Abstract 3586

Introduction:

Acute megakaryoblastic leukemia (AMKL) is a biologically heterogeneous variant of acute myeloid leukemia (AML), known as AML-M7 according to the FAB classification. The incidence of AML-M7 among pediatric AML in previously published studies ranged between 7 and 15% of the whole number. Differently from children with Down Syndrome (DS)-AMKL who have an excellent outcome with specifically designed, low-intensity chemotherapy, some studies have reported that the outcome of children with non-DS-AMKL is poor (Ribeiro et al, 1993; Athale et al, 2001). We analyzed the clinical/biological characteristics and outcome of children with non-DS-AMKL treated with the AIEOP AML 2002/01 Protocol.

Patients and Methods:

Between 12/2002 and 06/2011, 44 patients (pts) were included in the analysis. They were assigned to the high-risk (HR) group. Treatment was administered as follows: pts were given 2 courses of 7-day induction therapy (idarubicin, cytarabine and etoposide); if they obtained first complete remission (CR1), they proceeded to receive 2 consolidation courses based on high-dose cytarabine (HD-Ara-c), combined with either etoposide during the first course (AVE 3+4) or mitoxantrone during the second course (HAM 3+2). At the end of consolidation, pts with AMKL in CR1 were eligible to be treated with allogeneic (ALLO) HSCT from a HLA-identical relative, if available, or from alternative donors, namely unrelated donors or HLA-mismatched relatives. We analyzed CR, early death (ED) and induction failure (IF) rate of pts with AMKL; moreover, we calculated the cumulative incidence of relapse and the probability of both overall survival (OS) and event-free survival (EFS).

Results:

AMKL was diagnosed in 44 of 482 (9.1%) children and adolescents below 18 years of age without DS enrolled in the AIEOP AML 2002/01 protocol. Twenty-one pts were males and 23 females. Median age at diagnosis was 1.6 years (range 0.3–15), with 12/44 (27%) pts aged < 1 year. The median count of white blood cells (WBC) at diagnosis was 18,910×103/ml (range 2,490–59,000). Involvement of central nervous system (CNS) at diagnosis was documented in 5/44 (11%), whereas extramedullary disease in sites other than CNS was found in 3/44 pts (7%). Cytogenetic results were available in 34/44 (77%) pts; they showed t(1;22)(p13;q13) in 3/34 (9%) pts, either as sole abnormality or together with other numerical and structural changes and 11q23 rearrangements in 4/34 (12%). Complex karyotype was present in 5/34 (15%) pts. Three out of 44 pts (7%) showed FLT3-ITD mutation. The induction treatment outcome of pts with AMKL was as follows: the CR rate was 86%, while ED and IF rates were 5% and 9%, respectively. The cumulative incidence of leukemia recurrence was 27% [standard error (SE) 8.7]. Thirty out of 44 pts with AMKL (68%) received an ALLO HSCT in first CR. With a median follow up of 57 months (range 3–130), the 8-year OS and EFS of AMKL were 58.4% (SE 7.9) and 55.6% (SE 7.7), respectively, these values being similar to those of children with other variants of AML.

Conclusions:

Although AMKL is generally considered an AML variant characterized by a poor prognosis, our data indicate that an intensive treatment strategy including a wide use of HSCT in first CR results in a favorable outcome, more than 50% of pts being alive and disease-free. The main cause of treatment failure remains leukemia recurrence; once relapsed, children with AMKL have a dismal probability of being rescued by second-line treatment. Overall, our results compare favorably with previously published data on AMKL, including those recently reported by Rubnitz et al (Lancet Oncol 2010;11:543–52).

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution