Repression of Malignant Tumor Progression upon Pharmacologic IGF1R Blockade in a Mouse Model of Insulinoma

This article is featured in Highlights of This Issue, p. 671

Mounting evidence from cell culture experiments, animal models, and correlative studies in patients with cancer clearly show that the insulin-like growth factor (IGF)/IGF-1 receptor (IGF1R) system plays a central role in many cancer types (1, 2). Interfering with this system has emerged as a promising strategy to treat a variety of human cancers (3–5). IGFs (IGF-1 and IGF-2) exert their biologic function mainly by binding and stimulating the cell surface receptors IGF1R and insulin receptor (IR), both belonging to a family of phylogenetically conserved receptor tyrosine kinases (1). Hybrid IGF1R/IRs have also been observed, yet their specific functions remain elusive. The IGF2R/6-mannose receptor is a monomeric receptor which exclusively binds IGF-2 without apparent tyrosine kinase and signaling activities; it is thought to serve as a decoy receptor for IGF-2 (6). IGF1R and IR are heterotetrameric tyrosine kinase receptors consisting of 2 extracellular ligand-binding α-subunits and 2 β-subunits each composed of a short extracellular domain linked to the α-subunit by disulfide bridges, a transmembrane domain and an intracellular tyrosine kinase domain (7). Ligand–receptor interaction leads to receptor dimerization and tyrosine trans-phosphorylation of residues Y1131, Y1135, and Y1136 within the tyrosine kinase domains, promoting further tyrosine phosphorylation of the receptor and of various substrates. In particular, phosphorylation of the adaptor proteins insulin receptor substrate (IRS) and Shc, 2 major effectors of IGF1R signaling, leads to the activation of the phosphoinositide 3-kinases (PI3K), mitogen-activated protein kinase (MAPK), and 14-3-3 pathways (1, 8).

In normal physiology and embryonic development, IGF-1/IGF1R signaling is involved in regulating growth and homeostasis of a variety of tissues and organs and is therefore found ubiquitously activated. Serum IGF-1 levels, mainly of hepatic origin, are predominantly regulated by growth hormone. In the developing pancreas, IGF-1 and IGF-2 are produced by cells of the islets of Langerhans where they exert paracrine and autocrine functions critical for islet differentiation (9) and islet cell growth (10, 11). IGF-2 supports fibroblast growth factor (FGF)-2–mediated islet growth (12), whereas IGF-1 has been shown to directly protect β-cells from apoptosis (13, 14). However, several reports challenge a critical role of the IGF-1/IGF1R system for islet proliferation or protection. On one side, pancreas-specific deletion of IGF-1 results in enlarged islet mass and enhanced protection from chemically or dietary experimental diabetic state (15). On the other hand, β-cell–specific deletion of IGF1R does not alter total islet mass and only causes mild hyperinsulinemia (16).

In cancer biology, IGF1R has been found to be critical for the malignant transformation of cells by a number of oncogenes (17). In studies using the genetic ablation of IGF-1 expression, low IGF-1 levels provoke reduced growth and metastasis of xenografted tumors and increased resistance to carcinogen-induced tumorigenesis (18). Transgenic expression of IGF-1 in basal epithelial cells induces hyperplasia and well-differentiated adenocarcinomas of skin and prostate (19, 20). Transgenic expression of IGF-2 in the liver results in altered body composition and an increase in the development of a variety of cancer types (21). In population studies, high serum levels of IGF-1 have been correlated with an increased risk to develop prostate cancer and premenopausal breast cancer (22). An increased incidence of colorectal adenomas and cancer is seen in acromegaly where hypersecretion of growth hormone is accompanied by elevated IGF-1 levels (23).

The expression of either IGF-1 or IGF-2 is found highly upregulated in a variety of cancer types in patients and in mouse models of cancer (24). For example, during experimental insulinoma in Rip1Tag2 (RT2) mice (25), the expression of IGF-2 is upregulated concomitant with the onset of tumor cell proliferation (early hyperplasia; refs. 26, 27). Genetic ablation of IGF-2 expression in RT2 mice leads to increased tumor cell apoptosis and dramatically reduced tumor volumes. Conversely, the transgenic expression of IGF1R in β-tumor cells of RT2 mice (RT2;IGF1R) promotes progression to tumor malignancy and metastasis (28). Inhibition of the IGF system might therefore be a promising strategy in the treatment of insulinoma and other cancer types.

Several different experimental strategies in targeting the IGF/IGF1R system have been used, including reducing ligand availability by growth hormone antagonists, neutralizing antibodies against IGF-1 and IGF-2, recombinant IGF-binding proteins (IGFBP), reducing IGF1R expression by RNA interference, or inhibiting IGF1R activation by antibodies and small-molecule tyrosine kinase inhibitors (TKI), such as BMS754807, which are in clinical studies (5). IGF1R antibodies block activation of IGF1R and also trigger internalization and degradation by endocytosis and are already in phase I and II clinical trials.

Among the different TKIs, NVP-AEW541 has been used in several preclinical studies. It exhibits antitumoral activity upon transplantation of human neuroblastoma, human Ewing sarcoma, human multiple myeloma, and murine fibrosarcoma cells in immunodeficient mice (29–32). In vitro, IGF1R inhibition has been shown to induce apoptosis in cells relying heavily on IGF1R signaling. On the other hand, it has been shown that cells insensitive to IGF1R “mono-inhibition” can be sensitized to chemotherapeutic agents by IGF1R inhibition, as shown for small cell lung cancer cells (33), ovarian cancer cells (34), and gastrointestinal neuroendocrine tumor cells (35).

Here, we have used the RT2 transgenic mouse model of pancreatic β-cell carcinogenesis to investigate the biologic effects of IGF1R inhibition by NVP-AEW541. Treatment of RT2 mice with NVP-AEW541 moderately inhibits tumor progression but does not substantially attenuate tumor growth. The data indicate that inhibition of IGF1R alone is not sufficient to efficiently block insulinoma growth and suggest a compensating role of insulin receptor supporting IGF-mediated tumor cell survival and tumor growth.

The generation and phenotypic analysis of Rip1Tag2 and Rip7hIGF1R transgenic mice has been previously described (25, 28). AEW541 was kindly provided by Novartis Pharma AG. Mice were treated with either 50 mg/kg body weight AEW541, dissolved in 25 mmol/L tartaric acid (Sigma) or tartaric acid alone by twice daily oral administration according to the regimens described in Results. Tumor incidence per mouse was determined by counting all macroscopically detectable pancreatic tumors with a minimal diameter of 1 mm. Tumor volumes were calculated by measuring the tumor diameter and assuming a spherical shape of the tumors. All experimental procedures involving mice were conducted according to the guidelines of the Swiss Federal Veterinary Office (SFVO) and the regulations of the Cantonal Veterinary Office of Basel Stadt and are covered by licenses 1878 and 1908.

Pancreata from transgenic and control mice were isolated and fixed in 4% paraformaldehyde overnight, dehydrated, and embedded in paraffin. For analysis of cell proliferation, mice were injected intraperitoneally (i.p.) with 100 μg 5-bromo-2-deoxyuridine (BrdUrd; Sigma) per gram of body weight 2 hours before sacrifice. Histologic staging and grading of tumors was done in a blinded manner in hematoxylin/eosin-stained sections. The following antibodies were used: rabbit anti-mouse LYVE-1 (Reliatech), rat anti-mouse CD31 (clone MEC13.3, PharMingen), biotinylated mouse anti-BrdU (Zymed), rabbit anti-IGF1Rβ (Sc-713, Santa Cruz Biotechnology), and rabbit anti-phospho-IGF1R/IR (Cell Signaling Technology). Apoptosis was detected by terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling (TUNEL) kit according to manufacturer's instructions (Roche Applied Science). Immunohistochemistry was conducted by using biotinylated secondary antibodies followed by VECTASTAIN Elite ABC kit (Vectorlabs), and slides were developed either using diaminobenzidine (DAB; Dako) or 3-amino-9-ethylcarbazole (AEC; Vectorlabs) and counterstained by hematoxylin or eosin. Quantification of CD31 vessel density, proliferation, and apoptosis was carried out by counting positive events within a 40 × magnification high-power field (HPF) of randomly selected tumor areas and at least 10 fields were counted per mouse. All sections were analyzed with an Axioskop 2 Plus light microscope using Axiovision 3.1. Software (Zeiss).

Pancreatic tumors were isolated, and tumor cell lines were established as previously described (36). Established cell lines were cultured in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal calf serum (FCS), 100 U/mL penicillin and 100 μg/mL streptomycin sulphate (all reagents from Sigma). βT2 and βT1184 cell lines were derived from RT2 single-transgenic mice, whereas βT-IGF1R cell lines were derived from RT2;Rip7hIGF1R double-transgenic mice. For cell culture experiments, AEW541 was dissolved in dimethyl sulfoxide (DMSO). Inhibition of IGF1R phosphorylation by AEW541 was assessed by adding AEW541 3 hours before lysis to the cells. For βT2 cells, 50 ng/mL murine IGF-1 (PeproTech) was added 30 minutes before lysis and 2 mmol/L sodium orthovanadate was added 15 minutes before lysis.

Mice were treated with 4 doses of AEW541 or tartaric acid over 2 days as described above. Two hours after the last treatment, mice were injected i.v. with murine IGF-1 at 0.1 mg/kg body weight and killed 5 minutes later by cervical dislocation. Pancreatic insulinomas were quickly removed, shock-frozen in liquid nitrogen, ground in a precooled mortar containing dry ice, transferred to a conical tube, and stored on ice. After sublimation of dry ice, ground tumor pieces were lysed in HNIG buffer (50 mmol/L HEPES, pH 7.5, 150 mmol/L NaCl, 10% glycerol, 1% Triton X-100, 1.5 mmol/L MgCl, 1 mmol/L EDTA, 10 mmol/L sodium PP_i, 2 mmol/L sodium orthovanadate, 100 mmol/L NaF, 1 mmol/L phenylmethylsulfonylfluoride) containing protease inhibitors (Sigma), lysates cleared by centrifugation and used for immunoblotting.

For the determination of phospho-IGF1R levels, cells and tumors were lysed in HNIG buffer, and protein concentrations of lysates were determined by BCA assay (Pierce Biotechnology). Equal amounts of protein were resolved by SDS-PAGE and blotted on polyvinylidene difluoride (PVDF) membranes by semi-dry transfer.

βT2 or βT-IGF1R cell lines were incubated with AEW541 for 24 hours in DMEM containing either 2% (βT2) or 10% (βT-IGF1R) FCS. Detached and adherent cells were pooled and genomic DNA was quantified using propidium iodide (PI). Shortly, cells were washed with PBS, fixed in 4% paraformaldehyde, permeabilized using 0.1% saponine, and then stained in hypotonic citrate buffer (100 mmol/L Na-citrate, pH7.4) containing PI (50 μg/mL) and RNAse H (20μg/mL) overnight at 4°C. Cells were analyzed for PI staining in the PE channel in linear mode of a FACS Canto II flow cytometer (B&D).

In vitro wound-healing assays were conducted on confluent βT2 and βT-IGF1R cells. The media on the confluent cells were replaced with DMEM with 2% FBS media, and an area of cells was scraped off using a 200-μL pipette tip. Light microscopic images were taken at time 0, 24, 48, and 72 hours. Gap closure was analyzed by using an ImageJ program.

Total RNA was prepared from tumors of AEW541- or control-treated animals 3 hours after the last treatment using TRIzol, and reverse-transcribed with random hexamer primers using M-MLV reverse transcriptase (Sigma). cDNA was quantified on a ABI Prism 7000 light cycler (Applied Biosystems) using SYBR green PCR MasterMix (Fermentas) using the following primers: mVEGFA fw: 5′-ACTGGACCCTGGCTTTACTG-3′, rv: 5′-TCTGCTCTCCTTCTGTCGTG-3′; mIR fw: 5′-TGCTCATGCCCTAAGACTGAC-3′, rv: 5′-GATCTTCGCTTTCGGGATG-3′; and mIGF-2 fw: 5′-CGCTTCAGTTTGTCTGTTCG-3′, rv: 5′-GCAGCACTCTTCCACGATG-3′.

Statistical analysis was conducted using GraphPad Prism software (GraphPad Software Inc.).

To determine the efficacy of NVP-AEW541 (AEW541) in repressing proliferation and survival of insulinoma cells, we first tested the ability of AEW541 to induce apoptosis in the cultured insulinoma cell lines βT2 and βT-IGF1R, which have been derived from tumors of RT2 and RT2;RipIGF1R mice, respectively. Even though the βT-IGF1R cell line expresses higher levels of IGF1R as compared with βT2 (data not shown), both cell lines have a similar pattern in inhibition of IGF1R phosphorylation with a half-maximal inhibition below 0.3 μmol/L (Fig. 1A). This result indicates that AEW541 inhibits the mouse IGF1R (expressed by βT2 cells) as efficiently as the human IGF1R (expressed by βT-IGF1R cells). In sharp contrast, much lower levels of AEW541 are required to reduce viability of βT-IGF1R cells compared with βT2 cells (Fig. 1B), as determined by DNA integrity measurements using PI, indicating a stringent dependence of βT-IGF1R cells on functional IGF1R signaling. Whereas 5 μmol/L of AEW541 and reduced FCS levels (2%) in the culture medium are required to decrease the viability of βT2 cells by 50% after 24 hours, βT-IGF1R viability is already half-maximally reduced at 0.5 μmol/L AEW541 and in the presence of 10% FCS. Quantification of apoptosis by TUNEL assay revealed that apoptosis is the main cause for cell number reduction by AEW541 treatment (Fig. 1C; Supplementary Fig. S1). In βT2 cells grown in 10% FCS, cell death was only induced at 10 μmol/L AEW541, a concentration where also other receptor tyrosine kinases were inhibited (31).

To evaluate the effectiveness of IGF1R phosphorylation inhibition and assess short-term transcriptional changes, single-transgenic RT2 mice and double-transgenic RT2;RipIGF1R mice were treated for 2 days with AEW541 or vehicle control, and tumors were isolated 2 hours after the last treatment and then either analyzed for the phosphorylation status of IGF1R by immunoblotting or for mRNA levels of VEGF-A, IGF-2, or insulin receptor by quantitative reverse transcription PCR (qRT-PCR). During tumor progression in RT2 mice, VEGF-A is a key angiogenic factor inducing the angiogenic switch (37), IGF-2 is crucial for insulinoma cell survival (26), and IR-A, lacking 12 amino acids encoded by exon 11, has been shown to bind IGF-2 with comparable affinity to IGF1R (38) and to be critical for IGF-2 signaling in RT2 mice (39).

IGF1R phosphorylation was markedly reduced upon AEW541 treatment in tumors isolated from either single- or double-transgenic mice, however with varying efficacy (Fig. 2A). VEGF-A mRNA levels remained unchanged after 2 days of AEW541 treatment in mice of both genotypes (Fig. 2B). It should be noted that VEGF-A expression has been shown reduced in Ewing sarcoma and in multiple myeloma cells upon IGF1R inhibition (30, 40). In accordance with unchanged VEGF-A levels, we did not find a reduction in microvessel density in RT2 mice treated during ongoing angiogenesis (prevention trial; Fig. 2C). IGF-II mRNA levels were dramatically increased in tumors of both RT2 and RT2;RipIGF1R mice as compared with islets isolated from nontumor-bearing mice, as reported before (26, 28), yet IGF-2 expression was not affected by AEW541 treatment (Fig. 2B). Insulin receptor mRNA levels were found expressed at comparable levels in normal islets of Langerhans and in tumors of either genotype mice and were also not changed by AEW541 treatment (Fig. 2B). Furthermore, the relative abundance of IR-A and IR-B was unchanged between islets from nontumor-bearing mice and total tumor of either genotype, as assessed by semiquantitative PCR (Fig. 2D). These results indicate that the transgenic expression of IGF1R does not affect the expression of the major ligand of IGF1R in RT2 tumors, IGF-2, nor the expression of an additional receptor for IGF-2, IR-A.

Unlike for IGF1R targeting strategies involving antibodies, IGF1R is usually not internalized upon tyrosine kinase inhibition. Consistent with this notion, we also did not observe reduced IGF1R protein levels upon AEW541 treatment in either short- or long-term treated animals (Fig. 2A and data not shown). Immunohistochemical (IHC) staining for IGF1R in treated and untreated mice from both genotypes revealed intense staining of IGF1R expression at the tumor edges, with a patchy but uniform pattern within small tumors and a gradual loss inside large size tumors (Supplementary Fig. S2). Only tumors ≥1.5 mm in diameter were used for immunoblotting analysis, and we did not see a correlation of IGF1R or phospho-IGF1R level and tumor size. Such tumors appear as macroscopic tumors in histologic sections and display an apparently reduced IGR1R expression by IHC staining. Tumors smaller than 1 mm, showing relatively homogenous IGF1R expression by immunohistochemistry, were not analyzed for immunoblotting as such tumors cannot be isolated quickly enough to maintain protein phosphorylation status. RT2;RipIGF1R tumors expressed high levels of IGF1R and, similarly to RT2 mice, eventually lost expression in big tumors (Supplementary Fig. S2; ref. 28). In summary, no difference in IGF1R expression was found between size-matched tumors of mice treated with AEW541- or control-treated mice, assessed by either immunoblotting or IHC analysis.

To investigate the effects of IGF1R inhibition on tumor growth in vivo, we treated RT2 mice and RT2;RipIGF1R mice with AEW541 at a dose of up to 50 mg/kg twice daily in 3 different regimens. In a prevention trial, RT2 mice were treated from weeks 7 to 10 of age. In an intervention trial, RT2 mice were treated from weeks 10 to 12, 1 to 2 weeks before the single-transgenic RT2 mice would succumb to hypoglycemia due to high insulin levels produced by the developing insulinomas (Fig. 3F). RT2;RipIGF1R exhibits accelerated tumor progression and usually die around week 10 of age. Therefore, these mice were treated between weeks 6 and 9 (Fig. 3F). In a dosage escalation trial, RT2 mice were treated between weeks 9 and 11 at varying doses of AEW541, with tartaric acid as solvent control or not at all (Fig. 3F).

In all 3 regimens, body weights at the end of the experiments were not significantly different between control and AEW541-treated groups, and blood glucose levels were unaffected (data not shown). To our surprise, we did not observe any significant reduction of overall tumor burden between control and AEW541-treated RT2 mice (Fig. 3A) and also tumor incidence was unaltered (not shown). Consistent with the lack of any change in tumor growth, in neither of the treatment protocols, a significant decrease in tumor cell proliferation, as assayed by the incorporation of BrdUrd into chromosomal DNA during S-phase, nor a significant increase in tumor cell apoptosis, as determined by TUNEL assay, was observed (Fig. 3B and C, respectively).

We next determined potential changes in tumor progression by staging and grading individual tumors in AEW541- and control-treated mice on histologic pancreatic tissue sections. In RT2 mice, normal islets of Langerhans, hyperplastic islets, differentiated epithelial adenoma with well-defined margins to the exocrine pancreas, and invasive carcinoma with focal regions of invasion can be distinguished (Supplementary Fig. S3; refs. 28, 41). In mice treated in the early-stage prevention regimen, a significant reduction of average tumor malignancy was observed, dropping from 50% carcinoma to 30% carcinoma in AEW541- versus control-treated mice (Fig. 3D). This reduction was only significant in the prevention trial, even though a slight reduction of malignancy was also seen for the intervention regimen. In the dose-escalation experiment, a dose-dependent decrease of malignancy was apparent (Fig. 3E). These results indicate that IGF1R is more important in an early stage of RT2 tumor progression and that its effects are directed more toward the invasive program as opposed to a central mitogenic and/or antiapoptotic role.

As reported previously, insulinoma in RT2;IGF1R double-transgenic mice show a drastically accelerated tumor growth and enhanced tumor malignancy, as manifested in an almost complete absence of adenoma with well-defined margins and a potential direct transition from hyperplastic/angiogenic islets to invasive carcinoma (28). This supports a central role of IGF1R in defining the invasiveness of epithelial cancers and it also indicates a role of IGF1R in conveying accelerated tumor growth. Treatment of RT2;IGF1R mice with AEW541 for 3 weeks starting at 6 weeks of age (intervention trial), resulted in a moderate, yet not significant, reduction in tumor volume (Fig. 4A). No difference in the proliferation rate of tumor cells was apparent between AEW541- and control-treated animals (Fig. 4B). In contrast, the rate of apoptosis was significantly increased upon blockade of IGF1R using AEW541 (Fig. 4C). However, in RT2;IGF1R mice, AEW541 failed to repress malignant tumor progression as analyzed by the percentages of tumors in adenoma or carcinoma stages (Fig. 4D). Consistent with a tumor invasive function of IGF1R signaling, AEW541 treatment of βT-IGF1R cells reduced cell migration in scratch wound closure assays (Fig. 4E).

To gain insights into the molecular mechanisms underlying IGF1R-mediated tumor invasion and progression, we analyzed the expression of genes known to play critical roles in cell migration and invasion by RT-PCR array analysis. These experiments failed to reveal significant changes in the expression of cell migration or cell invasion genes (Supplementary Fig. S4). Validation of some of these results by individual gene RT-PCR analysis confirmed the lack of significant effects on gene expression by AEW541 (data not shown).

In this report, we show the efficacy of the receptor TKI NVP-AEW541 to inhibit phosphorylation of IGF1R in insulinoma cells in vitro and in tumors of transgenic mice in vivo. We observed in vitro IC₅₀ for IGF1R phosphorylation below 0.3 μmol/L for both the murine and the human forms of IGF1R, which is in agreement with the reported IC₅₀ of 0.15 μmol/L in solid-phase–binding assays and 0.1 μmol/L for cellular IGF1R phosphorylation (31). Tumor cell lines established from RT2;IGF1R mice were highly sensitive to inhibition of IGF1R, as an AEW541 concentration of 0.5 μmol/L was sufficient to induce massive apoptosis within 24 hours in full serum-containing medium. Among many kinases tested, this concentration does exclusively inhibit IGF1R and not the closely related insulin receptor, which has a cellular IC₅₀ of 2.3 μmol/L (31). In contrast, to obtain a similar apoptotic response in a cell line established from single-transgenic RT2 mice, a 10-fold higher concentration of AEW541 and reduced serum levels were needed. This result indicates that in cell lines derived from RT2 mice, expressing only endogenous murine IGF1R, another receptor, most likely insulin receptor, must be repressed to obtain a proapoptotic effect. This result is consistent with a previous report showing that the genetic ablation of insulin receptor in RT2 transgenic mice sensitizes pancreatic β-tumor cells to inhibition of IGF1R (39). The differential response between single-transgenic RT2 mice and double-transgenic RT2;IGF1R mice might be due to the fact that tumor cells from IGF1R-overexpressing tumors show oncogenic addiction to IGF1R in vitro. As soon as IGF1R is shut down, no other signals, for example, from 10% serum can protect cells from apoptosis. On the other hand, insulinoma cells derived from RT2 mice may use insulin receptor and IGF1R in parallel and also signal through heterotetrameric IGF1R/IR hybrid receptors.

Similar to the in vitro results, AEW541 was also capable of inhibiting IGF1R phosphorylation in single-transgenic RT2 and double-transgenic RT2;IGF1R mice in vivo. RT-PCR array expression analysis of genes involved in tumor progression and cell migration did not reveal any significant changes in the expression of tumor progression and cell migration genes upon treatment of Rip1Tag2 mice with AEW541, indicating that the changes in IGF1R signaling induced by AEW541 are subtle and require more detailed, for example, proteomic analysis of changes in signaling pathways. Notably, the expression of the potent angiogenic factor VEGF-A was not changed in AEW541-treated single- or double-transgenic mice or between IGF1R-overexpressing and normal tumors were unchanged, indicating that VEGF-A is not a direct target of IGF1R signaling in insulinoma. Accordingly, microvessel density, as assessed by CD31 IHC staining, was unchanged in long-term treated mice. In contrast to these results, VEGF-A expression has been found reduced upon IGF1R inhibition in Ewing sarcoma cells and in multiple myeloma (30, 40).

Two observations have lead to the concept that IGF-2 and IGF1R are critical mitogenic and antiapoptotic players during RT2 multistep carcinogenesis. First, tumors in RT2 mice dramatically upregulate IGF-2 during tumorigenesis, and if prevented from doing so by genetic ablation of IGF-2 expression, mice develop smaller and less malignant tumors, mainly due to increased rates of apoptosis (26). Second, transgenic overexpression of IGF1R dramatically accelerates tumor progression, manifested by increasing tumor volumes and increased malignancy of the tumors (28). However, treatment of RT2 mice with a monoclonal antibody against IGF1R has no significant effect on tumor burden. Only upon genetic ablation of IR-A specifically in β-tumor cells, anti-IGF1R antibody treatment results in reduced tumor growth and increased apoptosis (39). These results clearly show a functional compensation of insulin receptor for IGF1R in tumor cell survival via IGF stimulation. A similar functional contribution of insulin receptor to breast cancer growth and metastasis in vivo has been recently reported (42). High levels of insulin receptor expression in a variety of cancer types may thus hamper the therapeutic efficacy of the specific repression of IGF1R.

In conclusion, while inhibition of IGF1R signaling has demonstrable effects on tumor cell survival and tumor progression in the RT2 mouse model of pancreatic neuroendocrine tumor (insulinoma) and also in breast and other cancer cells, the efficiency of IGF1R inhibition as therapeutic strategy may be impeded by the compensatory activity of insulin receptor. On the basis of its physiologic and metabolic functions, insulin receptor does not qualify as therapeutic target. Future work is required to elucidate whether a therapeutic window exists to sensitize against IGF1R inhibition by repressing insulin receptor without harming its physiologic functions or, even better, whether there are downstream signaling effectors that serve specifically the protumorigenic functions of IGF1R and insulin receptor but not the physiologic functions of insulin receptor, molecules that could serve as appropriate therapeutic targets.

C. Garcia-Echeverria has been and F. Hofmann is an employee of Novartis Pharma AG. G. Christofori has a commercial research grant from Novartis Pharma AG. No potential conflicts of interests were disclosed by the other authors.

Conception and design: A. Zumsteg, F. Hofmann, G. Christofori

Development of methodology: A. Zumsteg, C. Caviezel, L. Pisarsky

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A. Zumsteg, L. Pisarsky, K. Strittmatter

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): A. Zumsteg, C. Caviezel, L. Pisarsky, G. Christofori

Writing, review, and/or revision of the manuscript: A. Zumsteg, C. Caviezel, L. Pisarsky, C. Garcia-Echeverria, F. Hofmann, G. Christofori

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): K. Strittmatter, G. Christofori

Study supervision: G. Christofori

The authors thank D. Hanahan for sharing Rip7hIGF1R transgenic mice and U. Schmieder, H. Antoniadis, and R. Jost for excellent technical support.

This work was supported by EU-FP7 TuMIC HEALTH-F2-2008-201662 and Novartis Pharma AG.

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