C4d Staining and Graft Survival
C4d deposition in peritubular capillaries (defined as C4d ≥2) was identified in 17.5% of HLA-incompatible biopsies at 1 year and graft survival for patients with and without C4d deposition was 61.9% versus 96.0%, respectively (P<0.001).
Figure 1 illustrates a flow diagram delineating the histopathological phenotypes of the HLA-incompatible patients, percentage representation of study cohort, and respective graft survival. The figure highlights the more histopathological observations that accrue, the worse the graft survival for HLA-incompatible recipients. For example, HLA-incompatible recipients with evidence of concomitant C4d deposition, transplant glomerulopathy, and microcirculation inflammation had 33.3% graft survival at median follow-up versus 100.0% for those not displaying any of these histopathological parameters.
Transplant Glomerulopathy and Graft Survival
Transplant glomerulopathy (TG), defined as cg greater than 0, was identified in 25.0% of HLA-incompatible kidney transplant recipients at 1 year, and graft survival with versus without TG was 66.7% versus 96.7%, respectively (P<0.001) (see Figure S1a, SDC,http://links.lww.com/TP/A894). Categorizing TG into individual grades, it was observed that compared to a cg score of 0 (i.e., no transplant glomerulopathy, graft survival 96.6%) a greater severity of transplant glomerulopathy grading demonstrated worse death-censored graft survival (cg score=1 vs. 2 or 3, graft survival=66.7% vs. 71.4%, respectively, P<0.001).
Concomitant presence of both C4d deposition and TG portended a poor outcome (see Figure S1b, SDC,http://links.lww.com/TP/A894). Graft survival was 33.3% versus 97.2% among patients with and without concomitant C4d and TG, respectively (P<0.001).
Microcirculation Inflammation (Glomerulitis or Peritubular Capillaritis) and Graft Survival
Glomerulitis was present in 60.0% of the 1-year biopsies, and its presence was associated with worse graft survival (82.1% vs. 98.1%, P=0.004). Peritubular capillaritis was present in 47.6% of the biopsies, and graft survival in those with versus without capillaritis was 88.1% versus 97.7%, respectively (P=0.091) (see Figures S1c and S1d, respectively, SDC,http://links.lww.com/TP/A894). We combined the glomerulitis and peritubular capillaritis scores into one (microcirculation inflammation score, or MCI score). Graft survival was 86.1% versus 98.0% with a MCI score greater than 2 versus less than or equal to 2, respectively, and this did not quite reach significance (P=0.054). Severity of microcirculation inflammation was associated with initial antibody strength, although only significantly different for glomerulitis rather than peritubular capillaritis (see Table 2).
See SDC 1,http://links.lww.com/TP/A894.
In C4d-negative HLA-incompatible recipients (82.6% of HLA-incompatible cohort), we investigated the importance of microcirculation inflammation as a predictor of allograft loss (Fig. 1B–D). Glomerulitis and peritubular capillaritis was prevalent at 47.0% and 43.6% of C4d-negative HLA-incompatible recipients, respectively (in 29.2% both were present together). Graft survival among C4d-negative recipients with or without glomerulitis was not significantly different (93.6% vs. 98.1%, P=0.271). Likewise, phenotypes consisting of C4d-negative staining with or without peritubular capillaritis had similar outcomes (94.1% vs. 97.7% graft survival, P=0.390). Combining both to attain the MCI score, MCI greater than 2 versus less than or equal to 2, there was no difference in outcomes of patients with C4d-negative 1-year biopsies who either did or did not show microcirculation inflammation (92.9% vs. 98.0%, P=0.291).
Donor-Specific Antibody Contribution to 1-Year Histology and Survival
Attributing TG and MCI as histopathological markers of antibody-mediated injury, we sought to determine how the development of these histological markers was related to DSA or rejection episodes. Presence of any DSA at 1 year increased the risk of TG at 1 year (prevalence of TG 30.4% vs. 7.7% if DSA present or absent, respectively, P=0.027). Prevalence of DSA at 1 year was significantly associated with the presence of MCI greater than 2 (58.7% vs. 42.9% if DSA present or absent, respectively, P=0.004).
Detectable circulating DSA at the time of biopsy increased the risk of any episode of AMR, AMR alone (no rejection vs. rejection, 39.7% vs. 69.2%, P=0.005), or mixed cellular and AMR (no rejection vs. rejection, 44.9% vs. 54.3%, P<0.001). The presence of DSA was also associated with overall (cellular or antibody-mediated, or both) rejection episodes (no rejection vs. rejection, 37.5% vs. 55.8%, P<0.001).
Although all recipients had DSA before desensitization, the persistence of DSA decreased over time; 82.3% of the patients at 1 month posttransplant had detectable DSA, whereas at 12 months it dropped to 53.6%.
See SDC 2,http://links.lww.com/TP/A894.
See SDC 3,http://links.lww.com/TP/A894.
To assess the independent association of histological markers and death-censored graft survival, we built multivariate models. All variables that showed significant association with the outcome measures in the univariate Cox models were considered for inclusion in the multivariate models. Accordingly, indication-based biopsy, glomerulitis, presence of C4d, transplant glomerulopathy, and chronicity score greater than 4 were the co-variables entered into the multivariate Cox proportional hazards models. Only having an indication-based biopsy was independently associated with higher risk of graft loss (hazard ratio 2.8, 95% CI 1.9–3.7, P=0.032). In a Cox regression model analyzing protocol biopsies only, no significant independent histopathological predictor for allograft loss was identified.
In this study, we have identified a series of histopathological phenotypes on 1-year biopsies and correlated them with allograft survival in HLA-incompatible live-donor kidney transplants. Characterizing the risk profile of each of these phenotypes adds to our understanding of the pathophysiology of antibody-mediated injury. Furthermore, it provides information necessary to alter the natural history of the progression of these microscopic lesions to irreversible allograft dysfunction by using therapeutic intervention. For example, the presence of microcirculation inflammation on a 1-year biopsy could justify intervention if there was also glomerulitis, whereas the results of this study are equivocal as to the need for intervention if the tissue is C4d negative or the peritubular capillaritis is isolated.
The utility of C4d scoring continues to evolve, with recent evidence questioning exclusive reliance upon C4d-based diagnostic frameworks for AMR. Evidence supporting changes to the current diagnostic classification have recently been mounting. For example, C4d demonstrates poor sensitivity for identifying AMR in both acute (6, 7) and chronic settings (8, 9). In the current study, however, C4d remains one of the most robust indicators of reduced allograft survival.
In 3-month protocol biopsies of a pre-sensitized deceased-donor cohort, Loupy et al. (6) demonstrated 49% prevalence of C4d-negative AMR (capillaritis, glomerulitis, and DSA) versus 31% C4d-positive AMR. The C4d-negative group subsequently developed more fibrosis and TG (and worse allograft function) at 1 year posttransplantation. Our data do not support a link between microcirculation inflammation and graft survival when C4d is negative. However, there are important differences in the methodology of the Loupy study and ours. Our index biopsy was taken at 1 year and our study population was recipients of live-donor kidneys desensitized before transplantation. Our data show that between 3 months and 12 months, the likelihood of finding DSA drops from 87.1% to 53.6%. The elimination or persistence of DSA may modulate the fate of these lesions. Also, our early (1- or 3-month) protocol biopsies with phenotypes showing C4d staining in the presence of DSA were treated with additional plasmapheresis and anti-CD20, and some of these have been converted to C4d-negative phenotypes, perhaps slowing or halting the progression to chronic AMR.
The number of patients with a C4d-negative, microcirculation inflammation–positive phenotype may be too small to be powered to show a difference. However, if one looks at the entire cohort, only glomerulitis, present in 60.0% of the 1-year biopsies, was associated with worse graft survival (82.1% vs. 98.1%, P=0.004). Peritubular capillaritis was present in a large portion (47.6%) of the biopsies and did not have a significant effect on graft survival (88.1% vs. 97.7%, respectively, P=0.091).
Other studies have shown a substantial fluctuation in C4d status in the first year after an HLA-incompatible transplant, suggesting a dynamic humoral process (10) or perhaps a variable expression in the level of the target HLA molecules. We have previously shown that DSA can suddenly appear or increase in strength in response to infection and surgical trauma (11). These findings support frequent DSA screening and biopsies triggered by inflammatory events and increases in DSA. Close surveillance permits early intervention. In our practice, after each treatment cycle for clinical or subclinical rejection, a 1-month follow-up biopsy is obtained, and if histologic and immunohistologic features of injury persist, then therapeutic intervention is repeated.
Microcirculation inflammation is emerging as an independent indicator of AMR, in the presence or absence of C4d. Although lacking specificity in the early period posttransplantation, this limitation should be minimized by 1 year posttransplant. Early differential diagnosis for microcirculation inflammation (12) includes T-cell-mediated rejection (21% with glomerulitis; 16% with peritubular capillaritis), glomerulonephritis (16% with glomerulitis), and acute tubular necrosis (24% with peritubular capillaritis). Sis et al. (12) computed a combined microcirculation inflammation score (glomerulitis+peritubular capillaritis) and found that it strongly indicated the presence of DSA and poor kidney allograft outcomes (in patients without pre-existing immunological barriers). Similarly, de Kort et al. (13) found an MCI score greater than 2 (in patients with de novo DSA) was associated with 21-fold increased risk for graft failure (95% CI 2.5–180.0, P=0.005).
In our study, glomerulitis seems to be a better predictor of premature graft loss than capillaritis or an MCI score greater than 2. Furthermore, patients with or without capillaritis when glomerulitis was not present had similar creatinine clearances and were at the same risk of developing proteinuria. Taken together, these findings bring into question the correlation of capillaritis to premature graft loss at least in the desensitization population. We have not routinely treated patients with biopsies that show capillaritis without positive C4d staining or severe glomerulitis, so we are probably seeing the unaltered natural history of the isolated capillaritis. Greater numbers and longer follow-up will be needed to determine whether peritubular capillaritis is associated with long-term outcome, but in our short-term to medium-term data, glomerular changes had a much stronger association.
The results of this study show that there is a greater burden of histopathologic features associated with chronic AMR with increasing strengths of the pre-desensitization crossmatch. Several studies have now shown that the incidence of TG is greater and long-term graft survival is reduced when the strength of the crossmatch before desensitization is at a cytotoxic level (14, 15). In our cohort, the presence of DSA is associated with an increased likelihood of finding TG and microcirculation inflammation on the 1-year biopsy but not proteinuria. Eliminating patients desensitized for high-titer DSA by finding more favorable donors (with lower DSA strength, requiring less desensitization) in paired donation pools would appear to have a dramatic favorable effect on outcomes.
Not surprisingly, the presence of both C4d and TG is associated with worse graft survival and had better predictive value than either C4d or TG alone, consistent with data from the compatible kidney transplant literature (16). While most closely associated with chronic AMR, TG can have diverse etiologies including hepatitis C and thrombotic microangiopathy (17). TG is thought to be progressive and currently has no proven treatment once established. The interval of progression, however, is variable, and if the DSA is successfully removed and microcirculation inflammation resolves, it could slow the progression to renal failure. The relationship between intervention and progression once lesions are established needs to be studied in this population.
The results from Stegall and colleagues (18), using prophylactic terminal complement inhibition with the anti-C5 specific antibody eculizumab for HLA-incompatible transplants, show a dramatic reduction in acute AMR in the first 3 months. However, a substantial number of patients went on to develop of TG or other features of chronic AMR (19). This could be explained by the discontinuation of complement blockade during the first year in many cases while there was still circulating DSA. Long-term complement inhibition could be used selectively in patients with persistent C4d deposition or early TG on surveillance biopsies with the goal of slowing or preventing further progression to chronic AMR.
This study has some limitations. Approximately a third of the incompatible kidney recipients over the period included in this study did not have biopsies around the 1-year posttransplantation window. This was largely a result of logistical issues and should not introduce systematic selection bias in this analysis (supported by comparative analysis of HLA-incompatible recipients included and excluded in study cohort—see Table S2, SDC,http://links.lww.com/TP/A894). However, we acknowledge there are some differences between included and excluded groups (e.g., follow-up posttransplant) that could potentially add bias to interpretation of our results. It could be argued that early biopsy analysis may be more beneficial, before the development of progressive and irreversible histology such as transplant glomerulopathy (20), although as discussed above the sensitivity and specificity of many histopathological markers are low immediately posttransplantation. Analyzing the evolution of histopathology over the first year post–HLA-incompatible kidney transplantation (and relationship to outcomes) was beyond the scope of this project but would be fascinating to explore further. The follow-up, while longer than most studies reported in the literature, is still limited for live-donor transplants. Finally, analyzing both indication and protocol biopsies together could skew our results, but we have taken measures to perform comparative analysis between the two cohorts. We also speculate that our protocol biopsy–only cohort is underpowered for the Cox regression analysis and may result in a type 2 statistical error (false-negative rate).
We conclude that histopathology from 1-year posttransplant biopsies can guide risk stratification among HLA-incompatible kidney transplant recipients. Further work is required to explore evolution of histopathology post–HLA-incompatible transplantation, as this would be predicted to be more instructive in guiding therapeutic intervention when it is likely to make a difference in disease progression. Our findings do not support a link between some phenotypes previously reported to lead higher rates of transplant glomerulopathy or poor outcomes in different populations with DSA and should provide a cautionary note during the Banff process to update diagnostic classification of antibody-mediated injury.
MATERIALS AND METHODS
This is a retrospective analysis of a prospective database maintained on all patients who underwent transplantation at the Johns Hopkins Hospital within the Incompatible Kidney Transplant Program who had DSA immediately before the initiation of desensitization. All patients were treated with a treatment protocol approved by the Johns Hopkins Institutional Review Board (IRB)—the IRB granted approval to convert the prospective clinical database to a research database (which undergoes annual review). Patients were offered the opportunity to enter into a kidney paired donation program as an alternative to receiving an incompatible allograft. Since the program began in 1998, 221 HLA-incompatible kidney transplants had been performed up to the end of 2010. Data was available from 1-year biopsies (protocol or indication based) in 124 HLA-incompatible kidney recipients and formed the study cohort.
All patients received plasmapheresis using COBE Spectra (Gambro BCT, Lakewood, CO) as previously described (21). One plasma volume was removed per treatment and replaced using either 5% albumin solution or fresh frozen plasma. CMVIg (Cytogam; MedImmune Inc., Gaithersburg, MD) was administered at 100 mg/kg after each plasmapheresis treatment. Escalating numbers of treatments were performed before and after transplantation dependent upon DSA level at baseline.
Mycophenolate mofetil (2 g/day) and tacrolimus (target serum level, 8–12 ng/mL) were initiated with the first plasmapheresis treatment before transplantation and then continued as maintenance therapy. Induction therapy consisted of the use of daclizumab, with an initial dose of 2 mg per kilogram and then 1 mg per kilogram every 2 weeks for a total of five doses, or antithymocyte globulin (Thymoglobulin; Genzyme) at a dose of 1.5 mg per kilogram per day for 5 days. Glucocorticoids, including dexamethasone, were administered at a dose of 100 mg intraoperatively and at a dose of 25 mg every 6 hours postoperatively for six doses, followed by tapering to 5 to 10 mg daily during a 3-month period.
Cellular rejection episodes (both clinical and subclinical) with Banff grades of 1A or 1B were treated with a 3-day pulse of dexamethasone 100 mg/day followed by a taper. If the Banff score was 2A, 2B, or 3, patients received a 7-day course of antithymocyte globulin. Both clinical and subclinical C4d-positive AMR, as per Banff criteria (22) in the presence of DSA, was treated with reinitiating plasmapheresis and IVIg until the flow crossmatch converted to negative or AMR resolved on histological examination, or both. C4d-negative histopathologic lesions (e.g., isolated peritubular capillaritis or glomerulitis) were not treated.
Crossmatch techniques including antiglobulin-enhanced lymphocytotoxicity (AHG-CDC) with T cells, one-wash CDC (1wCDC) with B cells, and flow cytometry with T and B cells were performed. If present, antibodies were characterized in tests of multianalyte beads bearing individual class I or class II phenotypes, or both (LIFEMATCH-ID; Gen-Probe, Stamford, CT), and, when needed for confirmation, tests of multianalyte beads bearing single antigens (LABscreen Single Antigens; One Lambda, Canoga Park, CA). Results were read on a Luminex fluroanalyzer with results expressed as mean fluorescence intensity. Titers for DSA were based upon IgG antibodies.
Reactions with test beads yielding mean fluorescent intensities (MFI) of greater than or equal to 1000 were considered positive. Although the sensitivity of detection varies for different HLA antibodies, in general, positive flow cytometric crossmatch tests correlated with the following MFI values from the solid-phase immunoassays: greater than or equal to 5000 on phenotype panels and greater than or equal to 10,000–15,000 on single antigen panels. Positive complement-dependent cytotoxicity crossmatch results were associated with greater than 10,000 MFI on phenotype panels. However, caution must be exercised when interpreting these results because of significant variation in the results of solid-phase assays within and especially between laboratories, and nothing replaces actually performing the crossmatch.
Histological and Immunohistologic Surveillance
Surveillance biopsies were obtained at 1 year posttransplantation. Indication biopsies were in context of allograft dysfunction (creatinine rise >20% or new-onset proteinuria, or both). Allograft biopsies were performed using an 18-gauge biopsy needle under ultrasound guidance and processed as previously described (23). C4d staining was performed on frozen tissue by indirect immunofluorescence using anti-human C4d antibody (Quidel, San Diego, CA) at 1:40 dilution, followed by secondary antibody (fluorescein-isothiocyanate-conjugated goat anti-mouse IgG; Jackson Immunoresearch Laboratories, West Grove, PA), or on paraffin-embedded tissue sections using a rabbit polyclonal anti-human antibody (American, catalog # 12-5000) at a 1:50 dilution, coupled with a biotin-free polymer detection system (Leica). Biopsies were evaluated for cell-mediated rejection and AMR. All pathological assessment was graded by reference to Banff 1997 criteria, taking account of 2007 updates (22). C4d staining was considered positive if present in greater than or equal to 50% of the peritubular capillaries with intensity greater than or equal to 1+ (C4d2–3). Banff qualifiers were used to score biopsies for the presence of glomerulitis: g greater than or equal to 1; peritubular capillaritis: ptc greater than or equal to 1; transplant glomerulopathy: cg greater than or equal to 1. Biopsies with immune complex glomerulonephritis were excluded from the analysis of glomerulitis and transplant glomerulopathy.
Statistical analysis was performed using standard software (SPSS version 20, Chicago, IL). Univariate and multivariate survival analyses were performed by Kaplan-Meier estimation and Cox regression proportional hazard modeling, respectively, to assess the effect of histological parameters upon the outcome of death-censored graft survival. All data was censored at 5 years and right-censored to account for sample loss before the final outcome (e.g., graft loss) was observed.
1. Montgomery RA, Warren DS, Segev DL, et al. HLA incompatible
renal transplantation. Curr Opin Organ Transplant
2012; 17: 386.
2. Montgomery RA, Lonze BE, King KE, et al. Desensitization
in HLA-incompatible kidney recipients and survival. N Engl J Med
2011; 365: 318.
3. Stegall MD, Gloor JM. Deciphering antibody-mediated rejection: new insights into mechanisms and treatment. Curr Opin Organ Transplant
2010; 15: 8.
4. Colvin RB. Antibody-mediated renal allograft rejection: diagnosis and pathogenesis. J Am Soc Nephrol
2007; 18: 1046.
5. Sharif A, Alachkar N, Kraus E. Incompatible kidney transplantation
—a brief overview of the past, present and future. QJM
2012; 105: 1141.
6. Loupy A, Suberbielle-Boissel C, Hill GS, et al. Outcome of subclinical antibody-mediated rejection in kidney transplant recipients with preformed donor-specific antibodies. Am J Transplant
2009; 9: 2561.
7. Haas M, Montgomery RA, Segev DL, et al. Subclinical acute antibody-mediated rejection in positive crossmatch renal allografts. Am J Transplant
2007; 7: 576.
8. Halloran PF, de Freitas DG, Einecke G, et al. An integrated view of molecular changes, histopathology
, and outcomes in kidney transplants. Am J Transplant
2010; 10: 2223.
9. Tinckam KJ, Djurdjev O, Magil AB. Glomerular monocytes predict worse outcomes after acute renal allograft rejection independent of C4d status. Kidney Int
2005; 68: 1866.
10. Loupy A, Hill GS, Suberbielle C, et al. Significance of C4d Banff scores in early protocol biopsies of kidney transplant recipients with preformed donor-specific antibodies (DSA). Am J Transplant
2011; 11: 56.
11. Locke JE, Zachary AA, Warren DS, et al. Proinflammatory events are associated with significant increases in breadth and strength of HLA-specific antibody. Am J Transplant
2009; 9: 2136.
12. 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: 1168.
13. de Kort H, Willicombe M, Brookes P, et al. Microcirculation inflammation associates with outcome in renal transplant patients with de novo donor-specific antibodies. Am J Transplant
2013; 13: 485.
14. Burns JM, Cornell LD, Perry DK, et al. Alloantibody levels and acute humoral rejection early after positive crossmatch kidney transplantation
. Am J Transplant
2008; 8: 2684.
15. Bentall A, Cornell LD, Gloor JM, et al. Five-year outcomes in living donor kidney transplants with a positive crossmatch. Am J Transplant
2013; 13: 76.
16. Kieran N, Wang X, Perkins J, et al. Combination of peritubular C4d and transplant glomerulopathy predicts late renal allograft loss. J Am Soc Nephrol
2009; 20: 2260.
17. Baid-Agrawal S, Farris AB 3rd, Pascual M, et al. Overlapping pathways to transplant glomerulopathy: chronic humoral rejection, hepatitis C infection, and thrombotic microangiopathy. Kidney Int
2011; 80: 879.
18. Stegall MD, Diwan T, Raghavalah S, et al. Terminal complement inhibition decreases antibody-mediated rejection in sensitized renal transplant recipients. Am J Transplant
2011; 11: 2405.
19. Stegall MD, Chedid MF, Cornell LD. The role of complement in antibody-mediated rejection in kidney transplantation
. Nat Rev Nephrol
2012; 8: 670.
20. Haas M, Mirocha J. Early ultrastructural changes in renal allografts: correlation with antibody-mediated rejection and transplant glomerulopathy. Am J Transplant
2011; 11: 2123.
21. Montgomery RA, Zachary AA, Racusen LC, et al. Plasmapheresis and intravenous immune globulin provides effective rescue therapy for refractory humoral rejection and allows kidneys to be successfully transplanted into cross-match-positive recipients. Transplantation
2000; 70: 887.
22. Solez K, Colvin RB, Racusen LC, et al. Banff 07 classification of renal allograft pathology: updates and future directions. Am J Transplant
2008; 8: 753.
23. Bagnasco SM, Tsai W, Rahman MH, et al. CD20-positive infiltrates in renal allograft biopsies with acute cellular rejection are not associated with worse graft survival. Am J Transplant
2007; 7: 1968.
Incompatible transplantation; Desensitization; HLA incompatible; ABO incompatible; Kidney transplantation; Protocol biopsy; Histopathology
Supplemental Digital Content
© 2014 by Lippincott Williams & Wilkins