Antibody Mediated Rejection of the Cardiac Allograft (Treatment Strategies in Cardiac Transplantation) Part 1

Introduction

Antibody mediated rejection (AMR), also known as B-cell mediated rejection or humoral rejection, of the cardiac allograft was first clinically described in the late 1980′s (Herskowitz et al., 1987) followed shortly thereafter by pathologic evidence to support a unique rejection process apart from cellular mechanisms (Hammond et al., 1989). This is in contrast to the progression of knowledge regarding cellular rejection, or T-cell mediated rejection, which was readily described in the early 1960′s and is the target of most current maintenance immunosuppression agents. Unfortunately, AMR remains poorly understood due, in large measure, to its complicated presentation, pathophysiology, diagnosis, and treatment. The lack of clarity regarding AMR has been compounded by multiple small studies in varying populations with a multitude of treatment modalities and combinations. Additionally, several new agents have been recently utilized or hypothesized to be of utility, with varying success.

Given the complexity of this process, lack of standardization in diagnosis, and multiple proposed treatment options, several professional organizations have endeavored to come to a consensus on the subject of AMR in heart transplant recipients. Most recently in 2011, the International Society for Heart and Lung Transplantation (ISHLT) published their outcomes from a consensus conference regarding AMR in heart transplantation (Kobashigawa et al., 2011) as well as a breakout group working formulation regarding pathologic diagnosis of AMR in heart transplantation (Berry et al., 2011). While these two documents provide some direction for practitioners and transplant providers, many questions remained unanswered and the rapid evolution of novel therapies and strategies for treatment will likely change the field of AMR in the heart transplant population dramatically.

This topic will look to lay a foundational knowledge of the pathophysiology, epidemiology, and diagnosis of AMR. Additionally, traditional therapies are described and evaluated with a highlight on the controversies surrounding their use; finally, novel and experimental therapies along with their potential impact on prevention and treatment of AMR are described.

Definitions

Antibody mediated rejection can be characterized in several different ways. First, it can be qualified based upon the temporal relationship it has to transplantation. Hyperacute AMR is a well known, well described process by which a patient has previously been exposed to some antigen that a donor expresses, and upon transplantation a rapid, immediate antibody response occurs leading to graft dysfunction and most often graft loss within 24 hours. Treatment of hyperacute AMR rarely reverses the process to salvage the graft. Acute AMR occurs sometime after the 24 hour postoperative period, and is generally rapid in onset; treatment strategies may be moderately effective. Chronic or late AMR is a newly recognized, poorly understood process that usually occurs greater than one year following transplantation and is thought to be very slow in progression with poor response to therapy.

Additionally, AMR can be described as either occurring due to pre-sensitization or is the result of de novo antibody production. De novo AMR occurs when a recipient lacks donor specific antibodies (DSA) and has a negative cross-match at the time of transplant, but subsequently develops AMR at some point after transplantation. Alternatively, if a patient has been previously exposed to antigens that a donor expresses, they are said to be pre-sensitized and typically receive prophylactic or empiric treatment in the peri-operative period. If, however, antibodies reappear at some point in the post-transplant period a renewed AMR may occur.

Pathophysiology

The immune system can generally be divided in to two main arms: the T cell, "cellular", arm, and the B cell, "humoral", arm. While these systems are complex and largely integrated, they do originate independently. B cells begin in the bone marrow as progenitor B cells and through activation by encounters with antigens mature through pro B cell, pre B cell, immature, and finally mature B cells. Activated mature B cells are also known as plasma cells and are essentially antibody factories. Antibodies are specific to a single antigen, such as proteins expressed on the surface of a transplanted organ, that are created to attach and signal other parts of the immune system to attack the foreign substance. This immune activation by antibody signaling ultimately damages the allograft. Damage is thought to occur via complement cascade-mediated fixation and activation, which actively damages the foreign material and also acts as a biochemical "amplifier", signaling other parts of the innate and adaptive immune systems such as neutophils, pro-inflammatory molecules and cytokines for example, to relocate to the site of antibody adhesion and attack. One of the more unique aspects of the B cell arm of the immune system is that it retains memory. Once a person has been exposed to an antigen presenting cell (usually from a foreign physiologic source such as an organ or transfusion) and mounts an immune response, a memory B cell is created that, without active intervention, will always exist and will mount a more-rapid response to subsequent antigen presentation from the same source.

These antigens can be portions of viruses, bacteria, or fungus. Human cells also express antigens; the most commonly identified of which are human leukocyte antigens (HLA). While a person does not usually attack itself and therefore tolerates their own HLAs, this is not true for other human tissues that express various antigens and are introduced in to a patient such as in solid organ transplantation.

Subsequently, the risk factors for development of AMR include anything that exposes patients to other human products and therefore creates more potential memory cells to respond to a transplanted organ. These include pregnancies, blood and blood product transfusions, repeat transplantation, and, specific to heart transplantation, the widespread and growing use of extracorporeal and intracorporeal mechanical circulatory support devices such as left ventricular assist systems (LVAS), bi-ventricular assist devices, total artificial hearts, extracorporeal membranous oxygenators (ECMO), or intra-aortic counterpulsators (Reed et al., 2006).

AMR has recently been described as occurring across a spectrum, from completely asymptomatic circulating antibody to clinically overt organ rejection with hemodynamic compromise, graft loss, and decreased survival (Takemoto et al., 2004). Additionally, AMR has been described to contribute significantly to cardiac allograft vasculopathy (CAV), and often occurs in conjunction with acute cellular rejection as so-called mixed rejection (Montgomery et al., 2004).

Epidemiology

The true incidence of AMR has been difficult to define given the lack of standardization in diagnosis; however, it is generally accepted that AMR plays a much larger role in overall graft and patient survival than previously appreciated. The reported incidence of de novo AMR varies widely based upon the definitions used and at which point on the spectrum a study defines AMR. Epidemiologic studies in centers that perform protocolized endomyocardial biopsies have shown a wide variability in incidence of 3 – 51% (Michaels et al., 2003; Shahzad et al., 2011). Not surprisingly, those institutions that include circulating antibodies without evidence of graft dysfunction had a higher reported incidence of AMR.

Additionally, as the boundaries of transplantation have been expanded in recent years, the number of patients who present for transplantation highly pre-sensitized to other human antigens is on the rise. Based on a survey of the patients who experienced AMR at some point after transplant from 46 heart transplant centers, 35% (114/324) of patients were pre-sensitized prior to transplant, and of those 32% (37/114 ) were treated to attempt to reduce the amount of circulating antibodies prior to transplantation (Kobashigawa et al., 2011).

Diagnosis

Significant effort has been placed on standardizing the diagnosis of AMR of the cardiac allograft within the past 3 – 5 years. These efforts highlight that clinical factors, immunologic criteria, and pathologic criteria all play important roles. In 2004, a general staging of AMR was developed (Table 1), as were criteria for diagnosis of AMR in heart transplant recipients (Table 2). More recently, the ISHLT proposed a preliminary pathologic grading scheme similar to the 2004 guidelines (Table 3) with one major difference: the ISHLT workgroup recognized AMR as a diagnosis that can be made without evidence of circulating antibodies or clinical dysfunction.

Immunologic screening

Antibody screening tests have been clinically available for many years. These tests determine circulating antibody, but do not address very low level antibodies or antibodies that may be active but not in circulation. Prior to solid-phase antibody (SPA) testing, the presence of antibodies was determined utilizing cell-based assays. The mainstay of testing was complement-dependent cytotoxicity (CDC) assays which involve incubating patient serum with cells of known HLA types, rabbit sera as a source of complement, and finally cell dyes to determine the amount of cell death that has occurred. The HLAs tested cover a very wide spectrum of known HLAs, however not all HLAs are tested. Limitations of this test included its lack of sensitivity and specificity (Berry et al., 2011). Unfortunately, the differences in clinical impact of circulating donor specific antibodies (DSAs), anti-HLA antibodies, or non-HLA antibodies, comparatively, have not been fully elucidated.

	Circulating Antibody	C4d Deposition	Tissue Pathology	Graft Dysfunction
Stage I: Latent Humoral Response	Present
Stage II: Silent Humoral Rejection	Present	Present
Stage III: Subclinical Humoral Rejeciton	Present	Present	Present
Stage IV. Humoral Rejection	Present	Present	Present	Present

Table 1. General AMR staging

Evidence of graft dysfunction	Present
Histologic evidence of tissue injury	*Endothelial swelling or denudation
	*Macrophages in capillaries
	Neutrophils in capillaries
	Interstitial edema, congestion and/or hemorrhage
Immunopathologic evidence for antibody action	Ig G, M, and/or A
	C3d and/or C4d and/or C1q in capillaries
	Fibrin in vessels
Serologic evidence of anti-HLA or other anti-donor antibody at time of biopsy	Present

* required histologic findings

Table 2. 2004 diagnostic criteria of acute AMR in heart transplant recipients

Category	Description	Definition
pAMR 0	Negative for pathologic AMR	Both histologic and immunopathologic studies are negative
pAMR 1 (H+)	Histopathologic AMR alone	Histologic findings present and immunopathologic studies negative
pAMR 1 (I+)	Immunopathol ogic AMR alone	Histologic findings negative and immunopathologic findings positive
pAMR 2	Pathologic AMR	Both histologic and immunopathologic findings present
pAMR 3	Severe pathologic AMR	Histologic findings of interstitial hemorrhage, capillary fragmentation, mixed inflammatory infiltrates, endothelial cell pyknosis, and/or karyorrhexis and marked edema

Table 3. 2011 ISHLT criteria for pathologic AMR

Solid Phase Antibody detection

The recent advent of SPA detection has revolutionized immunologic screening. The so-called Luminex® (LABScreen, One Lambda Inc., Canoga Park, CA) single antigen bead (SAB) assay panel provides a comprehensive assessment of individualized IgG and IgM HLA antibodies present in the recipient using a multiplex platform (El-Awar et al., 2005). These beads are coated with fluorescein-tagged antigens, which fluoresce in the presence of the known HLA antibody. The degree of fluorescence, defined in units of mean equivalents of soluble fluorochrome or mean fluorescent intensity, is directly proportional to the circulating amount of the HLA antibody in question. This quantitation is critical when determining which antibodies to exclude from the potential donor pool, and during the depletion process of DSA in the post transplant period.

Despite this improved specificity, the positive predictive value (PPV) of the Luminex assay for AMR remains poor (45%); however, the negative predictive value for AMR is quite good (100%) in a recent analysis (Chin et al., 2011). In an effort to improve the PPV of the Luminex-SAB assay, the Immunogenetics Laboratory at Stanford University spiked an otherwise ordinary Luminex assay sample with purified human Complement-1q (C1q) and ran the sample. The results of the assay revealed a significant decrease in background antibodies, and focused the assay only on those HLA antibodies able to fix C1q. This addition improved the assay’s PPV, dramatically, to 100% (Chin et al., 2011). This technique is currently in its infancy, but may result in enhanced utility of the Luminex assay over the decade to come.

Pathophysiology

Endomyocardial biopsies (EMB) at many centers are routinely performed in addition to those performed for any patient who exhibits signs and symptoms of graft dysfunction. It has been recognized that findings seen on histology are unique from those seen with acute cellular rejection or CAV. Some consensus regarding the findings for AMR was established recently. Pathologic findings are almost exclusively found in the capillary beds; common findings in AMR include endothelial swelling or denudation, deposition of macrophages or neutrophils in capillaries, and interstitial edema, congestion, and potentially hemorrhage in severe cases. Immunopathologic findings include deposition of IgG, M, or A, and positive staining for byproducts of the complement cascade including Complement-3d (C3d), Complemented (C4d), or C1q in the capillaries. Sometimes fibrin may also be found in the vessel beds. Table 3 outlines the grading criteria for pathologic AMR staging.

Treatment

When discussing treatment options, there are two major divisions for which these therapies have been studied. The first is for the removal of circulating antibodies prior to transplantation, a process known as desensitization; the second is for treatment of AMR, whether it be a reactivation of a previously sensitized patient or de novo AMR. Desensitization may be performed to either remove circulating antibody in the weeks to months prior to a transplant in an effort to allow for a larger donor pool in highly sensitized patients, or to mitigate the risk of AMR in the early postoperative period when a patient is known to have mismatched antigens such as is the case with ABO incompatible transplantation or positive cross-matches at the time of transplantation. Treatment may be performed at any point in the spectrum of AMR, from treatment of asymptomatic circulating antibodies to the treatment of clinically significant graft dysfunction caused by antibody-mediated activation of the immune system, with the goals of halting current damage, reverse signs and symptoms of AMR, and long-term to prevent the development of CAV and improve allograft and patient survival. Figure 1 contains a proposed treatment algorithm.

Fig. 1. AMR treatment algorithm. CDC, complement dependent cytotoxicity. DSA, donor specific antibody. IVIg, intravenous immunoglobulins. rATG, rabbit anti-thymocyte globulin. TPE, plasmapheresis.

Plasmapheresis

Plasmapheresis, or plasma exchange, has been used clinically for a variety of autoimmune conditions since the early 1970′s and is generally considered a cornerstone for treatment of AMR. It is a process which physically removes circulating antibodies along with many other circulating proteins; generally 7 – 14 plasmapheresis sessions at varying intervals (from daily to every 3 – 4 days) are required for substantial removal of antibodies. Each session generally lasts 2 – 4 hours. It is an invasive procedure in which a large-bore central venous catheter must be placed and extracorporeal separation of blood occurs via either centrifuge or filtration, antibodies are removed and discarded, and finally blood is returned to the patient.