The MEK1/2 inhibitor, selumetinib (AZD6244; ARRY-142886), enhances anti-tumour efficacy when combined with conventional chemotherapeutic agents in human tumour xenograft models

Abstract

Background: The Ras/RAF/MEK/ERK pathway is frequently deregulated in cancer and a number of inhibitors that target this pathway are currently in clinical development. It is likely that clinical testing of these agents will be in combination with standard therapies to harness the apoptotic potential of both the agents. To support this strategy, it has been widely observed that a number of chemotherapeutics stimulate the activation of several intracellular signalling cascades including Ras/RAF/MEK/ERK. The MEK1/2 inhibitor selumetinib has been shown to have anti-tumour activity and induce apoptotic cell death as a monotherapy.

Methods: The aim of this study was to identify agents, which would be likely to offer clinical benefit when combined with selumetinib. Here, we used human tumour xenograft models and assessed the effects combining standard chemotherapeutic agents with selumetinib on tumour growth. In addition, we analysed tumour tissue to determine the mechanistic effects of these combinations.

Results: Combining selumetinib with the DNA-alkylating agent, temozolomide (TMZ), resulted in enhanced tumour growth inhibition compared with monotherapies. Biomarker studies highlighted an increase in γH2A.X suggesting that selumetinib is able to enhance the DNA damage induced by TMZ alone. In several models we observed that continuous exposure to selumetinib in combination with docetaxel results in tumour regression. Scheduling of docetaxel before selumetinib was more beneficial than when selumetinib was dosed before docetaxel and demonstrated a pro-apoptotic phenotype. Similar results were seen when selumetinib was combined with the Aurora B inhibitor barasertib.

Conclusion: The data presented suggests that MEK inhibition in combination with several standard chemotherapeutics or an Aurora B kinase inhibitor is a promising clinical strategy.

Figure 1

A chronic dosing schedule of selumetinib increases levels of the pro-apoptotic protein Bim in HCT-116 xenografts. When tumours reached an average volume of 0.2 cm³, animals (n=5 per group) were dosed with either 25 mg kg^–1 per bid of selumetinib or vehicle control. Following treatment, tumours were excised after either 3 (acute) or 14 (chronic) doses and analysed for pharmacodynamic effects. Immunoblotting analysis of lysed tumour tissue was used to detect pERK1/2, total ERK1/2 and Bim-EL. Bim-EL levels were calculated in a ratio to GAPDH. The bar graph shows the average across the group ±s.e.m.

Figure 2

Selumetinib in combination with TMZ enhances DNA damage. When female nude mice bearing SW-620 human tumour xenografts reached an average volume of 0.2 cm³, animals were randomised (n=10 for controls and n=8 per treatment groups) and dosed with either selumetinib (25 mg kg^–1 per qd), TMZ 6.25 mg kg^–1 for 5 consecutive days, selumetinib+TMZ or vehicle controls. A subsequent study was performed in order to generate pharmacodynamic (PD) samples (n=5 per group) following exposure to the dosing regimens at the PD points highlighted by the white arrows on the (A) tumour growth inhibition graph; (B) representative IHC images for γH2A.X, cleaved caspase 3 and pHH3 at PD2. All error bars are ±s.e.m. ^***P<0.0005.

Figure 3

The sequence dependency of selumetinib when combined with docetaxel enhances apoptotic cell death. When female nude mice bearing HCT-116 human tumour xenografts reached an average volume of 0.2 cm³, animals were randomised (n=12 for controls and n=8 per treatment groups) and dosed with either (A) selumetinib (25 mg kg^–1 per qd), docetaxel 15 mg kg^–1 once weekly or both agents in continuous combination for 14 days or (B) following the sequence schedules shown and administered in (C) for one 7-day cycle. All error bars are ±s.e.m. ^**P<0.005, ^***P<0.0005.

Figure 4

The sequence dependency of selumetinib when combined with docetaxel enhances apoptotic cell death. Following the dosing schedules in Figure 3B, pharmacodynamic (PD) samples (n=4 per group) were taken at various points during the dosing cycles (PD1-7) as shown on the schedules in (A and B). Samples from both schedules were analysed by IHC for pHH3 and cleaved caspase 3 (C) representative IHC images for PD2 to illustrate the quantitation shown in (A). Abbreviations: C=control, D=docetaxel; S=selumetinib. All graphs represent the average data for each sample set ±s.e.m. ^***P<0.0005, ^**P<0.005, ^*P<0.05.

Figure 5

Sequence scheduling of selumetinib and the Aurora B kinase inhibitor, barasertib (AZD1152), results in tumour regression and increased cell death. (A) Two dosing schedules investigated in which barasertib was administered at 150 mg kg^–1 by mini-pump in 2 × 24-h infusions followed by a 24-h break and then selumetinib 25 mg kg^–1 per bid for 14 days or the reversal of this schedule, (B) when CaLu-6 human tumour xenografts reached an average of 0.2 cm³ animals were randomised into the relevant vehicle treated, monotherapy or combination groups and exposed to the sequence scheduling. A subsequent study was performed in order to generate pharmacodynamic (PD) samples (n=4 per group) at the timepoints, PD 1-3, highlighted in (A). (C and D) flow cytometric analysis was performed on disaggregated tumours in order to investigate levels of polyploidy and sub G1. Abbreviations: Con=control, Bara=barasertib; Selu=selumetinib. All bar graphs represent the average for each group ±s.e.m. ^*P<0.05, ^**P<0.005, ^***P<0.0005.