The type 1 IGF receptor (IGF1R) is a transmembrane tyrosine kinase that is frequently overexpressed by tumours, and mediates proliferation and apoptosis protection. IGF signalling also influences hypoxia signalling, protease secretion, tumour cell motility and adhesion, and thus can affect the propensity for invasion and metastasis. Therefore, the IGF1R is now an attractive anti-cancer treatment target. This review outlines the effects of IGF1R activation in tumour cells, and will describe the strategies that are available to block IGF signalling, both as investigational tools and as novel anti-cancer therapeutics. Design of specific IGF1R inhibitors has been problematic due to close homology with the insulin receptor, but recently it has proved possible to design selective IGF1R inhibitors. These compounds and IGF1R antibodies are showing promise in preclinical models of human cancer, and several agents are now in early phase clinical trials. Both classes of agents affect insulin receptor signalling, either by direct kinase inhibition or antibody-induced insulin receptor downregulation. This effect may lead to clinical toxicity, but could be therapeutically beneficial in blocking signalling via variant insulin receptors capable of a mitogenic response to IGF-II. Specificity for IGF1R targeting can be achieved by antisense and siRNA-mediated IGF1R downregulation; these approaches have undoubted utility as research tools, and may in future generate nucleic-acid-based therapeutics. It will be important to use data from preclinical and early clinical trials to establish the molecular correlates of sensitivity to IGF1R blockade, and the optimum means of combining this new approach with standard treatment modalities.
The insulin-like growth factor (IGF) signalling axis involves the coordinated function of two ligands, IGF-I and IGF-II, three cell surface receptors, at least six high affinity binding proteins and binding protein proteases (Baserga et al. 1997, Pollak et al. 2004). This signalling axis plays a pivotal role in normal growth and development (Liu et al. 1993), and is also implicated in mediating many aspects of the malignant phenotype in a variety of human malignancies (Khandwala et al. 2000, Moschos & Mantzoros 2002, Bohula et al. 2003a). This review will focus on the role of the type 1 IGF receptor (IGF1R) in tumour biology, summarising the principal effects of IGF1R signalling, and the strategies that are currently being developed for blocking IGF1R action as anti-cancer therapy.
Signalling through type 1 IGF receptor
The IGF receptor family comprises three trans-membrane proteins. The IGF1R is a transmembrane receptor tyrosine kinase (RTK) that is responsible for mediating IGF bioactivity. The IGF1R gene is located on chromosome 15q26 and encodes a single polypeptide of 1367 amino acids that is constitutively expressed in most cells. The IGF1R and insulin receptor (IR) share approximately 70% amino acid homology, and their genes exhibit striking homology in terms of size and exon organisation (Ullrich et al. 1986). There are two splice variants of the IR: the classical isoform (IR-B) regulates glucose uptake and is expressed principally in liver, muscle and adipose tissue, whereas the exon 11 variant (isoform A) is expressed by foetal tissues and some tumours, and binds IGF-II with high affinity to promote proliferation and protection from apoptosis (Sciacca et al. 2002). The type 2 IGF receptor is a monomeric transmembrane protein that lacks intrinsic signalling activity. It acts as a negative regulator of IGF activity by sequestration, endocytosis and degradation of IGF-II (Nolan et al. 1990).
Both the IGF1R and IR are synthesised as single-chain pre-proreceptors, with a 30-residue signal peptide that is cleaved co-translationally. En route to the membrane, proreceptors are glycosylated, folded and dimerised under the guidance of chaperone proteins before transport to the Golgi apparatus. Here, the receptors are processed at a tetrabasic Arg-Lys-Arg-Arg furin-protease cleavage site to generate the α (130–135 kDa) and β (90–97 kDa) subunits of the mature receptor (Ullrich et al. 1986). These assemble into disulphide-linked tetramers comprised of two extracellular α subunits, and two β subunits that contain extracellular and transmembrane domains and a cytoplasmic region, which includes the tyrosine kinase domain (Ward & Garrett 2004). IGF1R proreceptors can form hetero-dimeric complexes with insulin proreceptors, reflecting their close structural homology (Pandini et al. 1999).
The IGF1R is frequently overexpressed in tumours, including melanomas, cancers of the colon, pancreas and, as we have shown, prostate and kidney (Hellawell et al. 2002, Yuen et al. unpublished observations and reviewed in Bohula et al. 2003a). IGF1R overexpression can be the result of loss of tumour suppressors, including wild-type p53, BRCA1 and VHL (Werner & Roberts 2003, JSP Yuen, M Cockman, M Sullivan, A Protheroe, C Pugh, H Werner & VM Macaulay, unpublished observations). However, it must be acknowledged that the expression of the IGF1R is virtually ubiquitous, and the principal justification for considering the IGF1R as an anti-cancer treatment target is its function. The IGF1R is activated by IGFs present in the extracellular milieu, derived from endocrine, paracrine or autocrine sources. IGF1R activation leads to autophosphorylation on tyrosines 1131, 1135 and 1136 in the kinase domain, followed by phosphorylation of juxtamembrane tyrosines and carboxy-terminal serines. This is followed by recruitment of specific docking intermediates, including insulin-receptor substrate-1 (IRS-1), Shc and 14-3-3 proteins (Baserga et al. 1997, Pollak et al. 2004). These molecules link the IGF1R to diverse signalling pathways, allowing induction of growth, transformation, differentiation and protection against apoptosis.
The importance of the IGF1R in normal mammalian development is clear from studies in mice lacking functional receptors. IGF1R null mice are 45% of the size of wild-type littermates at birth, and die shortly due to severe organ hypoplasia (Liu et al. 1993). Mouse embryonic fibroblasts cultured from IGF1R null mice (R-cells) grow more slowly than wild-type fibroblasts, and are unable to proliferate under anchorage-independent conditions (Sell et al. 1994). Anchorage-independence in vitro is indicative of transformation potential and tumorigenicity in vivo, and it has been known for some time that R-cells are refractory to transformation by many oncogenes, including SV40 large T, activated RAS, epidermal growth factor receptor (EGFR), IR or platelet derived growth factor receptor (PDFGR) (Baserga et al. 1997). Recent studies indicate that the requirement of the IGF1R for transformation is related to its ability to tyrosine phosphorylate IRS-1 (DeAngelis et al. 2006).
Paradoxically, given its role in proliferation, the IGF1R can also induce differentiation in adipocytes, myoblasts, osteoblasts, neurons and haemopoietic cells (Valentinis et al. 1999). Following stimulation with IGFs, the SH2-containing adaptor protein Shc is recruited to phosphorylated tyrosine residue (Tyr950) in the juxtamembrane region of the IGF1R, and transduces differentiation signals through the Grb2/RAS/RAF/MAPK pathway. Since IRS-1 binds to the same phosphorylated Tyr950 and leads to proliferative phenotype, it is possible that commitment to proliferation or differentiation depends on competition between these two adaptor proteins for access to this binding site on the IGF1R. This hypothesis is supported by studies in murine haemopoietic and human prostate cancer cells that express low levels of IGF1R and lack IRS-1 (Reiss et al. 2000, Valentinis et al. 2000).
IGF1R activation protects cells from a variety of apoptosis-inducing agents, including osmotic stress, hypoxia and anti-cancer drugs (Dunn et al. 1997, Peretz et al. 2002). The level of expression offunctional IGF1R appears to be a critical determinant of resistance to apoptosis in vitro and in vivo (Resnicoff et al. 1995). The IGF1R suppresses apoptosis primarily through the phosphoinositide 3-kinase (PI3K) pathway. Following phosphorylation of IRS-1 by the activated IGF1R, PI3K is activated by binding its regulatory subunit to IRS-1 (Baserga et al. 1997). This interaction leads to an increase in the levels of phosphatidylinositol 3,4,5-triphosphate (PIP3), which leads to recruitment and activation of phosphoinositide-dependent kinase-1 and AKT/protein kinase B. The activated AKT induces inhibitory phosphorylation of pro-apoptotic factors, including the Bcl-2 family member Bad, members of the forkhead transcription factor family and caspase 9. In addition, AKT activation leads to increased expression of anti-apoptotic proteins, including Bcl-2, Bcl-x and NF-κB (Brazil et al. 2004). IGF effects on survival are also mediated by the activation of extracellular signal-regulated kinases (ERKs) and p38 MAPK, and by 14-3-3-dependent mitochondrial translocation of RAF (Peruzzi et al. 1999).
Since a functional IGF1R is required for transformation and promotes tumour cell growth and survival, it may be important for the early stages of tumour establishment. In recent years, it has also become apparent that IGF1R activation is capable of inducing additional effects that could influence the behaviour of tumours at a later stage in their natural history. IGF1R activation can protect cancer cells against the relatively hypoxic and nutrient-deprived microenvironment that pertains in large tumours (Peretz et al. 2002). Furthermore, IGFs and insulin enhance the hypoxic response by activation of AKT and MAPK signalling, leading to stabilisation of hypoxia-inducible factors HIF-1α and HIF-2α, and upregulation of vascular endothelial growth factor (Treins et al. 2005).
IGF1R activation or overexpression is associated with an increased propensity for invasion and metastasis. This effect is mediated by multiple signalling intermediates that influence invasive potential. IGF-induced phosphorylation of IRS-1 can influence the interaction between E-cadherin and β-catenin, enhancing β-catenin transcriptional activity and disconnecting E-cadherin from the actin cytoskeleton (Playford et al. 2000). Loss of E-cadherin expression or function is well recognised as causing disruption of cell–cell contacts and release of invasive tumour cells from primary epithelial tumours (Hazan et al. 2004). Similarly, tumour cell motility and invasive potential are influenced by crosstalk between the IGF axis and integrins (Shen et al. 2006), and by IGF-induced secretion of matrix metalloproteinases (Zhang et al. 2003). The importance of these functions is highlighted by the finding that IGF1R overexpression confers an invasive, metastatic phenotype in a murine model of pancreatic cancer (Lopez & Hanahan 2002). Furthermore, IGF1R inhibition is capable of inhibiting metastasis in vivo (Dunn et al. 1998, Samani et al. 2001).
Molecular strategies to target the IGF system
Receptor tyrosine kinases are now well-recognised targets for the treatment of cancer. While development of EGFR inhibitors began approximately 15 years ago, only recently, the IGF axis has been recognised as a drug target. The IGF axis could potentially be blocked at several different levels, including interference with the expression and function of ligands, binding proteins and receptors (Pollak et al. 2004, Yee 2006). This review will focus on the approaches currently available to inhibit IGF1R activity. However, it is important to remember that IGF-II bioactivity can also be mediated by tumour expression of IR-A and IGF1R:IR hybrid receptors. While the functional consequences of hybrid receptors have yet to be clarified, they appear to behave more like IGF1Rs than IRs and may represent an alternative route for IGF signalling (Frasca et al. 2003). While crossreactivity for the IR may be a disadvantage in terms of toxicity, it is probable that blockade of tumour cell IR-A could contribute to any anti-tumour effect of IGF1R inhibition, although this has yet to be proven.
Small molecule inhibitors
The design of IGF1R tyrosine kinase inhibitors is complicated by the fact that the kinase domain of the IGF1R shares 85% homology with that of the IR, and the ATP binding cleft is 100% conserved (Ullrich et al. 1986). However, there are subtle differences that may be exploited in drug design, and several groups are now generating agents that show promise as IGF1R inhibitors (Favelyukis et al. 2001, Garber 2005). NVP-AEW541 and NVP-ADW742 (Novartis) are inhibitors of the IGF1R kinase that appear to be equipotent for IGF1R and IR inhibition in vitro, but show selectivity for the IGF1R in intact cells. These agents inhibit tumour growth in models of fibrosar-coma, myeloma and Ewing’s sarcoma, and also enhance tumour cell chemosensitivity (Garcia-Echeverria et al. 2004, Mitsiades et al. 2004, Scotlandi et al. 2005). Unfortunately, the further development of these highly promising agents has been thwarted by toxicity, although this is thought not to be mechanism-based (Garber 2005). Inhibitors that bind to the active site of the enzyme have also been generated by Merck (Bell et al. 2005) and Bristol–Myers Squibb (Haluska et al. 2006); the former is selective for the IGF1R, the latter is equipotent for inhibition of the IGF1R and IR.
While inhibitors that bind in the ATP binding cleft are unlikely to show absolute specificity, inhibitors of substrate phosphorylation show greater potential for specific IGF1R inhibition. Several classes of catechol mimic act as substrate-competitive inhibitors of the IGF1R, and have been shown to inhibit IGF signalling and growth of human breast and prostate cancer cells in vitro (Blum et al. 2003, Youngren et al. 2005). One such agent, nordihydroguaiaretic acid, also blocks activation of the Her2 kinase, inhibits growth of IGF1R and Her2-positive murine breast cancer in vivo (Youngren et al. 2005), and is now in clinical trials. Larsson’s group have developed the cyclolignan picropodophyllin (PPP), which inhibits tyrosine phosphorylation of Y1136 in the activation loop of the IGF1R kinase domain. PPP does not affect the insulin receptor or compete with ATP in in vitro kinase assays, suggesting that it may inhibit IGF1R autophosphorylation at the substrate level (Girnita et al. 2004). This agent has been shown to induce tumour regression and inhibition of metastasis in several models of human cancer, and in vitro studies suggest development of only limited resistance in tumour cells after long-term PPP exposure (Girnita et al. 2006, Stromberg et al. 2006, Vasilcanu et al. 2006). Diarylureas also inhibit IGF1R activation, although their precise mode of action is presently unknown (Gable et al. 2006).
Neutralising antibodies that block receptor–ligand interactions are attractive agents against trans-membrane oncoproteins. Antibody-based therapies have great potential for clinical use, and the best example of this is trastuzumab (Herceptin; Genentech, San Francisco, CA, USA). This is now an established component of the treatment of Her2-positive metastatic breast cancer, and has recently shown significant activity in the adjuvant setting (Plosker & Keam 2006). The first available blocking IGF1R antibody was the monoclonal αIR3; indeed, this was the first antibody to any RTK to demonstrate anti-cancer activity in vivo (Arteaga et al. 1989). Subsequently, αIR3 has been shown to inhibit the growth of many different tumour types in vitro and in vivo (Macaulay et al. 1990, Furlanetto et al. 1993, Lahm et al. 1994, Bohula et al. 2003a). In cells that overexpress the IGF1R, αIR3 can act as an IGF1R agonist. Similar agonist effects have been observed in MCF-7 breast cancer cells treated with a humanised single chain anti-IGF1R antibody (IGF1R scFv-Fc), but nonetheless this antibody is capable of blocking growth of MCF-7 xenografts in vivo, and this has been shown to be related to antibody-induced IGF1R downregulation (Sachdev et al. 2003). In fact, this and other IGF1R antibodies, while being IGF1R-specific, can also induce insulin receptor downregulation due to co-downregulation of insulin receptors in IR:IGF1R dimers, or through endocytosis of insulin receptors in close proximity to IGF1R in membrane lipid rafts (Sachdev et al. 2006). The ability to induce IGF1R downregulation has been shown to be an integral component of the antitumour activity of several blocking antibodies including those developed by Pfizer (CP-751,871), (Cohen et al. 2005), Schering Plough (Wang et al. 2005) and ImClone (Wu et al. 2005). Most of these are fully humanised monoclonals that have shown significant activity in a range of tumour models, and several are in early clinical trials.
Expression of IGF1R-inhibitory proteins
Dominant-negative proteins interfere with the function of wild-type protein by either direct binding or competition for binding partners. The first described dominant negative IGF1R was 486/STOP, a secreted IGF1R with an intact ligand-binding domain that competes for ligand binding (D’Ambrosio et al. 1996). A truncated receptor (950/STOP) that retains the transmembrane domain also acts as a dominant negative by forming non-functional heterodimers with endogenous IGF1R. Adenoviral or plasmid-based delivery of dominant negative IGF1R shows anti-tumour effects and chemosensitisation in several tumour models (Dunn et al. 1998, Bohula et al. 2003a, Min et al. 2005).
Signalling via the IGF axis can be inhibited by forced expression of the IGF2R, leading to sequestration of IGF-II. Expression of full-length IGF2R inhibits the growth of human breast cancer xenografts in nude mice, and delays the onset of mammary tumours induced by IGF2 overexpression (Lee et al. 2003, Wise & Pravtcheva 2006). Expression of a soluble IGF2R inhibits the growth of choriocarcinomas in vivo (O’Gorman et al. 2002) and suppresses IGF-II-dependent adenoma growth in the ApcMin/+ mouse model of human colorectal cancer (Harper et al. 2006).
These approaches could be translated to clinical therapy, either through administration of recombinant IGF1R dominant negative or IGF2R protein, or through a gene therapy approach. From a therapeutic perspective, gene therapy strategies are presently limited by low efficiency of gene delivery and stability, although could become a realistic option for IGF1R blockade if highly stable, tissue-specific plasmid or viral constructs can be generated (Larsson et al. 2005).
The antisense approach involves introduction into the cell of antisense RNA molecules or antisense oligonucleotides (ASOs) which are binding to a complementary sequence in a target transcript. Gene expression is inhibited as a result of RNase H-mediated mRNA cleavage or translation block (Crooke 1999). Although there are no specific sequence restrictions for the use of antisense, this approach is strongly influenced by secondary structure in the target transcript, such that only 5–10% of any given mRNA is accessible to hybridise with ASOs (Sohail & Southern 2000). The translational start site is the conventional target for ASO design because this region of the transcript usually lacks extensive secondary structure to facilitate translation initiation (Crooke 1999). Antisense-mediated IGF1R downregulation has been shown to decrease cell growth, induce apoptosis and delay tumour formation in vivo in a number of experimental models (reviewed in Salisbury & Macaulay 2003).
There has been limited progress in the implementation of antisense technology as therapy (Gleave & Monia 2005). Phosphorothioate ASOs, in which non-bridging oxygens are replaced by sulphur, are more resistant to nuclease attack than unmodified (phosphodiester) DNA, while retaining the capacity to induce target-specific RNase H-mediated cleavage (Stein & Cohen 1988). Modification of the sugar ring confers increased nuclease resistance but does not allow activation of RNase H, a problem that can be circumvented by combining such modifications with unmodified or phosphorothioate DNA to create mixed backbone oligonucleotides (Wang et al. 2002). Several antisense agents have shown evidence of clinical activity, including those targeting Bcl-2 (Piro 2004) and c-Raf (Dritschilo et al. 2006). However, the efficacy of systemic administration of ASOs is limited by problems, including limited cellular uptake and non-specific toxicity due to poor target sequence specificity and protein binding (Krieg & Stein 1995). One route to overcome these problems is to use topical application, which has been developed for use of IGF1R antisense oligonucleotides to treat cutaneous hyperplasia in psoriasis (Fogarty et al. 2002). Another option is to use ex vivo transfection; applying this strategy to IGF1R downregulation in glioblastoma was shown to elicit a host response leading to protection from unmodified tumour cells. This approach was effective in an animal model of glioblastoma (Resnicoff et al. 1994), and also, impressively, in end-stage patients with malignant astrocytoma, (Andrews et al. 2001), which to date is the only published clinical study of IGF1R targeting.
RNA interference has been recognised since 1998 as a process in which long double-stranded RNAs (dsRNAs) are cleaved by the enzyme Dicer, a member of the RNase III family of dsRNA-specific ribonucleases, into short 21–23 nucleotide fragments containing two nucleotide single-stranded 3′ overhangs. The cleavage products, short interfering RNAs (siRNAs), are incorporated into a multi-protein RNA-inducing silencing complex (RISC) in which the antisense strand of the unwound siRNA acts as a guide to target the cleavage of a complementary mRNA (Fire et al. 1998). In mammalian cells, long dsRNAs can induce apoptosis, through the activation of dsRNA-dependent protein kinase and stimulation of interferon expression. In 2001, Tuschl’s laboratory demonstrated that this problem can be circumvented by direct introduction into the cell of siRNAs, rather than long dsRNAs, and this has enabled the use of gene silencing as a powerful research tool in mammalian cell biology (Elbashir et al. 2001).
When gene silencing was first introduced for use in mammalian cells, it was clear that not all siRNAs mediated effective knockdown of target proteins (Elbashir et al. 2001). The precise nature of the molecular interaction between siRNAs and target mRNA, and hence the factors regulating efficacy, were unclear. Given the known influence of transcript accessibility on ASO efficacy, we investigated effects of mRNA folding on the ability of siRNAs to silence the IGF1R. We used an oligonucleotide array-based screen to design siRNAs capable of interacting with accessible regions of IGF1R but not IR mRNA. The results indicated a clear correlation between the ability of siRNAs to block IGF1R expression and the accessibility of the target sequence; duplexes targeting inaccessible regions of the transcript caused minor IGF1R down-regulation, while siRNAs homologous to accessible targets induced profound sequence-specific IGF1R gene silencing and blocked IGF signalling in IGF-responsive human breast cancer cells. The same pattern of differential efficacy was observed in a range of human and murine cell lines, suggesting that the structural features dictating mRNA secondary structure are robust and conserved between different cell lines and species (Bohula et al. 2003b). The influence of mRNA folding has subsequently been confirmed by others (Kretschmer-Kazemi Far & Sczakiel 2003), and is consistent with a requirement for base pairing between mRNA and the antisense strand of the duplex. This study shed light on the mode of action of siRNAs, and also generated a series of potent and sequence-specific siRNAs that induce profound IGF1R knockdown without influencing IR expression. These agents are capable of blocking IGF1R expression, IGF signalling and survival of human and murine tumour cells (Bohula et al. 2003b, Rochester et al. 2005, Yeh et al. 2006).
While mRNA folding remains difficult to model accurately, subsequent investigations have clarified the nature of the primary sequence requirements for effective gene silencing. These include low G/C content, lack of inverted repeats, low internal stability at the 3′ end of the duplex and U at position 10; activated RISC cleaves target mRNA between nucleotides 10 and 11 relative to the 5′ end of the complementary targeting strand, and RISC, like most endonucleases, prefers to cleave 3′ of U (Reynolds et al. 2004). Careful siRNA design can also reduce the potential for induction of the interferon response by immunostimulatory motifs in synthetic siRNAs (Judge et al. 2005), and induction of off-target effects, in which an siRNA blocks translation of an mRNA to which it is imperfectly matched (Jackson & Linsley 2004).
Unmodified siRNAs are unsuitable for use in vivo, due to poor cellular uptake and susceptibility to degradation. Several studies have reported anti-tumour effects following in vivo administration of unmodified, naked siRNA (Filleur et al. 2003, Duxbury et al. 2004), but stabilised siRNAs would no doubt offer clear advantages for therapeutic use. Recent studies have used siRNAs incorporating modification of the DNA backbone and/or sugar ring, cholesterol conjugation or incorporation into stabilised nanoparticles. Encouragingly, it is now possible to achieve effective knockdown of a target gene following systemic administration of modified siRNAs in vivo (Zimmermann et al. 2006).
Predicting molecular determinants of sensitivity to IGF1R inhibition
Lessons learned from clinical studies of EGFR inhibitors such as gefitinib may provide valuable information to guide introduction of IGF1R inhibitors. In contrast to the situation with Herceptin, where Her2 overexpression correlates with response, EGFR expression levels do not correlate with response to EGFR inhibitor therapy. It is now clear that response to EGFR inhibition occurs in the context of acquired mutations in the ATP binding site of the EGFR kinase domain, and these occur typically in Asian women with adenocarcinomas of the lung, who have never smoked (Ready 2005). Recent studies have failed to identify mutations in the IGF1R kinase domain, equivalent to those in the EGFR, which might correlate with response to IGF1R TKIs. It has been suggested that autocrine production of IGF-II may serve as a marker of tumours that depend on IGF signalling (Pollak et al. 2004). Now that clinical trials have commenced, it will be possible to use genomic and proteomic techniques to identify molecular signatures of response, as has been accomplished for EGFR inhibitors (Alvarez et al. 2006).
It would be advantageous to anticipate, and if possible, circumvent, potential mechanisms of resistance to novel therapies, since this will allow more effective use in selected patient subgroups. Targeting of any growth factor receptor could be rendered ineffective by compensatory signalling via other receptor tyrosine kinases, and/or by the presence of mutations that activate downstream signalling pathways mediating growth and survival. Therefore, we investigated effects of IGF1R gene silencing (Bohula et al. 2003b) in situations that have implications for the range of tumour types that may be susceptible to IGF1R targeting in the clinic. The EGF system shares significant crosstalk with the IGF system (Adams et al. 2004), and in primary glioblastoma cells the IGF1R has been shown to mediate resistance to anti-EGFR therapy (Chakravarti et al. 2002). This prompted us to investigate whether apoptosis induced by IGF1R knockdown could be abrogated by compensatory signalling through the EGFR. In tumour cells that express relatively high EGFR, IGF1R knockdown induced no change in absolute levels of EGFR, but induced significant prolongation of EGF-induced phosphorylation of the EGFR and of jun N-terminal protein kinase (JNK), ERKs and signal transducers and activators of transcription 5 (STAT5). However, EGF-induced activation of AKT, p38 and STAT3 was suppressed, apoptosis was enhanced and cell survival inhibited, indicating that the EGF axis was unable to compensate for IGF1R loss, at least in vitro. However, it is possible that EGFR hyperphosphorylation following IGF1R knockdown might confer resistance in a small subpopulation of surviving cells in vivo. Although we found no evidence that the EGFR can compensate for IGF1R loss in vitro, it will be important, when IGF1R inhibitors reach the clinic, to assess the clinical relevance of these findings, which could potentially mediate resistance to IGF1R blockade. This work led us to identify a novel interaction between the IGF1R and EGFR, detectable in cultured cells and primary breast cancers, with evidence that the EGFR sequesters IGF1R and protects it from ubiquitin-mediated proteasomal degradation (J Riedemann, M Takiguchi, M Sohail & VM Macaulay, unpublished observations). These results are analogous to the reported effect of Her2 (ErbB2) on EGFR expression. Her2 contributes to an overall increase in EGFR half-life by reducing the rate of EGFR internalisation (Baulida et al. 1996), decreasing degradation and enhancing the rate of recycling (Worthylake et al. 1999). The clinical significance of an interaction between the IGF1R and EGFR is presently unclear; a similar interaction has been detected between Her2 and IGF1R and correlates with resistance to trastuzumab (Nahta et al. 2005).
In addition, we investigated whether sensitivity to IGF1R blockade is influenced by the presence of mutations that activate signalling pathways downstream of the IGF1R. In prostate cancer, there is a high incidence of deregulation of the PI3K-AKT pathway due to loss/mutation of the tumour suppressor PTEN (McMenamin et al. 1999). IGF1R gene silencing was capable of blocking cell survival, not only in prostate cancer cells that express wild-type PTEN, but also in cells that lack functional PTEN. The ability to inhibit survival in this situation was shown to be related to the fact that AKT phosphorylation remained sensitive to IGFs, despite the deregulation of AKT activation that follows PTEN loss, and also that IGF1R knockdown was capable of suppressing signalling through multiple pathways that influence survival (Rochester et al. 2005). We also assessed the effects of IGF1R knockdown in cells that show activation of components of the RAS/RAF/MAPK pathway. Mutations in NRAS or BRAF occur in 15 and 70% of human melanomas respectively, representing a more challenging obstacle to IGF1R targeting, since ERK phosphorylation is completely ligand-independent (Davies et al. 2002). Despite this, IGF1R knockdown inhibited survival and enhanced apoptosis with similar effects in isogenic cells over-expressing wild-type or mutant BRAF (Yeh et al. 2006). Thus, it appears that activation of the AKT or MAPK pathways is not sufficient to render tumour cells refractory to intervention at the receptor level. This is in contrast to reports indicating that resistance to the EGFR inhibitor Iressa can be conferred by activating KRAS mutation (Pao et al. 2005) and reduced PTEN expression (Kokubo et al. 2005). These findings are encouraging for the strategy of IGF1R targeting, and have implications for the spectrum of tumour types that may be susceptible to IGF1R inhibition in the clinic.
IGF1R inhibition as a route to chemosensitisation
IGFs are known to protect tumour cells against killing by cytotoxic drugs (Dunn et al. 1997). This effect can be attributed to the well-recognised ability of the IGF axis to suppress apoptosis, and also to an apparent ability to influence aspects of the DNA damage response (Macaulay et al. 2001, Trojanek et al. 2003). Consistent with this, sensitivity to chemotherapy and irradiation can be enhanced by various approaches to block the IGF axis, including small molecule inhibitors (Scotlandi et al. 2005), antibodies (Cohen et al. 2005), dominant negative IGF1R (Min et al. 2005), antisense (Macaulay et al. 2001) and siRNA (Rochester et al. 2005, Yeh et al. 2006). Initial clinical trials are now addressing the toxicity and efficacy of IGF1R blockade as monotherapy, and it will be important to follow these studies by investigating the potential of IGF1R inhibitors to influence chemo-sensitivity. The success of this approach should be judged not only by changes in short-term response rates, but also in the ability of IGF1R blockade to influence survival in combination with chemotherapy.
The development of IGF1R inhibitors and antibodies is now allowing testing of the clinical potential of the IGF1R as an anticancer treatment target. Initial studies will clarify the extent to which IGF1R inhibition can be tolerated in normal tissues, and the magnitude of toxicity, particularly glucose-intolerance, due to cross-reactivity with the insulin receptor. It will be important to collect data on the cellular and molecular characteristics of responding and non-responding tumours, to clarify the features associated with sensitivity to this approach. Even if IGF1R inhibitors prove to have limited activity as single agents, it will be important to evaluate effects on disease progression and chemosensitivity, given the important role of the IGF1R in invasion and metastasis, and in protecting from killing induced by cytotoxic drugs.
Research in the authors’ laboratory is supported by Cancer Research UK, Prostate Cancer Research Foundation, Prostate Cancer Research Campaign UK, British Urological Foundation and Kidney Research UK. We are grateful to Stephan Feller for comments on the manuscript. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
AlvarezJV Greulich H Sellers WR Meyerson M & Frank DA 2006 Signal transducer and activator of transcription 3 is required for the oncogenic effects of non-small-cell lung cancer-associated mutations of the epidermal growth factor receptor. Cancer Research663162–3168.
AndrewsDW Resnicoff M Flanders AE Kenyon L Curtis M Merli G Baserga R Iliakis G & Aiken RD 2001 Results of a pilot study involving the use of an antisense oligodeoxynucleotide directed against the insulin-like growth factor type I receptor in malignant astrocytomas. Journal of Clinical Oncology192189–2200.
BellIM Stirdivant SM Ahern J Culberson JC Darke PL Dinsmore CJ Drakas RA Gallicchio SN Graham SL Heimbrook DC et al.2005 Biochemical and structural characterization of a novel class of inhibitors of the type 1 insulin-like growth factor and insulin receptor kinases. Biochemistry449430–9440.
BohulaEA Salisbury AJ Sohail M Playford MP Riedemann J Southern EM & Macaulay VM 2003b The efficacy of small interfering RNAs targeted to the type 1 insulin-like growth factor receptor (IGF1R) is influenced by secondary structure in the IGF1R transcript. Journal of Biological Chemistry27815991–15997.
ChakravartiA Loeffler JS & Dyson NJ 2002 Insulin-like growth factor receptor I mediates resistance to anti-epidermal growth factor receptor therapy in primary human glioblastoma cells through continued activation of phosphoinositide 3-kinase signaling. Cancer Research62200–207.
CohenBD Baker DA Soderstrom C Tkalcevic G Rossi AM Miller PE Tengowski MW Wang F Gualberto A Beebe JS et al.2005 Combination therapy enhances the inhibition of tumor growth with the fully human anti-type 1 insulin-like growth factor receptor monoclonal antibody CP-751871. Clinical Cancer Research112063–2073.
DritschiloA Huang CH Rudin CM Marshall J Collins B Dul JL Zhang C Kumar D Gokhale PC Ahmad A et al.2006 Phase I study of liposome-encapsulated c-raf antisense oligodeoxyribonucleotide infusion in combination with radiation therapy in patients with advanced malignancies. Clinical Cancer Research121251–1259.
FilleurS Courtin A Ait-Si-Ali S Guglielmi J Merle C Harel-Bellan A Clezardin P & Cabon F 2003 SiRNA-mediated inhibition of vascular endothelial growth factor severely limits tumor resistance to antiangiogenic thrombospondin-1 and slows tumor vascularization and growth. Cancer Research633919–3922.
GableKL Maddux BA Penaranda C Zavodovskaya M Campbell MJ Lobo M Robinson L Schow S Kerner JA Goldfine ID et al.2006 Diarylureas are small-molecule inhibitors of insulin-like growth factor I receptor signaling and breast cancer cell growth. Molecular Cancer Therapeutics51079–1086.
GirnitaA All-Ericsson C Economou MA Astrom K Axelson M Seregard S Larsson O & Girnita L 2006 The insulin-like growth factor-I receptor inhibitor picropodophyllin causes tumor regression and attenuates mechanisms involved in invasion of uveal melanoma cells. Clinical Cancer Research121383–1391.
KokuboY Gemma A Noro R Seike M Kataoka K Matsuda K Okano T Minegishi Y Yoshimura A Shibuya M et al.2005 Reduction of PTEN protein and loss of epidermal growth factor receptor gene mutation in lung cancer with natural resistance to gefitinib (IRESSA). British Journal of Cancer921711–1719.
LahmH Amstad P Wyniger J Yilmaz A Fischer JR Schreyer M & Givel JC 1994 Blockade of the insulin-like growth-factor-I receptor inhibits growth of human colorectal cancer cells: evidence of a functional IGF-II-mediated autocrine loop. International Journal of Cancer58452–459.
MacaulayVM Salisbury AJ Bohula EA Playford MP Smorodinsky NI & Shiloh Y 2001 Downregulation of the type 1 insulin-like growth factor receptor in mouse melanoma cells is associated with enhanced radio-sensitivity and impaired activation of Atm kinase. Oncogene204029–4040.
MinY Adachi Y Yamamoto H Imsumran A Arimura Y Endo T Hinoda Y Lee CT Nadaf S Carbone DP et al.2005 Insulin-like growth factor I receptor blockade enhances chemotherapy and radiation responses and inhibits tumour growth in human gastric cancer xenografts. Gut54591–600.
MitsiadesCS Mitsiades NS McMullan CJ Poulaki V Shringarpure R Akiyama M Hideshima T Chauhan D Joseph M Libermann TA et al.2004 Inhibition of the insulin-like growth factor receptor-1 tyrosine kinase activity as a therapeutic strategy for multiple myeloma other hematologic malignancies and solid tumors. Cancer Cell5221–230.
PandiniG Vigneri R Costantino A Frasca F Ippolito A Fujita-Yamaguchi Y Siddle K Goldfine ID & Belfiore A 1999 Insulin and insulin-like growth factor-I (IGF-I) receptor overexpression in breast cancers leads to insulin/IGF-I hybrid receptor overexpression: evidence for a second mechanism of IGF-I signaling. Clinical Cancer Research51935–1944.
ResnicoffM Sell C Rubini M Coppola D Ambrose D Baserga R & Rubin R 1994 Rat glioblastoma cells expressing an antisense RNA to the insulin-like growth factor-1 (IGF-1) receptor are nontumorigenic and induce regression of wild-type tumors. Cancer Research542218–2222.
SachdevD Li SL Hartell JS Fujita-Yamaguchi Y Miller JS & Yee D 2003 A chimeric humanized single-chain antibody against the type I insulin-like growth factor (IGF) receptor renders breast cancer cells refractory to the mitogenic effects of IGF-I. Cancer Research63627–635.
SamaniAA Fallavollita L Jaalouk DE Galipeau J & Brodt P 2001 Inhibition of carcinoma cell growth and metastasis by a vesicular stomatitis virus G-pseudotyped retrovector expressing type I insulin-like growth factor receptor antisense. Human Gene Therapy121969–1977.
SellC Dumenil G Deveaud C Miura M Coppola D DeAngelis T Rubin R Efstratiadis A & Baserga R 1994 Effect of a null mutation of the insulin-like growth factor I receptor gene on growth and transformation of mouse embryo fibroblasts. Molecular and Cellular Biology143604–3612.
StrombergT Ekman S Girnita L Dimberg LY Larsson O Axelson M Lennartsson J Hellman U Carlson K Osterborg A et al.2006 IGF-1 receptor tyrosine kinase inhibition by the cyclolignan PPP induces G2/M-phase accumulation and apoptosis in multiple myeloma cells. Blood107669–678.
TrojanekJ Ho T Del Valle L Nowicki M Wang JY Lassak A Peruzzi F Khalili K Skorski T & Reiss K 2003 Role of the insulin-like growth factor I/insulin receptor substrate 1 axis in Rad51 trafficking and DNA repair by homologous recombination. Molecular and Cellular Biology237510–7524.
UllrichA Gray A Tam AW Yang-Feng T Tsubokawa M Collins C Henzel W Le Bon T Kathuria S Chen E et al.1986 Insulin-like growth factor I receptor primary structure: comparison with insulin receptor suggests structural determinants that define functional specificity. EMBO Journal52503–2512.
ValentinisB Romano G Peruzzi F Morrione A Prisco M Soddu S Cristofanelli B Sacchi A & Baserga R 1999 Growth and differentiation signals by the insulin-like growth factor 1 receptor in hemopoietic cells are mediated through different pathways. Journal of Biological Chemistry27412423–12430.
ValentinisB Navarro M Zanocco-Marani T Edmonds P McCormick J Morrione A Sacchi A Romano G Reiss K & Baserga R 2000 Insulin receptor substrate-1 p70S6K and cell size in transformation and differentiation of hemopoietic cells. Journal of Biological Chemistry27525451–25459.
WangH Wang S Nan L Yu D Agrawal S & Zhang R 2002 Antisense anti-MDM2 mixed-backbone oligonucleotides enhance therapeutic efficacy of topoisomerase I inhibitor irinotecan in nude mice bearing human cancer xenografts: in vivo activity and mechanisms. International Journal of Oncology20745–752.
WangY Hailey J Williams D Wang Y Lipari P Malkowski M Wang X Xie L Li G Saha D et al.2005 Inhibition of insulin-like growth factor-I receptor (IGF-IR) signaling and tumor cell growth by a fully human neutralizing anti-IGF-IR antibody. Molecular Cancer Therapeutics41214–1221.
WuJD Odman A Higgins LM Haugk K Vessella R Ludwig DL & Plymate SR 2005In vivo effects of the human type I insulin-like growth factor receptor antibody A12 on androgen-dependent and androgen-independent xenograft human prostate tumors. Clinical Cancer Research113065–3074.
YoungrenJF Gable K Penaranda C Maddux BA Zavodovskaya M Lobo M Campbell M Kerner J & Goldfine ID 2005 Nordihydroguaiaretic acid (NDGA) inhibits the IGF-1 and c-erbB2/HER2/neu receptors and suppresses growth in breast cancer cells. Breast Cancer Research and Treatment9437–46.
YuenJSP Cockman M Sullivan M Protheroe A Pugh C Werner H & Macaulay VM. The VHL tumour suppressor inhibits expression of the IGF1R and its loss contributes to IGF1R upregulation in clear cell renal cancer (unpublished observations).