Background: With the development of targeted therapeutics aimed at modulating signal transduction networks, there is a need for measuring the pharmacodynamic impact of these agents in relevant complex tissues at the single cell level. The ability to assess target coverage and selectivity is paramount ‐ first, for selection of candidate molecules to progress through the drug development process and second, once a candidate drug moves into clinical testing, for guiding dose and schedule selection.

Objectives: This study tested the ability to use single cell network profiling (SCNP), in which cells are perturbed with extracellular modulators and their response ascertained by multiparametric flow cytometry, to 1) measure the potency and selectivity of a panel of compounds with reported inhibitory effects on JAK/STAT signaling in whole blood versus fractionated peripheral blood mononuclear cells (PBMC) and 2) test the ability of SCNP to capture rare cells amid a heterogeneous population of cells.

Methods: We measured modulation of JAK/STAT, MAPK and NFkB signaling pathways in cell subsets after in vitro exposure of whole blood and PBMC to GM‐CSF (monocytes/neutrophils), IL‐2 (T‐cells), and CD40L (B‐cells) in the presence and absence of known JAK inhibitors. For evaluation of SCNP in rare cells, phosphorylation of STAT proteins in response to IL‐27 was examined on CD34+ CD11b− CD33− progenitors in healthy PBMCs.

Results: The relative selectivity of JAK3 versus JAK2 inhibition was simultaneously measured by plotting p‐Stat‐5 inhibition curves from gated T cells (JAK3 mediated) and monocytes (JAK2 mediated). Notably, CP‐690550, currently in clinical trials for a variety of autoimmune diseases, was 36‐fold more potent at inhibiting JAK3 compared to JAK2 activity, which is consistent with published reports. In addition, off‐target signaling effects outside the JAK/STAT pathway were observed (eg, Cucurbitacin I induced a dose‐dependent increase in p‐ERK in the monocyte subset and “Stat3 Inhibitor VII” blocked CD40L stimulated p‐ERK and pNFkB p65 in B cells). Importantly, a significant decrease in potency in all compounds except CP‐690550 was observed when studies were performed in whole blood, the clinically relevant tissue, versus fractionated PBMC. Notably, SCNP technology captured IL‐27 induced p‐Stat‐1 and p‐Stat‐3 responses and inhibition by Pyridone 6 in CD34+ CD11b− CD33− progenitor cells, a very rare (<40 cells; 0.11% of the total live cells) cell population in PBMC.

Conclusions: SCNP technology allowed for the evaluation of 1) compound activity in primary cells in a physiologically relevant environment (whole blood), 2) signal transduction target and off‐target effects in relevant cell populations, and 3) robust IC50 determination from rare subpopulations (<100 cells). Therefore, this technology is well suited for application in oncology drug development where assessment of candidate drug effects on rare cell populations (eg, cancer stem cells, minimal residual disease, and circulating tumor cells) is crucial and permits the rational design of clinical trials based on biologically active dose rather than the traditional maximum tolerated dose (MTD) design, which is better suited for cytotoxic, non‐targeted drugs.

Citation Information: Mol Cancer Ther 2009;8(12 Suppl):B182.