Using a mass spec-based phosphoproteomic screen of kidney cancer samples, researchers at Oregon Health & Science University's Knight Cancer Institute have identified a Src kinase signaling pathway active in a subset of renal cell carcinomas.
The findings, which were published this week in Science Translational Medicine, suggest the pathway could represent a new target for kidney cancer treatment and that Src inhibitors like Bristol-Myers Squibb's Sprycel – known generically as dasatinib – could be repositioned to treat the disease. Additionally, the scientists identified a series of proteins linked to the Src signaling pathway that could prove useful in identifying RCC patients most likely to respond to such treatment.
According to OHSU researcher George Thomas, the lead author on the study, while Src has been implicated in diseases like lung and colorectal cancer, it hasn't typically been associated with kidney cancer.
"It really wasn't something that intuitively I would have expected to come up," he told ProteoMonitor. "Our lab's focus is very much on targeted therapies, and certainly in that realm I don't think anyone had really looked at it as a potential target."
As the researchers examined the results of their phosphoproteomic screen, however, Src emerged as one of the most prominent differentially regulated proteins.
Using iTRAQ labeling and MS/MS analysis on an Applied Biosystems QSTAR Elite, the scientists compared phosphorylation profiles of two sets of RCC cells — those with and without wild-type expression of the von Hippel-Lindau protein, a tumor suppressor tied to a number of cancers including kidney cancer.
Thomas expected the screen to identify a large number of differentially phosphorylated proteins, he said, but, in fact, it identified just 22 such phospho-sites, many of which were linked to Src.
Looking at the data, it was "very clear that Src was there, and there were several substrates of Src that were also highlighted," he said. "So that gave me an indication that this was something we should look at."
At the beginning of the screen, Thomas noted, the researchers had expected it would require significant bioinformatics work to determine what pathways the identified proteins fit into, but "without even having to do much bioinformatics work you could see that [Src signaling] might be something interesting."
The group's unexpected find is an example of the advantages of unbiased techniques like phosphoproteomic screens, said University of California, Los Angeles, professor Thomas Graeber, who was not involved in the research but whose lab also focuses on cancer signaling and protein phosphorylation.
The project "really used an unbiased approach to get this initial lead as to what was happening in this particular subtype of cancer," he told ProteoMonitor. "From our point of view, we like examples that aren't always hypothesis-driven science. Sometimes just taking an unbiased look and following up on those leads – certainly with the technologies that exist now – can be very lucrative."
Src signaling is an interesting potential target for RCC particularly given the lack of effective therapies for the disease. The cancer is largely chemo- and radio-resistant, Thomas said, noting that until the advent of targeted angiogenesis inhibitors, cytokine-based treatments like interferon or interleukin – which have high toxicity – were the only effective therapies.
Even the angiogenesis inhibitors are only partially effective, he added. Since they target "the tumor [blood] vessels rather than the tumor itself, [they don't] cure the cancer. Most folks who start on that treatment will get some benefit, but it's certainly not a cure, and most people will become resistant at some point."
"What this study is trying to put forward is that there are subsets that could benefit from a Src inhibitor," he said.
The researchers identified five proteins – VHL, Src, pFAK, HIF, and CA-IX – whose levels could help stratify patients by their likelihood of responding to a Src inhibitor. Now, Thomas said, they hope to test the markers in a clinical trial to see if they "actually pan out."
"We're just starting communications with pharma about doing sort of a phase II trial," he said, noting, though, that his team has yet to enter any discussions with drugmakers regarding such a study. Because several drug companies have Src inhibitors already in phase II or III trials, data already exists on toxicity and maximum-tolerated doses, which should streamline efforts to test the biomarkers clinically, he added.
BMS's Sprycel, currently indicated for certain leukemia patients who no longer benefit from prior treatment , is an obvious drug to start with, Thomas said, "because it's been approved [by the US Food and Drug Administration], it has some history, we know what the toxicity is, what the doses are. So that would probably be the logical place to go."
The study's findings suggest potential for a commercial companion diagnostic for Src inhibitors targeting RCC, Thomas said, but, he cautioned, such a test "wouldn't be as easy or as clear-cut" as companion diagnostics like those measuring HER2 expression levels in breast cancer or Bcr-Abl expression in chronic myelogenous leukemia.
"Here it's going to be several markers, and you'll have to explore the relationship between them," he said. "So it's not going to be a yes-no answer. I think once we have some clinical data behind it you could then develop some sort of cutoffs and say that folks who have, say, greater than 75 percent expression in their tumors of proteins A,B, and C are the ones who respond best. You could then start making something a little more definite."
The study demonstrates that there's "no easy place to draw the line between the groups," Graeber said. "You do need to come up with single measures that will guide you as to whether patient should be treated with a certain approach or not, but inevitably that single measure is going to have a grey area. That remains a problem throughout the field, and the data highlights that."
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.