Aurora kinase A drives the evolution of resistance to third-generation EGFR inhibitors in lung cancer

Abstract

Although targeted therapies often elicit profound initial patient responses, these effects are transient due to residual disease leading to acquired resistance. How tumors transition between drug responsiveness, tolerance and resistance, especially in the absence of preexisting subclones, remains unclear. In epidermal growth factor receptor (EGFR)-mutant lung adenocarcinoma cells, we demonstrate that residual disease and acquired resistance in response to EGFR inhibitors requires Aurora kinase A (AURKA) activity. Nongenetic resistance through the activation of AURKA by its coactivator TPX2 emerges in response to chronic EGFR inhibition where it mitigates drug-induced apoptosis. Aurora kinase inhibitors suppress this adaptive survival program, increasing the magnitude and duration of EGFR inhibitor response in preclinical models. Treatment-induced activation of AURKA is associated with resistance to EGFR inhibitors in vitro, in vivo and in most individuals with EGFR-mutant lung adenocarcinoma. These findings delineate a molecular path whereby drug resistance emerges from drug-tolerant cells and unveils a synthetic lethal strategy for enhancing responses to EGFR inhibitors by suppressing AURKA-driven residual disease and acquired resistance.

Conflict of interest statement

COMPETING INTERESTS STATEMENT

H.J.H., L.R., A.D.S. and T.C.H. are employees of Clovis Oncology.

Figures

Figure 1.. EGFR mutant lung adenocarcinoma cells demonstrating acquired resistance to third-generation EGFR tyrosine kinase inhibitors are sensitive to Aurora kinase inhibition.

a Schematic of cell number throughout the process to generate acquired resistant EGFR mutant lung adenocarcinoma cell lines through continuous cell culture and stepwise dose escalation of either osimertinib or rociletinib from 10 nM to 1 uM over the course of 9 d. Cell lines and EGFR mutation are listed. b Mean relative proliferation of parental, osimertinib (denoted -OR) and rociletinib (denoted -RR) acquired resistant cell lines treated with the indicated agents and allowed to proliferate for 3 d. IC50 analysis of dose–response curves from n = 4 biologically independent samples. The IC50 for each cell line is indicated in parenthesis. c Immunoblot analysis showing activity of the EGFR, AKT and ERK as well as PARP cleavage in response to 24 h treatment (+) or not (−) with DMSO, osimertinib (1uM) or rociletinib (1uM) in parental or acquired resistant cell lines. Actin is loading control. cl. PARP = cleaved PARP. Experiment was perfomed twice with similar results. d Sorted results from a combinatorial drug screen across 94 drugs combined with 2uM rociletinib in H1975-RR cells. Synergy based on enhancement of growth inhibition compared to either drug along (see Methods). Screen was performed once. e Crystal violet staining of parental and osimertinib acquired resistant cell lines or f rociletinib acquired resistant cell lines 9 d after treatment with DMSO or the indicated drugs. Aurora kinase inhibitors are annotated with their relative targets in order of potency. Quantification (relative number of stained cells) is shown on the bottom right. c,e,f are representative of two independent experiments. Error bars are s.e.m. Full blots are shown in Supplementary Fig. 11.

Figure 2.. Activation of Aurora Kinase A is sufficient to cause resistance to EGFR-TKI and drug combinations induce apoptosis through BIM upregulation in vitro and in vivo.

a Immunoblot analysis of total and phosphorylated AURKA in PC9 and H1975 parental and acquired resistant cell lines treated with 1uM of the indicated inhibitors for 24 h. b Mean of cell proliferation of PC9 or H1975 cells transfected with plasmids expressing the indicated genes and treated 1uM EGFR-TKI for 72 h compared to DMSO treated cells performed in n=3 biologically independent samples. Significance based on comparison to LacZ control. c Immunoblot analysis 4 parental and 8 acquired resistant cell lines. Quantified intensities for pAURKA and TPX2 relative to the parental cell line is shown. d Proliferation compared to DMSO of PC9 and H1975 parental or acquired resistant cells treated with 1uM of osimertinib or rociletinib, 30nM of MLN8237 or the combination for 72 h. Mean over n=3 biologically independent samples. e Apoptosis measured by YO-PRO-1 positivity in the same models and drug treatments for 72 h. Shown is mean from n=3 biologically independent samples. f Mean tumor volume (mm3) of PC9-RR xenografts during treatment with rociletinib (100mg/kg), MLN8237 (10mg/kg) or the combination. n=10 tumors in the vehicle and rociletinib arm, and n=7 in the MLN8237 and combination arm. g Percent change in tumor volume compared to baseline for individual PC9-OR cell xenografts teated for 11 days with osimertinib (5mg/kg), MLN8237 (10mg/kg) or the combination. P-value comparing combo to single agent osimertinib treatment. h Immunoblot of lysates from parental PC9, PC9-OR and PC9-RR cells treated with the indicated inhibitors or DMSO for 24 h. i Proposed mechanism for the efficacy of the combination in acquired resistant cells. P-values based two-tailed Student’s t-test. Error bars represent s.e.m. Blots are representative of at least two independent experiments. Full blots shown in Supplementary Fig. 11.

Figure 3.. EGFR inhibition leads to the activation of TPX2 and AURKA during drug tolerance where it is necessary for survival and emergence of acquired resistance in vitro.

a Immunoblot of lysates from H1975 cells treated with osimertinib between 0 and 9 days and in H1975 osimertinib acquired resistant cells (H1975-OR) which have undergone continual treatment for >6 weeks. b Quantification of the indicated molecules in (a) normalized to actin and scaled by maximal intensity. c Immunoblot of indicated molecules lysates from drug tolerant PC9 or H1975 cells that persist after treatment with 1uM of the indicated agents for 9 d or in the parental cell line. d Immunoblot of PC9 drug tolerant persister (DTP) cells formed by treatment with 1uM of EGFR inhibitors erlotinib (first generation), afatinib (second generation), rociletinib or osimertinib (both third generation) for 9 d. e Growth of PC9 or H1975 cells in culture over time after plating and treated with the indicated drugs. Shown is mean over n=4 biologically independent samples per time point. f Time to resistance defined as days to reach exponential growth in various treatment conditions. n.a. = resistance not achieved after 90 d. g Clonogenic growth of PC9 and H1975 cells treated with either 1uM rociletinib or the combination of rociletinib and 30nM MLN8237 for 9 d. Images represents n=2 independent experiments. h Drug tolerant persister PC9 or H1975 cells were generated through 9 d of treatment with 1uM of osimertinib or rociletinib and then exposed to either 1uM EGFR-TKI alone or with that addition of 30nM MLN8237 for up to 7 d. Shown is mean of percent cells remaining relative to day 0 calculated from n=4 biologically independent samples per time point. Error bars s.e.m. c,d are representative images of at least two independent experiments. Full blots shown in Supplementary Fig. 11.

Figure 4.. Clinical potential of combined EGFR and Aurora kinase inhibition on residual disease and acquired resistance.

a Growth of patient derived xenograft derived from a patient with an EGFR L858R mutant lung adenocarcinoma displaying an incomplete response to erlotinib biopsied at the point of residual disease. Mice were treated with 100mg/kg rociletinib, 10mg/kg MLN8237 or the combination of rociletinib and MLN8237 for 30 d. P-value based on comparison of combination to rociletinib alone. Shown are averages over six tumors per arm. b Growth of individual PDX tumors treated 1mg/kg osimertinib, 10mg/kg MLN8237 or the combination for 30 d. P-value based on comparison of combination to osimertinib alone. Shown are averages over four tumors the osimertinib arm and six tumors in all others. c Immunoblot of lysates from individual PDX tumors harvested 30 d after the initiation of treatment with the indicated compounds. cl. PARP = cleaved PARP. Images represents n=4 independent tumors and was performed once. d Mean TPX2 levels from PDX tumors based on quantification of immunohistochemical (IHC) staining. e Representative image of TPX2 IHC from PDX tumors treated with the indicated agents. Images taken at 20× magnification. Scale bar, 100¼M. Data representative of six independent tumors. f Mean TPX2 IHC scores from 9 matched tumor samples from EGFR-mutant lung adenocarcinoma patients taken at the time of diagnosis and at the time of relapse to erlotinib. g TPX2 IHC score of patient tumor tissue obtained before (ending with A) and upon relapse on erlotinib (ending with B). Same numbers indicate tumors from same patient. Known genetic drivers of resistance based on genomic analysis are indicated. h Quantification of TPX2 IHC scores from human tumor samples taken from EGFR-mutant lung adenocarcinoma patients at the time of diagnosis (ending with the letter A) and at the time of relapse after treatment with either osimertinib or rociletinib (ending with letter B). i Mean TPX2 IHC score from all tumors taken from EGFR-mutant lung adenocarcinoma patients before treatment and after resistance to the indicated TKIs. All plots are mean with error bars as s.e.m. in panels a and b, and s.d. in panels d,f and i. P-values based two-tailed Student’s t-test. Full blot images are shown in Supplementary Fig. 11.