Modern targeted cancer therapies rely on the overexpression of tumor associated antigens with very little to no expression in normal cell types. Mesothelin is a glycosylphosphatidylinositol-anchored cell surface protein that has been identified in many different tumor types, including lung adenocarcinomas, ovarian carcinomas, and most recently in hematological malignancies, including acute myeloid leukemia (AML). Although the function of mesothelin is widely unknown, interactions with MUC16/CA125 indicate that mesothelin plays a role in the regulation of proliferation, growth, and adhesion signaling. Most research on mesothelin currently focuses on utilizing mesothelin to design targeted cancer therapies such as monoclonal antibodies, antibody–drug conjugates, chimeric antigen receptor T and NK cells, bispecific T cell engaging molecules, and targeted alpha therapies, amongst others. Both in vitro and in vivo studies using different immunotherapeutic modalities in mesothelin-positive AML models highlight the potential impact of this approach as a unique opportunity to treat hard-to-cure AML.

Relapse remains a major obstacle to achieving 100% overall survival rate in pediatric hematologic malignancies like acute lymphoblastic leukemia (ALL). Relapse often results from the development of chemoresistance. One of the mechanisms of chemoresistance involves ALL cell interactions with the bone marrow (BM) microenvironment, providing a sanctuary. This phenomenon is known as BM microenvironment-induced chemoprotection. Members of the transmembrane 4 superfamily (tetraspanins; TSPANs) are known to mediate microenvironmental interactions and have been extensively studied in solid tumors. Although the TSPAN family member CD81 is a minimal residual disease marker, its biological role in ALL is not well characterized. We show for the first time that CD81 knockout induces chemosensitivity, reduces cellular adhesion, and disrupts in vivo BM homing and engraftment in B-ALL. This chemosensitization is mediated through control of Bruton tyrosine kinase signaling and induction of p53-mediated cell death. We then show how CD81-related signaling can be disrupted by treatment with the epigenetic drug combination of DNA hypomethylating agent azacitidine (aza) and histone deacetylase inhibitor panobinostat (pano), which we previously used to sensitize ALL cells to chemotherapy under conditions that promote BM microenvironment-induced chemoprotection. Aza/pano-mediated modulation of CD81 surface expression is involved in decreasing BM load by promoting ALL cell mobilization from BM to peripheral blood and increasing response to chemotherapy in disseminated patient-derived xenograft models. This study identifies the novel role of CD81 in BM microenvironment-induced chemoprotection and delineates the mechanism by which aza/pano successfully sensitizes ALL cells via modulation of CD81.

Approximately 1%–2% of children with Down syndrome (DS) develop acute myeloid leukemia (AML) prior to age 5 years. AML in DS children (ML-DS) is characterized by the pathognomonic mutation in the gene encoding the essential hematopoietic transcription factor GATA1, resulting in N-terminally truncated short form of GATA1 (GATA1s). Trisomy 21 and GATA1s together are sufficient to induce transient abnormal myelopoiesis (TAM) exhibiting pre-leukemic characteristics. Approximately 30% of these cases progress into ML-DS by acquisition of additional somatic mutations. We employed disease modeling in vitro by the use of customizable induced pluripotent stem cells (iPSCs) to generate a TAM model. Isogenic iPSC lines derived from the fibroblasts of DS individuals with trisomy 21 and with disomy 21 were used. The CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats)/Cas9 system was used to introduce GATA1 mutation in disomic and trisomic iPSC lines. The hematopoietic stem and progenitor cells (HSPCs) derived from GATA1 mutant iPSC lines expressed GATA1s. The expression of GATA1s concomitant with loss of full-length GATA1 reduced the erythroid population, whereas it augmented megakaryoid and myeloid populations, characteristic of TAM. In conclusion, we have developed a model system representing TAM, which can be used for modeling ML-DS by stepwise introduction of additional mutations.

Acute myeloid leukemia (AML) in children remains deadly, despite the use of maximally intensive therapy. Because leukemia stem cells (LSCs) significantly contribute to chemoresistance and relapse, therapies that specifically target the LSCs are likely to be more beneficial in improving outcome. LSCs are known to have high telomerase activity and telomerase activity is negatively correlated with survival in pediatric AML. We evaluated the preclinical efficacy of imetelstat, an oligonucleotide inhibitor of telomerase activity in patient-derived xenograft (PDX) lines of pediatric AML. Imetelstat treatment significantly increased apoptosis/death of the LSC population in a dose-dependent manner in six pediatric AML PDX lines ex vivo, while it had limited activity on the stem cell population in normal bone marrow specimens. These results were validated in vivo in two distinct PDX models wherein imetelstat as single agent or in combination with chemotherapy greatly reduced the LSC percentage and prolonged median survival. Imetelstat combination with DNA hypomethylating agent azacitidine was also beneficial in extending survival. Secondary transplantation experiments showed delayed engraftment and improved survival of mice receiving imetelstat-treated cells, confirming the diminished LSC population. Thus, our data suggest that imetelstat represents an effective therapeutic strategy for pediatric AML.

Children with Down syndrome (DS) have a high risk for acute myeloid leukemia (DS-ML). Genomic characterization of DS-ML blasts showed the presence of unique mutations in GATA1, an essential hematopoietic transcription factor, leading to the production of a truncated from of GATA1 (GATA1s). GATA1s, together with trisomy 21, is sufficient to develop a pre-leukemic condition called transient abnormal myelopoiesis (TAM). Approximately 30% of these cases progress into DS-ML by acquisition of additional somatic mutations in a stepwise manner. We previously developed a model for TAM by introducing disease-specific GATA1 mutation in trisomy 21-induced pluripotent stem cells (iPSCs), leading to the production of N-terminally truncated short form of GATA1 (GATA1s). In this model, we used CRISPR/Cas9 to introduce a co-operating mutation in STAG2, a member of the cohesin complex recurrently mutated in DS-ML but not in TAM. Hematopoietic differentiation of GATA1 STAG2 double-mutant iPSC lines confirmed GATA1s expression and the loss of functional STAG2 protein, leading to enhanced production of immature megakaryocytic population compared to GATA1 mutant alone. Megakaryocyte-specific lineage expansion of the double-mutant HSPCs exhibited close resemblance to the DS-ML immunophenotype. Transcriptome analysis showed that GATA1 mutation resulted in downregulation of megakaryocytic and erythrocytic differentiation pathways and interferon α/β signaling, along with an upregulation of pathways promoting myeloid differentiation such as toll-like receptor cascade. The co-occurrence of STAG2 knockout partially reverted the expression of genes involved in myeloid differentiation, likely leading to enhanced self-renewal and promoting leukemogenesis. In conclusion, we developed a DS-ML model via hematopoietic differentiation of gene-targeted iPSCs bearing trisomy 21.