Chemoprevention of head and neck cancer by simultaneous blocking of epidermal growth factor receptor and cyclooxygenase-2 signaling pathways: preclinical and clinical studies

Abstract

Purpose: We investigated the efficacy and underlying molecular mechanism of a novel chemopreventive strategy combining EGF receptor (EGFR) tyrosine kinase inhibitor (TKI) with cyclooxygenase-2 inhibitor (COX-2I).

Experimental design: We examined the inhibition of tumor cell growth by combined EGFR-TKI (erlotinib) and COX-2I (celecoxib) treatment using head and neck cancer cell lines and a preventive xenograft model. We studied the antiangiogenic activity of these agents and examined the affected signaling pathways by immunoblotting analysis in tumor cell lysates and immunohistochemistry (IHC) and enzyme immunoassay (EIA) analyses on the mouse xenograft tissues and blood, respectively. Biomarkers in these signaling pathways were studied by IHC, EIA, and an antibody array analysis in samples collected from participants in a phase I chemoprevention trial of erlotinib and celecoxib.

Results: The combined treatment inhibited head and neck cancer cell growth significantly more potently than either single agent alone in cell line and xenograft models, and resulted in greater inhibition of cell-cycle progression at G1 phase than either single drug. The combined treatment modulated the EGFR and mTOR signaling pathways. A phase I chemoprevention trial of combined erlotinib and celecoxib revealed an overall pathologic response rate of 71% at time of data analysis. Analysis of tissue samples from participants consistently showed downregulation of EGFR, pERK, and pS6 levels after treatment, which correlated with clinical response.

Conclusion: Treatment with erlotinib combined with celecoxib offers an effective chemopreventive approach through inhibition of EGFR and mTOR pathways, which may serve as potential biomarkers to monitor the intervention of this combination in the clinic. Clin Cancer Res; 19(5); 1244-56. ©2013 AACR.

©2013 AACR.

Figures

Figure 1. Effects of erlotinib and celecoxib on growth of SCCHN cell lines

SCCHN cell lines Tu212 (A) and Tu686 (B) were treated with 1:2 serial dilutions of erlotinib (0-40 μM) and celecoxib (0-20 μM) as single agents and in combination as described in the Methods section. After incubation for 72 hours, SRB assay was used to determine the percentage of survival relative to the untreated cells. CIs at effective doses which resulted in 50% (ED50), 75% (ED75), and 90% (ED90) inhibitory rates (1 - survival rate) were calculated using CalcuSyn software. A CI value of >1 is antagonism, = 1 is additivity, and

Figure 1. Effects of erlotinib and celecoxib on growth of SCCHN cell lines

SCCHN cell lines Tu212 (A) and Tu686 (B) were treated with 1:2 serial dilutions of erlotinib (0-40 μM) and celecoxib (0-20 μM) as single agents and in combination as described in the Methods section. After incubation for 72 hours, SRB assay was used to determine the percentage of survival relative to the untreated cells. CIs at effective doses which resulted in 50% (ED50), 75% (ED75), and 90% (ED90) inhibitory rates (1 - survival rate) were calculated using CalcuSyn software. A CI value of >1 is antagonism, = 1 is additivity, and

Figure 2. Induction of G0/G1 arrest by erlotinib and celecoxib in SCCHN cells

SCCHN cell lines Tu212 (A) and Tu686 (B) were treated with erlotinib (1 μM), celecoxib (10 μM), and their combination for 24, 48, and 72 hours. Cell cycle analysis was performed by flow cytometry as described in the Methods section. The average percentages of the cell population arrested at G0/G1 at each time point are presented with standard deviation from three repeated experiments. Significant differences (p ≤ 0.05) in comparison of the combination treatment with the control and each of the single treatments at all time points are shown. (C) Western blot analyses of cell cycle regulatory proteins in both Tu212 and Tu686 cells treated for 24, 48, and 72 hours. The experiments were repeated three times.

Figure 2. Induction of G0/G1 arrest by erlotinib and celecoxib in SCCHN cells

SCCHN cell lines Tu212 (A) and Tu686 (B) were treated with erlotinib (1 μM), celecoxib (10 μM), and their combination for 24, 48, and 72 hours. Cell cycle analysis was performed by flow cytometry as described in the Methods section. The average percentages of the cell population arrested at G0/G1 at each time point are presented with standard deviation from three repeated experiments. Significant differences (p ≤ 0.05) in comparison of the combination treatment with the control and each of the single treatments at all time points are shown. (C) Western blot analyses of cell cycle regulatory proteins in both Tu212 and Tu686 cells treated for 24, 48, and 72 hours. The experiments were repeated three times.

Figure 2. Induction of G0/G1 arrest by erlotinib and celecoxib in SCCHN cells

SCCHN cell lines Tu212 (A) and Tu686 (B) were treated with erlotinib (1 μM), celecoxib (10 μM), and their combination for 24, 48, and 72 hours. Cell cycle analysis was performed by flow cytometry as described in the Methods section. The average percentages of the cell population arrested at G0/G1 at each time point are presented with standard deviation from three repeated experiments. Significant differences (p ≤ 0.05) in comparison of the combination treatment with the control and each of the single treatments at all time points are shown. (C) Western blot analyses of cell cycle regulatory proteins in both Tu212 and Tu686 cells treated for 24, 48, and 72 hours. The experiments were repeated three times.

Figure 3. Signal transduction pathways affected by erlotinib and celecoxib in SCCHN cells

Cell lysates were collected from SCCHN cell lines Tu212 and Tu686 after treatment with erlotinib (E: 1 μM), celecoxib (C: 10 μM), and their combination (EC) for 24, 48, and 72 hours. Untreated cells (NT) were used as a control at each time point. Western blot analyses were performed on total protein extracts from each of the time points to detect the expression levels of proteins involved in EGFR, AKT, mTOR, and COX-2 pathways. β-actin served as a loading control.

Figure 4. Inhibitory effects of erlotinib and celecoxib on HUVEC tubular formation and Matrigel invasion

(A) Effects of erlotinib and celecoxib on capillary formation by HUVECs were examined in vitro. HUVECs were pretreated with vehicle DMSO (Control), celecoxib (CCB: 10 μM), erlotinib (ER: 1 μM), and their combination (Comb) for 12 hours, followed by inoculation in 24-well plates pre-coated with Matrigel and incubation overnight. A representative image is shown in the figure. The average number of HUVEC capillary tube branches in 10 fields was counted using an Olympus inverted microscope (CKX40; Olympus, New York, NY). (B) HUVECs suspended in serum-free medium containing 0.1% BSA with or without DMSO (control), celecoxib (CCB: 10 μM), erlotinib (ER: 1 μM), or combination of erlotinib and celecoxib (Comb: 1 μM and10 μM, respectively) in triplicate were seeded in the invasion chamber. After 36-40 hours of incubation, the number of invaded cells was counted and quantified as the sum of 10 random fields under the microscope with 200X magnification. Experiments were repeated twice. * indicates statistical significance (p<0.05) of the treatment compared with control.

Figure 5. Effect of erlotinib and celecoxib on Tu212 xenograft tumor growth

Four groups of mice were orally gavaged with control (0.1% Tween 80 and 0.5% methylcellulose), erlotinib (75 mg/kg), celecoxib (50mg/kg), or the combination (n=8) of erlotinb (75 mg/kg) and celecoxib (50 mg/kg) for 6 days prior to a subcutaneous inoculation of 2 × 106 Tu212 cells. The animals were continuously gavaged with the agents 5 days a week for a total of 4 weeks. (A) Tumor volume was measured at the indicated time points. (B) Immunohistochemistry analyses shown as representative H&E staining (× 200) were performed for expression of proliferation marker Ki-67, mTOR substrate p-S6, and endothelium marker CD34. (C) Quantification of these biomarkers. * indicates statistical significance (p<0.05) of the treatment compared with control and ** indicates significant difference between the combination and either single agent.

Figure 6. Biomarker alterations upon treatment with the combination of erlotinib and celecoxib in patient tissue samples

(A) Immunohistochemical staining of EGFR, pERK, and p-S6 before and after the combined treatment. (B) Correlation of the change in expression of EGFR, pERK, and p-S6 with patients’ responses at the last clinical time point. Seven patients were included in the analysis. (C) Heat map of significant proteins (p

Figure 6. Biomarker alterations upon treatment with the combination of erlotinib and celecoxib in patient tissue samples

(A) Immunohistochemical staining of EGFR, pERK, and p-S6 before and after the combined treatment. (B) Correlation of the change in expression of EGFR, pERK, and p-S6 with patients’ responses at the last clinical time point. Seven patients were included in the analysis. (C) Heat map of significant proteins (p