The outcome of young adult (< 60 years) with acute myeloid leukemia (AML) still remains unsatisfactory. In fact, in spite of complete remission (CR) rates ranging from 60 to 80%, only 30–40% of young patients will be long-term survivors. Advances in biologic characterization of AMLs are expected to enhance a more realistic assessment of disease aggressiveness so that therapies will be delivered in the context of a stratified approach. Cytogenetic/genetic profile is the most relevant prognostic factor established at diagnosis. Nevertheless, it is well recognized that it cannot always reliably predict outcome in individual patients. Minimal residual disease (MRD) detection promises to be an efficient tool to establish on an individual basis the leukemia's susceptibility to treatment and guide delivery of risk-tailored therapies. A further element underlying the dismal long-term outcome of young patients with AML pertains the chance to get access to allogeneic stem cell transplantation (ASCT) when carrying high-risk features. The extensive use of ASCT option is precluded by the paucity of full matched family donor (25–30%). These premises are the background to the risk-adapted approach, developed at the Institute of Hematology, University Tor Vergata, based on the following strategies: 1) combination of upfront cytogenetics/genetics and MRD status (< or ≥3.5×10−4 residual leukemic cells as counted by flow cytometry) at the end of consolidation to determine risk assignment; 2) once a given patients was categorized as high-risk (due to the expression of an unfavorable karyotype, FLT3-ITD positivity or post-consolidation positive MRD status) and therefore selected as candidate for ASCT, the transplant procedure was given whatever the source of stem cells. The present analysis includes 30 high-risk patients treated according to this design (prospective cohort = PC) and, for comparative purposes, 55 consecutive high-risk patients treated in an “old fashion” design based on donor availability (retrospective cohort = RC). The PC included 4 patients with favorable-karyotype (FK) and a MRD+ status, 12 with intermediate kayotype (IK) and a MRD+ status, 5 with unfavorable karyotype (UK) and 9 with FLT3-ITD mutation. The RC included 8 FK/MRD+, 34 IK/MRD+, 1 UK and 12 with FLT3-ITD mutation. In the PC, 22 (73%) of 30 patients received ASCT (8 matched family donor, 7 matched unrelated donor, 7 haploidentical related donor), 8 did not due to relapse (6) or because too early (2). In the RC, 12 (22%) received ASCT (11 matched family donor, 1 haploidentical related donor) whereas 24 (44%) autologous SCT (AuSCT); 19 were not transplanted at all due to relapse (13) or mobilization failure (6). Therefore, using the risk-adapted approach, 73% of high-risk patients in the PC received ASCT versus 22% of those in the RC (p <0.001). With a median follow-up of 30 and 50 months for the PC and RC, respectively, DFS is 73% vs 15% (p=0.011), OS 69% vs 20% (p=0.020), CIR 21% vs 76% (p<0.001). Based on these results, the GIMEMA Group has activated a clinical trial (AML1310, ClinicalTrials.gov.Identifier NCT01452646) of “risk-adapted, MRD directed therapy for young adult with AML”. The trial relies on a stringent disease characterization at diagnosis in terms of cytogenetic/genetic definition and identification of “leukemia associated immunophenotype” for MRD assessment at the post-consolidation time-point. The 2 parameters are exploited to qualify the category of risk which the patients belong to: low vs intermediate vs high. All patients will receive induction and consolidation according to the previous GIMEMA LAM99P protocol. After the first consolidation, patients belonging to the low-risk category (CBF+ AML without c-Kit mutations, NPM1+FLT3-ITD- AML) will receive AuSCT and those with high-risk features (UK, FLT3-ITD mutations) ASCT. Patients with FLT3-TKD mutations or c-Kit mutated CBF+ AML and those belonging to the IK category will be stratified according to the post-consolidation MRD status and will receive AuSCT or ASCT. All patients who meet the criteria for high-risk definition will be offered ASCT regardless of the availability of a HLA identical sibling, therefore all the other sources of hematopoietic stem cells will be considered. Applying this strategy, we expect a 10% survival advantage at 24 months as compared to the historical control (LAM99P protocol) where OS at 2 years was 50%.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.