IGF-1R Resistance
Based on the experience in soft tissue sarcomas as well as in pancreatic cancer, it is clear that acquired resistance to IGF-1R inhibitors is common and eventually leads to disease progression. There are several proposed mechanisms of resistance to the IGF-1R targeted agents, which can be summarized into two mechanistic groups.
Based on the experience in soft tissue sarcomas as well as in pancreatic cancer, it is clear that acquired resistance to IGF-1R inhibitors is common and eventually leads to disease progression. There are several proposed mechanisms of resistance to the IGF-1R targeted agents, which can be summarized into two mechanistic groups.
The first group incorporates mechanisms that encourage the cells to adapt to alternative proliferative pathways by diminishing the activity of the IGF-1 pathway in cancer proliferation. Examples include the overexpression of IGFBPs, cross-talk between the IGF-1R and ErbB pathways, and HSP90.
• The bioavailability of circulating IGF-1 and IGF-2 is regulated by six IGFBPs as described earlier. Therefore, IGFBP overexpression has been implicated as a mechanism of resistance to the inhibition of IGF signaling. This has been demonstrated in vitro in sarcoma cell lines treated with BMS-536924 in which the overexpression of IGFBP-3 and IGFBP-6 correlated with resistance to IGF signaling inhibition.[39] Progressively increasing levels of IGFBPs have been noted in patients receiving IGF-1R inhibitors.[40] Whether this represents a resistance mechanism or simply a pharmacodynamic effect remains to be demonstrated.
• The IGF receptor family exhibits considerable crosstalk with the ErbB family of receptors (EGFR, HER2/neu, HER3, and HER4) and contributes to resistance to agents targeting these pathways. The co-inhibition of both pathways shows synergistic enhancement of apoptosis in several cancer cell lines. Data suggest that trastuzumab(Drug information on trastuzumab) (Herceptin) resistance correlates with IGF-1R overexpression and thus that the co-inhibition of the IGF-1 family and the HER family may overcome this resistance.[41] The converse may also be true: in response to IGF-1R inhibition with MK-0646 and BMS-754807, HER2 expression was upregulated.[41,42] Similar data have been reported regarding HER3 activation after treatment with the IGF-1R–directed monoclonal antibody AVE1642. In this case, siRNA downregulation of HER3 resulted in increased sensitivity to AVE1642.[43]
• Transcription factors directly in control of IGF1R gene expression include tumor suppressor genes p53 and BRCA1. Wild type p53 suppressed IGF1R promoter activity in vitro by 90% while also suppressing endogenous IGF1R mRNA levels. Tumor-derived mutant forms of p53 enhanced IGF1R expression.[44] These preclinical experiments have demonstrated that lack of inhibition or even stimulation of the IGF1R promoter by mutant p53 may accelerate tumor growth. The BRCA1 gene product has also been shown to inhibit IGF1R gene expression.[45] Mutant BRCA1 may lead to dysregulated gene expression, with higher expression of IGF1R in patients with mutant BRCA1 breast cancer.[46]
• IGF-1R, Akt and Raf are dependant on the chaperone protein HSP90, which preserves their stability. HSP90 expression is increased in cancer cells, and its inhibition may be a strategy to block mitogenesis. In Ewing sarcoma cells, HSP90 expression is increased; its inhibition by siRNAs reduced cancer cell survival in IGF-1R–refractory cell lines.[47]
• Many tumors have chronically elevated activity of PI 3-kinase–dependent signaling pathways, caused largely by oncogenic mutation of PI 3-kinase itself or loss of the opposing tumor suppressor, PTEN. In cell lines (glioblastoma, breast, and prostate), PTEN loss caused desensitization of IGF/insulin signaling pathways, suggesting that the efficacy of inhibitors of IGF-1R could be limited in patients with PTEN loss or oncogenic mutation of PI 3-kinase.[48]
There is currently a critical need for the identification of biomarkers in clinical specimens that could potentially be used to stratify patients for IGF-1R–directed therapies. Fortunately, there exists a considerable body of preclinical data as well as clinical experience in other cancer types that will be invaluable for developing IGF-1R–targeted therapies in gastrointestinal cancers.
Preclinical Studies of Potential Biomarkers of the IGF-1 Pathway
Zha et al investigated molecular predictors of the IGF-1R monoclonal antibody h10H5. They noted that in breast cancer and CRC, expression of IGF-1R is necessary for anti-tumor activity but by itself is not sufficient.[54] In their study, at least 1300 receptors per cell were needed for in vitro anti-tumor efficacy of h10H5. It is possible that oncogenic transformation requires expression of IGF-1R even at low levels, which may not be detectable by IHC. Therefore, IGF-1R expression by IHC may be a better negative predictor than positive predictor of response to anti–IGF-1R therapy, but this may not, by itself, have sufficient predictive value. Other preclinical studies (summarized in Table 3) suggest that activation of the IGF-1R pathway within cancer cells is required for anti-tumor efficacy of an agent targeting this pathway. For instance, IGF-1R needs IRS-1 in order to send mitogenic signals—and without IRS-1, IGF-1R sends a differentiation signal to the cells.[55] Thus, IRS-1 may be a requirement for IGF-1R activity and could potentially serve as a biomarker for sensitivity to agents that target IGF-1R. The Erk and Akt pathways are not activated by IGF-1R alone because of extensive cross-talk and may not accurately predict responsiveness to IGF-1R inhibition.
High IGF1R gene copy number was correlated with squamous cell histology and worse survival in lung cancer [56] Pitts et al, on the other hand, noted no IGF1R gene amplification in OSI-906–sensitive CRC cell lines.[57] But this OSI-906–sensitive line had an unbalanced gain of IGF1R based on ploidy. The cellular phenotype of IGF-1R–responsive cancer cells has not yet been defined. Squamous cell histology was considered as a possible phenotype that was responsive to IGF-1R inhibition. This was not proven to be the case in the figitumumab phase III study discussed above. Gualberto et al examined potential molecular and phenotypic determinants of figitumumab sensitivity in NSCLC. In their analysis, the mesenchymal phenotype was least likely to respond to IGF-1R inhibition.[58] However, this phenotype is also refractory to EGFR-targeted agents such as erlotinib,[59] and it remains to be determined whether mesenchymal differentiation is a prognostic marker rather than a predictive marker.
Therapeutic Agents
Mentioned in This Article
AMG 102
AMG 479
AMG 655
AVE1642
BIIB022
BMS-536924
BMS-754807
Carboplatin(Drug information on carboplatin)
Cetuximab (Erbitux)
Cixitumumab (IMC-A12)
Dalotuzumab (MK-0646)
Erlotinib (Tarceva)
Figitumumab
FOLFIRI (folinic acid, fluorouracil(Drug information on fluorouracil), and irinotecan)
Gemcitabine(Drug information on gemcitabine) (Gemzar)
h10H5
Irinotecan(Drug information on irinotecan) (Camptosar)
NVP-AEW541
OSI-906
Paclitaxel(Drug information on paclitaxel)
Panitumumab (Vectibix)
R1507
Sch717454
Sorafenib(Drug information on sorafenib) (Nexavar)
Trastuzumab (Herceptin
Brand names are listed in parentheses only if a drug is not available generically and is marketed as no more than two trademarked or registered products. More familiar alternative generic designations may also be included parenthetically.
Pharmacodynamic studies have investigated the modulation of the target pathways by IGF-1R inhibitors in clinical studies. Atzori et al investigated the modulation of pAkt, pMAPK, and pS6 kinase in paired biopsies of patients receiving the antibody MK-0646. They demonstrated significant inhibition of IGF-1R and IGF-1R signaling post-therapy in patients who received this therapy. Furthermore, elevation of IGF-1 levels were also noted secondary to IGF-1R inhibition.[60] However, paired tumor biopsies are not feasible in the vast majority of patients, and potential surrogate biomarkers that can be investigated in the blood may be invaluable in this setting. De Bono et al investigated the expression of IGF-1R on circulating tumor cells (CTC) in phase I trials of figitumumab as a single agent or with chemotherapy. Treatment with figitumumab alone decreased both total CTC count and IGF-1R expression. CTCs and IGF-1R–positive CTCs were most frequently detected in advanced prostate cancer patients. This non-invasive technique allows the evaluation of the therapeutic effect of targeted inhibitors on the IGF-1R–positive CTCs.[61] An obvious limitation is that CTCs are not easily detectable in all solid tumors, but this method may have implications in breast cancer and CRC.
Thus, the current data suggest that IGF-1R pathway activation within the tumor may predict responsiveness to specific inhibitors. In all probability, a reliable predictive panel for IGF-1R inhibitors will include multiple markers beyond just IGF-1R protein expression.
Summary
IGF-1R targeted therapy has been shown to be safe in many cancer types. Compelling preclinical data suggest that IGF-1 pathway inhibition may lead to clinical benefit in gastrointestinal cancers. Epidemiological studies have identified the IGF-1 pathway as an important risk factor for the development of several cancers. Experience with IGF-1R inhibitors in lung cancer and sarcoma has provided invaluable information that can be applied in the treatment of gastrointestinal malignancies. Preliminary data in CRC indicate a lack of clinically meaningful synergism between the IGF-1R and EGFR inhibitors in K-ras wt tumors. The activation of the IGF-1 pathway may also be predictive of response to the IGF-1R inhibitors. However, acquired resistance is common and not accounted for in the current clinical trials. The identification and validation of biomarkers of IGF-1 pathway activation in clinical samples must be emphasized so as to increase our chances of success with IGF-1R–targeted therapies in gastrointestinal cancers.
Financial Disclosure: Dr. Javle is a principal investigator on a clinical trial of MK-0646 (Merck).
A 47-year-old male with complaints of chronic abdominal pain and episodes of flushing presents for evaluation. What is your diagnosis?
