Targeting of Insulin-Like Growth Factor Type 1 Receptor in Ewing Sarcoma: Unfulfilled Promise or a Promising Beginning?

The selective targeting of oncogenic proteins and its translation into disease-specific clinical trials have largely driven the effort to identify new, singular targets that can be exploited for cancer therapy. Targeting the insulin-like growth factor type 1 receptor (IGF-1R) in Ewing sarcoma (ES) and the Ewing Sarcoma Family of Tumors (ESFT) certainly looked to fit the bill: a strong biologic rationale with a disease-defining translocation dependent on IGF signaling for transformation, preclinical evidence of antitumor effects with IGF-1R inhibition, and even impressive signals of clinical activity with IGF-1R inhibitors from early phase I studies. Yet in the accompanying articles, Pappo et al¹ and Juergens et al² report relatively disappointing results from two phase II trials evaluating IGF-1R–targeting monoclonal antibodies (R1507 [Roche, Basel, Switzerland] and figitumumab [Pfizer, New London, CT]), in which the treatment of 222 patients resulted in modest overall response rates (10% with R1507 and 14.2% with figitumumab) and poor median progression-free survival (1.3 months with R1507 and 1.9 months with figitumumab). We are now left with the same question with which we started: What is the potential of IGF-1R targeting for cancer therapy?

The IGF-1R pathway transduces extracellular signals intracellularly to mediate cell proliferation, growth, and survival. IGF-1R is activated on engagement by the growth factor ligands IGF-1 and IGF-2, resulting in receptor autophosphorylation. IGF-1R activity is also regulated by six IGF binding proteins (IGFBP1-6), which serve to either promote or antagonize IGF-1R signaling by binding with IGF ligands in circulation.³ This leads to the activation of multiple signaling cascades, including the phosphatidylinositol 3-phosphate kinase/Akt/mammalian target of rapamycin (mTOR) pathway (Fig 1), which when aberrantly activated promotes the oncogenic phenotype. Several lines of evidence have suggested that IGF-1R signaling is critical to the biology of ESFT. Expression of IGF ligands and IGF-1R in these tumors has long suggested that the pathway is activated via an autocrine loop.^4,5 Importantly, malignant transformation induced by the pathognomonic ESFT EWS/FL-1 fusion gene (product of t(11;22)) is dependent on IGF-1R.⁶ The EWS/FL-1 fusion also promotes IGF-1R activation by repressing the expression of IGFBP-3, a binding protein that negatively regulates IGF-1R signaling by sequestering IGF-1 in the serum.⁷ Many studies have also shown that drugs targeting IGF-1R inhibitors can elicit growth arrest in ES cells in vitro and in xenograft models.⁸

Fig 1.

Insulin-like growth factor type I (IGF-1R) signaling axis. IGF-1R signaling is modulated by circulating IGF ligands (IGF 1-2) and IGF binding proteins (IGFBPs 1-6). IGF engagement of IGF-1R results in receptor autophosphorylation and activation of several downstream signaling cascades, including phosphatidylinositol 3-phosphate kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR) pathway. Activation of mTOR complex 1 (TORC1) promotes mRNA translation and activates negative feedback signals to suppress upstream pathway signaling. 4EBP-1, eIF-4E-binding protein-1; eIF-4E, eukaryotic initiation factor 4E; Grb10, growth factor receptor-bound protein 10; IRS-1, insulin receptor substrate 1; PDK-1, phosphoinositide-dependent protein kinase 1; PIP2, phosphatidylinositol 4,5-bisphosphate; PIP3, phosphatidylinositol 3,4,5-triphosphate; PTEN, phosphatase and tensin homolog deleted on chromosome 10; RheB, Ras homolog enriched in brain; S6 kinase, p70 S6 kinase; S6 RP, S6 ribosomal protein; TSC1, hamartin; TSC2, tuberin.

However, the results of these phase II studies^1,2 now seem to reject the preclinical hypothesis that expression of EWS-containing fusion genes alone predicts for tumor susceptibility to IGF-1R targeting, leaving us to speculate whether better predictors of benefit remain to be identified. Although both groups correlated high baseline total and/or free IGF-1 serum levels to superior overall survival, IGF-1 did not significantly correlate to overall response rate or progression-free survival, suggesting that it may be a prognostic marker for patients with ES and ESFT and not a predictor of benefit with IGF-1R targeting. In fact, a recently published report concludes that high circulating IGF-1 levels portends a lower risk of disease progression and death among patients with ES and localized primary tumors who are not treated with IGF-1R inhibitors.⁹

Both groups have indicated future plans for analyzing archival tumor tissues for candidate predictors of benefit, providing the opportunity to validate proposed modes of resistance to IGF-1R targeting. Studies of rhabdomyosarcoma models have shown that careful quantification of IGF-1R expression reveals a direct correlation between levels of receptor expression and the antiproliferative effect of IGF-1R targeting.¹⁰ Expression of other IGF pathway components, including IGF-2, insulin receptor substrate 1 (IRS-1), and IRS-2, has also been correlated to tumor cell susceptibility.¹¹ Notably, therapeutic antibodies have been designed to minimize cross reactivity with closely related IRs to diminish the clinical risk of hyperglycemia. Evidence that IR signaling can contribute to the oncogenic phenotype¹² and mediate resistance to IGF-1R inhibition¹³ continues to emerge, raising the possibility that by minimizing metabolic toxicity, we may be inadvertently compromising clinical efficacy. From this perspective, small-molecule IGF-1R inhibitors may yield results distinct from therapeutic antibodies, because some of these drugs can inhibit IRs to various degrees. Lastly, dominance of alternate receptor tyrosine kinases (such as macrophage-stimulating 1 receptor¹⁴ and platelet-derived growth factor receptor¹⁵) also represents an alternate mechanism of resistance.

Nonetheless, to understand why the high expectations for these studies^1,2 were not fulfilled first requires examining the basic question: Did the IGF-1R–targeting antibodies elicit the intended biologic effects on the tumor? Without better noninvasive techniques, only serial tumor tissue analyses to assess the adequacy of target inhibition, activation status of downstream molecules (eg, Akt and mTOR phosphorylation), and activation of alternate signaling pathways can start to distinguish between inherent shortcomings of the preclinical models versus the clinical limitations of how IGF-1R signaling can be manipulated in vivo. Certainly, the costly barriers to developing, performing, and coordinating such difficult analyses are indisputable; however, prioritization of such studies could be valuable for interpreting clinical data and generating subsequent hypotheses.

Clinical results suggest the need for new directions, and preclinical data suggesting novel targeted combinations involving IGF-1R certainly abound. IGF-1R has been proposed to mediate resistance to a variety of therapies, including BRAF inhibitors in melanoma¹⁶ and Akt inhibitors.¹⁷ A promising approach in sarcoma remains combinatorial inhibition of IGF-1R and mTOR.^18,19 The rationale arises from the observation that inhibition of mTOR complex 1 with the drug rapamycin results in IGF-1R activation through abrogation of an mTOR-mediated negative feedback signal, which suppresses receptor-pathway activation through S6 kinase 1 activation²⁰ and stabilization of the protein Grb10 (Fig 1).^21,22 In fact, in the figitumumab study,² 29 patients with suboptimal responses to the antibody alone were then treated with the addition of rapamycin. This strategy is also being tested in an ongoing phase II clinical trial that includes ES and other IGF-1R expressing bone and soft tissue sarcomas (clinicaltrials.gov identifier NCT01016015). The hope is that these new directions will translate into more effective drug combinations so that the unfulfilled promise of IGF-1R targeting will one day be considered a promising beginning.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.

AUTHOR CONTRIBUTIONS

Manuscript writing: All authors

Final approval of manuscript: All authors