Donor-specific antibodies are associated with antibody-mediated rejection, acute cellular rejection, bronchiolitis obliterans syndrome, and cystic fibrosis after lung transplantation

http://www.jhltonline.org

Donor-specific antibodies are associated with antibody-mediated rejection, acute cellular rejection, bronchiolitis obliterans syndrome, and cystic fibrosis after lung transplantation Leonard J. Lobo, MD,a Robert M. Aris, MD,a John Schmitz, MD,b and Isabel P. Neuringer, MDc From the aDivision of Pulmonary and Critical Care Medicine, and bPathology and Laboratory Medicine, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina; and the cPulmonary and Critical Care Unit, Massachusetts General Hospital, Boston Massachusetts.

KEY WORDS: lung transplant; donor-specific antibodies; antibody-mediated rejection; cystic fibrosis; lung allograft dysfunction

BACKGROUND: Lung transplantation is limited by chronic lung allograft dysfunction. Acute cellular rejection (ACR) is a risk factor for allograft dysfunction; however, the role of antibody-mediated rejection (AMR) is not well characterized. METHODS: This was a retrospective review from 2007 to 2011 of lung transplant recipients with human leukocyte antigen (HLA) antibody testing using Luminex (Luminex Corp, Austin, TX) single-antigen beads. Statistics included Fisher’s exact test for significance. RESULTS: Donor-specific antibodies (DSA) developed in 13 of 44 patients. Of the 13 with DSA, 12 had cystic fibrosis compared with 18 of 31 in the non-DSA group (p ¼ 0.035). Of those with DSAs, 23.1% occurred within the first year, and 69.2% occurred between 1 and 3 years. Twelve of 13 DSA patients had anti-HLA DQ specificity compared with 2 of 31 non-DSA patients (p ¼ 0.0007). AMR developed in 10 of the 13 DSA patients compared with 1 of 31 non-DSA patients (p ¼ 0.0001). The DSA group experienced 2.6 episodes/patient of cellular rejection vs 1.7 episodes/patient in the non-DSA group (p ¼ 0.059). Bronchiolitis obliterans syndrome developed in 11 of 13 in the DSA group vs 10 of 31 in the non-DSA group (p ¼ 0.0024). In the DSA group, 11.5% HLAs matched compared with 20.4% in the non-DSA group (p ¼ 0.093). AMR developed in 11 of 22 patients in the non-DSA HLA group compared with 0 of 22 in the group without non-DSA HLA antibodies (p ¼ 0.002). Survival at 1 and 3 years was 92% and 36% in the DSA group, respectively, and 97% and 65% in the non-DSA group. CONCLUSIONS: DSAs and non-DSAs occur frequently after lung transplantation. DSAs are prevalent in the cystic fibrosis population and are associated with AMR, bronchiolitis obliterans syndrome, and possibly, ACR. J Heart Lung Transplant 2013;32:70–77 r 2013 International Society for Heart and Lung Transplantation. All rights reserved.

Lung transplantation is the ultimate therapeutic approach for the treatment of terminal lung disease.1 Despite improved survival rates during the first year after transplant, the longReprint requests: Leonard J. Lobo, MD, University of North Carolina, Division of Pulmonary and Critical Care Medicine, 130 Mason Farm Rd, Campus Box 7020, Chapel Hill, NC 27599. Phone: þ1-434-989-1530.. E-mail address: [email protected]

term survival has remained static.2 Long-term survival is limited by recurrent immunologic events, including primary graft dysfunction, acute vascular rejection, lymphocytic bronchiolitis, bronchiolitis obliterans syndrome (BOS), humoral rejection, and non-immune causes, including infection and gastroesophageal reflux. These processes contribute to chronic lung allograft dysfunction (CLAD) and morbidity.3

1053-2498/$ - see front matter r 2013 International Society for Heart and Lung Transplantation. All rights reserved. http://dx.doi.org/10.1016/j.healun.2012.10.007

Lobo et al.

Significance Of DSAs After Lung Transplantation

The alloimmune response of organ rejection is guided by T-cell recognition of the allograft’s major histocompatibility complexes, otherwise known as human leukocyte antigens (HLAs). HLA class I genes include A, B, and C loci, which are expressed on nucleated cells. HLA class II genes include DR, DQ, and DP loci, which are expressed primarily on B and T cells, monocytes, dendritic cells, and other antigenpresenting cells.4,5 Lung transplant recipients can develop a humoral response to the allograft, specifically the donor HLAs. Antibodies binding to the allo-HLAs activate the compliment cascade, with complement deposition causing endothelial cell injury, pro-inflammatory molecule production, and inflammatory cell recruitment.3 Patients with a negative pre-transplant panel-reactive antibody test can develop de novo donor and non–donor-specific anti-HLA antibodies after transplant, conferring an increased incidence of acute rejection, BOS, and death.4–6 Antibodymediated allograft rejection (AMR) and donor-specific antibodies (DSA) are readily detected now with improved assays. AMR is recognized in lung allograft rejection, described as latent, silent, sub-clinical, and humoral, accompanied by circulating DSAs and C4d deposition, which progress toward clinical allograft dysfunction.7 Tissue pathology may show capillaritis.8 A diagnosis of AMR requires some of the following criteria: (1) histologic evidence of capillary injury, (2) circulating DSAs in serum, (3) C4d staining of the sub-endothelium, and (4) clinical evidence of allograft dysfunction.3 The classic complement-fixing DSAs result in C4d deposition, which is a recognized feature of AMR in renal transplantation, although additional studies demonstrate AMR in the absence of C4d.9,10 C4d deposition also

71 occurs in ischemia–reperfusion injury and acute rejection.11,12 AMR remains an ill-defined process that leads to lung allograft dysfunction.13,14 Prevention and treatment are limited by the lack of a consensus diagnosis, nonstandardized thresholds for DSA positivity, the unknown role of non-DSA HLA antibodies, sensitive new assays, and the variability in C4d staining. Optimal monitoring and treatment strategies for humoral rejection remain unclear, because risk factors for de novo DSA production have yet to be determined. We hypothesize that class II DSAs are linked to cystic fibrosis (CF), AMR, and BOS based on early clinical observations, and we performed a retrospective study of lung transplant patients tested for DSAs to investigate our hypothesis (Figure 1).

Methods This was a retrospective review of 44 patients who received lung transplants from January 1, 2007, to April 1, 2011, and who underwent testing for HLA antibodies. Before transplant, all patients were screened for pre-formed antibodies using solid-phase assays (enzyme-linked immunosorbent assay, flow cytometry, and Luminex [Luminex Corp, Austin, TX]). Molecular typing was performed on recipients and on donors at the time of transplant. HLA antibodies were assayed periodically after transplant during surveillance bronchoscopies and/or during episodes of unclear clinical graft dysfunction. Recipients were screened after transplant for DSAs using solid-phase assays and single-antigen testing. A DSA was considered positive when the median fluorescence intensity was greater than 1,000. Data were collected for date of transplant, age, sex, diagnosis, cytomegalovirus (CMV) status, cadaveric HLA molecular typing, time to DSA and non-DSA HLA detection, HLA antibody class,

P=0.40

BOS P=0.0024

1.8 episodes/pt

ACR

No DSAs & No antiHLA abs (n=21)

P=0.56

1.6 episodes/pt 1.7 episodes/pt

No DSAs but +anti-HLA abs (n=10)

P=0.059

2.6 episodes/pt

No DSAs (n=31) DSA (n=13)

P=0.32

AMR P=0.0001

0%

20%

40%

60%

80%

100%

% of cohort with described finding

Figure 1 Cohort comparison of the differing patient sub-groups according to human leukocyte antigen (HLA) antibody production. The no donor-specific antibody (DSA) cohort (n ¼ 31) includes patients in both the No DSAs and No anti-HLA antibody group (n ¼ 21) in addition to the cohort with No DSA but positive anti-HLA antibodies (n ¼ 10). Combining groups was possible because there was no statistically significant difference between the 2 groups.

72

The Journal of Heart and Lung Transplantation, Vol 32, No 1, January 2013 patients, of whom all 13 had HLA class II antibodies and 9 had both HLA class I and II antibodies.

number of acute rejection episodes (A1-A3, including C4d staining when clinical graft dysfunction was present with no clear infection), number of biopsies, microbiology (virus, bacterial, fungal), BOS/CLAD staging, immunosuppressant regimens, and survival. Specimens from open lung biopsies were used if those from transbronchial biopsies were non-diagnostic and there was an unclear etiology of graft dysfunction (4 4 liters of supplemental oxygen above baseline, radiographic change, or a precipitous decline in pulmonary function testing). AMR was diagnosed based on de novo DSA production, progressive graft dysfunction, and the absence of a significant coinfection.8 A biopsy specimen with C4d staining was not needed for confirmation. Because coexisting acute cellular rejection (ACR) is possible, ACR was not used as an excluder, and steroid nonresponsiveness in a patient with ACR increased suspicion for AMR. We used standard immunosuppression, including prednisone, mycophenolate, and tacrolimus, with no induction therapy. When AMR was diagnosed, it was treated with varying combinations of plasma exchange, intravenous immunoglobulin G (IVIG), pulsedose steroids, rituximab, and bortezomib. Rituximab and bortezomib were held if there was a concern for active infection. Bortezomib was not used for treatment until after 2010. Data analysis included simple descriptive statistics and for comparing groups, Fisher’s exact test for significance. Analysis was done with GraphPad software (GraphPad Inc, La Jolla, CA).

DSAs were detected within 1 year in 23.1% of the patients and within 1 to 3 years in 69.2%. The median time to DSA development was 63.8 ⫾ 51.9 weeks in the 10 patients with graft dysfunction and 135 ⫾ 255.1 weeks in the 3 patients without graft dysfunction.

Results

Association of HLA DQ with DSA development

Results are summarized in Tables 1 and 2.

Anti-HLA DQ specificity was found in 12 of the 13 patients with DSAs, and 10 of those 12 had graft dysfunction. Further testing of DQ specificities in the 10 with graft dysfunction showed 3 with an anti-DQA1 DSA. Non-DSA HLA antibodies were found in 10 of the 31 patients with no DSAs, with only 2 having HLA DQ specificity. Thus, HLA DQ antibody production was significantly associated with DSA production (p ¼ 0.0007).

DSA production Since 2007, 51 patients received transplants and 44 were tested for DSAs. All patients were DSA-negative before transplant. DSAs were not assessed in 7 patients due to early infection-related deaths. DSAs developed in 13 of 44

Table 1

DSA development and pre-transplant diagnosis CF accounted for 64.7% of all patients who received allografts. Of the 44 patients assayed for DSAs, 30 received transplants for CF (68%), and 14 (32%) for non-CF etiologies, including chronic obstructive pulmonary disease, idiopathic pulmonary fibrosis, acute respiratory distress syndrome, pulmonary lymphangiectasia, sarcoid, bronchiectasis, a1-antitrypsin deficiency, and repeat transplant. A statistically significant higher proportion of CF patients were in the DSA group (12 of 13) than in the non-DSA group (18 of 31; p ¼ 0.035).

Time to DSA detection

Patient Demographics and Cohort Comparison Entire cohort

Variablea Age, years Range Mean Female, % Diagnosis Cystic fibrosis Other etiologies Median time to DSA production, weeks HLA panels assayed per patient r1 year after transplant r2 and 3 years after transplant Survival 1-year 3-year

DSA

(n ¼ 44)

Negative (n ¼ 31)

Positive (n ¼ 13)

17–67 37.9 ⫾ 14.3 43.2

21–67 41.7 ⫾ 14.7 38.7

17–42 28.8 ⫾ 7.9 46.2

30 (68.2) 14 (31.8) NA

18 (58.1) 13 (41.9) NA

12 (92.3) 1 (7.7) 79.6 ⫾ 142.6

1.3 ⫾ 0.8 2.8 ⫾ 1.1

0.98 ⫾ 0.7 1.8 ⫾ 1.2

2.0 ⫾ 1.6 5.3 ⫾ 2.8

41/43 (95) 15/28 (54)

29/30 (97) 11/17 (65)

12/13 (92) 4/11 (36)

DSA, donor-specific antibodies; HLA, human leukocyte antigen; NA, not applicable a Continuous data are shown with the ⫾ standard deviation, or as indicated, and categoric data as number (%) or as indicated. b By Fisher’s exact test.

p-valuesb

0.025

0.52 0.14

Lobo et al. Table 2

Significance Of DSAs After Lung Transplantation

73

Analysis of the Donor-Specific Antibodies and Histopathology in Patients With Positive Donor-Specific Antibodies

Pt

DSA latency (month)

De novo HLA antibodies

1

15

2

36

A1; B44,45,76,82; DQ6; DR4,15; DRW51 DQ2,4,7,8,9; DR7,9; DRW53

Clinical diagnosis

Pathology at time of graft dysfunction

AMR

Organizing DAD

AMR

No tissue obtained

DR53 (1865) DR7 (1614) DQ2 (4438)

AMR

AMR

C4d negative DAD, organizing phase, patchy distribution, A1 C4d positive

DSAa

MFIb

A1 DQ6 DQ2

A1 (1771) DQ6 (3114) DQ2 (5433)

DR53 DR7 DQ2

3

6

DQ2,4,7; DR1,14

4

1

A1,3,11,30,31,32,36,74; B7,8,27,57,75,76; C5 DQ2,4,7,8,9

A3

A3 (1894)

DQA1

DQA1 (4283)

DQ8 DR53 DQA1 DQ7 DQ2 DQ7 DQ2

DQ8 (1837) DR53 (1148) DQA1 (1315) NA NA

AMR AMR

Acute and early organizing DAD, B2 A3

DQ2 (12340)

AMR

AMR

DAD, exudative and organizing phase, focal submucosal bronchiolar fibroblastic proliferation, obliterative bronchiolitis. C3d and C4d negative A3

5

14

B44,45,76 DQ2,4,8,9; DRW53; DR4,9

6 7

18 30

8

18

DQ4,7; DR 1,4,103,10 C5 DQ2,4,7,8,9; DR4 B8,76 DQ2,4,7,8

9

3

A11 DQ7,8,9; DRW52

DQA1

DQA1 (4789)

AMR

10

2

AMR

10

A23 DQ7 DRB1

NA

11

DRB1 (4854)

Well

No graft dysfunction

12

24

DQ2

DQ2 (9246)

Well

13

31

A2,23,24,32,34,68,80; B45 DQ7 A23, B40,60, C3,4,10 DR7, DRB1 (7), DRW (DRB3,4, DR52,53), DQ8,3 A11; C5, DQ2, DP DQ2

C4d positive Patchy DAD, exudative and organizing phase, B1 C4d negative Marked DAD, consisting of fibrin deposition within airspaces and reactive type II pneumocytes AOBO

DQ2

DQ2 (2370)

Well

No graft dysfunction C4d positive No graft dysfunction

AMR, antibody-mediated rejection; DAD, diffuse alveolar damage; DSA, donor-specific antibodies; HLA, human leukocyte antigen; NA, not available. a DSA checked at time of biopsy b Median fluorescence intensities

DSA development, clinical AMR, and histopathology

DSA development and episodes of cellular rejection

AMR developed in 10 of 13 the DSA group compared with 1 of 31 in the non-DSA group (p ¼ 0.0001). Biopsy specimens at the time of AMR in the DSA group showed diffuse alveolar damage (DAD), without significant positive microbiology in 5 of 10 patients, ACR (A1-A3) in 5 patients, and lymphocytic bronchitis (B1-B2) in 3 patients. Of 10 patients with DSAs with clinical graft dysfunction, 5 had DAD on pathology whereas only 10 of 31 patients in the non-DSA group had DAD on histopathology, yielding a positive predictive value of 32.7% and a negative predictive value of 80.6% for DAD. The 3 patients with DSAs but no AMR had surveillance biopsies but no clinical graft dysfunction to correlate specimens with.

Episodes of cellular rejection were compared between the groups. Sampling was similar: 6.7 times/patient in the DSA group compared with 6.5 times/patient in the non-DSA group. There were 2.6 episodes/patient of cellular rejection in the DSA group (n ¼ 13) compared with 1.7 episodes/patient in the non-DSA group (n ¼ 31), thus trending toward significance (p ¼ 0.059).

DSA development and BOS Patients with BOS only included those with BOS stages 1–3 (International Society for Heart and Lung Transplantation

74

The Journal of Heart and Lung Transplantation, Vol 32, No 1, January 2013

clinical staging criteria).15 Those without BOS consisted of BOS Op or no BOS. BOS developed in 11 of the 13 in the DSA group compared with 10 of 31 in the non-DSA group. Thus, the presence of DSAs was associated with BOS (p ¼ 0.0024). Because BOS incidence increases with time after transplant, the average time alive after transplant was calculated. As of January 1, 2012, the average number of days alive after transplant per patient was 835 days in the DSA group and 924 days in the non-DSA group; thus, the DSA group had a higher incidence of BOS over a shorter period of time.

attributable to AMR with respiratory failure and the increased immunosuppression to treat AMR.

Non-DSA HLA antibody production

Five of 13 patients in the DSA group CMV mismatched (donorþ/recipient–), and 12 of 31 in the non-DSA group were CMV mismatched. CMV mismatch was not statistically significant between the DSA and non-DSA patients (p ¼ 0.96). CMV viral replication within the blood occurred in 2 of 13 DSA patients and in 7 of 24 non-DSA patients, which was not a statistically significant difference (p ¼ 0.87). The frequency of CMV monitoring was similar in both groups.

Non-DSA HLA antibodies developed in 22 of 44 patients. In the DSA group, 74 non-DSA HLA antibodies were detected in 12 of 13 patients, and 41were detected in 10 of 31 patients in the non-DSA group; thus, non-DSA HLA antibodies were detected to a significantly greater degree in the DSA group than in the non-DSA group (p ¼ 0.0006). There were 2.18 episodes/patient of A2 or greater histopathologic rejection (including C4d positivity) in the non-DSA HLA antibody cohort compared with 1.82 episodes/patient in the cohort without non-DSA HLA antibodies, which was not a statistically significant difference (p ¼ 0.9). BOS developed in 14 of 22 patients in the non-DSA HLA group compared with 7 of 22 in the group without non-DSA HLA antibodies, thus trending toward statistical significance (p ¼ 0.069). AMR developed in 11 of 22 patients in the non-DSA HLA group compared with 0 of 22 in the group without non-DSA HLA antibodies, which was a statistically significant difference (p ¼ 0.002).

DSA development and degree of HLA mismatch

Discussion

Patients in the DSA and non-DSA groups were evaluated for number of total HLA matches out of the total number of potential matches. Matching was performed at available loci between donor and recipient for HLA-A,- B, -C, -DR, and DQ, ranging from 4 loci to 10 possible loci. In the DSA group (n ¼ 13), 10 loci matched for HLA, but 77 loci did not match (11.5%). In the non-DSA group (n ¼31), 42 loci matched for HLA, but 164 did not match (20.4%). No statistical significance was noted between the groups (p ¼ 0.093).

Our understanding of AMR continues to evolve as evidence increasingly supports a role for AMR in acute and steroid-resistant rejection, CLAD, BOS, and worsened mortality.6,14,16 Adding to this evolution is the enhanced sensitivity of solid-phase assays for HLA antibodies, which has increased the detection of both donor-specific and nondonor specific HLA antibodies.17 This enables clinicians to detect low levels of antibodies before clinical consequences such as histologic changes and graft dysfunction. Currently, the lung community recognizes AMR as a clinical entity, with a progression from clinical allograft dysfunction to the detection of HLA antibodies in the absence of graft dysfunction.8 When positive, tissue histopathology for C4d may constitute an important confirmatory signal.6,12,18,19 In this single-center retrospective study, we identified additional risks of DSAs and AMR, including a CF diagnosis, HLA class II DSAs with anti-DQ specificity, BOS, and possibly, ACR. To date, no study has demonstrated an increased predisposition for AMR in CF after transplant, for which survival approaches 65% at 5 years.2 We identified CF patients at increased risk for DSA production, which may be related to the high percentage of patients who receive allografts at this center for CF (60%), to chronically elevated antibody levels to bacterial and fungal pathogens, to nonspecific antibody stimulation by infection, or to the contribution of gene modifiers in CF.20,21 The presence of autoimmunity and autoantibodies in CF with as-yetunknown cross-reactivities and pro-inflammatory immune responses in the presence of recipient HLA polymorphisms (HLA-DR4 and HLA-DR7) and persistent infections after transplant due to colonization may also contribute.22 It has

DSA development and degree of CMV mismatch

DSA development and survival As of January 1, 2012, 12 of 13 patients (92%) in the DSA group were alive at 1 year after transplant and 4 of 11 patients (36%) were alive at 3 years. In the non-DSA group, 29 of 30 patients (97%) were alive at 1 year and 11 of 17 (65%) were alive at 3 years. There was no statistical significant difference in survival at 1 (p ¼ 0.52) and 3 years (p ¼ 0.14) after transplant; however, survival was trending toward significance at 3 years after transplant.

Mortality analysis In the DSA group, 7 of 11 patients died within the first 3 years after transplant: 5 likely died of AMR and 2 died of severe sepsis after AMR-targeted therapy (plasma exchange, IVIG, pulse-dose steroids, rituximab, and bortezomib). In the non-DSA group, 6 of 17 patients died within 3 years: 5 died of BOS and 1 died of early infectious complications. The higher 3-year mortality in the DSA group was

Lobo et al.

Significance Of DSAs After Lung Transplantation

also been demonstrated that T-lymphocytes in CF patients exhibit a pattern of cytokine production different from that of normal T-lymphocytes, and this could be the direct influence of the CFTR gene on normal T-lymphocyte expression.23,24 The median time to DSAs detection was 557 days after transplant and all were de novo. Contrary to us, Hachem et al16 found that 56% of DSAs occurred during the first 90 days after transplant. Because the HLA testing in some of the patients at our center occurred at time of clinical graft dysfunction rather than during surveillance bronchoscopy, DSAs may have been present earlier. Our longer DSA latency period may also be attributed to higher prednisone dosing in the absence of induction therapy and the use of the anti–B-cell agent mycophenolate, in contrast to induction and azathioprine, which targets T cells to a greater extent than B cells. Combining our experience with Hachem et al,16 it is reasonable to assess for DSAs at the same intervals that current programs assess for cellular rejection, at 1, 3, 6, and 9 months and yearly after transplant, as well as during episodes of graft dysfunction. Monitoring frequency should be increased in the higher-risk populations such as CF patients and patients with pre-transplant anti-HLA antibodies. Takemoto et al7 described 4 categories of humoral response to a graft: 1. latent humoral response with circulating antibodies alone, with negative biopsy specimens and preserved graft function; 2. silent humoral reaction with circulating antibodies and C4d deposition but preserved graft function and no histologic changes; 3. sub-clinical humoral rejection with circulating antibodies, C4d deposition, and histologic changes but preserved graft function; and 4. humoral rejection with circulating antibodies, C4d deposition, histologic changes and graft dysfunction. As more is more learned about humoral rejection, we will probably realize that these are more like stages rather than categories and that patients will advance through the stages at different rates. Early detection and diagnosis of the latent humoral response may allow early intervention and prevent progression into silent humoral reaction, sub-clinical humoral reaction, and full humoral reaction. Although patients express anti-HLA class I and II DSAs independently and simultaneously, we identified an association between anti-HLA class II DSAs and clinical AMR that reinforces earlier findings in lung transplantation.14,16,23 In cardiac transplantation, anti-HLA class II DSAs directed against the DQ epitope constituted the predominant AMRassociated de novo DSA.24,25 Similarly, our study described DQ specificity in AMR. DQ specificity is further defined by the number of epitopes that are expressed on both the a and b chains of the HLA-DQ complex.26,27 Recent attention to the multilevel dimensions of DQ antigenicity have been recognized, such that a patient reacts with his or her own DQ-a or b chain when complexed with a non–self-DQ-a or

75 a non–self-DQ-b chain.28 Because additional DQ antigen typing will be performed as part of Organ Procurement and Transplantation Network/United Network of Organ Sharing allocation in the future, the potential for identification of DQ mismatch and virtual cross-match may ultimately require DQ-a as well as DQ-b chain specificities. As noted in our results, a recipient who is prone to make anti-HLA antibodies with DQ specificity is more likely to make DSAs. This could play a role in surveillance and treatment. If a patient has a positive pre-transplant panel reactive antibody with DQ specificity, consider antibodytargeted treatment before transplant to reduce the anti-HLA burden, matching the donor and recipient as closely as possible, augmenting the post-transplant immunosuppression, and increasing the HLA monitoring frequency. If a patient develops de novo anti-HLAs with DQ specificity after transplant, consider intensifying the immunosuppression and the HLA surveillance frequency. The data comparing clinical graft dysfunction and DAD on histopathology yielded a positive predictive value of 32.7% and a negative predictive value of 80.6% for DAD. Although DAD was not sensitive for AMR, the specificity of 80% suggests some benefit in helping to exclude AMR when DAD is absent. DAD also results from infection and gastroesophageal reflux; therefore, the clinical presentation is crucial. Given the association of ACR as antecedent or concurrent with AMR, the number of episodes of A2 or greater cellular rejection was determined in the DSA and non-DSA groups.6,12 The association between DSAs and cellular rejection trended toward significance. These findings parallel the findings of Hachem et al.15 The concurrence of ACR and AMR was demonstrated in renal transplantation and likely results from migration of DSA producing alloreactive B cells to the allograft, thus, presenting as lymphocytic infiltrates on the biopsy specimen. An association between AMR/DSAs and CLAD, as assessed by BOS staging, was identified. These findings concur with the study by Palmer et al,14 and others16,29 who found decreased BOS when AMR was treated with IVIG and/or rituximab. Evidence across solid-organ transplantation suggests that AMR confers worse long-term outcomes and predisposes to chronic allograft dysfunction.30–35 Prevention and treatment of AMR requires optimization to prevent long-term graft loss. The 1-year survival after transplant in the DSA group compared with the non-DSA group was similar; however, survival was greater in the non-DSA group at 3 years. This correlates to the higher prevalence of DSAs between years 1 and 3. Non-DSA HLA antibodies were associated with AMR and possibly BOS but with no clear association with cellular rejection. A consensus workshop found that non-DSA HLA antibodies with a strong signal had an association with graft failure across solid-organ transplantation.36 Possible mechanisms include cross-reactivity with the major proteins A B C OR DR/DQ/DP, mismatches at the allele level, and polymorphic epitopes with multiple targets. This raises awareness for whether non-DSA HLA antibodies require therapy.

76

The Journal of Heart and Lung Transplantation, Vol 32, No 1, January 2013

Although preceding community respiratory viral, CMV, and bacterial infections may be associated with ACR and BOS, limited data exist on any association between lung infection and DSA development.37–40 CMV infection may predispose to anti-endothelial cell antibody or non-HLA antibody development.41,42 Our study did not identify a significant association between CMV mismatch or active viral replication at any time with clinical AMR. HLA matching is limited in lung transplantation due to necessity for shortened ischemic times and lung availability. Lung transplant studies show benefits for matching at individual HLA loci and an association of BOS with mismatching; however, the relationship between AMR and HLA matching is unknown.43–45 The DSA and non-DSA groups were evaluated for HLA matching, and although there were only 11.5% HLAs matched in the DSA groups compared with 20.4% in the non-DSA group, the difference was not statistically significant. This study has several limitations. Firstly, this was a retrospective analysis during a period when our understanding of AMR was constantly evolving; thus, our laboratory analysis, surveillance regimens, and treatment algorithms changed throughout the course of the examined period. As a result, patients were not being routinely screened, and early DSAs, before graft dysfunction, may have been missed. Also, the risk for selection bias was notable. Secondly, the patient population was small, and the inability to control variables might have led to wrong conclusions. The size was, however, substantial enough to provide statistically significant data. Thirdly, most patients, when presenting with clinical graft dysfunction, were treated with broad-spectrum antimicrobials at the time of bronchoscopy, possibly masking infection. However, broad anti-microbial therapy with no clinical response was required before diagnosing AMR. Fourthly, we had a large proportion of CF patients. However, this was favorable because it allowed better analysis of DSA production and their effects on the CF patients. Fifthly, HLA antibody detection by solid-phase Luminex bead technology is extremely sensitive, with uncertain median fluorescence intensity cutoffs and multiple falsepositives.46 In addition, AMR may occur due to the presence of non-DSA antibodies, including k-a-tubulin and collagen IV in the lung, as well as anti-endothelial cell epitopes, for which assays are not yet clinically available.47–49 Lastly, the lack of a formal consensus for diagnosing clinical AMR, reliable tissue findings, and limitations surrounding C4d staining continue to hinder a clear determination of AMR. Our lack of C4d staining limited our clinical AMR confirmation, although this could represent non–C4d-related episodes of AMR, as reported in renal transplantation.50 The findings noted in this study should be cautiously interpreted but can add to the current understanding of AMR. In our study, the association between AMR/DSAs with CF, DSAs with clinical AMR/BOS, non-DSA HLA antibodies with AMR/BOS, and de novo HLA-DQ antibodies with AMR should lower the screening threshold. The identification of a common DQA1 allele may further

HLA testing at the allele level and forward investigation into the pathogenesis of antibodies to self-HLAs complexed to non–self-HLA antigens. Although the finding of diffuse alveolar damage for AMR lacked sensitivity, it still conferred an 80% specificity, which, in this pneumoniaprone population, may aid in clinical diagnosis. In summary, substantial knowledge may be gained from these findings with the hope to advance our understanding of AMR presentation, classification, and pathogenesis.

Disclosure statement 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. Orens JB, Garrity ER Jr. General overview of lung transplantation and review of organ allocation. Proc Am Thorac Soc 2009;6:13-9. 2. Christie JD, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplantation: twentyseventh official adult lung and heart-lung transplant report—2010. J Heart Lung Transplant 2010;29:1104-18. 3. Martinu T, Howell DN, Palmer SM. Acute Cellular Rejection and Humoral Sensitization in Lung Transplant Recipients. Semin Respir Crit Care Med 2010;2:179-88. 4. Klein J, Sato A. The HLA system. First of two parts. N Engl J Med 2000;343:702-9. 5. Klein J, Sato A. The HLA system. Second of two parts. N Engl J Med 2000;343:782-6. 6. Girnita AL, McCurry KR, Iacono AT, et al. HLA-specific antibodies are associated with high-grade and persistent recurrent lung allograft acute rejection. J Heart Lung Transplant 2004;23:1135-41. 7. Takemoto SK, Zeevi A, Feng S, et al. National conference to assess antibody-mediated rejection in solid organ transplantation. Am J Transplant 2004;4:1033-41. 8. Stewart S, Fishbein MC, Snell GI, et al. Revision of the 1996 working formulation for the standardization of nomenclature in the diagnosis of lung rejection. J Heart Lung Transplant 2007:26;1229–1242. 9. Sis B, Mengel M, Haas M, et al. Banff ’09 meeting report: antibody mediated graft deterioration and implementation of Banff working groups. Am J Transplant 2010;10:464-71. 10. Hass M. C4d-negative antibody-mediated rejection in renal allografts: evidence for its existence and effect on graft survival. Clin Neprol 2011;75:271-8. 11. Westall GP, Snell GI, McLean C, Kotsimbos T, Williams T, Magro C. C3d and C4d deposition early after lung transplantation. J Heart Lung Transplant 2008;27:722-8. 12. Magro CM, Pope Harman A, Klinger D, et al. Use of C4d as a diagnostic adjunct in lung allograft biopsies. Am J Transplant 2003;3:1143-54. 13. Magro CM, Ross P Jr, Kelsey M, Waldman WJ, Pope-Harman A. Association of humoral immunity and bronchiolitis obliterans syndrome. Am J Transplant 2003;3:1155-66. 14. Palmer SM, Davis RD, Hadjiliadis D, et al. Development of an antibody specific to major histocompatibility antigens detectable by flow cytometry after lung transplant is associated with bronchiolitis obliterans syndrome. Transplantation 2002;74:799-804. 15. Estenne M, Maurer JR, Boehler A, et al. Bronchiolitis obliterans syndrome 2001: an update of the diagnostic criteria. J Heart Lung Transplant 2002;21:297-310. 16. Hachem RR, Yusen RD, Meyers BF, et al. Anti-human leukocyte antigen antibodies and preemptive antibody-directed therapy after lung transplantation. J Heart Lung Transplant 2010;29:973-80. 17. Leffell MS, Zachary AA. Antiallograft antibodies: relevance, detection, and monitoring. Curr Opin Organ Transplant 2010;15:2-7.

Lobo et al.

Significance Of DSAs After Lung Transplantation

18. Ionescu DN, Girnita AL, Zeevi A, et al. C4d deposition in lung allografts is associated with circulating anti-HLA alloantibody. Transpl Immunol 2005;15:63-8. 19. Miller GG, Destarac L, Zeevi A, et al. Acute humoral rejection of human lung allografts and elevation of C4d in bronchoalveolar lavage fluid. Am J Transplant 2004;4:1323-30. 20. Collaco JM, Cutting GR. Update on gene modifiers in cystic fibrosis. Curr Opin Pulm Med 2008;14:559-66. 21. Aron Y, Polla BS, Bienvenu T, Dall’ava J, Dusser D, Hubert D. HLA class II polymorphism in cystic fibrosis. A possible modifier of pulmonary phenotype. Am J Respir Crit Care Med 1999;159: 1464-8. 22. Carter CJ. Pathogen and autoantigen homologous regions within the cystic fibrosis transmembrane conductance regulator (CFTR) protein suggest an autoimmune treatable component of cystic fibrosis. FEMS Immunol Med Microbiol 2011;62:197-214. 23. Hubeau C, Le Naour R, Abely M, et al. Dysregulation of IL-2 and IL-8 production in circulating T lymphocytes from young cystic fibrosis patients. Clin Exp Immunol 2004;135:528-34. 24. Moss RB. Lymphocytes in cystic fibrosis lung disease: a tale of two immunities. Clin Exp Immunol 2004;135:358-60. 25. Reinsmoen NL, Nelson K, Zeevi A. Anti-HLA antibody analysis and crossmatching in heart and lung transplantation. Transpl Immunol 2004;13:63-71. 26. Iniotaki-Theodoraki AG, Boletis JN, Trigas GC, Kalogeropoulou HG, Kostakis AG, Stavropoulos-Giokas CG. Humoral immune reactivity against human leukocyte antigen (HLA)-DQ graft molecules in the early post-transplant period. Transplantation 2003;75:1601. 27. Hourmant M, Cesbron-Gautier A, Terasaki P, et al. Frequency and clinical implications of development of donor-specific and non-donorspecific HLA antibodies after kidney transplantation. J Am Soc Nephrol 2005;16:2804. 28. Tambur AR, Leventhal JR, Friedewald JJ, Ramon DS. The complexity of human leukocyte antigen (HLA)-DQ antibodies and its effect on virtual crossmatching. Transplantation 2010;90:1117-24. 29. Appel JZ 3rd, Hartwig MG, Davis RD, Reinsmoen NL. Utility of peritransplant and rescue intravenous immunoglobulin and extracorporeal immunoadsorption in lung transplant recipients sensitized to hla antigens. Hum Immunol 2005;66:378-86. 30. Gloor J, Cosio F, Lager DJ, Stegall MD. The spectrum of antibodymediated renal allograft injury: implications for treatment. Am J Transplant 2008;8:1367-73. 31. Mao Q, Terasaki PI, Cai J, et al. Extremely high association between appearance of HLA antibodies and failure of kidney grafts in a fiveyear longitudinal study. Am J Transplant 2007;7:864-71. 32. Haas M, Montgomery RA, Segev DL, et al. Subclinical acute antibodymediated rejection in positive crossmatch renal allografts. Am J Transplant 2007;7:576-85. 33. Nair N, Ball T, Uber PA, Mehra MR. Current and future challenges in therapy for antibody-mediated rejection. J Heart Lung Transplant 2011;30:612-7. 34. Kfoury AG, Hammond ME. Controversies in defining cardiac antibody-mediated rejection: need for updated criteria. J Heart Lung Transplant 2010;29:389-94.

77 35. Montgomery RA, Hardy MA, Jordan SC, et al. Consensus opinion from the antibody working group on the diagnosis, reporting, and risk assessment for antibody-mediated rejection and desensitization protocols. Transplantation 2004;7:181-5. 36. Ozawa M, Terasaki PI, Castro R, et al. 4th International HLA and Immunogenetics Workshop Prospective Chronic Rejection Project: a three-year follow-up analysis. Clin Transpl 2007:255-60. 37. Kumar D, Husain S, Chen MH, et al. A prospective molecular surveillance study evaluating the clinical impact of communityacquired respiratory viruses in lung transplant recipients. Transplantation 2010;89:1028-33. 38. Vu DL, Bridevaux PO, Aubert JD, Soccal PM, Kaiser L. Respiratory viruses in lung transplant recipients: a critical review and pooled analysis of clinical studies. Am J Transplant 2011;11:1071-8. 39. Paraskeva M, Bailey M, Lewey BJ, et al. Cytomegalovirus replication within the lung allograft is associated with bronchiolitis obliterans syndrome. Am J Transplant 2011;11:2190-6. 40. Botha P, Archer L, Anderson RL, et al. Pseudomonas aeruginosa colonization of the allograft after lung transplantation and the risk of bronchiolitis obliterans syndrome. Transplantation 2008;85:771-4. 41. Toyoda M, Galfayan K, Galera OA, Petrosian A, Czer LS, Jordan SC. Cytomegalovirus infection induces anti-endothelial cell antibodies in cardiac and renal allograft recipients. Transpl Immunol 1997;5:104-11. 42. Costa C, Touscoz GA, Bergallo M. Non-organ-specific and antiendothelial antibodies in relation to CMV infection and acute rejection in renal transplant recipients. Clin Transplant 2010;24:488-92. 43. Brugiere O, Thabut G, Suberbielle C, et al. Relative impact of human leukocyte antigen mismatching and graft ischemic time after lung transplantation. J Heart Lung Transplant 2008;27:628-34. 44. Opelz G, S¨usal C, Ruhenstroth A, D¨ohler B. Impact of HLA compatibility on lung transplant survival and evidence for an HLA restriction phenomenon: a collaborative transplant study report. Transplantation 2010;90:912-7. 45. Chalermskulrat W, Neuringer IP, Schmitz JL, et al. Human leukocyte antigen mismatches predispose to the severity of bronchiolitis obliterans syndrome after lung transplantation. Chest 2003;123:1825. 46. Cecka JM. Current methodologies for detecting sensitization to HLA antigens. Curr Opin Organ Transplant 2011;16:398-403. 47. Goers TA, Ramachandran S, Aloush A, Trulock E, Patterson GA, Mohanakumar T. De novo production of K-alpha1 tubulin-specific antibodies: role in chronic lung allograft rejection. J Immunol 2008;180:4487-94. 48. Jackson AM, Lucas DP, Melancon JK, Desai NM. Clinical relevance and IgG subclass determination of non-HLA antibodies identified using endothelial cell precursors isolated from donor blood. Transplantation 2011;92:54-60. 49. Zhang Q, Reed EF. Non-MHC antigenic targets of the humoral immune response in transplantation. Curr Opin Immunol 2010;22: 682-8. 50. Sis B, Halloran PF. Endothelial transcripts uncover a previously unknown phenotype: C4d-negative antibody-mediated rejection. Curr Opin Organ Transplant 2010;15:42-8.