Despite improvements in stem cell transplant (SCT) procedures and supportive care, treatment of high risk (HR) AML patients remains a major challenge. In recent years, there have been few reports showing promising results of fast transplant (FTx) strategies applied to patients with uncontrolled AML. Aim of this retrospective analysis was to compare the FTx strategy to the well established approach of allogeneic SCT in complete remission (classical conditioning, CC) in high risk AML patients. Secondly, the impact of more extensive pretreatment in patients with uncontrolled AML was investigated.
We retrospectively analysed 111 consecutive patients (pts) with HR AML who received a first allogeneic SCT in our center between January 2000 and December 2009. HR AML was defined by (1) unfavorable cytogenetics (Mrozek and Bloomfield, Blood 2006), (2) AML secondary to myelodysplastic syndromes or prior radio/chemotherapy, (3) delayed response to induction chemotherapy, i.e. inadequate blast clearance on day 15 or persistent AML after the first course and/or (4) AML relapse.
Sixty pts who had achieved first (n=49), second (n=10) or third (n=1) complete remission (CR) at the time a donor with no or not more than one HLA mismatch was available were treated with a classical total body irradiation (TBI 12 or 8 Gy)-based (n=47) or intravenous busulfan (BU 16×0.8 mg/kg or 8×0.8mg/kg)-based (n=13) conditioning regimen (CC cohort). In contrast, 51 pts with active AML at the time of SCT received sequential treatment with intensive chemotherapy and reduced-intensity conditioning (RIC) as a fast-transplant strategy (FTx). The FLAMSA-RIC regimen (fludarabine (FLU), cytarabine, amsacrine from day -12 to -9, TBI 4 Gy on day - 5 (pts ≤ 60 yrs) or 4×0.8 mg/kg intravenous BU on day - 5 and -4 (pts > 60 yrs), and cyclophosphamide 40–60 mg/kg on day -3 and -2; Schmid C, Blood 2006) was applied to 26 pts. Twenty-five pts were treated with another course of intensive chemotherapy followed by FLU 30 mg/m2 from day -6 to -2 and melphalan 150 mg/m2 on day -3 given in aplasia (FLU/MEL; Platzbecker U, Leukemia 2006).
Patient's age ranged from 18 to 70 years and was significantly higher in the FTx than in the CC cohort (52 vs. 46 years; p=0.017). Of the leukemia-specific risk factors, the percentages of unfavorable cytogenetics did not differ between AML patients in the FTx and CC cohort (31 vs. 40%). Both groups were comparable with respect to the interval from first diagnosis of AML to SCT (median 244 vs. 247 days, range 17–2019 days), the use of unrelated donors (57 vs. 65%) and peripheral blood stem cells (93 vs. 88%). Graft versus host disease prophylaxis consisted of rabbit antithymocyte globulin which was equally balanced between the FTx and CC groups (47 vs. 53%), and cyclosporin A plus mycophenolate mofetil or methotrexate. In the FTx group, the median follow-up of surviving patients was significantly shorter than in the CC group (3.0 vs. 4.6 years, p=0.017).
Estimated median overall survival (OS) of the entire cohort was 5.7 years (95% CI, 4.8–6.6 years). Patients treated with CC showed significantly superior OS rates of 82%, 72% and 67% at 1, 2 and 4 years after SCT compared to 64%, 50% and 33% in the FTx group (p=0.001), due primarily to a significantly lower incidence of non-relapse mortality (10 vs. 25%, p=0.043). In contrast, the incidences of relapses after SCT were not different in both groups (23 vs. 35%).
Among the 51 patients of the FTx cohort, those transplanted with inaedequate blast clearance or persistent AML after the first induction course (n=27) had a better outcome than the more heavily pretreated patients with AML relapse (n=24). Median estimated OS in these groups was 4.0 vs.1.7 years and OS rates of 56 vs. 33% at 2 years (p=0.02). These groups did not differ significantly in terms of age, HCT-CI score (median 2.7) and bone marrow blast count at transplantation.
The fast transplant approach yields encouraging long-term survival in high-risk AML refractory to induction therapy, nearly approaching the results of patients transplanted in CR using classical conditioning. Among patients of the FTx cohort, relapsed AML with more extensive pretreatment had a significantly inferior outcome.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.