Background
HER2 overexpression is present in 13-30% of all breast cancers [
1,
2] and it correlates with poor disease outcome, high rates of metastasis and resistance to conventional treatment modalities [
1‐
5]. Trastuzumab (TZ; Herceptin
®), a monoclonal antibody that targets the HER2 receptor and interferes with its function is effective in treating some HER2-positive breast cancers [
6‐
8]. However, many patients with HER2-positive disease are insensitive to TZ both as first line treatment or following a relapse after conventional chemotherapy [
6‐
9]. Furthermore, the majority of patients with metastatic disease that initially respond to TZ ultimately develop clinically relevant resistance to this agent [
8,
9]. As TZ treatment has recently been expanded into the adjuvant setting [
10], intrinsic and acquired resistance represents an important clinical problem that urgently awaits a discovery of novel drugs and development of innovative drug combinations to improve outcome for patients with advanced HER2-positive and TZ refractory disease.
Numerous studies have demonstrated that HER2 is often co-expressed in breast cancers with epidermal growth factor receptor (EGFR) [
1,
5,
8,
11‐
16]. It has been established that dimerization of HER2 and EGFR generates a potent signaling response mediated primarily through activation of the phosphatidylinositol 3-kinase (PI3K)/AKT and the RAS-Raf-mitogen-activated protein kinase (MAPK) pathways that sustain cancer cell growth, proliferation and survival [
5,
8]. Co-expression of EGFR and HER2 in breast cancer cell lines has been shown to induce drug resistance, including resistance to TZ [
17,
18], and has been correlated with a negative prognosis for breast cancer patients [
1,
11]. These data suggested that EGFR constitutes an important therapeutic target in breast cancers and have prompted investigators to consider gefitinib (ZD1839, Iressa
®), a reversible small molecule inhibitor of the EGFR tyrosine kinase, for treatment of HER2 overexpressing and EGFR co-expressing breast malignancies [
19].
The preclinical data have demonstrated that gefitinib exerts positive therapeutic effects in models of HER2 overexpressing breast cancer which have been attributed to blocking activity of the PI3K/AKT and the MAPK pathways, increased apoptosis, induction of cytostasis through G
1/G
0 cell cycle arrest and downregulation of cyclin D1, as well as inhibiting angiogenesis [
12‐
14,
20,
21]. However, our previous study conducted in animals bearing HER2 overexpressing MCF7-HER2 and MDA-MB-435/LCC6-HER2 breast cancer xenografts showed that gefitinib monotherapy results in only modest reduction of tumor volume [
12]. The same study also showed that when gefitinib was used in combination with TZ the
in vivo efficacy has been improved as judged by inhibition of tumor growth, but the data obtained by measuring multiple endpoints of therapeutic activity revealed that the combination was not beneficial [
12]. These results have been recapitulated in a clinical trial demonstrating that the TZ and gefitinib combination should not be used for treatment in patients with HER2-positive breast cancer [
19].
More recently, it has been shown that HER2 overexpression in breast cancer is often associated with aberrant activation of the mTOR pathway [
22,
23]. mTOR is a major cellular signaling hub that integrates inputs from the upstream signaling pathways, including tyrosine kinase receptors, while also governing energy homeostasis and cellular responses to stress such as nutrient deprivation and hypoxia [
24,
25]. The mTOR kinase liaisons with either Raptor or Rictor proteins to form two functionally different complexes: rapamycin-sensitive mTOR complex 1 (mTORC1) and rapamycin-insensitive mTOR complex 2 (mTORC2) [
24,
25]. The most prominent downstream effectors of mTORC1 include ribosomal S6 kinase (S6K) and the eukaryotic translation initiation factor 4E-binding protein 1 (4E-BP1) which regulate the translation of ribosomal and cap-dependent proteins essential for cell growth and G
1 to S cell cycle progression [
24,
25]. mTORC2 is an Akt Ser473 kinase that is controlled by a feedback inhibitory loop mediated through S6K1 (p70S6K) [
24‐
29]. Because of its critical role in promoting cell growth, mTOR is considered an attractive target in cancer [
25,
30]. Everolimus (RAD001) and CCI-779 are two allosteric mTORC1 inhibitors that are in clinical development for various malignancies; however, single-agent therapy has only modest efficacy in the metastatic breast cancer setting [
31,
32]. These results have encouraged the investigation of mTORC1 inhibitors in combination with other targeted therapies such as aromatase inhibitors and HER2 targeting drugs. A Phase I/II trial of RAD001 in combination with TZ in refractory HER2 positive metastatic breast cancer have reported encouraging results with 34% of patients achieving clinical benefit [
33]. Interestingly, several preclinical studies documented that mTOR inhibitors combined with EGFR targeted agents increase efficacy of treatment in renal, lung, pancreatic, colon, prostate and HER2-negative breast cancer models [
34‐
36]. However, the therapeutic effects of EGFR and mTOR inhibitors in combination have not yet been broadly assessed in HER2 overexpressing breast cancers with different TZ sensitivity. Here, we show that the EGFR inhibitor gefitinib and the mTOR inhibitor RAD001 when used in combination improve effectiveness of the treatment in HER2 overexpressing breast cancers that results in impediment of cancer growth.
Methods
Cells, tumor xenografts and treatments
MCF7-HER2 cells were a gift from Dr. M. Alaoui-Jamali (McGill University, Montreal, Quebec, Canada) [
37], SKBR3 cells were purchased from American Type Culture Collection (ATCC) and JIMT-1 cells [
38] were purchased from German Collection of Microorganisms and Cell Culture (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ). All cell lines were tested Mycoplasma negative by PCR reaction. MCF7-HER2 cells were maintained in RPMI, SKBR-3 in McCoy's 5A and JIMT-1 in DMEM supplemented with L-glutamine and 10% fetal bovine serum. For
in vivo studies JIMT-1 and MCF7-HER2 cells were harvested in the exponential growth phase and 5 × 10
6 (JIMT-1) or 1 × 10
7 (MCF7-HER2) cells were injected subcutaneously (s.c.) on the back of female Rag2M immuno-compromised mice. Mice receiving MCF7-HER2 cells were implanted with 17-β-estradiol 60-day release tablets (IRA, Sarasota, FL) one day prior to tumor inoculation. Tumor growth was monitored twice a week; tumor sizes were calculated using the formula: 0.5 [length (mm)] × [width (mm)
2]. All agents were delivered as oral gavage. Treatment was initiated on day 17 and carried-out Monday through Friday (QDx5) for 28 (JIMT-1) or 25 (MCF7-HER2) days. RAD001 (a generous gift from Novartis) was diluted with vehicle (Novartis, content not disclosed) and aliquots were kept frozen for the course of treatment. RAD001 and vehicle aliquots were thawed 10-30 min before dosing animals and unused portions were discarded. Gefitinib (a generous gift from AstraZeneca) was solubilized in 0.5% Tween-80 in sterile milli-Q water (vehicle) and kept at 4°C. Gefitinib formulation was prepared weekly. Combination treated mice were dosed first with gefitinib followed by RAD001 four hours later. Tumors were harvested 30 min after the last dose and cut into two parts: one part was frozen in liquid nitrogen (N
2) for Western blot analysis and the second part was frozen in embedding medium and stored in -80°C for immunohistochemical processing. Animal protocols were approved by the University of British Columbia Animal Care Committee, and these studies were done in accordance with guidelines established by the Canadian Council on Animal Care.
Alamar Blue and IN Cell 1000 screening assays
Cells were plated under standard serum conditions (10% FBS) in their respective media in triplicate wells/condition in 96-well flat bottom plates (Optilux, Falcon, Becton-Dickinson). MCF7-HER2 and JIMT-1 cells were plated at densities of 5,000 or 1,500 cells/well for 72 and 144 h drug incubations, respectively. SKBR3 cells were plated at 15,000 and 4000 cells/well for 72 and 144 h, respectively. Cells were allowed to adhere overnight. Next day the cells were treated with gefitinib, RAD001 and combination of both drugs at a fixed molar ratio over a broad dose range to establish growth curves for a 72 h and 144 h read-out. Stock solutions of 20 mM gefitinib and 20 mM RAD001 were prepared in DMSO and stored in -80°C. Gefitinib and RAD001 stocks were diluted in medium with decreasing percentage of DMSO and 10× concentrated drugs were added to cells. The final concentration of DMSO in vehicle and drug treated cells was standardized to 0.5% (vol/vol) and the final media volume in 96-well plate wells was 200 μl. After 72 or 144 h incubation, Alamar Blue (Invitrogen, Burlington, ON, CA) was added to one set of plates to evaluate cell viability. Fluorescence was measured using the FLUOstar OPTIMA plate reader (BMG Labtechnologies, Germany) with 544 nm excitation and 590 nm emission filters. A second set of plates was stained with DRAQ5 (Biostatus, Shepshed, UK) and ethidium homodimer (ETH; Molecular Probes, Invitrogen) followed by imaging with IN Cell 1000 Analyzer (GE Healthcare). Ten images per well were acquired with 10× objective. Data analysis strategies were supported by enterprise level servers. Images were analyzed with IN Cell 1000 Investigator software using the Multi Target Analysis (MTA) module and data were reported as the percentage of dead cells normalized to vehicle control by subtracting the percentage of dead cells in the DMSO control from the percentage of dead cells in treated cultures.
Synergy Determination
Following drug treatment
in vitro, the number of viable cells was measured using the Alamar Blue assay as described above. Alamar Blue measures mitochondrial activity which is lost upon cell death. The data obtained with the Alamar Blue assay were normalized to the vehicle control and expressed as % viability. Next, these data were converted to Fraction affected (Fa; range 0-1), where Fa = 0 represents 100% viability and Fa = 1 represents 0% viability) and analyzed with the CompuSyn™ program (Biosoft, Ferguson, MO) based upon the Chou and Talalay median effect principle [
39]. This program calculates a combination index (CI) that is used to identify synergistic, additive, and antagonistic drug interactions.
Flow cytometry
Cells were plated in their respective media containing 10% FBS in T25 flasks or 6 cm diameter culture dishes and allowed to adhere overnight. The next day cells were treated with the indicated agents. After 72 h, supernatant from treated cells was transferred to a 14 ml tube and combined with adherent cells harvested with 0.25% Trypsin EDTA. For cell cycle analysis cells were washed twice with PBS and 2 × 106 cells/sample were fixed in 1.8 ml cold (-20°C) 70% ethanol followed by 1 h incubation on ice and 24 h incubation in -20°C. Cells were then pelleted and stained in PBS buffer containing 50 μg/ml propidium iodide (PI, Molecular Probes, Invitrogen) with 1 mg/ml RNase A (Sigma-Aldrich) and 0.1% Triton X-100 (Bio-Rad, Richmond, CA) for 15 min at 37°C followed by 1 h incubation on ice. For apoptosis analysis cells were washed twice with Hank's media without phenol red and pellets were resuspended in Annexin-V buffer containing anti-Annexin-V-FITC antibody (Caltag Laboratories, Burlingame, CA). Samples were then incubated on ice for 30 min and counterstained with PI at a final concentration of 1 μg/ml. Flow cytometric analysis was performed with FACSCalibur flow cytometer and acquired data were analyzed with the Cellquest software (Becton-Dickinson, San Jose, CA).
Western blotting
Cells were plated in T25 flasks or 6 cm culture dishes and after overnight adhesion treated with the indicated drugs. After 72 h cells were harvested in ice-cold PBS. Cell pellets were lysed in lysis buffer containing 50 mM Tris pH 7.4, 150 mM NaCl, 1% NP-40, 0.25% Na-deoxycholate, 1 mM EDTA, 0.1% SDS, and Mini Protease Inhibitor Cocktail tables (Roche Diagnostics, Mannheim, Germany). Tumors were homogenized in lysis buffer followed by sonication. After centrifugation (30 min at 13000 rpm) the protein concentration in the supernatant was quantified using the Pierce Micro BCA™ Assay Kit. 30 - 50 μg of total protein per sample was separated on precast 4-12% Bis-Tris gels (NuPage, Invitrogen) and transferred to NuPage 0.45 μm nitrocellulose membranes (Invitrogen). Membranes were blocked with 5% skim milk powder in TBS-T (150 mM NaCl, 50 mM Tris, 0.1% Tween-20, pH 7.4) and incubated overnight with primary antibodies in 5% BSA in TBS-T. The next day membranes were washed 3 times with TBS-T and incubated for 1 h with peroxidase-conjugated secondary antibodies (Promega) in TBS-T containing 5% skim milk. Membranes were washed 3 times with TBS-T and signals were detected by enhanced chemiluminescence (SuperSignal® West Pico Chemiluminescent Substrate, Thermo Scientific) on BioMax Light Film (Kodak). All antibodies used for Western blot analysis were from Cell Signaling Technology (Beverly, MA). The following phospho-specific antibodies were used: P-EGFR (Tyr1086 antibody #2220), P-HER2 (Tyr1221/1222 antibody #2243 or Tyr1248 antibody #2247), P-ERK1/2 (Thr202/Tyr204 antibody #9101), P-AKT (Ser473 antibody #4060), P-70S6K (Thr389 antibody #9206), P-ribosomal protein S6 (Ser235/236 antibody #2211). The β-actin antibody (Sigma-Aldrich) was used as a loading control. Films with visualized protein bands were digitized and the optical density (OD) of bands was measured using UN-SCAN-IT graph and gel digitizing software (Silk Scientific, Inc.). After background subtraction the optical density (OD) value for each individual protein band was corrected for β-actin loading and normalized to the vehicle control expressed as l. Western blot analysis was repeated 2 - 3 times to assure consistency of the results.
Immunohistochemistry, image acquisition and image analysis
10 μm cryosections were cut using a Cryostar HM560 (Microm International GmbH), air dried and then fixed in 50% (v/v) acetone/methanol for 10 min at room temperature. Endothelial cells were stained using a monoclonal antibody to PECAM/CD31 (BD Pharmingen) and fluorescent Alexa 647 secondary antibody (Invitrogen). Terminal deoxyribonucleotide transferase-mediated nick-end labeling (TUNEL) staining was used to label apoptotic cells (In situ Cell Death Detection kit, TMR red; Roche). Proliferating cells were stained using a polyclonal antibody to Ki67 (AbCam) followed by a peroxidase conjugated secondary antibody (Sigma-Aldrich) and metal-enhanced 3,3'-diaminobenzidine substrate (Pierce). Cell nuclei were labeled with Hoechst 33342 (8 ng/mL; Molecular Probes, Invitrogen) for 30 min at 37°C. At each stage of staining, whole tumor sections were imaged using a robotic fluorescence microscope, as previously described [
40,
41]. Automated tiling of adjacent microscope fields of view was completed to generate images of an entire tumor section at a resolution of 0.75 μm per pixel. All parameters stained on the same section were imaged separately using a monochrome camera and composite color images were generated using Adobe Photoshop (CS). Using NIH-Image
http://rsb.info.nih.gov/nih-image and user-supplied algorithms, digital images were superimposed, aligned and cropped to tumor tissue boundaries with staining artifacts removed. Confluent necrosis was subsequently cropped from images and the degree of necrotic tissue calculated as the proportion of necrotic pixels relative to all pixels. ImageJ software applications
http://rsb.info.nih.gov/ij/ and user-supplied algorithms were used to quantify the degree of staining above the thresholds determined to be > 10 standard deviations from background for CD31, TUNEL and Ki67, and data are reported as percent positive pixels of non-necrotic, viable tumor tissue. As a measure of tumor vascularization, the median distance of viable tissue to the nearest CD31-positive object (blood vessel) is reported (μm), such that a larger distance reflects a lower vascular density. Note that for CD31 analysis one JIMT-1 tumor was removed from each of the gefitinib and RAD001-gefitinib combination groups due to the presence of disproportionate necrosis; where only a narrow, avascular rim could be detected as viable tissue. To observe the location of proliferating cells in relation to blood vessels, Ki67 positive pixels were sorted based on their distance from CD31-positive vessels in 1.5 μm increments, and data are expressed as % positive Ki67 pixels relative to distance from vasculature (μm).
Statistical analysis
One-way ANOVA was used to assess differences among the treatment groups with an unpaired t-test (GraphPad Prism version 5.00). The obtained p values were adjusted for multiple comparisons using the Benjamini-Hochberg procedure (R version 2.11.1). Differences were considered significant at p ≤ 0.05.
Discussion
New treatment strategies are needed for patients with advanced HER2-positive breast cancers due to very limited therapeutic options available for fighting this disease. Encouraged by previously published reports demonstrating synergistic interactions between EGFR and mTOR inhibitors in various cancers [
34‐
36] we investigated the activity of the EGFR targeted drug gefitinib used in combination with a rapamycin analog, RAD001, in HER2 overexpressing and EGFR co-expressing breast cancer models with TZ sensitive or resistant phenotypes [
12,
38,
42,
43]. The rationale to study this combination in HER2 positive breast cancers has also been strengthened by a recent investigation of Miller
et al. who demonstrated that inhibition of PI3K and mTOR are required for the growth inhibitory effects of HER2 antagonists in HER2 overexpressing breast cancer and that inhibition of mTOR is an effective therapeutic strategy in TZ resistant breast cancer models [
48].
Our data showed that while SKBR3 cells were sensitive to gefitinib, JIMT-1 and MCF7-HER2 cells were gefitinib resistant; however, RAD001 was capable of sensitizing these cells to gefitinib. It is interesting to note that both JIMT-1 and MCF7-HER2 cell lines harbor PIK3CA mutations which have been associated with acquired resistance to EGFR kinase inhibitors but can also predict sensitivity towards mTOR inhibition [
43,
49,
50]. Together with our data, this may suggest that RAD001 is able to reverse gefitinib resistance in PIK3CA mutant tumors. Our data indicate that
in vitro gefitinib and RAD001 interact in a synergistic fashion, as shown by a mathematical model developed by Chou and Talaly [
39] and this synergy did not appear to be drug ratio dependent. The
in vivo efficacy of gefitinib and RAD001 was also greatly improved when these drugs were used in combination. Further validation of our results in other models of HER2 overexpressing and TZ resistant breast cancers such as MDA-MB-453, MDA-MB-361 or UACC893 would be crucial in order to determine if this combination is broadly effective in TZ resistant cancers. However, our results obtained using the JIMT-1 model do give an indication that the gefitinib and RAD001 combination was able to effectively target the cellular machinery that is indispensable for cancer cell growth despite the existence of multiple mechanisms contributing to the extreme TZ resistance of this cell line [
38,
42]. It should be noted that while the combination treatment did not result in regression of established tumors, this could be a consequence of our experimental design. We opted to use doses of gefitinib and RAD001 that on their own did not produce statistically significant (p > 0.05) reduction in tumor volume relative to vehicle treated controls, so that inhibition of tumor growth by the combination would be evident. Consequently, gefitinib given at 100 mg/kg resulted in a more potent reduction in MCF7-HER2 tumor volume than anticipated on its own, thus making the effect of the combination very modest.
The data obtained based on analysis of multiple endpoints after 72 h treatment suggest a contribution of cytostasis in the presence (in SKBR3 and JIMT-1 cells) or absence (in MCF7-HER2 cells) of cytotoxicity to the synergy between gefitinib and RAD001
in vitro. Treatment with the combination induced apoptosis only in JIMT-1 cells; however, it should be noted that Annexin V is a marker for the early apoptotic event so apoptosis may not be detected in SKBR3 and MCF7-HER2 cells after 72 h. Thus, a contribution of apoptosis to cytotoxicity at earlier time points is possible. Our findings are consistent with other reports demonstrating that gefitinib and RAD001 are cytostatic in nature [
12,
14,
25,
45] and that the degree of cytotoxicity triggered by these drugs is a cell type dependent phenomenon [
14,
48]. This perhaps reflects PIK3CA or other mutations in genes controlling cell growth, proliferation and survival [
43]. While the enhancement of cytostasis seen after 72 h in the combination (1 μM gefitinib with 5 nM RAD001) treated SKBR3 and JIMT-1 cells was confirmed by increased G
1/G
0 cell cycle arrest and decreased S phase relative to the single drugs, the combination failed to induce significant cell cycle changes in MCF7-HER2 cells despite growth inhibition in the absence of cytotoxicity. It has been reported that the parental MCF7 cell line expresses high levels of activated p70S6K and cyclin D1 [
51] which may have contributed to somewhat obscure cell cycle regulation, possibly resulting in longer time required to complete a cell cycle or perhaps a transient cell cycle block that was resolved before 72 h. Increased cytostasis by the gefitinib and RAD001 combination in the absence of increased cytotoxicity was also found
in vivo in JIMT-1 and MCF7-HER2 tumor xenografts. This may explain why the combination stabilized tumor growth and did not cause tumor regression. Interestingly, gefitinib increased levels of Ki67-positive cells in MCF7-HER2 tumors. These proliferating cells were present at similar frequency in proximal and longer (> 100 μM) distances from the blood vessels suggesting that tissue perfusion in gefitinib treated tumors was perhaps improved. In support, our previous study found that MCF7-HER2 tumors treated with gefitinib contain a greater proportion of functional Hoechst 33342 perfused vessels and this correlated with significantly increased tumor tissue oxygenation resulting in fewer hypoxic cells present [
52]. The study of Lu et al. also showed that positive therapeutic responses of cancer cells to EGFR-targeted therapy with cetuximab and gefitinib are associated with downregulation of hypoxia-inducable-factor-1α (HIF-1 α) [
53]. Furthermore, Hardee et al. reported that blockade of HER2 signaling in MCF7-HER2 tumors with TZ improved tumor tissue oxygenation and vascular architecture along with increased microvessel density [
54]. Thus, we speculate that gefitinib treatment perhaps resulted in vessel normalization. In turn, improved vessel functionality could be responsible for more efficient delivery of drugs to tumor tissue and increased cytostasis. This may explain why MCF7-HER2 tumors were more sensitive to gefitinib than JIMT-1 tumors, even though we found the opposite to be true
in vitro in MCF7-HER2 and JIMT-1 cells.
The most striking and consistent therapeutic effect of the combination noted
in vitro and
in vivo was greater inhibition of the mTOR pathway reflected by decreased P-p70S6K and P-S6 levels relative to the effects of the single drugs. These changes strongly correlated with better efficacy of the combination treatment. Accordingly, several reports suggested that P-p70S6K can be considered as a biomarker for monitoring treatment outcomes in patients receiving mTOR inhibitors [
25,
45,
55,
56]. While the combination did not decrease P-EGFR levels
in vitro compared to the single drugs, enhanced inhibition of P-EGFR by the combination
in vivo appeared to be a consistent molecular event in JIMT-1 and MCF7-HER2 tumors. This can be attributed to inhibition of P-EGFR by both gefitinib and RAD001. The latter effect was not reported in other studies and the mechanisms involved are unclear at this point. Improved inhibition of P-EGFR by the combination
in vivo may certainly play a role in downregulation of the mTOR pathway but how this is achieved without robust inhibition of AKT and ERK1/2 activity remains a question for further research. Interestingly, the
in vivo reduction in P-EGFR, P-HER2, P-p70S6K and P-S6 levels was mirrored by decreases in total expression of the corresponding proteins in different treatment groups. Similar correlations were observed
in vitro for selected proteins in gefitinib and/or combination treated MCF7-HER2 and JIMT-1 cells. These data suggest that inhibition of translation rates or perhaps changes in post-translational events regulating the expression of EGFR, HER2, p70S6K and S6 proteins may have contributed to decreased signaling in addition to direct effects on protein phosphorylation. In contrast, the expression of ERK1/2 and AKT
in vivo was not altered after different treatments indicating that changes in phosphorylation levels actually reflected activation status of these proteins.
It should be noted that despite greatly improved inhibition of the mTOR pathway by the gefitinib and RAD001 combination our data suggest lack of or only moderate inhibitory effects of the combination on P-AKT levels. This result can be explained by a RAD001 mediated negative feedback loop. It has been demonstrated that inhibition of mTORC1 by rapamycin analogs initiates p70S6K-dependent feedback signaling resulting in stimulation of mTORC2 and phosphorylation of AKT on Ser473 [
24‐
30]. Our
in vitro data show that after 72 h RAD001 increased P-AKT levels in all three cell lines, but addition of gefitinib to RAD001 was able to counteract this effect. RAD001 also enhanced P-ERK1/2 levels in SKBR3 and MCF7-HER2 cells and in JIMT-1 tumors and these results are in agreement with studies showing activation of ERK1/2 through a PI3K-dependent feedback loop following inhibition of mTORC1 in some human cancers [
57]. Again, addition of gefitinib to RAD001 counteracted activation of ERK1/2 in SKBR3 cells and in JIMT-1 tumors. Nonetheless, absence of robust inhibition of AKT and ERK1/2 activity
in vivo after treatment with the combination is of concern since it may provide cancer cells with a survival advantage and lead to development of drug resistance and escape from cytostasis which consequently would limit treatment efficacy [
3‐
5,
18,
58,
59]. Likewise, other investigators have shown that targeting HER2 and mTOR using the TZ and RAD001 combination inhibits growth of HER2 overexpressing cancers to a greater extent than single agents, but this treatment did not further reduce P-AKT or P-ERK1/2 levels, when compared to the single drug effects [
48]. Thus, combining drugs that inhibit function of EGFR/HER2 with dual PI3K/mTOR and MEK pathway inhibitors in order to abolish compensatory mechanisms may eliminate cancer cell survival and perhaps improve therapeutic effects in HER2-positive breast cancers [
59‐
62].
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
WHD designed experiments, performed flow cytometric analysis, analyzed data, interpreted results and wrote the draft of the manuscript. SAW performed Western blot experiments and helped with the final draft of the manuscript. MAQ supported the computer platform to perform HCS data analysis. LYW performed cell culture, HCS experiments and assisted in animal experiments. YF and GK performed Western blot experiments. AIK analyzed the HCS data. JHEB performed and interpreted the IHC and tumor mapping and AIM contributed to tumor mapping data interpretation. DM performed animal experiments. KAG and MBB conceived the study, participated in its design and helped with the manuscript draft. All authors read and approved the final manuscript.