Combined AKT and MEK Pathway Blockade in Pre-Clinical Models of Enzalutamide-Resistant Prostate Cancer

Abstract

Despite recent improvements in patient outcomes using newer androgen receptor (AR) pathway inhibitors, treatment resistance in castrate resistant prostate cancer (CRPC) continues to remain a clinical problem. Co-targeting alternate resistance pathways are of significant interest to treat CRPC and delay the onset of resistance. Both the AKT and MEK signaling pathways become activated as prostate cancer develops resistance to AR-targeted therapies. This pre-clinical study explores co-targeting these pathways in AR-positive prostate cancer models. Using various in vitro models of prostate cancer disease states including androgen dependent (LNCaP), CRPC (V16D and 22RV1) and ENZ-resistant prostate cancer (MR49C and MR49F), we evaluate the relevance of targeting both AKT and MEK pathways. Our data reveal that AKT inhibition induces apoptosis and inhibits cell growth in PTEN null cell lines independently of their sensitivity to hormone therapy; however, AKT inhibition had no effect on the PTEN positive 22RV1 cell line. Interestingly, we found that MEK inhibition had greater effect on 22RV1 cells compared to LNCaP, V16D or ENZ-resistant cells MR49C and MR49F cells. In vitro, combination AKT and MEK blockade had evidence of synergy observed in some cell lines and assays, but this was not consistent across all results. In vivo, the combination of AKT and MEK inhibition resulted in more consistent tumor growth inhibition of MR49F xenografts and longer disease specific survival compared to AKT inhibitor monotherapy. As in our in vitro study, 22RV1 xenografts were more resistant to AKT inhibition while they were more sensitive to MEK inhibition. Our results suggest that targeting AKT and MEK in combination may be a valuable strategy in prostate cancer when both pathways are activated and further support the importance of characterizing the dominant oncogenic pathway in each patient's tumor in order to select optimal therapy.

Fig 1. Effect of AKT and MEK inhibition on downstream signaling pathways an AR signaling pathway.

(A) Effect of AKT and MEK inhibition on downstream signaling pathways. Androgen dependent PCa cell line LNCaP, CRPC (V16D and 22RV) and ENZ resistant cell lines MR49C, MR49F cell lines were treated with AZD5363 1μM, PD0325901 20 μM alone or in combination for 48 hours. Total proteins were extracted and western blots were performed using AR, PSA and PI3K/AKT pathway signalling proteins as indicated. Representative blots of duplicate experiments are shown. (B-C) Effect of AKT and MEK inhibition on AR pathway. MR49C, MR49F and 22RV cell lines were treated with AZD5363 1μM, PD0325901 20 μM alone or in combination for 48 hours. RNA was extracted from different cell lines and quantitative real time were performed using Taqman probes for AR (B) and AR target gene PSA (C). Representative results of biologic duplicates with technical triplicates are shown.

Fig 2. Combination AZD5363 + PD325901 increases apoptosis.

(A-E) Propidium iodide flow cytometry cell cycle analysis of indicated cell lines shows increased apoptotic cell cycle fraction (SubG1/G0) (left panels). Means of triplicate experiments are plotted +/- SEM. Representative results of all cell cycle populations are shown in right panels. (F) Indicated cells were treated with AZD5363 1μM, PD0325901 20μM, or the combination for 48 hours. Proteins were extracted and western blot was performed using PARP antibody, vinculin was used as a loading control. Representative blots of two or more experiments are shown. (G) Caspase-3 activity in MR49C, MR49F, 22RV1, LNCaP and V16D cell lines treated with AZD5363 and/or PD0325901 for 24 hours. Mean fold change +/-SEM of pooled values from at least two biologic duplicates are shown.

Fig 3. Effect of AZD5363 and PD0325901 on cell viability.

(A-E) MR49C, MR49F, 22RV1, LNCaP and V16D cell lines were treated with AZD5363 and/or PD0325901 as indicated for 48 hours and cell viability was assessed using WST-1 assay. Pooled results of biologic triplicates with technical triplicates are shown. Combination indices are shown inset, with values <1 indicating synergy.

Fig 4. Effect of targeting MEK and AKT using PD0325901 and AZD5363 in vivo.

After establishment of tumors (200mm3) from subcutaneous injection of ENZ resistant MR49F cells in castrated mice under the pressure of 10mg/kg daily of ENZ, mice were treated with 100mg/kg, AZD5363 100mg/kg BID, PD0325901 5mg/kg QD or AZD5363 100mg/kg BID + PD0325901 5mg/kg QD. (A) Representative data of MR49F mean tumor volume over 4 weeks is shown. (B) Mean MR49F tumor growth velocity for each treatment group +/-SEM calculated using linear regression estimation of tumor growth velocity for each mouse. (C) Mean MR49F weekly PSA plotted as fold change from baseline for each treatment group +/-SEM. (D) Waterfall plot of individual PSA measurements after 3 weeks of treatment for all MR49F xenografts in the study. (E-F) Kaplan-Meier cancer specific survival and overall survival curves for treatment arms of MR49F xenografts. (G) Mean 22RV1 xenograft volume following treatment with vehicle, AZD5363 100mg/kg BID, selumetinib 25mg/kg QD or AZD5363 100mg/kg BID + selumetinib 25mg/kg QD. Mean fold change in tumor size is plotted +/-SEM. (H) Waterfall plot of 22RV1 xenograft tumor volumes after 6 weeks of treatment.

Fig 5. Analysis of tumors treated with combination MEK + Akt inhibition.

(A) Individual tumor growth curves for all mice in the study grouped by treatment. (B) Immunohistochemistry of pERK demonstrates detectable levels in 3 mice with early resistance to AZD5363 treatment (top); 3 other mice treated with AZD5363 are shown for comparison (bottom). (C) Staining of microarray specimens demonstrates that positive pERK staining was only evident in these mice. (D) The proportion of pS6 is decreased with AZD5363 and to a greater extent with combination AZD5363 + PD0325901. (E) Ki67 staining as a marker of proliferation. A tissue microarray was constructed using tumor samples in each group (7–9 per group). Mean scores of staining intensity graded by a blinded pathologist are shown +/- SEM.