Integrative analysis of drug response and clinical outcome in acute myeloid leukemia

Acute myeloid leukemia (AML) is a cancer of myeloid-lineage cells with limited therapeutic options. We previously combined ex vivo drug sensitivity with genomic, transcriptomic, and clinical annotations for a large cohort of AML patients, which facilitated discovery of functional genomic correlates. Here, we present a dataset that has been harmonized with our initial report to yield a cumulative cohort of 805 patients (942 specimens). We show strong cross-cohort concordance and identify features of drug response. Further, deconvoluting transcriptomic data shows that drug sensitivity is governed broadly by AML cell differentiation state, sometimes conditionally affecting other correlates of response. Finally, modeling of clinical outcome reveals a single gene, PEAR1, to be among the strongest predictors of patient survival, especially for young patients. Collectively, this report expands a large functional genomic resource, offers avenues for mechanistic exploration and drug development, and reveals tools for predicting outcome in AML.

Abstract 4538: Global and phosphoproteomic analysis of AML cell line response to phosphatase inhibitor treatment

In the United States, approximately 21,000 people are diagnosed with AML each year and over 10,000 AML related deaths are reported. Genomic analysis has revealed at least 11 genetic classes of AML, and more than 20 subsets can be assigned when considering the differentiation states of Leukaemic blasts. This variability has generated significant interest in the development of targeted therapies for AML. However, complex mutational patterns and a lack of pharmacological agents for most mutational events have challenged the development of effective treatments. Recent results from the Beat AML clinical trial revealed that response to drugs is associated with mutational status, including instances where drug sensitivity is specific to combinatorial mutational events.

To better understand the molecular mechanisms underpinning drug response, we performed global and phosphoproteomic analyses on 4 AML cell lines that carry mutations associated with AML: K562 with a BCR-ABL fusion, MOLM14 with a FLT3 mutation, HL60 with a RAS mutation, and CMK with a JAK3 mutation. These cell lines were each treated with a different kinase inhibitor drug for which they have a known sensitivity. Cell samples were collected prior to treatment and time points of 30 min, 3h, and 16 hrs. Frozen cell pellets were prepared for analysis using the workflow for multiplexed, deep-scale, proteome and phosphoproteome analysis developed by the NCI CPTAC program.

The goal of our Proteogenomic Translational Research Center is to develop drug response signatures allowing the use of baseline proteome and phosphoproteome measurements to predict patient drug response, providing a foundation for rational selection of combination therapies. Preliminary analysis of the data show a dynamic temporal response to drug treatment in both the global and phosphoproteome, Figure 1. Current efforts are focused on combining knowledge-driven and data-driven analysis approaches to identify the pathways associated with drug response.

Citation Format: Paul D. Piehowski, Jason E. McDermott, Joshua R. Hansen, Samantha L. Savage, Cristina E. Tognon, Anupriya Agarwal, Jeffrey W. Tyner, Brian J. Druker, Karin D. Rodland. Global and phosphoproteomic analysis of AML cell line response to phosphatase inhibitor treatment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4538.

Functional genomic landscape of acute myeloid leukaemia

The implementation of targeted therapies for acute myeloid leukaemia (AML) has been challenging because of the complex mutational patterns within and across patients as well as a dearth of pharmacologic agents for most mutational events. Here we report initial findings from the Beat AML programme on a cohort of 672 tumour specimens collected from 562 patients. We assessed these specimens using whole-exome sequencing, RNA sequencing and analyses of ex vivo drug sensitivity. Our data reveal mutational events that have not previously been detected in AML. We show that the response to drugs is associated with mutational status, including instances of drug sensitivity that are specific to combinatorial mutational events. Integration with RNA sequencing also revealed gene expression signatures, which predict a role for specific gene networks in the drug response. Collectively, we have generated a dataset-accessible through the Beat AML data viewer (Vizome)-that can be leveraged to address clinical, genomic, transcriptomic and functional analyses of the biology of AML.

IRS2 silencing increases apoptosis and potentiates the effects of ruxolitinib in JAK2V617F-positive myeloproliferative neoplasms

The recurrent V617F mutation in JAK2 (JAK2^V617F) has emerged as the primary contributor to the pathogenesis of myeloproliferative neoplasms (MPN). However, the lack of complete response in most patients treated with the JAK1/2 inhibitor, ruxolitinib, indicates the need for identifying pathways that cooperate with JAK2. Activated JAK2 was found to be associated with the insulin receptor substrate 2 (IRS2) in non-hematological cells. We identified JAK2/IRS2 binding in JAK2^V617F HEL cells, but not in the JAK2^WT U937 cell line. In HEL cells, IRS2 silencing decreased STAT5 phosphorylation, reduced cell viability and increased apoptosis; these effects were enhanced when IRS2 silencing was combined with ruxolitinib. In U937 cells, IRS2 silencing neither reduced cell viability nor induced apoptosis. IRS1/2 pharmacological inhibition in primary MPN samples reduced cell viability in JAK2^V617F-positive but not JAK2^WT specimens; combination with ruxolitinib had additive effects. IRS2 expression was significantly higher in CD34⁺ cells from essential thrombocythemia patients compared to healthy donors, and in JAK2^V617F MPN patients when compared to JAK2^WT. Our data indicate that IRS2 is a binding partner of JAK2^V617F in MPN. IRS2 contributes to increased cell viability and reduced apoptosis in JAK2-mutated cells. Combined pharmacological inhibition of IRS2 and JAK2 may have a potential clinical application in MPN.

The Colony-Stimulating Factor 3 Receptor T640N Mutation Is Oncogenic, Sensitive to JAK Inhibition, and Mimics T618I

PURPOSE

Colony-stimulating factor 3 receptor (CSF3R) mutations have been identified in the majority of chronic neutrophilic leukemia (CNL) and a smaller percentage of atypical chronic myeloid leukemia (aCML) cases. Although CSF3R point mutations (e.g., T618I) are emerging as key players in CNL/aCML, the significance of rarer CSF3R mutations is unknown. In this study, we assess the importance of the CSF3R T640N mutation as a marker of CNL/aCML and potential therapeutic target.

EXPERIMENTAL DESIGN

Sanger sequencing of leukemia samples was performed to identify CSF3R mutations in CNL and aCML. The oncogenicity of the CSF3R T640N mutation relative to the T618I mutation was assessed by cytokine independent growth assays and by mouse bone marrow transplant. Receptor dimerization and O-glycosylation of the mutants was assessed by Western blot, and JAK inhibitor sensitivity was assessed by colony assay.

RESULTS

Here, we identify a CSF3R T640N mutation in two patients with CNL/aCML, one of whom was originally diagnosed with MDS and acquired the T640N mutation upon evolution of disease to aCML. The T640N mutation is oncogenic in cellular transformation assays and an in vivo mouse bone marrow transplantation model. It exhibits many similar phenotypic features to T618I, including ligand independence and altered patterns of O-glycosylation--despite the transmembrane location of T640 preventing access by GalNAc transferase enzymes. Cells transformed by the T640N mutation are sensitive to JAK kinase inhibition to a similar degree as cells transformed by CSF3R T618I.

CONCLUSIONS

Because of its similarities to CSF3R T618I, the T640N mutation likely has diagnostic and therapeutic relevance in CNL/aCML.

What word learners do when input contradicts the mutual exclusivity assumption

In 4 studies, 3- to 5-year-olds heard 2 novel English labels each applied to the same novel object by a different adult. In all 4 studies, about half of the children accepted both labels, suggesting that hearing 2 labels applied to an object offers strong enough evidence for some to override mutual exclusivity. Nonetheless, about half of the children honored mutual exclusivity and hence accepted only 1 label; they also seemed to keep both labels in mind for at least a few minutes and settled on whichever label they re-encountered first. This strategy allows children to make an informed choice between 2 apparently equally good labels, without straining their limited memory capacities.