Antibody‐dependent cellular cytotoxicity of cetuximab against tumor cells with wild‐type or mutant epidermal growth factor receptor

Abstract

Cetuximab (Erbitux, IMC‐C225) is a monoclonal antibody targeted to the epidermal growth factor receptor (EGFR). To clarify the mode of antitumor action of cetuximab, we examined antibody‐dependent cellular cytotoxicity (ADCC) activity against several tumor cell lines expressing wild‐type or mutant EGFR. ADCC activity and complement‐dependent cytolysis activity were analyzed using the CytoTox 96 assay. ADCC activities correlated with the EGFR expression value (R = 0.924). ADCC activities were detected against all tumor cell lines, except K562 cells in a manner dependent on the cellular EGFR expression level, whereas complement‐dependent cytolysis activity was not detected in any of the cell lines. The ADCC activity mediated by cetuximab was examined in HEK293 cells transfected with wild‐type EGFR (293W) and a deletional mutant of EGFR (293D) in comparison with the mock transfectant (293M). ADCC activity was detected in 293W and 293D cells, in a cetuximab dose‐dependent manner, but not in 293M cells (<10%). These results indicate that ADCC‐dependent antitumor activity results from the degree of affinity of cetuximab for the extracellular domain of EGFR, independent of EGFR mutation status. These results suggest ADCC activity to be one of the modes of therapeutic action of cetuximab and to depend on EGFR expression on the tumor cell surface. (Cancer Sci 2007; 98: 1275–1280)

The epidermal growth factor receptor (EGFR) is a member of the ErbB family of receptors that is abnormally activated in many malignancies. EGFR is frequently overexpressed or abnormally activated in tumors. EGFR overexpression correlates with a worse outcome.^{( 1} , ^{2 )} Early studies with anti‐EGFR monoclonal antibodies (mAb) were shown to inhibit the growth of cancer cells bearing EGFR.^{( 3 )}

Cetuximab (IMC‐225, Erbitux) is a recombinant, human–murine chimeric mAb that is produced in mammalian (murine myeloma) cell culture and targeted specifically to EGFR. Cetuximab is composed of a murine Fv (EGFR‐binding) lesion and a human IgG1 heavy and κ light chain Fc (constant) region. In vitro studies have shown that cetuximab competes with endogenous ligands to bind with the external domain of EGFR. Cetuximab binds to EGFR with 10‐fold higher affinity than endogenous ligands (0.1–0.2 nM cetuximab vs 1 nM epidermal growth factor [EGF] or transforming growth factor (TGF)‐α, respectively).^{( 4 )} Cetuximab has shown promising preclinical and clinical activity in a variety of tumor types.^{( 5 )}

The anti‐tumor strategy is to direct mAb to the ligand‐binding extracellular domain and to prevent ligand binding and ligand‐dependent receptor inhibition. The use of humanized murine–human chimeric mAb of the IgG1 subtype is now well established for the treatment of human cancers. Treatment of advanced breast cancer with human epidermal growth factor receptor type 2 (HER‐2)‐specific trastuzumab (Herceptin) and of follicular non‐Hodgkin B‐cell lymphoma with CD20‐specific rituximab (Mabthera, Rituxan) has been shown to increase overall survival. Human IgG1 is thought to eliminate tumor cells via complement‐dependent cytolysis (CDC) and antibody‐dependent cellular cytotoxicity (ADCC), depending on the target, and also by direct pro‐apoptotic signaling or growth factor receptor antagonism. Clynes et al. suggested that ADCC is a major in vivo mechanism of IgG1 action.^{( 6 )} Recently, several mAb, including trastuzumab, which act predominantly via ADCC and CDC have been approved for the treatment of cancer patients. These include chimeric IgG1 mAb rituximab binding to the B‐cell differentiation antigen CD20 for the treatment of B‐cell lymphomas,^{( 7 )} humanized IgG1 mAb trastuzumab targeting HER‐2 overexpressed in a subgroup of breast cancers,^{( 8 )} and humanized IgG1 alemtuzumab (Campath) targeting the differentiation antigen CD52 for the treatment of B‐cell chronic lymphocytic leukemia.^{( 9 )}

We hypothesized that ADCC is a possible mode of action of cetuximab against EGFR‐expressing tumors. The present study was designed to clarify the role of cetuximab in ADCC and CDC activity, and to evaluate the relationship between EGFR expression status and cetuximab‐mediated ADCC and CDC activity.

Methods

Cell lines and cultures.  A human leukemia cell line (K562), a non‐small cell lung cancer (NSCLC) cell line (A549) and a human embryonic kidney cell line (HEK293) were obtained from the American Type Culture Collection (Manassas, VA, USA). Human NSCLC cell lines A431, PC‐9 and PC‐14 were obtained from Tokyo Medical University (Tokyo, Japan). Human NSCLC cell lines Ma‐1 and 11_18 were obtained from the National Cancer Center Research Institute (Tokyo, Japan). PC‐9 and Ma‐1 are known to contain E746_A750del, and 11_18 is known to contain L858R in tyrosine kinase domains of EGFR. The other cell lines are known to have wild‐type EGFR. K562, HEK293, A431, PC‐9, PC‐14, Ma‐1 and 11_18 cells were cultured in RPMI‐1640 (Sigma, St Louis, MO, USA) supplemented with 10% heat‐inactivated fetal bovine serum (FBS; Gibico BRL, Grand Island, NY, USA). A549 cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Carlsbad, CA) with 10% heat‐inactivated FBS.

Plasmid construction and transfection.  Construction of the mock expression plasmid vector (empty vector) and of the wild‐type EGFR and 15‐bp deletional EGFR (E746‐A750del type deletion) vectors, both of which possess the same deletion site as that observed in PC‐9 cells, have been described elsewhere.^{( 10 )} The plasmids were transfected into HEK293 cells and the transfectants were selected with Zeosin (Sigma). The stable transfectants (pooled cultures) of the empty vector, wild‐type EGFR and its deletion mutant were designated 293M, 293W and 293D cells, respectively.

Compound.  The mAb anti‐EGFR cetuximab (IMC‐225, Erbitux) was kindly provided by Bristol Myers Squibb (New York, NY, USA).

Analysis of EGFR expression on the cell surface.  Cell surface expression of EGFR in tumor cell lines was quantified using a flow cytometric system (BD LSR; Becton‐Dickinson, San Jose, CA, USA). The binding of cetuximab to tumor cell lines was titrated using FACS analysis. Cetuximab and another anti‐EGFR mAb (R‐1, sc‐101; Santa Cruz Biotechnology, Santa Cruz, CA, USA) were used as the primary antibodies. Then, 1 × 10⁶ tumor cells were incubated with 1 µg/mL cetuximab in 1% bovine serum albumin in phosphate‐buffered saline (PBS) for 30 min at room temperature. After the first reactions, the cell surface was stained with 10 µg/mL fluorescein‐conjugated antihuman IgG (Vector, Burlingame, CA, USA) for 45 min at room temperature in the dark. After the second reactions, the tumor cells were resuspended in 1 mL PBS. For analysis using the anti‐EGFR mAb, 1 µg EGFR mAb per 1 × 10⁶ tumor cells was used as the primary antibody. The secondary antibody was 10 µg/mL fluorescein‐conjugated antimouse IgG (Vector). A minimum of 2 × 10⁴ cells were analyzed by flow cytometry. Control experiments were carried out in the absence of primary antibodies. Data were analyzed with CellQuest software and the modifying program (Beckton Dickinson, CA, USA). The magnitude of surface expression of these proteins was indicated by the mean fluorescence intensity (MFI) of positively stained cells. The expression values were calculated as follows:

Expression value = (MFI of positively stained cells)/ (MFI of control cells).

The correlation between the expression of R‐1‐combined EGFR and that of cetuximab‐combined EGFR were calculated using simple regression analysis.

Cytotoxicity assays.  ADCC and CDC were examined using the CytoTox 96 Non‐Radioactive Cytotoxicity Assay (Promega, Madison, WI). For quantification of ADCC activity, peripheral blood mononuclear cells were isolated from healthy volunteers with Lymphocyte Separation Medium (Cappel, Aurora, OH, USA) and used as effector cells. The target cells were suspended in RPMI medium without FBS and plated in a 96‐well U‐bottom microtiter plate at 5 × 10³ cells/well. Cetuximab was added in triplicate to the individual wells at various concentrations from 0.001 to 10 µg/mL and effector cells were added at an effector : target cell ratio of 10:1. For quantification of CDC activity, human serum from a healthy volunteer was obtained as a compliment source. To yield a 1:3 final dilution, 50 µL serum was added. The plates were incubated for 4 h at 37°C, and the absorbance of the supernatants at 490 nm was recorded to determine the release of lactate dehydrogenase. The average of absorbance values for the culture medium background was subtracted from experimental release (A), target cell spontaneous release (B), effector cell spontaneous release (C) and target cell maximum release (D). The specific cytolysis percentage was calculated using the following formula:

Cytotoxicity (%) = (A – B – C)/(D – B) × 100.

The correlation between the expression of cetuximab‐combined EGFR and ADCC activity was calculated using a simple regression analysis.

Growth‐inhibition assay.  We used the 3‐(4,5‐dimethylthiazol‐2‐y)‐2,5‐diphenyltetrazolium bromide (MTT) assay to evaluate the cytotoxicity of various drug concentrations. Two hundred microliters of an exponentially growing cell suspension was seeded in a 96‐well microtiter plate, and cetuximab‐containing solution was added at various concentrations (from 0.001 to 100 µg/mL). Each experiment was carried out in triplicate for each drug concentration and independently three times.

Growth inhibitory assay for the combination of gefitinib and cetuximab‐mediated ADCC in the PC‐9 cell line.  We analyzed the growth inhibitory effect of the combination of gefitinib and cetuximab‐mediated ADCC in the PC‐9 cell line using the MTT assay. Two hundred microliters containing 1000 PC‐9 cells, and various concentrations of gefitinib, were seeded in a 96‐well microtiter plate. Then, 10 µL of cetuximab‐containing solutions of various concentrations (from 0.1 to 10 µg/mL) and 20 000 effector cells were added.

Western blotting.  PC‐9, PC‐14 and A549 cell lines were seeded in cell culture plates at a density of 6.0 × 10⁵ cells/plate and allowed to grow overnight in appropriate maintenance cell culture media for each cell line containing 10% heat‐inactivated FBS. The media were then replaced with RPMI‐1640 (Sigma) (PC‐9 and PC‐14) or DMEM without FBS, with or without cetuximab (10 and 100 µg/mL). The cells were incubated for a further 24 h and stimulated or not stimulated with EGF (100 ng/mL) under serum starvation conditions. Cells were washed with ice‐cold PBS and scraped immediately after adding 50 µL of M‐PER mammalian protein extraction reagent (Pierce Biotechnology, Rockford, IL, USA). The protein extracts were separated by electrophoresis on 7.5% sodium dodecylsulfate–polyacrylamide gels and transferred to nitrocellulose membranes by electroblotting. The membranes were probed with a mouse monoclonal antibody against EGFR (Transduction Laboratory, San Diego, CA, USA), and phosphor‐EGFR (specific for Tyr1068), Akt, phosphor‐Akt, p44/42 MAPK and phosphor‐p44/42 MAPK antibodies (Cell Signaling Technology, Beverly, MA, USA) as primary antibodies, followed by a horseradish peroxidase‐conjugated secondary antibody. The bands were visualized with an electrochemiluminescence reagent (ECL; Amersham, Piscataway, NJ, USA).

Results

Binding properties of cetuximab to tumor cell lines expressing EGFR.  The A431 cells expressed a high level of EGFR on their surfaces. Cell surface EGFR expression values of the PC‐9, PC‐14, A549, Ma‐1 and 11_18 cell lines were lower than that of A431. The MFI for the K562 cells was less than 10 (Table 1). A good correlation was observed between the binding of cetuximab and R‐1 antibody with a correlation coefficient of 0.999 (P < 0.001; Fig. 1a).

Table 1.

Epidermal growth factor receptor (EGFR) expression values and antibody‐dependent cellular cytotoxicity (ADCC) activity

Cell liine	EGFR expression (R‐1)	EGFR expression (cetuximab)	ADCC (%)
A431	286.2 ± 13.7	318.9 ± 98.2	30.7
PC‐9	9.7 ± 6.2	20.1 ± 10.2	20.1
PC‐14	17.6 ± 1.5	42.2 ± 8.6	26.8
A549	9.1 ± 1.9	19.1 ± 6.2	24.2
Ma‐1	13.8 ± 1.4	27.5 ± 2.9	22.3
11_18	6.1 ± 0.6	12.6 ± 1.1	15.5
K562	1.1 ± 0.4	2.8 ± 1.6	7.0
293M	3.7 ± 1.6	8.6 ± 3.2	8.2
293W	40.19 ± 6.2	39.73 ± 6.2	16.3
293D	55.21 ± 21.9	53.04 ± 8.2	18.9

Expression values and ADCC activity were calculated as described in the Materials and Methods section. The mean of expression values from three different experiments and standard deviations are shown. The values for cetuximab‐combined EGFR expression are shown for a concentration of 1 µg/mL.

Figure 1.

Epidermal growth factor receptor (EGFR) expression and cetuximab‐mediated antibody‐dependent cellular cytotoxicity (ADCC) activity in the tumor cell lines. (a) Correlation between the expression of cetuximab‐combined EGFR and R‐1‐combined EGFR. The values for cetuximab‐combined EGFR expression are shown for a concentration of 1 µg/mL. The correlation coefficient between the results of these assays was 0.999. (b) Cetuximab‐mediated ADCC activity in tumor cell lines at concentrations ranging from 0.001 to 10 µg/mL was determined using the CytoTox 96 Non‐Radioactive Cytotoxicity Assay. (c) Correlation between expression values of cetuximab‐combined EGFR and ADCC activity in the seven tumor cell lines. The values for cetuximab‐combined EGFR expression and cetuximab‐mediated ADCC activity are shown for a concentration of 1 µg/mL. The correlation coefficient between the results of these assays was 0.924. (d) Correlation between expression values of cetuximab‐combined EGFR and ADCC activity in transfected HEK293 cell lines. The correlation coefficient between the results of these assays was 0.952.

ADCC and CDC activities in tumor cell lines.  ADCC activities of cetuximab were detected in all tumor cell lines except K562 (Table 1; Fig. 1b). In the K562 cells, % ADCC activities were lower than 10% at all concentrations of cetuximab examined (from 0.001 to 10 µg/mL). ADCC activity mediated by cetuximab was highly correlated with the binding values of cetuximab to cells expressing EGFR (R = 0.924, P = 0.003; Fig. 1c). CDC activity was not detected in any of the cell lines in the cetuximab concentration range from 0.001 to 10 µg/mL.

Direct growth inhibitory effect of cetuximab on tumor cell lines.   Cetuximab showed no growth inhibitory effect in any of the cell lines examined, regardless of EGFR expression levels. Even the highest concentration of cetuximab (100 µg/mL) did not inhibit growth in any of the cell lines (Fig. 2).

Figure 2.

Growth inhibitory effect of cetuximab on non‐small cell lung cancer cell lines: (a) A431; (b) PC‐9; (c) PC‐14; (d) A549; (e) Ma‐1; (f) 11_18; and (g) K562. Cell growth was not inhibited at any concentration, even a high concentration (10 µg/mL). The figure shows the dose‐dependent growth inhibitory effect of gefitinib with various concentrations of cetuximab (0–10 µg/mL). Results are expressed as percentages of the untreated control value. The data shown are the mean + SD values from triplicate experiments.

ADCC activities of cetuximab against the cells transfected with wild‐type and mutant EGFR. EGFR expression was detected in 293W and 293D cells, but not in 293M cells (Table 1). The ADCC activity mediated by cetuximab in 293W and 293D cells was dose dependent. In contrast, ADCC activities in 293M cells were <10% at all concentrations of cetuximab tested (0.001–10 µg/mL). There was a good correlation between the ADCC activities and the levels of cetuximab binding to EGFR in the cells (R = 0.996, P = 0.055; Fig. 1d). These results indicate that ADCC depends on the level of cetuximab binding to EGFR, but not the mutation status of the EGFR tyrosine kinase domains.

Direct growth inhibitory effect of the combination of gefitinib and cetuximab‐mediated ADCC in the PC‐9 cell line.  The growth inhibitory effect in the PC‐9 cell line was shown by effector cells at a gefitinib exposure exceeding 0.01 µM and was concentration dependent (Fig. 3). When cetuximab was added, growth was inhibited in a cetuximab concentration‐dependent manner. An additive growth inhibitory effect was recognized between 0 and 0.01 µM of gefitinib. This additive growth inhibitory effect could not be evaluated at concentrations between 0.1 and 1.0 µM because of the strong inhibitory effect of gefitinib alone.

Figure 3.

Growth inhibitory effects of combining gefitinib and cetuximab‐mediated antibody‐dependent cellular cytotoxicity (ADCC). The figure shows dose‐dependent growth inhibitory effects of gefitinib with various concentrations of cetuximab (solid circle, 0 µg/mL; solid square, 0.1 µg/mL; open circle, 1.0 µg/mL; solid triangle, 10 µg/mL). Results are expressed as a percentage of the untreated control value. The data shown represent the median values of triplicate experiments.

Effect of cetuximab on phosphorylation of EGFR and its downstream signaling molecules in NSCLC cells.  Phosphorylation of EGFR was strongly expressed in PC‐9 regardless of EGF treatment, and the phosphorylation of EGFR continued the strong expression during cetuximab treatment. Phosphorylation of EGFR was slightly expressed in PC‐14 and A549 without EGF treatment, but the phosphorylation of EGFR was enhanced by the EGF treatment. Although the enhancement of phosphorylation was inhibited dose dependently by cetuximab, the phosphorylation was not completely inhibited at the highest concentration (10 µg/mL) of cetuximab. Phosphorylation of 44/42 MAPK and Akt was increased in all cell lines compared with the absence of EGF treatment. Although the increase in phosphorylation was diminished by adding cetuximab, phosphorylation was not completely inhibited at the highest concentration (10 µg/mL) of cetuximab (Fig. 4).

Figure 4.

Effects of cetuximab on phosphorylation of epidermal growth factor receptor (EGFR), Akt and p44/42 MAPK in non‐small cell lung cancer cell lines. (a) EGFR mutant cell line PC‐9 (with the E746_A750del mutation). (b) EGFR wild‐type cell line PC‐14. (c) EGFR wild‐type cell line A549. Cells were treated with cetuximab at the indicated concentrations for 24 h. Immunoblots of cellular protein were analyzed for phosphorylated and total EGFR, p44/42 MAPK and Akt. The experiments were repeated at least twice.

Discussion

Antibody therapies are a major approach in the treatment of various cancer types. Herein, we focused on the ADCC activity mediated by cetuximab against human lung cancer cells expressing wild‐type or mutant EGFR. Neither CDC nor direct growth inhibition mediated by cetuximab was detectable in our experiments.

Direct growth inhibition, ADCC and CDC mediated by antibodies are the modes of action of antibody therapies. We previously demonstrated that ADCC is the major mode of action of trastuzumab in breast cancer cell lines, even when used in combination with cisplatin.^{( 11 )} Cisplatin did not affect ADCC activity at the concentration for combined use in vitro. Clinical efficacies of cetuximab for various types of cancers have been demonstrated in many clinical studies using combinations with cytotoxic agents including cisplatin. Thus, ADCC is considered to be an important factor governing the efficacy of cetuximab.

Mukohara et al. reported that EGFR mutations in NSCLC cells are not associated with sensitivity to cetuximab in vitro.^{( 12 )} They focused on the direct growth inhibitory effect of cetuximab against lung cancer cells. We previously demonstrated that PC‐9 and 293 cells transfected with E746_A750del EGFR are hypersensitive to EGFR‐tyrosine kinase inhibitors.^{( 10 )} In contrast, we have demonstrated that ADCC activity mediated by cetuximab is not affected by EGFR mutation status in lung cancer cells or in 293 cells transfected with EGFR. Taken together, these observations indicate that cetuximab exerts its antitumor effects against human lung cancer cells independently of EGFR mutation status.

ADCC activity mediated by cetuximab has been demonstrated against 293 cells transfected with wild‐type and mutated EGFR. Higher ADCC activity against 293D cells compared with 293W cells was observed with cetuximab exposure (Fig. 1d; Table 1). However, ADCC was correlated with EGFR expression levels in these transfectants. The activity appears to depend on expression levels but not mutation status.

Approximately 30 mutations of EGFR have been reported in lung cancer.^{( 13} , ¹⁴ , ¹⁵ , ^{16 )} ADCC activity against PC‐9 cells with E746_A750del in exon 19, one of the common mutations, has been demonstrated herein. We also examined ADCC activity against another human lung cancer cell line, 11_18,^{( 17 )} with L858R in exon 21, which is another common mutation. Our results showed a strong positive correlation between ADCC activity and EGFR expression level, and that the impact on ADCC activity did not depend on the site of EGFR mutations.

Cetuximab is a chimeric antibody against the extracellular domain of EGFR. Other antitumor anti‐EGFR antibodies currently under investigation clinically include humanized antibodies.^{( 18 )} It remains unknown whether humanized and chimeric antibodies exert ADCC activity against lung cancer differentially, and this awaits future investigation.

Some investigators have reported on the predictive factor and enhancement of ADCC activity mediated by certain mAb other than cetuximab.^{( 19} , ²⁰ , ²¹ , ^{22 )} Important ADCC‐mediating effector cells that express receptors against the Fc region of IgG include monocytes and macrophages (FcγRI, IIa and IIb), granulocytes (FcgRII) and natural killer cells (FcγRIII).^{( 19 )} One group of researchers demonstrated single nucleotide polymorphisms of FcγRIII in individual patients correlating with rituximab‐dependent ADCC activity and the clinical response to rituximab.^{( 20 )} Carson et al. demonstrated that the natural killer cell‐mediated ADCC activity of breast cancer cell lines expressing HER2/neu, in the presence of trastuzumab, was markedly enhanced following stimulation with interleukin 2 and proposed the concurrent use of trastuzumab and interleukin‐2 therapy in patients with cancers expressing HER2/neu.^{( 21 )} However, from the view point of mAb but not effector cells, lack of fucosylation of the antibodies affects ADCC enhancement.^{( 22 )} Whether or not these factors enhance cetuximab‐mediated ADCC activity warrants further examination.

We showed additional growth inhibition by gefitinib and cetuximab in PC‐9 cells. PC‐9 cells had a deletional mutation in exon 19 of EGFR and hyper‐responsiveness to gefitinib. We think that cetuximab‐mediated ADCC increased the growth inhibition‐independent response to gefitinib. The ADCC activity could not be evaluated at higher concentrations of gefitinib (>0.1 µM) because PC‐9 cells were sufficiently inhibited at the higher concentrations. Additionally, we showed that some phosphorylations downstream of EGFR in NSCLC cell lines were mediated by cetuximab, although cetuximab had no growth inhibitory effect on the cell lines. We think that cetuximab‐combined EGFR inhibits binding of EGFR and its ligands, such as EGF, and that phosphorylation downstream of EGFR is inhibited as a consequence of the addition of cetuximab. We have shown that phosphorylation of 44/42 MAPK and Akt in NSCLC cell lines was increased by EGF treatment and decreased by then adding cetuximab. Phosphorylation of EGFR in PC‐14 and A549 cells was decreased with the addition of cetuximab, as in the 44/42 MAPK and Akt cell lines. Phosphorylation of EGFR in PC‐9 cells was strongly increased without ligands under serum starvation conditions and was not decreased by cetuximab. Phosphorylation that was independent of ligand binding to EGFR seem not to be controlled by cetuximab.

These results conclude that cetuximab has ADCC activity against tumor cells with EGFR expression, and ADCC activity depends on the degree of EGFR expression on tumor cell surfaces, additionally leading us to believe that cetuximab treatment has clinical activity in EGFR‐expressing tumor cells via cetuximab‐mediated ADCC.