Acute rejection of renal allografts is one of the major risk factors for chronic allograft nephropathy (CAN) and graft loss (1, 2). The diagnosis and treatment of acute rejection are based on the distinction between acute cellular- and antibody-mediated rejection (3). Of these two types of acute rejection, acute cellular rejection (ACR) is believed to be mediated by T cells, but recent reports suggested that B cells also participate in T-cell–mediated rejection (4, 5).
Two predominant subsets of B cells can be identified at the time of ACR. One is CD20+ B cells, which are specific for B cell only and expressed during mid-developing stages of B cells. The other subset is CD38+ cells, which are expressed in the late-lineage B cells such as plasmablasts and antibody-secreting plasma cells. Recent phenotypic analysis revealed that CD20+ B cells appear in cluster formations and CD38+ B cells appear in diffuse infiltrates into the renal interstitium (6, 7).
Impact of CD20+ cells on clinical outcome is still controversial. Some reports showed that infiltration of CD20+ B cells is correlated with steroid resistance and worse graft survival (6–9). Conversely, other reports showed no relationship between CD20+ B cells and clinical outcome (10–13). In contrast to CD20+ B cells, the clinical significance of infiltration by CD38+ B cells is not understood well. We attempted to examine the predictive value of the presence of CD38+ B cells alone or in combination with CD20+ B cells for rejection outcome and graft survival.
PATIENTS AND METHODS
Patients and Clinical Information
This study was approved by the Institutional Review Board of Seoul St. Mary's hospital. Between January 2000 and August 2008, 426 kidney transplants were performed in our transplantation center; 75 (17.6%) patients had biopsy-proven ACR according to the Banff '97 working classification of renal allograft pathology (14). Of these patients, 24 were excluded because of Banff grade III (n=5), treatment with rituximab (n=5), unavailable remnant tissues for observation (n=6), or BK or cytomegalovirus nephropathy (n=5). Lymphoproliferative disorders and CAN grade III were not present in these patients or biopsies. All biopsies were taken to evaluate episodes of graft dysfunction defined as serum creatinine increment by more than or equal to 10% from the baseline value. Patients were followed up from the date of transplantation to the date of nephrectomy, permanent dialysis, retransplantation, or patient death.
Relevant demographic and clinical information was collected retrospectively. Patients were classified into CD20+ or CD38+ patients if at least one of their biopsies met the positive criteria for CD20 or CD38 staining. We compared the baseline characteristics, late-onset and repeated ACR, steroid resistance, incomplete recovery after rejection treatment, and allograft survival according to positivity for CD20 and CD38 stains alone and in combination. If the first ACR developed more than 6 months after transplantation, it was defined as late-onset ACR. Patients who developed a second or third ACR were classified as repeated ACR. The estimated glomerular filtration rate was calculated by the modified diet in renal disease formula (15), and the baseline estimated glomerular filtration rate was calculated from the stable serum creatinine concentration 2 to 4 weeks before a rejection episode. Incomplete functional recovery was defined if the antirejection treatment did not recover graft function to within 10% of the baseline value. The allograft survival was censored in cases of patient death with a functioning graft.
Immunosuppression and Treatment of Acute Cellular Rejection
The initial immunosuppressant was cyclosporine A or tacrolimus (FK506) in combination with mycophenolate mofetil and prednisolone after transplantation. Basiliximab was used as an additional induction therapy in 23 (42.6%) patients. The initial dose of cyclosporine A was 10 mg/kg per day by oral route; the target trough levels were 200 to 400 ng/mL in the first 4 weeks and 100 to 200 ng/mL thereafter. The initial dose of FK506 was 0.16 mg/kg per day by oral route, and target trough levels were 8 to 15 ng/mL in the first 3 months and 3 to 8 ng/mL thereafter. Methylprednisolone (1 g/day) was administered by intravenous infusion on the day of transplantation and was then tapered to prednisolone at 30 mg/day on day 4 after transplantation. The initial dose of mycophenolate mofetil was 1.5 g/day, and the dose was modified to minimize the adverse effects such as diarrhea or leukopenia. During the study period, ACR was treated with three to four daily boluses of intravenous methylprednisolone (500 mg/day), followed by a 5- to 7-day oral steroid taper. Failure of serum creatinine concentration to decrease within 5 days or the requirement for antithymocyte globulin or muromonab-CD3 (OKT3) was classified as steroid resistance.
Renal Tissues and Histopathologic Evaluation
All 67 ACR biopsies were examined for the presence of CD20+ clusters, CD38+ infiltration, and diffuse C4d staining. Of the 67 biopsy samples, 54 biopsies were for first-time ACR, and the other specimens were repeated ACR biopsy samples (nine biopsies for the second time and four biopsies for the third time). Immunohistochemical (IHC) staining was performed using CD20 (Dako, Carpinteria, CA) and CD38 (Novocastra, Newcastle, UK) monoclonal antibodies in all rejection tissues. The positivity for each IHC stain was examined in a blinded fashion relative to the clinical information. Representative stains of CD20 and CD38 are shown in Figure 1(A and B). For each specimen, the single high-power field with the highest CD20 count was identified, and the threshold cell count more than 275 in the selected high-power field was defined as CD20+. Based on the extent of staining in nonclustering inflammatory cells, more than 30% infiltration of CD38-stained cell was defined as CD38+. Indirect immunofluorescence staining was performed using monoclonal antibodies against complement protein C4d (Biogenesis, Poole, UK) in 39 (58.2%) biopsies. In 28 (41.8%) biopsies where no C4d staining was performed on frozen sections, the sections were obtained from paraffin blocks and stained for IHC with C4d using a rabbit polyclonal antibody (Biogenesis). C4d positivity was defined as diffuse (>50%) and linear staining of peritubular capillaries.
Descriptive statistical values are presented as mean±SD. Continuous variables with a normal distribution were analyzed by Student's t test or analysis of variance; A P value less than or equal to 0.05 indicated significant differences. Nonparametric variables were compared using the Mann-Whitney U or Kruskal-Wallis test as appropriate. The chi-square test was used for categorical variables. Graft survival was analyzed by the Kaplan-Meier method with a log-rank test. Statistical analysis was performed using SPSS software version 15.0 (SPSS Inc., Chicago, IL).
Baseline Characteristics and Clinical Outcomes in the Study Population
The mean age of the entire sample was 39±12 years, and 31 patients (57.4%) were men. Two patients (3.7%) received a second kidney transplant, and eight patients (14.8%) received an allograft from a deceased donor. All patients were negative in the complement-dependent cytotoxicity crossmatch test, and four patients (7.4%) were positive in the B-cell flow cytometric crossmatch (B-FCXM) test. Twenty-three patients (42.6%) were CD20+, 25 (46.3%) were CD38+, and 15 (27.8%) were positive for both CD20 and CD38 (CD20+CD38+).
During the follow-up of a mean of 4 years after transplantation, 21 patients (38.9%) developed late-onset ACR and 11 (20.4%) developed repeated ACR. Ten patients (18.5%) experienced steroid-resistant rejection, and 29 (53.7%) experienced incomplete recovery from ACR. Eighteen patients (33.3%) progressed to immunologic (death censored) allograft loss.
Associations Between Banff Grade and CD20, CD38, and C4d Positivity in Renal Biopsies
Of the 67 biopsy specimens, 25 (37.3%) showed CD20+ cluster and 29 (43.3%) showed CD38+ cell infiltration (Table 1). The renal tissue biopsies from the second or third ACR displayed a similar degree of CD20+ and CD20− cell infiltration (P=0.924). The Banff grades for ACR and CAN did not differ between CD20+ and CD20− biopsies (P=0.375 and P=0.542, respectively). CD20+ biopsy samples were more likely to be simultaneously positive for CD38 than were CD20− biopsies (P=0.033). C4d positivity did not differ between CD20+ and CD20− samples (P=0.767). CD38 positivity were observed more frequently in the renal biopsies from the second or third ACR (P=0.011). The Banff ACR and CAN grades did not differ between CD38+ and CD38− patients (P=0.534 and P=0.062, respectively), but the rate of C4d positivity was greater in CD38+ biopsies than in CD38− biopsies (P=0.019).
Comparison of Baseline Characteristics and Clinical Outcomes Between CD20+ and CD20− Patients
The baseline clinical characteristics did not differ between CD20+ and CD20− patients (Table 2). The clinical outcomes of CD20+ and CD20− patients are compared (Fig. 2A–D). The incidence of late-onset and repeated ACR was 52.2% and 30.4%, respectively, in CD20+ patients, and it did not differ significantly from that of CD20− patients (P=0.085 for 29.0% in late-onset ACR, P=0.173 for 12.9% in repeated ACR). However, steroid resistance (39.1% vs. 3.2%, P=0.001) and incomplete recovery from ACR (87.0% vs. 29.0%, P<0.001) occurred significantly more frequently in CD20+ patients than in CD20− patients. The 5-year allograft survival rate was 41.5% in CD20+ patients and 75.4% in CD20− patients (Fig. 2E); Kaplan-Meier analysis showed that this rate was significantly lower in CD20+ than in CD20− patients (P=0.003).
Comparison of Baseline Characteristics and Clinical Outcomes Between CD38+ and CD38− Patients
The baseline characteristics did not differ between CD38+ and CD38− patients (Table 3). CD38+ patients were older at the time of transplantation, and positivity in the B-FCXM was observed in only CD38+ patients (P=0.029 vs. CD38− patients). The ACR characteristics were compared between CD38+ patients and CD38− patients (Fig. 3A–D). Late-onset ACR (P=0.017), repeated ACR (P=0.015), and incomplete recovery (P=0.002) occurred more frequently in CD38+ patients, but the incidence of steroid resistance did not differ between each other (P=0.159). The 5-year allograft survival rate was illustrated in Figure 3E. The survival rate was significantly lower in CD38+ patients (39.8%) than in CD38− patients (75.3%; P=0.011).
Baseline Characteristics in Four Subgroups Defined by CD20 and CD38 Positivity
Fifteen of the 54 (27.8%) patients exhibited both CD20 and CD38 positivity (Table 4). Eight (14.8%) patients were positive for CD20 alone, and 10 (18.5%) patients were positive for CD38 alone. The subgroups did not differ at baseline except for age and B-FCXM. CD20−CD38+ patients were older at transplantation than were CD20+CD38− patients (P=0.003). Of four patients positive on the B-FCXM test, two were CD20−CD38+ and two were CD20+CD38+ (P=0.015).
Clinical Outcomes in Patients With Both CD20 and CD38 Positivity
We compared the clinical outcomes between CD20+CD38+ patients and the other patients (Fig. 4A–D). The rates of late-onset (P=0.017) and repeated ACR (P=0.013), steroid resistance (P=0.003), and incomplete recovery (P<0.001) were significantly higher in CD20+CD38+ patients than in CD20−CD38− patients. The incomplete recovery rate was also higher in CD20+CD38+ patients than in CD20+CD38− and CD20−CD38+ patients (P=0.032 and P=0.001, respectively). CD20+CD38− patients were more resistant to steroid than were CD20−CD38− patients. However, the rates of late-onset or repeated ACR and incomplete recovery did not differ between CD20+CD38− patients, CD20−CD38+ patients, and CD20−CD38− patients.
The 5-year allograft survival rate was 24.0% in CD20+CD38+ patients (Fig. 4E). The allograft survival rate was significantly lower in CD20+CD38+ patients than in CD20−CD38−, CD20+CD38−, and CD20−CD38+ patients (P=0.001, P=0.049, and P=0.018, respectively). The graft survival rates did not differ between CD20−CD38−, CD20+CD38−, and CD20−CD38+ patients (all P values >0.05).
This study demonstrates high incidence of CD38+ (46.3%) and CD20+ B-cell infiltration. The presence of CD38+ B cells was associated with critical factors such as late onset ACR, repeated ACR, and incomplete recovery of graft function that determine the prognosis of ACR (16–20). Patients with CD20 or CD38+ B cells had a lower graft survival rate compared with patients lacking each B-cell type, and patients exhibiting positivity for both CD20 and CD38 had worse clinical outcomes than patients with CD20+CD38−. These findings suggest that CD38+ B-cell infiltration is not rare in ACR and that the positivity for CD38 alone or in combination with CD20 is important for determining the prognosis in renal transplant recipients with ACR.
In the previous reports, the presence of CD20+ B cells was associated with steroid resistance, incomplete recovery, and reduced graft survival (6–9). Our data for CD20+ B cells were consistent with these reports, and we found that CD38+ B cells is also associated with incomplete recovery and reduced graft survival. However, CD38+ B cells showed different relationships from CD20+ B cells in the other clinical parameters. Late-onset or repeated ACR was observed more frequently in CD38+ patients, but it was not observed in CD20+ patients. These findings suggest that the clinical impact of B cells differs according to the subtype and that staining for CD38 and CD20 may provide detailed information for predicting the long-term prognosis in patients with ACR.
Although the presence of CD20+ B cells was significantly associated with poorer clinical outcomes compared with the lack of CD20 positivity, this effect was more evident in patients who were simultaneously positive for CD38. The rate of incomplete recovery was 1.6-fold higher (62.5% vs. 100%), and the 5-year graft survival rate was approximately two third lower (80.3% vs. 24.0%) in CD20+CD38+ patients than in CD20+CD38− patients. These findings suggest that CD38+ infiltration is an important factor for determining the prognostic relevance of CD20+ B cells, and this may explain the negative results of some reports that did not evaluate the CD38 positivity in CD20+ ACR patients (10–13).
The patients with CD20+ B cells were not associated with late and repeated ACR, but this clinical association was demonstrated in patients who also exhibited CD38+ infiltration. Compared with CD20−CD38− patients, late-onset and repeated ACR were significantly greater in CD20+CD38+ patients. With regards to treatment responsiveness, CD20+CD38− and CD20+CD38+ patients showed similar rates of steroid resistance (37.5% vs. 40.0%), and both rates were significantly higher than that of CD20−CD38− patients. These findings suggest that CD20+ B cells were associated with late-onset and repeated ACR by the simultaneous positivity for CD38 and that the presence of CD20+ clusters is a reliable marker of steroid resistance regardless of CD38+ infiltration. Therefore, two B-cell types should be assessed at the same time to clarify the role of these B-cell subsets in patients with ACR.
The mechanism responsible for the role of CD38+ B cells in repeated or late-onset ACR is undetermined. One possible mechanism may relate to the B-cell developmental stages. CD38+ B cells are generated from memory B cells or ectopic germinal centers by reexposure to antigen (21, 22). Therefore, repeated and continuous antigen exposure might cause CD20+ B cells to differentiate into CD38+ B cells. If so, this may explain two findings in our study—the higher rates of CD38 positivity in CD20+ biopsies (60.0%) than in CD20− biopsies (18.4%) and the higher rates of CD38 positivity in the second or further ACR biopsies (34.5%) than in the first ACR biopsies (7.9%). Furthermore, many CD38+ B cells involved in graft rejection are plasmablasts and plasma cells (7), which can produce antidonor antibody in situ in the renal allograft (23, 24). We also found a high correlation between CD38 positivity and C4d deposition (44.8% C4d+ in CD38+ biopsies vs. 18.4% C4d+ in CD38− biopsies). This finding suggests that CD38+ B cells can cause severe graft injury through an antibody-mediated mechanism.
In this study, 15 patients (27.8%) exhibited positivity for both CD20 and CD38. These patients had the highest correlation between the four clinical outcomes and the worst allograft survival rates compared with all other patients. In addition, the allograft survival rates of patients with CD20+ or CD38+ alone did not differ from that of CD20−CD38− patients. These findings suggest that the positivity for both CD20 and CD38 is a powerful predictor of poor clinical outcome in renal transplant recipients with ACR and that CD20+ and CD38+ B cells were complementary to strengthen their prognostic relevance. Furthermore, these B cells expressing different IHC marker are linked to each other, because many CD20+ biopsies contained CD38+ infiltration. Therefore, the positivity for CD20 and CD38 should be considered as one consistent rather than a separate IHC marker.
The results of our study suggest that B cells play an important role in the clinical outcomes of ACR and that inclusion of B-cell subtypes in the assessment of ACR may be helpful in determining the therapeutic strategy. Rituximab has already shown to be effective against the refractory rejection with CD20+ clusters (25, 26). In advance, by demonstrating the association of CD38+ B cells with poorer clinical outcomes, we can consider another intervention for the effective reversal of immunologic injury. Bortezomib, a recently developed proteasome inhibitor, is effective in reducing the donor-specific antibody and plasma cell apoptosis (27, 28). Therefore, bortezomib may be useful in the treatment of late-onset or repeated ACR with CD38+ B-cell infiltration.
In conclusion, infiltration of CD38+ B cells is reliable marker for poor clinical outcomes and reduced graft survival in renal transplant recipients, and it is more prominent in combination with CD20+ B cells. The demonstration of CD38+ B cells is critical for determining the prognosis of CD20+ patients, and it could provide the opportunity to improve the long-term graft survival through the targeted rejection therapy.
1. Meier-Kriesche HU, Ojo AO, Hanson JA, et al. Increased impact of acute rejection on chronic allograft failure in recent era. Transplantation
2000; 70: 1098.
2. Flechner SM, Modlin CS, Serrano DP, et al. Determinants of chronic renal allograft rejection in cyclosporine-treated recipients. Transplantation
1996; 62: 1235.
3. 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.
4. Sarwal M, Chua MS, Kambham N, et al. Molecular heterogeneity in acute renal allograft rejection identified by DNA microarray profiling. N Engl J Med
2003; 349: 125.
5. Tsai EW, Rianthavorn P, Gjertson DW, et al. CD20+
lymphocytes in renal allografts are associated with poor graft survival in pediatric patients. Transplantation
2006; 82: 1769.
6. Zarkhin V, L Li, Sarwal M. “To B or not to B?” B-cells and graft rejection. Transplantation
2008; 85: 1705.
7. Zarkhin V, Kambham N, Li L, et al. Characterization of intra-graft B cells
during renal allograft rejection. Kidney Int
2008; 74: 664.
8. Hippen BE, DeMattos A, Cook WJ, et al. Association of CD20+
infiltrates with poorer clinical outcomes in acute cellular rejection
of renal allografts. Am J Transplant
2005; 5: 2248.
9. Martins HL, Silva C, Martini D, et al. Detection of B lymphocytes (CD20+
) in renal allograft biopsy specimens. Transplant Proc
2007; 39: 432.
10. 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.
11. Kayler LK, Lakkis FG, Morgan C, et al. Acute cellular rejection
-positive lymphoid clusters in kidney transplant patients following lymphocyte depletion. Am J Transplant
2007; 7: 949.
12. Doria C, di Francesco F, Ramirez CB, et al. The presence of B-cell nodules does not necessarily portend a less favorable outcome to therapy in patients with acute cellular rejection
of a renal allograft. Transplant Proc
2006; 38: 3441.
13. Eikmans M, Roos-van Groningen MC, Sijpkens YWJ, et al. Expression of surfactant protein-C, S100A8, S100A9, and B cell markers in renal allografts: Investigation of the prognostic value. J Am Soc Nephrol
2005; 16: 3771.
14. Racusen LC, Solez K, Colvin RB, et al. The Banff 97 working classification of renal allograft pathology. Kidney Int
1999; 55: 713.
15. Levey AS, Bosch JP, Lewis JB, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Ann Intern Med
1999; 130: 461.
16. Sijpkens YW, Doxiadis JJ, Mallat MJ, et al. Early versus late acute rejection episodes in renal transplantation. Transplantation
2003; 75: 204.
17. Opelz G. Critical evaluation of the association of acute with chronic graft rejection in kidney and heart transplant recipients. The Collaborative Transplant Study. Transplant Proc
1997; 29: 73.
18. Dickenmann MJ, Nikeleit V, Tsinalis D, et al. Why do kidney grafts fail? A long-term single center experience. Transpl Int
2002; 15: 508.
19. Humar A, Kerr S, Gillingham KJ, et al. Features of acute rejection that increase risk for chronic rejection. Transplantation
1999; 68: 1200.
20. Pascual M, Theruvath T, Kawai T, et al. Strategies to improve long-term outcomes after renal transplantation. N Engl J Med
2002; 346: 580.
21. Avery DT, Ellyard JI, Mackay F, et al. Increased expression of CD27 on activated human memory B cells
correlates with their commitment to the plasma cell lineage. J Immunol
2005; 174: 4034.
22. Cassese G, Lindenau S, de Boer B, et al. Inflamed kidneys of NZB/W mice are a major site for the homeostasis of plasma cells. Eur J Immunol
2001; 31: 2726.
23. Charney DA, Nadasdy T, Lo AW, et al. Plasma cell-rich acute renal allograft rejection. Transplantation
1999; 68: 791.
24. Xu X, Shi B, Cai M, et al. A retrospective study of plasma cell infiltrates in explanted renal allografts. Transplant Proc
2008; 40: 1366.
25. Lehnhardt A, Mengel M, Pape L, et al. Nodular B-cell aggregates associated with treatment refractory renal transplant rejection resolved by rituximab. Am J Transplant
2006; 6: 847.
26. Alausa M, Almagro U, Siddiqi N, et al. Refractory acute kidney transplant rejection with CD20
graft infiltrates and successful therapy with rituximab. Clin Transplant
2005; 19: 137.
27. Perry DK, Burns JM, Pollinger HS, et al. Proteasome inhibition causes apoptosis of normal human plasma cells preventing alloantibody production. Am J Transplant
2009; 9: 201.
28. Everly MJ, Everly JJ, Susskind B, et al. Bortezomib provides effective therapy for antibody- and cell-mediated acute rejection. Transplantation
2008; 86: 1754.