© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Clin Transplant 2014: 28: 1234–1243 DOI: 10.1111/ctr.12451
Clinical Transplantation
Dynamics of anti-human leukocyte antigen antibodies after renal transplantation and their impact on graft outcome de Souza PS, David-Neto E, Panajotopolous N, Agena F, Rodrigues H, Ronda C, David DR, Kalil J, Nahas WC, de Castro MCR. Dynamics of anti-human leukocyte antigen antibodies after renal transplantation and their impact on graft outcome. Abstract: The purpose of this study was to sequentially monitor antiHLA antibodies and correlate the results with antibody-mediated rejection (AMR), graft survival (GS), and graft function (GF). We collected sera from 111 kidney transplant recipients on transplant days 0, 7, 14, 30, 60, 90, 180, and 360 and analyzed PRA levels by ELISA. DSAs were analyzed by single-antigen beads in rejecting kidneys. At pretransplant, 79.3% of the patients were non-sensitized (PRA = 0%) and 20.7% were sensitized (PRA > 1%). After transplant, patients were grouped by PRA profile: no anti-HLA antibodies pre- or post-transplant (group HLApre /post ; n = 80); de novo anti-HLA antibodies posttransplant (group HLApre /post+; n = 8); sensitized pre-transplant/ increased PRA post-transplant (group HLApre+/post↑; n = 9); and sensitized pre-transplant/decreased PRA post-transplant (group HLApre+/post↓; n = 14). De novo anti-HLA antibodies were detected at 7–180 d. In sensitized patients, PRA levels changed within the first 30 d post-transplant. Incidence of AMR was higher in HLApre /post+ and HLApre+/post↑ than in HLApre /post , and HLApre+/post↓ (p < 0.001) groups. One-yr death-censored GS was 36% in group HLApre+/post↑, compared with 98%, 88% and 100% in groups HLApre /post , HLApre /post+, and HLApre+/post↓, respectively (p < 0.001). Excluding first-year graft losses, GF and GS were similar among the groups. In conclusion, post-transplant antibody monitoring can identify recipients at higher risk of AMR.
Patrıcia Soares de Souzaa,b, Elias David-Netoa,b, Nicolas Panajotopolousc, Fabiana Agenaa,b, He´lcio Rodriguesc, Carla Rondac, Daısa Ribeiro Davidd, Jorge Kalilc, Wiliam Carlos Nahasb and Maria Cristina Ribeiro de Castroa,b a
Renal Transplantation Service of Urology Division, bNephrology Department, c Laboratory of Immunology of Instituto do ~o (InCor) and dPathology Division, Coracß a ~o Paulo, Sao Paulo Hospital das Clınicas de Sa University, Sao Paulo, Brazil Key words: antibody-mediated rejection – anti-HLA antibodies – graft function – graft survival Corresponding author: Maria Cristina Ribeiro Rodrigues Alves de Castro, MD, Rua Jose Sobrinho Alves, 150/222, Sao Paulo, Sao Paulo, Brazil. Tel.: +55 11 3672-7628; fax: +55 11 2661-7238; e-mail: [email protected] Conflict of interest: None. Accepted for publication 13 August 2014
Antibody-mediated rejection (AMR) is a major problem in kidney transplantation. Technological advances in anti-human leukocyte antigen (HLA) antibody detection and the discovery of complement activation on the surface of endothelial cells in biopsy samples (C4d deposition) obtained from transplanted kidneys promoted a better definition of acute and chronic AMR (1, 2). In addition, there is a growing body of evidence showing a causal relationship between HLA antibodies and graft outcomes (3, 4). Therefore, post-transplant screening for anti-HLA antibodies could be an important tool for monitoring transplant recipients (5–7). However, it remains unclear which patients should be screened, which techniques are more effective,
1234
as well as when and how often anti-HLA antibody monitoring should be performed. Enzyme-linked immunosorbent assay (ELISA), a solid-phase immunoassay, has been found to be a useful and cost-effective screening tool for the pre- and post-transplant detection and serial measurement of anti-HLA antibodies. The use of ELISA permits early detection of post-transplant HLA sensitization in patients who had no anti-HLA antibodies at transplantation and can reveal substantial changes in panel-reactive antibody (PRA) levels among sensitized patients (8). Although many studies had evaluated the role of post-transplant anti-HLA antibodies, few have investigated anti-HLA antibody dynamics in the
Anti-HLA antibodies and graft outcome first year after kidney transplantation using sera collected at multiple time points (6, 9, 10). The objective of this study was to monitor anti-HLA antibodies sequentially by ELISA and to correlate the results with acute rejection episodes, graft survival, and renal function in the short- and longterm follow-up in groups of patients with different pre- and post-transplant antibody profiles. We also aimed to identify the post-transplant time points at which antibody monitoring could provide the most useful information. Patients and methods
This was a prospective cohort study conducted from July 2005 to July 2006 and including 111 consecutive kidney transplant recipients treated at our transplant center. The study was approved by the local ethics committee, and all participating patients gave written informed consent. All donor-recipient pairs were ABO-compatible. Anti-human globulin-enhanced complementdependent cytotoxicity negative T-cell crossmatches and NIH complement-dependent cytotoxicity B-cell cross-matches were required for all kidney transplant recipients at the time of transplant. All donors and recipients were HLA-A, -B, and -DR typed by low-resolution polymerase chain reaction-single-strand polymorphisms (One Lambda, Canoga Park, CA, USA). We did not evaluate HLA-DP, -DQ, and -Cw specificities. We applied the following exclusion criteria: patients undergoing simultaneous transplantation of another organ; participating in another study; younger than 18 yr of age; and evolving to graft loss or death within the first 24 h post-transplant. In the majority of cases, the induction therapy consisted of anti-interleukin (IL)-2 receptor antibodies. Antithymocyte globulin (ATG) was given in cases of retransplantation or to sensitized patients with pre-transplant higher than 50%, as well as when cold ischemia time exceeded 24 h. None of the patients received pre-transplant plasmapheresis or monoclonal anti-CD20 antibodies. The maintenance immunosuppression regimens consisted of prednisone, mycophenolic acid, and tacrolimus. Eight patients received alternative regimens consisting of prednisone, rapamycin, and cyclosporine. Study design
To evaluate PRA levels, we collected sequential blood samples immediately before transplantation and on post-transplant days 7, 14, 30, 60, 90, 180, and 360.
Patients were classified, on the basis of their pretransplant PRA profile, as non-sensitized (PRA = 0%) or sensitized (PRA > 1%). Of the 111 patients evaluated, 88 (79.3%) were classified as non-sensitized, the remaining 23 (20.7%) were classified as sensitized. A significant post-transplant change in PRA reactivity was defined as a higher than 10% variation in the PRA level in at least two consecutive samples (11). PRA variations lower or equal to 10% were not confirmed in subsequent samples. On the basis of the PRA profile within first year after transplantation (Fig. 1), sensitized and non-sensitized patients were divided into four groups, according to the change in the PRA level, as follows: 1. HLApre /post , comprising patients who were non-sensitized before transplantation and did not develop HLA antibodies in the first year after transplantation (n = 80); 2. HLApre /post+, comprising patients who were non-sensitized before transplantation and developed de novo antibodies in the first year after transplantation (n = 8); 3. HLApre+/post↑, comprising patients who were sensitized before transplantation and whose PRA levels increased in the first year after transplantation (n = 9); 4. HLApre+/post↓, comprising patients who were sensitized before transplantation and whose PRA levels decreased in the first year after transplantation (n = 14). All acute rejection episodes were biopsy-proven (BPAR) and classified according to Banff’ 97 criteria, updated in 2003. AMR was diagnosed when allograft dysfunction and two of the following three criteria were present: histological damage, peritubular capillary C4d staining, and presence of circulating DSA (1, 12): Biopsies were for cause and not indicated by the results of this study. Episodes of pure T-cell-mediated rejection (TCMR) were treated with pulses of methylprednisolone (500 mg/d for three d) associated or not to Thymoglobulin (7 mg/kg for 7–10 d; Genzyme Corporation, Framingham, MA, USA). AMR was treated with Thymoglobulin associated or not to high dose of human intravenous immunoglobulin or plasmapheresis, upon availability. Therapeutic success was defined by recovery of renal function to previous levels. Graft function was evaluated by estimated glomerular filtration rate, calculated by the Modification of Diet in Renal Disease (MDRD) equation (13). Proteinuria was evaluated by determining the urinary protein-to-creatinine ratio. Both were eval-
1235
de Souza et al.
HLApre−/post− ΔPRA: Class I: 0% Class II: 0%
HLApre−/post+ ΔPRA: Class I: 14-85% Class II: 38-100%
HLApre+/post↑
HLApre+/post↓
ΔPRA: Class I: 40-75% Class II: 58-67%
ΔPRA: Class I: 15-67% Class II: 16-76%
Fig. 1. Dynamics of anti-HLA antibodies after kidney transplantation. Pre- and post-transplant (Tx) dynamics of anti-human leukocyte antigen (HLA) antibodies, by panel-reactive antibody (PRA) level. HLApre /post , no anti-HLA antibodies pre- or post-transplant (n = 80); HLApre /post+, de novo anti-HLA antibodies post-transplant (n = 8); HLApre+/post↑, sensitized pretransplant/increased anti-HLA antibodies post-transplant (n = 9); HLApre+/post↓, sensitized pre-transplant/decreased anti-HLA antibodies post-transplant (n = 14). DPRA: Variation of panel reactive antibodies between the serum samples.
uated at the end of the first, second, and third years post-transplant. Graft failure was defined as return to dialysis. The study endpoints were to evaluate the impact that pre- and post-transplant PRA-ELISA profiles had on the incidence of acute rejection episodes, graft failure, and graft function. Data related to antibody profile and acute rejection were analyzed at the end of first year post-transplant, whereas those related to renal function and graft survival were analyzed up through the third year after surgery. Anti-HLA antibody detection
All pre- and post-transplant serum samples (n = 926) were screened by PRA-ELISA (LAT-M; One Lambda) to identify anti-HLA class I and class II antibodies of the immunoglobulin G isotype. In positive sera, PRA levels, expressed as a percentage, were measured by ELISA for class I
1236
and class II reagents (LAT 1240; One Lambda). Both PRA-ELISA tests were performed according to the manufacturer’s protocols (14). A sensitive solid-phase assay, using purified HLA class I and/or class II single-antigen beads (Luminex platform; LABScreen; One Lambda), was used in two circumstances: (i) to confirm the presence or absence of DSA in sera of all patients presenting BPAR; (ii) to confirm that anti-HLA antibodies were not present before transplantation and were in fact de novo post-transplant antibodies (in HLApre /post+ group patients). Commercial kits (LABScreen Single-antigen LS1A04 and LS2A01; One Lambda) were used in accordance with the manufacturer protocols. In brief (15), 5 lL of class I and class II beads was mixed with 20 lL of sera and the mixture was incubated in the dark for 30 min at room temperature. After three washes, 100 lL of 1:100 fluorescein isothiocyanate-conjugated goat anti-human IgG was added
Anti-HLA antibodies and graft outcome and the samples were incubated under the same conditions for 30 min. The microbeads were then washed twice and resuspended in 80 lL of fixing solution. Negative and positive controls were tested in the same manner. The tests were read on a flow cytometer (LABScreen Luminex 100; One Lambda). A mean fluorescent intensity above 1000 U was defined as a positive reaction.
Therefore, the final study sample comprised 111 kidney transplant recipients (57 males and 54 females), with a mean age of 43.6 13.0 yr. Of those 111 patients, 40 (34%) received a deceaseddonor kidney and 71 (66%) received a living-donor kidney. Ninety-six patients (86.5%) received their first transplants. Clinical data were obtained from our service electronic database.
Histology and immunopathology
Post-transplant dynamics of anti-HLA antibodies
The BPAR were defined according to Banff’ 97 criteria, updated in 2003 (1). Samples were stained for C4d by immunofluorescence (16). Immunohistochemistry (17) was used only when the material was not available for immunofluorescence. The intensity and distribution of C4d staining in the peritubular capillaries were graded as follows: negative staining (0–10%); focal staining (11–50%); and diffuse staining (> 50%). Any staining higher than 10% of the peritubular capillaries was considered positive (18).
There were no statistically significant differences among the four groups in terms of the following pre-transplant characteristics: recipient age, recipient gender, donor type, number of pregnancies, and number of blood transfusions, HLA mismatches, or maintenance immunosuppression regimens. In comparison with the HLApre / post and HLApre /post+ groups, the HLApre+/ post↑ and HLApre+/post↓ groups showed a higher number of cases of retransplantation (p = 0.02) and more frequent use of ATG as induction therapy (p < 0.001). Clinical and demographic data are presented in Table 1. In the HLApre /post+ group (n = 8), 12.5% of the patients were retransplants, 62.5% had pre-tranplant blood transfusions, and 100% of the women had previous pregnancy. Both pre-transplant and historic sera of these patients were evaluated using single-antigen beads to exclude pre-transplant DSA in this group of patients.
Statistical analysis
Continuous variables were expressed as mean standard deviation or as median and range. For continuous variables, means were compared using Student’s t-test. Observations between groups were compared using Fisher’s exact test and the Kruskal–Wallis test. One-way analysis of variance was used to compare continuous variables. Graft survival was assessed using Kaplan–Meier analysis and compared by log-rank test. In logistic regression analyses, several factors were tested in terms of their impact on AMR and graft loss: age; gender; donor type; number of previous transplants and blood transfusions; induction therapy based on HLA-A, -B, and -DR mismatches; maintenance immunosuppression therapy; and post-transplant PRA profile. Values of p < 0.05 were considered statistically significant. Data were analyzed with the SPSS software (SPSS Inc., Chicago, IL, USA), version 16.0 for Windows. Results
One hundred and thirty-nine kidney transplants were performed at our facility during the study period. A total of 28 recipients were excluded from the analysis: for undergoing simultaneous transplantation of another organ (n = 8); for participating in another study (n = 4); for being younger than 18 yr of age (n = 4); for evolving to graft loss or death within the first 24 h post-transplant (n = 2); and for declining to participate in the study (n = 10).
Time to PRA change
The mean time to PRA change was 38 8 d (median 14 d, p = 0.276). In non-sensitized patients, de novo anti-HLA antibodies were detected between post-transplant days 7 and 180 (Fig. 2A) in a bimodal curve: in 50% of these patients, de novo antibodies appeared in the first 14 post-transplant days and in other 50% they appear after the 60th post-transplant day. In sensitized patients, PRA levels changed within the first 90 d post-transplant (Fig. 2B). Among HLApre+/post↑ group, seven of nine (78%) patients showed an increase in PRA level by day 14; among HLApre+/ post↓ group, decrease of PRA occurred after day 30 in nine of 14 (65%) patients. Acute rejection
The overall incidence of BPAR was 22% (24 of the 111 patients evaluated) and occurred at a median of 13 d after transplantation (range, 5 to 340 d). As can be seen in Table 2, the rate of BPAR was significantly different among the four
1237
de Souza et al. Table 1. Clinical and demographic characteristics of the 111 kidney transplant recipients evaluated, by group Groups
Characteristics Age (yr), mean SD Female gender, n (%) Deceased donor, n (%) First transplant, n (%) Previous pregnancies (≥1), n (%) Previous transfusions (≥1), n (%) HLA-AB mismatches, n 0 1–2 3–4 HLA-DR mismatches, n 0 1 2 Induction therapy; n (%) Anti-IL-2 receptor antibodies, n (%) Antithymocyte globulin Maintenance therapy with Tac+mycophenolate, n (%) Post-transplant blood transfusion, n (%)
HLApre /post (n = 80)
HLApre /post+ (n = 8)
HLApre+/post↑ (n = 9)
HLApre+/post↓ (n = 14)
p
43 14 36 (45) 25 (31) 73 (91) 25 (70) 48 (60)
47 5 3 (37.5) 5 (62.5) 7 (87.5) 3 (100) 5 (62.5)
44 11 7 (78) 4 (44.5) 5 (56) 7 (100) 8 (89)
43 9 8 (57) 6 (43) 11 (78.5) 7 (87.5) 12 (86)
0.938 0.198 0.261 0.020a 0.254 0.132
2 44 34
0 7 1
0 7 2
0 9 5
0.517
28 36 16
4 3 1
2 4 3
7 5 2
0.819
75 (94) 5 (6) 6 (43)
8 (100) 0 (0) 7 (87.5)
3 (33) 6 (67) 8 (89)
6 (43) 8 (57) 14 (100)
8 (57)
5 (62.5)
6 (67)
4 (29)
Clin Transplant 2014: 28: 1234–1243 DOI: 10.1111/ctr.12451
Clinical Transplantation
Dynamics of anti-human leukocyte antigen antibodies after renal transplantation and their impact on graft outcome de Souza PS, David-Neto E, Panajotopolous N, Agena F, Rodrigues H, Ronda C, David DR, Kalil J, Nahas WC, de Castro MCR. Dynamics of anti-human leukocyte antigen antibodies after renal transplantation and their impact on graft outcome. Abstract: The purpose of this study was to sequentially monitor antiHLA antibodies and correlate the results with antibody-mediated rejection (AMR), graft survival (GS), and graft function (GF). We collected sera from 111 kidney transplant recipients on transplant days 0, 7, 14, 30, 60, 90, 180, and 360 and analyzed PRA levels by ELISA. DSAs were analyzed by single-antigen beads in rejecting kidneys. At pretransplant, 79.3% of the patients were non-sensitized (PRA = 0%) and 20.7% were sensitized (PRA > 1%). After transplant, patients were grouped by PRA profile: no anti-HLA antibodies pre- or post-transplant (group HLApre /post ; n = 80); de novo anti-HLA antibodies posttransplant (group HLApre /post+; n = 8); sensitized pre-transplant/ increased PRA post-transplant (group HLApre+/post↑; n = 9); and sensitized pre-transplant/decreased PRA post-transplant (group HLApre+/post↓; n = 14). De novo anti-HLA antibodies were detected at 7–180 d. In sensitized patients, PRA levels changed within the first 30 d post-transplant. Incidence of AMR was higher in HLApre /post+ and HLApre+/post↑ than in HLApre /post , and HLApre+/post↓ (p < 0.001) groups. One-yr death-censored GS was 36% in group HLApre+/post↑, compared with 98%, 88% and 100% in groups HLApre /post , HLApre /post+, and HLApre+/post↓, respectively (p < 0.001). Excluding first-year graft losses, GF and GS were similar among the groups. In conclusion, post-transplant antibody monitoring can identify recipients at higher risk of AMR.
Patrıcia Soares de Souzaa,b, Elias David-Netoa,b, Nicolas Panajotopolousc, Fabiana Agenaa,b, He´lcio Rodriguesc, Carla Rondac, Daısa Ribeiro Davidd, Jorge Kalilc, Wiliam Carlos Nahasb and Maria Cristina Ribeiro de Castroa,b a
Renal Transplantation Service of Urology Division, bNephrology Department, c Laboratory of Immunology of Instituto do ~o (InCor) and dPathology Division, Coracß a ~o Paulo, Sao Paulo Hospital das Clınicas de Sa University, Sao Paulo, Brazil Key words: antibody-mediated rejection – anti-HLA antibodies – graft function – graft survival Corresponding author: Maria Cristina Ribeiro Rodrigues Alves de Castro, MD, Rua Jose Sobrinho Alves, 150/222, Sao Paulo, Sao Paulo, Brazil. Tel.: +55 11 3672-7628; fax: +55 11 2661-7238; e-mail: [email protected] Conflict of interest: None. Accepted for publication 13 August 2014
Antibody-mediated rejection (AMR) is a major problem in kidney transplantation. Technological advances in anti-human leukocyte antigen (HLA) antibody detection and the discovery of complement activation on the surface of endothelial cells in biopsy samples (C4d deposition) obtained from transplanted kidneys promoted a better definition of acute and chronic AMR (1, 2). In addition, there is a growing body of evidence showing a causal relationship between HLA antibodies and graft outcomes (3, 4). Therefore, post-transplant screening for anti-HLA antibodies could be an important tool for monitoring transplant recipients (5–7). However, it remains unclear which patients should be screened, which techniques are more effective,
1234
as well as when and how often anti-HLA antibody monitoring should be performed. Enzyme-linked immunosorbent assay (ELISA), a solid-phase immunoassay, has been found to be a useful and cost-effective screening tool for the pre- and post-transplant detection and serial measurement of anti-HLA antibodies. The use of ELISA permits early detection of post-transplant HLA sensitization in patients who had no anti-HLA antibodies at transplantation and can reveal substantial changes in panel-reactive antibody (PRA) levels among sensitized patients (8). Although many studies had evaluated the role of post-transplant anti-HLA antibodies, few have investigated anti-HLA antibody dynamics in the
Anti-HLA antibodies and graft outcome first year after kidney transplantation using sera collected at multiple time points (6, 9, 10). The objective of this study was to monitor anti-HLA antibodies sequentially by ELISA and to correlate the results with acute rejection episodes, graft survival, and renal function in the short- and longterm follow-up in groups of patients with different pre- and post-transplant antibody profiles. We also aimed to identify the post-transplant time points at which antibody monitoring could provide the most useful information. Patients and methods
This was a prospective cohort study conducted from July 2005 to July 2006 and including 111 consecutive kidney transplant recipients treated at our transplant center. The study was approved by the local ethics committee, and all participating patients gave written informed consent. All donor-recipient pairs were ABO-compatible. Anti-human globulin-enhanced complementdependent cytotoxicity negative T-cell crossmatches and NIH complement-dependent cytotoxicity B-cell cross-matches were required for all kidney transplant recipients at the time of transplant. All donors and recipients were HLA-A, -B, and -DR typed by low-resolution polymerase chain reaction-single-strand polymorphisms (One Lambda, Canoga Park, CA, USA). We did not evaluate HLA-DP, -DQ, and -Cw specificities. We applied the following exclusion criteria: patients undergoing simultaneous transplantation of another organ; participating in another study; younger than 18 yr of age; and evolving to graft loss or death within the first 24 h post-transplant. In the majority of cases, the induction therapy consisted of anti-interleukin (IL)-2 receptor antibodies. Antithymocyte globulin (ATG) was given in cases of retransplantation or to sensitized patients with pre-transplant higher than 50%, as well as when cold ischemia time exceeded 24 h. None of the patients received pre-transplant plasmapheresis or monoclonal anti-CD20 antibodies. The maintenance immunosuppression regimens consisted of prednisone, mycophenolic acid, and tacrolimus. Eight patients received alternative regimens consisting of prednisone, rapamycin, and cyclosporine. Study design
To evaluate PRA levels, we collected sequential blood samples immediately before transplantation and on post-transplant days 7, 14, 30, 60, 90, 180, and 360.
Patients were classified, on the basis of their pretransplant PRA profile, as non-sensitized (PRA = 0%) or sensitized (PRA > 1%). Of the 111 patients evaluated, 88 (79.3%) were classified as non-sensitized, the remaining 23 (20.7%) were classified as sensitized. A significant post-transplant change in PRA reactivity was defined as a higher than 10% variation in the PRA level in at least two consecutive samples (11). PRA variations lower or equal to 10% were not confirmed in subsequent samples. On the basis of the PRA profile within first year after transplantation (Fig. 1), sensitized and non-sensitized patients were divided into four groups, according to the change in the PRA level, as follows: 1. HLApre /post , comprising patients who were non-sensitized before transplantation and did not develop HLA antibodies in the first year after transplantation (n = 80); 2. HLApre /post+, comprising patients who were non-sensitized before transplantation and developed de novo antibodies in the first year after transplantation (n = 8); 3. HLApre+/post↑, comprising patients who were sensitized before transplantation and whose PRA levels increased in the first year after transplantation (n = 9); 4. HLApre+/post↓, comprising patients who were sensitized before transplantation and whose PRA levels decreased in the first year after transplantation (n = 14). All acute rejection episodes were biopsy-proven (BPAR) and classified according to Banff’ 97 criteria, updated in 2003. AMR was diagnosed when allograft dysfunction and two of the following three criteria were present: histological damage, peritubular capillary C4d staining, and presence of circulating DSA (1, 12): Biopsies were for cause and not indicated by the results of this study. Episodes of pure T-cell-mediated rejection (TCMR) were treated with pulses of methylprednisolone (500 mg/d for three d) associated or not to Thymoglobulin (7 mg/kg for 7–10 d; Genzyme Corporation, Framingham, MA, USA). AMR was treated with Thymoglobulin associated or not to high dose of human intravenous immunoglobulin or plasmapheresis, upon availability. Therapeutic success was defined by recovery of renal function to previous levels. Graft function was evaluated by estimated glomerular filtration rate, calculated by the Modification of Diet in Renal Disease (MDRD) equation (13). Proteinuria was evaluated by determining the urinary protein-to-creatinine ratio. Both were eval-
1235
de Souza et al.
HLApre−/post− ΔPRA: Class I: 0% Class II: 0%
HLApre−/post+ ΔPRA: Class I: 14-85% Class II: 38-100%
HLApre+/post↑
HLApre+/post↓
ΔPRA: Class I: 40-75% Class II: 58-67%
ΔPRA: Class I: 15-67% Class II: 16-76%
Fig. 1. Dynamics of anti-HLA antibodies after kidney transplantation. Pre- and post-transplant (Tx) dynamics of anti-human leukocyte antigen (HLA) antibodies, by panel-reactive antibody (PRA) level. HLApre /post , no anti-HLA antibodies pre- or post-transplant (n = 80); HLApre /post+, de novo anti-HLA antibodies post-transplant (n = 8); HLApre+/post↑, sensitized pretransplant/increased anti-HLA antibodies post-transplant (n = 9); HLApre+/post↓, sensitized pre-transplant/decreased anti-HLA antibodies post-transplant (n = 14). DPRA: Variation of panel reactive antibodies between the serum samples.
uated at the end of the first, second, and third years post-transplant. Graft failure was defined as return to dialysis. The study endpoints were to evaluate the impact that pre- and post-transplant PRA-ELISA profiles had on the incidence of acute rejection episodes, graft failure, and graft function. Data related to antibody profile and acute rejection were analyzed at the end of first year post-transplant, whereas those related to renal function and graft survival were analyzed up through the third year after surgery. Anti-HLA antibody detection
All pre- and post-transplant serum samples (n = 926) were screened by PRA-ELISA (LAT-M; One Lambda) to identify anti-HLA class I and class II antibodies of the immunoglobulin G isotype. In positive sera, PRA levels, expressed as a percentage, were measured by ELISA for class I
1236
and class II reagents (LAT 1240; One Lambda). Both PRA-ELISA tests were performed according to the manufacturer’s protocols (14). A sensitive solid-phase assay, using purified HLA class I and/or class II single-antigen beads (Luminex platform; LABScreen; One Lambda), was used in two circumstances: (i) to confirm the presence or absence of DSA in sera of all patients presenting BPAR; (ii) to confirm that anti-HLA antibodies were not present before transplantation and were in fact de novo post-transplant antibodies (in HLApre /post+ group patients). Commercial kits (LABScreen Single-antigen LS1A04 and LS2A01; One Lambda) were used in accordance with the manufacturer protocols. In brief (15), 5 lL of class I and class II beads was mixed with 20 lL of sera and the mixture was incubated in the dark for 30 min at room temperature. After three washes, 100 lL of 1:100 fluorescein isothiocyanate-conjugated goat anti-human IgG was added
Anti-HLA antibodies and graft outcome and the samples were incubated under the same conditions for 30 min. The microbeads were then washed twice and resuspended in 80 lL of fixing solution. Negative and positive controls were tested in the same manner. The tests were read on a flow cytometer (LABScreen Luminex 100; One Lambda). A mean fluorescent intensity above 1000 U was defined as a positive reaction.
Therefore, the final study sample comprised 111 kidney transplant recipients (57 males and 54 females), with a mean age of 43.6 13.0 yr. Of those 111 patients, 40 (34%) received a deceaseddonor kidney and 71 (66%) received a living-donor kidney. Ninety-six patients (86.5%) received their first transplants. Clinical data were obtained from our service electronic database.
Histology and immunopathology
Post-transplant dynamics of anti-HLA antibodies
The BPAR were defined according to Banff’ 97 criteria, updated in 2003 (1). Samples were stained for C4d by immunofluorescence (16). Immunohistochemistry (17) was used only when the material was not available for immunofluorescence. The intensity and distribution of C4d staining in the peritubular capillaries were graded as follows: negative staining (0–10%); focal staining (11–50%); and diffuse staining (> 50%). Any staining higher than 10% of the peritubular capillaries was considered positive (18).
There were no statistically significant differences among the four groups in terms of the following pre-transplant characteristics: recipient age, recipient gender, donor type, number of pregnancies, and number of blood transfusions, HLA mismatches, or maintenance immunosuppression regimens. In comparison with the HLApre / post and HLApre /post+ groups, the HLApre+/ post↑ and HLApre+/post↓ groups showed a higher number of cases of retransplantation (p = 0.02) and more frequent use of ATG as induction therapy (p < 0.001). Clinical and demographic data are presented in Table 1. In the HLApre /post+ group (n = 8), 12.5% of the patients were retransplants, 62.5% had pre-tranplant blood transfusions, and 100% of the women had previous pregnancy. Both pre-transplant and historic sera of these patients were evaluated using single-antigen beads to exclude pre-transplant DSA in this group of patients.
Statistical analysis
Continuous variables were expressed as mean standard deviation or as median and range. For continuous variables, means were compared using Student’s t-test. Observations between groups were compared using Fisher’s exact test and the Kruskal–Wallis test. One-way analysis of variance was used to compare continuous variables. Graft survival was assessed using Kaplan–Meier analysis and compared by log-rank test. In logistic regression analyses, several factors were tested in terms of their impact on AMR and graft loss: age; gender; donor type; number of previous transplants and blood transfusions; induction therapy based on HLA-A, -B, and -DR mismatches; maintenance immunosuppression therapy; and post-transplant PRA profile. Values of p < 0.05 were considered statistically significant. Data were analyzed with the SPSS software (SPSS Inc., Chicago, IL, USA), version 16.0 for Windows. Results
One hundred and thirty-nine kidney transplants were performed at our facility during the study period. A total of 28 recipients were excluded from the analysis: for undergoing simultaneous transplantation of another organ (n = 8); for participating in another study (n = 4); for being younger than 18 yr of age (n = 4); for evolving to graft loss or death within the first 24 h post-transplant (n = 2); and for declining to participate in the study (n = 10).
Time to PRA change
The mean time to PRA change was 38 8 d (median 14 d, p = 0.276). In non-sensitized patients, de novo anti-HLA antibodies were detected between post-transplant days 7 and 180 (Fig. 2A) in a bimodal curve: in 50% of these patients, de novo antibodies appeared in the first 14 post-transplant days and in other 50% they appear after the 60th post-transplant day. In sensitized patients, PRA levels changed within the first 90 d post-transplant (Fig. 2B). Among HLApre+/post↑ group, seven of nine (78%) patients showed an increase in PRA level by day 14; among HLApre+/ post↓ group, decrease of PRA occurred after day 30 in nine of 14 (65%) patients. Acute rejection
The overall incidence of BPAR was 22% (24 of the 111 patients evaluated) and occurred at a median of 13 d after transplantation (range, 5 to 340 d). As can be seen in Table 2, the rate of BPAR was significantly different among the four
1237
de Souza et al. Table 1. Clinical and demographic characteristics of the 111 kidney transplant recipients evaluated, by group Groups
Characteristics Age (yr), mean SD Female gender, n (%) Deceased donor, n (%) First transplant, n (%) Previous pregnancies (≥1), n (%) Previous transfusions (≥1), n (%) HLA-AB mismatches, n 0 1–2 3–4 HLA-DR mismatches, n 0 1 2 Induction therapy; n (%) Anti-IL-2 receptor antibodies, n (%) Antithymocyte globulin Maintenance therapy with Tac+mycophenolate, n (%) Post-transplant blood transfusion, n (%)
HLApre /post (n = 80)
HLApre /post+ (n = 8)
HLApre+/post↑ (n = 9)
HLApre+/post↓ (n = 14)
p
43 14 36 (45) 25 (31) 73 (91) 25 (70) 48 (60)
47 5 3 (37.5) 5 (62.5) 7 (87.5) 3 (100) 5 (62.5)
44 11 7 (78) 4 (44.5) 5 (56) 7 (100) 8 (89)
43 9 8 (57) 6 (43) 11 (78.5) 7 (87.5) 12 (86)
0.938 0.198 0.261 0.020a 0.254 0.132
2 44 34
0 7 1
0 7 2
0 9 5
0.517
28 36 16
4 3 1
2 4 3
7 5 2
0.819
75 (94) 5 (6) 6 (43)
8 (100) 0 (0) 7 (87.5)
3 (33) 6 (67) 8 (89)
6 (43) 8 (57) 14 (100)
8 (57)
5 (62.5)
6 (67)
4 (29)
