center, retrospective study. J Heart Lung Transplant 2013;417:410-7. 27. Dipchand AI, Kirk R, Edwards LB, et al. The registry of the International Society for Heart and Lung Transplantation: sixteenth official pediatric heart transplantation report—2013; Focus theme: Age. J Heart Lung Transplant 2013;988:979-88.