Whether or not the benefits of antithymocyte globulin (ATG) on engraftment and GVHD are offset by increased risk of relapse, delayed T-cell recovery and increased infections remains controversial. We retrospectively studied the effect of ATG in 144 AML patients, 34 of whom received ATG, undergoing reduced intensity conditioning (RIC) umbilical cord blood transplantation (UCB) or HLA-matched sibling PBSC. ATG patients had not received intensive chemotherapy for 3 months before transplantation for UCB, 6 months for PBSC. There were no differences in engraftment between ATG and non-ATG patients. The cumulative incidences of TRM as well as acute and chronic GVHD in ATG-treated patients were not statistically different. ATG patients had significantly more infections between 46 and 180 days post transplantation. Unexpectedly, after adjusting for donor type, relapse was lower among ATG recipients (relative risk (RR) 0.5, 95% confidence interval (CI) 0.3–1.0, P=0.04). In summary, administration of ATG to AML patients undergoing RIC had no adverse impact on major clinical outcomes. ATG may be indicated for patients at higher risk of graft failure after allogeneic hematopoietic cell transplantation (allo-HCT).
Allogeneic hematopoietic cell transplantation (allo-HCT) is an effective treatment modality for high risk and advanced hematological malignancies, including AML.1,2 Given that only about one-third of patients will have a suitable HLA-matched sibling (SIB) donor, more unrelated donors are being utilized.3 Further, with reduced intensity conditioning (RIC) regimens now approaching nearly half of all allo-HCT performed, more patients are undergoing allo-HCT with lower treatment-related mortality (TRM) and acceptable engraftment.4, 5, 6 The success of RIC relies heavily on the graft-versus-tumor effect of the donor cells as the conditioning regimen may be less effective in eradicating the underlying malignancy.
Antithymocyte globulin (ATG) can target T cells to achieve in vivo T-cell depletion and has often been administered to patients considered high risk for graft rejection and GVHD, such as those receiving HLA-mismatched allografts. Patients who have not received intensive or highly immunosuppressive chemotherapy during the few months before transplantation7 and those receiving unrelated donor or HLA-mismatched grafts8 are also at higher risk of graft rejection. Including ATG as part of the conditioning regimen may overcome this barrier. However, like any form of T-cell depletion, ATG may have benefits on engraftment and GVHD that can be offset by increased risk of relapse, delayed T-cell recovery and increased infections. We studied the ability of ATG as administered via our institutional protocol to facilitate engraftment as part of the conditioning regimen in patients with AML undergoing RIC transplantation from umbilical cord blood (UCB) or SIB donors.
Patients and methods
Transplant and demographic data were prospectively collected from 2000 to 2010 on all adult (age greater than 18) patients undergoing RIC UCB transplantation or HLA-matched SIB-PBSC for AML at the University of Minnesota. Eligibility criteria for non-myeloablative transplantation, UCB graft selection and supportive care have been reported.9,10 Hematopoietic cell transplantation comorbidity index (HCT-CI) scores were reviewed and assigned retrospectively.11 Cytogenetic risk as poor, intermediate or favorable was based on the Southwest Oncology/Eastern Cooperative Oncology Group classification.12 Infection data were collected prospectively; however, all data were confirmed by retrospective review of the outpatient and inpatient records. An infection episode was defined as any infection confirmed by culture, histology, PCR or antigenemia for which treatment was initiated. Treatment protocols were approved by the Institutional Review Board of the University of Minnesota. All patients provided written informed consent in accordance with the Declaration of Helsinki before enrolment.
Conditioning regimen and ATG administration
Conditioning regimens have been previously reported.9,10,13 Between 2000 and 2009, all patients received fludarabine 40 mg/m2 i.v. on day −6 through day −2 for a total dose of 200 mg/m2 (reduced to 30 mg/m2 per day for those with limited renal function, defined as raw creatinine clearance <70 mg/min/m2, and those with previous cranial radiation), cyclophosphamide (CY) 50 mg/kg i.v. on day −6, and single dose 200-cGy TBI on day −1. Starting in October 2009, the fludarabine dose in the SIB-PBSC was reduced to 30 mg/m2 per day. On the basis of the institutional guidelines, equine ATG (ATGAM; Pfizer, New York, NY, USA) at 15 mg/kg i.v. every 12 h for six doses on days −6, −5 and −4 with methylprednisolone 1 mg/kg was administered to those not treated with combination chemotherapy within 6 months before transplantation for SIB-PBSC and 3 months for UCB transplantation, or who had not undergone prior autologous transplantation. GVHD prophylaxis consisted of bis cyclosporine (CYA), targeting a trough level of 200–400 ng/mL, and mycophenolate mofetil 2–3 g/day, starting on day −3. Mycophenolate mofetil was discontinued at day +30 and CYA was continued through day +100 and, if no evidence of GVHD, tapered to discontinue on day +180.
The outcome measures of interest included: OS, disease-free survival (DFS), TRM, relapse, grades II–IV and III–IV acute GVHD (aGVHD), chronic GVHD (cGVHD), infection-related mortality and cumulative density of infections (infections per 1000 person days) for fungal, bacterial and viral organisms. The diagnosis of aGVHD and cGVHD was based on standard clinical criteria with histopathologic confirmation where possible.14,15 Differences in categorical variables between groups were evaluated by χ2 test or Fisher’s Exact test when appropriate. For continuous measures, P-values were calculated by the non-parametric general Wilcoxon test. Patient comorbidities were scored uniformly according to the HCT-CI.11 Standard disease risk was defined as AML in first or second CR at the time of transplantation. Time from disease to transplantation was measured from the date of diagnosis for patients in first CR (CR1) and from the latest relapse before transplantation for patients in CR2 or subsequent CRs (CR3+). Kaplan–Meier curves were used to estimate OS and DFS.16 Cumulative incidence was used to estimate relapse, infection-related mortality, aGHVD and cGVHD treating non-event death as a competing risk. Infection-related death was defined as any death in which infection was a primary or contributing cause of death, unless the primary or contributing causes of death were disease relapse/progression or GVHD. The log-rank test was used for comparisons across use of ATG in the conditioning regimen. Infections were further classified by type (bacterial, viral and fungal) and by time period (0–45 days, 46–180 days and 181–365 days after transplant). The infection density takes into account all episodes per patient. The Mantel–Haenszel χ2 test for incidence densities was employed to test the infection density in the two study groups, with and without ATG.17 Multivariable analysis for DFS used Cox regression18 and for relapse, TRM and aGVHD and cGVHD used Fine and Gray regression.19 Potential confounders included donor type (SIB-PBSC vs UCB), patient CMV serostatus (negative vs positive), comorbidity HCT-CI at transplant (0 vs 1–2 vs 3+), cytogenetic risk (good/intermediate vs poor), age (<35 vs 35–50 vs 50+), time from last disease to transplant (both continuous and <100 days vs ⩾100 days) and disease risk (CR1/CR2 vs CR3+). Multivariate analysis of infections, the correlation of multiple events within each subject, was taken into account by an appropriate correction to the variance estimate, and an appropriate risk set was defined for each infection, completed by the conditional model of Prentice for multiple infections of a similar type.20 In this model, a patient is not at risk for a second infection until the occurrence of the first infection. All P-values were two-sided. Analyses were performed using the SAS 9.2 (SAS Institute, Cary, NC, USA) and R 2.11 statistical software (Open-source software environment, by the Statistical Department at the University of Auckland, New Zealand).
A total of 144 patients transplanted between 2000 and 2010 met the inclusion criteria. Table 1 summarizes demographic information stratified by the administration of ATG. There were no significant differences between study groups in regard to HLA matching, disease status at treatment, prior autologous transplantation, gender, stem cell source, recipient and donor CMV serostatus, HCT-CI and cytogenetic risk. The median age was greater in the ATG group at 59 vs 55 years (P=0.03). As expected by study design, time from diagnosis to last relapse to transplantation was greater in the ATG group (median 156 days vs 119, P<0.01), reflecting the planned indications for ATG.
Engraftment, survival and relapse
There were no differences in the cumulative incidence and median days to engraftment between ATG and non-ATG patients. Among UCB patients, 92% of those treated with ATG (N=28) engrafted at a median of 20 days (range 5–39) vs 92% (N=76) and a median of 11 days (range 0–38) for the non-ATG group. Comparatively, among SIB-PBSC patients, 100% of those treated with ATG (N=6) engrafted at a median of 9 days (range 2–11) vs 100% (N=34) and a median of 9 days (range 0–16) for the non-ATG patients. Patients who did or did not receive ATG as part of their conditioning regimen had a similar incidence of TRM at 2 years (Figure 1a). In multivariable models, after adjusting for HCT-CI and disease risk, we found no effect of the administration of ATG and the risk of TRM (Table 2). Patients with HCT-CI⩾3 had a 4.6-fold higher risk of TRM. The cumulative incidence of relapse at 2 years was 30% (95% confidence interval (CI) 14–46%) for the ATG group and 46% (95% CI 36–57%) for non-ATG group (P=0.09; Figure 1b). In multiple regression modeling, after adjusting as needed for donor type, ATG administration was an independent predictor of decreased risk of relapse (Table 2). The Kaplan–Meier probability of DFS for patients who did and did not receive ATG was similar at both 1 year (ATG 59% (95% CI 41–73) and no ATG 40% (95% CI 31–49)) and 5 years (ATG 34% (95% CI 19–50%) and no ATG 24% (95% CI 15–33); Figure 1c). Multiple regression of DFS showed that ATG administration had no significant influence on the risk of treatment failure (Table 2). However, a HCT-CI score of 3+ risk ratio (RR) of 1.8, poor cytogenetic risk RR of 1.5 and high disease risk RR of 2.9 were associated with inferior DFS.
The cumulative incidence of grades II–IV aGVHD was not statiscally different between patients who did or did receive ATG (29% vs 50%, P=0.11, Figure 1d). The risk of cGVHD was also similar in the ATG and no ATG groups (24% vs 30%, P=0.33, Figure 1e). ATG was not an independent predictor of aGVHD in multiple regression analysis; however, increasing age was associated with an increased risk of aGVHD (Table 2). UCB was associated with significantly lower risks of cGVHD (RR 0.5, P=0.02) but ATG did not influence the risks of cGVHD in multiple regression analysis.
ATG administration had no influence on the incidence of infection-related deaths at 6 months (15% vs 8%, P=0.25) and 1 year (21% vs 10%, P=0.12; Figure 1f). We also studied the cumulative density of infections, a method that takes into consideration multiple infections in an individual patient. Recipients of ATG had similar incidence of infection between days 0–30 and days 180–365, but a higher incidence of viral and fungal infections between days +46 and +180 (Figures 2a and c). However, the number of fungal infection events in particular was very low. In multiple regression analysis, the administration of ATG was not an independent predictor of the cumulative density of bacterial, viral or fungal infections over the first year post transplantation. Patients with high-risk disease had a higher risk of fungal infections (Table 3) and recipients of UCB and CMV seropositive patients had an increased risk of viral infections (Table 3); however, there were no independent predictors of bacterial infections.
In this comprehensive single center analysis, we observed that the administration of equine ATG as part of the conditioning regimen had no adverse effect on GVHD, TRM and survival, and comparable engraftment. An unexpected finding was that ATG was associated with a lower risk of relapse. We analyzed both short-term and long-term outcomes in patients with AML in the setting of ATG use.
There are three frequently used clinical preparations of ATG. Polyclonal ATG is the purified IgG fraction of sera from rabbits, horses or, more rarely, goats that are immunized with thymocytes or T-cell lines. The most widely used preparations include horse ATG and the ATG preparation utilized in our study (Atgam, Pfizer), which is derived from horses immunized with thymocytes. Other commonly used preparations include another equine product (Lymphoglobulin, Genzyme, Cambridge, MA, USA) and rabbit ATG (Thymoglobulin, Genzyme).21 In contrast, ATG-Fresenius (ATG-F; Fresenius Biotech, Graefelfing, Germany) is a polyclonal Ab that is produced against rabbit anti-Jurkat cell-line resembling active T lymphoblasts. Recent data comparing rabbit and equine ATG for the treatment of severe aplastic anemia clearly illustrate the varying efficacy and degree of lymphodepletion between these agents.22 Studying the effect of ATG on the outcomes of allo-HCT should take into consideration not only dosing and timing of administration but also biological differences among preparations.
In our study, administration of ATG had no statistically significant effect on the risk of aGVHD and cGVHD. The retrospective registry study by the Center for International Blood and Marrow Transplant Research (CIBMTR) also observed no effect of ATG in reducing the risk of aGVHD but a modest effect in lowering cGVHD risks.23 In contrast, a randomized study with ATG-F24,25 and the sequential randomized studies with rabbit-ATG demonstrated a lower risk of aGVHD and cGVHD in treated patients.26,27 However, the rabbit-ATG study observed a reduction in the risk of cGVHD only when low-dose and high-dose ATG groups were combined.26,27 The differences among the various studies reflect, at least in part, the varying patient populations and varying ATG preparations (equine vs rabbit vs ATG-F). Moreover, our study and the ATG-F study24 described uniform dosing of a single formulation, whereas the rabbit-ATG report26 studied two different ATG doses and the CIBMTR study analyzed different products (including alemtuzumab), doses and schedules.23 Admittedly, our study participants primarily underwent UCB transplantation, whereas the CIBMTR registry study as well as the sequential randomized studies with rabbit-ATG23–25 included only PBSC transplantation patients; thus, one should compare the two with some degree of caution.
The effect of ATG on TRM and survival is controversial. Although we observed no adverse effect on TRM and survival, a recent CIBMTR retrospective study reported that ATG or alemtuzumab resulted in a higher risk of relapse and TRM and poorer survival as compared with T-cell-replete grafts.23 However, this study includes heterogeneity in anti-T-cell product, donor type and immune suppressive regimens that might have influenced these findings. In contrast, our findings are consistent with randomized studies using rabbit ATG26 and ATG-F24 that found no effect of ATG on the risk of TRM and survival.
An increased risk of infectious complications is a well-recognized risk of ATG and other forms of T-cell depletion.23,28,29 We studied infection density, which considers multiple infectious episodes, and also studied three distinct post-transplantation periods recognizing that the risk of infection changes over time. Early post transplantation, there was no difference in infection density, perhaps because of neutropenia and other profound immunosuppressive states in the initial post-transplant month. Recipients of ATG had a somewhat higher risk of viral and fungal, but not bacterial, infections between days +46 and +180. However, the number of infectious events, in particular fungal infections was low, suggesting that caution is warranted in drawing definite conclusions. Similar to other reports,26 there was no significant increase in infection-related death; in particular, EBV and CMV infections were infrequent in the ATG recipients, specifically when adjusted for pre-transplant CMV serology. These data support the safety of ATG use but also underscore the importance of close monitoring of ATG, and perhaps all, allograft recipients for opportunist infections up to 6 months post transplantation.
Although not seen in earlier studies, after adjusting for donor type we observed lower relative risk of AML relapse among patients receiving ATG. In the CIBMTR study, patients receiving ATG were found to have a higher risk of relapse as compared with T-cell-replete grafts,23 yet the ATG-F randomized study observed no difference between treated and untreated patients.24,25 The CIBMTR study included patients who received both equine and rabbit ATG; however, the authors were unable to determine whether the effect on the risk of relapse was the same for the two ATG preparations. Whereas the differing results among studies may reflect biological differences in ATG preparations, schedules of administration, disease risk groups, donor types and graft sources, we speculate that the observed differences may also be because of inherent differences in the patient risk of relapse. Our indications for ATG were only for patients who received no intensive chemotherapy 3–6 months before transplantation, which possibly selected patients with a less aggressive phenotype and thus a longer pre-transplant remission. Regardless, the administration of ATG did not adversely influence the risk of relapse.
We believe that the results of our experience in the administration of equine ATG for RIC AML patients have clinical utility. However, this was a retrospective study and the lower relapse in the ATG group, as well as the lack of effect on GVHD, TRM and survival, may reflect sample size and number of events. This is likely most evident in the reduced risk of relapse seen in the ATG group in multivariate analysis that may be biased by patient selection and/or numbers.
In summary, we showed that the administration of equine ATG for patients with AML had no adverse effect on long-term outcomes. In this older patient population, our observations support the use of ATG to reduce the risk of graft rejection as part of non-myeloablative conditioning.
Appelbaum FR . Haematopoietic cell transplantation as immunotherapy. Nature 2001; 411: 385–389.
Burnett AK, Wheatley K, Goldstone AH, Stevens RF, Hann IM, Rees JH et al. The value of allogeneic bone marrow transplant in patients with acute myeloid leukaemia at differing risk of relapse: results of the UK MRC AML 10 trial. Br J Haematol 2002; 118: 385–400.
Fuchs E, O'Donnell PV, Brunstein CG . Alternative transplant donor sources: is there any consensus? Curr Opin Oncol 2013; 25: 173–179.
Giralt S, Estey E, Albitar M, van Besien K, Rondon G, Anderlini P et al. Engraftment of allogeneic hematopoietic progenitor cells with purine analog-containing chemotherapy: harnessing graft-versus-leukemia without myeloablative therapy. Blood 1997; 89: 4531–4536.
Sandmaier BM, McSweeney P, Yu C, Storb R . Nonmyeloablative transplants: preclinical and clinical results. Semin Oncol 2000; 2: 78–81.
Slavin S, Nagler A, Naparstek E, Kapelushnik Y, Aker M, Cividalli G et al. Nonmyeloablative stem cell transplantation and cell therapy as an alternative to conventional bone marrow transplantation with lethal cytoreduction for the treatment of malignant and nonmalignant hematologic diseases. Blood 1998; 91: 756–763.
Maris MB, Niederwieser D, Sandmaier BM, Storer B, Stuart M, Maloney D et al. HLA-matched unrelated donor hematopoietic cell transplantation after nonmyeloablative conditioning for patients with hematologic malignancies. Blood 2003; 102: 2021–2030.
Andersson BS, de Lima M, Thall PF, Wang X, Couriel D, Korbling M et al. Once daily i.v. busulfan and fludarabine (i.v. Bu-Flu) compares favorably with i.v. busulfan and cyclophosphamide (i.v. BuCy2) as pretransplant conditioning therapy in AML/MDS. Biol Blood Marrow Transplant 2008; 14: 672–684.
Brunstein CG, Barker JN, Weisdorf DJ, DeFor TE, Miller JS, Blazar BR et al. Umbilical cord blood transplantation after nonmyeloablative conditioning: impact on transplantation outcomes in 110 adults with hematologic disease. Blood 2007; 110: 3064–3070.
Warlick ED, Tomblyn M, Cao Q, Defor T, Blazar BR, Macmillan M et al. Reduced-intensity conditioning followed by related allografts in hematologic malignancies: long-term outcomes most successful in indolent and aggressive non-Hodgkin lymphomas. Biol Blood Marrow Transplant 2011; 17: 1025–1032.
Sorror ML, Maris MB, Storb R, Baron F, Sandmaier BM, Maloney DG et al. Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood 2005; 106: 2912–2919.
Slovak ML, Kopecky KJ, Cassileth PA, Harrington DH, Theil KS, Mohamed A et al. Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 2000; 96: 4075–4083.
Barker JN, Weisdorf DJ, DeFor TE, Blazar BR, Miller JS, Wagner JE . Rapid and complete donor chimerism in adult recipients of unrelated donor umbilical cord blood transplantation after reduced-intensity conditioning. Blood 2003; 102: 1915–1919.
Filipovich AH, Weisdorf D, Pavletic S, Socie G, Wingard JR, Lee SJ et al. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant 2005; 11: 945–956.
Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J et al. 1994 Consensus Conference on Acute GVHD Grading. Bone Marrow Transplant 1995; 15: 825–828.
Kaplan EL, Meier P . Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.
Mantel N, Haenszel W . Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 1959; 22: 719–748.
Cox DR . Regression models and life tables. J Royal Stat Soc B 1972; 34: 187–220.
Fine JP, Gray RJ . A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 1999; 94: 496–509.
Prentice RL, Williams BJ, Peterson AV . On the regression analysis of multivariate failure time data. Biometrika 1981; 68: 373–379.
Mohty M . Mechanisms of action of antithymocyte globulin: T-cell depletion and beyond. Leukemia 2007; 21: 1387–1394.
Scheinberg P, Nunez O, Weinstein B, Scheinberg P, Biancotto A, Wu CO et al. Horse versus rabbit antithymocyte globulin in acquired aplastic anemia. N Engl J Med 2011; 365: 430–438.
Soiffer RJ, Lerademacher J, Ho V, Kan F, Artz A, Champlin RE et al. Impact of immune modulation with anti-T-cell antibodies on the outcome of reduced-intensity allogeneic hematopoietic stem cell transplantation for hematologic malignancies. Blood 2011; 117: 6963–6970.
Finke J, Bethge WA, Schmoor C, Ottinger HD, Stelljes M, Zander AR et al. Standard graft-versus-host disease prophylaxis with or without anti-T-cell globulin in haematopoietic cell transplantation from matched unrelated donors: a randomised, open-label, multicentre phase 3 trial. Lancet Oncol 2009; 10: 855–864.
Socie G, Schmoor C, Bethge WA, Ottinger HD, Stelljes M, Zander AR et al. Chronic graft-versus-host disease: long-term results from a randomized trial on graft-versus-host disease prophylaxis with or without anti-T-cell globulin ATG-Fresenius. Blood 2011; 117: 6375–6382.
Bacigalupo A, Lamparelli T, Bruzzi P, Guidi S, Alessandrino PE, di Bartolomeo P et al. Antithymocyte globulin for graft-versus-host disease prophylaxis in transplants from unrelated donors: 2 randomized studies from Gruppo Italiano Trapianti Midollo Osseo (GITMO). Blood 2001; 98: 2942–2947.
Bacigalupo A, Lamparelli T, Barisione G, Bruzzi P, Guidi S, Alessandrino PE et al. Thymoglobulin prevents chronic graft-versus-host disease, chronic lung dysfunction, and late transplant-related mortality: long-term follow-up of a randomized trial in patients undergoing unrelated donor transplantation. Biol Blood Marrow Transplant 2006; 12: 560–565.
Brunstein CG, Weisdorf DJ, Defor T, Barker JN, Tolar J, van Burik JA et al. Marked increased risk of Epstein-Barr virus-related complications with the addition of antithymocyte globulin to a nonmyeloablative conditioning prior to unrelated umbilical cord blood transplantation. Blood 2006; 108: 2874–2880.
Haspel RL, Kao G, Yeap BY, Cutler C, Soiffer RJ, Alyea EP et al. Preinfusion variables predict the predominant unit in the setting of reduced-intensity double cord blood transplantation. Bone Marrow Transplant 2008; 41: 523–529.
This work was supported in part by grants from the National Cancer Institute P01 CA65493 (CGB, JEW), the Children’s Cancer Research Fund (JEW, TED), Leukemia and Lymphoma Society Scholar in Clinical Research Award, grant R6029–07 (CGB).
The authors declare no conflict of interest.
About this article
Cite this article
Hagen, P., Wagner, J., DeFor, T. et al. The effect of equine antithymocyte globulin on the outcomes of reduced intensity conditioning for AML. Bone Marrow Transplant 49, 1498–1504 (2014). https://doi.org/10.1038/bmt.2014.183
A systematic literature review of incidence, mortality, and relapse of patients diagnosed with chronic graft versus host disease
Expert Review of Hematology (2019)
Low dose anti-thymocyte globulin reduces chronic graft-versus-host disease incidence rates after matched unrelated donor transplantation
Leukemia & Lymphoma (2018)
The Role of Anti-Thymocyte Globulin or Alemtuzumab-Based Serotherapy in the Prophylaxis and Management of Graft-Versus-Host Disease
Comparison of Conditioning Regimens with or without Antithymocyte Globulin for Unrelated Cord Blood Transplantation in Children with High-Risk or Advanced Hematological Malignancies
Biology of Blood and Marrow Transplantation (2015)
Biology of Blood and Marrow Transplantation (2015)