The principle of AML treatment is based on “total cell kill” theory, to eradicate all the leukemic blasts that have expanded inside the patient’s body. This is attempted by multi-agent chemotherapy with optional use of hematopoietic stem cell transplantation (SCT) for high risk of relapse. In addition, risk stratification, means of optimizing therapy by evaluating relapse risk with known prognostic factors, and supportive care have contributed to improve survival rate in children with AML.

Treatment algorithm of pediatric AML proposed in the guideline of the Japanese Society of Pediatric Hematology and Oncology is shown in Fig. 3.1. ML-DS and APL are treated separately with each disease-specific protocol. The rest of the children with AML (de novo AML) are treated with appropriate risk-stratified therapies.

Down syndrome (DS) results from trisomy 21 and occurs in 1 in 700–1000 births [90, 91]. Although the incidence of neoplasms in DS does not differ significantly from that in the general population, the distribution of malignancies is different [92]. Patients with DS show a unique spectrum of malignancies, with a 10- to 20-fold higher risk of acute leukemia and a lower incidence of solid tumors [92, 93]. The most frequent form of leukemia during childhood, both with and without DS, is ALL, and the incidence of ALL in children with DS is approximately 20-fold higher than that in non-DS children [92]. However, the most marked increase in incidence in DS infants is AMKL, known as “myeloid leukemia associated with DS (ML-DS).” The development of ML-DS is closely linked to a preceding temporary form of neonatal leukemia called transient abnormal myelopoiesis (TAM). TAM and ML-DS are classified as an independent category as myeloid proliferations of Down syndrome in recent WHO classification [2, 94, 95]. Both generally manifest as megakaryoblastic phenotype, with TAM occurring at birth or within days of birth and being resolved in 1–2 months. About 1–2% of children with DS develop AML during the first 4 years of life. MDS often precedes AMKL and can last for several months. The clinical course of patients of MDS in DS appears to be relatively indolent, presents initially with a period of thrombocytopenia, and lacks significant increase of blasts. Findings from recent clinical and laboratory studies revealed that quantitative alterations of RUNX1, ETS2, and ERG genes (all located at 4-Mb region of chromosome 21) that resulted from constitutional trisomy 21 up-regulate GATA1 mutations and cause TAM [96] and that additional genetic alterations including those in epigenetic regulators and signaling molecules are involved in the progression from TAM to ML-DS [97]. Children with DS above the age of 5 with AML or MDS have different biology to ML-DS, such as non-megakaryoblastic phenotype and no GATA1 mutation of blasts, and should be considered not as “typical ML-DS,” but as “conventional” AML or MDS.

ML-DS has unique characteristics of higher sensitivity to chemotherapeutic agents. In vitro studies showed that ML-DS blasts were significantly more sensitive to chemotherapeutic drugs than non-DS AML cells [98]. ML-DS blasts are especially sensitive to cytarabine, possibly because of the effect of GATA1 mutations and trisomy 21 on the levels of cytarabine-metabolizing enzymes [99]. Before the 1990s, most patients with ML-DS received suboptimal therapy, resulting in poor outcomes. In 1992, high rates of EFS with intensive AML treatment were reported from the Pediatric Oncology Group (POG) [100]. After recognition of the favorable outcome of ML-DS patients treated with the AML protocol, recruitment to collaborative studies for ML-DS patients increased, but it became apparent that treatment-related toxicity was high in most series [101–103]. Since then, several collaborative groups have adapted their AML protocols for ML-DS by reducing the dosage of chemotherapeutic agents [103]. As a result, in recent clinical studies in several countries, ML-DS children were treated separately and less intensively than non-DS AML children. Recent clinical trials for ML-DS are summarized in Table 3.3 [101–109]. In Europe, the ML-DS 2006 study by the I-BFM-SG, which was based on the reduced intensity arm of the BFM 98 protocol with further elimination of etoposide in consolidation courses, was conducted. This protocol consists of four courses including high-dose cytarabine. The US AAML0431 study, conducted from 2007 to 2011 by the COG, which also included high-dose cytarabine, showed excellent outcomes [110]. The Japanese trials for ML-DS had investigated less intensive chemotherapy compared with those conducted in Western countries, and treatment outcomes were comparable [108, 109, 111]. Toronto group reported long-term results of an ultralow dose cytarabine-based regimens suggested that at least some population of ML-DS could be cured by minimum intensive therapy [106]. Predicting prognostic factors had been examined by several studies; however, no universal factors had been found to date [107, 109, 112]. On the other hand, it is reported that relapsed/refractory cases are rarely salvageable, with very limited role of allo-SCT [113]. In terms of treatment outcome, ML-DS is a heterogeneous disease and risk-oriented therapy is a reasonable strategy, so that finding an accurate method, perhaps MRD, for identifying a subgroup with poor prognosis is urgently needed. Moreover, new therapeutic approaches for relapsed/refractory cases using new drugs such as Wee 1 inhibitor [114], Aurora-kinase inhibitor [115], and histone deacetylase inhibitors [116] will be also needed.