Juvenile Myelomonocytic Leukemia and other myelodysplastic/myeloproliferative neoplasms

Juvenile Myelomonocytic Leukemia (JMML) is a clonal hematopoietic disorder that usually occurs in early childhood, characterized by hyperactivation of the RAS signaling pathway. About 90% of patients have mutations in 1 of 5 genes (PTPN11, NRAS, KRAS, NF1, CBL) that define genetically and clinically distinct subtypes of the disease, with a highly variable clinical course. Allogeneic hematopoietic stem cell transplantation (HSCT) remains the therapy of choice for most patients with JMML, although children with CBL mutations and few of those with N-RAS mutations may have spontaneous resolution of hematologic abnormalities. The results of HSCT in patients with JMML have progressively improved over time, but relapse is still an important cause of treatment failure.

Hybrid or Mixed Myelodysplastic/Myeloproliferative Disorders – Epidemiological Features and Overview

The WHO-category Myelodysplastic/Myeloproliferative neoplasms (MDS/MPNs) recognizes a unique group of clonal myeloid malignancies exhibiting overlapping features of myelodysplastic as well as myeloproliferative neoplasms. The group consists of chronic myelomonocytic leukemia (CMML), atypical chronic myeloid leukemia, BCR-ABL1-negative (aCML), juvenile myelomonocytic leukemia (JMML), myelodysplastic/myeloproliferative neoplasm with ringed sideroblasts and thrombocytosis (MDS/MPN-RS-T), and myelodysplastic/myeloproliferative neoplasms, unclassifiable (MDS/MPN-U). The most frequent entity in this category is CMML, while all other diseases are extremely rare. Thus, only very limited data on the epidemiology of these subgroups exists. An appropriate diagnosis and classification can be challenging since the diagnosis is still largely based on morphologic criteria and myelodysplastic as well as myeloproliferative features can be found in various occurrences. The diseases in this category share several features that are common in this specific WHO-category, but also exhibit specific traits for each disease. This review summarizes published data on epidemiological features and offers a brief overview of the main diagnostic criteria and clinical characteristics of the five MDS/MPN subgroups.

Establishment of Stable Cell Lines Representing Juvenile Myelomonocytic Leukemia with SHP2 Mutations

Protein tyrosine phosphatase SHP2 encoded by PTPN11 is a key regulator in growth factor and cytokine signaling. Overwhelming evidence suggests its vital role in hematopoietic stem cell function and hematopoiesis. As a bona fide proto-oncogene product, gain-of-function mutations of SHP2 cause hematological malignancies, most notably juvenile myelomonocytic leukemia (JMML) which bear somatic SHP2 mutations in 35% of cases. Numerous studies have utilized murine models to investigate the role of mutant SHP2 in hematopoiesis and leukemogenesis and successfully produced resembling myeloproliferative neoplasm (MPN) and even full-blown leukemia in recipient animals. However, mutant SHP2-transformed cell lines have not been generated. In the present study, we established oncogenic mutant SHP2-transformed cell lines from erythropoietin (EPO)-dependent HCD-57 erythroid leukemia cells. First, we generated recombinant retroviruses expressing SHP2-D61Y and SHP2-E76K, the two most common SHP2 mutants found in individuals with JMML, by using the pMSCV-IRES-GFP vector. We then infected HCD-57 cells with the recombinant retroviruses. Unlike the parent HCD-57 cells, the infected cells were able to grow in the absence of EPO as demonstrated by viable GFP-positive cells. We further performed semi-solid methylcellulose colony cultures and isolated single clones of EPO-independent HCD57 cells. The isolated clonal cells overexpressed mutant SHP2 and proliferate rapidly in the absence of EPO. In contrast, HCD57 cells infected with retroviruses expressing wild type SHP2 failed to survive in the absence of EPO, indicating only gain-of-function mutant forms of SHP2 have the cell-transformation capability. We also carried out parallel experiments with the pro-B Ba/F3 cell line that require interleukin 3 (IL3) for survival. Interestingly, over-expression of SHP2-D61Y and SHP2-E76K was not sufficient to give rise to IL3-indepdent Ba/F3 cells, suggesting that HCD57 cells have some unique properties making them susceptible for transformation by oncogenic SHP2 mutants. We further performed in vitro and in vivo characterization of transformed HCD57 cells. Cell signaling analyses revealed that both HCD57-SHP2-D61Y and HCD57-SHP2-E76Kcells exhibited aberrantly elevated levels of pERK and pAKT in the absence of cytokine stimulation, which was consistent with the notion that gain-of-function SHP2 mutants perturb growth control through deregulation of the Ras signaling pathway. Upon intravenous injection into immunodeficient mice, the SHP2 mutant-transformed HCD57 cells caused acute leukemia with markedly increased spleen. Finally, we screened a small molecule inhibitor library to identify compounds that may specifically target the SHP2 mutants. We found several tyrosine kinase inhibitors including dasatinib and trametinib potently inhibited HCD57-SHP2-D61Y and HCD57-SHP2-E76Kcells but not the parent HCD57 cells. At sub-micromolar concentrations, dasatinib and trametinib abolished elevated ERK and Akt activation caused by the SHP2 mutants. This study not only proves that gain-of function mutations of SHP2 are capable of fully transforming cells but also provides a unique cell system to study pathogenesis of SHP2 mutants and to identify specific inhibitors for drug development. Disclosures No relevant conflicts of interest to declare.

Correlation Analysis of DNA Methylation Level and Clinical Characteristics in JMML Patients

Objective: As a rare, aggressive pediatric myeloproliferative disease, juvenile myelomonocytic leukemia (JMML) encompassed both biological features of myelodysplastic syndrome and myeloproliferative neoplasm. Studies have shown that the methylation level in JMML patients is closely related to prognosis, and patients with high methylation level have poor prognosis. This study aimed to find clinical indicators that were associated with different methylation levels and prognosis. Methods: The clinical information of 24 JMML patients with DNA samples admitted to our center from December 2013 to May 2020 was retrospectively analyzed, and the DNA methylation level of their whole genome was detected. Results: The median age of onset was 14.5 months (0.1-153 months) among the 24 cases, including 17 males and 7 females. At diagnosis, the median WBC count was 27.1×10 9/L (6.2-98.1×10 9/L), and the median platelet count was 38×10 9/L (10-277×10 9/L). Chromosome karyotype abnormalities were found in 12.5% (3/24) of patients. Next-generation sequencing results showed that 79.2% (19/24) patients had at least one Ras pathway-related classical gene mutation, and 41.7% (10/24) patients had two or more somatic mutations. Genomic DNA methylation levels were divided into three groups: 10 cases in the hypomethylation group, 4 cases in the moderate methylation group, and 10 cases in the hypermethylation group. There were significant differences in age, platelets, PTPN11 gene mutation and the number of somatic mutations ≥2 in different methylation groups (P<0.05). The age of hypomethylated group was significantly lower than that of hypermethylated group (P<0.05), and the platelets of hypomethylated group was significantly higher than that of hypermethylated group (P<0.05). Patients ≤12m and platelets＞32×10 9/L had lower DNA methylation level (P<0.0001). The number of patients with PTPN11 gene mutation in the hypomethylated group was significantly lower than that in the hypermethylated group (P<0.05), and the number of patients with ≥2 mutations in the low and medium methylated groups was significantly lower than that in the hypermethylated group (P<0.05). Correlation analysis showed that hypermethylation level was significantly correlated with PTPN11 gene mutation and ≥2 somatic mutations (P<0.001). Conclusions: JMML patients with high methylation level in the DNA genome at diagnosis were older and with lower platelet levels, and hypermethylation were significantly correlated with high-risk prognostic factors such as PTPN11 gene mutation and ≥2 somatic mutations. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare. OffLabel Disclosure: Decitabine for treatment of children with JMML

Aberrant Histone Landscape in Juvenile Myelomonocytic Leukemia

Juvenile myelomonocytic leukemia (JMML) is a deadly pediatric cancer characterized by excessive accumulation of monocytes/macrophages with myelodysplastic features and splenomegaly. Hematopoietic stem cell transplantation (HSCT) is the only curative option, but relapse occurs in ~50% of the patients. Mutations in the RAS pathway are the major drivers and DNA hypermethylation is associated with severe clinical phenotype. DNA methylation is linked to histone modifications in normal cell development and cancer pathogenesis. Gene expression levels are programmed by histone modifications. However, a comprehensive characterization of histone modifications in JMML is yet to be elucidated. It was reported that the CD34+ hematopoietic stem and progenitors cells (HSPCs) possess leukemia initiation potential (Caye et al., 2015; Krombholz et al., 2016; Louka et al., 2021; Yoshimi et al., 2017). We hypothesized that dysregulated histone modifications contribute to leukemogenesis in JMML. To this goal, we investigated epigenetic landscape at the single-cell level in JMML HSPCs using cytometry by time-of-flight (EpiTOF). We isolated HSPCs from the spleens of JMML patients with PTPN11 mutations which were collected at diagnosis, before treatment (provided by the European Working Groups of Myelodysplastic Syndromes). The CD34+ cells from JMML Spleens (n=5) and Bone Marrow (BM n=1) alongside healthy controls (cord blood CB n=5; BM n=2) were stained with antibodies against 15 immunophenotypic and 46 key histone markers and subsequently analyzed on a mass cytometer (Fluidigm, Helios). We found that a group of key histone markers were significantly downregulated in JMML in comparison to healthy CBs, including H4K16ac, H3K23ac, H3K56ac, H3K36me1, H3K9me1 and H3K27me2 (Fig.1). Dimensionality reduction analyses showed a global reduction in histone acetylation in all JMML spleen samples (~70%) when compared to CBs (average >95%). Our study revealed significant dysregulation of key histone markers and their respective modifiers (histone modifying enzymes) within the leukemogenic CD34+ compartment (Fig.2). Our findings demonstrated an aberrant epigenetic regulation in JMML patient samples with concurrent loss of histone acetylation and some histone methylation markers, which have been reported as hallmarks of other cancers. Further characterization of the epigenetic landscape in JMML was performed using transposase-accessible chromatin with high-throughput sequencing (bulk ATAC-seq). Notably, ATAC-seq data revealed significantly lower expression of key histone acetyltranferases (HAT), such as Kat6b and Kat8 (Fig.2) , which specifically regulate acetylation of H3K23 and H4K16 respectively. Additionally, an increased activity of histone deacetylase, HDAC9, was also observed in these samples (Fig.2). Collectively, our data suggests aberrant activities of histone modifying enzymes (HMEs) in JMML HSPCs (Fig.2), with both increased HDAC and decreased HAT activities observed in our ATAC-data. This finding is concurrent with our EpiTOF findings of global loss of histone acetylation in JMML (Fig.1A). It is known that histone modifications are easier to reverse when compared to DNA methylation, and we believe targeted inhibition of specific HDACs that we have found to be upregulated in JMML (such as HDAC9) might hold great therapeutic potential. Additionally, investigation into the impact of altered chromatin structure of JMML HSPCs on gene expression is currently ongoing using single-cell RNA-seq. In conclusion, our study identified a novel mechanism of epigenetic dysregulation in putative JMML leukemia initiating cells related to imbalanced activities of histone modifying enzymes which could potentially represent future therapeutic targets. Figure 1 Figure 1. Disclosures Bertaina: AdicetBio: Membership on an entity's Board of Directors or advisory committees; Neovii: Membership on an entity's Board of Directors or advisory committees; Cellevolve Bio: Membership on an entity's Board of Directors or advisory committees.

Small Molecule Kinase Inhibitor Sunitinib Specifically Targets Juvenile Myelomonocytic Leukemia Cells with SHP2(PTPN11) Mutation

Juvenile myelomonocytic leukemia (JMML) is a highly fatal malignant disease in early childhood. It is still unknown of the specific pathogenesis, and there is shortage of effective targeted therapeutic approaches. Gain of function SHP2 mutation encoded by PTPN11 gene is found in approximately 35% of JMML patients, which maybe contributed to its pathogenesis. JMML patients with SHP2 mutation have lower survival rate and higher recurrence rate. All of the above make development of new therapies imperative. Currently, there is no stable cell line that can accurately reflect the characteristics of JMML abnormal cells for research on JMML. In this study, we established two leukemia cell lines that depend on mutated SHP2 for survival, and discovered promising drugs that targeted mutated-SHP2-dependent oncogenic signaling pathway through drug screening method. HCD-57 cells are murine erythroleukemia cells that solely depend on exogenic erythropoietin (EPO) for survival. We constructed SHP2-D61Y and SHP2-E76K transformed HCD-57 cell lines through retroviral vectors, the survival of which dependent on mutated SHP2 mediated signaling pathway. Based on these cells, we established a drug screening platform and screened small molecule compound library containing 2862 FDA-approved drugs and 1707 kinase inhibitors. We performed cell viability, flow cytometry, Wright-Giemsa staining, and western blot to evaluate cells after drug treatment. To further assess therapeutic potential, we established in-vivo transplantation model that SHP2-D61Y transformed HCD-57 cells were implanted into immunodeficient NCG mice, and verified the effectiveness of the in-vitro screened drugs. We found that the survival and proliferation of HCD-57 cells transduced by SHP2-D61Y and SHP2-E76K no longer required EPO, but completely relied on the abnormal activation of signaling pathway mediated by mutated SHP2. Western blot results showed that the phosphorylation status of ERK1/2 and AKT of HCD-57 cells expressing SHP2 mutation were abnormally increased, consistent with SHP2-mutated JMML. Thus, we have obtained the leukemia cell lines that can represent the characteristics of activated signaling pathway in JMML with SHP2 mutation. Through drug screening, we observed that drug sunitinib (Sutent ®) selectively inhibits SHP2-mutated HCD-57 cell lines. CCK-8-based cell viability assay demonstrated a dose-dependent inhibition of SHP2-D61Y and SHP2-E76K transformed HCD-57 cell and no effects on the parental HCD-57 cells. Live cell counting with trypan blue revealed that the proliferation of SHP2-mutated HCD-57 cells was totally halted after one day upon treatment with 250 nM sunitinib, whereas the HCD-57 cells were unaffected. Wright-Giemsa staining demonstrated that SHP2-mutated HCD-57 cells showed no normal morphology change and no mitotic activity under sunitinib treatment, otherwise parental HCD-57 cells showed normal mitotic activity. Sunitinib induced apoptosis and cell cycle arrest at G1 phase in SHP2-mutated HCD-57 cells by flow cytometry, but had little effect on the parental HCD-57 cells. Sunitinib effectively downregulates the phosphorylation of ERK and AKT in SHP2-mutated cells, revealing the mechanism of sunitinib targeting SHP2-mutated cells. In addition, after transplantation of SHP2-D61Y transformed HCD-57 cells for 3 weeks, the spleen of NCG mice increased from an average of 45 mg to more than 300 mg; flow cytometry analysis showed that the implanted cells accounted for over 75% of the total nucleated cells in the bone marrow and spleen. Compared with the vehicle control, the number of monocytes in these mice was reduced to the normal range by treatment with sunitinib, and the spleen weights were reduced by about 50%. Histochemical staining showed disappearance of the myeloid infiltration in the spleen, liver and bone marrow. The above results all indicate that sunitinib has strong in-vivo anti-leukemia activity. Furthermore, western blot analysis showed that the administration of sunitinib significantly inhibited the phosphorylation expression level of AKT and ERK, indicating the effectivity of sunitinib in vivo. In conclusion, our data demonstrated that HCD-57 cell line is an effective tool for studying oncogenic signaling pathway and screening drugs that targeted JMML with SHP2 mutation. Sunitinib can be an effective drug for the targeted treatment of JMML in the future. Disclosures No relevant conflicts of interest to declare.

MEK Inhibition Demonstrates Activity in Relapsed, Refractory Patients with Juvenile Myelomonocytic Leukemia: Results from COG Study ADVL1521

Background: Juvenile myelomonocytic leukemia (JMML) is a hematologic malignancy of infants and toddlers with both myelodysplastic and myeloproliferative features. The prognosis for patients (pts) with relapsed or refractory (r/r) JMML is poor and hematopoietic stem cell transplant (HCT) is the only curative therapy. The molecular hallmark of JMML is activation of the Ras/MAPK pathway. In preclinical studies, MEK inhibition was shown to be effective at reducing spleen sizes, restoring normal hematopoiesis, and extending survival compared to placebo in several genetically engineered mouse models of JMML. Trametinib is an orally bioavailable MEK1/2 inhibitor and is approved for treatment of several malignancies in adults. This Children's Oncology Group study (ADVL1521, NCT03190915) is the first clinical trial for pts with r/r JMML conducted in the United States. Pts are eligible if they have persistent clinical or molecular evidence of JMML after 1 cycle of high dose cytarabine, 2 cycles of a hypomethylating agent or HCT. Pts receive daily trametinib for up to 12 cycles (28 days) in the absence of disease progression or dose-limiting toxicity (DLT). Dosing is age-based with pts less than 6 years of age receiving 0.032mg/kg/day and those 6 years or older receiving 0.025mg/kg/day. An oral suspension is available for pts unable to swallow tablets. Using a Simon 2-stage design (10 pts in each stage), trametinib would be deemed effective if 3 or more pts achieved an objective response. Results: From 2018-2021, 9 pts were enrolled; all 9 were eligible and evaluable for toxicity and response. Each pt had a detectable Ras mutation at the time of enrollment and was monitored for response using clinical and molecular criteria developed by an international consensus panel (Niemeyer et al, 2015). Five pts were less than 2 years of age. Three patients had relapsed post-HCT prior to enrolling and 6 patients were refractory to a median of 1.5 prior therapies (range 1-3). Four pts had an objective response (1 clinical complete response (cCR), 3 clinical partial responses, (cPR); 2 pts had stable disease and 3 had progressive disease (Table 1). Both pts with stable disease completed the maximum 12 cycles permitted on study. Two pts who achieved a cPR proceeded to HCT. One patient who achieved a cCR remains on study. No molecular responses were achieved. There were no dose-limiting toxicities; one pt had grade 4 thrombocytopenia probably related to trametinib. Of the 8 patients who consented to correlative studies, 7 had DNA methylation testing, 6 had kinome profiling, and 5 had RNASeq testing performed on both pre- and post-trametinib paired samples. DNA methylation testing revealed stable intrapatient methylation signatures across diagnostic, relapse and post-trametinib timepoints using the international consensus criteria (Schönung et al, 2020). Integrated kinome and RNASeq analysis revealed downregulation of proteins and genes involved in Ras/MAPK signaling. Conclusions: In the first clinical trial for r/r JMML patients in the United States, 4 objective responses were observed among the initial 9 patients meeting the pre-defined criteria to deem trametinib effective. While clinical responses including resolution of splenomegaly, resolution of monocytosis and increased platelets counts were observed, no molecular responses were noted. The treatment of r/r JMML has historically depended on HCT. Recently, azacitidine has shown promise in treating r/r JMML. This trial demonstrates that trametinib is active in r/r JMML and has a favorable side effect profile. Ongoing analysis of extensive correlative testing results have revealed potential mechanisms of response and resistance to MEK inhibition. Future studies will focus on children with newly diagnosed JMML and combine trametinib with azacitidine with or without HCT. Figure 1 Figure 1. Disclosures Loh: MediSix therapeutics: Membership on an entity's Board of Directors or advisory committees. Barkauskas: Genentech: Current Employment.