Abstract 2705: Functional proteomics to reveal distinct signature for acute myeloid leukemia subtypes

Abstract

Abstract Patients with relapsed/ refractory AML have an overall survival rate <30%. Genomic approaches are being used for patients' stratification, but they have not been sufficient at predicting response to therapy. We propose an integrative approach to identify distinctive phosphorylated protein biomarkers in AML, map them to signaling networks, and compare these to pathways identified through analysis of gene expression. This study aims at identifying distinct phosphoproteomic signatures by mass spectrometry (MS) in different AML subtypes. Methods: Four AML cell lines and 40 bone marrow samples from patients with AML representing different AML subtypes are used. Samples were prepared using a MOAC-TiO2 stationary phase for phosphopeptide enrichment, and analysis on a Thermo LTQ Orbitrap Velos MS. Identified phosphorylated STY sites were cross-referenced against iPTMnet. Phosphoproteins were further matched to potential pathways using Ingenuity Pathway Analysis (IPA). Gene-expression profiling data from cell lines, available on HG-U133 Plus 2 arrays, were obtained from GEO and ArrayExpress. ANOVA was performed to identify significantly differentially expressed (fold-change ≥2) probe sets. The probes were mapped to their genes to identify over-represented canonical pathways. Results: 873 different phosphoproteins were identified among the 4 AML cell lines. Nearly 3,000 phosphopeptides were identified (868 in Kasumi-1, 861 in KG-1a, 681 in MV-4-11, and 917 in UT-7epo), of which 2,600 were matched to known sites in iPTMnet and nearly 200 identifications were novel phosphosites. 580 phosphosites were previously reported as known kinase targets (e.g., PRKCB, CDK2 and AKT1). From the top five most significant pathways identified by IPA, Telomerase Signaling was significant in 4 cell lines. DNA Methylation and Transcriptional Repression Signaling was significant in 3 out of 4 cell lines. The numbers of differentially expressed probe sets were: 1,788 mapped to 99 pathways in Kasumi-1; 1,310 mapped to 66 pathways in KG-1a cells; 2,699 mapped to 191 pathways in MV-4-11 cells; and for UT-7epo cells, 2,778 differentially expressed probe sets were mapped to 147 over-represented pathways. Conclusions: We identified phosphoproteomic signatures in AML subtypes. Ongoing studies are directed towards similar profiling of bone marrow of 40 AML patients. Phosphoproteomic profiling will greatly contribute to (1) stratify patients into risk groups, (2) better select targeted therapy based on activated protein pathways, and (3) better predict response to therapy. Citation Format: Patrick Pirrotte, Victoria David, Ritin Sharma, Daniel H. Wai, Cristina Tognon, Brian Druker, Eiman A. Aleem. Functional proteomics to reveal distinct signature for acute myeloid leukemia subtypes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2705.