Mechanisms of Acquired Resistance to Cetuximab: Role of HER (Erbb) Family Members

Home , Cetuximab

Oncogene (2008) 27, 3944–3956 & 2008 Nature Publishing Group All rights reserved 0950-9232/08 $30.00 www.nature.com/onc ORIGINAL ARTICLE Mechanisms of acquired resistance to cetuximab: role of HER (ErbB) family members

DL Wheeler1, S Huang1, TJ Kruser, MM Nechrebecki, EA Armstrong, S Benavente, V Gondi, K-T Hsu and PM Harari2

Department of Human Oncology, University of Wisconsin School of Medicine and Public Health, Madison, WI, USA

The epidermal growth factor receptor (EGFR) is a central Introduction regulator of proliferation and progression in human cancers. Five EGFR inhibitors, two monoclonal antibodies The epidermal growth factor receptor (EGFR) is a and three TKIs, have recently gained FDA approval in member of the HER family of receptor tyrosine kinases oncology (cetuximab, panitumumab, erlotinib, gefitinib and consists of four members: EGFR (ErbB1/HER1), and lapatinib). These strategies of EGFR inhibition HER2/neu (ErbB2), HER3 (ErbB3) and HER4 demonstrate major tumor regressions in approximately (ErbB4). Stimulation of the receptor through ligand 10–20% of advanced cancer patients. However, many binding activates the intrinsic receptor tyrosine kinase tumors eventually manifest acquired resistance to treat- and promotes receptor homo- or heterodimerization ment. In this study we established and characterized a with HER family members. EGFR activation leads to model to study molecular mechanisms of acquired downstream stimulation of several signaling cascades, resistance to the EGFR monoclonal antibody cetuximab. including MAPK and PI(3)K/Akt that influence cell Using high-throughput screening we examined the activity proliferation, angiogenesis, invasion and metastasis of 42 receptor tyrosine kinases in resistant tumor cells (Citri and Yarden, 2006). Aberrant expression or following chronic exposure to cetuximab. Cells developing activity of EGFR is identified in many human epithelial acquired resistance to cetuximab exhibited increased cancers. steady-state EGFR expression secondary to alterations Targeting EGFR has been intensely pursued in the in trafficking and degradation. In addition, cetuximab- last decade as a cancer treatment strategy. One resistant cells manifested strong activation of HER2, approach uses monoclonal antibodies (mAbs) to target HER3 and cMET. EGFR upregulation promoted in- the extracellular domain of the EGFR to block natural creased dimerization with HER2 and HER3 leading to ligand binding (Mendelsohn, 2003). Cetuximab (IMC- their transactivation. Blockade of EGFR and HER2 led C225, Erbitux) prevents receptor activation and dimeri- to loss of HER3 and PI(3)K/Akt activity. These data zation and ultimately induces receptor internalization suggest that acquired resistance to cetuximab is accom- and downregulation (Sunada et al., 1986). Cetuximab panied by dysregulation of EGFR internalization/degra- exhibits promising antitumor activity as monotherapy dation and subsequent EGFR-dependent activation of or in combination with chemotherapy and/or radiation HER3. Taken together these findings suggest a rationale in head and neck cancer and metastatic colorectal cancer for the clinical evaluation of combinatorial anti-HER (Cunningham et al., 2004; Baselga and Arteaga, 2005; targeting approaches in tumors manifesting acquired Bonner et al., 2006; Mendelsohn and Baselga, 2006). resistance to cetuximab. A second approach involves the use of small molecule Oncogene (2008) 27, 3944–3956; doi:10.1038/onc.2008.19; tyrosine kinase inhibitors (TKIs) that bind to the ATP- published online 25 February 2008 binding site in the tyrosine kinase domain (TKD) of the EGFR and HER2. Three anti-EGFR TKIs, erlotinib Keywords: EGFR; cetuximab; acquired resistance (OSI-774, Tarceva), gefitinib (ZD1839, Iressa) and lapatinib (GW572016, Tykerb) are approved by the FDA for use in oncology. Both approaches to EGFR inhibition show consider- able clinical promise. However, increasing evidence suggests that patients who initially respond to EGFR inhibitors may subsequently become refractory (Pao Correspondence: Dr PM Harari, Department of Human Oncology, University of Wisconsin Comprehensive Cancer Center, 600 Highland et al., 2005). Therefore, an improved understanding of Avenue, Madison, WI 53792-0600, USA. molecular mechanisms of acquired resistance to EGFR E-mail: [email protected] inhibitors may provide valuable leads to enhance the 1These authors contributed equally to this work. efficacy of this class of agents. The identification of 2This author holds research and/or consulting agreements with Amgen, AstraZeneca, Genentech and ImClone. catalytic domain EGFR mutations that predict response Received 17 October 2007; revised 2 January 2008; accepted 8 January to EGFR-TKIs in selected lung cancer patients 2008; published online 25 February 2008 represents a landmark development in the EGFR field Acquired resistance to EGFR inhibition DL Wheeler et al 3945 (Sequist et al., 2007). Although EGFR TKD mutations activation of alternative RTKs that share overlapping appear to correlate with response to the TKIs erlotinib signal transduction elements with the EGFR. To test and gefitinib, no such correlation exists for cetuximab this hypothesis, we screened the activity of a panel of response (Mukohara et al., 2005). activated RTKs using an antibody-based array from To study mechanisms of acquired resistance to R&D Systems (Minneapolis, MN, USA) as shown in cetuximab, we established a series of cetuximab-resistant Figure 1c. Following quantification of scanned images clones in vitro following long-term exposure to cetuxi- using ImageQuant software, the relative expression of mab in nonsmall cell lung cancer (NSCLC; H226) and specific phosphorylated RTKs between cetuximab- head and neck squamous cell carcinoma (HNSCC; resistant and parental cells was determined (Figure 1d). SCC-1) cell lines. Following establishment of stable The identical experimental approach was performed clones, we performed high-throughput screening to using the SCC1 cetuximab-resistant lines and parental examine the activity of 42 membrane receptor tyrosine control (Figures 1e and f). From this high-throughput kinases (RTKs). Through comparative analysis of screen, several phosphorylated RTKs were notably cetuximab-resistant versus parental lines, we identified upregulated in both cetuximab-resistant NSCLC and that EGFR along with HER2, HER3 and cMET are all HNSCC tumor lines including HER family members highly activated in the resistant clones. Further studies (EGFR, HER2 and HER3) and the hepatocyte growth suggest that acquired resistance to cetuximab reflects factor receptors (HGFR, cMET). These results dysregulation of EGFR internalization/degradation and indicated that these independent tumor cell lines, subsequent EGFR-dependent activation of HER3. chronically challenged with cetuximab, manifested highly similar patterns of altered RTK expression and/ or activation. To validate results of the phospho-RTK array in Results individual cetuximab-resistant clones we performed standard western blot analysis on the parental and Establishment of cetuximab-resistant lines cetuximab-resistant clones of H226 to measure levels of We established cetuximab-resistant tumor cell lines EGFR, HER2, HER3, cMET and members of their using the human NSCLC line NCI-H226 (H226) and downstream signaling cascades, including the phos- the HNSCC line UMSCC-1 (SCC1) to use as a model phorylated forms of MAPK and Akt. The results system to elucidate molecular mechanisms of acquired demonstrated that findings from the phospho-RTK resistance to cetuximab. These lines were chosen based array were consistent in all of the cetuximab-resistant on three primary criteria: (1) cetuximab is used in clones (Figure 2a). Although the activity of EGFR, therapy for both tumor types, (2) the cell lines are HER2, HER3 and cMET was increased relative to the sensitive to cetuximab and (3) the cell lines have no parental line, only EGFR steady-state expression was TKD mutations. To generate resistant lines, H226 and dramatically increased in cetuximab-resistant clones. SCC1 cells were continuously exposed to increasing Furthermore, analysis of EGFR binding partners using concentrations of cetuximab over 6 months. Following immunoprecipitation techniques indicated that EGFR the development of heterogeneous populations of displayed increased binding with HER2, HER3 and cetuximab-resistant cells we isolated individual sub- cMET as compared to the cetuximab-sensitive parental clones of cetuximab-resistant lines. This process resulted cells (Figure 2b). Taken together these data suggest that in six stable resistant clones for the H226 NSCLC line prolonged exposure to cetuximab induces upregulation designated HC1, HC4, HC5, HC6, HC7 and HC8. The of EGFR with resultant enhanced heterodimerization of sensitive parental line was designated HP. For the SCC1 EGFR with HER2, HER3 and cMET. HNSCC line six stable resistant clones were generated The increased expression and heterodimerization (SC1, SC2, SC5, SC6, SC7, SC8). As shown in activity of EGFR correlates with increased baseline Figure 1a, all HC clones displayed a robust cetux- activity of HER2, HER3, MAPK and Akt relative to imab-resistant phenotype when challenged with increas- the parental line (HP; Figure 2c, control lanes). To ing concentrations of cetuximab as compared to examine if cetuximab treatment could inhibit the activity parental controls. Similar results were observed with of EGFR and subsequent HER2 and HER3 activity we the SCC1 cetuximab-resistant clones (Figure 1b). challenged cetuximab-resistant cells (clone HC4) with Sequence analysis of the EGFR TKD in H226 cells increasing concentrations of cetuximab (20–320 nM)and after the establishment of resistant clones indicated no measured the phosphorylation of EGFR at tyrosine mutations developed during the selection process in 1173, relative to parental cells (HP). The results either the resistant or parental cells (data not shown). indicated that cetuximab had no effect on the autophosphorylation of EGFR or the activity of HER2, HER3 Upregulation of EGFR and activation of HER2, HER3 or Akt. In addition, cetuximab was able to block EGF and cMet induced activation of EGFR (Figure 2c, EGF þ CTX After successful establishment of cetuximab-resistant lanes) indicating that cetuximab retains the capacity to clones, we performed high-throughput comparative bind the receptor and inhibit EGF binding in cetux- analyses measuring phosphorylated RTKs in the resis- imab-resistant cells. Furthermore, cetuximab blocked tant versus parental lines to test the hypothesis that MAPK activity in cetuximab-resistant cells indicating acquired resistance to EGFR inhibition results from the that MAPK may play a minor role in resistance.

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3946

Figure 1 Phospho-receptor tyrosine kinase (RTK) array in NSCLC H226 and HNSCC SCC1 cells demonstrate upregulation of EGFR, HER2, HER3 and cMET. (a) Cetuximab dose response curve using NSCLC line NCI-H226 cetuximab-resistant clones. Cells were treated with the indicated concentration of cetuximab and growth was measured using the growth proliferation assay as described in the experimental procedures and plotted as a percentage of growth relative to the untreated control cells. Data points are represented as mean±s.e.m. (n ¼ 3). (b) Cetuximab dose response curve using HNSCC line SCC-1 cetuximab-resistant clones. Cells were treated with the indicated concentration of cetuximab and growth was measured using the growth proliferation assay as described in the experimental procedures and plotted as a percentage of growth relative to the untreated control cells. Data points are represented as mean±s.e.m. (n ¼ 3). (c–f) HER family members have increased activity in HNSCC and NSCLC cetuximab-resistant cells. Cetuximab- sensitive (parental) and cetuximab-resistant cells were grown to conﬂuence followed by protein extraction. The cell extracts were incubated with membranes containing antibodies to 42 different RTKs. The membranes were washed and incubated with a pan antiphosphotyrosine antibody to measure the levels of active receptor. Quantitation of phospho-RTK was completed using scanned images from ImageQuant software. Data points are represented as mean±s.e.m. (n ¼ 3).

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3947

Figure 2 EGFR is upregulated in cetuximab-resistant cells. (a) Characterization of expression of HER family members and downstream Akt and MAPK in cetuximab-resistant clones (HC1, HC4-HC8). Protein was collected and fractionated by SDS– polyacrylamide gel electrophoresis (PAGE) followed by immunoblotting for the indicated proteins. a-tubulin was used as loading control. HP; cetuximab-sensitive parental line, HC; cetuximab-resistant clones. (b) EGFR has increased association with HER2, HER3 and cMET in cetuximab-resistant cells. Cetuximab-resistant cells were harvested and EGFR was immunoprecipitated from the cetuximab-resistant clone HC4 and the parental HP cells with an anti-EGFR antibody. The immunoprecipitates were fractionated on SDS–(PAGE) followed by immunoblotting for the indicated proteins. (c) Cetuximab treatment does not modulate EGFR phosphorylation in cetuximab-resistant clone (HC4). Parental cells (HP) and cetuximab-resistant clone (HC4) were treated with vehicle (control) or increasing concentrations of cetuximab (CTX) for 24 h. Stimulation was with EGF (10 ng mlÀ1) and HRG (5 mM) for 45 min. Protein was collected and fractionated by SDS–(PAGE) followed by immunoblotting for the indicated proteins. a-Tubulin was used as a loading control.

Impaired endocytosis of EGFR by flow cytometry confirmed these findings and showed To examine underlying mechanisms of elevated EGFR that parental lines manifested rapid internalization of expression and activation in cetuximab-resistant cells we the EGFR following EGF stimulation with return to hypothesized that alterations in receptor trafficking may baseline within 24 h (Figure 3b). However, cetuximab- be involved. To test this hypothesis we analysed patterns resistant clones showed no receptor internalization in of ligand-stimulated EGFR internalization in resistant response to EGF stimulation. These results suggest that cells relative to parental controls. Using immunofluore- cetuximab-resistant cells have impaired ability to inter- scent staining, we found an accumulation of intra- nalize EGFR and this mechanism may contribute to the cellular endocytic vesicles staining for EGFR in parental increased expression and activity of EGFR in resistant cells following 45 min of EGF stimulation. However, cells. EGFR remained in the extracellular membrane and was To further investigate if cetuximab-resistant cells have not internalized in cetuximab-resistant cells, as shown in impaired ability to degrade EGFR, we examined Cbl- Figure 3a. Furthermore, quantifying the surface EGFR mediated ubiquitinylation of EGFR that is required for

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3948

Figure 3 Cetuximab-resistant cells have increased EGFR plasma membrane expression. (a) EGFR is internalized following EGF stimulation in parental (HP), but not in cetuximab-resistant cells (HC4). Internalization of EGFR was determined by immunofluorescent staining as described in Experimental procedures using FITC-conjugated anti-EGFR antibody 45 min following EGF (25 ng mlÀ1) stimulation. Internalized EGFR was indicated by the accumulation of intracellular endocytic vesicles staining for EGFR. (b) Cell surface EGFR is decreased following EGF stimulation in parental (HP), but not in cetuximab-resistant cells (HC4). The levels of EGFR on the cell surface were monitored following 0–24 h of EGF (25 ng mlÀ1) stimulation by flow cytometry as described in Experimental procedures. The numbers represent mean fluorescence intensity of anti-EGFR staining after subtracting IgG background fluorescence. (c) Lack of c-Cbl recruitment and EGFR ubiquitination in cetuximab-resistant cells following EGF stimulation. Cetuximab-resistant (HC4) and parental (HP) cells were either untreated (À) or treated with EGF (25 ng mlÀ1) for 45 min. Thereafter, cell lysates were prepared and analysed by immunoprecipitation (IP) and immunoblotting (IB) with the indicated antibodies.

lysosomal degradation of the receptor. Upon EGF subsequent activity of EGFR ultimately sending signals stimulation, the active EGFR recruits c-Cbl, an E3 to proliferative and anti-apoptotic pathways. To ubiquitin ligase, to tyrosine 1045. This interaction examine this hypothesis we challenged cetuximab- enhances receptor ubiquitination and thus directs the resistant cells (HC1, HC4, HC8) with a battery of small receptor to lysosomal degradation (Grovdal et al., molecule TKIs (Figure 4a), including erlotinib, gefitinib 2004). We examined this hypothesis by immuno- and the pan-HER irreversible inhibitor CI-1033. All precipitating EGFR and found no detectable levels of TKIs decreased EGFR phosphorylation at tyrosine c-Cbl associated with EGFR in cetuximab-resistant cells 1173. This downregulation of active EGFR correlated whereas the parental lines showed strong association of with a robust decrease in activity of HER2, HER3, c-Cbl after EGF stimulation (Figure 3c). Loss of cMET, Akt and MAPK. Active HER2, HER3 and Akt c-Cbl association with the EGFR led to significantly were undetectable by western blot in the parental lines. less EGFR ubiquitination after EGF stimulation in the To determine if TKIs had an impact on proliferative cetuximab-resistant versus parental cells. Sequence potential, we performed growth proliferation assays analysis of the c-Cbl binding site (Y1045) did not comparing parental and cetuximab-resistant clones indicate its mutations nor of its surrounding nucleotides. challenged with increasing concentrations of each TKI. However, Y1045 exhibited increased baseline and EGF This treatment led to profound growth inhibitory effects stimulated phosphorylation relative to the parental lines in cetuximab-resistant clones, responding in parallel to (data not shown). Therefore loss of c-Cbl binding their parental counterparts (Figure 4b). Recognizing the to EGFR in cetuximab-resistant cells must occur by relative nonspecificity of EGFR TKIs, and to examine if other mechanisms. Together these data suggest that EGFR alone was responsible for transphosphorylation impaired degradation may contribute to the increased of HER2, HER3 and cMET, we used siRNAs directed expression and activity of EGFR in cetuximab-resistant against the EGFR. Cetuximab-resistant cells treated cells and may be a potential mechanism of cetuximab with EGFR siRNA showed a dose dependent down- resistance. regulation of HER2, HER3 and cMET activity (Figure 5a) and this correlated with a strong inhibitory impact on the proliferative potential of the cetuximab- EGFR-dependent activation of HER2, HER3 and cMet resistant clones (Figure 5b). Taken together these data We hypothesized that impaired endocytosis of EGFR suggest that EGFR is responsible for transphosphoryla- was, in part, responsible for the increased levels and tion of HER2, HER3 and cMET. Furthermore, a loss of

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3949

Figure 4 TKIs retain capacity to inhibit HER signaling in cetuximab-resistant clones. (a) HER family members and downstream signaling molecules are attenuated after treatment with TKIs in cetuximab-resistant clones. Parental cells (HP) and cetuximab-resistant clones (HC1, HC4 and HC8) were treated with vehicle (DMSO), cetuximab (100 nM), erlotinib (10 mM), gefitinib (1 mM) or CI- 1033(500 nM) for 24 h. Protein lysates were fractionated on SDS–(PAGE) followed by western blotting for the indicated proteins. a-Tubulin was used as a loading control. (b) TKIs inhibit proliferation in cetuximab-resistant clones. Parental cells (HP) and cetuximab-resistant clones (HC1, HC4, HC8) were plated and allowed to adhere overnight. Growth was measured at 72 h after drug treatment using the growth proliferation assay as described in the experimental procedures and plotted as a percentage of growth relative to the untreated control cells. Treatment groups were cetuximab, erlotinib, gefitinib, CI-1033. Data points are represented as mean±s.e.m. (n ¼ 6). the activity of these molecules results in decreased (Figure 2a). To examine if increased cMET activity proliferative potential in cetuximab-resistant cells. resulted in cetuximab resistance, we used a selective small molecule cMET inhibitor, PHA665752 (PHA) to Inhibition of cMET activity in cetuximab-resistant cells inhibit cMET activity (Figure 6a; Christensen et al., Active cMET was detected in our preliminary studies 2003). Cetuximab-resistant cells (clone HC4) were examining alternative RTK activation in cells with treated with PHA and stimulated HGF prior to protein acquired resistance to cetuximab (Figure 1c) and these collection. PHA exhibited potent inhibition of cMET results were further verified by analysis of individual autophosphorylation (Y1230/1234/1235) but had no cetuximab-resistant clones for cMET activity effect on downstream signaling molecules Akt and

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3950 were transfected with the expression vector and treated with PHA665752 or in combination with cetuximab. The loss of cMET expression did not affect the proliferative potential of cetuximab-resistant cells, nor resensitize cells to the antiproliferative effect of cetuximab. Taken together these results suggest that despite upregulation, cMET activity does not signi- ﬁcantly contribute to the ability of cetuximab-resistant clones to proliferate in the presence of cetuximab.

Inhibition of HER2 or HER3 activity To further elucidate the roles of HER family members in acquired resistance to cetuximab we examined combinations of HER targeting agents. HER2 and HER3 signal primarily when heterodimerized to other HER family members, and HER2 is a preferred heterodimerization partner for HER3 (Citri et al., 2003). Therefore, we hypothesized that HER2/HER3 heterodimers may contribute to the cetuximab-resistant phenotype. To test this hypothesis we used 2C4, a monoclonal antibody that inhibits HER2 heterodimerization (Franklin et al., 2004). Parental and cetuximab-resistant cells (clone HC4 and HC8) were treated with cetuximab, 2C4 or the combination. Cetuximab alone had modest effects on HER2, HER3 or Akt activity whereas 2C4 treatment decreased activity of HER2, HER3 and Akt in resistant cells (Figure 7a). However, combining cetuximab with 2C4 a more potent suppression of HER3 and Akt was observed and this correlated with strong antiproliferative effects (Figure 7b). These data suggest that HER3 activity, dependent on EGFR and HER2, promotes survival in cetuximab-resistant cells. We find HER3 to be highly activated in our RTK Figure 5 HER2 and HER3 activity in cetuximab-resistant cells is array screening evaluation of cetuximab-resistant cells EGFR-dependent. (a) Effects of EGFR knockdown on HER2 and (Figure 1c). Indeed, recent reports suggest that trans- HER3 activity. HC4 cells were plated and treated with the vehicle DMSO (control), 1 mM erlotinib, 1 mM gefitinib, 1 nM of EGFR activation of HER3 is important in conferring resistance siRNA or 5 nM EGFR siRNA. 48 h later protein was collected and to EGFR inhibitors (Engelman et al., 2007; Sergina fractionated by SDS–(PAGE) and immunoblotted for the indicated et al., 2007). To examine whether the loss of HER3 proteins. (b) Effects of EGFR knockdown on proliferation of could resensitize cetuximab-resistant cells to cetuximab cetuximab-resistant cells. Cetuximab-resistant clone HC4 was we used siRNAs against HER3 (Figures 7c and d). plated and treated with control (DMSO), 100 nM cetuximab, 1 mM erlotinib, 1 mM gefitinib or 25 nM of EGFR siRNA. Seventy- These findings indicate that the loss of HER3 can two hours later growth proliferation was measured as outlined in resensitize cetuximab-resistant cells to cetuximab the experimental procedures. *Po0.05. therapy. Taken together these results indicate that cells may overcome EGFR inhibition by cetuximab through transactivation of HER2 and HER3 in an EGFR- MAPK activities. To determine the effects of PHA on dependent matter, and that HER3 signaling may proliferation we performed growth proliferation assays represent a critical signaling mediator in acquired on cetuximab-resistant clones using PHA alone or in resistance to cetuximab therapy (Figure 8). combination with cetuximab (Figure 6b). The parental cells responded briskly to cetuximab challenge but had no response to PHA. Furthermore, all cetuximab- resistant clones showed cetuximab resistance and no Discussion response to either PHA alone or in combination with cetuximab. This combination treatment had no effect on The development of acquired resistance to cancer the activity of HER3 or Akt (Figure 6c). To expand therapeutic drugs represents a major obstacle to upon these series of tests we designed small hairpin successful treatment. Tempering promise for the EGFR RNA (shRNA) vectors directed against the cMET open molecular targeting agents is the emergence of reports reading frame to knockdown the expression levels of that illustrate acquired resistance. It is therefore cMET in cetuximab-resistant cells (Figure 6d). Growth important to understand mechanisms of acquired proliferation assays were performed after transfection of resistance to anti-EGFR therapy in an effort to avoid, the cMET-shRNA expression vectors (Figure 6e). Cells overcome or diminish this phenomenon. In the current

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3951

Figure 6 Inhibition of cMET activity does not resensitize cetuximab-resistant cells to cetuximab. (a) PHA665752 inhibits cMET activity without affecting Akt or MAPK signaling. Cetuximab-resistant clone HC4 was treated with PHA665752 for 4, 8, 24 or 48 h. Cells were then stimulated with HGF for 45 min. Western blot shows inhibition of cMET activity up to 24 h when treated with PHA665752. (b) PHA665752 alone or in combination with cetuximab does not inhibit growth of cetuximab-resistant cells. Growth was measured at 72 h after drug treatment using the growth proliferation assay as described in the experimental procedures and plotted as a percentage of growth relative to the untreated control cells. Treatment groups were cetuximab (CTX), PHA665752 (PHA) or the combination of the two drugs. HP, cetuximab-sensitive parental line; HC, cetuximab-resistant clones. Data points are represented as mean±s.e.m. (n ¼ 3). (c) Western blotting of cetuximab-resistant clone (HC4) cells after treatment with cetuximab, PHA665752 or the combination. Cetuximab-resistant clone HC4 was treated with cetuximab (100 nM), PHA665752 (100 nM) or the combination. 24 h later protein was collected and fractionated by SDS–(PAGE) and immunoblotted for the indicated proteins. (d) Western blotting of cetuximab-resistant clone (HC4) cells after transient transfection with cMET shRNA vectors. Control is empty vector control; 1, 2 and 4 mg of vector were transfected. Cells were collected 48 h after treatment. Forty-eight hours after transfection protein was collected and fractionated by SDS–(PAGE) and immunoblotted for the indicated proteins. (e) Loss of cMET does not affect growth of cetuximab- resistant cells (HC4). Cells were transfected with cMET shRNA vectors followed by treatment with vehicle (shRNA only), PHA665752 (shRNA þ PHA group) or PHA665752 or cetuximab (shRNA þ PHA þ CTX group). Growth was measured at 72 h after treatment using the growth proliferation assay as described in the experimental procedures and plotted as a percentage of growth relative to the untreated control cells. Data points are represented as mean±s.e.m. (n ¼ 6). study, we investigate mechanisms of acquired resistance the EGFR, no such correlation between EGFR muta- to cetuximab using newly established cetuximab-resis- tion and cetuximab resistance exists. Our results tant tumor clones following long-term exposure to implicate a novel EGFR-dependent mechanism in cetuximab. Using high-throughput screening to examine acquired resistance to EGFR-targeted monoclonal the activity of 42 RTKs, we found that EGFR, HER2, antibody therapy. Evidence suggests that loss of c-Cbl- HER3 and cMET were highly activated in cetuximab- regulated EGFR degradation contributes to EGFR resistant clones derived from either HNSCC or NSCLC overexpression in cancer cells. Several recent reports lines (Benavente et al., 2004; Gondi et al., 2006). We directly implicate c-Cbl in regulation of the EGFRvIII present data to suggest that acquired resistance to mutant (Han et al., 2006; Grandal et al., 2007). These cetuximab reﬂects dysregulation of EGFR internaliza- reports demonstrate that hypophosphorylation of re- tion/degradation and subsequent EGFR-dependent sidue Y1045 of the EGFRvIII leads to defective activation of HER2 and HER3. Furthermore, it appears downregulation mediated by c-Cbl. Other studies that HER3 activity, dependent on EGFR and HER2, examining EGFR TKD mutants in NSCLC indicate represents a critical step for cells to escape from that the TKI-sensitizing mutation EGFRD746–750 cetuximab inhibition. reduces phosphorylation at Y1045 and may in part be Although acquired resistance to TKI therapy has responsible for increased signaling from this mutant been attributed to mutations in the cytoplasmic tail of receptor (Furukawa et al., 2007). Consistent with these

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3952

Figure 7 Inhibition of HER2 or HER3 activity sensitizes cetuximab-resistant cells. (a) Combinatorial treatment of cetuximab- resistant clones HC4 and HC8 with cetuximab and 2C4 leads to loss of HER3 and AKT activity. Parental cells (HP) and cetuximab- resistant clones HC4 and HC8 were treated with cetuximab (CTX), 2C4 (2C4) or the combination. Protein lysates were fractionated on SDS–(PAGE) followed by western blotting for the indicated proteins. a-Tubulin was used as a loading control. (b) Combinatorial treatment of cetuximab-resistant clones with cetuximab and 2C4 leads to growth inhibition. Parental cells (HP) and cetuximab- resistant clones (HC4, HC8) were plated and allowed to adhere overnight. Growth was measured at 72 h after drug treatment using the growth proliferation assay as described in the experimental procedures and plotted as a percentage of growth relative to the untreated control cells. Treatment groups were cetuximab (100 nM), 2C4 (100 nM) or the combination. Data points are represented as mean±s.e.m. (n ¼ 6). *Po0.05. (c) Western blotting of cetuximab-resistant clone (HC4) cells after transient transfection with HER3 siRNA. Treatments are scrambled siRNA (50 nM), cetuximab (100 nM), HER3 siRNA (50 nM) or the combination. Forty-eight hours after treatment protein was collected and fractionated by SDS–(PAGE) and immunoblotted for the indicated proteins. (d) HER3 knockdown resensitizes cetuximab-resistant cells to cetuximab. Cells were transfected with scrambled (50 nM, control), 100 nM cetuximab, HER3 (50 nM) siRNA in combination with 100 nM cetuximab. Growth was measured at 72 h after treatment using the growth proliferation assay as described in the experimental procedures and plotted as a percentage of growth relative to the untreated control cells. Data points are represented as mean±s.e.m. (n ¼ 6). *Po0.05.

findings our data suggests that cells chronically exposed and Bottaro, 2006). In addition, cMET has recently to cetuximab have dysregulated c-Cbl-mediated EGFR been implicated in acquired resistance to gefitinib. degradation (Figure 3c). These results suggest that Engelman et al. (2007) reported that NSCLC HCC827 increased expression and activity of the EGFR in cells chronically exposed to gefitinib in vitro led to the cetuximab-resistant cells may be caused, in part, by amplification of cMET resulting in the constitutive dysregulated receptor internalization and sorting. activation of the HER3-Akt signaling pathway. Unlike Increasing evidence identifies cMET as an attractive the above findings, we did not observe protein target for molecularly targeted cancer therapy (Peruzzi amplification of cMET expression (Figure 2a). We

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3953

Figure 8 Model of acquired resistance to cetuximab and therapeutic strategies to overcome resistance. Chronic exposure to cetuximab leads to dysregulation of c-Cbl mediated internalization and degradation of the EGFR. This increased steady-state expression of EGFR serves to bind and activate HER2 or HER3 and thereby maintain signaling to MAPK and Akt pathways in the presence of cetuximab. TKIs targeting EGFR and HER2 lead to blockade of HER3 transactivation and subsequent pro-survival signals to Akt. In addition, combinatorial monoclonal antibody therapy using cetuximab and 2C4 leads to inactivation of HER3 and downregulation of its signaling pathways. found that cetuximab-resistant clones had constitutively cetuximab-sensitivity. Although increased EGFR activ- phosphorylated cMET in the absence of its ligand, ity in cetuximab-resistant cells results in increased HGF. This overactivity of cMET appeared to be dimerization and activation of HER2 and HER3, the dependent on dimerization with EGFR and an increased role in which HER family ligands plays a role in association of cMET with the EGFR was demonstrated resistance remains to be investigated. These findings are by co-immunoprecipitation (Figure 2b). TKIs and consistent with several recent reports that suggest HER3 siRNA directed against the EGFR led to a dramatic plays an important role in regulating tumor response to decrease in cMET phosphorylation (Figures 4a and 5a). EGFR or HER2 targeting agents. Erjala et al. (2006) Although EGFR TKIs lead to the decrease of cMET, demonstrated a correlation between HER3 expression MAPK and Akt activities, small molecule inhibitors and gefitinib resistance. Similarly, Fujimoto et al. (2005) targeting cMET directly (PHA665752) did not lead to pointed to HER3 as a potentially important mediator of the loss of MAPK or Akt, two critical downstream EGFR inhibition by gefitinib in NSCLC cells. In HER2- molecules of the EGFR and cMET signaling pathways. driven breast cancer cells, Sergina et al. (2007) demon- Furthermore, PHA could not resensitize cetuximab- strated that activation of HER3 and downstream PI3K/ resistant cells to cetuximab therapy. These data indicate AKT signaling can evade the inhibition by gefitinib both that although EGFR heterodimerizes with and activates in vitro and in vivo. Furthermore, knocking down HER3 cMET, this may not represent a critical determinant of signaling by siRNA restores potent pro-apoptotic acquired resistance to cetuximab. The specific conse- activity of gefitinib in breast cancer cells. Zhou et al. quences of the EGFR/cMET interaction in acquired (2006) reported that heregulin-dependent HER3 activa- resistance to cetuximab remain to be defined. tion contributes to gefitinib resistance in NSCLC A549 Another major consequence of the increased EGFR cells. Inhibiting heregulin release by targeting ADAM activity in cetuximab-resistant cells is the binding and proteases restored gefitinib sensitivity. Collectively, activation of HER2 and HER3 (Figure 2). HER2 these studies indicate that HER3 mediates tumor activity is EGFR dependent in these cells, as siRNA response to EGFR TKIs and activation of HER3 against EGFR can knock down HER2 activity signaling help cells to escape the antitumor effects of (Figure 5a). However, it appears that once active, these agents. Our data indicate that cetuximab resistance HER2 can transactivate HER3, as the HER2 hetero- appears to be dependent on HER3 signaling. Taken dimerization inhibitor 2C4 partially inhibits HER3 together these data suggest that HER3 signaling may be signaling (Figure 7a). Furthermore, it appears that a critical shared pathway for acquired resistance to HER3 signaling is more crucial, as our results cetuximab or TKIs directed against the EGFR. (Figure 7d) demonstrate that direct targeting of HER3 A potential limitation of the current study is by siRNA in cetuximab-resistant cells can restore that cetuximab-resistant cells were developed in vitro.

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3954 Previous studies suggest that altered control of angio- were obtained from Life Technologies Inc. genesis may play a role in acquired resistance to (Gaithersburg, MD, USA). cetuximab in vivo (Ciardiello et al., 2004). Nevertheless, the current findings are highly consistent with recent Compounds reports of in vivo acquired resistance to the HER2 Gefitinib (ZD1839, Iressa) was provided by AstraZeneca targeting antibody herceptin. Ritter et al. (2007) indicate (Macclesfield, UK). Erlotinib (OSI-774, Tarceva) was that cells chronically challenged with herceptin exhibit provided by Genentech Inc. (San Francisco, CA, USA). elevated levels of activated EGFR as well as increased Cetuximab (C225, Erbitux) was provided by ImClone EGFR/HER2 dimers. Further, their results imply that Systems Inc. (New York, NY, USA). 2C4 was provided cells resistant to herceptin retain a brisk antiproliferative by Genentech Inc.. CI-1033 and PHA665752 were response to EGFR TKIs. These corroborative findings provided by Pfizer Global Research Inc. (Ann Arbor, suggest that the development of acquired resistance to MI, USA). HER family targeting antibodies, in vitro or in vivo, result in a dysregulation and increased dependence on HER family signaling. Antibodies and proteins The findings herein suggest potential clinical implica- All antibodies were purchased from commercial sources tions when considering optimal cancer therapeutic as indicated below: AKT, EGFR, p-EGFR (Tyr1173), strategies. In this model we demonstrate that acquired HER3, MET and HRP-conjugated goat-anti-rabbit resistance to the mAb cetuximab occurs via an EGFR- IgG, goat-anti-mouse IgG and donkey-anti-goat IgG dependent mechanism. Therefore, targeting the EGFR antibodies were obtained from Santa Cruz Biotechno- by another mechanism, for example, with EGFR TKIs logy Inc. (Santa Cruz, CA, USA). pHER2(Y1221/1222), such as gefitinib and erlotinib, may retain effectiveness pHER3(Y1289), pAKT(S473), p-MAPK (Thr202/ for tumors manifesting cetuximab resistance (Figure 4; Tyr204) and c-Cbl were obtained from Cell Signaling Huang et al., 2004; Matar et al., 2004). Furthermore, Technology (Beverly, MA, USA). HER2 was purchased EGFR-dependent transactivation of HER2 and HER3 from Neomarker (Fremont, CA, USA), pCMET appears to play an important role in resistance to (Y1230/1234/1235) from Biosource International cetuximab since the HER2 heterodimerization inhibiting (Camarillo, CA, USA), tubulin from Calbiochem (San antibody 2C4 can resensitize resistant cells to cetuximab Diego, CA, USA). Purified recombinant EGF and HRG (Figures 7a and b). Several, clinical trials utilizing were purchased from R&D Systems. Annexin V-FITC combination therapy approaches directed against HER apoptosis detection kit was obtained from BD Bios- family members are in progress (Reid et al., 2007), and ciences Pharmingen (San Diego, CA, USA). Anti- our results suggest this may be an effective strategy to ubiquitin and all other chemicals were purchased from prevent or overcome acquired resistance to cetuximab. Sigma (St Louis, MO, USA). In conclusion this model system of acquired resistance to cetuximab identifies candidate RTKs that contribute Establishment of acquired resistance to EGFR inhibitors to the underlying molecular mechanisms of acquired Over a period of 6 months, tumor cells in culture were resistance. The findings presented suggest that chronic continuously exposed to increasing concentrations of therapy with cetuximab results in dysregulation of cetuximab. Commencing with the IC50 of cetuximab, EGFR processing and subsequent activation of HER2, the exposure dose was progressively doubled every HER3, cMET, MAPK and Akt. Acquired resistance to 10–14 days until eight dose doublings had been cetuximab may result from the activation of RTKs, such successfully achieved. In parallel, controlled parental as HER2 and HER3, which share overlapping prolif- cells were exposed to the phosphate-buffered saline erative and survival signaling pathways. These findings (PBS) vehicle. The established resistant cell lines were carry clinical implications and suggest that novel then maintained in continuous culture with the maxi- combinations of molecularly targeted therapies blocking mally achieved dose of cetuximab. both EGFR and other HER family members may provide a strategy to overcome acquired resistance and enhance the effectiveness of EGFR-targeted therapy. Phospho-receptor tyrosine kinase array We screened the activity of a panel of RTKs using an antibody-based array from R&D Systems. This phospho-RTK array specifically screens for membrane- Experimental procedures associated proteins/receptors. Antibodies against 42 different RTKs were prespotted in duplicate on nitro- Cell lines cellulose membranes, and cell lysates from our EGFR- The human NSCLC line NCI-H226 was obtained from inhibitor-resistant or parental cells were incubated with ATCC (Manassas, VA, USA) and maintained in 10% the membrane. Thereafter, a pan antiphosphotyrosine fetal bovine serum (FBS) in RPMI. UM-SCC1 cells antibody was used to detect the phosphorylated tyrosine were obtained from University of Michigan and on activated RTKs. Following quantification of scanned maintained in 10% FBS in Dulbecco’s modified images using ImageQuant software the relative expres- Eagle’s medium (DMEM) supplemented with 1% sion of specific phosphorylated RTKs between cetux- hydrocortisone. Cell culture media and supplements imab-resistant and parental H226 cells was determined.

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3955 Cell proliferation assay with PBS, cells were incubated with FITC-conjugated Exponentially growing cells were seeded in six well antibody for 30 min followed by flow cytometry plates. Following treatment, monolayers were washed analysis. with PBS and fixed/stained with 0.5% crystal violet. Plates were air dried overnight and dye was eluted with Immunofluorescent staining of EGFR 0.1 M sodium citrate (pH 4.2) in ethanol (1:1). Elution Cells (5000 cells per well) were seeded on a four-well was transferred to 96 well plates, and the absorbance glass chamber slide (Nalgene Nunc, Naperville, IL, was read at 540 nm to determine cell growth. USA). Following EGF stimulation, cells were washed with PBS and fixed with 4% paraformaldehyde for Immunoblotting analysis 10 min. Cells were permeabilized with PBS containing Following treatment, cells were lysed with Tween-20 0.1% Triton-X 100 and 1% BSA for 20 min and blocked lysis buffer (50 mM HEPES, pH 7.4, 150 mM NaCl, with 3% BSA in PBS fro 15 min. Cells were incubated 0.1% Tween-20, 10% glycerol, 2.5 mm EGTA, 1 mM with anti-EGFR antibody (1:100) for 2 h followed by À1 EDTA, 1 mM DTT, 1 mM PMSF and 10 mgml of 30 min FITC-conjugated antibody. Following washes leupeptin and aprotinin), sonicated and protein was with PBS, slides were mounted using ProLong antifade quantitated using a standard Bradford absorbance mounting solution (Molecular Probes, Carlsbad, CA, assay. Equal amounts of protein were fractionated USA). Fluorescence microscopy and photography were by SDS–PAGE. Thereafter, proteins were transferred carried out using FITC filter array on an Olympus to PVDF membrane and analysed by incubation with fluorescent microscope (Scope BX51). the appropriate primary antibody. Proteins were detected via incubation with HRP-conjugated secondary antibodies and ECL chemiluminescence detection RNA interference studies system. All siRNAs were obtained from Dharmacon (Lafayette, CO, USA). Transfection of Dharmacon siRNA into cell Immunoprecipitation lines was achieved by using Dharmafect 1 reagent. Cells were lysed with NP-40 lysis buffer (50 mM HEPES (pH 7.4), 150 mM NaCl, 1% NP-40, 0.5% deoxycholic acid, 10% glycerol, 2.5 mM EGTA, 1 mM EDTA, 1 mM Abbreviations DTT, 1 mM PMSF and 10 mgmlÀ1 of leupeptin and aprotinin). Cell lysates containing 0.5 mg of protein were DMSO, dimethyl sulfoxide; EGFR, epidermal growth factor incubated overnight with 2 mg of anti-EGFR antibody receptor; ELISA, enzyme-linked immunosorbent assay; HGF, followed by 30 ml of proteinA/G agarose beads for 2 h. hepatocyte growth factor; mAb, monoclonal antibody; MAPK, mitogen-activated protein kinase; NSCLC, nonsmall The immunoprecipitates were pelleted and washed three 0 times with lysis buffer and three times with PBS. The cell lung cancer; PI(3)K, phosphatidylinositol 3 -kinase; RTK, captured immunocomplexes were then boiled in 2 receptor tyrosine kinase; SCC, squamous cell carcinoma; Â TKD, tyrosine kinase domain; TKI, tyrosine kinase inhibitor. SDS sample buffer for 5 min and subjected to immuno- blot analysis as described above. Acknowledgements Detection of surface EGFR level by flow cytometry We thank AstraZeneca (gefitinib), Genentech (erlotinib, 2C4), Following treatment, cells were washed in PBS ImClone (cetuximab), OSI (erlotinib) and Pfizer Global and harvested by trypsinization. A cell suspension Research (PHA665752 and CI-1033) for supplying EGFR- 5 containing 1 Â 10 cells in 100 ml PBS containing 1% and cMET-targeting agents. This work was supported in part BSA was incubated with 2 mg of either control IgG or by NIH R01 CA 113448-01 (PMH) and American Cancer anti-EGFR antibodies for 1 h at 4 1C. Following washes Society Postdoctoral Fellowship (DLW).

References

Baselga J, Arteaga CL. (2005). Critical update and emerging trends in Ciardiello F, Bianco R, Caputo R, Caputo R, Damiano V, Troiani T epidermal growth factor receptor targeting in cancer. J Clin Oncol et al. (2004). Antitumor activity of ZD6474, a vascular endothelial 23: 2445–2459. growth factor receptor tyrosine kinase inhibitor, in human cancer Benavente S, Huang S, Armstrong E, Chi A, Chinnaiyan P, cells with acquired resistance to antiepidermal growth factor Harari PM. (2004). Establishment of acquired resistance to receptor therapy. Clin Cancer Res 10: 784–793. epidermal growth factor receptor (EGFR) inhibitors in human Citri A, Skaria KB, Yarden Y. (2003). The deaf and the dumb: the tumor cell lines. AACR Meeting Abstracts 2004: 1230a. biology of ErbB-2 and ErbB-3. Exp Cell Res 284: 54–65. Bonner JA, Harari PM, Giralt J, Azarnia N, Shin DM, Cohen RB Citri A, Yarden Y. (2006). EGF-ERBB signalling: towards the systems et al. (2006). Radiotherapy plus cetuximab for squamous-cell level. Nat Rev Mol Cell Biol 7: 505–516. carcinoma of the head and neck. N Engl J Med 354: 567–578. Cunningham D, Humblet Y, Siena S, Khayat D, Bleiberg H, Christensen JG, Schreck R, Burrows J, Kuruganti P, Chan E, Santoro A et al. (2004). Cetuximab monotherapy and cetuximab Le P et al. (2003). A selective small molecule inhibitor of c-Met plus irinotecan in irinotecan-refractory metastatic colorectal cancer. kinase inhibits c-Met-dependent phenotypes in vitro and N Engl J Med 351: 337–345. exhibits cytoreductive antitumor activity in vivo. Cancer Res 63: Engelman JA, Zejnullahu K, Mitsudomi T, Song Y, Hyland C, 7345–7355. Park JO et al. (2007). MET ampliﬁcation leads to geﬁtinib resistance

Oncogene Acquired resistance to EGFR inhibition DL Wheeler et al 3956 in lung cancer by activating ERBB3 signaling. Science 316: (ZD1839) and the monoclonal antibody cetuximab (IMC-C225): 1039–1043. superiority over single-agent receptor targeting. Clin Cancer Res 10: Erjala K, Sundvall M, Junttila TT, Zhang N, Savisalo M, Mali P et al. 6487–6501. (2006). Signaling via ErbB2 and ErbB3 associates with resistance Mendelsohn J. (2003). Antibody-mediated EGF receptor blockade as and epidermal frowth factor receptor (EGFR) amplification with an anticancer therapy: from the laboratory to the clinic. Cancer sensitivity to EGFR inhibitor gefitinib in head and neck squamous Immunol Immunother 52: 342–346. cell carcinoma cells. Clin Cancer Res 12: 4103–4111. Mendelsohn J, Baselga J. (2006). Epidermal growth factor receptor Franklin MC, Carey KD, Vajdos FF, Leahy DJ, De Vos AM, targeting in cancer. Semin Oncol 33: 369–385. Sliwkowski MX. (2004). Insights into ErbB signaling from the Mukohara T, Engelman JA, Hanna NH, Yeap BY, Kobayashi S, structure of the ErbB2-pertuzumab complex. Cancer Cell 5: Lindeman N et al. (2005). Differential effects of gefitinib and 317–328. cetuximab on non-small-cell lung cancers bearing epidermal growth Fujimoto N, Wislez M, Zhang J, Iwanaga K, Dackor J, Hanna AE factor receptor mutations. J Natl Cancer Inst 97: 1185–1194. et al. (2005). High expression of ErbB family members and their Pao W, Miller VA, Politi KA, Riely GJ, Somwar R, Zakowski MF ligands in lung adenocarcinomas that are sensitive to inhibition of et al. (2005). Acquired resistance of lung adenocarcinomas to epidermal growth factor receptor. Cancer Res 65: 11478–11485. gefitinib or erlotinib is associated with a second mutation in the Furukawa M, Nagatomo I, Kumagai T, Yamadori T, Takahashi R, EGFR kinase domain. PLoS Med 2: 1–11. Yoshimura M et al. (2007). Gefitinib-sensitive EGFR lacking Peruzzi B, Bottaro DP. (2006). Targeting the c-Met signaling pathway residues 746–750 exhibits hypophosphorylation at tyrosine residue in cancer. Clin Cancer Res 12: 3657–3660. 1045, hypoubiquitination, and impaired endocytosis. DNA Cell Biol Reid A, Vidal L, Shaw H, De Bono J. (2007). Dual inhibition of 26: 178–185. ErbB1 (EGFR/HER1) and ErbB2 (HER2/neu). Eur J Cancer 43: Gondi V, Huang S, Benavente S, Armstrong E, Harari PM. (2006). 481–489. Potential mechanisms of acquired Resistance to EGFR inhibitors. Ritter CA, Perez-Torres M, Rinehart C, Guix M, Dugger T, Engelman AACR Meeting Abstracts 2006: 294b. JA et al. (2007). Human breast cancer cells selected for resistance to Grandal MV, Zandi R, Pedersen MW, Willumsen BM, van Deurs B, trastuzumab in vivo overexpress epidermal growth factor receptor Poulsen HS. (2007). EGFRvIII escapes down-regulation due to and ErbB ligands and remain dependent on the ErbB receptor impaired internalization and sorting to lysosomes. Carcinogenesis network. Clin Cancer Res 13: 4909–4919. 28: 1408–1417. Sequist LV, Bell DW, Lynch TJ, Haber DA. (2007). Molecular Grovdal LM, Stang E, Sorkin A, Madshus IH. (2004). Direct predictors of response to epidermal growth factor receptor interaction of Cbl with pTyr 1045 of the EGF receptor (EGFR) is antagonists in non-small-cell lung cancer. J Clin Oncol 25: required to sort the EGFR to lysosomes for degradation. Exp Cell 587–595. Res 300: 388–395. Sergina NV, Rausch M, Wang D, Blair J, Hann B, Shokat KM et al. Han W, Zhang T, Yu H, Foulke JG, Tang CK. (2006). Hypopho- (2007). Escape from HER-family tyrosine kinase inhibitor therapy sphorylation of residue Y1045 leads to defective downregulation of by the kinase-inactive HER3. Nature 445: 437–441. EGFRvIII. Cancer Biol Ther 5: 1361–1368. Sunada H, Magun BE, Mendelsohn J, MacLeod CL. (1986). Huang S, Armstrong EA, Benavente S, Chinnaiyan P, Harari PM. Monoclonal antibody against epidermal growth factor receptor is (2004). Dual-agent molecular targeting of the epidermal growth internalized without stimulating receptor phosphorylation. Proc factor receptor (EGFR): combining anti-EGFR antibody with Natl Acad Sci USA 83: 3825–3829. tyrosine kinase inhibitor. Cancer Res 64: 5355–5362. Zhou BB, Peyton M, He B, Liu C, Girard L, Caudler E et al. (2006). Matar P, Rojo F, Cassia R, Moreno-Bueno G, Di Cosimo S, Targeting ADAM-mediated ligand cleavage to inhibit HER3 and Tabernero J et al. (2004). Combined epidermal growth factor EGFR pathways in non-small cell lung cancer. Cancer Cell 10: receptor targeting with the tyrosine kinase inhibitor gefitinib 39–50.

Oncogene