Dr. Andrew Godwin, a researcher at Fox Chase Cancer Center in Philadelphia, Penn., presented data at the 2008 American Society of Clinical Oncology conference (ASCO) indicating that the Insulin-like Growth Factor 1 Receptor (IGF-1R) may have a role in GIST oncogenesis. Up to now, most research has focused on the mutant c- KIT or PDGFRα proteins that have been identified as the initiating pathogenic event in over 80 percent of GIST tumors. However there has always been five to ten percent of GIST that have no mutation and are called “Wild-type”. While some of these GISTs have responded to c-KIT inhibitors, it is clear that another mechanism has been involved in their growth. Godwin’s research focused on these wild-type tumors, which occur in adults but also mainly in female, pediatric GIST patients.
Driving Godwin’s team has been the awareness that wild-type GIST patients have historically had poorer prognosis in GIST trials measuring progression-free survival and overall survival. This observation raised the question of what alternative mechanisms were at work. Since 2003, several research papers have pointed to IGF-1R as a potential research target. It has also been observed that infrequently, GIST patients have experienced disruptions in their glucose metabolism, exhibited by low glucose symptoms, possibly due to insulin-like factors produced in the tumors. Recent publications discussing GIST metabolism, gene profiling and DNA copy analysis have preceded this latest research on IGF-1R in GIST.
IGF-1R is a tyrosine kinase similar to c-KIT but with important differences. Unlike c-KIT, IGF-1R is made up of two parts: an α part outside the cell and a β part inside the cell. Like c-KIT, two similar configurations must join together to become active (in this case one α+β joins with another α+β). There are two ligands, or proteins, that join the extracellular part of the paired receptors to activate the intra-cellular kinase portion of the pair. The ligands are IGF-1 and IGF-2.
As c-KIT is similar to PDGFRA, IGF- 1R is very similar to the insulin receptor (Believe it or not, the insulin receptor does not have an acronym; it is called the “insulin receptor”). A challenge for inhibitor designers is to design a drug that inhibits IGF-1R, but not the insulin receptor. Inhibiting either receptor can disrupt glucose metabolism, so IGF-1R inhibitor clinical trials may have precautions for metabolic side-effects.
Like c-KIT, IGF-1R is involved in cell growth and tumor formation and has been identified as a potential target in a number of cancers. Unlike c-KIT, there is presently no known mutation driving IGF-1R. It is important to note that key to the Godwin study was the availability of frozen tumor samples from collaborating researchers and from patients and families.
By comparing “normal” mutant GIST tumors with Wild-type, Dr. Godwin was able to establish significant increased IGF-1R expression in Wild-type GIST by multiple methods:
Using western blotting whereby tumor cell contents were filtered through a matrix and separated by molecular weight, Wild-type had higher expression of IGF- 1R than mutant GIST (10 fold)
Using immunohistochemical analysis (IHC), in which tumor samples are stained and interpreted by microscopic analysis, Godwin’s team found that wild-type GIST scored significantly higher IGF-1R levels among a team of analysts blinded to the mutation status. Blinding removed subjective bias in a test that can be highly subjective.
Further tumor specimen analysis using mRNA based assays indicated an even higher (80 fold) difference in IGF-1R expression with wild-type GIST— much higher than mutant.
Using Fluorescent In Situ Hybridization (FISH), whereby snips of DNA designed to match a target sequence are given a fluorescent tag and allowed to mix with tumor sample DNA, Godwin’s team found that Wild-type cells had multiple copies of the region that produces IGF-1R, more so than mutant tumor samples. This was further confirmed using Quantitative Polymerase Chain Reaction (qPCR) analysis, whereby regions of test DNA are replicated and compared to determine relative copy number differences. Interestingly, while IGF-1R copy numbers were higher in Wild-type, Godwin also found a high number of copies in mutant GIST indicating a possible area of investigation. Godwin’s team then tested an IGF-1R inhibitor against mutant GIST cell lines. They used a compound from Novartis called NVP-AEW541 (NVP), which is highly specific for IGF-1R. They found that:
NVP can shut down common GIST signaling pathways downstream from IGF-1R.
Inhibiting IGF-1R reduces cell growth and causes apoptosis in exon 11 and exon 13 mutant GIST cell lines similar to high concentrations of imatinib.
There was no synergy observed in combining NVP and imatinib (Synergy has been reported in another more recent paper).
NVP probably also inhibits other offtarget GIST pathways. In his presentation, Godwin pointed out that IGF-1R is involved in a complex of pathways (“a lot more complicated than you actually can imagine”), with major interactions at multiple levels varying between proteins and even cell types.
IGF-1R is activated in both mutant and wild-type GIST.
IGF-1R is over expressed in a subset of adult and pediatric wild-type GIST (not in 100%, but in a significant fraction).
Gene amplification or ligand overexpression may be the contributing oncogenic effect.
Targeting IGF-1R can induce cytotoxicity in mutant GIST cell lines.
Godwin’s group has developed wildtype cell lines and they are now testing IGF-1R inhibitors currently in clinical trials against these lines.
Plans are ongoing to develop clinical trials of IGF-1R inhibitors in GIST.
Godwin’s presentation referenced two recent papers on the subject of IGF-1R in GIST.
Dr. Chi Tarn, a member of the Godwin team was lead author on a paper providing details on much of the material in the oral presentation at ASCO (Tarn et al. 8387-92).
A paper by an Italian research team identified the IGF-1 and IGF-2 ligand expression as associated with GIST relapse in a series of 94 tumors. Sample expression levels were rated as “Absent”, “Moderate” and “Strong”. They found that 80 percent of tumors express one or the other ligand. There was significant correlation with high mitotic index, larger size, higher risk, metastatic disease and relapse. There was also significance between the increasing levels worsening disease-free survival. GIST genotype (exon) did not matter. This study also used blinded assays due to the subjectivity of the test results. One caution is that the level of IGF-1 and IGF-2 staining is often low and hard to measure. Using the same IHC methodology, Godwin’s team found no difference in the levels of these ligands between mutant and wild-type GIST tumor and noted the difficulty in measurement as a possible reason. These results, and Godwin’s in vitro analysis, both point to the possibility of a role for IGF-1R inhibition in mutant GIST. They also may open the door to screening patients for markers (IGF-1 and IGF2), potentially indicating treatment options (Braconi et al.).
A recent paper by researchers at Memorial Sloan-Kettering in New York, has also identified IGF-1R as overexpressed in pediatric GIST (Agaram et al. 3204-15).
For GIST patients, IGF-1R inhibition will continue to be an area of interest, especially for those without c-KIT or PDGFRa mutations. Also, while other signal pathways have not yet produced another Gleevec, it is encouraging to see new possibilities emerge and move forward from the lab. While pending success in the clinic, these results also point to the growing importance of genotyping.
Agaram, N. P., et al. “Molecular characterization of pediatric gastrointestinal stromal tumors.” Clin Cancer Res 14.10 (2008): 3204-15.
Braconi, C., et al. “Insulin-like growth factor (IGF) 1 and 2 help to predict disease outcome in GIST patients.” Ann.Oncol (2008).
Tarn, C., et al. “Insulin-like growth factor 1 receptor is a potential therapeutic target for gastrointestinal stromal tumors.” Proc Natl.Acad.Sci.U.S.A 105.24 (2008): 8387- 92.