Myelodysplastic syndromes (MDSs) often occur in older adults with significant comorbidities. Therefore, a reduced-toxicity conditioning regimen may be more suitable than standard conditioning regimens before allogeneic blood stem cell transplantation. Here, we retrospectively compare the outcome of a treosulfan-based conditioning regimen with standard myeloablative TBI-based conditioning regimens in patients (pts) with MDS. A total of 48 pts with MDS were included in the study, of which 29 (60%) pts received TBI-based and 19 (40%) pts received a treosulfan-based conditioning regimen. A significantly lower relapse incidence (5% vs 34% at 3 years, P=0.019) resulting in a better, but not statistically significant relapse-free survival (RFS) (57% vs 31%, P=0.086) was observed after treosulfan-based conditioning. In pts with increased risk for significant side effects due to comorbidities (haematopoietic stem cell transplantation specific comorbidity index), the estimated 3-year RFS was significantly better in the treosulfan group: 54% (95% confidence interval (CI), 17–90%) compared with pts in the TBI group: 11% (95% CI, 0–44%; log-rank test P=0.0455). Treosulfan-based conditioning therapy is a feasible and effective regimen for pts with MDS, especially in pts with preexisting comorbidities.
Allogeneic haematopoietic stem cell transplantation (HCT) for patients (pts) with myelodysplastic syndrome (MDS) is a potentially curative treatment. However, as the majority of pts with MDS are of older age, they may not be eligible for standard high-dose myeloablative conditioning regimens due to severe side effects. Although recently introduced dose-reduced conditioning regimens allow transplantation of older pts, the lower intensity may result in an increased relapse risk. A conditioning regimen using treosulfan and fludarabine combines reduced toxicity with full intensity.1, 2, 3, 4, 5, 6
Treosulfan is a water-soluble bifunctional alkylating cytotoxic agent and has been evaluated in the therapy of solid tumours, including ovarian cancer.7 Because of the well-known advantages of a low non-haematologic toxicity profile, immunosuppressive as well as antileukaemic activity and broad stem cell toxicity, high-dose treosulfan was considered as an alternative agent in conditioning regimens before allogeneic HCT. The nucleoside analogue fludarabine is being increasingly used for reduced-intensity conditioning regimens. Fludarabine potentiates alkylator-induced cell killing through inhibition of DNA repair mechanisms.8 It has a favourable toxicity profile and its immunosuppressive effects, particularly on T-cell activity, make the substance suitable as a component of alternative conditioning regimens before allogeneic HCT.
As the treosulfan and fludarabine combination has been shown to be a promising treatment alternative to standard as well as reduced-intensity conditioning in a number of phase I and II studies,1, 2, 3, 4, 5, 6 we retrospectively compared pts suffering from MDS either treated with treosulfan-based conditioning regimens or TBI-based standard conditioning before allogeneic HCT. As a major negative impact of preexisting comorbid conditions on survival after allogeneic HCT was reported in earlier studies by Sorror et al.,9, 10 we considered the haematopoietic stem cell transplantation specific comorbidity index (HCT-CI) in the analysis.
Materials and methods
A retrospective analysis of treosulfan-based conditioning in comparison with standard conditioning was performed. Pts with acute myeloid leukaemia, MDS and non-Hodgkin lymphoma who received first allogeneic stem cell transplantation from an HLA identical donor between 1 January 1999 and 31 December 2005 at the Department of Bone Marrow Transplantation at the University Hospital Essen (n=234), and at the Division of Hematology and Oncology of the University of Rostock (n=62) were included for evaluation. Here, we report about the subgroup analysis of 48 pts with MDS (Essen n=39, Rostock n=9). The median follow-up of those pts surviving was 5.8 years (range, 2.5–9 years).
In total, 37 (77%) pts with primary MDS and 11 (23%) pts with therapy-related MDS were transplanted from HLA-compatible related (n=18) or matched unrelated (n=30) donors. The median patient age was 47.5 years (range, 18–68 years). A total of 27 (56%) pts were <50 years, whereas 21 (44%) pts were >50 years. The HCT-CI as described by Sorror et al.8, 9 was ⩽2 in 67% of pts and >2 in 33% pts.
In all, 29 (60%) pts received a TBI-based conditioning regimen with a total dose of at least 10 Gy (2.5 Gy on 4 consecutive days up to 3 × 4 Gy). Of these, 10 (21%) pts were treated with TBI/fludarabine and 19 (40%) pts with the combination of TBI and an alkylating agent. Nineteen (40%) pts received a treosulfan-based conditioning regimen (16 (33%) pts with treosulfan/fludarabine and 3 (6%) pts with treosulfan/CY). In cases of unrelated donors, rabbit antithymocyte globulin (10 mg/kg BW, day −4 to day −2) or alemtuzumab (50 mg total dose) was added. Acute GVHD prophylaxis was CYA based in all pts. A short course MTX was part of immunosuppressive therapy in all pts with the exception of pts with treosulfan/CY-based conditioning.
All HCT recipients were managed in laminar airflow rooms and received prophylaxis for bacterial, viral and fungal infections as well as pneumocystis carinii pneumonia according to local standards. A summary of the patient characteristics is shown in Table 1.
Evaluation of response
As surrogate parameter for toxicity after SCT, mucositis, morphine requirement (defined as at least 10 mg of morphine equivalent once after transplantation), onset of veno-occlusive disease and bilirubinaemia were evaluated according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTC AE Version 3.0). Acute and chronic GVHD were graded according to the standard criteria.11, 12 Chronic GVHD was evaluated in pts with a survival follow-up of at least 100 days.
Primary end points were non-relapse mortality (NRM), OS, relapse-free survival (RFS) and cytological incidence of relapse (RI). NRM was defined as the probability of dying without previous occurrence of relapse or persisting disease. Pts alive without relapse or persisting disease were censored at last contact. Persisting disease/relapse were considered as competing events. The NRM rates were estimated using cumulative incidence curves. OS was calculated from HCT to death of any cause. RFS was considered to be the time from HCT to death or disease progression/relapse. Survival curves and probabilities were estimated using the Kaplan–Meier method with the date of HCT as a starting point.
For statistical comparison of time-to-event data without competing events, the log-rank test was applied. Cumulative incidence curves in the presence of competing risks were compared using the test of Gray.
Comparison of ordinal variables between treatment arms were performed using the exact Wilcoxon–Mann–Whitney test. Fisher's exact test was applied in case of dichotomous variables. All reported P-values are two-sided and to be interpreted on an exploratory perspective.
Statistical analysis was performed using SAS (SAS Institute Inc., Cary, NC, USA) software package version 9.1.3 and R version 2.2.1 or higher (The R Foundation for statistical computing; http://www.r-project.org) on Windows platform.
The frequency of grade I/II mucositis was comparable between the TBI group and treosulfan group with 51.7% vs 57.9%, respectively. However, 20.7% vs 10.5% of pts developed CTC grade III/IV mucositis. Of note, in treosulfan-treated pts no grade IV mucositis was observed. However, the difference was not statistically significant (exact Mann–Whitney test, P=0.292).
None of the pts suffered from veno-occlusive disease. Bilirubinaemia grade III/IV developed in one patient of both groups, respectively, whereas 95.8% of pts had normal levels of bilirubin. Analgesia with opiates was required in 75.9% (22 pts) of pts after conditioning with TBI and in 57.9% (11 pts) of pts after treosulfan-based conditioning (Fisher's exact test, P=0.218).
The rates of acute GVHD until day +100 before any DLI or immunotherapeutic intervention were 72.4% (I–II° 58.6%, III–IV° 13.8%) in the TBI group and 58% (I–II° 36.9%, III–IV° 21.1%) in the treosulfan group (Fisher's exact test, P=0.001). The observed distribution of organ involvement of acute GVHD is shown in Table 2.
Chronic GVHD before any DLI or immunotherapeutic intervention occurred in 56% of pts after conditioning with TBI and in 72% of pts after treosulfan-based conditioning (Fisher's exact test, P=0.348). The proportion of extensive GVHD was 28% vs 50%, respectively (Fisher's exact test, P=0.204).
Incidence of cytological relapse
The RI at 100 days, 1 and 3 years after transplant (Figure 1a) was 3% (95% confidence interval (CI), 0–10%), 14% (95% CI, 1–26%) and 34% (95% CI, 16–53%) after TBI—compared with 0% (95% CI, 0–0%) at 100 days and 1 year and 5% (95% CI, 0–15%) 3 years after the treosulfan-based regimen (Gray test, P=0.019). Stratified by HCT-CI, no differences in RI were detected (Gray test, P=0.788). In the subgroup analysis, the RI after 3 years for pts with HCT-CI<2 was 35% (95% CI, 13–57%) after TBI conditioning and 0% (95% CI, 0–0%) after conditioning with treosulfan (Figure 1b, Gray test P=0.026) and for pts with HCT-CI>2 was 33% (95% CI, 5–61%) vs 14% (95% CI, 0–38%; Gray test P=0.348), respectively.
NRM and causes of death
The cumulative incidences of NRM (Figure 2a) at 100 days, 1 and 3 years after transplant were 14% (95% CI, 1–26%), 31% (95% CI, 14–48%) and 34% (95% CI, 17–52%) in the TBI group compared with 5% (95% CI, 0–15%), 26% (95% CI, 7–46%) and 38% (95% CI, 16–60%) in the treosulfan group (Gray test P=0.764). Of note, pts with therapy-related MDS had a lower 3-year NRM (19%; 95% CI, 0–42%) compared with pts with primary MDS (41%; 95% CI, 24–57%; Gray test P=0.111), whereas no differences were observed in NRM stratified by HCT-CI (Gray test P=0.725). However, pts with a HCT-CI>2 (Figure 2b) appeared to have a higher NRM in the TBI group with 22% (95% CI, 0–48%) vs 0% in the treosulfan group at day 100 and 56% (95% CI, 24–87%) vs 32% (95% CI, 0–65%) after 36 months (Gray test P=0.336), respectively.
All of the 10 pts in the TBI group with a transplantation-related cause of death died from infection, whereas in the treosulfan group four pts died from infectious disease and two pts from chronic GVHD. One patient in each group died from relapse.
Kaplan–Meier estimates of OS are illustrated in Figure 3. The estimated 1- and 3-year OS (Figure 3a) for pts receiving a TBI-based conditioning was 62% (95% CI, 44–80%) and 34% (95% CI, 14–55%) and for pts with a treosulfan-based conditioning was 74% (95% CI, 54–93%) and 57% (95% CI, 34–80%; log-rank test P=0.1764). For those pts with an HCT-CI<2, the estimated 3-year OS was 40% (95% CI, 16–64%) in the TBI group compared with 58% (95% CI, 31–86%) in the treosulfan group (P=0.545, log-rank test) and for those with a HCT-CI>2 (Figure 3b), 22% (95% CI, 0–56%) in the TBI group compared with 54% (95% CI, 17–90%) in the treosulfan group (P=0.151, log-rank test).
The estimated 1- and 3-year RFS for pts receiving a TBI-based conditioning was 55% (95% CI, 36–74%) and 31% (95% CI, 10–52%) and for pts with a treosulfan-based conditioning 74% (95% CI, 54–93%) and 57% (95% CI, 34–80%; log-rank test P=0.086). No differences were observed in RFS stratified by HCT-CI (P=0.579, log-rank test), but in the subgroup analysis for pts at high risk with an HCT-CI>2 (Figure 4), the estimated 3-year RFS was significantly better in the treosulfan group 54% (95% CI, 17–90%) compared with pts in the TBI group 11% (95% CI, 0–44%; log-rank test P=0.046). However, no difference in the RFS was observed for pts with an HCT-CI<2.
For many years, pts with MDS have received standard-intensity conditioning regimens. Improved understanding of transplantation biology in recent years has led to the development of dose-reduced conditioning regimens, which have made it possible to offer allogeneic HCT to pts in higher-age groups and/or with comorbidities. However, the 3-year OS still ranges from 20 to 48%, with a resulting 3-year RFS between 22 and 40% and a 3-year NRM ranging from 22 to 37%.13, 14, 15, 16 The combination of treosulfan and fludarabine has recently been shown to be a toxicity reduced but full-intensity conditioning approach1, 2, 17 and first disease specific phase II protocols have shown its feasibility.3, 5, 18, 19
In this retrospective analysis, we evaluated the role of treosulfan-based reduced-toxicity conditioning in comparison with TBI-based standard conditioning and incorporated the HCT-CI in the assessment of outcomes after allogeneic HCT for pts with MDS.
Surprisingly and despite the small number of pts, the relapse rate was significantly lower after treosulfan-based conditioning compared with the standard high-dose TBI-based approach (5% at 3 years vs 34%, P=0.019). This may be the result of antitumour activity through allogeneic immunity as a beneficial effect of chronic GVHD in MDS pts was reported,20 and the rate of chronic GVHD was higher in the treosulfan group (TBI 56% vs treosulfan 72%, P=0.348). On the other hand, in a prospective treosulfan/fludarabine phase II study,19 the cumulative incidence of chronic GVHD (59%) was in the range of the historic TBI group in this evaluation and the relapse rate after treosulfan/fludarabine conditioning still remained considerably low with 16% after 2 years, suggesting a beneficial cytotoxic effect of the treosulfan-based conditioning regimen. Another difference between the two treatment groups has been the allocation of secondary/therapy-related MDS types almost exclusively into the treosulfan group. However, one would rather be inclined to anticipate a higher relapse rate in secondary MDS types.
Despite the predominant allocation of secondary/therapy related MDS into the treosulfan group, the NRM was not increased in comparison to the TBI group. In pts with comorbidities (HCT-CI>2), treosulfan-based conditioning was associated with a significantly better 3-year RFS (54% vs 11%; log-rank test P=0.0455) and a trend to a better 3-year OS (54% vs 22%; log-rank test P=0.1514) compared with TBI-based conditioning.
Drawbacks of the present study are the retrospective character and the small sample size, which allow only a limited interpretation of the results. Nevertheless, the long follow-up as well as the confirmation of the results by the prospective phase II study indicate that the treosulfan-based conditioning might combine the desired effects of reduced toxicity and high antileukaemic activity: because of the higher incidence of MDS in pts of higher age, conditioning with a treosulfan-based regimen may, therefore, offer a promising alternative to standard conditioning regimens, but prospective randomized studies are required to validate the superiority. A larger prospective randomized study in AML and MDS pts currently under way in several European countries may help to further define the role of treosulfan and fludarabine in allogeneic HSCT.
Casper J, Knauf W, Kiefer T, Wolff D, Steiner B, Hammer U et al. Treosulfan and fludarabine: a new toxicity-reduced conditioning regimen for allogeneic hematopoietic stem cell transplantation. Blood 2004; 103: 725–731.
Casper JK, Wolff D, Knauf W, Blau IW, Ruutu T, Volin L et al. Allogeneic hematopoietic stem cell transplantation in patients with hematologic malignancies after dose-escalated treosulfan/fludarabine conditioning. J Clin Oncol 2010 (e-pub ahead of print).
Holowiecki J, Giebel S, Wojnar J, Krawczyk-Kulis M, Markiewicz M, Holowiecka-Goral A et al. Treosulfan and fludarabine low-toxicity conditioning for allogeneic haematopoietic stem cell transplantation in chronic myeloid leukaemia. Br J Haematol 2008; 142: 284–292.
Kröger N, Shimoni A, Zabelina T, Schieder H, Panse J, Ayuk F et al. Reduced-toxicity conditioning with treosulfan, fludarabine and ATG as preparative regimen for allogeneic stem cell transplantation (alloSCT) in elderly patients with secondary acute myeloid leukemia (sAML) or myelodysplastic syndrome (MDS). Bone Marrow Transplantation 2006; 37: 339–344.
Schmidt-Hieber M, Blau IW, Trenschel R, Andreesen R, Stuhler G, Einsele H et al. Reduced-toxicity conditioning with fludarabine and treosulfan prior to allogeneic stem cell transplantation in multiple myeloma. Bone Marrow Transplant 2007; 39: 389–396.
Shimoni A, Hardan I, Shem-Tov N, Rand A, Yerushalmi R, Nagler A . Fludarabine and treosulfan: a novel modified myeloablative regimen for allogeneic hematopoietic stem-cell transplantation with effective antileukemia activity in patients with acute myeloid leukemia and myelodysplastic syndromes. Leuk Lymphoma 2007; 48: 2352–2359.
Gropp M, Meier W, Hepp H . Treosulfan as an effective second-line therapy in ovarian cancer. Gynecol Oncol 1998; 71: 94–98.
Gandhi V, Plunkett W . Cellular and clinical pharmacology of fludarabine. Clin Pharmacokinet 2002; 41: 93–103.
Sorror ML, Maris MB, Storer B, Sandmaier BM, Diaconescu R, Flowers C et al. Comparing morbidity and mortality of HLA-matched unrelated donor hematopoietic cell transplantation after nonmyeloablative and myeloablative conditioning: influence of pretransplantation comorbidities. Blood 2004; 104: 961–968.
Sorror ML, Sandmaier BM, Storer BE, Maris MB, Baron F, Maloney DG et al. Comorbidity and disease status based risk stratification of outcomes among patients with acute myeloid leukemia or myelodysplasia receiving allogeneic hematopoietic cell transplantation. J Clin Oncol 2007; 25: 4246–4254.
Glucksberg H, Storb R, Fefer A, Buckner CD, Neiman PE, Clift RA et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HL-A-matched sibling donors. Transplantation 1974; 18: 295–304.
Atkinson K, Horowitz MM, Gale RP, Lee MB, Rimm AA, Bortin MM . Consensus among bone marrow transplanters for diagnosis, grading and treatment of chronic graft-versus-host disease. Committee of the International Bone Marrow Transplant Registry. Bone Marrow Transplant 1989; 4: 247–254.
Barrett AJ, Savani BN . Allogeneic stem cell transplantation for myelodysplastic syndrome. Semin Hematol 2008; 5: 49–59.
Laport GG, Sandmaier BM, Storer BE, Scott BL, Stuart MJ, Lange T et al. Reduced-intensity conditioning followed by allogeneic hematopoietic cell transplantation for adult patients with myelodysplastic syndrome and myeloproliferative disorders. Biol Blood Marrow Transplant 2008; 14: 246–255.
Sierra J, Pérez WS, Rozman C, Carreras E, Klein JP, Rizzo JD et al. Bone marrow transplantation from HLA-identical siblings as treatment for myelodysplasia. Blood 2002; 100: 1997–2004.
Lim Z, Brand R, Martino R, van Biezen A, Finke J, Bacigalupo A et al. Allogeneic hematopoietic stem-cell transplantation for patients 50 years or older with myelodysplastic syndromes or secondary acute leukemia. J Clin Oncol 2009; 28: 405–411.
Nemecek ER, Sorror ML, Doney KC, Scott BL, Kovacsovics TJ, Maziarz RT et al. A regimen of treosulfan and fludarabine followed by allogeneic hematopoietic cell transplantation (HCT) for high-risk hematologic malignancies is associated with low treatment-related mortality. Blood 2008; 112: 1029 (ASH Annual Meeting Abstracts 2994).
Beelen DW, Gromke T, Trenschel R, Hilgendorf I, Wolff D, Elmaagacli AH et al. Treosulfan compared to total body irradiation-based preparative regimens before allogeneic stem cell transplantation for acute myeloid leukemia: a retrospective long-term study. Blood 2008; 112: 748 (ASH Annual Meeting Abstracts 2149).
Ruutu T, Beelen DW, Finke J, Holowiecki J, Uharek L, Kienast J et al. Reduced-toxicity conditioning with treosulfan and fludarabine in allogeneic haematopoietic stem cell transplantation for myelodysplastic syndromes: final results of an international prospective phase II trial. Bone Marrow Transplantation 2009; 43: 556 (Suppl 1) (abstract O300).
Valcárcel D, Martino R, Caballero D, Martin J, Ferra C, Nieto JB et al. Sustained remissions of high-risk acute myeloid leukemia and myelodysplastic syndrome after reduced-intensity conditioning allogeneic hematopoietic transplantation: chronic graft-versus-host disease is the strongest factor improving survival. J Clin Oncol 2008; 26: 577–584.
U Pichlmeier is an employee of medac GmbH.
About this article
Cite this article
Hilgendorf, I., Wolff, D., Gromke, T. et al. Retrospective analysis of treosulfan-based conditioning in comparison with standard conditioning in patients with myelodysplastic syndrome. Bone Marrow Transplant 46, 502–509 (2011) doi:10.1038/bmt.2010.153
- haematopoietic stem cell transplantation
- myelodysplastic syndrome
- reduced-toxicity conditioning
Disposition of treosulfan and its active monoepoxide in a bone marrow, liver, lungs, brain, and muscle: Studies in a rat model with clinical relevance
European Journal of Pharmaceutical Sciences (2017)
European Society for Blood and Marrow Transplantation Analysis of Treosulfan Conditioning Before Hematopoietic Stem Cell Transplantation in Children and Adolescents With Hematological Malignancies
Pediatric Blood & Cancer (2016)
Fludarabine and treosulfan compared with other reduced-intensity conditioning regimens for allogeneic stem cell transplantation in patients with lymphoid malignancies
Bone Marrow Transplantation (2015)
Treosulfan-based conditioning regimens for allogeneic HSCT in children with acute lymphoblastic leukaemia
Annals of Hematology (2015)
Conditioning with treosulfan and fludarabine for patients with refractory or relapsed non-Hodgkin lymphoma
Molecular and Clinical Oncology (2014)