Dual ALK and EGFR inhibition targets a mechanism of acquired resistance to the tyrosine kinase inhibitor crizotinib in ALK rearranged lung cancer

Abstract

Introduction: The multitargeted tyrosine kinase inhibitor (TKI) crizotinib is active against ALK translocated non-small-cell lung cancer (NSCLC); however acquired resistance invariably develops over time. ALK mutations have previously been implicated in only a third of resistant tumors. We sought to evaluate alternative mechanisms of resistance and preclinical strategies to overcome these in a cell line driven by EML4-ALK.

Methods: We selected the NSCLC cell line NCI-H3122 (H3122: EML4-ALK E13;A20) and derived resistant variants that were able to grow in the presence of 1 μM crizotinib. These were analyzed for ALK mutations, sensitivity to crizotinib in combination with other TKIs, and for activation of alternative tyrosine kinases.

Results: All H3122 crizotinib resistant (CR) clones lacked amplification or mutations in the kinase domain of ALK. To evaluate if possible alternative kinases functioned as "bypass" tracks for downstream signaling activation in these resistance cells, we performed of phosho-receptor tyrosine kinase array that demonstrated that CR clones had higher phospho-EGFR signals than H3122 cells before and after exposure to crizotinib. A functional approach of dual ALK TKI (with crizotinib) with combinatory TKI inhibition was used as a secondary screen for possible targets. Crizotinib+erlotinib (reversible EGFR TKI) and crizotinib+afatinib (irreversible EGFR/ERBB2 TKI) were able to inhibit the growth of H3122 CR clones, confirming EGFR activation as a mechanism of resistance. The removal of crizotinib from the culture media re-sensitized CR cells to crizotinib.

Conclusions: We identified activation of EGFR as a mechanism of resistance to crizotinib in preclinical models of ALK translocated NSCLC. If EGFR activation is confirmed as a predominant mechanism of ALK TKI-induced resistance in patient-derived tumors, the use of ALK plus EGFR TKIs could be explored for this important cohort of NSCLCs.

Keywords: Anaplastic lymphoma kinase; Crizotinib; Epidermal growth factor receptor; Kinase inhibitor; Non-small-cell lung cancer; Tyrosine kinase.

FIGURE 1

H3122 crizotinib resistant (CR) cells. A. H3122 cells are sensitive to crizotinib (1μM) and western blot analysis demonstrates inactivation of ALK, AKT and ERK 1/2 phosphorylation after 6 hours of exposure to crizotinib. B. Proliferation assay evaluating the inhibitory effect of 1μM crizotinib in H3122 cells and three H3122 CR cells: CR_A, CR_B and CR_C. The 50% inhibitory concentration (IC₅₀) was calculated using dose-response curves (data not shown) and indicated in paranthesis. C. Western blot analysis of protein extracts from H3122 CR cells demonstrates inactivation of ALK phosphorylation but persistence of AKT and ERK 1/2 phosphorylation after 6 hours of exposure to crizotinib. Crizotinib was removed from the growth media of H3122 CR cells two days prior to commencing the experiment.

FIGURE 2

Activation of EGFR in H3122 CR cells. A. Phospho-receptor tyrosine kinase (RTK) array of H3122 and H3122 CR cells exposed to 6 hours of control conditions (DMSO) or 1μM crizotinib. In all H3122 CR cells, EGFR was the main signal observed (black arrows). B. Western blot analysis of protein extracts from H3122 and H3122 CR cells demonstrates increased levels of phosphorylated EGFR and upregulation of EGFR in all H3122 CR cells after 6 hours of exposure to control conditions (DMSO, indicated as minus sign) or crizotinib (1μM, indicated as positive sign). Crizotinib was removed from the growth media of H3122 CR cells two days prior to commencing the experiment.

FIGURE 3

Proliferation assay evaluating multitargeted tyrosine kinase inhibitors (TKIs) in combination with crizotinib in H3122 NSCLC cell lines. Cells were plated at a density of 3000–5000 cells/well for H3122 and CR clones. All experiments were performed in triplicate. A. Inhibitory profile of imatinib (1μM) in combination with crizotinib (1μM). B. Inhibitory profile of sorafenib (1μM) in combination with crizotinib (1μM). C. Inhibitory profile of erlotinib (1μM) in combination with crizotinib (1μM). D. Inhibitory profile of afatinib (1μM) in combination with crizotinib (1μM). Results are displayed in bar columns with standard deviation in relation to cell viability. Treatment with DMSO (indicated as a concentration of 0) was used as the standard for 100% cell viability in each cell line. * indicates a p-value <0.05 (see text for exact p-values) for treated cells.

FIGURE 4

Combination of EGFR and ALK TKIs in H3122 CR_A cells inhibit downstream targets. Western blot analysis of protein extracts from H3122 CR_A cells after 6 hours of exposure to control conditions (DMSO), crizotinib (1μM), afatinib (1μM) or the combination of 1μM crizotinib with 1μM afatinib. The extracts obtained in the combination experiment demonstrate inactivation of ALK, EGFR, AKT and ERK 1/2 phosphorylation; while the treatment with single agents was unable to inhibit all these signals simultaneously. Crizotinib was removed from the growth media of DMSO and afatinib alone conditions a day prior to commencing the experiment.

FIGURE 5

Continuous exposure to crizotinib is important to maintain acquired resistance to crizotinib in H3122 CR_A cells. A. Proliferation assay evaluating the inhibitory effect of 1μM crizotinib in H3122 CR_A cells that were maintained in growth media with 1μM crizotinib or H3122 CR_A cells that were grown in growth media without crizotinib for 4 weeks (indicated in parenthesis). Treatment with DMSO (indicated as a minus sign) was used as the standard for 100% cell viability in each cell line. B. Western blot analysis of protein extracts from H3122 CR_A cells demonstrates decreased levels of phosphorylated EGFR and downregulation of EGFR in H3122 CR_A cells that were grown in the absence of crizotinib in the growth media when compared to H3122 CR_A cells grown in the presence of crizotinib, after re-exposure to crizotinib (1μM) for 6 hours. C. Proliferation assay evaluating the inhibitory effect of 1μM crizotinib in the presence or absence of EGF (at 100 ng/mL). Treatment with DMSO (indicated as a minus sign) was used as the standard for 100% cell viability in each cell line.