- Email: [email protected]
Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan
Human Immunology xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
www.ashi-hla.org
journal homepage: www.elsevier.com/locate/humimm
Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan Robert P. Carroll M.D., Ph.D. a,b,c,d,⇑, Michael Riceman a,b, Christopher M. Hope Ph.D. a,b, Anna Zeng a,b, Sue Deayton B(Sc) c, Greg D. Bennett B(Sc) c, P.T. Coates M.D., Ph.D. a,b,c a
Central Northern Adelaide Renal and Transplantation Services (CNARTS), Adelaide, Australia The University of Adelaide, Department of Medicine, Adelaide, Australia c Australian Red Cross Blood Service (ARCBS), Adelaide, Australia d University South Australia, Adelaide, Australia b
a r t i c l e
i n f o
Article history: Received 2 March 2016 Revised 8 August 2016 Accepted 13 August 2016 Available online xxxx Keywords: Antibodies Biomarker Rejection Renal Transplantation
a b s t r a c t Angiotensin II type 1 antibodies (AT1Rab) can mediate antibody mediated rejection (AMR). Pre transplant AT1Rab levels, and risk of rejection were assessed in Kidney Transplant Recipients (KTR) transplanted in our centre from 2013 to 2014 (n = 145). 14/145 (9.7%) KTR experienced antibody mediated rejection (AMR). The Hazard Ratio for AMR = 3.7 [95% CI 2–26] (p = 0.009) for KTR with AT1Rab levels >17.5 U/ ml. 6/11 of KTR with levels >25 U/ml experienced AMR. In 2015 (n = 80) KTR were transplanted and 6/80 KTR experienced rejection (2 AMR and 4 TCMR with vascular lesions). 7/80 of KTR had AT1Rab 17.5–25 U/ml and none experienced rejection and were induced with ATG and candesartan. 7/80 had AT1Rab 25–40 U/ml and received pre and post-operative plasma exchange, ATG and candesartan and 1/7 experienced TCMR with a vascular lesion. This perioperative regimen may alter the risk of rejection in patients with high levels of AT1Ab and further studies are needed. Ó 2016 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.
1. Introduction Transplantation across a negative complement dependent cytotoxicity (CDC) cross-match, in the presence of anti-Human Leucocyte Antigen (HLA) Donor Specific Antibodies (DSA) is associated with inferior graft survival and antibody mediated rejection (AMR) compared to transplantation where DSA are not present [1–3]. However, acute humoral rejection or AMR may occur in HLA identical transplantation [4] and it has been proposed that
Abbreviations: AMR, antibody mediated rejection; ATG, anti-thymocyte globulin; AT1R, angiotensin II type 1 receptor; AT1Rab, angiotensin II type 1 receptor antibody; AUC, area under the curve; CDC, complement dependent cytotoxicity; DSA, donor specific antibody; HLA, human leukocyte antigen; KTR, kidney transplant recipient; MFI, mean fluorescence index; ROC, receiver operator characteristics; TCMR, cell mediated rejection. ⇑ Corresponding author at: Central Northern Adelaide Renal and Transplantation Services (CNARTS), Level 9, East Wing, Royal Adelaide Hospital, Adelaide 5000, Australia. E-mail address: [email protected] (R.P. Carroll).
non-HLA antibodies, such as anti-angiotensin II type-1 receptor antibodies (AT1Rab) may mediate AMR. The observation of AT1Rab causing acute humoral rejection with a vascular component was first described by Dragun et al. in 2005 [5] and they have now been associated with any rejection type and graft failure independent of HLA mismatch or DSA [5–8]. The angiotensin II type-1 receptor (AT1R) is a G-protein coupled receptor that can be activated by angiotensin II and can be stimulated by agonistic AT1Rab [5]. AT1Rab activate the AT1R and the downstream effects are to activate nuclear factor-jb, inducing an inflammatory response [5]. It has been hypothesised that ischaemia-reperfusion injury increases the expression of donor AT1R on vascular smooth-muscle and endothelial cells, predisposing the graft to injury by pre-existing AT1Rab [9]. Binding of an AT1R antagonist such as a sartan to AT1R changes its conformation and prevents the binding of AT1Rab. This is the rationale for using sartans together with plasma exchange and Antithymocyte globulin (ATG) to treat AMR mediated by AT1Rab.
http://dx.doi.org/10.1016/j.humimm.2016.08.009 0198-8859/Ó 2016 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009
2
R.P. Carroll et al. / Human Immunology xxx (2016) xxx–xxx
We retrospectively analysed Kidney Transplant Recipients (KTR) from our centre to determine whether the presence of pretransplant AT1Rab was associated with AMR and TCMR as define by Banff Classifications [10].
2. Materials and methods From 2013 to 2014 there were 163 KTR transplanted at our centre. Eighteen KTR were excluded because they received preemptive plasma exchange and IVIG to achieve a negative crossmatch after DSA (MFI >4000) were detected (Fig. 1). The remainder (n = 145) were transplanted across a negative T and B cells cross match and in 16/145 (11%) patients there was at least one DSA (MFI 1500-4000) that did not require the KTR to undergo preemptive plasma exchange and therefore were included in the study. 9 of these 16 sensitized KTR received ATG induction and the remainder together with DSA negative patients received antiIL-2R alpha antibodies. During the period between 2013 and 2014 there were 44 biopsy proven rejection episodes, including 30 TCMR and 14 AMR (Fig. 1). There were 5 KTR with AMR that concurrently had sufficient interstitial infiltrate and tubulitis scores to fulfil TCMR criteria (Banff 13), these KTR were classified as AMR when analysing AT1R effects on AMR and TCMR. Although a negative value in the ELISA used to measure AT1Rab is defined as <2.5 U/ml literature suggests KTR with, AT1Rab levels >10 U/ml can be considered at increased risk of any rejection and KTR with levels >17.5 U/ml are consider at higher risk of AMR. Based on the high levels of AMR and TCMR with vascular lesions (Banff 2A/2B) in patients with high levels of AT1Rab in the 2013– 2014 cohort, we instituted a change in practice for those patients transplanted in 2015. All patients with AT1Rab >17.5 U/ml were treated with three doses of ATG (total dose 3–4.5 mg/kg) instead of our standard unit policy of anti-IL-2R alpha antibodies and were given pre and post operative candesartan (between 4 and 16 mg per day as tolerated). For KTR with AT1Rab >25 U/ml the above procedure was followed with the addition of 1 plasma volume, plasma exchange pre-op and two plasma exchanges post op. In 2015 (85 patients were transplanted), three patients with AT1Rab 17.5, 20 and 21 U/ml were not treated per protocol: two
163 KTR
18 Excluded (Desensised)
44 Rejecon Episodes
145 Included
7 Sub-clinical Rejecon
94 Non Rejecon
30 T Cell Mediated Rejecon (TCMR) 14 Anbody Mediated Rejecon (AMR) Fig. 1. Flow diagram of Kidney Transplant Recipient (KTR) Included and Excluded from study: There were 163 KTR entries between 2013 and 2014, 18 of these KTR were excluded as they underwent desensitisation, which included plasma exchange and IVIG treatments. There were 44 who rejected, 7 who had subclinical rejection and 94 who did not have a rejection episode. Of those who rejected there were 30 Cell Mediated Rejections (TCMR) as classified with Banff09 criteria and 14 Antibody Mediated Rejection (AMR) episodes classified with Banff13criteria.
developed AMR; and one vascular rejection. 2 patients received Eculizumab (AT1Rab 19 and 6 U/ml) and neither experienced rejection. These 5 KTR were excluded from the subsequent analysis. DSA were assessed in all KTR in routine pre-transplant sera and at 14 and 28 days post transplant. In addition KTR experiencing rejection or acute graft dysfunction also had sera tested for de novo DSA. DSA were detected by first using a Lifecodes LifeScreen, Class I and Class II Identification Kit (Immucor Transplant Diagnostics, GA, USA) to identify any anti-HLA antibodies, if positive a secondary test using a Lifecodes single antigen (Class I and Class II) beads (Immucor Transplant Diagnostics) was performed to identify antibody specificity. Both these kits were performed following the manufacturers protocols. DSA positivity was determined by a Mean Fluorescence Index (MFI) P1500 as measured by a Luminex platform. Anti-AT1R antibodies were measured pre transplant and at time of rejection using an ELISA Immunoassay kit (One Lambda, Canoga Park, CA, USA) following published protocol (Reinsmoen NL, 2010) and performed by the Australian Red cross Blood service (ARCBS) in Adelaide.
3. Results 3.1. 2013–2014 historical observational cohort All patients underwent a protocol biopsy at creatinine nadir or for cause and were graded using Banff criteria 2009 for TCMR and 2013 for AMR. 44/145 (30%) KTR experienced a rejection episode, consisting of 14 AMR and 30 TCMR (Fig. 1). The KTR with AMR (n = 14) were more likely to; receive a live donor (p = 0.06), have had shorter cold ischemia time (p = 0.08), have had multiple previous transplants (p = 0.06) and have had
Table 1 Patient Demographics cohort 2013–2014. Demographics
Non-AMR
AMR (BANFF13)
Number, n Male Gender, n (%) Median Age (IQR) Multiple Transplants, n (%) BDD, n (%) Live Donor, n (%) DCD, n (%) AT1Rab, U/ml Any DSA (MFI >300), n (%) DSA (MFI 1500–4000), n (%) de Novo DSA (MFI 1500) HLA MM <2, n (%) HLA MM 3–4, n (%) HLA MM 5–6, n (%) Ischaemic Time, Median (IQR) ATG induction, n (%) DGF (>7 days), n (%) CMR, n (%) Sub-clinical rejection, n (%) C4d Positive, n (%)
131 87 (66) 51 (41–61) 14 (11) 97 (74) 19 (15) 14 (11) 8(0–40) 26 (20) 14 (11) 7 (5) 25 (19) 40 (31) 66 (50) 12 (3–34) 8 (6) 41 (31) 30 (23) 7 (5) 3 (2)
14 9 (64) 54 (42–60) 4 (29) 9 (64) 5 (36) 0 (0) 15(6–40) 2 (14) 2 (14) 0 (0) 2 (14) 5 (36) 7 (50) 6 (4–15) 1 (7) 6 (43) 5 (36) 0 (0) 2 (14)
ESRF ADPKD, n (%) Alports, n (%) Diabetes, n (%) Hypertension, n (%) IgA Nephropathy, n (%) Reflux, n (%) Other, n (%)
20 (15) 6 (5) 25 (19) 10 (8) 21 (16) 8 (6) 41 (31)
5 0 2 3 2 0 2
(36) (0) (14) (21) (14) (0) (14)
IQR = interquartile range, BDD = brain dead donor, DCD = donations after cardiac death, ATRab = Angiotensin II type 1 receptor antibody, DSA = donor specific antibody, MFI = mean fluorescence index, HLA = human leukocyte antigen, MM = missmatches, ATG = anti-thymocyte globulin, DGF = delayed graft function, CMR = cell mediated rejection, ADPKD = autosomal dominant polycystic kidney disease.
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009
R.P. Carroll et al. / Human Immunology xxx (2016) xxx–xxx
3
adult polycystic kidney disease (p = 0.06) compared to non-AMR KTR (n = 131, Table 1) When AT1Rab levels in Banff defined rejection groups are compared, those 14 KTR with AMR had the highest median value (15.4 U/ml), which was significantly higher than KTR with no rejection episodes (7.9 U/ml, p < 0.001), Fig. 2. Levels of AT1Rab levels in those experiencing any rejection generally fell from a median of 24 U/ml pre transplant to 13 U/ml at the time of rejection before therapy for rejection was commenced. The ability of AT1Rab to predict AMR and TCMR was assessed in Receiver Operator Characteristics (ROC) curves. The AMR ROC curve had an Area Under the Curve (AUC) = 0.79, p < 0.001 with a sensitivity of 43% and specificity of 96% at value of >25 U/ml (Fig. 3A). The TCMR ROC curve had AUC 0.63, p = 0.019 with sensitivity 26% and specificity 88% at 17.5 U/ml (Fig. 3B). Time to AMR stratified by AT1Rab levels is shown in Fig. 4. Importantly, no KTR with >17.5 U/ml had detectable DSA, with only 1 developing de novo DSA. There were only 2 AMR episodes in those KTR with DSA (>1500) i.e. 2/14 and these 2 KTR had AT1Rab levels of 9 U/ml and 16 U/ml. 4. Prospective intervention cohort There were no immunological or demographic differences between the 2013–14 and 2015 cohorts other than the following: ATG induction increased from 6% to 25% (p < 0.001) as ATG was part of the intervention; rejection rates fell from 27% to 8% (p = 0.004); and rejection in KTR >17.5 U/ml fell from 58% to 7% (p = 0.002). See Table 2 and Fig. 5. There were slight differences in the composition of concomitant DSA and ATR1ab levels between the cohorts (see Table 3). Of those who experienced AMR: both were C4d positive on biopsy and had HLA DSA 900 and 4000 MFI with AT1Rab 16 and 6 U/ml respectively; both required ATG and PEX to terminate the rejection. The other 4 episodes of TCMR with a vascular lesion had i and t scores (1–3) with ptc and g scores (0). They had AT1Rab of <2.5, <2.5, 6, and 40 U/ml with no HLA DSA pre or post transplantation. 7 patients with AT1Rab >25 U/ml were given the agreed protocol and 1 KTR developed TCMR with a vascular lesion after the initial course of ATG and PEX and required addition doses of ATG/PEX to terminate the rejection episode. An additional 7 KTR with levels 17.5–25 U/ml were given ATG and candesartan and none developed rejection. Hyperkalaemia requiring haemodialysis was not found in those treated with candesartan. Delayed graft function (lack of spontaneous fall in serum creatinine within 7 days and/or the require-
Fig. 3. Receiver operator characteristics curve for pre-transplant angiotensin II type 1 receptor levels predicting antibody mediated rejection post-transplant in 145 Kidney Transplant Recipients (KTR) 2013–2014 cohort: A ROC with accuracy of 79% (p < 0.001) in determining post-transplant rejection with pre-transplant AT1Rab. The trade-off value was >25 U/ml with sensitivity of 43% and specificity of 96%.
ment for haemodialysis within 7 days) was similar between those treated with candesartan 4/14 (29%) and KTR not treated with candesartan (excluding rejection and DCD groups) 13/71 (18%) (p = 0.465 Fisher’s Exact Test). In spite of CMV prophylaxis, CMV tissue invasive disease occurred in 4/14 KTR with >17.5 U/ml AT1Rab treated with ATG. 3 of these KTR received a CMV positive graft and were CMV negative at the time of transplant. All became CMV viraemia negative when treated with treatment dose valganciclovir. An additional 2 patients were transiently viraemic. 1/14 of patients was transiently BK viraemic.
Fig. 2. Pre-transplant anti-angiotensin II type-1 receptor antibody (AT1RAb) levels in Kidney Transplant Recipients (KTR) 2013–2014 cohort: Pre-transplant AT1RAb levels were quantitated in 145 KTR; 44 KTR who suffered rejection episodes; 14 KTR with Antibody Mediated Rejection (AMR) and 4 KTR in this group also had concurrent Cell Mediated Rejection (TCMR), 30 KTR with TCMR and 7 KTR subclinical borderline rejection, and 94 KTR with no rejection episodes. Those 14 KTR with AMR had a greater number of AT1Rab U/ml than the 94 KTR with no rejection (p < 0.01) using Kruskal-Wallis analysis.
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009
4
R.P. Carroll et al. / Human Immunology xxx (2016) xxx–xxx
prevalence of BK viraemia and or nephropathy was no different between the groups (p = 0.678). The incidence of CMV disease was therefore more common in those treated with ATG (p = 0.01). The prevalence of CMV+ (donor) to CMV (recipient) was different between the AT1Rab >17.5 U/ml group (35%) and the 60 KTR who did not received ATG (10%) and this reached statistical significance (p = 0.043 Fisher’s Exact Test). When assessing the CMV disease in only CMV+ (donor) to CMV (recipient) the prevalence between AT1Rab group 3/5(60%) was not statistically difference to the non ATG group 2/7 (29%) p = 0.558 Fisher’s Exact Test
5. Discussion Fig. 4. Cumulative incidence of antibody mediated rejection (AMR) in 145 Kidney Transplant Recipients (KTR) 2013–2014 cohort: KTR were stratified by antiangiotensin II type-1 receptor antibodies (AT1Rab) levels at; AT1Rab >25 U/ml (n = 12) dark grey line, AT1Rab >17.5 U/ml (n = 22) light grey, AT1Rab <17.5 U/ml (n = 123) black line. Those KTR with AT1Rab levels >17.5 U/ml had a Hazard Ratio [95% Confidence Interval of 3.7 [2–26] (p = 0.009) compared to those <17.5 U/ml. the 12 KTR with >25 U/ml had HR [95%C.I.] = 8 [12–501] (p < 0.001).
Table 2 Patient Demographics cohort 2015. Demographics
2013–2014
2015
Number, n Male Gender, n (%) Median Age (IQR) Multiple Transplants, n (%) BDD, n (%) Live Donor, n (%) DCD, n (%) ATRab >17.5 U/ml Any DSA (MFI >300), n (%) DSA (MFI 1500–4000), n (%) de Novo DSA (MFI 1500) HLA MM <2, n (%) HLA MM 3–4, n (%) HLA MM 5–6, n (%) Ischaemic Time, Median (IQR) ATG induction, n (%) DGF (>7 days), n (%) Any rejection, n (%) Sub-clinical rejection, n (%) C4d Positive, n (%) Rejection ATRab >17.5 U/ml
145 98 (68) 51 (41–61) 18 (12) 106 (73) 21 (14) 14 (10) 24 (17) 28 (19) 16 (11) 7 (5) 27 (19) 45 (31) 75 (50) 12 (3–34) 9 (6) 47 (32) 40 (27) 7 (5) 5 (3) 14 (58)
80 55 (69) 49 (40–60) 8 (10) 60 (75) 12 (15) 8 (10) 14(18) 16 (20) 3(4) 4 (5) 18 (23) 21(25) 41 (51) 10 (2–27) 20 (25) 22 (28) 6 (8) 0 (0) 2 (3) 1 (7)
ESRF ADPKD, n (%) Alports, n (%) Diabetes, n (%) Hypertension, n (%) IgA Nephropathy, n (%) Reflux, n (%) Other, n (%)
25 (17) 6 (4) 27 (19) 13 (9) 23 (16) 8 (6) 43 (30)
9 (11) 1 (1) 14 (13) 6 (8) 12 (15) 9 (11) 32 (40)
IQR = interquartile range, BDD = brain dead donor, DCD = donations after cardiac death, ATRab = Angiotensin II type 1 receptor antibody, DSA = donor specific antibody, MFI = mean fluorescence index, HLA = human leukocyte antigen, MM = missmatches, ATG = anti-thymocyte globulin, DGF = delayed graft function, ADPKD = autosomal dominant polycystic kidney disease. The bold values in the only statistically significant different results between the groups (p < 0.05).
Of the 6 KTR treated with ATG in 2015 for non AT1Rab reasons, there was one viraemia episode and one tissue invasive disease event. The remaining 2015 cohort not treated with ATG (n = 60) experienced two episodes of CMV tissue invasive disease (both in CMV positive donor to CMV negative recipients) and 5 episodes of viraemia. One (2%) experienced BK nephropathy and 9 (15%) experienced sustained viraemia >10,000 copies for >2 weeks. The
In our historical observational 2013–2014 cohort, we found that high levels of AT1Rab (>17.5 U/ml and >25 U/ml) pre transplant in the absence of HLA DSA significantly associated with the development of AMR (Table 1 and Fig. 4). Although all patients were tested at 14 or 28 post transplant we only subsequently tested for de novo HLA DSA at time of clinical rejection or deterioration in graft function. Within the confines of this antibody screening policy we did not find evidence of de novo HLA DSA in KTR with AMR associated with high levels of AT1Rab. Whereas the levels of de novo DSA were 5% in those with any rejection episode or clinical deterioration without high (>17.5 U/ml) AT1Rab. Compared to other centres, the prevalence of AMR in our historical 2013–2014 cohort may seem high. However over half of KTR had 5 or 6 HLA antigen mismatches, 20% had a DSA at the time of transplantation and >10% were regrafts. Half of KTR who developed AMR (n = 12) had AT1Rab levels >25 U/ml. This indicates that, at least in part, that pre-transplant AT1Rab levels can identify KTR at risk of developing AMR despite having negative T or B cell crossmatch and no detectable pre-transplant or de novo DSA. However, there were also KTR with low levels of AT1Rab who developed AMR (n = 6). There were two KTR with >40 U/ml who experienced no rejection (Fig. 2) and have stable graft function at 12 months with no proteinuria who are not taking candesartan. Although stratification of immunological risk by AT1Rab predicts risks of AMR (Fig. 4), there are only small proportions (8– 15%) of KTR within these relatively high levels of 17.5 U/ml and 25 U/ml (Fig. 2 and Fig. 5). Almost half of all AMR occurs in those KTR with levels of 5–15 U/ml. However given that 60% of KTR with >25 U/ml experienced AMR and 30% with >17.5 U/ml experienced AMR, and needed prolonged courses of ATG and PEX to terminate the humoral rejection [11] we altered the our induction protocol to utilise ATG/candesartan and or PEX for transplants occurring in 2015. As shown in Table 2 other than ATG/candesartan use there were no major pre transplant demographic/immunological differences between the two cohorts other than the fall in both cellular and humoral rejection episodes (compare Figs. 2 and 5). We did not find any increase in hypotensive of hyperkalaemia events but the doses of candesartan were modified on a daily basis such that all patients continued to received 4–16 mg per day. There was no increased risk of delayed graft function in those treated with candesartan. So although the introduction of ATG may have reduced rejection rates, unfortunately, CMV tissue invasive disease was more prevalent in AT1Rab >17.5 U/ml treated with ATG compared to KTR not treated with ATG. However this observation is probably explained by the higher than expected prevalence of CMV+ donor to CMV recipient in the AT1Rab group. Limitations of this single centre study include the fact that we did not measure other minor HLA antibodies or other non HLA anti
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009
R.P. Carroll et al. / Human Immunology xxx (2016) xxx–xxx
5
Fig. 5. Pre-transplant anti-angiotensin II type-1 receptor antibody (AT1RAb) levels in Kidney Transplant Recipients (KTR) in 2015 cohort: Pre-transplant AT1RAb levels were quantitated in 80 KTR; 6 KTR who suffered rejection episodes; 2 KTR with Antibody Mediated Rejection (AMR), 4 KTR with vascular rejection (Banff 2A/2B) and 74 KTR with no rejection episodes.
Table 3 Distribution of donor specific antibodies (DSA) with Angiotensin Receptor Antiboidy levels in patients transplanted in the 2013–2014 and 2015 cohorts. Demographics
2013–2014
2015
Number, n
145
80
DSA +ve ATRab, n (%) <10 U/ml 10–17.5 U/ml >17.5 U/ml
20/76 (21) 8/18 (31) 0/23 (0)
10/43 (19) 3/10 (23) 3/11 (21)
ATRab = Angiotensin II type 1 receptor antibody, DSA = donor specific antibody, MFI = mean fluorescence index. Table showing the prevalence of any measure DSA MFI >300 units stratified by ATRab level. There was no statistically significant difference in the distribution of ATRab and DSA between the historical 2013–14 cohort compared to the 2015 cohort apart from the DSA+ in the >17.5 units groups which was more common in 2015 than 2013–14 (p = 0.047 Fischer’s Exact Test).
bodies such as MICA and MICB which have been shown to associated with humoral rejection. With relatively small cohorts there can be year to year variation in immunological risk and indeed although there were no statistically significant differenced in the rates of HLA DSA in terms of MFI, the MFI’s were higher in the 2013–14 cohort compared to 2015 cohort, but this did not reach statistical significance (p = 0.078). Also as shown in Table 3 the distribution of DSA in the >17.5 unit AT1Rab was different between the two cohorts but this observation may result from relatively small numbers in these sub groups. Year to year variation in rejection could in part explain our findings but according to the ANZDATA the rates of rejection over the last 15 years at our centre have remained between 25% and 30% per annum and a fall to 8% in 2015 is a departure from our centre’s long term average. Although future studies are needed it would seem from these data that AT1Rab risk stratification, peri operative candesartan and PEX can modify patient outcomes at least in the short term follow up of this study.
Acknowledgements The Jacquot Research Establishment Award from the Australian and New Zealand Society of Nephrology, The Mark Cocks Award from Teamlife Australia. References [1] M. El Ters, J. Grande, M. Keddis, E. Rodrigo, B. Chopra, P. Dean, M. Stegall, F. Cosio, Kidney allograft survival after acute rejection, the value of follow-up biopsies, Am. J. Transplant. 13 (2013) 2334. [2] C. Lefaucheur, A. Loupy, D. Vernerey, J.P. Duong-Van-Huyen, C. Suberbielle, D. Anglicheau, J. Vérine, T. Beuscart, D. Nochy, P. Bruneval, D. Charron, M. Delahousse, J.P. Empana, G. Hill, D. Glotz, C. Legendre, X. Jouven, Antibodymediated vascular rejection of kidney allografts: a population-based study, The Lancet 381 (2013) 313. [3] S. Fidler, A. Irish, W. Lim, P. Ferrari, C. Witt, F. Christiansen, Pre-transplant donor specific anti-HLA antibody is associated with antibody-mediated rejection, progressive graft dysfunction and patient death, Transpl. Immunol. 28 (2013) 148. [4] G. Opelz, Non-HLA transplantation immunity revealed by lymphocytotoxic antibodies, The Lancet 365 (2005) 1570. [5] D. Dragun, D. Müller, J.H. Bräsen, L. Fritsche, M. Nieminen-Kelhä, R. Dechen, U. Kintscher, B. Rudolph, J. Hoebeke, D. Eckert, I. Mazak, R. Plehm, C. Schönemann, T. Unger, K. Budde, H.H. Neumayer, F. Luft, G. Wallukat, Angiotensin II type 1receptor activating antibodies in renal-allograft rejection, N. Engl. J. Med. 352 (2005) 558. [6] M. Giral, Y. Foucher, A. Dufay, et al., Pretransplant sensitization against angiotensin II type 1 receptor is a risk factor for acute rejection and graft loss, Am. J. Transpl. Off. J. Am. Soc. Transpl. Am. Soc. Transpl. Surg. 13 (10) (2013) 2567. [7] M. Banasik, M. Boratynska, K. Koscielska-Kasprzak, et al., The influence of nonHLA antibodies directed against angiotensin II type 1 receptor (AT1R) on early renal transplant outcomes, Transpl. Int. Offic. J. Eur. Soc. Organ Transpl. 27 (10) (2014) 1029. [8] M. Taniguchi, L.M. Rebellato, J. Cai, et al., Higher risk of kidney graft failure in the presence of anti-angiotensin II type-1 receptor antibodies, Am. J. Transpl. Off. J. Am. Soc. Transpl. Am. Soc. Transpl. Surg. 13 (10) (2013) 2577. [9] N. Reinsmoen, Role of angiotensin II type 1 receptor-activating antibodies in solid organ transplantation, Hum. Immunol. 74 (2013) 1474. [10] M. Haas, B. Sis, L.C. Racusen, et al., Banff 2013 meeting report: inclusion of c4dnegative antibody-mediated rejection and antibody-associated arterial lesions, Am. J. Transpl. Off. J. Am. Soc. Transpl. Am. Soc. Transpl. Surg. 14 (2) (2014) 272. [11] A. Fuss, C.M. Hope, S. Deayton, et al., C4d negative antibody mediated rejection with high anti-angiotensin II type I receptor antibodies in absence of donor specific antibodies, Nephrology (2015).
Disclosure The authors of this manuscript declare no conflicts of interest.
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009
Contents lists available at ScienceDirect
www.ashi-hla.org
journal homepage: www.elsevier.com/locate/humimm
Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan Robert P. Carroll M.D., Ph.D. a,b,c,d,⇑, Michael Riceman a,b, Christopher M. Hope Ph.D. a,b, Anna Zeng a,b, Sue Deayton B(Sc) c, Greg D. Bennett B(Sc) c, P.T. Coates M.D., Ph.D. a,b,c a
Central Northern Adelaide Renal and Transplantation Services (CNARTS), Adelaide, Australia The University of Adelaide, Department of Medicine, Adelaide, Australia c Australian Red Cross Blood Service (ARCBS), Adelaide, Australia d University South Australia, Adelaide, Australia b
a r t i c l e
i n f o
Article history: Received 2 March 2016 Revised 8 August 2016 Accepted 13 August 2016 Available online xxxx Keywords: Antibodies Biomarker Rejection Renal Transplantation
a b s t r a c t Angiotensin II type 1 antibodies (AT1Rab) can mediate antibody mediated rejection (AMR). Pre transplant AT1Rab levels, and risk of rejection were assessed in Kidney Transplant Recipients (KTR) transplanted in our centre from 2013 to 2014 (n = 145). 14/145 (9.7%) KTR experienced antibody mediated rejection (AMR). The Hazard Ratio for AMR = 3.7 [95% CI 2–26] (p = 0.009) for KTR with AT1Rab levels >17.5 U/ ml. 6/11 of KTR with levels >25 U/ml experienced AMR. In 2015 (n = 80) KTR were transplanted and 6/80 KTR experienced rejection (2 AMR and 4 TCMR with vascular lesions). 7/80 of KTR had AT1Rab 17.5–25 U/ml and none experienced rejection and were induced with ATG and candesartan. 7/80 had AT1Rab 25–40 U/ml and received pre and post-operative plasma exchange, ATG and candesartan and 1/7 experienced TCMR with a vascular lesion. This perioperative regimen may alter the risk of rejection in patients with high levels of AT1Ab and further studies are needed. Ó 2016 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.
1. Introduction Transplantation across a negative complement dependent cytotoxicity (CDC) cross-match, in the presence of anti-Human Leucocyte Antigen (HLA) Donor Specific Antibodies (DSA) is associated with inferior graft survival and antibody mediated rejection (AMR) compared to transplantation where DSA are not present [1–3]. However, acute humoral rejection or AMR may occur in HLA identical transplantation [4] and it has been proposed that
Abbreviations: AMR, antibody mediated rejection; ATG, anti-thymocyte globulin; AT1R, angiotensin II type 1 receptor; AT1Rab, angiotensin II type 1 receptor antibody; AUC, area under the curve; CDC, complement dependent cytotoxicity; DSA, donor specific antibody; HLA, human leukocyte antigen; KTR, kidney transplant recipient; MFI, mean fluorescence index; ROC, receiver operator characteristics; TCMR, cell mediated rejection. ⇑ Corresponding author at: Central Northern Adelaide Renal and Transplantation Services (CNARTS), Level 9, East Wing, Royal Adelaide Hospital, Adelaide 5000, Australia. E-mail address: [email protected] (R.P. Carroll).
non-HLA antibodies, such as anti-angiotensin II type-1 receptor antibodies (AT1Rab) may mediate AMR. The observation of AT1Rab causing acute humoral rejection with a vascular component was first described by Dragun et al. in 2005 [5] and they have now been associated with any rejection type and graft failure independent of HLA mismatch or DSA [5–8]. The angiotensin II type-1 receptor (AT1R) is a G-protein coupled receptor that can be activated by angiotensin II and can be stimulated by agonistic AT1Rab [5]. AT1Rab activate the AT1R and the downstream effects are to activate nuclear factor-jb, inducing an inflammatory response [5]. It has been hypothesised that ischaemia-reperfusion injury increases the expression of donor AT1R on vascular smooth-muscle and endothelial cells, predisposing the graft to injury by pre-existing AT1Rab [9]. Binding of an AT1R antagonist such as a sartan to AT1R changes its conformation and prevents the binding of AT1Rab. This is the rationale for using sartans together with plasma exchange and Antithymocyte globulin (ATG) to treat AMR mediated by AT1Rab.
http://dx.doi.org/10.1016/j.humimm.2016.08.009 0198-8859/Ó 2016 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009
2
R.P. Carroll et al. / Human Immunology xxx (2016) xxx–xxx
We retrospectively analysed Kidney Transplant Recipients (KTR) from our centre to determine whether the presence of pretransplant AT1Rab was associated with AMR and TCMR as define by Banff Classifications [10].
2. Materials and methods From 2013 to 2014 there were 163 KTR transplanted at our centre. Eighteen KTR were excluded because they received preemptive plasma exchange and IVIG to achieve a negative crossmatch after DSA (MFI >4000) were detected (Fig. 1). The remainder (n = 145) were transplanted across a negative T and B cells cross match and in 16/145 (11%) patients there was at least one DSA (MFI 1500-4000) that did not require the KTR to undergo preemptive plasma exchange and therefore were included in the study. 9 of these 16 sensitized KTR received ATG induction and the remainder together with DSA negative patients received antiIL-2R alpha antibodies. During the period between 2013 and 2014 there were 44 biopsy proven rejection episodes, including 30 TCMR and 14 AMR (Fig. 1). There were 5 KTR with AMR that concurrently had sufficient interstitial infiltrate and tubulitis scores to fulfil TCMR criteria (Banff 13), these KTR were classified as AMR when analysing AT1R effects on AMR and TCMR. Although a negative value in the ELISA used to measure AT1Rab is defined as <2.5 U/ml literature suggests KTR with, AT1Rab levels >10 U/ml can be considered at increased risk of any rejection and KTR with levels >17.5 U/ml are consider at higher risk of AMR. Based on the high levels of AMR and TCMR with vascular lesions (Banff 2A/2B) in patients with high levels of AT1Rab in the 2013– 2014 cohort, we instituted a change in practice for those patients transplanted in 2015. All patients with AT1Rab >17.5 U/ml were treated with three doses of ATG (total dose 3–4.5 mg/kg) instead of our standard unit policy of anti-IL-2R alpha antibodies and were given pre and post operative candesartan (between 4 and 16 mg per day as tolerated). For KTR with AT1Rab >25 U/ml the above procedure was followed with the addition of 1 plasma volume, plasma exchange pre-op and two plasma exchanges post op. In 2015 (85 patients were transplanted), three patients with AT1Rab 17.5, 20 and 21 U/ml were not treated per protocol: two
163 KTR
18 Excluded (Desensised)
44 Rejecon Episodes
145 Included
7 Sub-clinical Rejecon
94 Non Rejecon
30 T Cell Mediated Rejecon (TCMR) 14 Anbody Mediated Rejecon (AMR) Fig. 1. Flow diagram of Kidney Transplant Recipient (KTR) Included and Excluded from study: There were 163 KTR entries between 2013 and 2014, 18 of these KTR were excluded as they underwent desensitisation, which included plasma exchange and IVIG treatments. There were 44 who rejected, 7 who had subclinical rejection and 94 who did not have a rejection episode. Of those who rejected there were 30 Cell Mediated Rejections (TCMR) as classified with Banff09 criteria and 14 Antibody Mediated Rejection (AMR) episodes classified with Banff13criteria.
developed AMR; and one vascular rejection. 2 patients received Eculizumab (AT1Rab 19 and 6 U/ml) and neither experienced rejection. These 5 KTR were excluded from the subsequent analysis. DSA were assessed in all KTR in routine pre-transplant sera and at 14 and 28 days post transplant. In addition KTR experiencing rejection or acute graft dysfunction also had sera tested for de novo DSA. DSA were detected by first using a Lifecodes LifeScreen, Class I and Class II Identification Kit (Immucor Transplant Diagnostics, GA, USA) to identify any anti-HLA antibodies, if positive a secondary test using a Lifecodes single antigen (Class I and Class II) beads (Immucor Transplant Diagnostics) was performed to identify antibody specificity. Both these kits were performed following the manufacturers protocols. DSA positivity was determined by a Mean Fluorescence Index (MFI) P1500 as measured by a Luminex platform. Anti-AT1R antibodies were measured pre transplant and at time of rejection using an ELISA Immunoassay kit (One Lambda, Canoga Park, CA, USA) following published protocol (Reinsmoen NL, 2010) and performed by the Australian Red cross Blood service (ARCBS) in Adelaide.
3. Results 3.1. 2013–2014 historical observational cohort All patients underwent a protocol biopsy at creatinine nadir or for cause and were graded using Banff criteria 2009 for TCMR and 2013 for AMR. 44/145 (30%) KTR experienced a rejection episode, consisting of 14 AMR and 30 TCMR (Fig. 1). The KTR with AMR (n = 14) were more likely to; receive a live donor (p = 0.06), have had shorter cold ischemia time (p = 0.08), have had multiple previous transplants (p = 0.06) and have had
Table 1 Patient Demographics cohort 2013–2014. Demographics
Non-AMR
AMR (BANFF13)
Number, n Male Gender, n (%) Median Age (IQR) Multiple Transplants, n (%) BDD, n (%) Live Donor, n (%) DCD, n (%) AT1Rab, U/ml Any DSA (MFI >300), n (%) DSA (MFI 1500–4000), n (%) de Novo DSA (MFI 1500) HLA MM <2, n (%) HLA MM 3–4, n (%) HLA MM 5–6, n (%) Ischaemic Time, Median (IQR) ATG induction, n (%) DGF (>7 days), n (%) CMR, n (%) Sub-clinical rejection, n (%) C4d Positive, n (%)
131 87 (66) 51 (41–61) 14 (11) 97 (74) 19 (15) 14 (11) 8(0–40) 26 (20) 14 (11) 7 (5) 25 (19) 40 (31) 66 (50) 12 (3–34) 8 (6) 41 (31) 30 (23) 7 (5) 3 (2)
14 9 (64) 54 (42–60) 4 (29) 9 (64) 5 (36) 0 (0) 15(6–40) 2 (14) 2 (14) 0 (0) 2 (14) 5 (36) 7 (50) 6 (4–15) 1 (7) 6 (43) 5 (36) 0 (0) 2 (14)
ESRF ADPKD, n (%) Alports, n (%) Diabetes, n (%) Hypertension, n (%) IgA Nephropathy, n (%) Reflux, n (%) Other, n (%)
20 (15) 6 (5) 25 (19) 10 (8) 21 (16) 8 (6) 41 (31)
5 0 2 3 2 0 2
(36) (0) (14) (21) (14) (0) (14)
IQR = interquartile range, BDD = brain dead donor, DCD = donations after cardiac death, ATRab = Angiotensin II type 1 receptor antibody, DSA = donor specific antibody, MFI = mean fluorescence index, HLA = human leukocyte antigen, MM = missmatches, ATG = anti-thymocyte globulin, DGF = delayed graft function, CMR = cell mediated rejection, ADPKD = autosomal dominant polycystic kidney disease.
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009
R.P. Carroll et al. / Human Immunology xxx (2016) xxx–xxx
3
adult polycystic kidney disease (p = 0.06) compared to non-AMR KTR (n = 131, Table 1) When AT1Rab levels in Banff defined rejection groups are compared, those 14 KTR with AMR had the highest median value (15.4 U/ml), which was significantly higher than KTR with no rejection episodes (7.9 U/ml, p < 0.001), Fig. 2. Levels of AT1Rab levels in those experiencing any rejection generally fell from a median of 24 U/ml pre transplant to 13 U/ml at the time of rejection before therapy for rejection was commenced. The ability of AT1Rab to predict AMR and TCMR was assessed in Receiver Operator Characteristics (ROC) curves. The AMR ROC curve had an Area Under the Curve (AUC) = 0.79, p < 0.001 with a sensitivity of 43% and specificity of 96% at value of >25 U/ml (Fig. 3A). The TCMR ROC curve had AUC 0.63, p = 0.019 with sensitivity 26% and specificity 88% at 17.5 U/ml (Fig. 3B). Time to AMR stratified by AT1Rab levels is shown in Fig. 4. Importantly, no KTR with >17.5 U/ml had detectable DSA, with only 1 developing de novo DSA. There were only 2 AMR episodes in those KTR with DSA (>1500) i.e. 2/14 and these 2 KTR had AT1Rab levels of 9 U/ml and 16 U/ml. 4. Prospective intervention cohort There were no immunological or demographic differences between the 2013–14 and 2015 cohorts other than the following: ATG induction increased from 6% to 25% (p < 0.001) as ATG was part of the intervention; rejection rates fell from 27% to 8% (p = 0.004); and rejection in KTR >17.5 U/ml fell from 58% to 7% (p = 0.002). See Table 2 and Fig. 5. There were slight differences in the composition of concomitant DSA and ATR1ab levels between the cohorts (see Table 3). Of those who experienced AMR: both were C4d positive on biopsy and had HLA DSA 900 and 4000 MFI with AT1Rab 16 and 6 U/ml respectively; both required ATG and PEX to terminate the rejection. The other 4 episodes of TCMR with a vascular lesion had i and t scores (1–3) with ptc and g scores (0). They had AT1Rab of <2.5, <2.5, 6, and 40 U/ml with no HLA DSA pre or post transplantation. 7 patients with AT1Rab >25 U/ml were given the agreed protocol and 1 KTR developed TCMR with a vascular lesion after the initial course of ATG and PEX and required addition doses of ATG/PEX to terminate the rejection episode. An additional 7 KTR with levels 17.5–25 U/ml were given ATG and candesartan and none developed rejection. Hyperkalaemia requiring haemodialysis was not found in those treated with candesartan. Delayed graft function (lack of spontaneous fall in serum creatinine within 7 days and/or the require-
Fig. 3. Receiver operator characteristics curve for pre-transplant angiotensin II type 1 receptor levels predicting antibody mediated rejection post-transplant in 145 Kidney Transplant Recipients (KTR) 2013–2014 cohort: A ROC with accuracy of 79% (p < 0.001) in determining post-transplant rejection with pre-transplant AT1Rab. The trade-off value was >25 U/ml with sensitivity of 43% and specificity of 96%.
ment for haemodialysis within 7 days) was similar between those treated with candesartan 4/14 (29%) and KTR not treated with candesartan (excluding rejection and DCD groups) 13/71 (18%) (p = 0.465 Fisher’s Exact Test). In spite of CMV prophylaxis, CMV tissue invasive disease occurred in 4/14 KTR with >17.5 U/ml AT1Rab treated with ATG. 3 of these KTR received a CMV positive graft and were CMV negative at the time of transplant. All became CMV viraemia negative when treated with treatment dose valganciclovir. An additional 2 patients were transiently viraemic. 1/14 of patients was transiently BK viraemic.
Fig. 2. Pre-transplant anti-angiotensin II type-1 receptor antibody (AT1RAb) levels in Kidney Transplant Recipients (KTR) 2013–2014 cohort: Pre-transplant AT1RAb levels were quantitated in 145 KTR; 44 KTR who suffered rejection episodes; 14 KTR with Antibody Mediated Rejection (AMR) and 4 KTR in this group also had concurrent Cell Mediated Rejection (TCMR), 30 KTR with TCMR and 7 KTR subclinical borderline rejection, and 94 KTR with no rejection episodes. Those 14 KTR with AMR had a greater number of AT1Rab U/ml than the 94 KTR with no rejection (p < 0.01) using Kruskal-Wallis analysis.
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009
4
R.P. Carroll et al. / Human Immunology xxx (2016) xxx–xxx
prevalence of BK viraemia and or nephropathy was no different between the groups (p = 0.678). The incidence of CMV disease was therefore more common in those treated with ATG (p = 0.01). The prevalence of CMV+ (donor) to CMV (recipient) was different between the AT1Rab >17.5 U/ml group (35%) and the 60 KTR who did not received ATG (10%) and this reached statistical significance (p = 0.043 Fisher’s Exact Test). When assessing the CMV disease in only CMV+ (donor) to CMV (recipient) the prevalence between AT1Rab group 3/5(60%) was not statistically difference to the non ATG group 2/7 (29%) p = 0.558 Fisher’s Exact Test
5. Discussion Fig. 4. Cumulative incidence of antibody mediated rejection (AMR) in 145 Kidney Transplant Recipients (KTR) 2013–2014 cohort: KTR were stratified by antiangiotensin II type-1 receptor antibodies (AT1Rab) levels at; AT1Rab >25 U/ml (n = 12) dark grey line, AT1Rab >17.5 U/ml (n = 22) light grey, AT1Rab <17.5 U/ml (n = 123) black line. Those KTR with AT1Rab levels >17.5 U/ml had a Hazard Ratio [95% Confidence Interval of 3.7 [2–26] (p = 0.009) compared to those <17.5 U/ml. the 12 KTR with >25 U/ml had HR [95%C.I.] = 8 [12–501] (p < 0.001).
Table 2 Patient Demographics cohort 2015. Demographics
2013–2014
2015
Number, n Male Gender, n (%) Median Age (IQR) Multiple Transplants, n (%) BDD, n (%) Live Donor, n (%) DCD, n (%) ATRab >17.5 U/ml Any DSA (MFI >300), n (%) DSA (MFI 1500–4000), n (%) de Novo DSA (MFI 1500) HLA MM <2, n (%) HLA MM 3–4, n (%) HLA MM 5–6, n (%) Ischaemic Time, Median (IQR) ATG induction, n (%) DGF (>7 days), n (%) Any rejection, n (%) Sub-clinical rejection, n (%) C4d Positive, n (%) Rejection ATRab >17.5 U/ml
145 98 (68) 51 (41–61) 18 (12) 106 (73) 21 (14) 14 (10) 24 (17) 28 (19) 16 (11) 7 (5) 27 (19) 45 (31) 75 (50) 12 (3–34) 9 (6) 47 (32) 40 (27) 7 (5) 5 (3) 14 (58)
80 55 (69) 49 (40–60) 8 (10) 60 (75) 12 (15) 8 (10) 14(18) 16 (20) 3(4) 4 (5) 18 (23) 21(25) 41 (51) 10 (2–27) 20 (25) 22 (28) 6 (8) 0 (0) 2 (3) 1 (7)
ESRF ADPKD, n (%) Alports, n (%) Diabetes, n (%) Hypertension, n (%) IgA Nephropathy, n (%) Reflux, n (%) Other, n (%)
25 (17) 6 (4) 27 (19) 13 (9) 23 (16) 8 (6) 43 (30)
9 (11) 1 (1) 14 (13) 6 (8) 12 (15) 9 (11) 32 (40)
IQR = interquartile range, BDD = brain dead donor, DCD = donations after cardiac death, ATRab = Angiotensin II type 1 receptor antibody, DSA = donor specific antibody, MFI = mean fluorescence index, HLA = human leukocyte antigen, MM = missmatches, ATG = anti-thymocyte globulin, DGF = delayed graft function, ADPKD = autosomal dominant polycystic kidney disease. The bold values in the only statistically significant different results between the groups (p < 0.05).
Of the 6 KTR treated with ATG in 2015 for non AT1Rab reasons, there was one viraemia episode and one tissue invasive disease event. The remaining 2015 cohort not treated with ATG (n = 60) experienced two episodes of CMV tissue invasive disease (both in CMV positive donor to CMV negative recipients) and 5 episodes of viraemia. One (2%) experienced BK nephropathy and 9 (15%) experienced sustained viraemia >10,000 copies for >2 weeks. The
In our historical observational 2013–2014 cohort, we found that high levels of AT1Rab (>17.5 U/ml and >25 U/ml) pre transplant in the absence of HLA DSA significantly associated with the development of AMR (Table 1 and Fig. 4). Although all patients were tested at 14 or 28 post transplant we only subsequently tested for de novo HLA DSA at time of clinical rejection or deterioration in graft function. Within the confines of this antibody screening policy we did not find evidence of de novo HLA DSA in KTR with AMR associated with high levels of AT1Rab. Whereas the levels of de novo DSA were 5% in those with any rejection episode or clinical deterioration without high (>17.5 U/ml) AT1Rab. Compared to other centres, the prevalence of AMR in our historical 2013–2014 cohort may seem high. However over half of KTR had 5 or 6 HLA antigen mismatches, 20% had a DSA at the time of transplantation and >10% were regrafts. Half of KTR who developed AMR (n = 12) had AT1Rab levels >25 U/ml. This indicates that, at least in part, that pre-transplant AT1Rab levels can identify KTR at risk of developing AMR despite having negative T or B cell crossmatch and no detectable pre-transplant or de novo DSA. However, there were also KTR with low levels of AT1Rab who developed AMR (n = 6). There were two KTR with >40 U/ml who experienced no rejection (Fig. 2) and have stable graft function at 12 months with no proteinuria who are not taking candesartan. Although stratification of immunological risk by AT1Rab predicts risks of AMR (Fig. 4), there are only small proportions (8– 15%) of KTR within these relatively high levels of 17.5 U/ml and 25 U/ml (Fig. 2 and Fig. 5). Almost half of all AMR occurs in those KTR with levels of 5–15 U/ml. However given that 60% of KTR with >25 U/ml experienced AMR and 30% with >17.5 U/ml experienced AMR, and needed prolonged courses of ATG and PEX to terminate the humoral rejection [11] we altered the our induction protocol to utilise ATG/candesartan and or PEX for transplants occurring in 2015. As shown in Table 2 other than ATG/candesartan use there were no major pre transplant demographic/immunological differences between the two cohorts other than the fall in both cellular and humoral rejection episodes (compare Figs. 2 and 5). We did not find any increase in hypotensive of hyperkalaemia events but the doses of candesartan were modified on a daily basis such that all patients continued to received 4–16 mg per day. There was no increased risk of delayed graft function in those treated with candesartan. So although the introduction of ATG may have reduced rejection rates, unfortunately, CMV tissue invasive disease was more prevalent in AT1Rab >17.5 U/ml treated with ATG compared to KTR not treated with ATG. However this observation is probably explained by the higher than expected prevalence of CMV+ donor to CMV recipient in the AT1Rab group. Limitations of this single centre study include the fact that we did not measure other minor HLA antibodies or other non HLA anti
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009
R.P. Carroll et al. / Human Immunology xxx (2016) xxx–xxx
5
Fig. 5. Pre-transplant anti-angiotensin II type-1 receptor antibody (AT1RAb) levels in Kidney Transplant Recipients (KTR) in 2015 cohort: Pre-transplant AT1RAb levels were quantitated in 80 KTR; 6 KTR who suffered rejection episodes; 2 KTR with Antibody Mediated Rejection (AMR), 4 KTR with vascular rejection (Banff 2A/2B) and 74 KTR with no rejection episodes.
Table 3 Distribution of donor specific antibodies (DSA) with Angiotensin Receptor Antiboidy levels in patients transplanted in the 2013–2014 and 2015 cohorts. Demographics
2013–2014
2015
Number, n
145
80
DSA +ve ATRab, n (%) <10 U/ml 10–17.5 U/ml >17.5 U/ml
20/76 (21) 8/18 (31) 0/23 (0)
10/43 (19) 3/10 (23) 3/11 (21)
ATRab = Angiotensin II type 1 receptor antibody, DSA = donor specific antibody, MFI = mean fluorescence index. Table showing the prevalence of any measure DSA MFI >300 units stratified by ATRab level. There was no statistically significant difference in the distribution of ATRab and DSA between the historical 2013–14 cohort compared to the 2015 cohort apart from the DSA+ in the >17.5 units groups which was more common in 2015 than 2013–14 (p = 0.047 Fischer’s Exact Test).
bodies such as MICA and MICB which have been shown to associated with humoral rejection. With relatively small cohorts there can be year to year variation in immunological risk and indeed although there were no statistically significant differenced in the rates of HLA DSA in terms of MFI, the MFI’s were higher in the 2013–14 cohort compared to 2015 cohort, but this did not reach statistical significance (p = 0.078). Also as shown in Table 3 the distribution of DSA in the >17.5 unit AT1Rab was different between the two cohorts but this observation may result from relatively small numbers in these sub groups. Year to year variation in rejection could in part explain our findings but according to the ANZDATA the rates of rejection over the last 15 years at our centre have remained between 25% and 30% per annum and a fall to 8% in 2015 is a departure from our centre’s long term average. Although future studies are needed it would seem from these data that AT1Rab risk stratification, peri operative candesartan and PEX can modify patient outcomes at least in the short term follow up of this study.
Acknowledgements The Jacquot Research Establishment Award from the Australian and New Zealand Society of Nephrology, The Mark Cocks Award from Teamlife Australia. References [1] M. El Ters, J. Grande, M. Keddis, E. Rodrigo, B. Chopra, P. Dean, M. Stegall, F. Cosio, Kidney allograft survival after acute rejection, the value of follow-up biopsies, Am. J. Transplant. 13 (2013) 2334. [2] C. Lefaucheur, A. Loupy, D. Vernerey, J.P. Duong-Van-Huyen, C. Suberbielle, D. Anglicheau, J. Vérine, T. Beuscart, D. Nochy, P. Bruneval, D. Charron, M. Delahousse, J.P. Empana, G. Hill, D. Glotz, C. Legendre, X. Jouven, Antibodymediated vascular rejection of kidney allografts: a population-based study, The Lancet 381 (2013) 313. [3] S. Fidler, A. Irish, W. Lim, P. Ferrari, C. Witt, F. Christiansen, Pre-transplant donor specific anti-HLA antibody is associated with antibody-mediated rejection, progressive graft dysfunction and patient death, Transpl. Immunol. 28 (2013) 148. [4] G. Opelz, Non-HLA transplantation immunity revealed by lymphocytotoxic antibodies, The Lancet 365 (2005) 1570. [5] D. Dragun, D. Müller, J.H. Bräsen, L. Fritsche, M. Nieminen-Kelhä, R. Dechen, U. Kintscher, B. Rudolph, J. Hoebeke, D. Eckert, I. Mazak, R. Plehm, C. Schönemann, T. Unger, K. Budde, H.H. Neumayer, F. Luft, G. Wallukat, Angiotensin II type 1receptor activating antibodies in renal-allograft rejection, N. Engl. J. Med. 352 (2005) 558. [6] M. Giral, Y. Foucher, A. Dufay, et al., Pretransplant sensitization against angiotensin II type 1 receptor is a risk factor for acute rejection and graft loss, Am. J. Transpl. Off. J. Am. Soc. Transpl. Am. Soc. Transpl. Surg. 13 (10) (2013) 2567. [7] M. Banasik, M. Boratynska, K. Koscielska-Kasprzak, et al., The influence of nonHLA antibodies directed against angiotensin II type 1 receptor (AT1R) on early renal transplant outcomes, Transpl. Int. Offic. J. Eur. Soc. Organ Transpl. 27 (10) (2014) 1029. [8] M. Taniguchi, L.M. Rebellato, J. Cai, et al., Higher risk of kidney graft failure in the presence of anti-angiotensin II type-1 receptor antibodies, Am. J. Transpl. Off. J. Am. Soc. Transpl. Am. Soc. Transpl. Surg. 13 (10) (2013) 2577. [9] N. Reinsmoen, Role of angiotensin II type 1 receptor-activating antibodies in solid organ transplantation, Hum. Immunol. 74 (2013) 1474. [10] M. Haas, B. Sis, L.C. Racusen, et al., Banff 2013 meeting report: inclusion of c4dnegative antibody-mediated rejection and antibody-associated arterial lesions, Am. J. Transpl. Off. J. Am. Soc. Transpl. Am. Soc. Transpl. Surg. 14 (2) (2014) 272. [11] A. Fuss, C.M. Hope, S. Deayton, et al., C4d negative antibody mediated rejection with high anti-angiotensin II type I receptor antibodies in absence of donor specific antibodies, Nephrology (2015).
Disclosure The authors of this manuscript declare no conflicts of interest.
Please cite this article in press as: R.P. Carroll et al., Angiotensin II type-1 receptor antibody (AT1Rab) associated humoral rejection and the effect of peri operative plasma exchange and candesartan, Hum. Immunol. (2016), http://dx.doi.org/10.1016/j.humimm.2016.08.009