http://www.jhltonline.org
ORIGINAL CLINICAL SCIENCE
Late antibody-mediated rejection after heart transplantation: Mortality, graft function, and fulminant cardiac allograft vasculopathy Guillaume Coutance, MD,a Salima Ouldamar, MD,a Philippe Rouvier, MD,b Samir Saheb, MD,c Caroline Suberbielle, MD, PhD,d Nicolas Bréchot, MD,e Sarah Hariri, MD,f Guillaume Lebreton, MD,a Pascal Leprince, MD, PhD,a and Shaida Varnous, MDa From the aDepartment of Cardiac and Thoracic Surgery, Cardiology Institute; bDepartment of Pathology, and; c Department of Hemo-biotherapies, Pitié Salpêtrière Hospital, University of Paris VI; dLaboratory of Immunology and Histocompatibility—CIB-HOG, Saint Louis Hospital, and; eDepartment of Medical Reanimation; and the fDepartment of Cardiac Anesthesia and Reanimation, Cardiology Institute, Pitié Salpêtrière Hospital, University of Paris VI, Paris, France.
KEYWORDS: antibody-mediated rejection; donor-specific anti-HLA antibodies; heart transplantation; survival; cardiac allograft vasculopathy
BACKGROUND: Late antibody-mediated rejection (AMR) after heart transplantation is suspected to be associated with a poor short-term prognosis. METHODS: A retrospective single-center observational study was performed. Late AMR was defined as AMR occurring at least 1 year after heart transplantation. The study included all consecutive patients with proven and treated late acute AMR at the authorsʼ institution between November 2006 and February 2013. The aim was to analyze the prognosis after late AMR, including mortality, recurrence of AMR, left ventricular ejection fraction, and cardiac allograft vasculopathy (CAV). Selected endomyocardial biopsy specimens obtained before AMR were also blindly reviewed to identify early histologic signs of AMR. RESULTS: The study included 20 patients treated for late AMR. Despite aggressive immunosuppressive therapies (100% of patients received intravenous methylprednisolone, 90% received intravenous immunoglobulin [IVIg],85% received plasmapheresis, 45% received rituximab), the prognosis remained poor. Survival after late AMR was 80% at 1 month, 60% at 3 months, and 50% at 1 year. All early deaths (o3 months, n ¼ 8) were directly attributable to graft dysfunction or to complication of the intense immunosuppressive regimen. Among survivors at 3 months (n ¼ 12), histologic persistence or recurrence of AMR, persistent left ventricular dysfunction, and fulminant CAV were common (33%, 33%, and 17% of patients). Microvascular inflammation was detected in at least 1 biopsy specimen obtained before AMR in 13 patients (65%). CONCLUSIONS: Prognosis after late AMR is poor despite aggressive immunosuppressive therapies. Fulminant CAV is a common condition in these patients. Microvascular inflammation is frequent in endomyocardial biopsy specimens before manifestation of symptomatic AMR. J Heart Lung Transplant ]]]];]:]]]–]]] r 2015 International Society for Heart and Lung Transplantation. All rights reserved.
Reprint requests: Shaida Varnous, MD, Service de Chirurgie Thoracique et Cardio-Vasculaire, Institut de Cardiologie, Groupe hospitalier Pitié-Salpêtrière, 47-83, boulevard de lʼHôpital 75651 Paris Cedex 13. Telephone: þ33 1/42-16-56-90. Fax: þ33 1/42-16-55-76. E-mail address: [email protected]
Heart transplantation is the gold standard treatment for patients with advanced heart failure.1 Although improvements in immunosuppressive therapies have significantly reduced the frequency of acute cellular rejection (ACR), the
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.2015.03.002
2
The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]
incidence of antibody-mediated rejection (AMR) remains almost unchanged.2 AMR is now recognized as a major cause for graft loss after solid organ transplant.3 As preformed donor-specific anti-HLA antibodies (DSA), de novo DSA contribute to poor outcomes by increasing the risk of allograft rejections, cardiac allograft vasculopathy (CAV), and death.4,5 Late AMR may be associated with a worse prognosis compared with early AMR.6,7 However, little is known concerning late AMR after heart transplantation. In a previous cohort of 15 patients, Hodges et al4 showed that late AMR was associated with poor short-term prognosis despite aggressive immunosuppressive therapies. We performed a single-center retrospective study to analyze the characteristics of patients with late AMR, the treatment they received, and their prognosis.
pre-transplant lymphocytotoxicity assay and/or enzyme-linked immunosorbent assay techniques revealed the presence of antiHLA antibodies (n ¼ 2). Pre-transplant complement-dependent cytotoxic crossmatch was negative for these patients. Since 2008, post-transplant DSA have been screened using Luminex screening kits 1, 3, 6, and 12 months after transplantation; yearly thereafter; and in the presence of allograft dysfunction. DSA were classified as “certain de novo DSA” if pre-transplant Luminex did not revealed pre-formed DSA, as “probable de novo DSA” if pre-transplant low sensibility screening tests (enzymelinked immunosorbent assay, lymphocytotoxicity assay) were negative for class I and class II anti-HLA antibodies, and as “mixed DSA” if pre-formed and de novo DSA were identified. “Cumulative mean fluorescence intensity (MFI)” represents the sum of MFI of all DSA at the diagnosis of AMR for each patient.
Endomyocardial biopsy
Methods Patients We performed a retrospective single-center observational study. All consecutive patients with proven and treated late acute AMR at our institution between November 2006 and February 2013 were included. The institutional review board approved the protocol, and informed consent was obtained.
Objectives The primary objective was to analyze mortality after late AMR. A secondary objective was to describe allograft injury after AMR in patients alive at 3 months, including recurrence or persistence of histologic features of AMR, evolution of left ventricular (LV) systolic function, and CAV. We also analyzed biopsy specimens obtained before AMR to detect early signs of AMR.
Diagnosis of late AMR The diagnosis of AMR was based on the presence of at least 2 of the 3 following criteria: cardiac allograft dysfunction, evidence of development of DSA, and histologic signs of AMR according to International Society for Heart and Lung Transplantation (ISHLT) guidelines.3 Late AMR was defined as AMR diagnosed 41 year after transplantation. Cardiac allograft dysfunction was defined as the presence of at least 1 of the following criteria: signs of heart failure (symptoms or signs of low cardiac output and/or pulmonary or systemic congestion), requirement for inotropic drugs and/or LV dysfunction on transthoracic echocardiography using Simpson technique (left ventricular ejection fraction [LVEF] o0.45 or 425% decrease of LVEF from baseline).
Immunology De novo DSA was defined as an anti-HLA antibody directed against donor-specific antigen not present at the time of transplantation. Since 2008, the detection of pre-transplant and posttransplant anti-HLA antibodies has been based on Luminex mixed class I and II Antibody Screening kits (One Lambda, Canoga Park, CA). Patients with a positive screen are characterized for HLA class I and/or class II antibody specificity using LABScreen Single Antigen beads. For patients who received their transplants before 2008 (n ¼ 16), retrospective Luminex screening was performed if
Routine endomyocardial biopsies (EMBs) are performed 2 years after transplantation at our center (every 6 months for years 2–5, then every year). “Reference EMB” refers to the EMB at the time of AMR diagnosis. Additional biopsies were performed for clinical indications. Selected EMB specimens obtained before AMR (“preAMR EMB”—1, 3, 6, and 12 months after transplantation; then every year; and in case of rejection or allograft dysfunction) and all EMB specimens after AMR (“post-AMR EMB”) were retrospectively and blindly reviewed to detect early microvascular inflammation and recurrence and/or persistence of AMR after treatment. Positive and negative controls were included with each run. Biopsy specimens were processed and examined according to current standards. Standard serial sections were cut from formalinfixed paraffin-embedded EMB specimens and stained with hematoxylin-eosin-saffron for diagnosis. Immunofluorescence for C4d was performed on all reference and post-AMR EMB specimens (frozen section; C4d monoclonal 1/100; Quidel Corporation; Polyclonal rabbit anti-mouse fluorescein isothiocyanate; Dako). Only capillary staining for C4d was assessed. EMB specimens were assessed for ACR according to the revised 2004 ISHLT criteria8 and for histologic and immunologic features of AMR according to the 2011 ISHLT consensus conference.3 Microvascular inflammation was defined as intravascular macrophages and lymphocytes in 410% of myocardial capillaries.
CAV At our institution, routine coronary angiograms are performed 1 year after heart transplantation and then every 2 years or in case of unexplained LV dysfunction. Staging of CAV was performed using the recommended ISHLT nomenclature.9 Fulminant CAV was defined as a lesion 470% within 1 year of a benign angiogram (o30% previously) according to ISHLT experts.9
Immunosuppression protocol after heart transplantation Immunosuppression therapy after heart transplantation was based on an induction therapy for all patients, using mostly anti-thymocyte globulin or basiliximab for patients with low immunologic risk and high risk of infectious complications (particularly patients with ventricular assist devices). Prophylactic immunosuppressive
Coutance et al.
Late AMR After Heart Transplantation
treatment includes calcineurin inhibitors (cyclosporine or tacrolimus), mycophenolate mofetil (MMF), and corticosteroids with dosage adjustment as recommended by ISHLT guidelines.10 Everolimus could be introduced for a clinical indication, but we never interrupted calcineurin inhibitor basal immunosuppression (low-dose calcineurin inhibitor or MMF discontinuation).
Statistical analyses Results are presented as mean ⫾ SD for continuous variables with normal distribution, as median and interquartile range for continuous variables with non-normal distribution, and as number and percentage for categorical variables. Cumulative survival curves for the time-to-event analyses were constructed according to the Kaplan-Meier method. Comparison of LVEF before and after AMR was done using Studentʼs paired t-test. Univariate analyses were conducted to determine if different factors could predict the risk of early death. Statistical significance was determined if the null hypothesis could be rejected at the p o 0.05 level.
3
Treatment of antibody-mediated rejection All patients received intravenous methylprednisolone infusion. There were 18 patients treated with IVIg (2 g/kg during 4 days), 17 were treated with plasmapheresis (2–11 sessions), and 9 were treated with at least 1 infusion of rituximab. The combination of methylprednisolone, IVIg, and plasmapheresis was given to 8 patients; 8 other patients received rituximab in addition to that combination. Thymoglobulin was given to 3 patients, mainly for the treatment of a concomitant significant ACR. The prophylactic immunosuppressive regimen was increased. Among patients alive 3 months after the diagnosis, the dosage of calcineurin inhibitor was increased for 2 patients, and 5 were switched from cyclosporine to tacrolimus. Mean MMF daily dose before AMR was 1,320 ⫾ 1,100 mg compared with 1,860 ⫾ 840 mg 3 months after the diagnosis (paired t-test, p o 0.01).
Clinical outcomes
Results Characteristics of included patients This study included 20 patients treated for late acute AMR at our institution between November 2006 and February 2013. Clinical characteristics of patients are presented in Table 1. Most patients were men (n ¼ 16; 80%). The mean age at diagnosis of AMR was 45 ⫾ 16 years. The mean and median times between transplantation and late AMR were 4.9 ⫾ 2.8 years and 4.3 years (range 3.3–6.4). AMR episodes concerned a first heart transplantation for all but 1 patient.
Diagnosis of late AMR All patients except 1 presented with signs of allograft dysfunction (dyspnea on minimal exertion, n ¼ 18; clinical signs of heart failure, n ¼ 12; cardiogenic shock, n ¼ 6). Transthoracic echocardiography revealed significant LV systolic dysfunction compared with baseline in 14 patients (Table 1). DSA were present in all tested patients (n ¼ 19); 7 patients were known to undergo DSA testing before AMR. Median cumulative MFI at diagnosis was 13,766 (4,577–26,000). DSA were de novo DSA for most patients (certain de novo DSA, n ¼ 4; probable de novo DSA, n ¼ 13). In 2 patients (“mixed DSA”), class I DSA were identified in pre-transplant serum samples, but patients developed very high titers of class II DSA at the time of AMR (cumulative class II DSA MFI ¼ 14,376 and 9,589). All adequate biopsy specimens revealed histologic or immunologic features of AMR except for 1 patient (Patient 14, ACR0, pAMR0). Histologic signs of ACR were present in 10 biopsy specimens (7 grade 1R, 2 grade 2R, and 1 grade 3R). Diagnosis was based on the association of allograft dysfunction, identification of DSA, and histologic signs of AMR in 16 patients (80%).
Survival During the follow-up period, 10 patients (50%) died. Survival after late AMR was 80% at 1 month, 60% at 3 months, and 50% at 1 year (Figure 1). Median survival was 0.96 years. All patients admitted to the intensive care unit for cardiogenic shock secondary to AMR died during their hospitalization (n ¼ 6). All early deaths (o3 months, n ¼ 8) were directly attributable to graft dysfunction (cardiogenic shock, multiorgan failure) or to complication of the intense immunosuppressive regimen. Severe infectious complications occurred in 5 patients during hospitalization and were directly responsible for 2 deaths (severe Pneumocystis infection, septic shock) and 3 clinical degradations precipitating death. Neither histologic grade (pAMR 0, 1, or 2) nor levels of DSA at diagnosis (cumulative, maximal, or number 43,000) were significant markers of the risk of in-hospital mortality.
Allograft function in patients alive at 3 months (n ¼ 12) At least 2 endomyocardial biopsies were performed after AMR in all patients (range 2–22, median 3). Histologic analyses revealed persistent signs of AMR in all follow-up EMB specimens for 1 patient—pAMR 1 (h) in 4 consecutive EMB specimens—without allograft dysfunction. Asymptomatic histologic AMR relapse was diagnosed in 3 patients (25%) 1 to 7 months after the initial diagnosis (pAMR 1). One of these patients (Patient 9) developed persistent histologic AMR—pAMR1 (h) in 10 consecutive EMB specimens—with progressive decline in LV function. Among patients alive at 3 months, 6 had LV systolic dysfunction at the time of the diagnosis. During follow-up, 3 patients (25%) had persistent LV dysfunction, whereas LV function normalized in the other 3 patients. However, 1 of these patients experienced recurrent and then persistent AMR leading to progressive decline in LV function. LV
4
Table 1
Characteristics of Included Patients
Patient
CAV worsening LVEF normalization Post-AMR EMB normalization AMR relapse (histologic) AMR relapse (clinical)
2
3
4
5
6
7
8
9
10
M 53 3
F 55 6.5
M 35 4.2
M 58 1.1
M 54 14.2
M 56 4.4
M 67 3.8
F 27 5.3
M 25 8.5
M 26.4 7.3
CAV 0 No No TAC/MMF/CS
CAV 0 No No CYA/CS
CAV 0 No Yes CYA/MMF
CAV 0 No Yes CYA/MMF/CS
CAV 0 No No CYA/AZA/CS
CAV 1 Yes (1) No CYA/EVL/CS
CAV 1 No No TAC/MMF/ CS
CAV 0 No No CYA/ MMF/ CS
CAV 0 No Yes TAC/MMF/CS
CAV 1 Yes (1) No CYA/MMF/CS
Yes 20
Yes 40
Yes 40
Yes 30
Yes 40
Yes 25
Yes 0
Yes 20
Yes 30
Yes 0
638 638 0
15,931 6,740 2
8,142 8,142 1
6,746 6,746 1
26,769 13,076 2
4,659 4,139 1
312 312 0
17,365 7,383 3
19,222 14,678 1
36,422 8,479 5
1R1A pAMR 2
0R pAMR 2
2R3A pAMR 1 (h)
0R pAMR 1 (h)
0R 1R1A pAMR 1 (i) pAMR 2
2R3A NA pAMR 1 (i) NA
1R1B pAMR 1 (h)
0R pAMR 1 (i)
Yes 5 Yes No IVMP
Yes 2 No No IVMP
Yes 5 Yes No IVMP-TMG
Yes 7 Yes No IVMP
Yes 10 Yes No IVMP
Yes 10 Yes Yes (2) IVMP
Yes 5 Yes No IVMP
No — Yes No IVMP
No — Yes No IVMP
Yes 4 Yes Yes IVMP
Death 90 Pneumocystis
Death 3 Acute rejection NA No NA — —
Alive 319 —
Death 24 Acute rejection NA No NA — —
Death 33 MOF
Alive 1437 —
Death 352 Stroke
Alive 781 —
Alive 1740 —
NA No Yes — —
Yes (CAV2) NA No — Yes Yes Yes No No No
No ? ? — —
No Yes Yes Yes þþþ Yes
Death 227 Sudden death (MI) Yes (CAV3) — Yes No No
No No No — —
? Yes No No No
The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]
Baseline characteristics Gender Age at diagnosis of AMR (years) Time from transplant - AMR (years) CAV stage ACR 4 1R before AMR Known de novo DSA Immunosuppression at time of AMR Diagnosis of AMR Cardiac allograft dysfunction Clinical deterioration Absolute change in EF Immunology Cumulative MFI of DSA MFI of DSA maximal DSA with MFI 43,000 (n) Histology and IHC ACR AMR Treatment Plasmapheresis Number of cycles IVIg Rituximab Others Outcomes Survival Survival (days) Cause of death
1
CAV worsening LVEF normalization Post-AMR EMB normalization AMR relapse (histologic) AMR relapse (clinical)
12
13
14
15
16
17
18
19
20
M 19 1.8
F 30 5
M 53 3.4
M 56 6.3
M 44 5.8
F 22 3.3
M 59 3.2
M 64 2.7
M 29 6.5
M 60 3.4
CAV 0 Yes (1) Yes CYA/MMF/CS
CAV 0 Yes (1) Yes CYA/MMF/CS
CAV 0 No No CYA/MMF/CS
CAV 2 No No CYA/EVL/CS
CAV 0 No Yes CYA/MMF/ CS
CAV 0 No No CYA/MMF/ CS
CAV 1 No No CYA/MMF/ CS
CAV 0 No Yes CYA/MMF/CS
CAV 1 No No CYA/MMF/CS
CAV 0 No Yes CYA/EVL/CS
Yes 30
Yes 0
Yes 40
Yes 40
No 20
Yes 35
Yes 0
No 0—LVH
Yes 35
Yes 0
19,792 10,867 3
49,395 12,864 5
NA NA NA
40,911 8,937 5
13,766 8,960 1
32,570 12,202 6
10,914 5,353 2
1,465 1,465 0
4,577 2,333 0
1,039 1,039 0
3R3B pAMR 2
0R pAMR 2
1R1A pAMR1 (i)
0R pAMR 0
0R 0R pAMR 1 (h) pAMR 2
1R1B 1R1A pAMR 1 (h) pAMR 1(h)
0R pAMR 1 (h)
1R1B pAMR 1 (h)
Yes 5 Yes Yes IVMP-TMG
Yes 5 Yes No IVMP
No — No No IVMP
Yes 5 Yes No IVMP
Yes 5 Yes Yes (2) IVMP
Yes 11 Yes Yes (3) TMG
Yes 5 Yes Yes IVMP
Yes 5 Yes No IVMP
Yes 6 Yes Yes IVMP
Yes 5 Yes Yes IVMP
Death 50 Acute rejection NA No No — —
Alive 771 —
Death 4 Acute rejection — No — — —
Death 17 Acute rejection NA No Yes — —
Alive 461 —
Alive 858 —
Alive 437 —
Alive 327 —
Alive 486 —
No — No No No
No — Yes No No
Death 46 Acute rejection NA No No — —
No — No Persistent No
Yes (CAV 2) NA No No Yes Yes No Yes No No
Late AMR After Heart Transplantation
Baseline characteristics Gender Age at diagnosis of AMR (years) Time from transplant - AMR (years) CAV stage ACR 4 1R before AMR Known de novo DSA Immunosuppression at time of AMR Diagnosis of AMR Cardiac allograft dysfunction Clinical deterioration Absolute change in EF Immunology Cumulative MFI of DSA MFI of DSA maximal DSA with MFI 43,000 (n) Histology and IHC ACR AMR Treatment Plasmapheresis Number of cycles IVIg Rituximab Others Outcomes Survival Survival (days) Cause of death
11
Coutance et al.
Patient
NA — Yes No No
ACR, acute cellular rejection; AMR, antibody-mediated rejection; AZA, azathioprine; CAV, cardiac allograft vasculopathy; CS, corticosteroids; CYA, cyclosporine A; DSA, donor-specific anti-HLA antibodies; EF, ejection fraction; EMB, endomyocardial biopsy; EVL, everolimus; F, female; IHC, immunohistochemistry; IVIg, intravenous immunoglobulin; IVMP, intravenous methylprednisolone; LVEF, left ventricular ejection fraction; LVH, left ventricular hypertrophy; M, male; MFI, mean fluorescence intensity; MI, myocardial infarction; MMF, mycophenolate mofetil; MOF, multiorgan failure; NA, not available; TAC, tacrolimus; TMG, thymoglobulin.
5
6
The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]
Figure 1 Survival after the diagnosis of late antibodymediated rejection (AMR).
function was significantly lower at the end of follow-up compared with baseline (baseline, LVEF ¼ 0.68 ⫾ 0.05; end of follow-up, LVEF ¼ 0.57 ⫾ 0.06, paired t-test, p o 0.01). There were 9 patients who had at least 1 coronary angiogram before and after late AMR. Significant worsening in CAV after AMR was found in 3 patients, including 2 cases of fulminant CAV. Patient 10 had stable CAV stage 1 for years, and Patient 15 did not present with any angiographic signs of CAV (angiograms performed 6 and 5 years after transplantation and 18 and 8 months before late AMR in Patient 10 and Patient 15, respectively). Post-AMR angiograms were performed 1 month after AMR and revealed fulminant CAV (Figure 2).
Pre-AMR endomyocardial biopsy specimens We blindly reviewed 158 “pre-AMR” EMB specimens in 19 patients (8.3 ⫾ 3 specimens per patient). We found histologic patterns of microvascular inflammation in at least 1 EMB specimens in 13 patients (68%). For 13 patients (26 EMB specimens), histologic analysis was performed in EMB specimens obtained r1 year before clinical AMR. Microvascular inflammation was present in at least 1 biopsy specimen for 7 patients (54%).
Discussion We described a retrospective cohort from a single heart transplantation institution of 20 patients treated for late AMR mostly secondary to de novo DSA. Despite an aggressive immunosuppressive regimen, the prognosis remained poor. Survival was 60% at 3 months and 50% at 1 year. Early deaths were attributable to graft dysfunction or to complication of the intensive immunosuppressive therapies. Among the patients alive at 3 months, signs of allograft injury were observed during follow-up: 33% had histologic signs of recurrent or persistent AMR, 33% had LV dysfunction, and 17% experienced fulminant CAV. This study outlines the severity of symptomatic late AMR despite aggressive immunosuppressive therapies and agrees with results previously published by Hodges et al.4 Compared with their cohort, we included similar patients
and found similar prognosis. Until 2011, ISHLT guidelines included the requisite criterion of allograft dysfunction in the definition of AMR in heart transplantation. Almost all patients treated in both cohorts were symptomatic or had LV systolic dysfunction (only 1 patient in each cohort without allograft dysfunction). Survival after 1, 6, and 12 months was 80%, 67%, and 53% for the cohort in the Hodges et al study4 and 80%, 60%, and 50% for our cohort. In both cohorts, early deaths were mostly due to allograft failure. However, in our cohort, 2 deaths resulted from infectious complications during immunosuppressive treatment (1 opportunistic infection and 1 septic shock). Similar complications were described after treatment of AMR in kidney transplant recipients.6 Immunosuppressive treatment of AMR is based on antibody removal and therapy to suppress further synthesis; it is challenging, and no consensus guidelines exist. As emphasized by a recent survey, the treatment of AMR is heterogeneous across hospitals.11 A combination of intravenous steroids, IVIg, and plasmapheresis is the most validated strategy12 and the most suggested initial therapy.11 However, the very high mortality after late AMR emphasizes the limits of current immunotherapies. Whether the treatment of AMR should be intensified or introduced earlier is still matter of debate. Strategies to prevent AMR are required. The adjustment of routine immunosuppression might be an option, but no studies have evaluated the efficacy of any routine immunosuppressive regimen in preventing AMR.3 However, it was suggested that tacrolimus/MMF was superior to cyclosporine/MMF in decreasing any treated rejection.13 This association might be a better regimen to decrease the development of AMR. Based on our local experience since 2008, we closely monitor DSA and discuss individually prophylactic treatment of AMR. In patients with pre-transplant DSA, monitoring of DSA is performed every month during the first year and then every year. For others, DSA are monitored at 1, 6, and 12 months after transplantation and then every year. We now routinely use rituximab to reinforce immunosuppressive treatment of AMR, using doses recommended for the treatment of lymphoproliferative disorders. Bortezomib, a proteasome inhibitor usually prescribed for refractory multiple myeloma, has shown promising results in small studies to treat persistent AMR.14,15 In their cohort, Hodges et al4 reported the use of bortezomib for 2 patients. However, our own experience is very limited concerning bortezomib or total lymphoid irradiation for the treatment of persistent AMR. Photopheresis also showed objective evidence for efficacy in the treatment of rejection with hemodynamic compromise or recurrent rejection.16 In 2011, the clinical definition for AMR (cardiac dysfunction) was no longer believed to be required because of recent publications demonstrating an increased development of CAV and inferior survival in patients with asymptomatic biopsy-proven AMR.3,17,18 However, it is still unknown whether treating asymptomatic AMR is beneficial or not.3 Our study emphasizes that AMR pathology grades do not appear to correlate to clinical severity of presentation and prognosis; this is consistent
Coutance et al.
Late AMR After Heart Transplantation
7
Figure 2 Fulminant cardiac allograft vasculopathy (CAV). Coronary angiograms of Patient 10. (A) Baseline coronary angiogram performed in May 2012 showing smooth ramus and left anterior descending artery and moderate infiltration of the left circumflex artery. CAV stage 1. (B–D) Coronary angiogram performed in December 2012, 1 month after the diagnosis of AMR, revealing significant stenosis (90%) of the left anterior descending artery with infiltration of diagonal arteries, occlusion of the ramus and the first marginal artery, and infiltration of the second marginal artery. The right coronary artery was smooth. CAV stage 3.
with the findings from the AMR consensus conference that cardiac dysfunction and/or DSA might mandate more aggressive AMR therapies.3 Blinded retrospective analysis of EMB specimens revealed that approximately 66% of our patients had evidence of microvascular inflammation before the diagnosis of AMR. This sign should alert the clinicians to monitor closely DSA, LVEF, and symptoms. Rapidly progressive CAV was described as humoral rejection in the form of vasculitis. In 1987, arterial vasculitis was described in EMB specimens of patients with graft failure and was associated with a poor prognosis.19 Later,
Berglin et al20 reported 5 cases of cardiac rejection with vasculitis and severe heart failure, which were treated effectively with plasmapheresis. Although no CAV worsening after late AMR was reported in the cohort of Hodges et al.,4 we described 2 cases of fulminant CAV diagnosed in the month after late AMR. In our cohort, the prevalence of fulminant CAV after AMR was 17% of patients alive at 3 months. However, we might have underestimated the degree of CAV progression because only 9 of the 20 patients had at least 1 coronary angiogram before and after late AMR.
8
The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]
Coronary-related deaths could represent a significant part of early mortality after late AMR. Hodges et al4 reported 2 early deaths resulting from CAV and myocardial infarction, respectively, despite the fact that no patient had angiographic progression that would explain the acute allograft dysfunction.4 Because of the severity of LV dysfunction secondary to late AMR and the significant prevalence of fulminant CAV after late AMR, we suggest that early coronary angiograms should be routinely performed in such patients. This study has several limitations. First, the observational retrospective design is subject to biases inherent in such studies. However, symptomatic late AMR after heart transplantation secondary to de novo DSA is a rare condition, and we believe that after exhaustive research we included all treated patients. Second, the detection of pre-transplantation HLA antibodies varied across time. The most sensitive technique based on single-antigen bead assay was used at our institution beginning in 2008. We classified DSA as certain de novo, probable de novo, and mixed DSA. Finally, only 1 specialist retrospectively analyzed EMB specimens obtained before AMR, and we did not review EMB specimens in a control group of patients with DSA and no AMR or patients without DSA and AMR. Interobserver agreement and statistical relevance (e.g., sensitivity, specificity) of microvascular inflammation to predict symptomatic AMR could not be evaluated. In conclusion, despite an aggressive immunosuppressive regimen, the prognosis after late AMR remains poor. Mortality was 40% at 3 months, which emphasizes the limits of current immunotherapies. Asymptomatic microvascular inflammation in EMB specimens before manifestation of symptomatic AMR is common. Among survivors at 3 months, histologic persistence or recurrence of AMR and LV dysfunction was frequent (33% of patients). Fulminant CAV occurred in 2 patients. Coronary-related deaths after AMR are common. Regular coronary angiograms may be considered in these patients.
Disclosure statement The authors acknowledge Catherine Aubailly for her valuable help in collecting data. None of the authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose.
References 1. McMurray JJ, Adamopoulos S, Anker SD, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012;33:1787-847.
2. Sureshkumar KK, Hussain SM, Marcus RJ, et al. Proteasome inhibition with bortezomib: an effective therapy for severe antibody mediated rejection after renal transplantation. Clin Nephrol 2012;77:246-53. 3. Kobashigawa J, Crespo-Leiro MG, Ensminger SM, et al. Report from a consensus conference on antibody-mediated rejection in heart transplantation. J Heart Lung Transplant 2011;30:252-69. 4. Hodges AM, Lyster H, McDermott A, et al. Late antibody-mediated rejection after heart transplantation following the development of de novo donor-specific human leukocyte antigen antibody. Transplantation 2012;93:650-6. 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;11:312-9. 6. Dörje C, Midtvedt K, Holdaas H, et al. Early versus late acute antibody-mediated rejection in renal transplant recipients. Transplantation 2013;96:79-84. 7. Kfoury AG, Snow GL, Budge D, et al. A longitudinal study of the course of asymptomatic antibody-mediated rejection in heart transplantation. J Heart Lung Transplant 2012;31:46-51. 8. 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;24:1710-20. 9. Mehra MR, Crespo-Leiro MG, Dipchand A, et al. International Society for Heart and Lung Transplantation working formulation of a standardized nomenclature for cardiac allograft vasculopathy—2010. J Heart Lung Transplant 2010;29:717-27. 10. Costanzo MR, Dipchand A, Starling R, et al. The International Society of Heart and Lung Transplantation Guidelines for the care of heart transplant recipients. J Heart Lung Transplant 2010;29:914-56. 11. Chih S, Tinckam KJ, Ross HJ. A survey of current practice for antibodymediated rejection in heart transplantation. Am J Transplant 2013; 13:1069-74. 12. Gubensek J, Buturovic-Ponikvar J, Kandus A, et al. Plasma exchange and intravenous immunoglobulin in the treatment of antibody-mediated rejection after kidney transplantation: a single-center historic cohort study. Transplant Proc 2013;45:1524-7. 13. Kobashigawa JA, Miller LW, Russell SD, et al. Tacrolimus with mycophenolate mofetil (MMF) or sirolimus vs. cyclosporine with MMF in cardiac transplant patients: 1-year report. Am J Transplant 2006;6:1377-86. 14. Eckman PM, Thorsgard M, Maurer D, Kim Y, Alloway RR, Woodle ES. Bortezomib for refractory antibody-mediated cardiac allograft rejection. Clin Transpl 2009:475-8. 15. Idica A, Kaneku H, Everly MJ, et al. Elimination of post-transplant donor-specific HLA antibodies with bortezomib. Clin Transpl 2008: 229-39. 16. Kirklin JK, Brown RN, Huang ST, et al. Rejection with hemodynamic compromise: objective evidence for efficacy of photopheresis. J Heart Lung Transplant 2006;25:283-8. 17. Kfoury AG, Hammond MEH, Snow GL, et al. Cardiovascular mortality among heart transplant recipients with asymptomatic antibody-mediated or stable mixed cellular and antibody-mediated rejection. J Heart Lung Transplant 2009;28:781-4. 18. Wu GW, Kobashigawa JA, Fishbein MC, et al. Asymptomatic antibody-mediated rejection after heart transplantation predicts poor outcomes. J Heart Lung Transplant 2009;28:417-22. 19. Herskowitz A, Soule LM, Ueda K, et al. Arteriolar vasculitis on endomyocardial biopsy: a histologic predictor of poor outcome in cyclosporine-treated heart transplant recipients. J Heart Transplant 1987;6:127-36. 20. Berglin E, Kjellström C, Mantovani V, Stelin G, Svalander C, Wiklund L. Plasmapheresis as a rescue therapy to resolve cardiac rejection with vasculitis and severe heart failure. A report of five cases. Transpl Int 1995;8:382-7.
ORIGINAL CLINICAL SCIENCE
Late antibody-mediated rejection after heart transplantation: Mortality, graft function, and fulminant cardiac allograft vasculopathy Guillaume Coutance, MD,a Salima Ouldamar, MD,a Philippe Rouvier, MD,b Samir Saheb, MD,c Caroline Suberbielle, MD, PhD,d Nicolas Bréchot, MD,e Sarah Hariri, MD,f Guillaume Lebreton, MD,a Pascal Leprince, MD, PhD,a and Shaida Varnous, MDa From the aDepartment of Cardiac and Thoracic Surgery, Cardiology Institute; bDepartment of Pathology, and; c Department of Hemo-biotherapies, Pitié Salpêtrière Hospital, University of Paris VI; dLaboratory of Immunology and Histocompatibility—CIB-HOG, Saint Louis Hospital, and; eDepartment of Medical Reanimation; and the fDepartment of Cardiac Anesthesia and Reanimation, Cardiology Institute, Pitié Salpêtrière Hospital, University of Paris VI, Paris, France.
KEYWORDS: antibody-mediated rejection; donor-specific anti-HLA antibodies; heart transplantation; survival; cardiac allograft vasculopathy
BACKGROUND: Late antibody-mediated rejection (AMR) after heart transplantation is suspected to be associated with a poor short-term prognosis. METHODS: A retrospective single-center observational study was performed. Late AMR was defined as AMR occurring at least 1 year after heart transplantation. The study included all consecutive patients with proven and treated late acute AMR at the authorsʼ institution between November 2006 and February 2013. The aim was to analyze the prognosis after late AMR, including mortality, recurrence of AMR, left ventricular ejection fraction, and cardiac allograft vasculopathy (CAV). Selected endomyocardial biopsy specimens obtained before AMR were also blindly reviewed to identify early histologic signs of AMR. RESULTS: The study included 20 patients treated for late AMR. Despite aggressive immunosuppressive therapies (100% of patients received intravenous methylprednisolone, 90% received intravenous immunoglobulin [IVIg],85% received plasmapheresis, 45% received rituximab), the prognosis remained poor. Survival after late AMR was 80% at 1 month, 60% at 3 months, and 50% at 1 year. All early deaths (o3 months, n ¼ 8) were directly attributable to graft dysfunction or to complication of the intense immunosuppressive regimen. Among survivors at 3 months (n ¼ 12), histologic persistence or recurrence of AMR, persistent left ventricular dysfunction, and fulminant CAV were common (33%, 33%, and 17% of patients). Microvascular inflammation was detected in at least 1 biopsy specimen obtained before AMR in 13 patients (65%). CONCLUSIONS: Prognosis after late AMR is poor despite aggressive immunosuppressive therapies. Fulminant CAV is a common condition in these patients. Microvascular inflammation is frequent in endomyocardial biopsy specimens before manifestation of symptomatic AMR. J Heart Lung Transplant ]]]];]:]]]–]]] r 2015 International Society for Heart and Lung Transplantation. All rights reserved.
Reprint requests: Shaida Varnous, MD, Service de Chirurgie Thoracique et Cardio-Vasculaire, Institut de Cardiologie, Groupe hospitalier Pitié-Salpêtrière, 47-83, boulevard de lʼHôpital 75651 Paris Cedex 13. Telephone: þ33 1/42-16-56-90. Fax: þ33 1/42-16-55-76. E-mail address: [email protected]
Heart transplantation is the gold standard treatment for patients with advanced heart failure.1 Although improvements in immunosuppressive therapies have significantly reduced the frequency of acute cellular rejection (ACR), the
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.2015.03.002
2
The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]
incidence of antibody-mediated rejection (AMR) remains almost unchanged.2 AMR is now recognized as a major cause for graft loss after solid organ transplant.3 As preformed donor-specific anti-HLA antibodies (DSA), de novo DSA contribute to poor outcomes by increasing the risk of allograft rejections, cardiac allograft vasculopathy (CAV), and death.4,5 Late AMR may be associated with a worse prognosis compared with early AMR.6,7 However, little is known concerning late AMR after heart transplantation. In a previous cohort of 15 patients, Hodges et al4 showed that late AMR was associated with poor short-term prognosis despite aggressive immunosuppressive therapies. We performed a single-center retrospective study to analyze the characteristics of patients with late AMR, the treatment they received, and their prognosis.
pre-transplant lymphocytotoxicity assay and/or enzyme-linked immunosorbent assay techniques revealed the presence of antiHLA antibodies (n ¼ 2). Pre-transplant complement-dependent cytotoxic crossmatch was negative for these patients. Since 2008, post-transplant DSA have been screened using Luminex screening kits 1, 3, 6, and 12 months after transplantation; yearly thereafter; and in the presence of allograft dysfunction. DSA were classified as “certain de novo DSA” if pre-transplant Luminex did not revealed pre-formed DSA, as “probable de novo DSA” if pre-transplant low sensibility screening tests (enzymelinked immunosorbent assay, lymphocytotoxicity assay) were negative for class I and class II anti-HLA antibodies, and as “mixed DSA” if pre-formed and de novo DSA were identified. “Cumulative mean fluorescence intensity (MFI)” represents the sum of MFI of all DSA at the diagnosis of AMR for each patient.
Endomyocardial biopsy
Methods Patients We performed a retrospective single-center observational study. All consecutive patients with proven and treated late acute AMR at our institution between November 2006 and February 2013 were included. The institutional review board approved the protocol, and informed consent was obtained.
Objectives The primary objective was to analyze mortality after late AMR. A secondary objective was to describe allograft injury after AMR in patients alive at 3 months, including recurrence or persistence of histologic features of AMR, evolution of left ventricular (LV) systolic function, and CAV. We also analyzed biopsy specimens obtained before AMR to detect early signs of AMR.
Diagnosis of late AMR The diagnosis of AMR was based on the presence of at least 2 of the 3 following criteria: cardiac allograft dysfunction, evidence of development of DSA, and histologic signs of AMR according to International Society for Heart and Lung Transplantation (ISHLT) guidelines.3 Late AMR was defined as AMR diagnosed 41 year after transplantation. Cardiac allograft dysfunction was defined as the presence of at least 1 of the following criteria: signs of heart failure (symptoms or signs of low cardiac output and/or pulmonary or systemic congestion), requirement for inotropic drugs and/or LV dysfunction on transthoracic echocardiography using Simpson technique (left ventricular ejection fraction [LVEF] o0.45 or 425% decrease of LVEF from baseline).
Immunology De novo DSA was defined as an anti-HLA antibody directed against donor-specific antigen not present at the time of transplantation. Since 2008, the detection of pre-transplant and posttransplant anti-HLA antibodies has been based on Luminex mixed class I and II Antibody Screening kits (One Lambda, Canoga Park, CA). Patients with a positive screen are characterized for HLA class I and/or class II antibody specificity using LABScreen Single Antigen beads. For patients who received their transplants before 2008 (n ¼ 16), retrospective Luminex screening was performed if
Routine endomyocardial biopsies (EMBs) are performed 2 years after transplantation at our center (every 6 months for years 2–5, then every year). “Reference EMB” refers to the EMB at the time of AMR diagnosis. Additional biopsies were performed for clinical indications. Selected EMB specimens obtained before AMR (“preAMR EMB”—1, 3, 6, and 12 months after transplantation; then every year; and in case of rejection or allograft dysfunction) and all EMB specimens after AMR (“post-AMR EMB”) were retrospectively and blindly reviewed to detect early microvascular inflammation and recurrence and/or persistence of AMR after treatment. Positive and negative controls were included with each run. Biopsy specimens were processed and examined according to current standards. Standard serial sections were cut from formalinfixed paraffin-embedded EMB specimens and stained with hematoxylin-eosin-saffron for diagnosis. Immunofluorescence for C4d was performed on all reference and post-AMR EMB specimens (frozen section; C4d monoclonal 1/100; Quidel Corporation; Polyclonal rabbit anti-mouse fluorescein isothiocyanate; Dako). Only capillary staining for C4d was assessed. EMB specimens were assessed for ACR according to the revised 2004 ISHLT criteria8 and for histologic and immunologic features of AMR according to the 2011 ISHLT consensus conference.3 Microvascular inflammation was defined as intravascular macrophages and lymphocytes in 410% of myocardial capillaries.
CAV At our institution, routine coronary angiograms are performed 1 year after heart transplantation and then every 2 years or in case of unexplained LV dysfunction. Staging of CAV was performed using the recommended ISHLT nomenclature.9 Fulminant CAV was defined as a lesion 470% within 1 year of a benign angiogram (o30% previously) according to ISHLT experts.9
Immunosuppression protocol after heart transplantation Immunosuppression therapy after heart transplantation was based on an induction therapy for all patients, using mostly anti-thymocyte globulin or basiliximab for patients with low immunologic risk and high risk of infectious complications (particularly patients with ventricular assist devices). Prophylactic immunosuppressive
Coutance et al.
Late AMR After Heart Transplantation
treatment includes calcineurin inhibitors (cyclosporine or tacrolimus), mycophenolate mofetil (MMF), and corticosteroids with dosage adjustment as recommended by ISHLT guidelines.10 Everolimus could be introduced for a clinical indication, but we never interrupted calcineurin inhibitor basal immunosuppression (low-dose calcineurin inhibitor or MMF discontinuation).
Statistical analyses Results are presented as mean ⫾ SD for continuous variables with normal distribution, as median and interquartile range for continuous variables with non-normal distribution, and as number and percentage for categorical variables. Cumulative survival curves for the time-to-event analyses were constructed according to the Kaplan-Meier method. Comparison of LVEF before and after AMR was done using Studentʼs paired t-test. Univariate analyses were conducted to determine if different factors could predict the risk of early death. Statistical significance was determined if the null hypothesis could be rejected at the p o 0.05 level.
3
Treatment of antibody-mediated rejection All patients received intravenous methylprednisolone infusion. There were 18 patients treated with IVIg (2 g/kg during 4 days), 17 were treated with plasmapheresis (2–11 sessions), and 9 were treated with at least 1 infusion of rituximab. The combination of methylprednisolone, IVIg, and plasmapheresis was given to 8 patients; 8 other patients received rituximab in addition to that combination. Thymoglobulin was given to 3 patients, mainly for the treatment of a concomitant significant ACR. The prophylactic immunosuppressive regimen was increased. Among patients alive 3 months after the diagnosis, the dosage of calcineurin inhibitor was increased for 2 patients, and 5 were switched from cyclosporine to tacrolimus. Mean MMF daily dose before AMR was 1,320 ⫾ 1,100 mg compared with 1,860 ⫾ 840 mg 3 months after the diagnosis (paired t-test, p o 0.01).
Clinical outcomes
Results Characteristics of included patients This study included 20 patients treated for late acute AMR at our institution between November 2006 and February 2013. Clinical characteristics of patients are presented in Table 1. Most patients were men (n ¼ 16; 80%). The mean age at diagnosis of AMR was 45 ⫾ 16 years. The mean and median times between transplantation and late AMR were 4.9 ⫾ 2.8 years and 4.3 years (range 3.3–6.4). AMR episodes concerned a first heart transplantation for all but 1 patient.
Diagnosis of late AMR All patients except 1 presented with signs of allograft dysfunction (dyspnea on minimal exertion, n ¼ 18; clinical signs of heart failure, n ¼ 12; cardiogenic shock, n ¼ 6). Transthoracic echocardiography revealed significant LV systolic dysfunction compared with baseline in 14 patients (Table 1). DSA were present in all tested patients (n ¼ 19); 7 patients were known to undergo DSA testing before AMR. Median cumulative MFI at diagnosis was 13,766 (4,577–26,000). DSA were de novo DSA for most patients (certain de novo DSA, n ¼ 4; probable de novo DSA, n ¼ 13). In 2 patients (“mixed DSA”), class I DSA were identified in pre-transplant serum samples, but patients developed very high titers of class II DSA at the time of AMR (cumulative class II DSA MFI ¼ 14,376 and 9,589). All adequate biopsy specimens revealed histologic or immunologic features of AMR except for 1 patient (Patient 14, ACR0, pAMR0). Histologic signs of ACR were present in 10 biopsy specimens (7 grade 1R, 2 grade 2R, and 1 grade 3R). Diagnosis was based on the association of allograft dysfunction, identification of DSA, and histologic signs of AMR in 16 patients (80%).
Survival During the follow-up period, 10 patients (50%) died. Survival after late AMR was 80% at 1 month, 60% at 3 months, and 50% at 1 year (Figure 1). Median survival was 0.96 years. All patients admitted to the intensive care unit for cardiogenic shock secondary to AMR died during their hospitalization (n ¼ 6). All early deaths (o3 months, n ¼ 8) were directly attributable to graft dysfunction (cardiogenic shock, multiorgan failure) or to complication of the intense immunosuppressive regimen. Severe infectious complications occurred in 5 patients during hospitalization and were directly responsible for 2 deaths (severe Pneumocystis infection, septic shock) and 3 clinical degradations precipitating death. Neither histologic grade (pAMR 0, 1, or 2) nor levels of DSA at diagnosis (cumulative, maximal, or number 43,000) were significant markers of the risk of in-hospital mortality.
Allograft function in patients alive at 3 months (n ¼ 12) At least 2 endomyocardial biopsies were performed after AMR in all patients (range 2–22, median 3). Histologic analyses revealed persistent signs of AMR in all follow-up EMB specimens for 1 patient—pAMR 1 (h) in 4 consecutive EMB specimens—without allograft dysfunction. Asymptomatic histologic AMR relapse was diagnosed in 3 patients (25%) 1 to 7 months after the initial diagnosis (pAMR 1). One of these patients (Patient 9) developed persistent histologic AMR—pAMR1 (h) in 10 consecutive EMB specimens—with progressive decline in LV function. Among patients alive at 3 months, 6 had LV systolic dysfunction at the time of the diagnosis. During follow-up, 3 patients (25%) had persistent LV dysfunction, whereas LV function normalized in the other 3 patients. However, 1 of these patients experienced recurrent and then persistent AMR leading to progressive decline in LV function. LV
4
Table 1
Characteristics of Included Patients
Patient
CAV worsening LVEF normalization Post-AMR EMB normalization AMR relapse (histologic) AMR relapse (clinical)
2
3
4
5
6
7
8
9
10
M 53 3
F 55 6.5
M 35 4.2
M 58 1.1
M 54 14.2
M 56 4.4
M 67 3.8
F 27 5.3
M 25 8.5
M 26.4 7.3
CAV 0 No No TAC/MMF/CS
CAV 0 No No CYA/CS
CAV 0 No Yes CYA/MMF
CAV 0 No Yes CYA/MMF/CS
CAV 0 No No CYA/AZA/CS
CAV 1 Yes (1) No CYA/EVL/CS
CAV 1 No No TAC/MMF/ CS
CAV 0 No No CYA/ MMF/ CS
CAV 0 No Yes TAC/MMF/CS
CAV 1 Yes (1) No CYA/MMF/CS
Yes 20
Yes 40
Yes 40
Yes 30
Yes 40
Yes 25
Yes 0
Yes 20
Yes 30
Yes 0
638 638 0
15,931 6,740 2
8,142 8,142 1
6,746 6,746 1
26,769 13,076 2
4,659 4,139 1
312 312 0
17,365 7,383 3
19,222 14,678 1
36,422 8,479 5
1R1A pAMR 2
0R pAMR 2
2R3A pAMR 1 (h)
0R pAMR 1 (h)
0R 1R1A pAMR 1 (i) pAMR 2
2R3A NA pAMR 1 (i) NA
1R1B pAMR 1 (h)
0R pAMR 1 (i)
Yes 5 Yes No IVMP
Yes 2 No No IVMP
Yes 5 Yes No IVMP-TMG
Yes 7 Yes No IVMP
Yes 10 Yes No IVMP
Yes 10 Yes Yes (2) IVMP
Yes 5 Yes No IVMP
No — Yes No IVMP
No — Yes No IVMP
Yes 4 Yes Yes IVMP
Death 90 Pneumocystis
Death 3 Acute rejection NA No NA — —
Alive 319 —
Death 24 Acute rejection NA No NA — —
Death 33 MOF
Alive 1437 —
Death 352 Stroke
Alive 781 —
Alive 1740 —
NA No Yes — —
Yes (CAV2) NA No — Yes Yes Yes No No No
No ? ? — —
No Yes Yes Yes þþþ Yes
Death 227 Sudden death (MI) Yes (CAV3) — Yes No No
No No No — —
? Yes No No No
The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]
Baseline characteristics Gender Age at diagnosis of AMR (years) Time from transplant - AMR (years) CAV stage ACR 4 1R before AMR Known de novo DSA Immunosuppression at time of AMR Diagnosis of AMR Cardiac allograft dysfunction Clinical deterioration Absolute change in EF Immunology Cumulative MFI of DSA MFI of DSA maximal DSA with MFI 43,000 (n) Histology and IHC ACR AMR Treatment Plasmapheresis Number of cycles IVIg Rituximab Others Outcomes Survival Survival (days) Cause of death
1
CAV worsening LVEF normalization Post-AMR EMB normalization AMR relapse (histologic) AMR relapse (clinical)
12
13
14
15
16
17
18
19
20
M 19 1.8
F 30 5
M 53 3.4
M 56 6.3
M 44 5.8
F 22 3.3
M 59 3.2
M 64 2.7
M 29 6.5
M 60 3.4
CAV 0 Yes (1) Yes CYA/MMF/CS
CAV 0 Yes (1) Yes CYA/MMF/CS
CAV 0 No No CYA/MMF/CS
CAV 2 No No CYA/EVL/CS
CAV 0 No Yes CYA/MMF/ CS
CAV 0 No No CYA/MMF/ CS
CAV 1 No No CYA/MMF/ CS
CAV 0 No Yes CYA/MMF/CS
CAV 1 No No CYA/MMF/CS
CAV 0 No Yes CYA/EVL/CS
Yes 30
Yes 0
Yes 40
Yes 40
No 20
Yes 35
Yes 0
No 0—LVH
Yes 35
Yes 0
19,792 10,867 3
49,395 12,864 5
NA NA NA
40,911 8,937 5
13,766 8,960 1
32,570 12,202 6
10,914 5,353 2
1,465 1,465 0
4,577 2,333 0
1,039 1,039 0
3R3B pAMR 2
0R pAMR 2
1R1A pAMR1 (i)
0R pAMR 0
0R 0R pAMR 1 (h) pAMR 2
1R1B 1R1A pAMR 1 (h) pAMR 1(h)
0R pAMR 1 (h)
1R1B pAMR 1 (h)
Yes 5 Yes Yes IVMP-TMG
Yes 5 Yes No IVMP
No — No No IVMP
Yes 5 Yes No IVMP
Yes 5 Yes Yes (2) IVMP
Yes 11 Yes Yes (3) TMG
Yes 5 Yes Yes IVMP
Yes 5 Yes No IVMP
Yes 6 Yes Yes IVMP
Yes 5 Yes Yes IVMP
Death 50 Acute rejection NA No No — —
Alive 771 —
Death 4 Acute rejection — No — — —
Death 17 Acute rejection NA No Yes — —
Alive 461 —
Alive 858 —
Alive 437 —
Alive 327 —
Alive 486 —
No — No No No
No — Yes No No
Death 46 Acute rejection NA No No — —
No — No Persistent No
Yes (CAV 2) NA No No Yes Yes No Yes No No
Late AMR After Heart Transplantation
Baseline characteristics Gender Age at diagnosis of AMR (years) Time from transplant - AMR (years) CAV stage ACR 4 1R before AMR Known de novo DSA Immunosuppression at time of AMR Diagnosis of AMR Cardiac allograft dysfunction Clinical deterioration Absolute change in EF Immunology Cumulative MFI of DSA MFI of DSA maximal DSA with MFI 43,000 (n) Histology and IHC ACR AMR Treatment Plasmapheresis Number of cycles IVIg Rituximab Others Outcomes Survival Survival (days) Cause of death
11
Coutance et al.
Patient
NA — Yes No No
ACR, acute cellular rejection; AMR, antibody-mediated rejection; AZA, azathioprine; CAV, cardiac allograft vasculopathy; CS, corticosteroids; CYA, cyclosporine A; DSA, donor-specific anti-HLA antibodies; EF, ejection fraction; EMB, endomyocardial biopsy; EVL, everolimus; F, female; IHC, immunohistochemistry; IVIg, intravenous immunoglobulin; IVMP, intravenous methylprednisolone; LVEF, left ventricular ejection fraction; LVH, left ventricular hypertrophy; M, male; MFI, mean fluorescence intensity; MI, myocardial infarction; MMF, mycophenolate mofetil; MOF, multiorgan failure; NA, not available; TAC, tacrolimus; TMG, thymoglobulin.
5
6
The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]
Figure 1 Survival after the diagnosis of late antibodymediated rejection (AMR).
function was significantly lower at the end of follow-up compared with baseline (baseline, LVEF ¼ 0.68 ⫾ 0.05; end of follow-up, LVEF ¼ 0.57 ⫾ 0.06, paired t-test, p o 0.01). There were 9 patients who had at least 1 coronary angiogram before and after late AMR. Significant worsening in CAV after AMR was found in 3 patients, including 2 cases of fulminant CAV. Patient 10 had stable CAV stage 1 for years, and Patient 15 did not present with any angiographic signs of CAV (angiograms performed 6 and 5 years after transplantation and 18 and 8 months before late AMR in Patient 10 and Patient 15, respectively). Post-AMR angiograms were performed 1 month after AMR and revealed fulminant CAV (Figure 2).
Pre-AMR endomyocardial biopsy specimens We blindly reviewed 158 “pre-AMR” EMB specimens in 19 patients (8.3 ⫾ 3 specimens per patient). We found histologic patterns of microvascular inflammation in at least 1 EMB specimens in 13 patients (68%). For 13 patients (26 EMB specimens), histologic analysis was performed in EMB specimens obtained r1 year before clinical AMR. Microvascular inflammation was present in at least 1 biopsy specimen for 7 patients (54%).
Discussion We described a retrospective cohort from a single heart transplantation institution of 20 patients treated for late AMR mostly secondary to de novo DSA. Despite an aggressive immunosuppressive regimen, the prognosis remained poor. Survival was 60% at 3 months and 50% at 1 year. Early deaths were attributable to graft dysfunction or to complication of the intensive immunosuppressive therapies. Among the patients alive at 3 months, signs of allograft injury were observed during follow-up: 33% had histologic signs of recurrent or persistent AMR, 33% had LV dysfunction, and 17% experienced fulminant CAV. This study outlines the severity of symptomatic late AMR despite aggressive immunosuppressive therapies and agrees with results previously published by Hodges et al.4 Compared with their cohort, we included similar patients
and found similar prognosis. Until 2011, ISHLT guidelines included the requisite criterion of allograft dysfunction in the definition of AMR in heart transplantation. Almost all patients treated in both cohorts were symptomatic or had LV systolic dysfunction (only 1 patient in each cohort without allograft dysfunction). Survival after 1, 6, and 12 months was 80%, 67%, and 53% for the cohort in the Hodges et al study4 and 80%, 60%, and 50% for our cohort. In both cohorts, early deaths were mostly due to allograft failure. However, in our cohort, 2 deaths resulted from infectious complications during immunosuppressive treatment (1 opportunistic infection and 1 septic shock). Similar complications were described after treatment of AMR in kidney transplant recipients.6 Immunosuppressive treatment of AMR is based on antibody removal and therapy to suppress further synthesis; it is challenging, and no consensus guidelines exist. As emphasized by a recent survey, the treatment of AMR is heterogeneous across hospitals.11 A combination of intravenous steroids, IVIg, and plasmapheresis is the most validated strategy12 and the most suggested initial therapy.11 However, the very high mortality after late AMR emphasizes the limits of current immunotherapies. Whether the treatment of AMR should be intensified or introduced earlier is still matter of debate. Strategies to prevent AMR are required. The adjustment of routine immunosuppression might be an option, but no studies have evaluated the efficacy of any routine immunosuppressive regimen in preventing AMR.3 However, it was suggested that tacrolimus/MMF was superior to cyclosporine/MMF in decreasing any treated rejection.13 This association might be a better regimen to decrease the development of AMR. Based on our local experience since 2008, we closely monitor DSA and discuss individually prophylactic treatment of AMR. In patients with pre-transplant DSA, monitoring of DSA is performed every month during the first year and then every year. For others, DSA are monitored at 1, 6, and 12 months after transplantation and then every year. We now routinely use rituximab to reinforce immunosuppressive treatment of AMR, using doses recommended for the treatment of lymphoproliferative disorders. Bortezomib, a proteasome inhibitor usually prescribed for refractory multiple myeloma, has shown promising results in small studies to treat persistent AMR.14,15 In their cohort, Hodges et al4 reported the use of bortezomib for 2 patients. However, our own experience is very limited concerning bortezomib or total lymphoid irradiation for the treatment of persistent AMR. Photopheresis also showed objective evidence for efficacy in the treatment of rejection with hemodynamic compromise or recurrent rejection.16 In 2011, the clinical definition for AMR (cardiac dysfunction) was no longer believed to be required because of recent publications demonstrating an increased development of CAV and inferior survival in patients with asymptomatic biopsy-proven AMR.3,17,18 However, it is still unknown whether treating asymptomatic AMR is beneficial or not.3 Our study emphasizes that AMR pathology grades do not appear to correlate to clinical severity of presentation and prognosis; this is consistent
Coutance et al.
Late AMR After Heart Transplantation
7
Figure 2 Fulminant cardiac allograft vasculopathy (CAV). Coronary angiograms of Patient 10. (A) Baseline coronary angiogram performed in May 2012 showing smooth ramus and left anterior descending artery and moderate infiltration of the left circumflex artery. CAV stage 1. (B–D) Coronary angiogram performed in December 2012, 1 month after the diagnosis of AMR, revealing significant stenosis (90%) of the left anterior descending artery with infiltration of diagonal arteries, occlusion of the ramus and the first marginal artery, and infiltration of the second marginal artery. The right coronary artery was smooth. CAV stage 3.
with the findings from the AMR consensus conference that cardiac dysfunction and/or DSA might mandate more aggressive AMR therapies.3 Blinded retrospective analysis of EMB specimens revealed that approximately 66% of our patients had evidence of microvascular inflammation before the diagnosis of AMR. This sign should alert the clinicians to monitor closely DSA, LVEF, and symptoms. Rapidly progressive CAV was described as humoral rejection in the form of vasculitis. In 1987, arterial vasculitis was described in EMB specimens of patients with graft failure and was associated with a poor prognosis.19 Later,
Berglin et al20 reported 5 cases of cardiac rejection with vasculitis and severe heart failure, which were treated effectively with plasmapheresis. Although no CAV worsening after late AMR was reported in the cohort of Hodges et al.,4 we described 2 cases of fulminant CAV diagnosed in the month after late AMR. In our cohort, the prevalence of fulminant CAV after AMR was 17% of patients alive at 3 months. However, we might have underestimated the degree of CAV progression because only 9 of the 20 patients had at least 1 coronary angiogram before and after late AMR.
8
The Journal of Heart and Lung Transplantation, Vol ], No ], Month ]]]]
Coronary-related deaths could represent a significant part of early mortality after late AMR. Hodges et al4 reported 2 early deaths resulting from CAV and myocardial infarction, respectively, despite the fact that no patient had angiographic progression that would explain the acute allograft dysfunction.4 Because of the severity of LV dysfunction secondary to late AMR and the significant prevalence of fulminant CAV after late AMR, we suggest that early coronary angiograms should be routinely performed in such patients. This study has several limitations. First, the observational retrospective design is subject to biases inherent in such studies. However, symptomatic late AMR after heart transplantation secondary to de novo DSA is a rare condition, and we believe that after exhaustive research we included all treated patients. Second, the detection of pre-transplantation HLA antibodies varied across time. The most sensitive technique based on single-antigen bead assay was used at our institution beginning in 2008. We classified DSA as certain de novo, probable de novo, and mixed DSA. Finally, only 1 specialist retrospectively analyzed EMB specimens obtained before AMR, and we did not review EMB specimens in a control group of patients with DSA and no AMR or patients without DSA and AMR. Interobserver agreement and statistical relevance (e.g., sensitivity, specificity) of microvascular inflammation to predict symptomatic AMR could not be evaluated. In conclusion, despite an aggressive immunosuppressive regimen, the prognosis after late AMR remains poor. Mortality was 40% at 3 months, which emphasizes the limits of current immunotherapies. Asymptomatic microvascular inflammation in EMB specimens before manifestation of symptomatic AMR is common. Among survivors at 3 months, histologic persistence or recurrence of AMR and LV dysfunction was frequent (33% of patients). Fulminant CAV occurred in 2 patients. Coronary-related deaths after AMR are common. Regular coronary angiograms may be considered in these patients.
Disclosure statement The authors acknowledge Catherine Aubailly for her valuable help in collecting data. None of the authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose.
References 1. McMurray JJ, Adamopoulos S, Anker SD, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012;33:1787-847.
2. Sureshkumar KK, Hussain SM, Marcus RJ, et al. Proteasome inhibition with bortezomib: an effective therapy for severe antibody mediated rejection after renal transplantation. Clin Nephrol 2012;77:246-53. 3. Kobashigawa J, Crespo-Leiro MG, Ensminger SM, et al. Report from a consensus conference on antibody-mediated rejection in heart transplantation. J Heart Lung Transplant 2011;30:252-69. 4. Hodges AM, Lyster H, McDermott A, et al. Late antibody-mediated rejection after heart transplantation following the development of de novo donor-specific human leukocyte antigen antibody. Transplantation 2012;93:650-6. 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;11:312-9. 6. Dörje C, Midtvedt K, Holdaas H, et al. Early versus late acute antibody-mediated rejection in renal transplant recipients. Transplantation 2013;96:79-84. 7. Kfoury AG, Snow GL, Budge D, et al. A longitudinal study of the course of asymptomatic antibody-mediated rejection in heart transplantation. J Heart Lung Transplant 2012;31:46-51. 8. 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;24:1710-20. 9. Mehra MR, Crespo-Leiro MG, Dipchand A, et al. International Society for Heart and Lung Transplantation working formulation of a standardized nomenclature for cardiac allograft vasculopathy—2010. J Heart Lung Transplant 2010;29:717-27. 10. Costanzo MR, Dipchand A, Starling R, et al. The International Society of Heart and Lung Transplantation Guidelines for the care of heart transplant recipients. J Heart Lung Transplant 2010;29:914-56. 11. Chih S, Tinckam KJ, Ross HJ. A survey of current practice for antibodymediated rejection in heart transplantation. Am J Transplant 2013; 13:1069-74. 12. Gubensek J, Buturovic-Ponikvar J, Kandus A, et al. Plasma exchange and intravenous immunoglobulin in the treatment of antibody-mediated rejection after kidney transplantation: a single-center historic cohort study. Transplant Proc 2013;45:1524-7. 13. Kobashigawa JA, Miller LW, Russell SD, et al. Tacrolimus with mycophenolate mofetil (MMF) or sirolimus vs. cyclosporine with MMF in cardiac transplant patients: 1-year report. Am J Transplant 2006;6:1377-86. 14. Eckman PM, Thorsgard M, Maurer D, Kim Y, Alloway RR, Woodle ES. Bortezomib for refractory antibody-mediated cardiac allograft rejection. Clin Transpl 2009:475-8. 15. Idica A, Kaneku H, Everly MJ, et al. Elimination of post-transplant donor-specific HLA antibodies with bortezomib. Clin Transpl 2008: 229-39. 16. Kirklin JK, Brown RN, Huang ST, et al. Rejection with hemodynamic compromise: objective evidence for efficacy of photopheresis. J Heart Lung Transplant 2006;25:283-8. 17. Kfoury AG, Hammond MEH, Snow GL, et al. Cardiovascular mortality among heart transplant recipients with asymptomatic antibody-mediated or stable mixed cellular and antibody-mediated rejection. J Heart Lung Transplant 2009;28:781-4. 18. Wu GW, Kobashigawa JA, Fishbein MC, et al. Asymptomatic antibody-mediated rejection after heart transplantation predicts poor outcomes. J Heart Lung Transplant 2009;28:417-22. 19. Herskowitz A, Soule LM, Ueda K, et al. Arteriolar vasculitis on endomyocardial biopsy: a histologic predictor of poor outcome in cyclosporine-treated heart transplant recipients. J Heart Transplant 1987;6:127-36. 20. Berglin E, Kjellström C, Mantovani V, Stelin G, Svalander C, Wiklund L. Plasmapheresis as a rescue therapy to resolve cardiac rejection with vasculitis and severe heart failure. A report of five cases. Transpl Int 1995;8:382-7.
