Since the first Banff meeting in 1991, the diagnosis and

Transcription

1 CLINICAL AND TRANSLATIONAL RESEARCH Acute Cellular Rejection: Impact of Donor-Specific Antibodies and C4d Michelle Willicombe, 1,5 Candice Roufosse, 2 Paul Brookes, 3 Adam G. McLean 1, Jack Galliford, 1 Tom Cairns, 1 Terry H. Cook, 4 and David Taube 1 Background. Mixed rejection in kidney transplantation consists of histologic and/or serological evidence of both cellular and humoral components. As it is not confined to a distinct category in the Banff classification, how to best manage these patients is not clear. The aim of this study was to determine the incidence and outcome of morphological T-cellYmediated rejection (TCMR) with a humoral component, defined as the presence of either DSA or C4d, compared with the outcome of pure TCMR. Methods. We retrospectively studied 922 consecutive renal transplant recipients and analyzed patients with TCMR according to the evidence of a humoral component. Results. A total of 147 cases of morphological TCMR were analyzed. Of these, 92 (62.6%) had pure TCMR and 55 (37.4%) had mixed TCMR on the index biopsy. On univariant analysis, diffuse C4d (odds ratio [OR]=10.9, 95% confidence interval [CI]=1.8Y66.9, P=0.01) and DSA positivity at the time of index biopsy (OR=2.8, 95% CI=1.2Y6.6, P=0.02) were associated with allograft loss, whereas arteritis (OR=0.5, 95% CI=0.2Y1.2, P=0.11) and glomerulitis (OR=0.9, 95% CI=0.4Y2.1, P=0.8) were not. Arteritis was associated with subsequent antibody-mediated rejection (OR=4.9, 95% CI=1.1Y20.8, P=0.03), and glomerulitis was associated with the development of transplant glomerulopathy (OR=10.7, 95% CI=3.1Y37.1, PG0.01). On the multivariate analysis, only patients with C4d and DSA were at risk of graft failure (OR=4.9, 95% CI=2.0Y12.0, PG0.01) in the medium term. Conclusion. TCMR with a humoral component has a worse prognosis when compared with pure TCMR. As such, it is important to test for alloantibody in cases of morphological TCMR to optimize patient management. Such cases might benefit from more aggressive immunotherapy. Keywords: Acute cellular rejection, Mixed rejection, C4d, Donor-specific antibodies, Glomerulitis. (Transplantation 2014;97: 433Y439) A.G.M. has received a grant/research support from Astellas pharma. The other authors declare no funding or conflicts of interest. 1 Imperial College Kidney and Transplant Centre, Hammersmith Hospital, London, UK. 2 Department of histopathology, Imperial College NHS Trust, London, UK. 3 Department of Histocompatibility and Immunogenetics, Imperial College NHS Trust, London, UK. 4 Centre for Complement and Inflammation Research, Imperial College, London, UK. 5 Address correspondence to: Michelle Willicombe, Imperial College Kidney and Transplant Centre, Hammersmith Hospital, London, UK W12 0HS. Michelle.Willicombe@imperial.nhs.uk M.W. collected and analyzed the clinical data along with writing the article. P.B. contributed to the generation of patient DSA data. C.R. and T.C. contributed by reporting the histology. J.G., A.G.M., and T.C. participated in research design, generation of patient material, and reviewing the article. D.T. supervised the research and was involved at all levels. Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal s Web site ( Received 30 May Revision requested 17 June Accepted 3 September Copyright * 2014 by Lippincott Williams & Wilkins ISSN: /14/ DOI: /01.TP f Since the first Banff meeting in 1991, the diagnosis and classification of renal transplant rejection has been continually adapting according to technological advances and available evidence. The drive behind this perpetual focus is that acute rejection is still associated with inferior allograft survival, and chronic rejection is responsible for the failure of allograft longevity despite the use of potent immunosuppressive agents (1, 2). Antibody-mediated rejection (AMR) is associated with a worse prognosis than T-cellYmediated rejection (TCMR); however, the different types of rejection are not mutually exclusive and can coexist, with TCMR often being reported as seen in conjunction with AMR (3Y7). Because mixed rejection is not confined to a discrete category in the Banff classification, it is often ill-defined in the literature (8). According to Banff, tubulointerstitial rejection is the hallmark of TCMR, whereas AMR is defined by the presence of acute tissue injury with concurrent C4d staining and circulating donor-specific antibodies (DSAs) (8). Two specific histologic features require special emphasis; vascular lesions are polysemous being implicated in both cellular- and antibody-mediated rejection and glomerulitis is another ambiguous lesion currently classified by Banff as being associated with AMR only. However, studies have shown that Transplantation & Volume 97, Number 4, February 27,

2 434 Transplantation & Volume 97, Number 4, February 27, 2014 glomerulitis can occur in TCMR with the type of infiltrating cell determining the underlying pathogenesis (9Y11). Few studies have specifically analyzed the outcomes of morphological TCMR with a humoral component, defined as C4d positivity and/or DSA, compared with pure TCMR. The importance of correctly diagnosing rejection type has therapeutic implications because TCMR is often responsive to corticosteroids, whereas AMR requires more aggressive therapy directed at DSA removal along with the prevention of antibody production or its action. Mixed rejection treated as conventional pure TCMR may increase the risk of chronic active antibody rejection, transplant glomerulopathy (TG), and allograft failure (7). The aims of this study were as follows: to determine the frequency of TCMR with a humoral component; to compare the outcomes in terms of the subsequent development of AMR, TG, or allograft failure; and to analyze the impact of the different Banff grades and ambivalent lesions on outcomes. RESULTS We retrospectively analyzed 147 (15.9%) of 922 consecutive transplant recipients who had an indicative biopsy with morphological TCMR. Patient demographics are shown in Table 1. The median time to rejection was 6.70 months (range=4.91y8.26 months), with a median follow-up post index biopsy of months (range=26.74y35.59 months). There was a trend to worse allograft survival in patients who had rejection after the first 3 months after transplantation, with allograft survival being 87.6%, 51.7%, and 46.7% in the early rejection (G3 months), intermediate rejection (3Y12 months), and late rejection (912 months) groups, respectively (P=0.051). Of the 147 recipients, 92 (62.6%) had pure TCMR(C4d- DSAj) and 55 (37.4%) had mixed rejection, defined as TCMR with evidence of a humoral component: 13 (8.8%) had C4d+DSAjTCMR, 33 (22.4%) had C4djDSA+TCMR, and 9 (6.1%) had C4d+DSA+TCMR. Patients who were sensitized or receiving a second allograft or more were more likely to have a mixed rejection episode. Of the 92 patients with pure TCMR,6(6.5%)hadreceivedaregraftcomparedwith14 (25.5%) of the 55 patients with mixed TCMR (P=0.002). Seventeen (18.5%) of the 92 pure TCMR cases compared with 28 (50.9%) of the 55 mixed TCMR cases developed in sensitized patients (PG0.001). Outcomes by DSA Status Of the 147 recipients, 42 (28.6%) had DSA detected at the time or before the index TCMR episode and 19 (40.4%) had preformed DSA, and 23 (18.0%) of 128 had developed de novo DSA. Patients with DSA were at an increased risk of allograft failure, with allograft survival being 72.6% and 30.5% in the DSAj and DSA+ groups, respectively (P=0.021; Fig. 1A). Patients with DSA were more likely to develop AMR, with an AMR-free survival of 92.1% and 74.4% in the DSAj and DSA+ groups, respectively (P=0.0002). There was no difference in the risk of TG between the patients with DSAj and DSA+ TCMR, with TGfree survival being 82.6% and 82.1%, respectively (P=0.11). However, following the index biopsy, 9 patients who were initially DSAj subsequently became DSA+, 5 of whom developed TG. Therefore, the detection of DSA considering all time points in a patient with TCMR was associated with TG. Eleven (21.6%) of 51 DSA+ patients compared with 4 (4.2%) of 96 DSAj patients who developed TG (P=0.0004). Analyzing the effect of DSA HLA class on outcomes, of the 42 cases, 23 (54.8%) had HLA class I DSA, 11 (26.2%) had HLA class II DSA, and 8 (19.0%) had both HLA class I and II DSA. Patients with both class I and II DSA were at the highest risk of subsequent allograft failure. Five (62.5%) of 8 patients with class I and II DSA lost their graft compared with 3 (27.3%) of 11 patients with class II DSA, 4 (17.4%) of 23 patients with class I DSA, and 17 (16.2%) of 105 DSAj patients (P=0.0002). Outcomes of patients with nonydonor-specific HLA antibody are shown in the Supplemental Data (see SDC, Outcomes by C4d Status Twenty-two (15.0%) of 147 patients were C4d positive on the index TCMR biopsy. Of these patients, 6 (27.3%) had diffuse C4d and 16 (72.7%) had focal C4d. Diffuse C4d was strongly associated with DSA. Of the 6 patients with diffuse C4d, 4 had DSA at the time of biopsy (P=0.03); the 2 remaining patients had DSA detected subsequently. Focal C4d was not associated with DSA; 5 (31.3%) of 16 patients with TCMR and focal C4d had DSA compared with 33 (26.4%) of 125 C4d-negative patients having DSA (P=0.77). Patients with diffuse C4d were at the highest risk of allograft failure, AMR, and TG. Allograft survival in the diffuse C4d, focal C4d-positive, and C4d-negative patients was 33.3%, 64.5%, and 73.1%, respectively (P=0.033; Fig. 1B). AMRfree survival was 16.7%, 81.2%, and 93.7% (PG0.001) and TG-free survival was 25.0%, 84.4%, and 85.8% (P=0.003) in the diffuse, focal, and negative groups, respectively. Outcomes by Banff Grade and Arteritis Of the 147 patients, 20 (13.6%) had Banff borderline, 107 (72.8%) had Banff I, 19 (12.9%) had Banff II, and 1 (0.7%) had Banff III TCMR. Vascular involvement was not associated with allograft failure (hazard ratio [HR]=0.46, 95% confidence interval [CI]=0.18Y1.20, P=0.11) or TG (HR=1.21, 95% CI=0.32Y4.64, P=0.78). However, arteritis was associated with the risk of subsequent AMR (HR=4.87, 95% CI=1.14Y20.83, P=0.03). Arteritis was not associated with DSA at the time of index biopsy, with 38 (29.9%) of 127 Banff BL/I cases having DSA compared with 4 (20.0%) of 20 Banff II/III cases (P=0.44). However, patients with vascular involvement were more likely to develop subsequent DSA, with 5 (5.6%) of 89 Banff BL/I cases subsequently developing DSA compared with 4 (25.0%) of 16 Banff II/III cases (P=0.029). Arteritis was not associated with C4d, with 17 (13.4%) of 127 Banff BL/I cases being C4d positive compared with 4 (20.0%) of 20 Banff II/III cases being C4d positive (P=0.49). Impact of Glomerulitis and Capillaritis Of the 147 patients, 25 (17.0%) had glomerulitis. Glomerulitis was not associated with allograft failure (odds ratio [OR]=0.89, 95% CI=0.37Y2.13, P=0.80) or AMR (OR=0.64, 95% CI=0.18Y2.23, P=0.48). However, glomerulitis was associated with the development of TG (OR=10.67, 95% CI=3.07Y37.08, P=0.0002). Overall, glomerulitis was not associated with C4d staining, which was found in 4 (16.0%) of 25

3 * 2014 Lippincott Williams & Wilkins Willicombe et al. 435 TABLE 1. Patient demographics at the time of transplantation Overall (n=147) Pure group (n=92) Mixed group (n=55) P a Sex, n (%) Male 104 (70.7) 70 (76.1) 34 (61.8) 0.10 Female 43 (29.3) 22 (23.9) 21 (38.2) Age (yr), meantsd 46.38T T T Ethnicity, n (%) Caucasian 72 (49.0) 46 (50.0) 26 (47.3) 0.70 Indoasian 45 (30.6) 28 (30.4) 17 (30.9) Afrocaribbean 24 (16.3) 13 (14.1) 11 (20.0) Other 6 (4.1) 5 (5.4) 1 (1.8) Cause of renal failure, n (%) APKD b 14 (9.5) 7 (7.6) 7 (12.7) 0.35 Diabetes 25 (17.0) 19 (20.7) 6 (22.2) GN c 49 (33.3) 28 (30.4) 20 (21.7) Urological 9 (6.1) 6 (6.5) 3 (5.5) Other 13 (8.8) 7 (7.6) 7 (12.7) Unknown 37 (25.2) 25 (27.2) 12 (21.8) Type of allograft, n (%) Deceased donor 69 (46.9) 40 (43.5) 29 (52.7) 0.50 Living donor 67 (45.6) 44 (47.8) 23 (41.8) SPK 11 (7.5) 8 (8.7) 3 (5.5) Graft number, n (%) First 127 (86.4) 86 (93.5) 41 (74.5) Second or more 20 (13.6) 6 (6.5) 14 (25.5) Induction, n (%) Alemtuzumab 113 (76.9) 73 (79.3) 40 (72.7) 0.47 IL-2RA 34 (23.1) 19 (20.7) 15 (27.3) Sensitization, n (%) Nonsensitized 102 (69.4) 75 (81.5) 27 (49.1) d G0.001 Sensitized 26 (17.7) 17 (18.5) 9 (16.4) Preformed DSA 19 (12.9) 0 (0) 19 (34.5) HLA mismatch, meantsd 3.44T T T Time to rejection (mo), median (95% CI) 6.70 (4.91Y8.26) 6.45 (4.62Y8.84) 6.87 (4.46Y8.47) 0.69 Follow-up (yr), meantsd 2.74T T T a Comparison between pure and mixed groups. b Adult polycystic kidney disease. c Glomerulonephritis. d For analysis preformed DSA classified as sensitized. CI, confidence interval; DSA, donor-specific antibody; IL-2RA, interleukin 2 receptor agonist; SD, standard deviation; SPK, simultaneous pancreas and kidney transplant. glomerulitis cases compared with 18 (14.8%) of 122 patients without glomerulitis (P=0.30). We also found no correlation between DSA and glomerulitis, with 7 (28.0%) of 25 patients being DSA+ compared with 35 (28.5%) of 122 patients without glomerulitis having DSA (P=0.92). Immunohistochemical staining for CD3 + and CD68 + cells revealed that, of the 25 cases, 8 (32.2%) consisted predominantly of CD3 + cells, 6 (24.0%) consisted of CD68 + cells, and 11 (44.0%) had both CD3 + and CD68 + cells. Only patients with CD68 + cells were at risk of developing TG, with a TG-free survival in the glomerulitisnegative, CD3 +,CD68 +,andcd3 + /CD68 + groups being 90.0%, 87.5%, 44.4%, and 48.6%, respectively (P=0.0004; Fig. 2). Of the 6 CD68 + glomerulitis cases, 1 (16.7%) had diffuse C4d and 5 (83.3%) had DSA. CD68 + glomerulitis was strongly associated with DSA when compared with glomerulitis-negative or CD3 + glomerulitis patients who had 35 (29.2%) of 120 and 2 (25.0%) of 8 DSA positivity, respectively (P=0.009). Of the 147 patients, 8 (5.4%) had peritubular capillaritis (PTC+). Only one of these patients subsequently lost his/her graft; therefore, any statistical inference is not possible. However, PTC+ was associated with further risk of TCMR, with a subsequent TCMR-free survival of 49.8% and 33.3% in the PTCj and PTC+ groups, respectively (P=0.048), and an AMR-free survival of 86.9% and 87.5% in the PTCj and PTC+ groups, respectively (P=0.74). Capillaritis was associated with the increased risk of subsequent TG, with a TG-free survival of 82.1% and 70.0% in the PTCj and PTC+ groups, respectively (P=0.012).

4 436 Transplantation & Volume 97, Number 4, February 27, 2014 Outcomes by Allograft Function There was no difference in allograft function before or at thetimeofindexbiopsybetweenthepureandmixedgroups. The mean best/baseline serum creatinine was 143.5T77.6 Kmol/L in the pure group and 171.3T144.2 Kmol/L in the mixed group (P=0.13). There was also no difference in allograft function after index biopsy between the two groups, not accounting for failed allografts, as shown in the Supplemental Data (see SDC, The proportion of patients with a significant interstitial fibrosis or tubular atrophy (ci and/or ct score Q2) was not different between the mixed and the pure groups, with 21 (22.8%) of the 92 pure group and 11 (20.0%) of the 55 mixed group having a ct and/or ci score of 2 or higher (P=0.62) on the index biopsy. FIGURE 2. Transplant glomerulopathy (TG)Yfree survival by infiltrating cell type. CD68 + glomerulitis was associated with an increased risk of TG. TG-free survival in the glomerulitis-free, CD3 +, CD68 +, and combined CD3 + / CD68 + glomerulitis was 90.0%, 87.5%, 44.4%, and 48.6% (PG0.001). Multivariable Analysis of Factors Associated With a Poor Prognosis in TCMR Table 2 shows the multivariate analysis of factors associated with adverse allograft outcomes after TCMR. C4d+ DSA+TCMR was significantly associated with allograft loss (OR=4.89, 95% CI=1.99Y11.99, PG0.001). Risk of subsequent morphological AMR was associated with C4d+DSA+ TCMR (OR=11.68, 95% CI=2.91Y46.85, PG0.001), DSA at the time of TCMR (OR=4.35, 95% CI=1.12Y16.90, P=0.035), and arteritis (OR=9.38, 95% CI=2.44Y36.03, P=0.001). Whereas factors at the time of index TCMR associated with the risk of developing TG included diffuse C4d (OR=4.81, 95% CI=1.17Y19.80, P=0.031) and glomerulitis (OR=6.61, 95% CI=2.24Y19.54, PG0.001). Analysis was repeated to incorporate subsequent DSA detection and AMR episodes along with the clinical findings at the time of index TCMR. TG was found to be associated with glomerulitis (OR=8.87, 95% CI=2.77Y28.47, PG0.001), DSA positivity at any time point (OR=4.11, 95% CI=1.23Y13.77, P=0.02), and AMR (OR=8.05, 95% CI=2.13Y30.43, P=0.002). AMR after TCMR was associated with allograft failure (OR=12.99, 95% CI=5.49Y30.74, PG0.001). FIGURE 1. Allograft survival by DSA and C4d status at the time of index TCMR episode. (A) patients with DSAs were at an increased risk of subsequent allograft failure: allograft survival was 72.6% and 30.5% in the DSA-negative and DSA-positive groups, respectively (P=0.021). (B) patients with diffuse C4d with histologic TCMR were at an increased risk of subsequent allograft failure, with allograft survival in the diffuse C4d, focal C4d, and C4d-negative groups being 33.3%, 64.5%, and 73.1%, respectively (P=0.033). DSA, donor-specific antibody; TCMR, T-cellYmediated rejection. Comparison of Pure TCMR, C4d+ and/or DSA+ TCMR, Pure AMR, and Coexisting Morphological AMR and TCMR As a final analysis, we compared the allograft survival after pure TCMR and diffuse C4d+ and/or DSA+ TCMR with a control group of patients who had histologic evidence of pure AMR or coexisting morphological AMR plus TCMR on a first rejection episode. After the index biopsy, allograft survival was 71.9%, 29.1%, 55.2%, and 32.1% in the pure TCMR, diffuse C4d and/or DSA+ TCMR, pure AMR, and coexisting AMR plus TCMR groups, respectively (PG0.001; Fig. 3). Patients with diffuse C4d and/or DSA+ TCMR had significantly worse allograft survival than patients with pure TCMR (P=0.027). There was a trend toward worse graft

5 * 2014 Lippincott Williams & Wilkins Willicombe et al. 437 TABLE 2. Multivariate analysis of the factors associated with allograft failure, morphological AMR, and TG after TCMR according to DSA status and histologic findings at the time of index rejection episode Allograft failure AMR TG Variable OR (95% CI) P OR (95% CI) P OR (95% CI) P C4d+DSA+ a 4.89 (1.99Y11.99) G (2.91Y46.85) G0.001 V NS DSA+ TCMR V NS 4.35 (1.12Y16.90) V NS Diffuse C4d V NS V NS 4.81 (1.17Y19.80) Arteritis V NS 9.38 (2.44Y36.03) V NS Glomerulitis V NS V NS 6.61 (2.24Y19.54) G0.001 a Includes focal and diffuse C4d. V, factors not retained in the model; AMR, antibody-mediated rejection; CI, confidence interval; DSA, donor-specific antibody; IL-2RA, interleukin 2 receptor agonist; NS, not significant; OR, odds ratio; SD, standard deviation; TCMR, T-cellYmediated rejection; TG, transplant glomerulopathy. survival in the pure AMR group when compared with the diffuse C4d and/or DSA+ TCMR group, but this failed to reach significance (P=0.08). Patients with coexisting AMR plus TCMR had inferior survival than the diffuse C4d and/ or DSA+ TCMR group (P=0.0001). Regarding the cellular component of the patients with TCMR, there was a significantly higher proportion of severe (QBanff II) cellular rejection episodes in the group with coexisting AMR plus TCMR. Seven (70.0%) of 10 patients with coexisting AMR plus TCMR had a severe cellular component compared with 6 (13.6%) of the 44 diffuse C4d and/or DSA+ TCMR group and 12 (13.0%) of the 92 pure TCMR group (PG0.001). FIGURE 3. Comparison of allograft survival after pure TCMR, TCMR with a humoral component, pure AMR, and coexisting AMR and TCMR. After the index biopsy, allograft survival was 71.9%, 29.1%, 55.2%, and 32.1% in the pure TCMR, diffuse C4d and/or DSA+ TCMR, pure AMR, and coexisting AMR plus TCMR groups, respectively (PG0.001). Patients with diffuse C4d and/or DSA+ TCMR had inferior allograft survival than patients with pure TCMR (P=0.027); there was no significant difference compared with patients with pure AMR (P=0.08) and superior survival compared with the coexisting AMR plus TCMR group (PG0.001). AMR, antibody-mediated rejection; DSA, donor-specific antibody; TCMR, T-cellYmediated rejection. DISCUSSION In our experience, we have found that TCMR with a humoral component is associated with inferior allograft outcomes when compared with pure TCMR. This has important clinical implications, and patients who have rejection with the morphological appearance of TCMR should be investigated for a humoral element, allowing management to be tailored accordingly. We have shown that morphological TCMR with DSA and/or diffuse C4d positivity is associated with AMR, TG, and allograft failure. As such, patients with evidence of alloantibody with morphological TCMR might benefit from a more aggressive immunotherapy directed at the antibody to improve outcomes. There have been few studies that have included data on TCMR biopsies with a humoral component that have incorporated DSA detection using sensitive luminex methods. In the early landmark C4d studies, Feucht et al. (12) showed that C4d-positive rejection strongly correlated with allograft loss, with the majority of grafts having combined histologic features of cellular rejection and what was then described as vascular rejection. Nickeleit et al. (13), in their analysis of 398 allograft biopsies, found that 43% of biopsies with interstitial rejection were C4d positive. In their series, more cases of C4d-positive rejection (regardless of histologic features) were treated with antilymphocytic agents, and as a result, they found no difference in clinical outcomes between the C4d-positive and -negative groups. Mauiyyedi et al. (4) found that the outcomes of patients with C4d+ TCMR and also C4d positivity with mixed morphological features of AMR and TCMR correlated closely with pure C4d+ AMR. In more recent studies, Al-Aly et al. (5) showed that it is not uncommon for patients to exhibit mixed morphological features of TCMR and AMR, with inferior outcomes to the pure forms of either category. Similarly, Everly et al. (7) found that mixed TCMR episodes were associated with an intermediate allograft survival between the pure TCMR and AMR groups, and Kayler et al. (14) established that even patients with focal C4d and TCMR might have underlying humoral involvement leading to steroid unresponsiveness. TCMR is generally considered a reversible lesion, with few cases being considered as steroid unresponsive (15, 16). Historically, studies have demonstrated that the prognosis of allografts with TCMR is dependent on arterial involvement. Nickeleit et al. (15) looked specifically at arterial

6 438 Transplantation & Volume 97, Number 4, February 27, 2014 lesions in patients undergoing TCMR and found that those patients with vascular involvement were more likely to be steroid unresponsive, and in particular, all those with fibrinoid necrosis lost their grafts within a year. Haas et al. (16) compared the outcomes of patients with TCMR and moderate arteritis against a group with TCMR and severe arteritis and found that patients in the latter category had a worse prognosis. More recent studies employing the use of microarrays have shown that pure TCMR can be easily reversed regardless of vascular involvement as long as it is not complicated by AMR (17Y19). Traditionally, glomerular inflammation is considered to be a lesion of AMR and has been previously shown to be an independent poor prognostic feature (11, 20Y22). Others have found that only monocytic glomerulitis is associated with C4d positivity and AMR, with T cells being the predominant cell in C4d-negative glomerulitis (11, 21). In an indepth study of glomerulitis, Batal et al. (9) found that glomerulitis could be detected in 47% of DSAjC4djTCMR and 62% of C4djDSAj borderline rejection episodes. They further defined the impact of grading glomerulitis, with higher g scores being more likely to be associated with C4d and DSA (9). Sis et al. (23) found that microcirculation injury, both glomerulitis and capillaritis, may be found in TCMR, and in the absence of DSA, the prognosis was good. We found that 18.4% of TCMR episodes had glomerulitis, and the prognosis was related to the cell type. It should be highlighted that the majority of our patients received alemtuzumab induction, and it has been reported that rejection episodes after alemtuzumab may consist of a higher proportion of C4d+ cases. In a retrospective descriptive study, Zhang et al. (24) found that 47.2% of ACR cases after alemtuzumab were C4d+. It has also been reported that the infiltrate after alemtuzumab induction in pure TCMR consists mainly of monocytes and that such cases respond well to corticosteroid therapy (25). We did not establish an increased risk of mixed rejection in our alemtuzumab-treated patients compared with the IL-2 receptor antagonist (IL-2RA)Yinduced patients. We have shown that, in our center s experience, patients with TCMR with a humoral component have inferior allograft outcomes when compared with pure TCMR, a finding that has therapeutic implications. Given the relationship between the humoral and cellular components of the immune system, the classic division of the histologic categories may be an oversimplification. It may be more appropriate for morphological TCMR with a humoral component to be treated in a more aggressive manner similar to AMR, which might improve long-term outcomes. MATERIALS AND METHODS Patients We retrospectively analyzed all patients who had received a kidney transplant at Imperial College Kidney and Transplant Centre between 2005 and We identified all patients who had an indicative biopsy that demonstrated histologic features of TCMR. We excluded patients receiving an antibody-incompatible transplant, defined as either blood group incompatibility or those patients with a positive CDC or T-cell FCXM. We did include those patients with DSA detected by luminex-only before transplant. We further excluded from our analysis patients found to have evidence of coexisting BK or pyelonephritis on their index biopsy, surveillance biopsies, and patients with previous biopsy-proven AMR. Immunosuppression All patients received monoclonal antibody induction with either alemtuzumab (Campath-1H; Cambridge, MA) 30 mg perioperatively, daclizumab (Zenax; Roche, Inc., NJ) 2 mg/kg on days 0 and 14 or basiliximab (Simulect; Novartis Pharma Corp, NJ) 20 mg on days 0 and 4, as described previously (26). Maintenance immunosuppression in the alemtuzumab group consisted of tacrolimus (FK) monotherapy. Patients induced with an IL-2RA received FK with the addition of mycophenolate mofetil. All patients received 1 week of corticosteroids only. Histopathology and C4d Staining All biopsy specimens were examined by light microscopy and were classified using the Banff 07 Classification of Renal Allograft Pathology (8). For analysis purposes, TCMR cases were categorized as borderline (BL), grade I (Banff IA and IB), grade II (Banff IIA and IIB), or grade III (Banff III). BL cases suspicious for TCMR were included in our cohort if treated. Classification of pure TCMR and mixed TCMR considered in this studied is shown below: 1. Pure TCMR: C4djDSAj morphological TCMR 2. TCMR with a humoral component/mixed rejection: a. C4d+DSAj morphological TCMR b. C4djDSA+ morphological TCMR c. C4d+DSA+ morphological TCMR AMR was defined as the presence of glomerulitis and/or peritubular capillaritis and/or thrombotic microangiopathy with C4d positivity and DSA. Coexisting AMR plus TCMR was defined as a C4d+DSA+ biopsy with morphological AMR plus TCMR by Banff criteria. TG was defined as the presence of double contours by light microscopy, with the absence of immune complex deposition by immunofluorescence and/or electron microscopy. Immunohistochemistry for C4d was performed on paraffin sections from all biopsies, C4d was classified as negative (PTC staining G10%), focal (11%Y50%) or diffuse (950%). A g score of 1 or higher was considered positive for glomerulitis, and immunohistochemical studies were carried out on positive cases to determine the presence of CD3 and CD68 cells. A ptc score of 1 or higher was considered positive for peritubular capillaritis. Detection of Anti-HLA DSAs Patients were routinely screened for DSA before and after transplant at 1, 3, 6, and 12 months and then at 12 monthly intervals thereafter or when clinically indicated using LABScreen mixed beads (One Lambda, Canoga Park, CA). Patients with a positive screen had the specificity of their anti- HLA antibody identified using LABScreen single-antigen beads. It is laboratory protocol to type for HLA antigens ja, jb, jcw, jdr, and YDQ. HLA DP was typed when clinically relevant. Patients were considered to have DSA+ TCMR when they had their DSA detected before the rejection episode or up to 10 days after rejection. A mean fluorescence index greater than 300 was considered positive. Treatment of TCMR In patients who received alemtuzumab induction, TCMR was treated with mycophenolate mofetil and oral corticosteroids (30 mg/d weaned to 10 mg by 3 months and continued thereafter) T 3500 mg methylprednisolone. Patients induced with an IL-2RA were treated with corticosteroids (Tmethylprednisolone). At the discretion of the transplant team, some patients received 2 g/kg intravenous immunoglobulin (Vigam; Bio Products Laboratory, Hertfordshire, UK) or alemtuzumab. There was no difference in the treatment methods between the pure and mixed groups: 30 (54.5%) of the 55 mixed group and 56 (60.9%) of the 92 pure group were treated with conventional corticosteroids alone (P=0.56).

7 * 2014 Lippincott Williams & Wilkins Willicombe et al. 439 Statistical Analysis All analyses were performed using MedCalc version Comparisons of means and frequencies of normally distributed variables were calculated using t-tests and W 2 /Fisher exact tests. Kaplan-Meier survival analysis was used to calculate the time of event from index biopsy, and statistical significance was determined by log-rank testing. Cox proportional regression plots were used for multivariable analyses, and variables with a significance level of PG0.1 on univariate analysis were included in the multivariable analysis using a stepwise method selection. PG0.05 was deemed statistically significant. ACKNOWLEDGMENTS The authors thank the work contributed by the transplant clinic staff, the Leslie Brent laboratory, and the Histocompatibility and Immunogenetics laboratory staff. REFERENCES 1. El-Zoghby ZM, Stegall MD, Lager DL, et al. Identifying specific causes of kidney allograft loss. Am J Transplant 2009; 9: Meier-Kriesche HU, Schold JD, Srinivas TR, et al. Lack of improvement in renal allograft survival despite a marked decrease in acute rejection rates over the most recent era. Am J Transplant 2004; 4: Willicombe M, Roufosse C, Brookes P, et al. Antibody-mediated rejection after alemtuzumab induction: incidence, risk factors, and predictors of poor outcome. Transplantation 2011; 92: Mauiyyedi S, Crespo M, Collins AB, et al. Acute humoral rejection in kidney transplantation: II. Morphology, immunopathology, and pathologic classification. J Am Soc Nephrol 2002; 13: Al-Aly Z, Reddivari V, Moiz A, et al. Renal allograft biopsies in the era of C4d staining: the need for change in the Banff classification system. Transpl Int 2008; 21: Nickeleit V, Andreoni K. The classification and treatment of antibody-mediated renal allograft injury: where do we stand? Kidney Int 2007; 71: Everly MJ, Everly JJ, Arend LJ, et al. Reducing de novo donor-specific antibody levels during acute rejection diminishes renal allograft loss. Am J Transplant 2009; 9: Solez K, Colvin RB, Racusen LC, et al. Banff 07 classification of renal allograft pathology: updates and future directions. Am J Transplant 2008; 8: Batal I, Lunz JG, Aggarwal N, et al. A critical appraisal of methods to grade transplant glomerulitis in renal allograft biopsies. Am J Transplant 2010; 10: Magil AB. Infiltrating cell types in transplant glomerulitis: relationship to peritubular capillary C4d deposition. Am J Kidney Dis 2005; 45: Fahim T, Bohmig GA, Exner M, et al. The cellular lesion of humoral rejection: predominant recruitment of monocytes to peritubular and glomerular capillaries. Am J Transplant 2007; 7: Feucht HE, Schneeberger H, Hillebrand G, et al. Capillary deposition of C4d complement fragment and early renal graft loss. Kidney Int 1993; 43: Nickeleit V, Zeiler M, Gudat F, et al. Detection of the complement degradation product C4d in renal allografts: diagnostic and therapeutic implications. J Am Soc Nephrol 2002; 13: Kayler LK, Kiss L, Sharma V, et al. Acute renal allograft rejection: diagnostic significance of focal peritubular capillary C4d. Transplantation 2008; 85: Nickeleit V, Vamvakas EC, Pascual M, et al. The prognostic significance of specific arterial lesions in acute renal allograft rejection. JAm Soc Nephrol 1998; 9: Haas M, Kraus ES, Samaniego-Picota M, et al. Acute renal allograft rejection with intimal arteritis: histologic predictors of response to therapy and graft survival. Kidney Int 2002; 61: Mueller TF, Einecke G, Reeve J, et al. Microarray analysis of rejection in human kidney transplants using pathogenesis-based transcript sets. Am J Transplant 2007; 7: Reeve J, Einecke G, Mengel M, et al. Diagnosing rejection in renal transplants: a comparison of molecular- and histopathology-based approaches. Am J Transplant 2009; 9: Famulski KS, Einecke G, Sis B, et al. Defining the canonical form of T- cellymediated rejection in human kidney transplants. Am J Transplant 2010; 10: Papadimitriou JC, Drachenberg CB, Munivenkatappa R, et al. Glomerular inflammation in renal allografts biopsies after the first year: cell types and relationship with antibody-mediated rejection and graft outcome. Transplantation 2010; 90: Magil AB, Tinckam K. Monocytes and peritubular capillary C4d deposition in acute renal allograft rejection. Kidney Int 2003; 63: Einecke G, Sis B, Reeve J, et al. Antibody-mediated microcirculation injury is the major cause of late kidney transplant failure. Am J Transplant 2009; 9: Sis B, Jhangri GS, Riopel J, et al. A new diagnostic algorithm for antibody-mediated microcirculation inflammation in kidney transplants. Am J Transplant 2012; 12: Zhang PL, Malek SK, Blasick TM, et al. C4d positivity is often associated with acute cellular rejection in renal transplant biopsies following Campath-1H (Alemtuzumab) induction. Ann Clin Lab Sci 2007; 37: Zhang PL, Malek SK, Prichard JW, et al. Acute cellular rejection predominated by monocytes is a severe form of rejection in human renal recipients with or without Campath-1H (alemtuzumab) induction therapy. Am J Transplant 2005; 5: Chan K, Taube D, Roufosse C, et al. Kidney transplantation with minimized maintenance: alemtuzumab induction with tacrolimus monotherapyvan open label, randomized trial. Transplantation 2011; 92: 774.