Polyclonal antilymphocyte globulins have been widely used as induction therapy in renal transplantation. With the advent of more effective immunosuppression in the cyclosporine A (CsA) era, the need for antilymphocyte globulins, particularly as induction treatment, has been questioned. Many transplant centers continue to use some type of antilymphocyte antibody in recipients with delayed graft function or with a high risk of acute rejection. Support for induction therapy is found in a meta-analysis studying the effect of antilymphocyte antibody induction therapy on renal allograft survival in patients treated with CsA-based triple immunosuppression therapy (1).
Nevertheless, polyclonal antithymocyte globulins (ATGs) are powerful immunosuppressive agents that may expose the transplant recipients to a significantly increased risk of infection in the early posttransplant period and malignancy in the late period. Therefore, the benefit to risk ratio is questionable in low-risk patients (2).
Different ATG preparations are commercially available, but little is known about the clinical equivalence of these different polyclonal antilymphocyte globulins in terms of long-term efficacy and safety. Recently, some studies have demonstrated that Thymoglobulin (SangStat, Fremont, CA), a rabbit anti-human thymocyte globulin, is superior to AT-GAM, a horse anti-human thymocyte globulin, for prevention of acute rejection (3, 4). Few prospective studies have compared Thymoglobulin and ATG Fresenius (ATGF), two rabbit-derived ATGs, in renal transplantation (5, 6). These studies failed to show any significant difference between the two agents. Nevertheless, the interpretation of these trials is largely hampered by the small size of the study population and the short period of follow-up.
We performed a retrospective, single-center study to compare the occurrence of different posttransplant events according to the use of ATGF or Thymoglobulin in renal transplant recipients.
SUBJECTS AND METHODS
Study Population
This retrospective, single-center study evaluates the respective influence of two polyclonal ATGs on the occurrence of different posttransplant events. Participants in the study were 194 consecutive patients who underwent renal transplantation in our center between June 1993 and April 2001.
The mode of dialysis (hemodialysis, peritoneal dialysis, or none) and its duration before transplantation were recorded. Age, gender, history of cardiovascular disease, malignancy, and diabetes mellitus, and prior renal transplantation and panel reactive antibody were also analyzed as covariates.
Data concerning relevant donors (age, serum creatinine [sCt] level, and collapsus during reanimation) were collected. Information on kidney transplant (cold ischemia and human leukocyte antigen compatibility status) was also gathered.
Immunosuppressive Treatment
From June 1993 to April 2001, 236 renal transplantations were performed in our center. Forty two patients were included in multi-center trials and excluded from this analysis. A total of 194 patients received our local immunosuppressive protocol. Induction consisted of a short course of polyclonal ATGs with ATGF (day 0: 9 mg/kg; days 1–4: 3 mg/kg) or Thymoglobulin (day 0: 5 mg/kg; days 1–4: 2 mg/kg until May 1996; day 0: 3 mg/kg, days 1–4: 1 mg/kg June 1996–May 1998; day 0: 2 mg/kg, days 1–4: 1 mg/kg from June 1998). The induction protocol in our center was initially commenced with ATGF. This product was not available in France from February 1995 to August 1996. During this period we decided to use Thymoglobulin, as previously mentioned. Since August 1996, the two polyclonal globulins were available and used alternatively (every second patient to either drug) to prevent any problem in provision of these agents by the manufacturers. The dose of Thymoglobulin was decreased over time because of early adverse events including fever and leukopenia.
A total of 180 patients received CsA and 14 patients received Tacrolimus. These anticalcineurin inhibitors were introduced once sCt concentration had decreased to half of the pretransplant level, but not after day 10 posttransplant. A total of 151 patients received azathioprine and 43 patients received mycophenolate mofetil. All patients received steroids according to the same schedule.
Posttransplant Events
All patients had the same schedule of posttransplant follow-up. After discharge from the hospital, they had at least weekly visits in our clinic until month 4 posttransplant. Afterward, they were seen twice per month in the first year and then monthly.
Acute Rejection
Acute rejection was searched for in any case of sCt elevation. Only biopsy-proven rejections were considered. Acute rejection was defined and recorded according to Banff classification.
Cytomegalovirus Disease
No patient received anti-cytomegalovirus (CMV) prophylaxis until May 2000. Thereafter, CMV-negative recipients with CMV-positive donors received valacyclovir for a 3-month period. CMV infection was surveyed by serum CMV polymerase chain reaction assay once per week until month 4 posttransplant. CMV disease was defined by a positive serum polymerase chain reaction assay or antigenemia in patients manifesting relevant clinical or biologic symptoms (e.g., fever, leukopenia, and liver enzyme abnormalities). All patients were treated with ganciclovir. No patient received preemptive treatment.
Malignancy
All cases of malignancies were recorded, and cutaneous cancers were excluded.
Death
All patients who were deceased and the causes of death were assessed through medical records.
Statistical Analysis
Arithmetic mean was calculated and expressed as ± standard deviation. Using log-rank tests on Kaplan Meier nonparametric estimates of the survival without event distribution (CMV disease, acute rejection, malignancy, graft loss [death censored], and death), we selected variables with a P value lower than or equal to 0.20. The selected variables were included into a Cox proportional hazards model, and a backward stepwise selection process was performed (this time at a classic α=0.05). Age was split into two classes separated by its median (51 years). Because the dose of Thymoglobulin had varied with time, the variables were also calculated for each of the different therapeutic periods (see Discussion). Results are expressed as relative risk (RR) and 95% confidence interval (CI), with a P value testing the null hypothesis (RR=1). Therefore when the P value is less than 0.05, the RR is significantly different from 1, either greater than 1 (i.e., risk is increased) or less than 1 (i.e., risk is decreased).
RESULTS
Study Population
The characteristics of the study population are summarized in Tables 1 and 2. A total of 129 patients received ATGF, and 65 patients received Thymoglobulin. The pre-transplant demographic and clinical variables were similar in the two groups (Table 1). Also, donor characteristics were not different in these two treatment groups (Table 2). The average follow-up period was 1,749 and 1,405 days for patients receiving ATGF and Thymoglobulin, respectively.
;)
Table 1: Demographic and clinical characteristics of the study population at baseline
Table 2: Donor characteristics
Posttransplant Events
Cytomegalovirus Disease.
Thirty patients (23%) who had received ATGF and 24 patients (37%) who had received Thymoglobulin developed CMV disease (P =0.02) (Table 3). In monovariate analysis, age (P =0.008), male gender (P =0.13), dialysis duration (P =0.08), smoking status (P =0.04), donor and recipient CMV status (P =0.009), and Thymoglobulin (P =0.04) were associated with CMV disease. The same proportion of patients in the two treatment groups had received antiviral prophylaxis.
Table 3: Incidence of outcomes according to polyclonal antithymocyte globulins
After backward stepwise selection, the variables that remained in the Cox proportional hazards model (i.e., variables linked to CMV disease with P <.05) were age, Thymoglobulin, and donor and recipient CMV status (Table 4). Cox regression analysis revealed that Thymoglobulin (RR 2.16, CI 95% [1.04–4.48]) was an independent predictor of the subsequent development of CMV disease.
Table 4: Cox multivariate analysis of endpoints*
Malignancy
Five patients (3.9%) who had received ATGF and eight patients (12.3%) who had received Thymoglobulin developed posttransplant malignancy (P =.01). This corresponds to 9 and 38 malignancies in 1,000 patients per year in the ATGF and Thymoglobulin groups, respectively. Notably, patients in the Thymoglobulin group developed malignancy significantly earlier than patients in the ATGF group (364±245 days vs. 653±365 days; P=0.008).
The malignancy sites were the following: prostate (1), lymphoma (2), kidney (1), and breast (1) in the ATGF group, and prostate (1), lymphoma (3), bladder (1), tongue (1), lung (1), and liver (1) in the Thymoglobulin group. Lymphoma represented 38.5% of all malignancies. There was a trend toward a higher incidence of lymphoma in the Thymoglobulin group compared with the ATGF group (4.6% vs. 1.5%; P=0.1). In monovariate analysis, age (P =0.14) and Thymoglobulin (P =0.01) were found to be associated with increased incidence of malignancy.
After backward stepwise selection, the variables that remained in the Cox proportional hazards model (i.e., variables linked to malignancy with P <0.05) were age and Thymoglobulin (Table 4). Cox regression analysis revealed that Thymoglobulin (RR 2.16, CI 95% [1.04–4.48]) was an independent predictor of the subsequent development of malignancy.
Death
Five patients (3.9%) who had received ATGF and nine patients (13.8%) who had received Thymoglobulin died during the follow-up period (P =0.005). This corresponds to 9 and 42 deaths in 1,000 patients per year in the ATGF and Thymoglobulin groups, respectively. The rate of immunosuppression-related death (malignancy and infection) was 3.6 and 23.3 in 1,000 patients per year in the ATGF and Thymoglobulin groups, respectively.
The causes of death were the following: trauma (1), cirrhosis (1), cerebrovascular disease (1), and systemic infections (2) in the ATGF group, and unknown (2), malignancy (4), cerebrovascular disease (1), abdominal aneurysm (1), and infection (1) in the Thymoglobulin group.
In monovariate analysis, age (P =0.01), history of cardiovascular disease (P =0.005), and Thymoglobulin (P =0.008) were associated with increased frequency of death.
After backward stepwise selection, the variables that remained in the Cox proportional hazards model (i.e., variables linked to cardiovascular events with P <0.05) were age and Thymoglobulin (Table 4). Cox regression analysis revealed that Thymoglobulin (RR 4.14, CI 95% [1.36–12.6]) was an independent predictor of the death.
Graft Loss (Death Censored)
The rate of graft loss was not different in the two treatment groups. The only predictive factor of graft loss was a 1-month posttransplant sCt greater than the median value (RR 4.14, CI 95% [1.36–12.6]).
Acute Rejection
The rate of acute rejection was not different in the two treatment groups. We could not identify any predictive factor of acute rejection.
DISCUSSION
The main conclusion of our retrospective study is that induction treatment with Thymoglobulin in renal transplant recipients carries an increased risk of CMV disease and post-transplant malignancy compared with ATGF, without any significant beneficial effect on the rate of acute rejection. We found that Thymoglobulin-treated patients demonstrated a twofold increased risk of CMV disease compared with ATGF-treated patients. Krogsgaard et al. (7) have previously reported similar results regarding CMV infection in heart transplant recipients. In this study, the authors reported that patients who were treated with Thymoglobulin developed CMV disease more frequently than patients who were treated with ATGF. In renal transplantation, Norrby et al. (5) also found a greater, although not significant, incidence of CMV disease in Thymoglobulin-treated recipients compared with ATGF-treated recipients.
In our study, posttransplant malignancies were also more frequent in patients who received Thymoglobulin compared with patients who received ATGF. Of note, this difference was mainly because of an increased incidence of lymphoma in the Thymoglobulin group. Several studies have demonstrated that the incidence of malignancy is increased after organ transplantation (8, 9), and that the use of polyclonal ATGs has been shown to be a risk factor for the subsequent development of lymphoma (10). Nevertheless, to date no study has reported a difference in the rate of posttransplant lymphoma with the use of different polyclonal agents.
We also observed an increased incidence of death in the Thymoglobulin group compared with the ATGF group. Indeed, patients who were treated with Thymoglobulin showed a fourfold increased risk of death compared with patients who were treated with ATGF. This increased risk was even greater when only immunosuppression-related deaths were considered. These results indicate that Thymoglobulin is a more aggressive immunosuppressive agent than ATGF and carries a significantly increased risk of viral and neoplasic complications.
Thymoglobulin is a purified immunoglobulin solution produced by the immunization of rabbits with human thymocytes. The solution is pasteurized to increase viral safety, concentrated to 5 mg/mL, and provided in lyophilized form. Thymoglobulin is known to consist of antibodies specific for T-cell epitopes including CD2, CD3, CD4, CD8, CD11a, CD18, CD25, human leukocyte antigen-DR, and human leukocyte antigen class I (11). ATGF is also a rabbit immune globulin. It is, however, obtained by immunizing rabbits with cells of the Jurkat line, an immortalized human T-lymphocyte cell line. The differences in preparation may account for differences in inhibition activities against lymphocyte antigens. Indeed, some studies have demonstrated different inhibition activities of these two polyclonal antibodies against specific lymphocyte antigens (12–15). Thymoglobulin has a higher inhibition activity of specific T-cell CD antigen than ATGF, and blocking of CD4 is especially more potent with Thymoglobulin (14). The two agents also have different inhibition activities against B cells (13, 14). B-cell depletion has been shown to be more intense with Thymoglobulin than with ATGF (16). These bioimmunologic differences might have clinical consequences in term of immunosuppression-related complications, as found in this study.
Few prospective studies have compared Thymoglobulin and ATGF in renal transplantation. Norrby et al. (5) reported the short-term follow-up of 90 renal transplant recipients randomized to receive ATGF or Thymoglobulin. They failed to show any difference in the incidence of acute rejection and infectious complications, and patient and graft survival. Ourahma et al. (6) compared the efficacy and safety of these agents in 70 renal transplant recipients. At 3 months post-transplant, there was a trend toward a lesser incidence of graft loss in ATGF-treated patients. Both the small size of the study population and the short period of follow-up hamper the magnitude of the results. Compared with these reports, our study includes a greater number of patients and provides a much longer period of follow-up.
There are limitations to our study. First, the dose of Thymoglobulin has been modified over time. However, once the incidence of the endpoints was evaluated for each dose period, the results were found to be similar with the final global one (Tables 5 and 6). Indeed, the great variations in dose and duration of ATG therapy from one center to another makes it difficult to find an exact dose equivalence between Thymoglobulin and ATGF. Globally, the recommended total dose for Thymoglobulin in the literature varies from 10 to 40 mg/kg. In our study, the total dose of Thymoglobulin varied from 17 mg/kg to 6 mg/kg over time. Thus, even when using the maximal doses, total exposure to Thymoglobulin was in the recommended range for all of the patients.
Table 5: Incidence of the endpoints in Thymoglobulin group according to different dose periods of Thymoglobulin
Table 6: Incidence of endpoints in ATGF group according to different dose periods of Thymoglobulin
Second, a selection bias could not be excluded because of the retrospective nature of the study. Nevertheless, patients who were treated with Thymoglobulin did not differ from those who were treated with ATGF at baseline, and our results were adjusted for the major donor and recipient covariates associated with posttransplant outcomes.
Third, a bias of difference in evaluation, although unlikely, could not be excluded. Indeed, all patients were followed by the same physicians in our clinic with the same systematic schedule of evaluation. All data were assessed through medical records, and the rate of missing data was low (<3%) in both treatment groups.
This retrospective study, which concerns two widely used immunosuppressive agents, might be considered a preliminary report that needs to be confirmed by prospective randomized trials with long periods of follow-up.
REFERENCES
1. Szczech LA, Berlin JA, Aradhye S, et al. Effect of antilymphocyte induction therapy on renal allograft survival: a meta analysis. J Am Soc Nephrol 1997; 8: 1771.
2. Meier-Kriesche H, Arndorfer JA, Kaplan B. Association of antibody induction with short- and long-term cause-specific mortality in renal transplant recipients. J Am Soc Nephrol 2002; 13: 769.
3. Gaber AO, First MR, Tesi RJ, et al. Results of the double-blind, randomised, multicenter, phase III clinical trial of thymoglobulin versus AT-GAM in the treatment of acute graft rejection episodes after renal transplantation. Transplantation 1998; 66: 29.
4. Brennan DC, Flavin K, Lowell JA, et al. A randomised, double-blinded comparison of Thymoglobulin versus Atgam for induction immunosuppressive therapy in adult renal transplant recipients. Transplantation 1999; 67: 1011.
5. Norrby J, Olausson M. A randomised clinical trial using ATG Fresenius or ATG Merieux as induction therapy in kidney transplantation. Transplant Proc 1997; 29: 3135.
6. Ourahma S, Talon D, Barrou B, et al. A prospective study on efficacy and tolerance of antithymocyte globulin Fresenius versus Thymoglobuline Merieux after renal transplantation. Transplant Proc 1997; 29: 2427.
7. Krogsgaard K, Boesgaard S, Aldershvile J, et al. Cytomegalovirus infection rate among heart transplant patients in relation to anti-thymocyte immunoglobulin induction therapy. Copenhagen Heart Transplant Group. Scand J Infect Dis 1994; 26: 239.
8. Kyllönen L, Salmella K, Pukkala E. Cancer incidence in a kidney-transplanted population. Transpl Int 2000; 13(Suppl): S394.
9. Brunner FP, Landais P, Selwood NH. Malignancies after renal transplantation: the EDTA-ERA registry experience. European Dialysis and Transplantation Association-European Renal Association. Nephrol Dial Transplant 1995; 10(Suppl 1): 74.
10. Opelz G, Henderson R. Incidence of non-Hodgkin lymphoma in kidney and heart transplant recipients. Lancet 1993; 342(8886–8887): 1514.
11. Bonnefoy-Bernard N, Flacher M, Revillard JP. Anti-proliferative effect of antilymphocyte globulins on B-cells and B-cell lines. Blood 1992; 79: 2164.
12. Bonnefoy-Berard N, Vincent C, Revillard JP. Antibodies against functional leukocyte surface molecules in polyclonal antilymphocyte and antithymocyte globulins. Transplantation 1991; 51: 669.
13. Rebellato LM, Gross U, Verbanac KM, et al. A comprehensive definition of the major antibody specifications in polyclonal rabbit antithymocyte globulins. Transplantation 1994; 57: 685.
14. Bourdage JS, Hamlin DM. Comparative polyclonal antithymocyte globulin and antithymocyte/antilymphoblast globulin anti-CD antigen analysis by flow cytometry. Transplantation 1995; 59: 1194.
15. Wakabayashi H, Cook DJ, Posner M, et al. Batch to batch variations in rabbit antithymocyte globulin preparations. Transplant Proc 1991; 23: 1125.
16. Muller TF, Grebe SO, Neumann MC, et al. Persistent long-term changes in lymphocyte subsets induced by polyclonal antibodies. Transplantation 1997; 64: 1432.
