Complement activation mediates cetuximab inhibition of non-small cell lung cancer tumor growth in vivo

Background Cetuximab, an antibody targeting the epidermal growth factor receptor (EGFR), increases survival in patients with advanced EGFR-positive non-small cell lung cancer when administrated in combination with chemotherapy. In this study, we investigated the role of complement activation in the antitumor mechanism of this therapeutic drug. Results EGFR-expressing lung cancer cell lines were able to bind cetuximab and initiate complement activation by the classical pathway, irrespective of the mutational status of EGFR. This activation led to deposition of complement components and increase in complement-mediated cell death. The influence of complement activation on the activity of cetuximab in vivo was evaluated in xenografts of A549 lung cancer cells on nude mice. A549 cells express wild-type EGFR and have a KRAS mutation. Cetuximab activity against A549 xenografts was highly dependent on complement activation, since complement depletion completely abrogated the antitumor efficacy of cetuximab. Moreover, cetuximab activity was significantly higher on A549 cells in which a complement inhibitor, factor H, was genetically downregulated. Conclusions We demonstrate for the first time that the in vivo antitumor activity of cetuximab can be associated with a complement-mediated immune response. These results may have important implications for the development of new cetuximab-based therapeutic strategies and for the identification of markers that predict clinical response.


Background
Lung cancer accounts for more than 25% of all cancer deaths in United States [1]. Non-small cell lung cancer (NSCLC) represents about 80% of all lung cancers. Current treatment options consist of surgical resection, platinum-based doublet chemotherapy, and radiation. Unfortunately, despite these therapies, the prognosis remains poor. Recent advances in the understanding of the molecular pathogenesis of the disease have led to the development of molecular targeted therapies for NSCLC [2]. Bevacizumab, a monoclonal antibody to vascular endothelial growth factor, and erlotinib, a small-molecule tyrosine kinase inhibitor (TKI) of epidermal growth factor receptor (EGFR), are targeted agents approved in the treatment of NSCLC [3]. The clinical efficacy of cetuximab, a humanized monoclonal antibody against the extracellular domain of EGFR, has also been evaluated. A randomized phase III trial has recently shown significantly prolonged survival of advanced NSCLC patients who received cetuximab in combination with platinumbased chemotherapy as first-line treatment [4]. Conversely, combinations of gefitinib or erlotinib, EGFR tyrosine kinase inhibitors (TKIs), with standard chemotherapy in advanced NSCLC have failed to show clinical benefit [5][6][7][8]. Another remarkable observation is that, in contrast to the evidence for TKI treatment, KRAS mutation status does not appear to be predictive of response to cetuximab in NSCLC [9][10][11]. These data strongly suggest clinically relevant differences between the mechanisms of action of EGFR-TKIs and cetuximab [12]. In this sense, it has been suggested that immune mechanisms may con-tribute to the antitumor activity of cetuximab [13]. In particular, cetuximab, alone or in combination with other antibodies, may elicit immunological responses such as antibody-dependent cellular cytotoxicity (ADCC) or complement activation [14][15][16][17].
A better understanding of the mechanisms that govern cetuximab antitumor activity is necessary to optimize its therapeutic efficacy and to identify those patients who are going to benefit from the treatment. In the current report we investigated the influence of the activation of complement in the action of cetuximab in an in vivo animal model. We also explored the possibility of enhancing complement activation in an attempt to increase the clinical efficacy of cetuximab.

Sera
Normal human serum (NHS) was used as the source of complement. A pool of sera from ten healthy donors was prepared. Heat inactivated NHS (HI-NHS) was obtained by incubation of the serum at 56°C for 30 minutes.

KRAS mutations
Human KRAS codon 12 mutations were assessed by sequencing. Genomic DNA was subjected to PCR amplification with the following set of intronic primers: sense 5'-CGATACACGTCTGCAGTCAA-3', antisense 5'-GGTCCTGCACCAGTAATATGC-3'. The PCR products were sequenced using the Big Dye Terminator V1.1 Cycle Sequencing Kit (Applied Biosystems) according to the protocol supplied by the manufacturer.

C1q fixation
A polystyrene 96-well plate was coated with 30 to 2000 ng of antibody per well in 100 μl of 50 mM sodium bicarbonate (pH 8.3) during one hour at room temperature. After washing, the plate was blocked overnight at 4°C with Tris-buffered saline (TBS) containing 1% bovine serum albumin, and 0.1% Tween 20. After washing, normal human serum, used as the source of C1q, was added in 100 μl of veronal buffer [1.8 mM barbital, 3.1 mM barbituric acid, 141 mM sodium chloride, 0.5 mM MgCl 2 and 0.15 mM CaCl 2 (pH 7.4)] and incubated for 30 min at 37°C. The plate was washed and the assay was developed with a rabbit anti-human C1q antibody (1:500; Dako), a goat anti-rabbit antibody coupled to horseradish peroxidase (1:1,000; Sigma-Aldrich) and O-phenylenediamine dihydrochloride (Sigma-Aldrich). A human IgG1 antibody (Sigma-Aldrich) was used as a positive control. The anti-factor H monoclonal antibody OX24, and heat inactivated NHS (56°C for 30 minutes) were used as negative controls. Cetuximab was kindly provided by Merck KGaA. OX24 was obtained as previously described [18].

Complement-mediated cell death
Complement-mediated cell death is associated with DNA fragmentation [19]. DNA fragmentation was evaluated using a method previously described [20]. In brief, 7 × 10 5 cells were resuspended in 0.2 ml of RPMI medium containing 30% NHS (or HI-NHS), and treated with 40 μg/ml of cetuximab for 24 hours at 37°C. Afterwards, cells were pelleted and fixed in 2 ml of cold 70% ethanol for 60 min at 4°C. Cells were centrifuged, washed twice with PBS, resuspended in 0.5 ml of PBS, and incubated with 10 μl of 1 mg/ml RNase A (Sigma) during 1 hour at 37°C. Finally, 5 μl of 1 mg/ml 7-aminoactinomycin D (Sigma) were added and, after incubation in the dark for 15 min at room temperature, cells were analyzed by flow cytome-try. The percentage of DNA giving fluorescence below the G 1 /G 0 peak was taken as measure of cell death.

Cell proliferation assay
Cells (2,000 A459 cells/well or 4,000 HCC827 cells/well) were seeded in 96-well plates in RPMI medium supplemented with 10% FBS. Twenty-four hours later, cells were treated with cetuximab at different concentrations, and incubated for another 24 hours. Ten microliters of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) at 5 g/l (Sigma) was added to each well and incubated for 4 hours at 37°C. Later, 100 μl of dimethyl sulfoxide was added to each well to dissolve the MTT-formazan crystals. Absorbance was measured at 540 nm, with a reference filter at 690 nm, using a microplate reader.

Xenograft model
Care of the animals was in accord with our institution guidelines. A549 cells (15 × 10 6 ) were mixed 1:1 with Growth Factor Reduced Matrigel Matrix (BD Biosciences), and injected subcutaneously on the right flank of 4-6 week old female athymic nude mice (Harlan Laboratories, Italy). Athymic mice are immunodeficient and cannot develop a complete adaptive immune response, but have complement and NK cell activities. Tumor growth was measured every 2-3 days. Tumor volume was calculated using the formula: Volume = length × width 2 × 0.5. Tumors were allowed to reach about 200-250 mm 3 before randomization. When indicated, complement was depleted with cobra venom factor (CVF), as previously described [21].

Statistical analysis
Data were analyzed by Student's t-test. A p value of less than 0.05 was considered as statistically significant.

Results and Discussion
We used three different lung cancer cell lines in the study: A549, HCC827, and H187 cells. A549 cells express wildtype EGFR and have a KRAS mutation (Fig. 1A-B). HCC827 cells express mutated-EGFR and wild-type KRAS. H187 cells do not have detectable levels of EGFR. We first analyzed the capacity of cetuximab to fix com-plement. C1q from NHS was able to bind to wells coated with cetuximab ( Fig. 2A), but not to wells coated with OX24, an IgG1 monoclonal antibody unable to fix complement [18]. C1q binding to cetuximab was inhibited by heat inactivation. We next assessed the capacity of cetuximab to fix complement on the membrane of lung cancer cells. We used 40 μg/ml of cetuximab (at the FDAapproved dosing level, according to the manufacturer, the peak and trough plasma concentrations of cetuximab range from 168 to 235 μg/ml and from 41 to 85 μg/ml, respectively). In cells expressing EGFR, cetuximab treatment initiated complement activation, irrespective of the mutational status of EGFR ( Fig. 2B-C). Complement activation by the antibody-dependent classical pathway results in C3b-deposition, which leads to the formation of the cytolytic C5b-C9 complex [22]. Treatment of EGFRexpressing cells with cetuximab and NHS led to the deposition of these complement components and to complement-mediated cell death (Fig. 3A-B). In contrast, Kimura et al previously reported that complement dependent cytotoxicity was not detected in A549 cells treated with cetuximab [14], and Dechant et al found that complement activation only occurred when cetuximab was combined with other EGFR antibodies [17]. In both studies cetuximab was used at 10 μg/ml. We confirmed that there is not complement-mediated cell death with 10 μg/ml of cetuximab, but higher concentrations are required (data not shown). As mentioned above, the concentration used in the present study (40 μg/ml) is similar to the lowest concentration found in plasma from patients treated with the standard dosing of cetuximab. Treatment of EGFR-negative H187 cells with 40 μg/ml of cetuximab did not affect C3b-deposition, C5b-C9 complex formation, or complement-mediated cell death (Fig.  It is important to realize that lysis is not the only biological consequence of complement activation in vivo. For example, complement activation releases anaphylatoxins that promote proinflammatory responses. In addition, the interaction of iC3b with CR3 (CD11b/CD18) on immune cells mediates complement-dependent cellular cytotoxicity (CDCC). Therefore, cetuximab-mediated complement activation may generate a powerful antitumor response in an in vivo setting. To test this hypothesis, we used A549 cells, which express wild-type EGFR and have a KRAS mutation. In in vitro studies, A549 cells are insensitive to the blockade of EGFR signaling and resistant to cetuximab [23]. This resistance may be a consequence of the presence of the KRAS mutation. However, cetuximab has some antitumor effect in vivo in A549 xenografts grown in nude mice [24]. This observation strongly suggests that cetuximab operates by in vivo mechanisms of action other than inhibition of EGFR-signaling. We first confirmed the resistance of A549 cells to cetuximab using an MTT assay (Fig 4A). Afterwards, we grew A549 xenografts in athymic mice and treated them with cetuximab. As previously reported [24], tumor growth was partially inhibited by cetuximab (Fig. 4B). Interestingly, explanted tumors from mice treated with cetuximab had high levels of C3 deposition, demonstrat-  Complement-dependent antitumor activity of cetuximab in A549 xenografts. A, MTT assay for the in vitro antitumor activity of cetuximab in A549 and HCC827 lung cancer cell lines (with wildtype and mutant-EGFR, respectively). Cell proliferation was determined after 24 h treatment using an MTT assay. The graph shows mean ± SEM. B, A549 cells were injected s.c. on athymic nude mice. When tumors reached 200-250 mm 3 , mice (n = 10) were randomized into two groups: one was treated with cetuximab and the other with PBS. Cetuximab was administered i.p. at 0.5 mg per mouse three times a week and tumor growth was measured every 2-3 days. Data represent mean ± SEM (*, P < 0.05). C, Representative example of the deposition of C3-related fragments, determined by immunofluorescence, in explanted tumors from each experimental group. D, A549 xenografts were generated as above. Animals (n = 18) were divided into three groups. In one group, complement was depleted with CVF. Cetuximab treatment was administrated to the CVF-treated group and to a second group with full complement activity. The last group received PBS. Data represent mean ± SEM (*, P < 0.05). ing the capacity of cetuximab to activate complement in vivo (Fig. 4C). The contribution of complement in the reduction of tumor growth was confirmed using complement-deficient mice. Complement depletion was achieved with cobra venom factor (CVF). Serum C3 levels were reduced to less than 10% as evaluated by ELISA (data not shown). Complement depletion completely abrogated the activity of cetuximab in A549 xenografts (Fig. 4D).
The efficacy of complement-activating antibodies could be potentiated by the induction of immune effectors suppressed by the tumor microenvironment. In this line, we tried to increase the efficacy of cetuximab by eliminating the control exerted by factor H, a complement inhibitor that prevents an efficient immune response against lung cancer cells [21]. Incubation of A549 cells with NHS and cetuximab in the presence of OX24 significantly increased the deposition of C3-related fragments (Fig.  5A). OX24 is an antibody that inhibits the binding of factor H to surface-bound C3b. To evaluate the contribution of factor H to the control of complement-mediated cetuximab activity, we used A549 cells in which the expression of factor H was stably downregulated by siRNA [21]. The antitumor activity of cetuximab was significantly higher in factor H-deficient cells than in cells transfected with a control siRNA (Fig. 5B). Cetuximab-mediated tumor growth reduction was more than 90% in factor H-deficient cells, compared to around 50% in control cells. In vivo antitumor activity of cetuximab was again associated with an increase in the deposition of C3-related fragments (Fig. 5C). In conclusion, these results demonstrate for the first time that a complement-mediated immune response induced by cetuximab participates in the control of tumor cell growth in vivo. This mechanism of action provides numerous opportunities to enhance the efficacy of therapeutical monoclonal antibodies [25].
Human tumors in immunocompromised mouse models do not entirely behave as syngeneic tumors in immunocompetent hosts. For example, although athymic mice have normal complement activity, they cannot develop an adequate adaptive immune response. In an intact immune system, both ADCC and CDCC would likely be important mediators of cetuximab antitumor activity (Fig. 6). Thus, the efficacy of cetuximab against human tumors might be attenuated in our immunocompromised animal model. Still, our data clearly indicate that complement activation is sufficient to elicit an intense response able to induce more than 90%-reduction in the tumor growth of cells resistant to blockade of the EGFR-signaling pathway. It would be also possible that the activation of complement may contribute to the antitumor effect of cetuximab in cells sensitive to the blockade of this pathway.
It would be interesting to evaluate the immune mechanisms through which complement activation is able to control tumor growth. A direct effect of complement on the target cells may be accompanied by the recruitment and activation of cellular effectors such as lymphokine activated killer (LAK) cells and natural killer (NK) cells. In addition, the combination of cetuximab with modifiers of the immune response may be considered an attractive therapeutic approach to enhance the clinical efficacy of cetuximab against lung, colorectal, and head and neck carcinomas. For example, clinical trials are underway using cetuximab in combination with β-glucans to treat colorectal cancer. β-glucans are polysaccharides that collaborate with complement-activating antibodies in the elimination of tumors [26]. Cetuximab may also be combined with other monoclonal antibodies to potentiate complement activation [17]. Modifications in the Fcregion of the monoclonal antibody can also be attempted   to increase the ability of the antibody's activation of complement [27].
In an unselected population, cetuximab treatment, as many other biological agents, only benefits a small percentage of NSCLC patients [28]. The body of knowledge that has paved the way for more targeted use of cetuximab has been recently reviewed [29]. Many groups are also focusing their research on markers of response (or resistance) to cetuximab. A more rational use of EGFRtargeted agents, such as cetuximab, should provide benefits for patients and health-care providers alike by sparing patients unnecessary treatment and allowing better use of health-care resources [29]. However, at present, no molecular biomarker is available to predict response to cetuximab in NSCLC. Based on our results, polymorphisms or other genetic traits that modulate the immune function may condition the activity of cetuximab, serving as markers to predict clinical response.

Conclusions
There is a pressing need to elucidate the molecular mechanisms involved in cetuximab activity in order to improve its clinical efficacy and to better select patients who would benefit from cetuximab treatment. In this paper we demonstrate that a complement-mediated immune response participates in the inhibition of tumor cell growth by cetuximab in vivo. This observation may have important clinical implications in the development of new cetuximab-based therapeutic strategies and in the identification of markers that predict clinical response. Figure 6 Effector mechanisms triggered by cetuximab. Cetuximab can inhibit EGFR-intracellular signaling or induce an immune response. These two antitumor mechanisms may act in conjunction. In certain conditions, such as in the presence of KRAS mutations, the immune response may be particularly important. The activation of the immune system by cetuximab can lead to both antibody-dependent cellular cytotoxicity (ADCC) and complement activation. Complement activation results in complement-dependent cellular cytotoxicity (CDCC) and/or complement-dependent cytotoxicity (CDC). CDCC is mediated by the interaction of C3b and its receptors in effector cells (e.i. phagocytes and NK cells). The release of anaphylatoxins, such as C5a, would further increase the recruitment and activation of effector cells. The three main immune consequences of complement activation are summarized in the boxes.