Purpose: Targeting the epidermal growth factor receptor (EGFR) is a validated approach to treat cancer. In non–small cell lung cancer (NSCLC), EGFR contains somatic mutations in 10% of patients, which correlates with increased response rates to small molecule inhibitors of EGFR. We analyzed the effects of the monoclonal IgG1 antibody Erbitux (cetuximab) in NSCLC xenografts with wild-type (wt) or mutated EGFR.

Experimental Design: NSCLC cell lines were grown s.c. in nude mice. Dose-dependent efficacy was established for cetuximab. To determine whether combination therapy produces tumor regressions, cetuximab was dosed at half-maximal efficacy with chemotherapy used at maximum tolerated dose.

Results: Cetuximab showed antitumor activity in wt (A549, NCI-H358, NCI-H292) and mutated [HCC-827 (delE746-A750), NCI-H1975 (L858R, T790M)] EGFR-expressing xenografts. In the H292 model, cetuximab and docetaxel combination therapy was more potent to inhibit tumor growth than cetuximab or docetaxel alone. Cisplatin augmented efficacy of cetuximab to produce 6 of 10 regressions, whereas 1 of 10 regressions was found with cetuximab and no regression was found with cisplatin. Using H1975 xenografts, gemcitabine increased efficacy of cetuximab resulting in 12 of 12 regressions. Docetaxel with cetuximab was more efficacious with seven of nine regressions compared with single treatments. Cetuximab inhibited autophosphorylation of EGFR in both H292 and H1975 tumor lysates. Exploring the underlying mechanism for combination effects in the H1975 xenograft model, docetaxel in combination with cetuximab added to the antiproliferative effects of cetuximab but was the main component in this drug combination to induce apoptosis.

Conclusions: Cetuximab showed antitumor activity in NSCLC models expressing wt and mutated EGFR. Combination treatments increased the efficacy of cetuximab, which may be important for the management of patients with chemorefractory NSCLC.

Targeted therapy in the treatment of cancer has made significant progress over the last few years. Deregulation of receptor tyrosine kinases through overexpression or activating mutations has frequently been observed in human cancer leading to induction of proliferation, inhibition of cell death, angiogenesis, and migration (13). The ErbB family of receptor tyrosine kinases consists of the epidermal growth factor receptor (EGFR/ErbB1), ErbB2 (HER2/neu), ErbB3, and ErbB4. EGFR and HER2 were implicated in the pathogenesis of various epithelial cancers such as colorectal, head and neck, pancreatic, lung, and breast cancers (46).

Non–small cell lung cancer (NSCLC) is characterized by high incidence and mortality rates and rapid emergence of resistance against commonly used chemotherapeutics. Because EGFR is overexpressed in 65% to 100% of NSCLC cases (7), novel targeted therapeutics were developed that block activation of EGFR (8). Cetuximab is a chimeric IgG1 monoclonal antibody that blocks ligand binding to EGFR, leading to a decrease in receptor dimerization, autophosphorylation, and activation of signaling pathways (9, 10). In preclinical mouse models, cetuximab showed antitumor effects in renal (11), pancreatic (12, 13), colorectal (14), and lung (1517) tumors. Clinical use of cetuximab is approved for late-stage chemorefractory colorectal cancer and for squamous cell carcinoma of the head and neck. Gefitinib and erlotinib are small-molecule inhibitors of EGFR. Both drugs compete for the ATP-binding site in the kinase domain of EGFR and are used against pancreatic cancer and NSCLC (8, 18).

Several somatic mutations in EGFR have been identified in a subset of NSCLC patients (10-15%), which were characterized by adenocarcinoma histology, oriental origin, female gender, and never-smoker status. Importantly, the patients with mutated EGFR responded better to treatment with gefitinib than those without mutations (1921). The clinically identified changes included short amino acid deletions and point mutations in the ATP-binding pocket of the tyrosine kinase domain encoded by exons 18 to 21. These findings suggested that the observed correlation between expression of EGFR mutants and better response rate to gefitinib is due to hypersensitivity of the EGFR mutants to inhibitors of the kinase domain. The tumor cells depend on the continued activity of the EGFR pathway to maintain their malignant phenotype, also called “oncogene addiction” (22). Subsequently, acquired resistance to gefitinib and erlotinib has been studied in patient tumor samples and was attributed to a secondary point mutation (T790M) in the kinase domain of EGFR (23, 24). The genetics of resistance is recapitulated in the human NSCLC cell line NCI-H1975, which contains both the gefitinib-sensitizing L858R and the resistant T790M point mutations (24) and is resistant to gefitinib and erlotinib in vitro (2426).

Demonstration of preclinical efficacy of cetuximab in NSCLC xenograft mouse models is limited to two studies using NCI-H226 (15) and NCI-H292 cells (16). Furthermore, cetuximab inhibited lung tumor growth in transgenic mice overexpressing an activating allele of EGFR (17). To better understand the potential of cetuximab in NSCLC, we tested cetuximab in a panel of NSCLC xenografts expressing wild-type (wt) EGFR and found antitumor effects in A549, NCI-H358, and H292 tumors. In addition, HCC-827 cells, which contain an activating exon 19 deletion mutation in EGFR (delE746-A750) and which were sensitive to cetuximab in vitro (27), were highly sensitive to cetuximab in our in vivo studies. The effects of cetuximab on gefitinib-resistant H1975 cells (L858R and T790M), either in vitro or in vivo, have not been described previously. We show that H1975 xenografts retained dose-dependent sensitivity to cetuximab. Cetuximab was tested in combination with the cytidine analogue gemcitabine (Gemzar), the microtubule-stabilizing agent docetaxel (Taxotere), platinum-based alkylating agents cisplatin (Platinol) and carboplatin (Paraplatin), and the antifolate pemetrexed (Alimta). Strong effects were found in both H292 and H1975 mouse models with cetuximab in combination with gemcitabine or docetaxel. Finally, we found that docetaxel augmented the antiproliferative effects of cetuximab but was mostly responsible for the induction of apoptosis in H1975 tumor extracts.

Cell lines. All NSCLC cell lines were from the American Type Culture Collection (Manassas, VA). The cells were maintained at 37°C in a 5% CO2 incubator in RPMI 1640 (Invitrogen, Carlsbad, CA) containing 10% fetal bovine serum (HyClone, Logan, UT), 2 mmol/L GlutaMAX (Invitrogen), 1 mmol/L sodium pyruvate (Invitrogen), and 10 mmol/L HEPES (Sigma, St. Louis, MO).

Sequencing of EGFR. Exons 18 to 21 of EGFR were sequenced from genomic DNA isolated from H292 cells using methods and sequencing primers described previously (19, 20).

S.c. NSCLC xenograft models. All experiments and procedures involving mice were done according to the U.S. Department of Agriculture, Department of Health and Human Services, and NIH policies regarding the humane care and use of laboratory animals. Eight-week-old female athymic (nu/nu) mice from Charles River Laboratories (Wilmington, MA) were housed under pathogen-free conditions in microisolator cages with laboratory chow and water ad libitum. NSCLC xenografts were established by injecting s.c. into the left flanks 2 × 107 (A549, NCI-H358, HCC-827), 5 × 106 (Calu-6, NCI-H292), 3 × 105 (NCI-H460), or 1 × 107 (NCI-H1975) cells per mouse mixed 1:1 in Matrigel (100% Matrigel for A549; Collaborative Research Biochemicals, Bedford, MA). Tumors were allowed to reach 200 to 400 mm3 for efficacy studies (10-12 mice per group) or 350 to 450 mm3 for mechanism of action studies (6 mice per group) before randomization.

Treatments and tumor measurements. Cetuximab was produced by ImClone System manufacturing facility (Branchburg, NJ). Human IgG antibodies were from Sigma and diluted in 0.9% USP saline. Gemcitabine was from SynChem OHG (Kassel, Germany) and prepared in 0.9% USP saline containing 12.5 mg/mL mannitol and 0.781 mg/mL sodium acetate. Docetaxel was from Fluka (Buchs, Switzerland) and prepared in 0.9% USP saline containing 4% ethanol and 10% polysorbate 80. Carboplatin and cisplatin were from Sigma and prepared in 0.9% USP saline. Pemetrexed was purchased from Briarwood Pharmacy (Jamaica, NY) and dissolved in 0.9% USP saline. Cetuximab followed by chemotherapeutics was administered i.p. at 20 μL per gram of body weight. Caliper measurements were used to calculate tumor volumes using the formula V = π/6 (length × width × width). Antitumor effects are expressed as %T/C (treated versus control), dividing the tumor volumes from treatment groups with the control groups and multiplied by 100. For mechanism studies, docetaxel was injected up to 12 h before cetuximab.

Statistical analysis. Tumor volumes were analyzed using repeated measures ANOVA in the JMP Statistical Discovery package (v.5.1; SAS Institute Inc., Cary, NC). Frequencies of tumor regressions across four treatment groups were analyzed by χ2 test. Statistical significance between band intensities from immunoblots and absorbance units from histology and ELISA analysis was determined by Student t test (one-tailed distribution). *P = 0.05 to 0.01 (significant), **P = 0.01 to 0.001 (very significant), and ***P < 0.001 (extremely significant).

Antibodies for Western blotting. Antibodies against phospho-EGFR Tyr1068 and Tyr845 and phosphorylated mitogen-activated protein kinase (phospho-MAPK) were from Cell Signaling Technology (Beverly, MA). Antibodies against phospho-EGFR Tyr1086 and Tyr1173 were from BioSource International (Camarillo, CA). Anti-EGFR antibody was from Upstate (Lake Placid, NY). Anti–β-actin antibody was from Sigma. Anti-mouse IgG peroxidase-linked whole antibody and anti-rabbit IgG peroxidase-linked species-specific whole antibody were from Amersham (Piscataway, NJ).

Tumor harvesting and Western blotting. Xenograft tumors were excised from euthanized mice and snap frozen in liquid nitrogen. Lysates were prepared in lysis buffer (50 mmol/L Tris-HCl, 150 mmol/L NaCl, 1% Triton X-100, 100 mmol/L sodium orthovanadate, 1 mmol/L EDTA, Complete Protease Inhibitor Cocktail Tablets; Roche Applied Science, Indianapolis, IN) using a dounce homogenizer. Protein concentration was determined using BCA Protein Assay Reagent (Pierce, Rockford, IL). Lysates were run at 50 μg per lane on 4% to 12% Bis-Tris gels (Invitrogen) and transferred to nitrocellulose membranes (Invitrogen). Western blot band intensity quantification was done using MultiGauge software v2.3 (FujiFilm Medical Systems, Stanford, CT). To account for differences in protein loading, all band intensities were corrected for β-actin.

Histologic analysis of phospho-EGFR and apoptosis markers. Tumors were fixed in 10% neutral-buffered formalin, incubated in 30% sucrose, and embedded in paraffin for sectioning and immunohistochemical staining of phospho–Tyr1086 of human EGFR (Invitrogen) and cleaved caspase-3 (Cell Signaling). DNA fragmentation (apoptosis) was detected with ApopTag Plus peroxidase in situ apoptosis kit (Chemicon, Temecula, CA). Quantification of phospho-EGFR, cleaved caspase-3, and ApopTag was done with a Zeiss Axiocam mounted on a Zeiss universal microscope (Carl Zeiss, Thornwood, NY). Morphometric analysis of digital images was done with Image-Pro Plus software (Media Cybernetics, Silver Spring, MD) to determine the average mean density of phospho-EGFR staining or the percentage of cells positive for colored reaction product (ApopTag or caspase-3) in 100 randomly selected tumor cells per section using five sections per tumor.

Phospho–histone H3 and cleaved caspase-3 assays. Lysates were prepared in the same buffer used for Western blotting. Fifty micrograms of extracts were analyzed in the PathScan phospho–histone H3 (Ser10) sandwich ELISA (Cell Signaling). Two hundred micrograms were used for the PathScan cleaved caspase-3 sandwich ELISA (Cell Signaling) following manufacturer's instructions. In brief, extracts were mixed with 100 μL of sample diluent and incubated in antibody-coated microwell strips. One hundred microliters of biotinylated phospho–histone H3 or cleaved caspase-3 detection antibodies were added to each well. Binding was detected with 100 μL of horseradish peroxidase–linked streptavidin antibody and 100 μL of TMB substrate solution. The colored reaction product was measured in a standard ELISA plate reader at 450 nm.

Effects of cetuximab on growth of human NSCLC xenografts. Seven human NSCLC cell lines were established as s.c. xenograft tumor models in nude mice. Cetuximab at 4 and 40 mg/kg administered thrice per week showed strong antitumor efficacy in vivo using the adenocarcinoma cell lines A549 and H358 (Fig. 1A and B). NCI-H460 (large cell) and Calu-6 (anaplastic carcinoma) xenografts did not respond to cetuximab (Fig. 1C and data not shown). Tumor growth of the squamous H292 cell line, which was susceptible to cetuximab in vivo (16), was inhibited in a dose-dependent way (twice per week), allowing us to define an IC50 for cetuximab of ∼0.4 mg/kg (Fig. 1D). All five cell lines express wt EGFR (28), which we confirmed for H292 cells by sequencing exons 18 to 21 from genomic DNA (data not shown). EGFR protein was readily detectable by immunoblotting in A549 and H292 cells, whereas H460 and Calu-6 cells expressed moderate levels of EGFR. Surprisingly, the cetuximab-sensitive H358 line expressed very low levels of EGFR (data not shown). Small deletions and point mutations in exons 19 to 21 of EGFR were reported in human NSCLC tumor samples, which correlated with clinical benefit to treatment with small-molecule inhibitors of EGFR (1921). The adenocarcinoma cell line HCC-827 harbors such a deletion in EGFR (delE746-A750). Dosing of HCC-827 xenografts with 4 or 40 mg/kg of cetuximab (twice per week) completely inhibited tumor growth. A dose-response was established by lowering cetuximab to 0.4 mg/kg (Fig. 1E). Last, the gefitinib-resistant adenocarcinoma cell line H1975 (L858R and T790M) was established as a xenograft mouse model. Treatment of H1975 tumors with cetuximab (twice per week) produced dose-dependent growth inhibition. Cetuximab at 0.4 mg/kg was identified as an approximate IC50 in the H1975 xenograft model because of its half-maximal antitumor effects. Higher doses completely inhibited tumor growth for 2 weeks followed by moderate growth inhibition (Fig. 1F).

Fig. 1.

Effects of cetuximab on NSCLC xenografts. A, A549 cells were injected s.c. into nude mice with randomization (n = 10) once tumors reached >250 mm3. Human IgG (40 mg/kg, i.p.) and cetuximab (4 and 40 mg/kg, i.p.) were administered thrice per week. Bars, SEM repeated measures ANOVA indicated statistically significant effects (P < 0.005) for both cetuximab groups. B, same with NCI-H358, thrice per week, P < 0.001 for both cetuximab groups. C, same with NCI-H460, cetuximab (6 and 60 mg/kg), twice per week. D, same with NCI-H292, cetuximab (0.4, 4, and 40 mg/kg), twice per week, P < 0.1 for 0.4 mg/kg, otherwise P < 0.001. Human IgG and 0.4 mg/kg cetuximab groups were euthanized after day 22 due to weight loss. E, same with HCC-827 harboring mutated EGFR (delE746-A750), twice per week, P < 0.05 for 0.4 mg/kg, otherwise P < 0.001. F, same with NCI-H1975 harboring mutated EGFR (L858R and T790M), twice per week, P < 0.05 for 0.4 mg/kg, otherwise P < 0.001. The human IgG group was terminated after day 13 due to ulceration of tumors.

Fig. 1.

Effects of cetuximab on NSCLC xenografts. A, A549 cells were injected s.c. into nude mice with randomization (n = 10) once tumors reached >250 mm3. Human IgG (40 mg/kg, i.p.) and cetuximab (4 and 40 mg/kg, i.p.) were administered thrice per week. Bars, SEM repeated measures ANOVA indicated statistically significant effects (P < 0.005) for both cetuximab groups. B, same with NCI-H358, thrice per week, P < 0.001 for both cetuximab groups. C, same with NCI-H460, cetuximab (6 and 60 mg/kg), twice per week. D, same with NCI-H292, cetuximab (0.4, 4, and 40 mg/kg), twice per week, P < 0.1 for 0.4 mg/kg, otherwise P < 0.001. Human IgG and 0.4 mg/kg cetuximab groups were euthanized after day 22 due to weight loss. E, same with HCC-827 harboring mutated EGFR (delE746-A750), twice per week, P < 0.05 for 0.4 mg/kg, otherwise P < 0.001. F, same with NCI-H1975 harboring mutated EGFR (L858R and T790M), twice per week, P < 0.05 for 0.4 mg/kg, otherwise P < 0.001. The human IgG group was terminated after day 13 due to ulceration of tumors.

Close modal

Chemotherapy increases antitumor efficacy of cetuximab in NSCLC xenografts. H292 (wt EGFR) xenografts were dosed with 1.5 mg/kg cetuximab (twice per week) to achieve half-maximal antitumor effects. For combination treatments, five chemotherapy drugs were used at their maximum tolerated dose and injected according to a clinically relevant dosing schedule of once per week. Cetuximab monotherapy caused tumor growth delay for ∼17 days with a tumor inhibition rate of 65% T/C after 42 days of treatment (Fig. 2A and E). Gemcitabine at 125 mg/kg (once per week) had no effect on tumor growth but when combined with cetuximab a strong inhibition of tumor growth was observed up to day 42 with 22% T/C (Fig. 2A). Statistical analysis by repeated measures ANOVA revealed that the combination protocol added a significant benefit over monotherapy treatments (Fig. 2E). Furthermore, the combination treatment led to tumor regressions in 4 of 11 tumors, whereas cetuximab caused 1 of 10 regression. No regression was observed in either the saline or the gemcitabine group (Fig. 2E). The statistical significance of the regression frequencies across all four treatment groups was analyzed by a χ2 test and found to be significant for the combination group (Fig. 2E). The combination group was followed in the absence of treatment after day 42 to determine whether the tumor reductions constituted tumor cures or simply tumor growth inhibition. The size of the tumors increased without treatment, indicating relapse of tumor growth (Fig. 2A).

Fig. 2.

Combination effects of cetuximab with chemotherapy in NCI-H292 xenografts. A, NCI-H292 cells were injected s.c. into nude mice with randomization (n = 12) once tumors reached 200 mm3. Treatment with saline, cetuximab (1.5 mg/kg, i.p., twice per week), gemcitabine (125 mg/kg, i.p., once per week), or combination lasted for 42 d. Horizontal bar, treatment stopped. Bars, SEM. B, cetuximab with docetaxel (15 mg/kg, i.p., once per week). C, cetuximab with cisplatin (5 mg/kg, i.p., once per week). D, cetuximab with carboplatin (25 mg/kg, i.p., once per week). Treatment stopped, not done. E, summary table for (A-D) listing effects on tumor volume (%T/C) and incidence of tumor regressions (number of regressions / number of mice per group). Statistical analysis done at termination of the saline group: ns, not significant; *, P = 0.05 to 0.01; **, P = 0.01 to 0.001; ***, P < 0.001. %T/C was analyzed by repeated measures ANOVA: 1, versus saline; 2, versus cetuximab; 3, versus gemcitabine; 4, versus docetaxel; 5, versus cisplatin; 6, versus carboplatin. Numbers of regressions between groups were analyzed by χ2 test: 7, versus saline and both monotherapies.

Fig. 2.

Combination effects of cetuximab with chemotherapy in NCI-H292 xenografts. A, NCI-H292 cells were injected s.c. into nude mice with randomization (n = 12) once tumors reached 200 mm3. Treatment with saline, cetuximab (1.5 mg/kg, i.p., twice per week), gemcitabine (125 mg/kg, i.p., once per week), or combination lasted for 42 d. Horizontal bar, treatment stopped. Bars, SEM. B, cetuximab with docetaxel (15 mg/kg, i.p., once per week). C, cetuximab with cisplatin (5 mg/kg, i.p., once per week). D, cetuximab with carboplatin (25 mg/kg, i.p., once per week). Treatment stopped, not done. E, summary table for (A-D) listing effects on tumor volume (%T/C) and incidence of tumor regressions (number of regressions / number of mice per group). Statistical analysis done at termination of the saline group: ns, not significant; *, P = 0.05 to 0.01; **, P = 0.01 to 0.001; ***, P < 0.001. %T/C was analyzed by repeated measures ANOVA: 1, versus saline; 2, versus cetuximab; 3, versus gemcitabine; 4, versus docetaxel; 5, versus cisplatin; 6, versus carboplatin. Numbers of regressions between groups were analyzed by χ2 test: 7, versus saline and both monotherapies.

Close modal

To verify that the increase in efficacy is not limited to the cetuximab and gemcitabine combination, cetuximab was tested with docetaxel, cisplatin, carboplatin, and pemetrexed. Docetaxel at 15 mg/kg reached an antitumor effect of 34% T/C with 1 of 10 regression (Fig. 2B). The combination with cetuximab and docetaxel was superior over monotherapy treatments resulting in an 11% T/C. The incidence of regressions in the combination group was high with 9 of 10 regressions (Fig. 2E). Combination treatment with cetuximab and cisplatin (5 mg/kg) was efficacious at 13% T/C with 6 of 10 regressions. Cisplatin reached 42% T/C with no regressions (Fig. 2C and E). The combination of cetuximab and carboplatin (25 mg/kg) inhibited tumor growth at 19% T/C with 4 of 10 regressions (Fig. 2D and E). This was unexpected because carboplatin did not affect tumor growth and cetuximab only caused tumor growth delay. Finally, pemetrexed at 150 mg/kg (twice per week) marginally enhanced the effects of cetuximab and was only modestly efficacious as monotherapy (data not shown). Weight loss occurred in all mice due to H292 tumor burden, which was increased slightly by carboplatin and severely by cisplatin (data not shown).

Chemotherapy increases efficacy of cetuximab in NSCLC xenografts with mutations in EGFR. To extend our findings on antitumor efficacy with cetuximab and chemotherapy in a wt EGFR tumor model, we tested the same combinations in the H1975 xenograft model (L858R and T790M double mutation). Cetuximab was dosed at a concentration close to the IC50, 1 mg/kg (twice per week), in combination with chemotherapies used at their maximum tolerated dose (once per week). Combination effects of cetuximab with gemcitabine were evaluated after 22 days when the saline group was terminated due to ulceration of tumors. The combination treatment showed a lasting antitumor effect of 7% T/C (Fig. 3A and E). Cetuximab had a half-maximal effect of 49% T/C, whereas gemcitabine at 125 mg/kg was slightly more effective with 26% T/C. Statistical analysis by repeated measures ANOVA documented the benefit of combination over monotherapies (Fig. 3E). Furthermore, cetuximab with gemcitabine produced 12 of 12 regressions, whereas fewer regressions were observed in the monotherapy treatments and no regression in the controls. Frequencies of regressions were analyzed by χ2 test and found to be highly significant (Fig. 3E). No toxicity was associated with these treatments based on the absence of body weight loss. Treatment was stopped after day 33 to assess cure rate versus growth inhibition. All 12 tumors in the cetuximab and gemcitabine combination group resumed growth by day 60, indicating relapse of tumor growth (data not shown).

Fig. 3.

Combination effects of cetuximab with chemotherapy in NCI-H1975 xenografts. A, NCI-H1975 cells with mutated EGFR (L858R and T790M) were injected s.c. into nude mice with randomization (n = 12) once tumors reached 250 mm3. Treatment with saline, cetuximab (1 mg/kg, i.p., twice per week), gemcitabine (125 mg/kg, i.p., once per week), or combination lasted for 33 d. The saline group was terminated after 22 d due to tumor size. Bars, SEM. B, cetuximab with docetaxel (20 mg/kg, i.p., once per week). C, cetuximab with cisplatin (5 mg/kg, i.p., once per week). D, cetuximab with carboplatin (25 mg/kg, i.p., once per week). The saline and carboplatin groups were terminated at day 22 due to tumor size and ulcerations. E, summary table for (A-D) listing effects on tumor volume (%T/C) and incidence of tumor regressions (number of regressions / number of mice per group). Statistical analysis done at termination of the saline group: *, P = 0.05 to 0.01; **, P = 0.01 to 0.001; ***, P < 0.001. %T/C was analyzed by repeated measures ANOVA: 1, versus saline; 2, versus cetuximab; 3, versus gemcitabine; 4, versus docetaxel. Numbers of regressions between groups were analyzed by χ2 test: 7, versus saline and both monotherapies.

Fig. 3.

Combination effects of cetuximab with chemotherapy in NCI-H1975 xenografts. A, NCI-H1975 cells with mutated EGFR (L858R and T790M) were injected s.c. into nude mice with randomization (n = 12) once tumors reached 250 mm3. Treatment with saline, cetuximab (1 mg/kg, i.p., twice per week), gemcitabine (125 mg/kg, i.p., once per week), or combination lasted for 33 d. The saline group was terminated after 22 d due to tumor size. Bars, SEM. B, cetuximab with docetaxel (20 mg/kg, i.p., once per week). C, cetuximab with cisplatin (5 mg/kg, i.p., once per week). D, cetuximab with carboplatin (25 mg/kg, i.p., once per week). The saline and carboplatin groups were terminated at day 22 due to tumor size and ulcerations. E, summary table for (A-D) listing effects on tumor volume (%T/C) and incidence of tumor regressions (number of regressions / number of mice per group). Statistical analysis done at termination of the saline group: *, P = 0.05 to 0.01; **, P = 0.01 to 0.001; ***, P < 0.001. %T/C was analyzed by repeated measures ANOVA: 1, versus saline; 2, versus cetuximab; 3, versus gemcitabine; 4, versus docetaxel. Numbers of regressions between groups were analyzed by χ2 test: 7, versus saline and both monotherapies.

Close modal

Cetuximab in combination with docetaxel (20 mg/kg) resulted in an antitumor effect of 16% T/C. Cetuximab reached 46% T/C, whereas docetaxel was efficacious during the first half of the study but only reached 60% T/C at day 33 (Fig. 3B and E). The benefit of the combination was supported by the frequencies of tumor regressions with seven of nine regressions in the cetuximab and docetaxel combination group (Fig. 3E). Toxicity was minimal based on absence of weight loss. However, a tumor relapse study for the cetuximab and docetaxel combination group after day 33 could not be done because mice were euthanized with obstructed bowels. Cetuximab with cisplatin at 5 mg/kg (Fig. 3C) or with carboplatin at 25 mg/kg (Fig. 3D) inhibited tumor growth at 48% T/C for both combinations, and tumor regressions were observed at low frequency (Fig. 3E). A similar effect could be obtained in both experiments with cetuximab alone. Cisplatin and carboplatin were only weakly efficacious, although both drugs were used at their maximum tolerated dose (Fig. 3E). This indicated that platinum-based agents did not increase the benefits of cetuximab in H1975 xenografts. Lastly, pemetrexed at 100 mg/kg (twice per week) was less efficacious than cetuximab and the combination did not augment the effect of cetuximab (data not shown).

Cetuximab inhibits phosphorylation of EGFR in NSCLC xenografts with wt and mutated EGFR. The mechanism of action of cetuximab has been studied in pancreatic (12), epidermoid (29), and colon (30) xenograft tumor models and involved inhibition of phosphorylation of EGFR, MAPK, and AKT. To elucidate how cetuximab inhibited H292 tumor growth, we determined the change in phosphorylation of four EGFR tyrosine residues in xenograft extracts. Quantitative immunoblot analysis showed that cetuximab at 40 mg/kg reduced phosphorylation of Tyr845 and Tyr1173 by 57% and 34% after 3 days of treatment (Fig. 4A and C). Both effects reached statistical significance by Student t test. Phosphorylation of Tyr1068 and Tyr1086 was reduced by a smaller degree. Furthermore, we found equal levels of total EGFR in saline- and cetuximab-treated tumor samples (Fig. 4E), indicating that cetuximab inhibited H292 tumor growth by reducing EGFR kinase activity and not through degradation of EGFR.

Fig. 4.

Reduction in phospho-EGFR with cetuximab in NSCLC xenografts. A, s.c. tumors were established with NCI-H292 cells in nude mice and treated with saline or cetuximab (40 mg/kg, i.p.). After 3 d, tumors were collected and phosphorylated tyrosine residues of EGFR were visualized by immunoblotting. Columns, average phosphorylation intensities corrected for β-actin (loading control) using five or six xenograft tumors per treatment group; bars, SEM. *, P = 0.05 to 0.01; **, P = 0.01 to 0.001. B, same analysis as in (A) with NCI-H1975 xenograft tumors. ***, P < 0.001. C, representative immunoblots from two or three NCI-H292 xenografts per treatment probed for the indicated phospho-EGFR sites used for quantitative analysis. D, same as in (C) with three NCI-H1975 xenografts per treatment. E, total EGFR was visualized by immunoblotting after 3 d of treatment with saline or cetuximab (40 mg/kg, i.p.). Columns, average intensities corrected for β-actin (loading control) using six xenograft tumors per treatment group.

Fig. 4.

Reduction in phospho-EGFR with cetuximab in NSCLC xenografts. A, s.c. tumors were established with NCI-H292 cells in nude mice and treated with saline or cetuximab (40 mg/kg, i.p.). After 3 d, tumors were collected and phosphorylated tyrosine residues of EGFR were visualized by immunoblotting. Columns, average phosphorylation intensities corrected for β-actin (loading control) using five or six xenograft tumors per treatment group; bars, SEM. *, P = 0.05 to 0.01; **, P = 0.01 to 0.001. B, same analysis as in (A) with NCI-H1975 xenograft tumors. ***, P < 0.001. C, representative immunoblots from two or three NCI-H292 xenografts per treatment probed for the indicated phospho-EGFR sites used for quantitative analysis. D, same as in (C) with three NCI-H1975 xenografts per treatment. E, total EGFR was visualized by immunoblotting after 3 d of treatment with saline or cetuximab (40 mg/kg, i.p.). Columns, average intensities corrected for β-actin (loading control) using six xenograft tumors per treatment group.

Close modal

No previous report addressed the question whether cetuximab inhibits kinase activity of mutated EGFR in vivo using xenograft models. We treated H1975 tumors with saline or 40 mg/kg of cetuximab for 3 days followed by immunoblotting with antibodies specific for phosphorylated residues of EGFR. Cetuximab caused a decrease in phosphorylation of Tyr1068, Tyr1086, and Tyr1173 between 31% and 54% (Fig. 4B and D). Phosphorylation of EGFR on Tyr845, which is phosphorylated by the cytoplasmic dual-specificity protein kinase Src, was reduced by cetuximab, implying an indirect mechanism involving signaling components regulated by EGFR. All effects were statistically significant by Student's t test. In contrast to H292, total EGFR was reduced by 30% in the H1975 xenograft model after 3 days of treatment with cetuximab (P = 0.07; Fig. 4E). This finding indicates that the mechanism for cetuximab-mediated inhibition of H1975 tumor growth is through both inhibition of kinase activity and reduction of mutated EGFR protein levels.

Docetaxel enhances inhibitory effects of cetuximab on EGFR and MAPK pathways in NSCLC xenografts. Our efficacy studies showed that docetaxel enhanced the antitumor effects of cetuximab in H292 and H1975 xenograft models. We wondered whether the combination of cetuximab and docetaxel would lead to a stronger decrease in phospho-EGFR compared with cetuximab alone. H1975 xenografts were chosen because phosphorylation of all four tyrosine residues was reduced by cetuximab (see Fig. 4B). Specifically, changes in Tyr1086 were analyzed because this site regulates ras/MAPK and phosphatidylinositol 3-kinase/AKT signaling pathways. Immunohistochemical analysis of tumor sections from saline-treated H1975 xenografts revealed strong Tyr1086 staining localized to the cell membranes of tumor cells (Fig. 5A). Cetuximab at 10 mg/kg caused a reduction in membrane staining after 3 days, whereas intracellular staining for Tyr1086 increased (Fig. 5A). This supports the notion that cetuximab leads to internalization followed by inactivation of mutated EGFR in H1975 tumors. Docetaxel at 30 mg/kg had no effect but enhanced the activity of cetuximab in the combination leading to elimination of cell membrane and cytoplasmic staining of phospho-EGFR Tyr1086 (Fig. 5A). These observations were confirmed by quantitative morphometric measurements of the mean phosphorylation intensity of Tyr1086 in tumor sections (Fig. 5B). Statistical analysis by Student's t test showed a significant effect of cetuximab versus saline, whereas the combination treatment was significantly different from saline and both monotherapies.

Fig. 5.

Reduction in phospho-EGFR, phospho-MAPK, and phospho-histone with cetuximab and docetaxel in NCI-H1975 xenografts. A, s.c. tumors were established with NCI-H1975 cells in nude mice and treated with saline, cetuximab (10 mg/kg, i.p.), docetaxel (30 mg/kg, i.p.), or combination. After 3 d, tumors were embedded in paraffin for immunohistochemical staining of phospho-Tyr1086 of human EGFR. Representative sections from the tumor periphery at ×400 magnification. Size bar, 25 μm. B, quantitation of human Tyr1086 staining shown in (A). Morphometric measurement of average mean density (absorbance units) of colored reaction product in 100 randomly selected tumor cells per section using five sections per tumor. Columns, average phosphorylation intensities using six tumors per treatment group; bars, SEM. *, P = 0.05 (versus saline); ***, P < 0.001 (versus saline, cetuximab, or docetaxel). C, tumors were collected 24 h after treatment with saline, cetuximab (1 mg/kg, i.p.), docetaxel (20 mg/kg, i.p.), or combination and immunoblotted for phospho-Tyr1086. Columns, average phosphorylation intensities corrected for β-actin (loading control) using five or six tumors per treatment group; bars, SEM. *, P = 0.05 to 0.01 (versus saline). D, same tumors as in (C) were immunoblotted for phospho-MAPK with signals corrected for β-actin. **, P = 0.01 to 0.001 (versus saline); *, P = 0.05 to 0.01 (versus cetuximab or docetaxel). E, representative immunoblots from two NCI-H1975 xenografts per treatment group probed for phospho-Tyr1086, EGFR, phospho-MAPK, and β-actin used for quantitative analysis. F, tumors were collected 3 d after treatment with saline, cetuximab (10 mg/kg, i.p.), docetaxel (30 mg/kg, i.p.), or combination. Phospho–histone H3 was measured by ELISA. Columns, average absorbance units (450 nm) using five or six tumors per treatment group; bars, SEM. *P = 0.05 to 0.01 (versus saline or cetuximab).

Fig. 5.

Reduction in phospho-EGFR, phospho-MAPK, and phospho-histone with cetuximab and docetaxel in NCI-H1975 xenografts. A, s.c. tumors were established with NCI-H1975 cells in nude mice and treated with saline, cetuximab (10 mg/kg, i.p.), docetaxel (30 mg/kg, i.p.), or combination. After 3 d, tumors were embedded in paraffin for immunohistochemical staining of phospho-Tyr1086 of human EGFR. Representative sections from the tumor periphery at ×400 magnification. Size bar, 25 μm. B, quantitation of human Tyr1086 staining shown in (A). Morphometric measurement of average mean density (absorbance units) of colored reaction product in 100 randomly selected tumor cells per section using five sections per tumor. Columns, average phosphorylation intensities using six tumors per treatment group; bars, SEM. *, P = 0.05 (versus saline); ***, P < 0.001 (versus saline, cetuximab, or docetaxel). C, tumors were collected 24 h after treatment with saline, cetuximab (1 mg/kg, i.p.), docetaxel (20 mg/kg, i.p.), or combination and immunoblotted for phospho-Tyr1086. Columns, average phosphorylation intensities corrected for β-actin (loading control) using five or six tumors per treatment group; bars, SEM. *, P = 0.05 to 0.01 (versus saline). D, same tumors as in (C) were immunoblotted for phospho-MAPK with signals corrected for β-actin. **, P = 0.01 to 0.001 (versus saline); *, P = 0.05 to 0.01 (versus cetuximab or docetaxel). E, representative immunoblots from two NCI-H1975 xenografts per treatment group probed for phospho-Tyr1086, EGFR, phospho-MAPK, and β-actin used for quantitative analysis. F, tumors were collected 3 d after treatment with saline, cetuximab (10 mg/kg, i.p.), docetaxel (30 mg/kg, i.p.), or combination. Phospho–histone H3 was measured by ELISA. Columns, average absorbance units (450 nm) using five or six tumors per treatment group; bars, SEM. *P = 0.05 to 0.01 (versus saline or cetuximab).

Close modal

Tumor lysates from H1975 xenografts treated for 24 h with 1 mg/kg of cetuximab, 20 mg/kg of docetaxel, and combination were subjected to immunoblotting of phospho-EGFR Tyr1086 and phospho-MAPK. Cetuximab caused a 35% decrease in phospho-EGFR, whereas docetaxel led to a 23% reduction (Fig. 5C and E). The combination lowered phosphorylation on Tyr1086 by 47%. This effect was statistically significant compared with saline but did not reach significance versus monotherapies. Cetuximab lowered phospho-MAPK by 23%, whereas docetaxel failed to have an effect (Fig. 5D and E). The combination reduced phospho-MAPK by 43%, which was statistically significant compared with saline and both monotherapies. The signals for phospho-AKT from immunoblots or ELISA were below detection levels in H1975 tumor extracts (data not shown). Lastly, a strong decrease in the proliferation marker phospho–histone H3 was noticed by ELISA using xenografts treated for 3 days with the combination of cetuximab and docetaxel (Fig. 5F). Altogether, these data emphasize the point that cetuximab interfered with EGFR/MAPK signaling in H1975 tumors and that this effect was augmented by docetaxel, leading to an overall decrease in tumor proliferation.

Cetuximab does not increase the effects of docetaxel on apoptosis in NSCLC xenografts. Staining of H1975 tumor sections treated with cetuximab and docetaxel showed degenerative tumor cells with shrunken and dissociated nuclei (see Fig. 5A). Such pyknotic nuclei contain condensed chromatin characteristic of programmed cell death. We asked whether the combination of cetuximab and docetaxel was more potent to induce apoptosis than monotherapy alone. Immunohistochemical staining of tumor sections with an ApopTag apoptosis kit revealed a low incidence of cells with fragmented DNA in saline-treated H1975 xenografts. Morphometric measurement of the percentage of cells positive for ApopTag showed that cetuximab at 10 mg/kg produced a 2.2-fold increase in apoptosis after 3 days of treatment, whereas 30 mg/kg docetaxel increased cell death by 3.4-fold (Fig. 6A and B). The combination of cetuximab and docetaxel was only slightly more potent than docetaxel without reaching statistical significance (Fig. 6A). This indicated that induction of apoptosis in xenograft tumors treated with the combination is mediated by the chemotherapy without any additional benefits from cetuximab.

Fig. 6.

Induction of apoptosis markers with cetuximab and docetaxel in NCI-H1975 xenografts. A, s.c. tumors were established with NCI-H1975 cells in nude mice and treated with saline, cetuximab (10 mg/kg, i.p.), docetaxel (30 mg/kg, i.p.), or combination. After 3 d, tumors were embedded in paraffin for immunohistochemical staining of fragmented DNA (ApopTag). Morphometric measurement of the percentage of cells with colored reaction product (ApopTag) in the tumor periphery in 100 randomly selected tumor cells per section using five sections per tumor. Columns, average percentage of ApopTag-positive cells using six tumors per treatment group; bars, SEM. ***, P < 0.001 (versus saline; 1, versus cetuximab). B, representative sections stained for ApopTag used for quantitative analysis. Magnification, ×400. Size bar, 50 μm. C, same analysis as in (A) with immunohistochemical staining of cleaved (activated) caspase-3. **, P = 0.01 to 0.001 (versus saline); ***, P < 0.001 (versus saline); *, P = 0.05 to 0.01 (versus cetuximab). D, representative sections stained for cleaved caspase-3 used for quantitative analysis. Magnification, ×400. Size bar, 50 μm. E, same as in (A), measuring cleaved (activated) caspase-3 by ELISA. Columns, average absorbance units (450 nm) using five or six tumors per treatment group; bars, SEM. **, P = 0.01 to 0.001 (versus saline); *, P = 0.05 to 0.01 (versus cetuximab).

Fig. 6.

Induction of apoptosis markers with cetuximab and docetaxel in NCI-H1975 xenografts. A, s.c. tumors were established with NCI-H1975 cells in nude mice and treated with saline, cetuximab (10 mg/kg, i.p.), docetaxel (30 mg/kg, i.p.), or combination. After 3 d, tumors were embedded in paraffin for immunohistochemical staining of fragmented DNA (ApopTag). Morphometric measurement of the percentage of cells with colored reaction product (ApopTag) in the tumor periphery in 100 randomly selected tumor cells per section using five sections per tumor. Columns, average percentage of ApopTag-positive cells using six tumors per treatment group; bars, SEM. ***, P < 0.001 (versus saline; 1, versus cetuximab). B, representative sections stained for ApopTag used for quantitative analysis. Magnification, ×400. Size bar, 50 μm. C, same analysis as in (A) with immunohistochemical staining of cleaved (activated) caspase-3. **, P = 0.01 to 0.001 (versus saline); ***, P < 0.001 (versus saline); *, P = 0.05 to 0.01 (versus cetuximab). D, representative sections stained for cleaved caspase-3 used for quantitative analysis. Magnification, ×400. Size bar, 50 μm. E, same as in (A), measuring cleaved (activated) caspase-3 by ELISA. Columns, average absorbance units (450 nm) using five or six tumors per treatment group; bars, SEM. **, P = 0.01 to 0.001 (versus saline); *, P = 0.05 to 0.01 (versus cetuximab).

Close modal

These findings were supported by a second marker of apoptosis, cleaved caspase-3, using immunohistochemical staining of tumor sections, and ELISA analysis of tumor extracts. Morphometric quantitation of cleaved caspase-3 in H1975 sections showed an increase in the percentage of positive cells with cetuximab and docetaxel monotherapies between 2.1- and 3.5-fold (Fig. 6C and D). As observed previously, the combination displayed the same potency to induce apoptosis as docetaxel. Lastly, an ELISA capturing cleaved caspase-3 in H1975 tumor extracts confirmed equal activity between the cetuximab and docetaxel combination and docetaxel monotherapy (Fig. 6E). In summary, the data indicated that docetaxel is the main driver for induction of apoptosis in H1975 xenografts.

Although the EGFR antibody cetuximab is approved for clinical use in the treatment of colorectal and head and neck cancer, its potential benefit in NSCLC is still emerging (8). Cetuximab is evaluated as first- and second-line therapy in NSCLC patients in combination with standard chemotherapy. To support this effort, we are demonstrating preclinical antitumor efficacy of cetuximab using NSCLC cell lines with either wt or mutated EGFR. Cetuximab showed antitumor effects in A549, H358, and H292 xenografts grown in nude mice but failed to inhibit tumor growth in H460 and Calu-6 tumor models (all expressed wt EGFR). A dose-response study allowed the identification of an IC50 concentration for cetuximab in our H292 xenografts of ∼0.4 mg/kg. This compares favorably with small-molecule inhibitors of EGFR, such as gefitinib or erlotinib, which have published IC50 concentrations between 8 and 100 mg/kg depending on the wt-EGFR–expressing tumor model (15, 28, 31, 32). Our results complement previous reports describing activity of cetuximab at 1 mg/mouse (∼40 mg/kg) in H292 xenografts (16) and half-maximal efficacy at 0.2 mg/mouse (∼8 mg/kg) in a H226 tumor model (15). Our work supports the notion that EGFR expression levels in NSCLC do not predict responsiveness of tumors to cetuximab. H358 cells displayed low levels of EGFR but responded well to cetuximab. Conversely, H460 and Calu-6 cells expressed EGFR at moderate levels but were resistant to cetuximab. It was reported that signaling via ErbB2 and ErbB3 associated with resistance to gefitinib but not with resistance to cetuximab in head and neck squamous cell carcinoma cells (33). We analyzed ErbB2 expression levels by immunoblotting in NSCLC cell lines and found the highest levels of ErbB2 in the cetuximab-sensitive lines H292, HCC-827, and H1975 (data not shown). It should be noted that the presence of ErbB2 or ErbB3 is not expected to influence the responsiveness of cells to cetuximab because Erb family members require heterodimerization with EGFR for activation (6). Because cetuximab blocks ligand-mediated heterodimerization of EGFR with ErbB2 or ErbB3, expression of these family members would not be expected to result in decreased sensitivity. It is possible that mutations in signaling molecules downstream of EGFR such as ras, raf, or PTEN might modulate efficacy of cetuximab. The present study does not answer the contributions of ras because mutations in ras were present in A549 and NCI-H358 cell lines but absent in NCI-H292 cells. Nevertheless, an antitumor response to cetuximab was observed in all three lines grown as xenograft tumors.

Although NSCLC patients have few treatment options and the response rate is generally low, a subset of patients with activating mutations in EGFR can achieve remarkable responses with gefitinib and erlotinib. Clinical benefits can last 2 to 3 years until patients develop resistance. The molecular determinants in responding patients are small exon 19 deletions and exon 21 point mutations in the ATP-binding pocket of EGFR. The NSCLC cell line HCC-827 contains an activating exon 19 deletion mutation and displays in vitro sensitivity to gefitinib, erlotinib, and cetuximab (27). We grew HCC-827 cells as xenografts and established an IC50 concentration for cetuximab of 0.4 mg/kg. Furthermore, cetuximab, as well as erlotinib, showed efficacy against lung tumors in transgenic mice overexpressing an activated allele of EGFR (17). Hence, our findings show that preclinical models of NSCLC with activating EGFR mutations are as sensitive to cetuximab as they are to small-molecule inhibitors. A second point mutation in exon 20 (T790M) prevents small-molecule inhibitors of EGFR from binding to the ATP-binding cleft, thereby creating resistance (34, 35). Knowing the existence of clinically relevant resistance mechanisms against gefitinib and erlotinib, cetuximab was reported to have limited in vitro activity in Ba/F3 cells overexpressing a gefitinib-resistant allele of EGFR containing the L858R and T790M double mutation (26). However, in our in vivo xenograft tumor studies using the gefitinib- and erlotinib-resistant NSCLC cell line H1975 (L858R and T790M), cetuximab was clearly efficacious with dose-response studies defining an approximate IC50 value for cetuximab of 0.4 mg/kg. In conclusion, the response to cetuximab between xenograft mouse models that express wt or mutant EGFR is similar based on IC50 values presented in this study.

Combined antitumor activity of cetuximab with standard chemotherapy is well documented in colorectal (14), pancreatic (12), and lung (16) mouse models. In addition, gefitinib worked synergistically with gemcitabine in a head and neck carcinoma mouse model (36). Combinations with gefitinib and paclitaxel showed antitumor effects in a renal mouse model (37), whereas sequence-specific dosing of gefitinib and docetaxel was analyzed in a bladder cancer model (38). These data, together with data presented in this study, indicate that inhibition of EGFR together with blockade of cellular processes that are involved in proliferation, cell division, or cell cycle control resulted in increased antitumor activity. A notable exception was the lack of better efficacy of the combination of cetuximab with platinum-based agents in H1975 xenografts. It is currently unknown whether this is caused by mutant EGFR or downstream signaling networks that are differentially affected by mutant EGFR. A clinical analysis of NSCLC patients with wt or mutated EGFR did not uncover differential response rates to platinum agents (39). Our preclinical combination data are thus far in agreement with human clinical trial data. Several phase II trials have reported antitumor effects with cetuximab as monotherapy (40) or in combination with gemcitabine, docetaxel, cisplatin, and carboplatin in chemotherapy-naïve and chemotherapy–refractory/resistant NSCLC patients (8, 41, 42). An exploratory study using 38 tumor specimens showed that two of three patients with mutant EGFR experienced stable disease in a cetuximab monotherapy trial enrolling pretreated NSCLC patients (40, 43). It remains to be determined whether clinical response rates to cetuximab combination treatments in NSCLC are influenced by EGFR mutations.

Basal levels of EGFR kinase activity were higher in H1975 xenografts compared with H292 tumors, whereas total EGFR levels were comparable. Higher activity in H1975 cells was most likely caused by the constitutively active kinase domain of EGFR. Cetuximab reduced kinase activity in both H292 and H1975 tumor extracts. The reduction in H1975 cells is noteworthy because EGFR is activated in a ligand-independent fashion through mutations in the kinase domain. We conclude that mutant EGFR kinase activity in H1975 xenografts was induced by circulating ligand. Because H1975 cells are heterozygous for mutated EGFR, cetuximab might inhibit EGFR expressed from the wt allele. This interaction could inhibit mutated EGFR through impaired heterodimerization. However, such an explanation seems unlikely because the copy number of the mutated allele is in excess over the wt allele in H1975 cells (24). We noticed a 30% reduction in total EGFR in H1975 tumors treated with cetuximab. Although the effect missed statistical significance, it points out a difference in the response of wt and mutant EGFR to cetuximab. Internalization and degradation of the receptor could explain why constitutively activated, ligand-independent mutant EGFR was susceptible to cetuximab in H1975 xenografts. Such a model is supported by immunohistochemical analysis, which revealed a transition of Tyr1086 staining from the cell membrane to the cytoplasm after cetuximab treatment. The addition of docetaxel enhanced the cetuximab-mediated decrease in phospho-EGFR although docetaxel did not affect phosphorylation of EGFR. Inhibition of the EGFR pathway by combination therapy also caused lowered levels of phospho-MAPK and ultimately a reduction in phosphorylated histone H3. Conversely, our studies looking at markers of apoptosis did not support a similar statement about combined activity of cetuximab and docetaxel. In H1975 tumors, docetaxel is the sole driver of apoptosis even in the presence of cetuximab. Altogether, antitumor efficacy of the combination of cetuximab and docetaxel was mediated by a decrease in proliferation with a concomitant increase in apoptosis. This treatment protocol highlights the benefits of combining two agents with different mechanisms of action.

Acquisition of drug resistance to the reversible inhibitors gefitinib and erlotinib appears upon somatic mutation of tyrosine T790M positioned in the kinase domain of EGFR. In fact, inherited susceptibility to lung cancer has been described in humans and is associated with germ line transmission of the T790M mutation (44). One additional patient-associated EGFR mutation resistant to gefitinib is an insertion mutation in exon 20 (45). Our research suggests that cetuximab might be a clinically viable option for NSCLC patients with the L858R/T790M double mutation or the T790M germ line mutation. The effects of cetuximab on the exon 20 insertion mutation have not been tested yet. Alternative approaches to overcome gefitinib resistance have been reported and involve irreversible small-molecule inhibitors of EGFR, such as HKI-272 or EKB-569 (25, 46) and CL-387,785 (23, 26, 45). To prevent the emergence of the T790M mutation, it has been suggested to combine gefitinib and cetuximab as an anti-EGFR strategy. Such an approach is partly based on lessons learned from the treatment of chronic myeloid leukemia and gastrointestinal stromal tumors with imatinib (47). A common resistance mechanism to imatinib is the T315I mutation in the Abl tyrosine kinase domain of Bcr-Abl (48, 49). This mutation leads to a structural change in Abl, which is similar to the one in EGFR caused by the T790M mutation. Combining imatinib with a reagent that inhibits the T315I mutation in newly diagnosed chronic myeloid leukemia would not only inhibit Bcr-Abl but also prevent the clonal selection and spread of tumor cells that have acquired the resistance mutation. This treatment paradigm can be applied to NSCLC patients that are diagnosed with gefitinib-sensitizing mutations in exons 19 or 21. In fact, combinations of gefitinib and cetuximab were tested in preclinical lung and epidermoid carcinoma mouse models and showed superior and more durable efficacy over monotherapy (15, 29).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Erik Corcoran, Jessica Kearney, Chris Damoci, and Ling Ling Rong for help with NSCLC mouse models and histology, and Cindy Wang, Robin Rolser, and Jacqueline Doody for sequencing EGFR in NSCLC cell lines.

1
Hanahan D, Weinberg RA. The hallmarks of cancer.
Cell
2000
;
100
:
57
–70.
2
Blume-Jensen P, Hunter T. Oncogenic kinase signalling.
Nature
2001
;
411
:
355
–65.
3
Gschwind A, Fischer OM, Ullrich A. The discovery of receptor tyrosine kinases: targets for cancer therapy.
Nat Rev Cancer
2004
;
4
:
361
–70.
4
Schlessinger J. Cell signaling by receptor tyrosine kinases.
Cell
2000
;
103
:
211
–25.
5
Mendelsohn J, Baselga J. The EGF receptor family as targets for cancer therapy.
Oncogene
2000
;
19
:
6550
–65.
6
Hynes NE, Lane HA. ERBB receptors and cancer: the complexity of targeted inhibitors.
Nat Rev Cancer
2005
;
5
:
341
–54.
7
Franklin WA, Veve R, Hirsch FR, Helfrich BA, Bunn PA, Jr. Epidermal growth factor receptor family in lung cancer and premalignancy.
Semin Oncol
2002
;
29
:
3
–14.
8
Ettinger DS. Clinical implications of EGFR expression in the development and progression of solid tumors: focus on non-small cell lung cancer.
Oncologist
2006
;
11
:
358
–73.
9
Goldstein NI, Prewett M, Zuklys K, Rockwell P, Mendelsohn J. Biological efficacy of a chimeric antibody to the epidermal growth factor receptor in a human tumor xenograft model.
Clin Cancer Res
1995
;
1
:
1311
–8.
10
Graham J, Muhsin M, Kirkpatrick P. Cetuximab.
Nat Rev Drug Discov
2004
;
3
:
549
–50.
11
Prewett M, Rothman M, Waksal H, Feldman M, Bander NH, Hicklin DJ. Mouse-human chimeric anti-epidermal growth factor receptor antibody C225 inhibits the growth of human renal cell carcinoma xenografts in nude mice.
Clin Cancer Res
1998
;
4
:
2957
–66.
12
Bruns CJ, Harbison MT, Davis DW, et al. Epidermal growth factor receptor blockade with C225 plus gemcitabine results in regression of human pancreatic carcinoma growing orthotopically in nude mice by antiangiogenic mechanisms.
Clin Cancer Res
2000
;
6
:
1936
–48.
13
Overholser JP, Prewett MC, Hooper AT, Waksal HW, Hicklin DJ. Epidermal growth factor receptor blockade by antibody IMC-C225 inhibits growth of a human pancreatic carcinoma xenograft in nude mice.
Cancer
2000
;
89
:
74
–82.
14
Prewett MC, Hooper AT, Bassi R, Ellis LM, Waksal HW, Hicklin DJ. Enhanced antitumor activity of anti-epidermal growth factor receptor monoclonal antibody IMC-C225 in combination with irinotecan (CPT-11) against human colorectal tumor xenografts.
Clin Cancer Res
2002
;
8
:
994
–1003.
15
Huang S, Armstrong EA, Benavente S, Chinnaiyan P, Harari PM. Dual-agent molecular targeting of the epidermal growth factor receptor (EGFR): combining anti-EGFR antibody with tyrosine kinase inhibitor.
Cancer Res
2004
;
64
:
5355
–62.
16
Raben D, Helfrich B, Chan DC, et al. The effects of cetuximab alone and in combination with radiation and/or chemotherapy in lung cancer.
Clin Cancer Res
2005
;
11
:
795
–805.
17
Ji H, Li D, Chen L, et al. The impact of human EGFR kinase domain mutations on lung tumorigenesis and in vivo sensitivity to EGFR-targeted therapies.
Cancer Cell
2006
;
9
:
485
–95.
18
Siegel-Lakhai WS, Beijnen JH, Schellens JH. Current knowledge and future directions of the selective epidermal growth factor receptor inhibitors erlotinib (Tarceva) and gefitinib (Iressa).
Oncologist
2005
;
10
:
579
–89.
19
Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib.
N Engl J Med
2004
;
350
:
2129
–39.
20
Paez JG, Janne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy.
Science
2004
;
304
:
1497
–500.
21
Pao W, Miller V, Zakowski M, et al. EGF receptor gene mutations are common in lung cancers from “never smokers” and are associated with sensitivity of tumors to gefitinib and erlotinib.
Proc Natl Acad Sci U S A
2004
;
101
:
13306
–11.
22
Weinstein IB. Cancer. Addiction to oncogenes—the Achilles heal of cancer.
Science
2002
;
297
:
63
–4.
23
Kobayashi S, Boggon TJ, Dayaram T, et al. EGFR mutation and resistance of non-small-cell lung cancer to gefitinib.
N Engl J Med
2005
;
352
:
786
–92.
24
Pao W, Miller VA, Politi KA, et al. Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the EGFR kinase domain.
PLoS Med
2005
;
2
:
e73
.
25
Kwak EL, Sordella R, Bell DW, et al. Irreversible inhibitors of the EGF receptor may circumvent acquired resistance to gefitinib.
Proc Natl Acad Sci U S A
2005
;
102
:
7665
–70.
26
Kobayashi S, Ji H, Yuza Y, et al. An alternative inhibitor overcomes resistance caused by a mutation of the epidermal growth factor receptor.
Cancer Res
2005
;
65
:
7096
–101.
27
Amann J, Kalyankrishna S, Massion PP, et al. Aberrant epidermal growth factor receptor signaling and enhanced sensitivity to EGFR inhibitors in lung cancer.
Cancer Res
2005
;
65
:
226
–35.
28
Thomson S, Buck E, Petti F, et al. Epithelial to mesenchymal transition is a determinant of sensitivity of non-small-cell lung carcinoma cell lines and xenografts to epidermal growth factor receptor inhibition.
Cancer Res
2005
;
65
:
9455
–62.
29
Matar P, Rojo F, Cassia R, et al. Combined epidermal growth factor receptor targeting with the tyrosine kinase inhibitor gefitinib (ZD1839) and the monoclonal antibody cetuximab (IMC-C225): superiority over single-agent receptor targeting.
Clin Cancer Res
2004
;
10
:
6487
–501.
30
Luo FR, Yang Z, Dong H, et al. Prediction of active drug plasma concentrations achieved in cancer patients by pharmacodynamic biomarkers identified from the geo human colon carcinoma xenograft model.
Clin Cancer Res
2005
;
11
:
5558
–65.
31
Sirotnak FM, Zakowski MF, Miller VA, Scher HI, Kris MG. Efficacy of cytotoxic agents against human tumor xenografts is markedly enhanced by coadministration of ZD1839 (Iressa), an inhibitor of EGFR tyrosine kinase.
Clin Cancer Res
2000
;
6
:
4885
–92.
32
Raben D, Helfrich B, Bunn PA, Jr. Targeted therapies for non-small-cell lung cancer: biology, rationale, and preclinical results from a radiation oncology perspective.
Int J Radiat Oncol Biol Phys
2004
;
59
:
27
–38.
33
Erjala K, Sundvall M, Junttila TT, et al. Signaling via ErbB2 and ErbB3 associates with resistance and epidermal growth factor receptor (EGFR) amplification with sensitivity to EGFR inhibitor gefitinib in head and neck squamous cell carcinoma cells.
Clin Cancer Res
2006
;
12
:
4103
–11.
34
Johnson BE, Janne PA. Epidermal growth factor receptor mutations in patients with non-small cell lung cancer.
Cancer Res
2005
;
65
:
7525
–9.
35
Pao W, Miller VA. Epidermal growth factor receptor mutations, small-molecule kinase inhibitors, and non-small-cell lung cancer: current knowledge and future directions.
J Clin Oncol
2005
;
23
:
2556
–68.
36
Chun PY, Feng FY, Scheurer AM, Davis MA, Lawrence TS, Nyati MK. Synergistic effects of gemcitabine and gefitinib in the treatment of head and neck carcinoma.
Cancer Res
2006
;
66
:
981
–8.
37
Sumitomo M, Asano T, Asakuma J, Asano T, Horiguchi A, Hayakawa M. ZD1839 modulates paclitaxel response in renal cancer by blocking paclitaxel-induced activation of the epidermal growth factor receptor-extracellular signal-regulated kinase pathway.
Clin Cancer Res
2004
;
10
:
794
–801.
38
Kassouf W, Luongo T, Brown G, Adam L, Dinney CP. Schedule dependent efficacy of gefitinib and docetaxel for bladder cancer.
J Urol
2006
;
176
:
787
–92.
39
Lee KH, Han SW, Hwang PG, et al. Epidermal growth factor receptor mutations and response to chemotherapy in patients with non-small-cell lung cancer.
Jpn J Clin Oncol
2006
;
36
:
344
–50.
40
Hanna N, Lilenbaum R, Ansari R, et al. Phase II trial of cetuximab in patients with previously treated non-small-cell lung cancer.
J Clin Oncol
2006
;
24
:
5253
–8.
41
Thienelt CD, Bunn PA, Jr., Hanna N, et al. Multicenter phase I/II study of cetuximab with paclitaxel and carboplatin in untreated patients with stage IV non-small-cell lung cancer.
J Clin Oncol
2005
;
23
:
8786
–93.
42
Robert F, Blumenschein G, Herbst RS, et al. Phase I/IIa study of cetuximab with gemcitabine plus carboplatin in patients with chemotherapy-naive advanced non-small-cell lung cancer.
J Clin Oncol
2005
;
23
:
9089
–96.
43
Tsuchihashi Z, Khambata-Ford S, Hanna N, Janne PA. Responsiveness to cetuximab without mutations in EGFR.
N Engl J Med
2005
;
353
:
208
–9.
44
Bell DW, Gore I, Okimoto RA, et al. Inherited susceptibility to lung cancer may be associated with the T790M drug resistance mutation in EGFR.
Nat Genet
2005
;
37
:
1315
–6.
45
Greulich H, Chen TH, Feng W, et al. Oncogenic transformation by inhibitor-sensitive and -resistant EGFR mutants.
PLoS Med
2005
;
2
:
e313
.
46
Carter TA, Wodicka LM, Shah NP, et al. Inhibition of drug-resistant mutants of ABL, KIT, and EGF receptor kinases.
Proc Natl Acad Sci U S A
2005
;
102
:
11011
–6.
47
Baselga J. Targeting tyrosine kinases in cancer: the second wave.
Science
2006
;
312
:
1175
–8.
48
Gorre ME, Mohammed M, Ellwood K, et al. Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification.
Science
2001
;
293
:
876
–80.
49
Shah NP, Tran C, Lee FY, Chen P, Norris D, Sawyers CL. Overriding imatinib resistance with a novel ABL kinase inhibitor.
Science
2004
;
305
:
399
–401.