Research at the Oregon Health and Science University in Portland may lead to a diagnostic that could help physicians unsatisfied with Gleevec’s performance against gastrointestinal stromal tumors to switch to Sutent.
“We wondered if the same [genes of response for Gleevec] would hold true for Sutent, and what we found is that it’s exactly opposite,” according to Michael Heinrich, who led the study. Heinrich previously studied mutations linked to Gleevec response in GIST.
Gleevec, made by Novartis, is a first-line treatment for GIST while Sutent, developed by Pfizer, is second line. There are no other treatments for GIST, which is an orphan disease. Pfizer funded Heinrich’s research.
Patients who have failed Gleevec therapy whose tumors harbored cancer-causing mutations in the C-Kit gene’s exon 9 region, as well as tumors with no known activating mutation, respond much better as a group to Sutent than do patients with a C-kit exon 11 mutation, he said.
If the current research is validated, a diagnostic might help clinicians decide whether and when to switch to Sutent patients showing some resistance to Gleevec.
Heinrich said that if a patient has a C-kit exon 9 mutation or a tumor without mutations at all he “would be influenced to switch earlier to Sutent because I think that the possibility of a really good benefit of the drug would be much higher.”
However, he noted, “If the patient has a C-kit exon 11 [mutated] tumor, I would be less hopeful of a very good response to Sutent, and I might be inclined to continue Gleevec, escalate the dose of Gleevec and hold onto Gleevec longer before switching over, because I think [Sutent] would not be as effective as a salvage therapy.”
Gastrointestinal stromal tumor is designated an orphan disease by the US Food and Drug Administration, with about 5,000 cases diagnosed each year, according to the American Cancer Society.
But even though the indication is rather small, “in GIST, it does make sense to look at the mutations and see which patient might be more likely to respond to one or the other” drug, Charles Baum, vice president and oncology development head at Pfizer, told Pharmacogenomics Reporter this week.
Asked whether Pfizer has explored genetic diagnostic options for guiding Sutent treatment, Baum said “there have been some companies that came to us, and we’re willing to cooperate with efforts like that,” but Pfizer is not primarily engaged in diagnostics, and would not pursue them solo.
Heinrich’s study followed patients who had failed Gleevec, and research supporting the use of a diagnostic to guide GIST treatment would have to begin with patients who had not taken either drug.
Pfizer has “started discussion with some investigators to do that kind of a study, but it’s not settled yet.” However, Baum said such a study might begin within about a year and could have published data in about two to three years.
Of course, the final outline for the best course of GIST treatment may not resemble today’s positioning of Gleevec as first-line therapy and Sutent as second line. “Does it make sense to use one or the other? There are people who suggested putting both of the drugs together as one of the strategies to avoid the development of mutations that cause resistance,” said Baum. These strategies will be explored over “the next few years,” he added.
Currently, only the OHSU Molecular Diagnostics Center, ARUP Laboratories, and MD Anderson Cancer Center offer C-kit mutation testing for cancer, said Heinrich. At OHSU, investigators use “an HPLC and sequencing combined approach” with Transgenomic’s Wave platform, while ARUP uses melting-curve analysis, he said.
Transgenomic, Novartis, and ARUP could not be reached for comment in time for this publication.
Two main factors would complicate the job of creating a simple mutation-analysis kit: mutation variety and tumor sample purity. Several of the C-kit mutations involve deletions, duplications, internal repeats, and other complications, while the number of mutant cells in a tumor sample can be as low as 10 percent, according to Heinrich.
About 90 percent of patients with GIST have tumors with a genetic mutation believed to cause the cancer, and of these about 85 percent are mutations of the C-kit gene, said Heinrich.
About five percent of all GI stromal tumors carry a mutation in the platelet-derived growth factor-alpha, or PDGF-alpha, gene, which is very similar to C-kit. “Ten percent of the tumors are unexplained — we call them ‘wild-type’ tumors because they don’t have any mutations, at least of the kinases, but we don’t know what gives rise to the tumors, or how closely they are related to the mutant tumors,” he said.
Within C-kit, mutations in exon 11 account for about two-thirds of genetic changes related to GIST, while mutations in exon 9 and elsewhere within C-kit account for the rest, Heinrich said.
“We wondered if the same [genes of response for Gleevec] would hold true for Sutent, and what we found is that it’s exactly opposite.”
Clinical symptoms are the most commonly used piece of evidence that clinicians use to decide whether to continue treatment with Gleevec or whether to increase the dose, although using positron emission tomography to measure glucose use seems to more quickly show tumor resistance to the drug, said Baum.
Genetic testing is used even less frequently than imaging to help inform GIST treatment decisions with the two drugs, said Baum. “Although people know about the [C-kit exon 9 and 11 mutation] information now, it’s a little bit more technically challenging for all of the medical centers to do that type of testing,” he said, referring to both secondary resistance mutations and primary mutations. “I don’t think you could call it a typical diagnostic test that’s done on all patients, at this point.”
However, there may be a growing demand for GIST-related genetic testing. “I think that [this kind of assay is] getting more interest now that [Heinrich has] shown that this is something you can do, and there’s more interest on the part of patients, et cetera, in having the testing done,” said Baum.
The research could have implications elsewhere. “Recently there have been publications [in which] there are C-kit mutations in melanoma — in a minority of them — and that those patients may respond to [Gleevec], so there are a number of trials looking at mutant melanoma to see whether [Gleevec] would be effective,” said Heinrich.
Heinrich speculated that some melanomas may also be sensitive to Sutent, but added that he was not aware of any clinical trials investigating the subject.
The American Cancer Society estimated that about 60,000 US adults were diagnosed with melanoma in 2005.