Fig. 3
From: Defining subpopulations of differential drug response to reveal novel target populations
Divergent response exemplified with PI3K inhibition in comparison to MEK and BRAF inhibitors. a Scatter plot showing subpopulations that exhibit divergent pharmacological response to RDEA119-2 (MEK inhibitor) and GSK2126458 (PI3K inhibitor). Dashed lines indicate 20th percentile of log(IC50) values for each drug. BRAF and KRAS mutations were found enriched in the associated subpopulations. Enriched mutations, cancer tissue types, and/or expression pathway markers are labelled beside each subpopulation. Enriched expression pathway markers represent either the activated or inactive pathways in the subpopulation. Each pathway was named “-up” for activated pathways or “-down” for inactive pathways. b Subpopulations from comparison of PLX4720-2 (BRAF inhibitor) and PI-103 (PI3K inhibitor) responses. c Individual cell line responses from PLX4720-2 and PI-103 coloured by the subpopulations they were grouped in. d Workflow illustrating cell lines being tested with individual inhibitors and their joint pharmacological patterns of response. Drug pairs with divergent response and BRAF mutant subpopulations suggest a target for drug combination therapies, which are validated in cell lines25 and patient-derived tumour xenograft (PDX) models.26 e In vitro synergistic effect of combining MEK inhibitors with PI3K or BRAF inhibitors in all cell lines and just those with BRAF mutations. This was also compared to measured synergy for all combinations tested in cell lines. f In vivo effect of combining BRAF inhibitors with PI3K or MEK inhibitors. Measured response is the change in tumour volume following treatment. BRAF and MEK inhibitor combinations have also been shown to be effective in the patients28
