With the advent of sensitive assays such as single antigen bead assay, we can now easily detect donor-specific anti–human leukocyte antigen (HLA) antibody (DSA) through serially monitoring. It has recently been determined that 20% of all primary transplant recipients will develop DSA in the first 5 years after transplantation, with the highest period of risk in the first 12 months after transplantation (1). We also now know that 7% to 9% of patients will fail per year once the DSA appears (1, 2). However, just knowing the risk of failure is not the complete picture of the clinical scenario. Rather, knowing how the graft function changes temporally before, at the onset of, and after DSA would be helpful in assessing when an intervention is necessary to improve or prevent dysfunction. Thus, the primary aim of the current study was to examine the posttransplantation correlation between DSA appearance and its relationship to change in allograft function.
RESULTS
All adult, non-HLA identical, pretransplantation DSA-negative, crossmatch (XM)–negative, ABO-compatible, primary kidney transplants recipients who were transplanted between 1999 and 2006 were identified. Of the 189 patients, 175 had adequate longitudinal data sample points for allograft function analysis and were included in this study. Of these 175 patients, 42 patients had de novo DSA and had at least 5-year posttransplantation follow-up or lost their allograft.
Risk of Failure in DSA-Positive Patients
The patients’ characteristics are presented in Table 1. The baseline posttransplantation estimated glomerular filtration rate (eGFR) in those who would eventually develop DSA was nearly identical to those who never developed DSA (median eGFR 57 vs. 55 mL/min, respectively; P=0.76). Conversely, at the last follow-up, the eGFR was less than 40 mL/min in a higher proportion of transplant patients with DSA (20 of 42) than in those without DSA (40 of 133; P=0.037). Additionally, a higher proportion patients failed if they had DSA (11 of 42) compared with if they were without DSA (15 of 133; P=0.02). This equates to DSA increasing the odds of allograft failure by 2.8 times (odds ratio, 2.8; 95% confidence interval, 1.2–6.7).
;)
TABLE 1: Patient demographics
Stable Allograft Function Correlates with Freedom from DSA
Finding that the DSA-positive patients had a significantly worse function over time led to a close interrogation of when the eGFR decline commenced in this group. In the DSA-positive cohort, the eGFR remained stable during the DSA-negative period (time from transplantation to the last DSA-negative sera sample collection) and is shown in Figure 1A. The median time to DSA from transplantation was 17.7 months. Only four patients’ eGFR significantly declined in this period (eGFR decrease >25%). One patient had recurrent focal segmental glomerulosclerosis. Another was found to have polyoma virus nephropathy. The third had medical complications unrelated to the kidney including cerebrovascular event and pneumonia that resulted in a drop in eGFR. The fourth patient was found to have medication noncompliance.
FIGURE 1: Glomerular function trends before (A) and after (B) DSA onset. Before DSA onset, allograft function is relatively stable (Wilcoxon signed rank P<not significant [NS]). After DSA onset, allograft function significantly declines by 2 years post-DSA (Wilcoxon signed rank P<0.001).
Early Allograft Dysfunction After DSA Onset
After DSA onset, however, allograft function significantly declined for the group of DSA-positive patients as a whole by 1 and 2 years post-DSA (P<0.001 for both; Fig. 1B). However, not all of the patients had allograft dysfunction. Figure 2A shows 12 patients that allograft dysfunction by 1 year post-DSA (8 patients with >25% drop in their eGFR but still have a functioning allograft and 4 patients with allograft failure by 1 year post-DSA). The remaining 30 patients remained stable (≤25 decline in eGFR) at 1 year post-DSA (Fig. 2B). By 2 years post-DSA, only two new patients from the stable group developed an eGFR decline (compared with time of DSA onset) of more than 25%. The remaining 28 patients continued to be stable.
FIGURE 2: Glomerular function trends in DSA-positive patients stratified by those who are stable (A) and those how have early dysfunction/failure (B) by 1 year post DSA.
Characteristics of Early GFR Decline After DSA Onset
Table 2 divides the patients based on the eGFR drop at 1 year post-DSA. Most baseline characteristics did not differ between stable allograft dysfunction (≤25 decline in eGFR by 1 year post-DSA; n=30), early allograft dysfunction (with a functioning allograft but >25% drop in eGFR by 1 year post-DSA; n=8), and early failure (allograft failure in the first year after DSA onset; n=4) patients. The only variable that was significantly different was the rate of antibody-mediated rejection (AMR; detected via for-cause biopsies). In those patients considered stable, the rates of acute AMR after DSA onset were significantly lower than the early allograft dysfunction and early failure groups (P=0.03). All patients in the early failure group had AMR before failure. Of note, early failure was rapid with a median time to allograft loss from DSA onset of 2.4 months (1–10 months).
TABLE 2: Change in GFR between DSA onset to 1-year post- DSA
Predicting Late GFR Decline After DSA Onset
Early failure is strongly tied to AMR and may be difficult to address due to its rapid onset, leaving very little time to treat. However, late failure (allograft failure more than 1 year after DSA onset) may be preventable. If trends in eGFR decline are slow and can predict late failure, strategic monitoring and treatment may save a patient from needing retransplantation. After excluding all those patients with early failure, we analyzed the trends in GFR on the remaining 38 patients to determine if early allograft dysfunction could predict late failure. As shown in Figure3A, having an eGFR decline of more than 25% by 1 year post-DSA (early allograft dysfunction group) leads to a higher late allograft failure rate when compared with stable patients (P<0.001). Early allograft dysfunction predicts failures at 3 years post-DSA. In addition, early allograft dysfunction precedes late failure by almost 1 year, allowing time for clinical intervention. We further divided these patients into groups based low (<45 mL/min; n=11) or normal (≥45 mL/min; n=31) eGFR at DSA onset. In the patients with an eGFR of 45 mL/min or more, 24 patients were stable and 7 patients had early allograft dysfunction (Fig. 3B). Between these two groups, those with early allograft dysfunction were still more likely to fail despite the high starting eGFR (P<0.001). This is important given that this group is the least likely to receive treatment because of their normal eGFR at the time of DSA onset. In the low eGFR group, the sample size does not permit us to conduct adequate analysis and is therefore not included here.
FIGURE 3: Allograft survival trend based on the degree of dysfunction by 1 year post-DSA. A, all patients are shown and those with an eGFR decline by 1 year post DSA are likely to progress to failure (P<0.001). B, A DSA cohort with eGFR above 45 mL/min at DSA onset is shown to still progress to failure if early dysfunction occurs in the first post-DSA year (log-rank P<0.001).
DISCUSSION
Long-term kidney allograft survival still remains a problem. Many studies indicate that circulating alloantibodies precede allograft failure and are likely a major cause of failure (2–11). This study confirms ours and other previous findings by showing that allograft function declines over time in those with circulating DSA more than in those without circulating DSA. Although other causes of allograft failure do exist such as hypertension, diabetes, recurrent disease, and infection their incidence in kidney transplant is proportionally smaller than DSA-associated allograft failure (1, 8, 12, 13). Thus, it seems that the development of DSA is the leading indicator of those who will have future poor allograft function.
In addition to confirming that DSA is a problem in renal transplant, this study extends prior research by looking at the change in allograft function over time as it relates to DSA onset. We have previously reported on this transplant patient population showing that that DSA appears before failure (1) but have not looked longitudinally to determine the strength of the correlation of DSA and temporal allograft function. Before DSA, our current analysis shows that little to no change in allograft function occurs and that nearly all patients have the same function as they did at baseline immediately after transplantation. This lack of eGFR change is similar to the patients who never develop DSA. Conversely, after DSA onset, the eGFR significantly declined in the DSA-positive cohort. This decline was evident as early as 1 year post-DSA and in many allograft failure ensued by 2 years post-DSA. Together, this trend of eGFR is evidence that DSA is temporally related to allograft failure.
However, despite showing a temporal relationship between DSA onset and eGFR decline, we also found a large population of patients with DSA present in their serum that were stable for at least 2 to 3 years after DSA appearance. Dividing the DSA-positive cohort at 1 year post-DSA into those stable, those with early allograft dysfunction, and those with early failure, we were able to see that patients who developed acute AMR after DSA went on to early failure. Given that the early failure occurred shortly after DSA onset, the only means to improve early failure is to adequately treat AMR (14, 15). Recent evidence from multiple reports indicates that, along with histologic resolution, DSA removal may be essential to achieving this goal (14–16). Future prospective studies are needed to see if reduction/removal of DSA is both beneficial and possible. Furthermore, these studies will be needed to determine which therapeutic agents/regimens are best suited to achieve DSA removal.
One important clinical pearl from this analysis was that late failure could be predicted by early allograft dysfunction. An eGFR decline of more than 25% within in the first year after DSA led to a dismal 42% survival probability by 3 years post-DSA onset. Conversely, the lack of early allograft dysfunction post-DSA led to 92% survival at 3 years post-DSA. Furthermore, in those patients with excellent allograft function at the time of DSA onset (eGFR >45 mL/min), the difference survival was even more drastic. This indicates that function at the time of DSA should not lead the treating physician to think that the patient will be okay. Additionally, the finding that decline precedes the eventual failure by nearly 1 year, gives the treating physician a window of opportunity to address the DSA with modalities that may include a combination of maintenance immunosuppression escalation and intermittent plasma cell–directed therapy.
Despite the intriguing findings and their possible implications for a primary renal transplant, these findings must be tempered due to a few study limitations. First, DSA is likely the major but may not be the only cause of allograft loss in these patients and future understanding of allograft loss is needed to determine the impact of other concomitant medical problems. Second, the DSA cohort sample size is small and a repeat analysis in larger data sets is needed to confirm these findings. Finally, protocol biopsy studies in this cohort were not available and would be helpful to see if minor changes intragraft correlate with allograft stability/instability.
Globally interpreting these data, we believe that the findings in this study are extremely important and timely to clinical trial design and therapeutic decision-making. First, patients who never develop DSA are not at the same risk for long-term allograft failure when compared with those who have DSA. This suggests that, in DSA-negative patients, drastic changes in immunosuppression and new immunosuppressive agent development should not be a research priority. Trials focusing on the entire transplant cohort are of little use in improving future outcomes. Rather, new therapeutic agents and clinical trials focused on the treatment of subpopulations of patients such as those with alloantibodies (and possibly autoantibodies) should be the priority and will likely be necessary to improve outcomes. Second, DSA monitoring is important to determine which patients are at eventual risk of allograft loss. Third, because many patient are stable after DSA, studies are needed to further elucidate the mechanistic/immunologic reason for allograft stability in these patients. Finally, monitoring eGFR decline in DSA-positive patients for early allograft dysfunction post-DSA may be a way to determine those most likely to have late allograft failure giving a treating physician a window of opportunity to treat.
MATERIALS AND METHODS
Patients
We enrolled all renal transplant patients receiving a living-donor or deceased-donor transplant between March 1999 and March 2006. All patients underwent a standard pretransplantation evaluation. At the time of transplantation, all patients were tested for reactivity to their donor via complement-dependent cytotoxicity XM. Flow cytometric XM was performed on all living-donor transplants. Patients’ pretransplantation sera were tested using LABScreen single antigen beads to detect for alloantibodies that would be considered DSA. Tissue typing was performed using both serology and polymerase chain reaction-single-specific-primer methods for HLA-A, -B, -DR, and -DQ antigens. We excluded all patients found to have donor-reactive alloantibodies present in circulation (and detected via XM or single antigen bead assay). DP antigen typing was not conducted on the majority of patients. Therefore, data regarding DP antigens or antibodies are not included in this report.
Study Protocol
Testing and the use of patient data were approved by the East Carolina University Brody School of Medicine Institutional Review Board for human studies. All clinical and research activities are consistent with the Principles of the Declaration of Istanbul.
Immunosuppression
Per protocol, patients with a panel-reactive antibody less than 20% and without delayed graft function received daclizumab induction, whereas patients with a panel-reactive antibody more than 20% or delayed graft function received rabbit antithymocyte globulin induction. Maintenance immunosuppression included a calcineurin inhibitor, a mycophenolic acid derivative, and a prednisone taper starting at the time of transplantation, which was reduced to and maintained at a level of 5 mg per day by 1 month after transplantation. Patients remained on maintenance immunosuppression at similar intensity throughout the study period lowering only for cases of suspected drug toxicity. No new therapeutic agents were added to treat DSA. Additional immunosuppressive agents were only added in cases of rejection. Rejection episodes were initially treated with steroids. Rabbit antithymocyte globulin was used to treat biopsy-proven acute cellular (T-cell) rejection. If the biopsy was consistent with AMR, the patient was also treated with plasmapheresis.
Rejection Pathology
Acute rejection was defined as an increase in serum creatinine at least 20% above baseline serum creatinine with histologic evidence on renal allograft biopsy by Banff 1997 criteria (update 2005) (17, 18).
Anti–HLA-Specific IgG Antibody Monitoring and Testing
Pretransplantation sera was tested with LABScreen Single Antigen class I and II beads (One Lambda, Canoga Park, CA). Posttransplantation patients were routinely monitored at 1, 3, 6, 9, and 12 months, annually, and when clinically indicated, for HLA class I and II antibodies development using LABScreen Mixed beads (One Lambda). Samples that tested positive on LABScreen Mixed beads were also tested using LABScreen Single Antigen class I and II beads (One Lambda) to determine antibody specificity. If a patient was found to be positive on LABScreen Single Antigen, all previous samples tested with the LABScreen Mixed antigen product were tested via the single antigen platform. All LABScreen tests were performed according to the manufacturer’s protocol. HLA antibodies were analyzed as mean fluorescence intensity (MFI) values. Because high-resolution typing was not conducted on all patients, DSA in this article are reported based on low-resolution typing. Low-resolution typing is DNA-based typing result at the level of the digits comprising the first field in the DNA-based nomenclature (e.g., A*01). For MFI reporting in the case of a DSA that had multiple single antigen bead alleles (i.e., A2), MFI of the highest allele bead was reported. De novo DSA were considered positive if it was a new IgG antibody not present at time of transplantation and the normalized intensity via single antigen bead of 1000 MFI or greater.
Outcomes and Definitions
The major outcome investigated in this series of consecutive renal transplant patients was the change in allograft function in the period before and after DSA appearance. Serum creatinine values from each patient within 1 to 2 months after transplantation (baseline), at time of last DSA negative sample, and at 1 and 2 years post-DSA onset were used in the analysis. All serum creatinine values were converted to eGFR using the four variable Modification of Diet in Renal Disease equation (age, race, gender, and serum creatinine). The two time periods analyzed were the DSA-negative period defined as the period from transplantation to the date of the last DSA-negative sample and the DSA-positive period as the date of the last DSA-negative sample to 1 or 2 years post-DSA or to last follow-up. Allograft loss was defined as a return to dialysis.
Statistical Methods
All statistical analyses were performed using Stata/MP version 10.1 (College Station, TX). A two-sided P<0.05 was considered statistically significant. Observations between groups were compared using the Fisher’s exact or chi-square test for categorical variables. Unpaired t test, one-way analysis of variance, or Kruskal–Wallis test were used for continuous variables. Comparison of change in eGFR was assessed between two time points using the Wilcoxon signed rank test. Kaplan–Meier analysis was used to determine probability of survival with a log-rank used to compare groups.
REFERENCES
Everly MJ, Rebellato LM, Haisch CE, et al. Incidence and impact of de novo donor-specific alloantibody in primary renal allografts.
Transplantation 2013; 95: 410.
Bentall A, Gloor J, Cornell LD, et al. Five year outcomes of positive crossmatch kidney transplants: insight into late graft injury.
Am J Transplant 2012; 12: 227.
Terasaki PI. Humoral theory of transplantation.
Am J Transplant 2003; 3: 665.
Worthington JE, Martin S, Al-Husseini DM, et al. Posttransplantation production of donor HLA-specific antibodies as a predictor of renal transplant outcome.
Transplantation 2003; 75: 1034.
Lachmann N, Terasaki PI, Schonemann C. Donor-specific HLA antibodies in chronic renal allograft rejection: a prospective trial with a four-year follow-up.
Clin Transpl 2006: 171.
Ozawa M, Rebellato LM, Terasaki PI, et al. Longitudinal testing of 266 renal allograft patients for HLA and MICA antibodies: Greenville experience.
Clin Transpl 2006: 265.
Smith RN, Kawai T, Boskovic S, et al. Four stages and lack of stable accommodation in chronic alloantibody-mediated renal allograft rejection in cynomolgus monkeys.
Am J Transplant 2008; 8: 1662.
Einecke G, Sis B, Reeve J, et al. Antibody-mediated microcirculation injury is the major cause of late kidney transplant failure.
Am J Transplant 2009; 9: 2520.
Loupy A, Hill GS, Jordan SC. The impact of donor-specific anti-HLA antibodies on late kidney allograft failure.
Nat Rev Nephrol 2012; 8: 348.
Sellares J, de Freitas DG, Mengel M, et al. Understanding the causes of kidney transplant failure: the dominant role of antibody-mediated rejection and nonadherence.
Am J Transplant 2012; 12: 388.
Wiebe C, Gibson IW, Blydt-Hansen TD, et al. Evolution and clinical pathologic correlations of de novo donor-specific HLA antibody post kidney transplant.
Am J Transplant 2012; 12: 1157.
El-Zoghby ZM, Stegall MD, Lager DJ, et al. Identifying specific causes of kidney allograft loss.
Am J Transplant 2009; 9: 527.
Nankivell BJ, Borrows RJ, Fung CL, et al. The natural history of chronic allograft nephropathy.
N Engl J Med 2003; 349: 2326.
Everly MJ, Everly JJ, Arend LJ, et al. Reducing de novo donor-specific antibody levels during acute rejection diminishes renal allograft loss.
Am J Transplant 2009; 9: 1063.
Everly MJ, Rebellato LM, Ozawa M, et al. Beyond histology: lowering human leukocyte antigen antibody to improve renal allograft survival in acute rejection.
Transplantation 2010; 89: 962.
Lefaucheur C, Nochy D, Andrade J, et al. Comparison of combination plasmapheresis/IVIg/anti-CD20 versus high-dose IVIg in the treatment of antibody-mediated rejection.
Am J Transplant 2009; 9: 1099.
Racusen LC, Colvin RB, Solez K, et al. Antibody-mediated rejection criteria—an addition to the Banff 97 classification of renal allograft rejection.
Am J Transplant 2003; 3: 708.
Solez K, Colvin RB, Racusen LC, et al. Banff ‘05 Meeting Report: differential diagnosis of chronic allograft injury and elimination of chronic allograft nephropathy (‘CAN’).
Am J Transplant 2007; 7: 518.
