Epithelial-mesenchymal transition status is a remarkable biomarker for the combination treatment with avutometinib and defactinib in KRAS-mutated non-small cell lung cancer

Abstract

Background: Recent therapeutic strategies for KRAS-mutated cancers that inhibit the MAPK pathway have attracted considerable attention. The RAF/MEK clamp avutometinib (VS-6766/CH5126766/RO5126766/CKI27) is promising for patients with KRAS-mutated cancers. Although avutometinib monotherapy has shown clinical activity in patients with KRAS-mutated cancers, effective combination strategies will be important to develop.

Methods: Using a phosphorylation kinase array kit, we explored the feedback mechanism of avutometinib in KRAS-mutated NSCLC cells, and investigated the efficacy of combining avutometinib with inhibitors of the feedback signal using in vitro and in vivo experiments. Moreover, we searched for a biomarker for the efficacy of combination therapy through an in vitro study and analysis using the The Cancer Genome Atlas Programme dataset.

Results: Focal adhesion kinase (FAK) phosphorylation/activation was increased after avutometinib treatment and synergy between avutometinib and FAK inhibitor, defactinib, was observed in KRAS-mutated NSCLC cells with an epithelial rather than mesenchymal phenotype. Combination therapy with avutometinib and defactinib induced apoptosis with upregulation of Bim in cancer cells with an epithelial phenotype in an in vitro and in vivo study.

Conclusions: These results demonstrate that the epithelial-mesenchymal transition status may be a promising biomarker for the efficacy of combination therapy with avutometinib and defactinib in KRAS-mutated NSCLC.

Fig. 1. Avutometinib activates FAK in KRAS-mutated NSCLC cells.

a The results of human RTK phosphorylation array. NCI-H358 cells are treated with or without avutometinib (150 nmol/L) for 72 h. b NCI-H358 and NCI-H2122 cells are treated with avutometinib (150 nmol/L) for indicated times, and each protein is detected in cell lysates using western blotting. c The indicated cells are treated with nonspecific control (siSCR) or siRNA specific to FAK (siFAK) for 72 h in the presence or absence of avutometinib (150 nmol/L), and cell viability is determined using CCK-8 assay. The cell viability rate of cells treated with siFAK in the presence of avutometinib is compared with that of cells treated with siSCR. p values are calculated using one-way ANOVA. *: p < 0.001. d SiSCR or siFAK are introduced into the indicated cells. After 24 h, the cells are incubated with or without avutometinib (150 nmol/L) for 72 h and lysed, and the indicated proteins are detected using western blotting. e The indicated cells are incubated with or without avutometinib (150 nmol/L) for 24 h, and quantitative RT-PCR is used to assess the expression of IGF-1 at the mRNA level in indicated cells. p values are calculated using student’s t-tests. f The indicated cells are treated with avutometinib (150 nmol/L) for indicated times, and each protein is detected in cell lysates using western blotting.

Fig. 2. EMT status correlates with synergistic effects of defactinib in combination with avutometinib.

a The KRAS-mutated NSCLC cell lines NCI-H358, NCI-H2122, NCI-H441, A549, SW1573, Calu-1, NCI-H23, Calu-6, and HOP-62 are lysed, and the indicated proteins are detected using western blotting. The mutation status of KRAS in these cell lines obtained from the database are shown. b The combination index with avutometinib and defactinib in the indicated cell lines is calculated using CompuSyn®. c Based on the result of (a) and (b), cells are further divided into groups with high and low expression of phosphorylation of FAK, KRAS G12C, and non-G12C point mutations; comparative data are shown (p = 0.392, p = 0.879, respectively; p values are calculated using student’s t-tests). d Hierarchical clustering and heatmap of EMT signature gene expression levels in the indicated cell lines from the Cell Model Passports dataset. e Distribution of the 9 KRAS-mutated NSCLC cell lines in the epithelial-mesenchymal score space following the GSVA scoring. f The correlation coefficients between the calculated EMT scores and the combination index are examined. The white boxes and black circles in (e) and (f) correspond to epithelial and mesenchymal-type cell lines, respectively, as shown in (a). EMT epithelial-mesenchymal transition, NSCLC Non-small cell lung cancer, GSVA Gene set variation analysis.

Fig. 3. Defactinib sensitises epithelial phenotype KRAS-mutated non-small cell lung cancer cells to avutometinib.

Cells are cultured with avutometinib (150 nmol/L) in the presence or absence of defactinib (1 mol/L) for 6 h (a), 24 h (c), and 72 h (d). The cells are lysed, and the indicated proteins detected using western blotting. b The sub G1 population of the indicated cells treated with avutometinib (15  nmol/L) in the presence or absence of defactinib (1 mol/L) is analysed via flow cytometry using FACS Accuri. The sub G1 rate of the combination therapy is compared to that of each single treatment in the indicated cells. p values are calculated using one-way ANOVA.

Fig. 4. Induction of epithelial-mesenchymal transition reduces the effectiveness of the combination treatment of avutometinib and defactinib.

a Representative microscopic images of NCI-H358 cells and H358-TGF-β cells are shown. b The cells are lysed, and the indicated proteins detected using western blotting. c Cell growth is determined using CCK-8 assay. d Cell cycle is defined as collection reduces via flow cytometry using FACS Accuri. e The combination index with avutometinib and defactinib in cells is calculated using CompuSyn®. The combination index of NCI-H358 is compared with that of H358-TGF-β. p values are calculated using student’s t-tests. f The cells are treated with avutometinib (150 nmol/L) in the presence or absence of defactinib (1 μmol/L) for 6 h. The cells are lysed, and the indicated proteins detected using western blotting.

Fig. 5. Combination of defactinib with avutometinib inhibits the growth of NCI-H358 tumours in vivo.

a NCI-H358 cell-line-derived xenograft (CDX) model mice, after randomisation, are treated daily by oral administration for 28 days with vehicle, defactinib (10.0 mg/kg), avutometinib (0.3 mg/kg), or a combination with defactinib (10.0 mg/kg) and avutometinib (0.3 mg/kg) (n = 6 per group). Tumour volumes are measured over time from the start of treatment, and the results are shown (mean ± SEM). b Images of NCI-H358 CDX tumours at the end of the experiment are shown. c NCI-H358 CDX tumours weights in each group at the end of the experiment are compared. p values are calculated using one-way ANOVA. d The NCI-H358 CDX model mice are treated with avutometinib (0.3 mg/kg) with or without defactinib (10.0 mg/kg) for 4 days. After collection, the tumours are lysed, and the indicated proteins detected using western blotting. e Representative images of hematoxylin & eosin staining and immunohistochemical staining with antibodies specific for human Ki-67 and TUNEL of tumours collected after 4 days of the treatment with each drug on NCI-H358 CDX model are shown. f The percentage of the Ki-67- and TUNEL-positive cells in each group is calculated from the average of the five evaluated areas. These are compared in each group (mean ± SD). p values are calculated using one-way ANOVA.