Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Sep 1.
Published in final edited form as: J Thorac Oncol. 2014 Sep;9(9):1285–1293. doi: 10.1097/JTO.0000000000000245

Co-targeting c-Met and COX-2 leads to enhanced inhibition of lung tumorigenesis in a murine model with heightened airway HGF

Laura P Stabile 1, Mary E Rothstein 2, Christopher T Gubish 3, Diana E Cunningham 4, Nathan Lee 5, Jill M Siegfried 6
PMCID: PMC4134355  NIHMSID: NIHMS608280  PMID: 25057941

Abstract

Background

The hepatocyte growth factor (HGF)/c-Met pathway is often dysregulated in non-small cell lung cancer (NSCLC). HGF activation of c-Met induces cyclooxygenase-2 (COX-2), resulting in downstream stimulation by PGE2 of additional pathways. Targeting both c-Met and COX-2 might lead to enhanced anti-tumor effects by blocking signaling upstream and downstream of c-Met.

Methods

Effects of crizotinib or celecoxib alone or in combination were tested in NSCLC cells in vitro and in mice transgenic for airway expression of human HGF (HGF TG).

Results

Proliferation and invasion of NSCLC cells treated with a combination of crizotinib and celecoxib was significantly lower compared to single treatments. Transgenic mice showed enhanced COX-2 expression localized to preneoplastic areas following exposure to the tobacco carcinogen 4-(methylnitrosoamino)-1-(3-pyridyl)-1-butanone (NNK), which was not present without carcinogen exposure. This shows that COX-2 activity is present during lung tumor development in a high HGF environment. Following NNK treatment, a significant decrease in the number of lung tumors per animal was observed after 13 week treatments of crizotinib, celecoxib or the combination compared to placebo (P<0.001). With combination treatment, the number of tumors was also significantly lower than single agent treatment (P<0.001). In the resulting lung tumors, P-c-Met, COX-2, PGE2, and P-MAPK were significantly down-modulated by combination treatment compared to single treatment. Expression of the epithelial-mesenchymal transition (EMT) markers E-cadherin and snail were also modulated by combination treatment.

Conclusions

In the presence of high HGF, dual inhibition of c-Met and COX-2 may enhance anti-tumor effects. This combination may have clinical potential in NSCLCs with high HGF/c-Met expression or EMT phenotype.

Keywords: HGF, COX-2, c-Met, lung cancer, crizotinib, celecoxib

Introduction

Targeted therapies are increasingly being used in the treatment of lung cancer, and this approach is most beneficial in patients with dysregulation of the specific pathway being inhibited.1, 2 In the case of inhibitors of the c-Met pathway, clinical benefit has been seen in non-small cell lung cancer (NSCLC) patients with either dysregulated c-Met expression (through gene amplification or over-expression)3 or with dysregulated ligand production (increased hepatocyte growth factor [HGF] expression).4 We recently showed that increased HGF serum levels are very common in lung cancer patients5, and increased HGF serum level was found by us and others to be a biomarker for presence of lung cancer.5–7 We also have reported increased susceptibility to development of lung cancer in a transgenic mouse that over-expresses HGF in the airways.8 This increased tumorigenic effect in response to a tobacco carcinogen can be overcome by use of a neutralizing antibody to HGF.8

One of the downstream effects of c-Met signaling in response to HGF is induction of COX-29–12, the inducible rate-limiting enzyme responsible for production of prostaglandin E2 (PGE2). PGE2 mediates pro-tumor effects through paracrine stimulation of inflammatory cells as well as through autocrine stimulation of its receptors on tumor cells.13, 14 One of the important pro-tumor effects of PGE2 in NSCLC is repression of E-cadherin expression13, leading to stimulation of epithelial-mesenchymal transition (EMT), a key mechanism that promotes invasion and tumor progression.15 We found in NSCLC cells that COX-2 was commonly induced following HGF treatment, with subsequent heightened release of PGE2.9 PGE2 in turn caused release of ligands for the EGFR pathway that led to activation of EGFR signaling along with a delayed activation of c-Met that was not dependent on HGF.9 Further mechanistic studies in our laboratory showed that functional c-Met was needed for maximal invasion in response to EGFR activation, which was not blocked by an HGF neutralizing antibody16, showing that c-Met activation through lateral signals from other tyrosine kinase receptors, independent of HGF, contributes to the pro-tumor effects of c-Met.

In this report, we tested the relevance of our published mechanistic data on lateral signaling to determine if enhanced anti-tumor effects could be achieved by blocking the activation of c-Met by HGF using a c-Met small molecule tyrosine kinase inhibitor, along with blockade of c-Met downstream reinforcing signaling that is initiated by PGE2 using a COX-2 inhibitor. Effects of dual inhibition were first evaluated in cell culture using NSCLC cell lines and then were validated in an animal model with increased HGF expression in the airways that also shows evidence of increased COX-2 activity in response to carcinogen treatment. Dual treatment with the c-Met small molecule tyrosine kinase inhibitor (TKI) crizotinib and the COX-2 inhibitor celecoxib resulted in greater anti-tumor effects in both systems. In the animal model, evidence of reduced downstream c-Met signaling was found and biomarkers associated with EMT were also maximally inhibited by combination treatment. This combination has potential clinical benefit for lung cancer patients with dysregulated HGF/c-Met pathways and/or evidence of an EMT phenotype.

Materials and Methods

Reagents and Cell Culture

NSCLC cell lines 201T and 273T were established in our laboratory from primary tissue and maintained at 37°C in 5% CO2.17 All cells were grown in Basal Medium Eagle (BME) (Invitrogen Life Technologies, Inc.) with 10% fetal bovine serum (FBS) and 2mM L-glutamine. All cell lines were validated by genotyping within 2 months of conducting the experiments. Crizotinib was purchased from ChemieTek (Indianapolis, IN). Celecoxib in capsule form was obtained from the University of Pittsburgh Cancer Institute Pharmacy. Celecoxib was dissolved in DMSO and the molarity of the stock solution was adjusted based on the pharmaceutical formulation. HGF was purchased from R&D Systems (Minneapolis, MN).

Cell Proliferation Assay

NSCLC cells were plated in 96-well plates at 1×104 cells/well and grown in BME supplemented with 10% FBS. Experimental treatments were added as indicated in the figure legend. Cell proliferation was monitored after 72 hr using the CellTiter 96 Aqueous One Solution Cell Proliferation Assay Kit (Promega Corporation). Metabolically active cells were labeled with MTS tetrazolium for 2 hrs and measured at 490nM using an absorbance plate reader (Bio-Rad Laboratories). The data were analyzed by Prism4 (Graph Pad software) to determine the IC25 and IC50 concentrations of each inhibitor.

Matrigel Invasion Assay

Invasion assays were conducted using 24-well Matrigel-coated transwell chambers (BD Biosciences). NSCLC cells (1×105) were plated in full serum in the upper chamber. Both the upper and lower chambers contained drug treatment. Non-invading cells in the upper chamber were removed after 24 hr by cotton swab, and invading cells were fixed and stained using the Diff-Quik staining solution kit, according to the manufacturer’s instructions (VWR International, Radnor, PA). The number of invading cells was counted at 10× magnification. Invaded cell number was normalized to cell proliferation at the same time point for each treatment. The mean ± SE was calculated from 3 independent experiments.

Protein Extraction and Western Analysis

Lung cancer cells were grown to 85% confluency in T75 flasks. Protein extraction, quantitation and Western analysis were performed as described previously.9

Murine Tumorigenesis Model

All mice used for the experiments were human HGF transgenic (FVB/N strain; HGF TG), permissive in the airways under control of the Clara cell secretory protein (CCSP) promoter, and heterozygous for the transgene with high copy number as described previously.8 These mice show enhanced lung tumorigenesis compared to their wild-type littermates and express 3–4 times the amount of HGF in the airways compared to wild type littermates.8 Breeding and identification of transgenic mice were as described previously.8 Mice were given a total of eight i.p. injections of 3mg NNK (15µg/µl) over 4 weeks. Celecoxib (50mg/kg), crizotinib (40mg/kg) or the combination was administered five times per week by oral gavage from week 3 until week 15. At the end of week 15, animals were sacrificed and lungs were formalin inflated. Dose of crizotinib was based on our previously published results in reducing lung carcinogenesis.5 Celecoxib dose was based on our prior observations in lung cancer xenograft studies (data not shown); the dose used falls in the low-dose range according to other models.18 We did not observe any systemic toxicity of either the NNK treatments or the therapeutic treatments. Mice did not show weight loss, hair loss, bruising or general toxicity during these experiments. Tumors on the surface of the lungs were counted under a dissecting microscope and tumor size (defined as surface area, mm2) was measured using Motic Images 2000 software. Animal care was in strict compliance with the institutional guidelines established at the University of Pittsburgh. This protocol was approved by the University of Pittsburgh IACUC committee (Protocol Number 12057507).

Immunohistochemistry

Mouse lungs were fixed in 10% buffered formalin. Lung samples were paraffin embedded, sliced and mounted on slides. Slides were stained with the following antibodies: PGE2 (1:750; Ab2318; Abcam), COX-2 (1:100; 160107; Cayman Chemical), P-MAPK (1:400; 4370; Cell Signaling Technology), E-cadherin (1:500; 160107; Cayman Chemical), Snail (1:500; Ab53519; Abcam) and P-c-Met (1:200; 3077; Cell Signaling Technology). All antibodies react with mouse antigens. Quantitation for COX-2, PGE2, P-MAPK and P-c-Met was performed in 3–5 tumors per treatment group on a +1 to +4 scaling system: +1= 1–25% positive stained cells, +2= 26–50% positive stained cells, +3= 51–75% positive stained cells, +4=76–100% positive stained cells. All positive cells were scored including epithelial cells and immune cells. Immune cells were determined by histological examination of morphological characteristics by a certified pathologist of a serial H&E section from the same sample. The number of positive cells was counted for E-cadherin and snail. Scale bar= 50 or 100 µM depending on figure.

Statistical Analysis

For in vitro studies, ANOVA was used. Effect of in vitro drug combinations was tested for synergy using the method of Chou and Talalay.19 In the mouse model, Poisson regression was used to compare the mean number of tumors per mouse in each treatment group, and linear contrasts were used to determine which groups differed significantly from one another. To examine the differences in tumor sizes between treatment groups, we used a linear mixed effects regression model. For immunohistochemical quantitation, ANOVA was used. All statistical tests were two-sided with the threshold for statistical significance defined as P<0.05. Analyses were conducted with SAS v. 9.2 (Cary, NC).

Results

Anti-tumor effects of combined crizotinib and celecoxib

Because COX-2 action is downstream of c-Met signaling induced by HGF as well as upstream of HGF-independent lateral signaling to c-Met in NSCLC9, we examined whether the combination of inhibiting the c-Met pathway with crizotinib and the COX-2 pathway with celecoxib would have enhanced anti-proliferative effects in NSCLC. We first tested the combination in cell culture. The IC25 and IC50 of each inhibitor was first determined in two NSCLC cell lines with normal c-Met copy number: 201T cells which show high c-Met expression and low COX-2 basal expression9, and 273T cells which show low c-Met expression and high COX-2 basal expression (Figure 1A, inset).9 Both cell lines show induction of COX-2 above basal levels upon treatment with HGF.9 The IC25 for crizotinib was 4.3 µM in 201T cells and 2.5 µM in 273T cells; the IC50 for crizotinib was 7.8 µM and 4.6 µM in 201T and 273T cells, respectively. For celecoxib, the IC25 and IC50 in 201T cells were 23.6 and 48.6 µM, and 38.6 µM and 80.5 µM in 273T cells. The effect of combining each inhibitor at its IC25 or IC50 concentration (determined from prior concentration-response curves) was examined. At either the IC25 or IC50 concentration, increased anti-proliferation effects were observed for the combination that were statistically significant compared to single agent treatment. Figure 1A shows proliferation results at the IC50 concentrations of celecoxib and crizotinib. In both 201T and 273T cells, the combination resulted in 18–22% relative cell proliferation compared to control, while cell proliferation relative to control was 43–50%, as expected with each drug alone at its calculated IC50. Supplemental Figure 1 shows proliferation results at the IC25 concentration of each drug. A statistical test for synergy was negative and showed that the effects on proliferation of the combination were additive. Similar results were observed in an additional NSCLC cell line, H23 (data not shown).

Figure 1. Proliferation and invasion with single and combination treatments of celecoxib and crizotinib.

Figure 1

Open in a new tab

A) 201T and 273T lung cancer cells were treated with celecoxib or crizotinib alone or in combination using the IC50 concentrations of each drug for 72 hrs. MTS assay was performed to determine relative cell proliferation with each treatment. Control was set to 100. The results show significant differences between the combination treatment and each single agent as well as the control. Maximum decreases were observed with combination treatment. Similar results were obtained with the IC50 and IC25 conditions. *** P<0.001, ANOVA. B) 201T and 273T lung cancer cells were plated on Matrigel invasion chambers and were treated with celecoxib or crizotinib alone or in combination using the IC50 concentrations of each drug for 24 hrs. Cells were fixed, stained and counted for number of invading cells. The combination treatment maximally inhibited cellular invasion in both cell lines. *P<0.05; **P<0.005, ***P<0.001, ANOVA. n.s.= non-significant.

Since c-Met and PGE2 are key mediators of invasion, a phenotype that is driven largely by the EMT process13, 20, 21, we also examined the effect of the combination of celecoxib and crizotinib on invasion in 201T and 273T cells, using the same concentrations of celecoxib and crizotinib as shown for Fig. 1A. As seen in Figure 1B, the number of cells that invaded through the Matrigel matrix in 24 hrs was significantly decreased by 40% with celecoxib alone and by 44% with crizotinib alone in each cell line, while combination treatment decreased the number of invading cells by 73% in 201T cells (P<0.01 compared to celecoxib alone and P<0.05 compared to crizotinib alone). In 273T cells, the combination decreased the number of invading cells by 60% (P<0.05 compared to celecoxib alone and n.s. compared to crizotinib alone). 273T cells showed a higher degree of variation in the extent of invasion in replicates compared to 201T cells. The invasion phenotype showed additive (but not synergistic) effects with the drug combination as observed for the proliferation endpoint.

COX-2 is observed in preneoplasias formed after NNK treatment in HGF transgenic mice

We then sought to demonstrate the benefit of dual crizotinib/celecoxib treatment in an animal model of enhanced pulmonary HGF activity using an HGF TG mouse with transgene expression controlled by the Clara cell secretory protein8, modeling the increased levels of HGF that we have observed in many lung cancer patients.5 HGF TG mice express 3–4 times the amount of HGF in bronchoalveolar lavage fluid compared to wild type littermates and show enhanced c-Met pathway activity.8 These mice are more susceptible to both preneoplasia and lung tumor formation after NNK exposure.5, 8 The preneoplastic lesions that develop are associated with the lung parenchyma and contain closely packed epithelial cells with reduced nuclear to cytoplasmic ratio and papillary growth, resembling adenomatous hyperplasia.8 To determine if the COX-2 protein is a relevant target during the process of lung tumor development following NNK exposure in HGF TG mice, we examined lungs of these animals with and without NNK treatment before the onset of lung tumors. At 10 weeks after treatment, aged-matched HGF TG lungs exposed to placebo show no evidence of preneoplasia either near airways or in the peripheral lung, and little detectable COX-2 expression by immunohistochemistry(Figure 2A). However, lungs of HGF TG mice 10 weeks after NNK treatment show distinct highly positive COX-2 protein expression by IHC, which was localized in and around preneoplastic areas (Figure 2B, black arrow and high magnification inset). The epithelial cells within these areas demonstrate COX-2 positivity and highly elevated COX-2 protein is also observed in infiltrating leukocytes (red arrows). This suggests that elevated COX-2 expression is part of a host response in the microenvironment of preneoplasias. In contrast, although COX-2 expression was induced in preneoplastic areas by NNK treatment in wild type littermates, the extent of induction and the infiltration by leukocytes was much lower compared to HGF TG animals (Figures 2C and D). The infiltrating leukocytes appear to be a mixture of lymphocytes, macrophages and neutrophils. In addition, inflammatory cells in bronchioalveolar lavage fluid showed a 2.5- and 4-fold increase in the number of macrophages and neutrophils, respectively, in fluid of HGF TG mice compared to wild type mice (Supplemental Table 1). This observation shows that increased COX-2 expression is one of the selective changes occurring in the lungs of HGF TG mice following tobacco carcinogen exposure, and validates the HGF TG mouse as a model to test dual therapy.

Figure 2. Representative COX-2 staining in lungs of HGF TG mice with and without NNK treatment.

Figure 2

Open in a new tab

A) Control aged-matched lungs from HGF TG animals and C) wild type animals with no NNK treatment. B) HGF TG and D) wild type mouse lungs after animals were treated with 6mg NNK and preneoplastic lesions were allowed to form for 10 wks. These lesions are analogous to human atypical adenomatous hyperplasia. Lungs were formalin inflated and stained for COX-2 by IHC. Representative images are shown. Red arrows represent COX-2 positive infiltrating leukocytes. Black arrows represent area of high powered magnification.

Combination treatment with crizotinib and celecoxib reduces lung tumorigenesis in response to the tobacco carcinogen NNK in HGF TG mice

Combination treatment was then compared in a prevention of lung tumorigenesis experiment in which HGF TG mice were exposed to 24 mg total NNK over a period of 4 weeks and were treated with placebo, celecoxib alone, crizotinib alone, or the combination. Crizotinib (40mg/kg), celecoxib (50mg/kg), combined crizotinib and celecoxib, or placebo control was administered by oral gavage daily from week 3 until week 15, at which time lung tumors were evaluated. Concentrations were based on previous published studies in NSCLC xenografts. Table 1 shows the group sizes, mean number of tumors per animal, the range of tumors per animal, and mean and range of lung tumor size for each treatment group. Group sizes were increased for single and dual treatments because of anticipated low tumor yields that would make it difficult to subsequently evaluate tumor biomarkers. Both number of tumors per animal and range of tumor numbers were lower with single treatment compared to control, and were lowest with combination treatment. A highly significant decrease in the number of lung tumors per animal was observed with crizotinib treatment (mean 3.6, range 2–6), celecoxib treatment (mean 4.3, range 3–6) and combination treatment (mean 1.5, range 0–3) compared to placebo control (mean 9.9, range 6–13; P<0.001, Poisson regression). Celecoxib alone reduced lung tumor numbers by 57% (Figure 3A), while crizotinib reduced tumor numbers by 64%, and combination treatment reduced the number of tumors per mouse by 85%. The number of tumors in the combination treatment group was also significantly lower than either single agent treatment (P<0.001). Statistical assessment for synergy was negative based on exponential parameter estimates, and the effect of combination therapy was shown to be additive. Hyperplastic lesions were also present at this time point even when adenomas were present. We histologically examined the H&E stained lung sections for the number of hyperplastic lesions by treatment group. The combination treatment resulted in significantly fewer hyperplasia areas per lung (mean= 2.4; median= 2.0) compared to control (mean= 4.1; median= 4.0, P<0.001). Each single agent treatment group had decreased lesions with trends towards significance.

Table 1.

Results of NNK Exposure in HGF TG Mice

control celecoxib crizotinib combo
number of animals 10 13 16 21
total number of tumors 92 54 53 24
mean # of tumors per animal 9.98 4.33 3.60 1.50
# of tumors per animal (min, max) (6,13) (3,6) (2,6) (0,3)
mean tumor size (mm2) 0.38 0.22 0.20 0.15
tumor size std. dev. 0.37 0.10 0.17 0.06
tumor size (min, max) (0.10, 2.24) (0.07, 0.69) (0.03, 1.25) (0.07, 0.29)
Open in a new tab

Figure 3.

Figure 3

Open in a new tab

A) Boxplot of the number of tumors per mouse by treatment group analyzed by Poisson regression analysis. B) Boxplot of the log (tumor size mm2) by treatment group. A mixed effects model was used for analysis of tumor size. Horizontal line in box of boxplots represents the median. + represents the mean. **P<0.001, *P<0.01, ns= non-significant.

Lung tumors were also smaller in all treatment groups compared to control, and the range of tumor sizes was also smaller with treatments. As seen in Figure 3B, a significant decrease in mean tumor size was observed with crizotinib (mean 0.20mm2), celecoxib (mean 0.22mm2) and combination (mean 0.15mm2) compared to placebo (mean 0.38mm2, P<0.001, mixed effects modeling). Size of tumors with crizotinib treatment was more variable but was still significantly smaller than the control group. With combination treatment, the mean size of tumors (0.06mm2) was also significantly smaller than celecoxib alone (0.10mm2), but did not reach significance compared to mean size of crizotinib alone group (0.17mm2), most likely due to the large variation observed with crizotinib alone.

Biomarkers of Prevention Effect

Biomarkers that may be modulated along with therapeutic effect of single or combination treatments were examined by immunohistochemisty in lung tumors that were present at the time of sacrifice. Biomarkers in the c-Met and COX-2 signaling pathways examined included P-AKT, P-P38, P-c-Met, P-STAT3, COX-2, PGE2, and P-MAPK. Several of these proteins were previously shown to be modulated by combining celecoxib with the EGFR inhibitor erlotinib in head and neck cancer models.22 Four biomarkers (P-c-Met, COX-2, PGE2, and P-MAPK) were optimally down-modulated by combination treatment, compared to single treatment (Figure 4). P-c-Met was detected in control tumors with a mean staining score of +3. The mean score was reduced to +2 in celecoxib treated tumors and +1 in crizotinib treated tumors, demonstrating that inhibiting either COX-2 or c-Met reduced phosphorylation of c-Met. P-c-Met expression in combination treated tumors was barely detectable (mean score 0.5), demonstrating that inhibiting both drug targets resulted in optimal suppression of c-Met phosphorylation. Level of PGE2 in lung tumors was high (mean IHC score of +4) in both control and crizotinib-treated tumors, while mean PGE2 IHC score was reduced to +3 by celecoxib treatment and to an average of +1 by the combination. Similarly the mean COX-2 IHC score was +3 in control and crizotinib treated tumors and was reduced to +2 by celecoxib. Combination treatment reduced this score further to a mean of +1. COX-2 staining was observed in both the malignant and stromal compartments. Mean P-MAPK IHC score was reduced from +4 in controls to +3 with treatment with crizotinib or celecoxib alone, and reduced further to +2 with combination treatment. P-AKT, P-P38, and P-STAT3 showed little or modulation with treatment (data not shown), suggesting that P-MAPK is the main signaling molecule downstream of c-Met and PGE2 that is affected by dual treatment, consistent with the observed maximum down-regulation of P-c-Met, COX-2 and P-MAPK by combination treatment in the cell lines used for the proliferation and invasion studies (Supplemental Figure 2).

Figure 4. Biomarkers modulated by treatment.

Figure 4

Open in a new tab

P-c-Met, PGE2, COX-2 and P-MAPK expression was examined by IHC in tumors arising from different treatment groups. Lungs were fixed in 10% buffered formalin. Lung samples were paraffin embedded, sliced and mounted on slides. Antibodies and dilutions used are as follows: P-c-Met (1:200; 3077; Cell Signaling Technology), PGE2 (1:750; Ab2318; Abcam), COX-2 (1:100; 160107; Cayman Chemical) and P-MAPK (1:400; 4370; Cell Signaling Technology). Representative staining for each marker and treatment group is shown. Quantitation was performed from 3–5 tumors per treatment group on a +1 to +4 scaling system, and determining the mean score. +1= 1–25% positivity, +2= 26–50% positivity, +3= 51–75% positivity, +4 76–100% positivity. Scale bar= 100µm.

Because induction of EMT is an important downstream effect of COX-2 activation in lung cancer that contributes to invasion and metastasis13, we examine the EMT marker E-cadherin, which is important in maintaining intercellular adhesion. We also examined two transcription factors that regulate E-cadherin: snail13 and Zeb-1.19 E-cadherin and snail were both found to be modulated by both single and dual treatment when compared to control: E-cadherin expression was significantly up-regulated 3.3-fold (P<0.001) following treatment with celecoxib, 4-fold (P<0.001) after crizotinib, and 4.2-fold after combination treatment (P<0.001). The E-cadherin repressor, snail, was down-regulated 30% by both celecoxib and crizotinib (P<0.001) and was down-regulated 46% by dual treatment (P<0.001) (Figure 5A and B). Although E-cadherin expression was the highest and snail expression was lowest with combination treatment, expression of these markers in the combination treatment was not statistically different from single agent treatments. In this model, Zeb-1 was not found to be modulated by either single or dual treatment.

Figure 5. EMT was modulated by single and dual treatment.

Figure 5

Figure 5

Open in a new tab

Expression of E-cadherin and Snail was examined using immunohistochemistry in tumors arising from different treatment groups. Antibodies and dilutions used are as follows: E-cadherin (1:500; 610182; BD Biosciences), Snail (1:500; Ab53519; Abcam). Representative staining for each marker and treatment group is shown. Quantitation was performed from 3–5 high-powered tumor fields per area by counting the number of positive cells per field. ***, P<0.001; *, P<0.05. Scale bar= 50µm.

Discussion

Use of combination therapies offers the advantage of dual inhibition of molecules with pro-tumor effects. Dual inhibition may allow for greater anti-proliferative effects, as well as reduce or delay the emergence of resistance. Knowledge of how the two targets of interest interact may help identify good candidates for combination, as well as define the patient population most likely to respond. The mechanistic rationale for this study was based on our prior published data in NSCLC cell lines that COX-2 was an important downstream signaling molecule of the c-Met receptor tyrosine kinase.9 We commonly observed COX-2 induction in NSCLC in response to HGF, and examination of downstream events showed that PGE2 produced by COX-2 triggered activation of the EGFR pathway, leading to long-term activation of c-Met through EGFR lateral signaling that was independent of HGF.8 We found that this lateral pathway was very important for maximal induction of invasion.16 Thus c-Met and COX-2 participate in a signaling network that leads to reinforcing signaling of both EGFR and c-Met itself.8, 16 In the current study, as predicted by our published model, we showed that dual inhibition of c-Met with the small molecule TKI crizotinib and of COX-2 with the selective COX-2 inhibitor celecoxib, resulted in significantly less proliferation and invasion in NSCLC cells in culture compared to single treatment. These increased effects were additive in nature. Although blocking HGF with a neutralizing antibody has also been successful as part of a dual inhibitor strategy5, 8, a c-Met TKI has the advantage of additionally blocking HGF-independent signaling of c-Met caused by lateral signaling from other tyrosine kinases.8, 16, 23 Although crizotinib can also target ALK in NSCLC with ALK rearrangements, ALK rearrangement is not present in the lung cancer cells or mouse model used in these experiments. The concentration of crizotinib necessary to observe effects on cell proliferation in vitro was higher than the reported IC50 for inhibiting c-Met phosphorylation. One factor that may account for this observation is that NSCLC release ligands for EGFR and we have previously documented that EGFR signaling can substitute for c-Met signaling when c-Met is inhibited, making NSCLC cells more resistant to c-Met inhibition. In addition, Ron and Tie inhibition could also be involved in the effect of crizotinib.

To validate use of dual therapy in an in vivo model, we utilized an HGF TG mouse that expresses human HGF under the control of the CCSP promoter. This model is preferential to a human tumor xenograft model because HGF is a paracrine factor that is produced almost exclusively by stromal cells in lung tumors, and murine HGF produced by the stroma of human tumor xenografts is not well recognized by human c-Met, whereas human HGF is able to activate murine c-Met. The HGF TG mouse exhibits increased local HGF production in the lungs and increased susceptibility to both preneoplasia and lung cancer after carcinogen exposure8. Our prior observations showed that circulating HGF and the EGFR ligand amphiregulin are often elevated in lung cancer patients compared to smokers without lung cancer.5 In addition, the role of c-Met and EGFR lateral signaling suggests that EGFR can substitute for c-Met signaling and vice versa.16 Many NSCLCs with wild type EGFR are driven by both EGFR and HGF.

In this study we also showed that the target of celecoxib, COX-2, was highly expressed in the lungs of HGF TG mice within 10 weeks after exposure to the carcinogen NNK, and COX-2 expression was localized to preneoplasias that arose from NNK treatment. Some COX-2 protein localized to the lung epithelia itself in these preneoplastic lesions but most of it was found localized to inflammatory cells infiltrating these lesions. Inhibition of COX-2 expressed in infiltrating inflammatory cells should prevent release of PGE2 which is known to stimulate pro-tumor processes such as release of EGFR ligands and cytokines by tumor cells. By short circuiting COX-2, celecoxib could prevent reinforcing pro-tumor interactions in the tumor microenvironment. Inflammation is expected in response to NNK, but since T cells, macrophages and neutrophils express c-Met24, HGF present in the airways of TG mice may also drive infiltration of leukocytes. HGF is a known inflammatory molecule25 and COX-2 induction in response to HGF is part of that inflammatory process.10 Furthermore, tumor associated macrophages derived from primary lung tumors express high levels of both COX-2 and HGF.26 High HGF in the pulmonary environment is accompanied by presence of pulmonary COX-2 in the context of tobacco carcinogen exposure, suggesting that COX-2 is a rational target for combination with a c-Met inhibitor. Our observations are consistent with the literature showing that pulmonary inflammation is an important lung cancer risk factor27 and is often observed in smokers with chronic obstructive pulmonary disease who are at increased lung cancer risk.28 Moreover, serum or tissue HGF levels are high in many inflammatory diseases.29, 30

The combination of celecoxib and crizotinib yielded an additive inhibitory effect on lung tumor formation in which the resulting tumors were also smaller, and phosphorylation of c-Met was optimally reduced compared to celecoxib or crizotinib alone. In addition, the resulting tumors displayed optimal reduction of both COX-2 and PGE2 compared to single treatment. This suggests that both the induction of COX-2 that occurs through c-Met phosphorylation and the production of PGE2 by COX-2 were blunted by combination treatment. In contrast, single treatments resulted in a lesser degree of inhibition of these two targets by their respective inhibitors. The extent of activation of MAPK was also maximally reduced by the combination of celecoxib and crizotinib compared to single treatment, suggesting that the c-Met and PGE2 downstream signaling networks were disrupted. Because the process of EMT has a strong effect on the invasive and progressive nature of lung cancer, we also examined EMT markers in lung tumors that arose despite single or dual treatment. E-cadherin, which plays an important role in maintenance of adherent junctions and which acts to inhibit tumor metastasis31, was induced by both crizotinib and celecoxib, and was also highly induced by combination treatment. Snail, a transcription factor that suppresses E-cadherin31, was reduced by both crizotinib and celecoxib, and was further reduced by combination treatment. Reduction in the EMT phenotype may be a biomarker of response to c-Met and COX-2 inhibitors, alone or in combination. Changes in PGE2 might also be informative. An mRNA signature for NSCLC that consists of genes sorting with low E-cadherin and high snail expression, among other genes including several regulated by HGF, was recently published.31 This gene signature or other markers of EMT might also be valuable in classifying patients with an EMT phenotype who would respond to a c-Met/COX-2 combination targeted strategy, or in evaluating extent of response to such a combination. Recently, several mesenchymal markers including E-cadherin in combination with c-Met expression have been shown to be surrogate markers of response to c-Met inhibitors in SCLC patients.32

We have previously shown that the HGF transgene in this mouse strain produces both KRAS mutant and wild type tumors with a KRAS mutant frequency of approximately 40%.5 We analyzed a subset of lung tumors from this experiment for KRAS mutation. While we did observe a slight decrease in the KRAS mutation frequency rate with crizotinib treatment, but not with celecoxib treatment, this was not statistically significant. It has recently been shown that KRAS mutation is a mechanism of crizotinib resistance33 and it has been suggested that KRAS mutations may exist in separate subclonal populatons that lack ALK gene rearrangement and that resistance is the result of emergence of a second driver oncogenic driver. Whether or not this combination treatment would be beneficial in KRAS mutant tumors requires further exploration.

These results show that in a pulmonary environment with heightened HGF/c-Met pathway activity that models what is observed in many lung cancer patients, dual inhibition of the c-Met and COX-2 pathways may enhance anti-tumor effects. Such a combination has potential clinical benefit for lung cancer patients with dysregulated c-Met/COX-2 pathways, especially those that show evidence of an EMT phenotype. It is important to note that in this animal model, experimental treatment was administered concurrently with the tobacco carcinogen and represents a prevention rather than a therapeutic setting. Whether or not these same responses would be observed in the setting of previously established advanced lesions is not known. Future studies should test this combination in an animal model after lung adenocarcinomas are formed. Since COX-2 is induced downstream of other oncogenic drivers such as EGFR, combining COX-2 inhibition with targeted therapies based on specific molecular aberrations may be a general strategy for increasing efficacy in cancer. The combination of EGFR inhibitors with celecoxib has been examined in several clinical trials. This combination showed efficacy in head and neck cancer34, 35, and down-modulation of EGFR pathway markers correlated with response to combined treatment. In NSCLC, several trials have examined gefitinib or erlotinib in combination with celecoxib.36, 37 Although an increased response to the combination was not observed in unselected patients, both high tumor COX-2 expression38 and reduction in circulating markers of EMT39 have emerged as potential biomarkers of combination efficacy. Our results suggest that patients with tumor dysregulation of the HGF/c-Met pathway, perhaps in consort with evidence of COX-2 activity and/or EMT phenotype, might benefit from dual inhibition of the c-Met and COX-2 pathways. Clinical testing of such inhibitors in combination for lung cancer treatment or prevention is warranted.

Supplementary Material

1

Supplemental Figure 1. A) 201T and 273T lung cancer cells were treated with celecoxib or crizotinib alone or in combination using the IC25 concentrations of each drug for 72 hrs. MTS assay was performed to determine relative cell proliferation with each treatment. Control was set to 100. The results show significant differences between the combination treatment and each single agent as well as the control. Maximum decreases were observed with combination treatment. Similar results were obtained with the IC50 and IC25 conditions. * P<0.05; **P<0.001;*** P<0.001, ANOVA

2

Supplemental Figure 2. 273T lung cancer cells were serum deprived for 48 hr followed by treatment with 50ng/ml HGF for 10 min or overnight pre-treatment with 25µM celecoxib or 1µM crizotinib or a combination of the inhibitors. Immunoblots were performed for P-MAPK, T-MAPK, P-c-Met, T-Met, COX-2 and actin. Representative immunoblots are shown.

3

Acknowledgements

The authors wish to thank Beatriz Kanterewicz for technical assistance with Western blotting and cell culture.

Sources of Support: Supported by NCI R01 CA79882 to JMS and NCI SPORE in Lung Cancer P50 CA090440 to JMS. This research utilized the Animal Facility and Biostatistics Facility of the UPCI, which were supported by the UPCI Cancer Center Support Grant P30 CA047904.

Abbreviations

HGF

hepatocyte growth factor

COX-2

cyclooxygenase-2

NSCLC

non-small cell lung cancer

NNK

4-(methylnitrosoamino)-1-(3-pyridyl)-1-butanone

EMT

epithelial-mesenchymal transition

TKI

tyrosine kinase inhibitor

CCSP

Clara cell secretory protein

Footnotes

Conflict of Interest: None

Contributor Information

Laura P. Stabile, Department of Pharmacology and Chemical Biology, University of Pittsburgh, Pittsburgh, PA 15213, las22@pitt.edu; 412-623-2015

Mary E. Rothstein, Department of Pharmacology and Chemical Biology, University of Pittsburgh, Pittsburgh, PA 15213, meb63@pitt.edu; 412-623-7821

Christopher T. Gubish, Department of Pharmacology and Chemical Biology, University of Pittsburgh, Pittsburgh, PA 15213, cgubish@pitt.edu; 412-623-7821

Diana E. Cunningham, Department of Biostatistics, University of Pittsburgh, Pittsburgh PA 15213, Del12345@gmail.com; 412-600-6383

Nathan Lee, Department of Pharmacology and Chemical Biology, University of Pittsburgh, Pittsburgh, PA 15213, nathan.lee@imua.ksbe.edu; 412-623-7821.

Jill M. Siegfried, Department of Pharmacology, University of Minnesota and University of Pittsburgh, jsiegfri@umn.edu; 612-624-2584

References

  • 1.Paez JG, Janne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science. 2004;304:1497–1500. doi: 10.1126/science.1099314. [DOI] [PubMed] [Google Scholar]
  • 2.Ou SH, Kwak EL, Siwak-Tapp C, et al. Activity of crizotinib (PF02341066), a dual mesenchymal-epithelial transition (MET) and anaplastic lymphoma kinase (ALK) inhibitor, in a non-small cell lung cancer patient with de novo MET amplification. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2011;6:942–946. doi: 10.1097/JTO.0b013e31821528d3. [DOI] [PubMed] [Google Scholar]
  • 3.Goldman JW, Laux I, Chai F, et al. Phase 1 dose-escalation trial evaluating the combination of the selective MET (mesenchymal-epithelial transition factor) inhibitor tivantinib (ARQ 197) plus erlotinib. Cancer. 2012;118:5903–5911. doi: 10.1002/cncr.27575. [DOI] [PubMed] [Google Scholar]
  • 4.Surati M, Patel P, Peterson A, et al. Role of MetMAb (OA-5D5) in c-MET active lung malignancies. Expert Opin Biol Ther. 2011;11:1655–1662. doi: 10.1517/14712598.2011.626762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Stabile LP, Rothstein ME, Keohavong P, et al. Targeting of Both the c-Met and EGFR Pathways Results in Additive Inhibition of Lung Tumorigenesis in Transgenic Mice. Cancers (Basel) 2010;2:2153–2170. doi: 10.3390/cancers2042153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Pine SR, Mechanic LE, Enewold L, et al. Increased levels of circulating interleukin 6, interleukin 8, C-reactive protein, and risk of lung cancer. J Natl Cancer Inst. 2011;103:1112–1122. doi: 10.1093/jnci/djr216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Ujiie H, Tomida M, Akiyama H, et al. Serum hepatocyte growth factor and interleukin-6 are effective prognostic markers for non-small cell lung cancer. Anticancer Res. 2012;32:3251–3258. [PubMed] [Google Scholar]
  • 8.Stabile LP, Lyker JS, Land SR, et al. Transgenic mice overexpressing hepatocyte growth factor in the airways show increased susceptibility to lung cancer. Carcinogenesis. 2006;27:1547–1555. doi: 10.1093/carcin/bgl003. [DOI] [PubMed] [Google Scholar]
  • 9.Siegfried JM, Gubish CT, Rothstein ME, et al. Signaling pathways involved in cyclooxygenase-2 induction by hepatocyte growth factor in non small-cell lung cancer. Mol Pharmacol. 2007;72:769–779. doi: 10.1124/mol.107.034215. [DOI] [PubMed] [Google Scholar]
  • 10.Ogunwobi OO, Liu C. Hepatocyte growth factor upregulation promotes carcinogenesis and epithelial-mesenchymal transition in hepatocellular carcinoma via Akt and COX-2 pathways. Clin Exp Metastasis. 2011;28:721–731. doi: 10.1007/s10585-011-9404-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Lee YH, Suzuki YJ, Griffin AJ, et al. Hepatocyte growth factor regulates cyclooxygenase-2 expression via beta-catenin, Akt, and p42/p44 MAPK in human bronchial epithelial cells. Am J Physiol Lung Cell Mol Physiol. 2008;294:L778–L786. doi: 10.1152/ajplung.00410.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Jones MK, Sasaki E, Halter F, et al. HGF triggers activation of the COX-2 gene in rat gastric epithelial cells: action mediated through the ERK2 signaling pathway. FASEB J. 1999;13:2186–2194. doi: 10.1096/fasebj.13.15.2186. [DOI] [PubMed] [Google Scholar]
  • 13.Dohadwala M, Yang SC, Luo J, et al. Cyclooxygenase-2-dependent regulation of E-cadherin: prostaglandin E(2) induces transcriptional repressors ZEB1 and snail in non-small cell lung cancer. Cancer Res. 2006;66:5338–5345. doi: 10.1158/0008-5472.CAN-05-3635. [DOI] [PubMed] [Google Scholar]
  • 14.Sandler AB, Dubinett SM. COX-2 inhibition and lung cancer. Semin Oncol. 2004;31:45–52. doi: 10.1053/j.seminoncol.2004.03.045. [DOI] [PubMed] [Google Scholar]
  • 15.Neil JR, Johnson KM, Nemenoff RA, et al. Cox-2 inactivates Smad signaling and enhances EMT stimulated by TGF-beta through a PGE2-dependent mechanisms. Carcinogenesis. 2008;29:2227–2235. doi: 10.1093/carcin/bgn202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Dulak AM, Gubish CT, Stabile LP, et al. HGF-independent potentiation of EGFR action by c-Met. Oncogene. 2011;30:3625–3635. doi: 10.1038/onc.2011.84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Siegfried JM, Krishnamachary N, Gaither Davis A, et al. Evidence for autocrine actions of neuromedin B and gastrin-releasing peptide in non-small cell lung cancer. Pulm Pharmacol Ther. 1999;12:291–302. doi: 10.1006/pupt.1999.0210. [DOI] [PubMed] [Google Scholar]
  • 18.Okada T, Takigawa N, Kishino D, et al. Selective cyclooxygenase-2 inhibitor prevents cisplatin-induced tumorigenesis in A/J mice. Acta medica Okayama. 2012;66:245–251. doi: 10.18926/AMO/48564. [DOI] [PubMed] [Google Scholar]
  • 19.Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul. 1984;22:27–55. doi: 10.1016/0065-2571(84)90007-4. [DOI] [PubMed] [Google Scholar]
  • 20.Siebzehnrubl FA, Silver DJ, Tugertimur B, et al. The ZEB1 pathway links glioblastoma initiation, invasion and chemoresistance. EMBO Mol Med. 2013;5:1196–1212. doi: 10.1002/emmm.201302827. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Lin X, Shang X, Manorek G, et al. Regulation of the Epithelial-Mesenchymal Transition by Claudin-3 and Claudin-4. PloS one. 2013;8:e67496. doi: 10.1371/journal.pone.0067496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Fu S, Rivera M, Ko EC, et al. Combined Inhibition of Epidermal Growth Factor Receptor and Cyclooxygenase-2 as a Novel Approach to Enhance Radiotherapy. Journal of cell science & therapy. 2011;1 doi: 10.4172/2157-7013.s1-002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Han C, Michalopoulos GK, Wu T. Prostaglandin E2 receptor EP1 transactivates EGFR/MET receptor tyrosine kinases and enhances invasiveness in human hepatocellular carcinoma cells. J Cell Physiol. 2006;207:261–270. doi: 10.1002/jcp.20560. [DOI] [PubMed] [Google Scholar]
  • 24.Galimi F, Cottone E, Vigna E, et al. Hepatocyte growth factor is a regulator of monocyte-macrophage function. J Immunol. 2001;166:1241–1247. doi: 10.4049/jimmunol.166.2.1241. [DOI] [PubMed] [Google Scholar]
  • 25.Kaibori M, Yanagida H, Nakanishi H, et al. Hepatocyte growth factor stimulates the induction of cytokine-induced neutrophil chemoattractant through the activation of NF-kappaB in rat hepatocytes. J Surg Res. 2006;130:88–93. doi: 10.1016/j.jss.2005.09.025. [DOI] [PubMed] [Google Scholar]
  • 26.Wang R, Zhang J, Chen S, et al. Tumor-associated macrophages provide a suitable microenvironment for non-small lung cancer invasion and progression. Lung cancer. 2011;74:188–196. doi: 10.1016/j.lungcan.2011.04.009. [DOI] [PubMed] [Google Scholar]
  • 27.Azad N, Rojanasakul Y, Vallyathan V. Inflammation and lung cancer: roles of reactive oxygen/nitrogen species. J Toxicol Environ Health B Crit Rev. 2008;11:1–15. doi: 10.1080/10937400701436460. [DOI] [PubMed] [Google Scholar]
  • 28.Brody JS, Spira A. State of the art. Chronic obstructive pulmonary disease, inflammation, and lung cancer. Proc Am Thorac Soc. 2006;3:535–537. doi: 10.1513/pats.200603-089MS. [DOI] [PubMed] [Google Scholar]
  • 29.Nayeri F, Nilsson I, Hagberg L, et al. Hepatocyte growth factor levels in cerebrospinal fluid: a comparison between acute bacterial and nonbacterial meningitis. The Journal of infectious diseases. 2000;181:2092–2094. doi: 10.1086/315506. [DOI] [PubMed] [Google Scholar]
  • 30.Kern MA, Bamborschke S, Nekic M, et al. Concentrations of hepatocyte growth factor in cerebrospinal fluid under normal and different pathological conditions. Cytokine. 2001;14:170–176. doi: 10.1006/cyto.2001.0875. [DOI] [PubMed] [Google Scholar]
  • 31.Byers LA, Diao L, Wang J, et al. An epithelial-mesenchymal transition gene signature predicts resistance to EGFR and PI3K inhibitors and identifies Axl as a therapeutic target for overcoming EGFR inhibitor resistance. Clinical cancer research : an official journal of the American Association for Cancer Research. 2013;19:279–290. doi: 10.1158/1078-0432.CCR-12-1558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Canadas I, Rojo F, Taus A, et al. Targeting epithelial-to-mesenchymal transition with Met inhibitors reverts chemoresistance in small cell lung cancer. Clinical cancer research : an official journal of the American Association for Cancer Research. 2014;20:938–950. doi: 10.1158/1078-0432.CCR-13-1330. [DOI] [PubMed] [Google Scholar]
  • 33.Doebele RC, Pilling AB, Aisner DL, et al. Mechanisms of resistance to crizotinib in patients with ALK gene rearranged non-small cell lung cancer. Clinical cancer research : an official journal of the American Association for Cancer Research. 2012;18:1472–1482. doi: 10.1158/1078-0432.CCR-11-2906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Shin DM, Zhang H, Saba NF, et al. Chemoprevention of head and neck cancer by simultaneous blocking of epidermal growth factor receptor and cyclooxygenase-2 signaling pathways: preclinical and clinical studies. Clinical cancer research : an official journal of the American Association for Cancer Research. 2013;19:1244–1256. doi: 10.1158/1078-0432.CCR-12-3149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Saba N, Hurwitz SJ, Kono S, et al. Chemoprevention of Head and Neck Cancer with Celecoxib and Erlotinib: Results of a Phase 1b and Pharmacokinetic Study. Cancer prevention research. 2013 doi: 10.1158/1940-6207.CAPR-13-0215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Gadgeel SM, Ruckdeschel JC, Heath EI, et al. Phase II study of gefitinib, an epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI), and celecoxib, a cyclooxygenase-2 (COX-2) inhibitor, in patients with platinum refractory non-small cell lung cancer (NSCLC) Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2007;2:299–305. doi: 10.1097/01.JTO.0000263712.61697.69. [DOI] [PubMed] [Google Scholar]
  • 37.Reckamp KL, Krysan K, Morrow JD, et al. A phase I trial to determine the optimal biological dose of celecoxib when combined with erlotinib in advanced non-small cell lung cancer. Clinical cancer research : an official journal of the American Association for Cancer Research. 2006;12:3381–3388. doi: 10.1158/1078-0432.CCR-06-0112. [DOI] [PubMed] [Google Scholar]
  • 38.Fidler MJ, Argiris A, Patel JD, et al. The potential predictive value of cyclooxygenase-2 expression and increased risk of gastrointestinal hemorrhage in advanced non-small cell lung cancer patients treated with erlotinib and celecoxib. Clinical cancer research : an official journal of the American Association for Cancer Research. 2008;14:2088–2094. doi: 10.1158/1078-0432.CCR-07-4013. [DOI] [PubMed] [Google Scholar]
  • 39.Reckamp KL, Gardner BK, Figlin RA, et al. Tumor response to combination celecoxib and erlotinib therapy in non-small cell lung cancer is associated with a low baseline matrix metalloproteinase-9 and a decline in serum-soluble E-cadherin. Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer. 2008;3:117–124. doi: 10.1097/JTO.0b013e3181622bef. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1

Supplemental Figure 1. A) 201T and 273T lung cancer cells were treated with celecoxib or crizotinib alone or in combination using the IC25 concentrations of each drug for 72 hrs. MTS assay was performed to determine relative cell proliferation with each treatment. Control was set to 100. The results show significant differences between the combination treatment and each single agent as well as the control. Maximum decreases were observed with combination treatment. Similar results were obtained with the IC50 and IC25 conditions. * P<0.05; **P<0.001;*** P<0.001, ANOVA

NIHMS608280-supplement-1.TIF (67KB, TIF)
2

Supplemental Figure 2. 273T lung cancer cells were serum deprived for 48 hr followed by treatment with 50ng/ml HGF for 10 min or overnight pre-treatment with 25µM celecoxib or 1µM crizotinib or a combination of the inhibitors. Immunoblots were performed for P-MAPK, T-MAPK, P-c-Met, T-Met, COX-2 and actin. Representative immunoblots are shown.

NIHMS608280-supplement-2.TIF (148.7KB, TIF)
3
NIHMS608280-supplement-3.docx (13.6KB, docx)

RESOURCES