Humoral rejection describes the production of potentially damaging antibodies by recipient immune cells targeted against the allograft. Antibody-mediated rejection (AMR) in lung transplantation is a diagnostic challenge. Notwithstanding, AMR is increasingly being identified as a driver of both acute and chronic lung allograft dysfunction (CLAD). Human leukocyte antigen (HLA) molecules are the major, although not the only, transplant antigens, and donor-specific antibodies (DSA) are largely described as targeting HLA molecules. It remains unclear whether B cells or the more differentiated plasma cell is the main source for anti-HLA DSA; this becomes critically important when considering appropriately targeted therapies in the treatment of AMR.
Anti-HLA DSA may be present at the time of transplant (sensitized recipient) or develop de novo following lung transplantation, and may be specific to HLA-class I (HLA-A, HLA-B, HLA-C) and/or HLA Class II (HLA-DP, HLA-DQ, HLA-DR). Transitioning from the serological to the molecular determination of HLA alleles has demonstrated the highly polymorphic nature of these transplant antigens , and an appreciation of HLA epitopes [2▪▪].
Solid phase assays such as the Luminex platform provide qualitative and quantitative data on anti-HLA DSA. The mean fluorescence intensity (mfi) is used by the clinician as a surrogate for the magnitude of the antibody response, but is not an accurate measure of the circulating DSA. Likewise, measuring DSA in the blood may not represent the effect that the antibody exerts in the lung allograft. A low mfi may be falsely reassuring if the DSA is absorbed within the lung allograft attached to its cognate HLA ligand. High levels (mfi) of DSA are not necessarily of concern if they are not bound within the lung allograft, and they fail to amplify the immune response via activation of the complement system. The C1q assay is a modification of the standard Luminex assay that aims to distinguish complement-fixing (and, therefore, injurious) from noncomplement-fixing DSA. The clinical utility of the C1q assay has been demonstrated in renal transplantation , and similar studies are awaited in lung transplantation. DSA can also be alloreactive via complement-independent pathways. Natural killer cells and macrophages can be directly activated via the low-affinity fragment crystallizable (Fc) receptor and cause cell lysis in a process termed antibody-independent cell-mediated cytotoxicity.
Sensitizations to HLA remains a major impediment to the success of solid-organ transplantation with implications for both access to suitable organs and posttransplant outcomes. More than 1600 HLA class I and II alleles have been identified and there is clear evidence that sensitizing events such as blood transfusion, pregnancy, and prior transplantation can result in the formation of anti-HLA antibodies.
Technological advancements, and in particular the introduction of solid phase assays has highlighted the relative insensitivity of CDC assays and led to detection of greater numbers of HLA antibodies. In turn, a greater proportion of individuals being considered for lung transplantation are now being identified as sensitized, impacting both their access to transplantation, and potentially posttransplant outcomes [4▪,5▪▪,6]. The increased sensitivity of solid phase testing, however, has allowed the detection of very low levels of HLA antibodies, the impact of which is as yet undefined. There is currently no consensus regarding the mfi threshold at which HLA antibodies should be regarded as significant and potentially deleterious further complicating pretransplant and posttransplant management.
Pretransplant HLA antibodies have been associated with both acute and chronic effects on the lung allograft including refractory acute rejection , lymphocytic bronchiolitis , AMR [4▪], chronic allograft dysfunction [6,8,9▪], as well as worse overall survival [4▪,9▪]. A study examining the characteristics of pretransplant DSA showed that patients with complement fixing DSAs of higher mfi exhibited poorer survival at 1 year although this had no additional impact on incidence of acute rejection or bronchiolitis obliterans syndrome (BOS) [9▪].
Genetic variations in HLA antigens are known as alleles and, approximately, 1270 HLA-A, HLA-B and HLA-C and 340 HLA-DR, -DP and -DQ variants have been characterized . However, by comparing the detailed structure of the different alleles, it is now recognized that HLA antigens have specific areas called epitopes that are the actual targets of DSA. In turn, the DSA response to an epitope has been further analyzed by stereochemical modelling of key protein antigen-antibody amino acid combinations to describe critical sequences called eplets [2▪▪].
Eplets have been extensively explored and defined by Duquesnoy [10▪], Approximately, 270 HLA-A, HLA-B, and HLA-C eplets are described with 219 being ‘functional’, with a further 51 buried within the HLA antigen, unreachable by antibody, and, therefore, ‘nonfunctional’. Additionally, approximately, 147 HLA-DR, 81 HLA-DQ, 58 HLA-DP, and 67 major histocompatability complex class-I-related chain A eplets are known [11▪]. It is notable that many alleles will share eplets, and, therefore, those quite different alleles might all be alloimmunological targets of a single DSA that recognizes this single eplet. This detail explains why sensitization with a previous allograft or blood transfusion can induce anti-HLA antibodies [and a resultant high Panel Reactive Antibody (PRA) status] to many seemingly nonseen HLA alleles .
Duquesnoy [2▪▪] has created a computer algorithm to characterize the extent of epitope matching between a recipient and their donor blood product or allograft. It is downloadable from the http://www.HLAMatchmaker.net website free of charge. It is useable as a quantitative tool to determine the immunological load of a given HLA mismatch.
Matchmaker has been used to highlight the likelihood of success utilizing a donor for highly sensitized patients, or those in which decreasing the extent of potential sensitization may be of subsequent benefit to enable easier matching of later grafts or transfusions. Pediatric renal transplant recipients  and conditions requiring recurrent platelet transfusions are two such examples . Further to this, there is evidence in renal transplantation that tighter Class II epitope matching is associated with lower rates of de-novo DSA, with the likely correlate of improved long-term graft survival . Thresholds of tolerable epitope mismatches were also noted by these authors – 10 for HLA-DR and 17 for HLA-DQ. A further study has recently confirmed these findings and assumptions – noting on multivariate assessment an odds ratio of 2.84 for transplant glomerulopathy wherein more than 27 HLA-DR and HLA-DQ eplet mismatches were present (4.62 above 43 mismatches) [16▪]. As a continuous variable the odds ratio increased 25% for every 10 additional mismatches. In both studies, these findings provide significant information beyond that available by simply considering HLA antigen mismatching [15,16▪]. Although the concept of very detailed profiling of all donors to enable very sensitized individuals to find a unique match has been mooted [17▪▪], the cost, feasibility, and overall utility to the wider large waiting list pool likely outweighs this [18▪▪].
Our own experience in evaluating Matchmaker in lung transplantation also suggests a future role . Looking at HLA-A, HLA-B, and HLA-DR antigens we noted that the average number of donor–recipient mismatches was, approximately, 4.5/6, essentially clustered as 3, 4, or 5. In the same patient cohort, the number of HLA-A, HLA-B, and HLA-DR eplet mismatches laid out linearly from 10 to 45, already providing a more sensitive tool for analysis. In univariate analysis Class I eplet mismatches did not predict future CLAD, however, HLA-DR and HLA-DQ eplets did predict CLAD with an odds ratio of 2.1 wherein greater than 40 eplet mismatches were present. By contrast, the development of CLAD was not predicted by the degree of HLA mismatch.
Variations in the absolute level of antigenicity (i.e., strength of antibody binding to epitopes) and immunogenicity (i.e., strength of the resultant alloinflammatory downstream effect) will be important to explore in future studies, as these may reveal specific eplet mismatches to avoid or specific therapeutic targets for monoclonal antibodies. For an individual potential recipient it may prove feasible to calculate the eplet score against a particular donor, as additional immunologic information that may direct toward or away from that donor–recipient combination. The concept of specifically monitoring or targeting with augmented immunosuppression, patients with overall higher eplet mismatch immunological load should also be studied.
ANTIBODY-MEDIATED REJECTION AFTER LUNG TRANSPLANTATION
The challenge of defining AMR is illustrated by the International Society of Heart and Lung Transplant's (ISHLT) delayed recognition of its very existence. Not mentioned in the 1990 and 1996 iteration of the ISHLT grading system for pulmonary allograft rejection, and only referred to obliquely in the 2007 revision of the classification, it is only in 2012 that the Pathology Council of ISHLT listed the histologic and immunologic criteria of pulmonary AMR. The lung transplant communities approach to AMR has largely mirrored diagnostic paradigms established in renal transplantation; namely a reliance on the Banff criteria of presence of anti-HLA DSAs, C4d deposition signifying complement activation, characteristic histological features centered on the capillary endothelium, and allograft dysfunction .
The diagnosis of pulmonary AMR requires a multidisciplinary approach in which the bedside assessment integrates with the histological interpretation of the lung biopsy and the immunologic evaluation of the presence of DSA . Most lung transplant physicians would agree that the patient with an unexplained drop in lung function, anti-HLA DSA, positive C4d staining, and neutrophilic capillaritis has ‘definite’ AMR. Reflecting the limitations of the histology (and immunohistochemistry), the more common scenario is the patient with DSA who may or may not have graft dysfunction; two entities that could be defined, respectively, as either clinical or subclinical AMR. The pulmonary council of ISHLT has convened a working group on AMR with a view to reaching consensus on the diagnostic features of AMR. A working definition of AMR is essential to facilitate epidemiological studies, between center comparisons of patient outcomes and allow interventional studies. As we now recognize different phenotypes of CLAD that include BOS, restrictive allograft syndrome, and pleuroparenchymal fibroelastosis, it will be of interest to see whether any of these are predominantly driven by antibody-mediated immune activation.
Positive C4d staining, particularly around lung capillaries, is used as biomarker for activation of the classical pathway of complement system, which in itself is suggestive of binding of an anti-HLA DSA with its cognate ligand, nonself HLA within the lung allograft. C4d deposition can be detected by either immunofluorescence or immunohistochemistry (IHC). The diagnostic utility of complement C4d deposition has recently questioned because of the poor reproducibility of immunofluorescence and IHC staining. Roden et al.[22▪] analysed over 200 transbronchial biopsy samples showing low correlation between immunofluorescence and IHC staining, and poor agreement between pathologists with regard to positive IHC [agreement 46.6%, 95% confidence interval (CI) 40.4–53.0%, κ = 0.13] and positive immunofluorescence staining (agreement 81.4%, 95% CI 68.7–89.7%, κ = 0.18). Recognizing the challenges in interpreting C4d staining in the lung allograft, the pathology council of the ISHLT is creating a central repository of slides that demonstrate agreed histological and immunohistochemical features consistent with AMR. Other confounders to the interpretation of C4d staining  include: positive staining can also be seen in primary graft dysfunction and infection; C4d staining may represent activation of the complement system by the mannose–lectin pathway; and an inconsistent association with the presence of anti-HLA DSA . The immune system can also be activated by anti-HLA DSA via complement-independent pathways. Cytotoxic natural killer cells and macrophages can be activated by DSA via the Fc receptor; a scenario that represents immunopathology but by definition will be C4d negative. Finally, evidence that we should consider C4d staining as a minor, not major, criterion in the diagnosis of AMR comes from the 2013 Banff meeting on renal transplantation in which C4d-negative AMR is now recognized in the general classification of AMR [24▪▪].
Histologic features of antibody-mediated rejection
The histologic features of AMR are not specifically defined, although the Pathology Council of ISHLT have recently released a summary statement on AMR . Suggestive patterns of AMR include ‘neutrophilic capillaritis’ described as ‘a patchy or diffuse process composed of dense neutrophilic septal infiltrates associated with neutrophilic karyorrhectic debris and fibrin with or without platelet–fibrin thrombi in the microvasculature, alveolar hemorrhage, and flooding of neutrophils into adjacent airspaces’ . The pathologists did, however, note that many of the histologic features of AMR are nonspecific and may also be seen in acute cellular rejection, ischemic reperfusion injury, and infection. Histology alone does not cement the clinical diagnosis of AMR but rather is an adjunct to the clinical presentation of a patient with DSA.
There is no clear definition of the clinical features that characterize AMR and such features are often nonspecific and potentially confounded by coexisting diseases and processes. Symptoms potentially include dyspnea and exercise limitation with clinical signs of lung crepitations and evidence on spirometry and radiology of allograft dysfunction. Spirometric changes may include evidence of new restrictive physiology [26,27▪] and radiological demonstration of new diffuse infiltrates or focal fibrosis may also be evident [21,26,27▪].
AMR may present acutely with allograft dysfunction and be accompanied by evidence of supportive diagnostic features such as circulating DSA and neutrophilic infiltration and C4d staining on lung biopsy . It may also present in a more chronic fashion, with the development of chronic dyspnea and restrictive physiology on lung function testing  in the setting of circulating DSA without histological evidence to make a firm diagnosis. In our experience, it is this scenario that appears more commonly with AMR usually diagnosed on the basis of clinical suspicion in the setting of unexplained allograft dysfunction, evidence of circulating DSA and nonspecific histological features [27▪]
TREATMENT OF ANTIBODY-MEDIATED REJECTION
No randomized controlled trials are currently available in lung transplantation to guide therapeutic strategies. Evidence is predominantly confined to case reports and small series and as such, approaches are based primarily on emerging renal transplant data.
Elevated PRA levels and evidence of HLA antibodies pretransplant has been shown to lead to early and late morbidity and mortality. Prevention, through the avoidance of known DSA HLA targets at the time of transplant remains a primary strategy [5▪▪,27▪] but limits the sensitized potential recipient's access to transplantation [4▪]. The use of preoperative desensitization techniques to reduce DSA loads with plasmapheresis, antithymocyte globulin (ATG), intravenous immunoglobulin (IVIG), and mycophenolate has recently been reported with reasonable tolerability and reduction of early rejection episodes and equivalent posttransplant outcomes to unsensitized patients [29▪]. One strategy utilizing plasmapheresis, corticosteroids, bortezomib, and rituximab lead to only 50% of patients receiving transplants and no improvements in outcomes when compared with untreated sensitized patients [5▪▪].
The treatment of established AMR has included combinations of corticosteroids, IVIG, plasmapheresis, ATG, and rituximab and varying results have been reported. We have approached the treatment of AMR with mycophenolate, intravenous corticosteroids and IVIG, escalating to plasmapheresis and rituximab wherein improvement is not forthcoming or not sustained [26,27▪].
In renal transplantation, bortezomib-based AMR treatments have led to depletion of plasma cells producing DSA, reduction of DSA levels and improvements in histological changes and renal function . Other therapies include eculizumab a monoclonal antibody complement activation inhibitor  and extracorporeal photopheresis . There are no robust data in lung transplantation to currently support their routine use.
The success of AMR treatment strategies appears to be somewhat related to a decrease in target DSA mfi although this is not always true [27▪,28]. In the absence of randomized trial data there is limited evidence for the efficacy of any of these treatments for AMR in lung transplantation. It, therefore, remains difficult to make any clear conclusions as to their role in the treatment of AMR especially when taking into account their complexity, significant side-effects, and expense [21,33].
AMR influences both short-term and long-term outcomes following lung transplantation. The limitations of the diagnostic tools that define the condition impact upon the clinical bedside assessment of AMR, and highlights the need for consensus on how AMR is defined. Only then can we better understand the true impact of AMR and how best to treat it.
Financial support and sponsorship
Conflicts of interest
G.P.W. has received research support from One Lambda, Inc. For the remaining authors none were declared.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
1. Tait B, Susal C, Gebel H, et al. Consensus guidelines on the testing and clinical management issues associated with HLA and non-HLA antibodies in transplantation. Transplantation 2013; 95:19–47.
2▪▪. Duquesnoy R. HLA epitope based matching for transplantation. Transpl Immunol 2014; 31:77–81.
Excellent review and introduction to HLA eplets and how they may better define immune sensitization.
3. Loupy A, Lefaucheur C, Vernerey D, et al. Complement-binding anti-HLA antibodies and kidney-allograft survival. N Engl J Med 2013; 369:1215–1226.
4▪. Kim M, Townsend K, Wood I, et al. Impact of pretransplant anti-HLA antibodies on outcomes in lung transplant candidates. Am J Transpl 2014; 189:1234–1239.
Large single center study defining the relationship between pretransplant sensitization and time to transplant, as well as the posttransplant incidence of AMR.
5▪▪. Snyder L, Gray A, Reynolds J, et al. Antibody desensitization therapy in highly sensitized lung transplant recipients. Am J Transpl 2014; 14:849–856.
Interesting study demonstrating that while an aggressive pretransplant desensitization protocol had minimal impact on lowering levels of sensitization, satisfactory clinical outcomes could be achieved in the same cohort of patients.
6. Brugiere O, Suberbielle C, Thabut G, et al. Lung transplantation
in patients with pretransplantation donor-specific antibodies
detected by Luminex assay. Transplantation 2013; 95:761–765.
7. Girnita A, McCurry K, Iacona A, et al. HLA-specific antibodies are associated with high-grade and persistent-recurrent lung allograft acute rejection. J Heart Lung Transpl 2004; 23:1135–1141.
8. Girnita A, Duquesnoy R, Yousem S, et al. HLA-specific antibodies are risk factors for lymphocytic bronchiolitis and chronic lung allograft dysfunction
. Am J Transplant 2005; 5:131–138.
9▪. Smith J, Ibrahim M, Newell H, et al. Pretransplant donor HLA-specific antibodies: characteristics causing detrimental effects on survival after lung transplantation
. J Heart Lung Transplant 2014; 33:1074–1082.
(with [27▪]). One of a number of recent studies defining the relationship between pretransplant sensitization and subsequent clinical outcomes.
10▪. Duquesnoy R. Update of the HLA class I eplet database in the website based registry of antibody-defined HLA epitopes. Tissue Antigens 2014; 83:382–390.
(with [11▪]). Updated list of eplets with respect to both class I and class II HLA.
11▪. Duquesnoy R, Marrari M, Tambur A, et al. First report on the antibody verification of HLA-DR, HLA-DQ and HLA-DP epitopes recorded in the HLA epitope registry. Human Immunol 2014; 98:1097–1103.
(with [10▪]). Updated list of eplets with respect to both class I and class II HLA.
12. Tambur A, Rosati J, Roitberg S, et al. Epitope analysis of HLA-DQ antigens: what does the antibody see? Transplantation 2014; 98:157–166.
13. Valentini R, Nehlsen-Cannarella S, Gruber S, et al. Intravenous immunoglobulin, HLA allele typing and HLAMatchmaker facilitate successful transplantation in highly sensitized pediatric renal allograft recipients. Ped Transplant 2007; 11:77–81.
14. Pai S, Lo S, Lin Tsai S, et al. Epitope-based matching for HLA-alloimmunized platelet refractoriness in patients with hematologic diseases. Transfusion 2010; 50:2318–2327.
15. Wiebe C, Pochinco D, Blydt-Hansen T, et al. Class II HLA epitope matching. A strategy to minimize de novo donor-specificantibody development and improve outcomes. Am J Transpl 2013; 13:3114–3122.
16▪. Sapir-Pichhadze R, Tinkam K, Quach K, et al. HLA-DR and -DQ eplet mismatches and transplant glomerulopathy: a nested case–control study. Am J Transpl 2015; 15:137–148.
One of a number of recent studies defining the clinical applicability of defining immune sensitization in terms of eplet mismatches between donor and recipient.
17▪▪. Duquesnoy R, Kamoun M, Baxter-Lowe L, et al. Should HLA mismatch acceptability for sensitized candidates bre determined at the high-resolution rather than the antigen level? Am J Transplant 2015; 15:923–930.
(with [18▪▪]) Two opinion pieces discussing the pros and cons of integrating eplet scores into routine clinical practice.
18▪▪. Cecka J, Reed E, Zachary A. HLA high-resolution typing for sensitized patients: a solution in search of a problem? Am J Transplant 2015; 15:855–856.
(with [17▪▪]). Two opinion pieces discussing the pros and cons of integrating eplet scores into routine clinical practice.
19. Chin N, Westall G, Paraskeva M, et al. Challenges inherent to the diagnosis of antibody-mediated rejection
in lung transplantation
. Respirol Case Rep 2015; 3:36–39.
20. Takemoto S, Zeevi A, Feng S, et al. National conference to assess antibody-mediated rejection
in solid organ transplantation. Am J Transpl 2004; 4:1033–1041.
21. Westall G, Snell G. Antibody-mediated rejection
in lung transplantation
: fable, spin or fact? Transplantation 2014; 98:927–930.
22▪. Roden A, Maleszewski J, Yi E, et al. Reproducibility of complement C4d deposition by immunofluorescence and immunohistochemistry in lung allograft biopsies. J Heart Lung Transpl 2014; 33:1223–1232.
Study highlighting the limitations of assessing C4d staining in lung transplant recipients.
23. Roberts J, Barrios R, Cagle P, et al. The presence of anti-HLA donor-specific antibodies
in lung allograft recipients does not correlate with C4d immunofluorescence in transbronchial biopsy specimens. Arch Path Lab Med 2014; 138:1053–1058.
24▪▪. Haas M, Sis B, Racusen L, et al. Banff 2013 meeting report: inclusion of C4d-negative antibody-mediated rejection
and antibody-associated arterial lesions. A J Transpl 2014; 14:272–283.
Recent consensus meeting in renal transplantation highlighting the limitations of C4d staining, and a recognition that C4d negative AMR represents a defined clinical entity.
25. Berry G, Burke M, Andersen C, et al. Pathology of pulmonary antibody-mediated rejection
: 2012 update from the Pathology Council of the ISHLT. J Heart Lung Transpl 2013; 32:14–21.
26. Fuller J, Paraskeva M, Thompson B, et al. A spirometric journey following lung transplantation
. Respirol Case Rep 2014; 2:120–122.
27▪. Otani S, Davis A, Cantwell L, et al. Evolving experience of treating antibody-mediated rejection
following lung transplantation
. Transplant Immunol 2014; 31:75–80.
(with [9▪]). One of a number of recent studies defining the relationship between pretransplant sensitization and subsequent clinical outcomes.
28. Witt C, Gaut J, Yusen R, et al. Acute antibody-mediated rejection
after lung transplantation
. J Heart Lung Transpl 2013; 32:1034–1040.
29▪. Tinkham K, Keshavjee S, Chaparro C, et al. Survival in sensitized lung transplant recipients with perioperative desensitization. Am J Transpl 2015; 15:417–426.
Study includes details of a desensitization protocol for sensitized wait-listed patients that is associated with equivalent posttransplant survival compared with unsensitized patients.
30. Walsh R, Alloway R, Girnita A, et al. Proteasome inhibitor-based therapy for antibody-mediated rejection
. Kid Int 2012; 81:1–5.
31. Commereuc M, Karras A, Amrein C, et al. Successful treatment of acute thrombotic microangiopathy by eculizumab after combined lung and kidney transplantation. Transplantation 2013; 96:358–359.
32. Baskaran G, Tiriveedhi V, Ramachandran S, et al. Efficacy of extracorporeal photopheresis in clearance of antibodies to donor-specific and lung-specific antigens in lung transplant recipients. J Heart Lung Transpl 2014; 33:950–956.
33. Snell G, Westall G, Paraskeva M. Immunosuppression and allograft rejection following lung transplantation
: evidence to date. Drugs 2013; 73:1793–1813.