Treosulfan added to fludarabine demonstrated an improvement in event-free and overall survival compared with busulfan/fludarabine as a treatment for patients with acute myeloid leukaemia or myelodysplastic syndrome who are scheduled for allogeneic hematopoietic stem cell transplantation but are considered ineligible for standard myeloablative conditioning.
Dietrich Beelen, MD, PhD
Treosulfan added to fludarabine demonstrated an improvement in event-free and overall survival (OS) compared with busulfan/fludarabine as a treatment for patients with acute myeloid leukaemia (AML) or myelodysplastic syndrome (MDS) who are scheduled for allogeneic hematopoietic stem cell transplantation (HSCT) but are considered ineligible for standard myeloablative conditioning (MAC), according to phase III findings presented at the 2019 European Society for Blood and Marrow Transplantation Annual Meeting.1
Data also suggested that the agent could also be a standard preparative regimen for those with these diseases who have comorbidities and/or are elderly.
“The large randomized trial demonstrated that the modified treosulfan/fludarabine regimen is significantly noninferior to the busulfan/fludarabine reduced intensity regimen in elderly and/or comorbid patients,” explained Dietrich Beelen, MD, PhD, director, Department of Bone Marrow Transplantation, University Hospital of Duisburg-Essen in Essen, Germany. “[Event-free survival, relapse-free survival], and complete donor chimerism point to a potential benefit with the treosulfan regimen in elderly and comorbid patients with AML or MDS.”
Beelen presented the full analysis of data from this prospective, randomized, international multicenter phase III trial (NCT00822393; EudraCT-No: 2008-002356-18) comparing a treosulfan-based myeloablative procedure to an established busulfan-based, reduced-intensity conditioning regimen prior to allogeneic HSCT in elderly and comorbid patients.
“Allogeneic HSCT remains a challenge, particularly in elderly and comorbid patients with AML and MDS, because this patient population has an increased risk of non-relapse mortality when treated with standard myeloablative regimens,” said Beelen.
Trial recruitment had been terminated early after confirmatory interim analysis of data from 476 patients demonstrated significant noninferiority of treosulfan to busulfan, thereby meeting the primary study objective.2
The phase III study enrolled patients with MDS or AML in complete hematological remission who were scheduled for allogeneic HSCT and were ineligible for MAC due to age or comorbidity. The patients were aged 50 to 70 years or younger than 18 years with a Hematopoietic Cell Transplantation Comorbidity (HCTC) index >2. Patients were also required to have Karnofsky Performance Index ≥60% and a related or unrelated donor with ≥9/10 matched HLA class I or II allele identities. Patients were stratified according to the transplant center, risk (low/intermediate versus high/very high, according to European LeukemiaNet classification), and by matched related donor (MRD) versus matched unrelated donor (MUD) type.
Patients were randomized 1:1 and treated with treosulfan at 10 g/m2 intravenously (IV) daily on days 4 to 2 (n = 283) or busulfan at 3.2 mg/kg IV daily on days 4 to 3 (n = 268). Both regimens were combined with fludarabine at 30 mg/m2 IV daily on days 6 to 2 and standardized pre and post transplant immunosuppressive treatment with cyclosporin and short course methotrexate. MUD patients also received anti thymocyte globulin.
The median age in both groups was 60 years (range, 31 to 70); the median HCT-Comorbidity Index was 3 (QI-Q3: 1-4). Slightly more patients in the busulfan (95.8%) than treosulfan (94.0%) arms were aged ≥50 years and 59.0% versus 58.2% of patients had HCT-CI ≥3, respectively. The majority of patients were male.
AML and MDS were diagnosed in 57.5% and 42.5% of patients in the busulfan arm and in 70.5% and 29.5% of patients in the treosulfan group. Of these, 47.2% on busulfan and 53.2% of those on treosulfan were high risk due to an unfavorable first complete remission (CR) or more than 1 CR (patients with AML) or according to the Revised International Prognostic Scoring System (patients with MDS).
Transplant characteristics were similar between groups. The predominant donor type was MUD (75.4% busulfan and 76.4% of treosulfan arms, respectively) and the stem cell source was peripheral blood for 97.9% of patients receiving busulfan and 97.3% of those on treosulfan.
Event free survival (EFS), including relapse, graft failure, or death, within 24 months served as the primary study endpoint; secondary endpoints were acute non-hematological toxicities (days -6 to +28), engraftment and donor chimerism, relapse incidence, non-relapse mortality (NRM), relapse-free survival (RFS), and acute versus chronic graft-versus-host disease (GVHD).
At a median 29-month follow-up, results showed that EFS and OS significantly favored treosulfan; the 24-month EFS rate was 51.2% with busulfan versus 65.7% with treosulfan (HR, 0.64; P = .0000001). Additionally, EFS was greater with treosulfan than busulfan across donor type (MUD, HR, 0.63; MRD, HR, 0.93), disease types (AML, HR, 0.72; MDS, HR, 0.66), age (<50 years, HR, 0.74; ≥50 years, HR, 0.69), and HCT-CI score (≤2, HR, 0.68; >2, HR, 0.70).
The 24-month OS rates were 60.2% with busulfan versus 72.7% with treosulfan, (HRcrude, 0.71; HRadjusted 0.64; P = .0037). Treosulfan OS rates were greater across all parameters save MRD type (HR, 1.16).
The 24-month NRM rates were 20.4% with busulfan versus 12.0% with treosulfan, (HRadjusted, 0.63; P = .0343). All parameters evaluated, including disease type, age, and donor type favored treosulfan.
At 28 days post-allogeneic HSCT, neutrophil recovery was similar (>0.5 x 109 cells/L) with both regimens, but complete donor type chimerism was 83.0% with busulfan compared with 93.2% with treosulfan (P = .0159).
More patients receiving busulfan (57.2%) experienced acute GVHD than those treated with treosulfan (52.0%); however, the rates of grade 3/4 GVHD were similar.
Regarding safety, both groups showed similar frequencies of early adverse events, with 96.1% and 92.6% of patients receiving busulfan and treosulfan experiencing any grade of Common Terminology Criteria for Adverse Events (CTCAEs), respectively. Grade 3/4 CTCAEs occurred in 53.4% and 54.8% of patients on busulfan and treosulfan, respectively; of these, 29.0% and 26.7% were determined as regimen-related, respectively.
Serious adverse events (SAEs) occurred in 7.1% of patients on busulfan and 8.5% of those on treosulfan, respectively, with 2.8% versus 4.8% of these being determined as life-threatening; 2.1% and 3.0% of SAEs in the respective arms were fatal.
The incidence of primary graph failure was 0.4% in both groups, with only busulfan-treated patients (2.9%) experiencing secondary graph failure.
“The study results in terms of the secondary endpoints of OS, NRM, and EFS point to a potential benefit with the treosulfan-based regimen that suggest this new regimen has the potential to become a standard preparative regimen prior to allogeneic HCT in this target population,” Beelen concluded.