Introduction: Juvenile myelomonocytic leukemia (JMML) is a rare myelodysplastic/myeloproliferative neoplasm that occurs in infancy and early childhood, characterized by excessive myelomonocytic cell proliferation and granulocyte-macrophage colony-stimulating factor hypersensitivity. Over 80% of JMML patients have been reported to harbor mutually exclusive somatic and/or germline mutations in canonical RAS pathway genes (e.g., PTPN11, NF1, NRAS, KRAS, and CBL), and previous studies have identified several poor prognostic factors, including the presence of two or more mutations, acute myeloid leukemia-like (AML-like) gene expression profiles, and selected gene promoter methylation profiles. However, the molecular pathogenesis in 10%-20% of patients and the relationships among biomarkers have not been well defined. This study aimed to assess the molecular pathogenesis in patients and the relationships among several biomarkers and outcomes.Patients and Methods: We performed comprehensive genetic analyses in 150 JMML patients, including 15 with Noonan syndrome-associated myeloproliferative disorder (NS/MPD), using whole-exome sequencing (n = 69, 46%), targeted deep sequencing (n = 92, 61%), RNA sequencing (n = 129, 86%), and genome-wide methylation analysis (n = 106, 71%).Results: We identified ALK/ROS1 tyrosine kinase fusions (DCTN1-ALK, RANBP2-ALK, and TBL1XR1-ROS1) in three of 16 patients (18%) who lacked canonical RAS pathway mutations. Crizotinib, an ALK/ROS1 inhibitor, markedly suppressed ALK/ROS1 fusion-positive JMML cell proliferation in vitro . Therefore, we administered crizotinib to a patient with the RANBP2-ALK fusion gene. This patient achieved complete molecular remission after the administration of crizotinib for 51 days and successfully bridged to allogeneic stem cell transplantation. This patient survived without disease recurrence for 15 months after transplantation.We performed a genome-wide methylation analysis in 106 patients and combined the data with repository data from normal CD34+ samples (n = 5) and AML CD34+CD38− samples (n = 14) for an unsupervised consensus clustering of CpG methylation profiles, which yielded two distinct subgroups, hypermethylation profile (n = 45, 42%) and hypomethylation profile (n = 61, 58%). All AML samples had a hypermethylation profile, whereas all normal CD34+ and NS/MPD samples had a hypomethylation profile. A gene set enrichment analysis revealed marked enrichment of AML-associated genes in the hypermethylation cluster. The 5-year OS rate was 46.0% [95% confidence interval (CI), 31.0%-59.8%] among patients with a hypermethylation profile and 73.4% (95% CI, 57.6%-84.1%) among those with a hypomethylation profile (p = 0.0028). The 5-year TFS was significantly poorer in the hypermethylation group than in the hypomethylation group [2.2% (95% CI, 0.2%-10.1%) vs. 41.2% (95% CI, 27.1%-54.8%), p = 3.1 × 10−10]. Differential gene expression analysis for the hypermethylation and hypomethylation profiles revealed overexpression of LIN28B and its downstream target HMGA2 in the hypermethylation profile.Lastly, we integrated the molecular analyses of JMML, including genome-wide methylation profiling, RNA sequencing, and resequencing data and found that the hypermethylation profile was highly correlated with the most established risk factors for JMML, including older age, higher HbF level, lower platelet count, PTPN11/NF1 mutations, presence of two or more mutations, LIN28B overexpression, and AML-like expression profiles, as well as poor survival.Discussion: We identified recurrent activated ALK/ROS1 fusions in JMML patients without canonical RAS pathway gene mutations. Crizotinib represents a promising candidate drug for treating JMML in patients with ALK/ROS1 fusions. In addition, we established a risk profiling system based on genome-wide methylation analysis and showed that the hypermethylation profile is highly associated with poor survival and most established poor prognostic biomarkers for JMML. These finding will contribute to the development of precision treatments for JMML patients. DisclosuresNo relevant conflicts of interest to declare.