Abstract A73: Integrating in vitro signaling and drug response data from single cell network profiling (SCNP) to inform on individual's AML biology.

Background: SCNP is a multiparametric flow cytometry-based assay that allows for simultaneous measurement of multiple signaling nodes (stimulus intracellular signaling molecule) in cell subsets from heterogeneous tissues, such as whole blood or PBMC (Kornblau et al. CCR 2010). SCNP has proven useful for functional pathway analysis of AML patient (pt) samples without the need for cell isolation. Objectives/Design: 1) Functionally characterize intrinsic and extrinsic signals implicated in the growth/survival of primary AML cells; 2) Concurrently assess the in vitro effects of a wide variety of kinase inhibitors (KIs) and cytotoxic agents on those signaling networks; and 3) Evaluate any association between signaling profiles and effects upon cell cycle and survival. Methods: Cryopreserved AML samples (N=8; 3 pre-induction, 3 post-induction, 2 relapsed) and healthy bone marrow (BM) samples (N=5) were analyzed via SCNP in 2 experimental arms: Arm 1 assessed basal and growth factor (GF) (SCF, FLT3L, G-CSF, IL-3, TPO) induced signaling in the JAK/STAT, PI3K/mTor, and MEK/ERK pathways ± specific KIs (GDC-0941, BEZ235, AZD6244, AC220, and CP-690550). Signaling readouts were measured simultaneously in multiple myeloid subsets (present in each individual AML sample) identified by expression of CD34, CD117 (cKit), CD45, and light scatter properties. Arm 2 assessed the cytotoxic and cytostatic impact of commercial and investigational drugs (n=23) as single agents and in selected combinations after 48-hr exposure at clinically relevant concentrations. Measures included cell viability (AQUA), apoptosis (cPARP), S/G2 phase (Cyclin B1), M-Phase (p-HistoneH3), and DNA damage (H2AX). Results: In Arm 1, analysis of signaling patterns within the multiple myeloid subsets in individual AML samples demonstrated inter patient morphologic and functional heterogeneity (i.e., GF and/or KIs hypo/hypersensitivity) contrasting with the more homogeneous cellular and signaling patterns observed in the healthy BM. In Arm 2, using unsupervised hierarchical clustering, broad heterogeneity in drug sensitivity (as measured by both cytostasis and apoptosis) among the same AML samples was also observed, again contrasting with the relatively homogeneous responses observed in healthy BM samples. Of note, cluster analysis of apoptosis and cytostasis data grouped together drugs which share mechanisms of action such as JAK inhibitors (CYT387, CP-690550, INCB018424); mTor inhibitors (BEZ235, RAD001); and chemotherapeutics (AraC, clofarabine, etoposide) thus supporting the biologic relevance of the assay. Combined analyzes of Arm 1 and 2 data revealed one AML sample with a distinguishing signaling profile (unresponsive to G-CSF and FLT3L and hypersensitivity to TPO) which was refractory, as measured by induced apoptosis, to most treatments except for bortezomib and an HSP90 inhibitor. Conclusion: SCNP allows for simultaneous assessment of signaling and apoptotic/cytostatic effects of anti-AML drugs. Integrating these data provides insight into the aberrant tumor biology underlying each individual AML pt sample, potentially informing pt-specific strategies for treatment and selection of rational drug combinations. The clinical relevance of the above observations is the focus of ongoing clinical studies. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr A73.