Portland, Ore.

Mutational analysis of gastrointestinal tumors gives clinicians new tool to predict outcomes

An international team of researchers has concluded that lab testing can reveal ahead of time just how well patients with certain gastrointestinal cancers will respond to therapy with the drug Gleevec, according to study results presented at the annual meeting of the American Society of Clinical Oncology.

Patients with a favorable lab result had an 84 percent chance of a partial remission in response to Gleevec. Only 10 percent of patients with an unfavorable lab result had a partial remission.

"These results demonstrate that the most important predictor of tumor shrinkage during Gleevec therapy is not age or tumor size but rather the specific type of mutation causing the tumor," said Michael Heinrich, M.D., member of the Oregon Health & Science University (OHSU) Cancer Institute, associate professor of medicine at the Portland VA Medical Center, and co-principal investigator of the study.

Gastrointestinal stromal tumors (GISTs), which occur in the stomach and intestines, are diagnosed in up to 10,000 Americans annually. Surgery is the treatment of choice for localized tumors, but in many patients the tumor recurs and spreads elsewhere, particularly to the liver. Up until two years ago, at this metastatic stage, the disease was invariably fatal because chemotherapy, radiation and surgery are ineffective in advanced cases. More recently, however, physicians have shown that metastatic GISTs are responsive to Gleevec.

Gleevec, technically known as a tyrosine kinase inhibitor, was first introduced for the treatment of chronic myelogenous leukemia (CML), the growth of which is driven by a mutant tyrosine kinase enzyme called BCR-ABL. Gleevec is a potent inhibitor of the BCR-ABL enzyme, effectively shutting down the growth of leukemia cells with only relatively minor side effects.

Gleevec also inhibits a related tyrosine kinase called KIT. The presence of mutant KIT protein in GISTs led researchers to suspect that Gleevec would be an effective treatment for this cancer.

"The majority of GI stromal tumors have a mutant form of KIT that acts like a gas pedal stuck to the floor, providing a constant stimulus for GIST cells to grow," Heinrich said.

A fast-tracked clinical trial proved this hypothesis, and the drug was approved by the FDA in 2002 for patients with advanced GIST. This clinical trial was performed at the OHSU Cancer Institute, Dana-Farber Cancer Institute, Fox Chase Cancer Center and the University Hospital of Helsinki (Finland).

"Treatment with Gleevec inhibited KIT activity much like turning off this engine driving tumor cell growth," Heinrich said.

While the majority of the GIST patients treated in the Gleevec trial benefited from the drug, a subset did not respond well, and the latest research was designed to reveal why that was so. The researchers analyzed DNA samples of the GIST tumors from 127 patients enrolled in the trial to determine whether a KIT tyrosine kinase mutation was present and whether the specific type of mutation had an impact on drug response.

The results revealed three distinct subsets of GIST. Those with "exon 11" mutations of KIT responded well to Gleevec; 84 percent of patients with such tumors had a partial remission (greater than 50 percent shrinkage of their tumors), and half of the members of this group are still benefiting from the drug after 22 months. GISTs that lacked a KIT mutation, however, did not respond well to Gleevec. Only 10 percent of patients in this group had a partial remission, and all had failed therapy in less than a year. GIST with an "exon 9" mutation of KIT responded in an intermediate fashion in the trial, with half of the patients failing treatment after 187 days.

"KIT mutation status not only predicted the likelihood of Gleevec response, it was also the most important predictor of duration of response and overall patient survival in the trial," Heinrich said.

Further studies by Heinrich and his colleagues also revealed a new wrinkle in the GIST story. Among the tumors that lacked a KIT mutation, some had a mutation in a different (but closely related) tyrosine kinase called PDGFRA. While mutations in this kinase were found in only 4.7 percent of GIST cases in the Gleevec clinical trial, the drug was effective against some of these PDGFRA mutations.

"We are in the process of studying the PDGFRA mutations, but so far preliminary data suggest that there are parallels with KIT - that is, if a PDGFRA mutation is present, then Gleevec response can be predicted by the exact type of mutation," Heinrich said.

The study results suggest that clinical testing for specific mutation type may help clinicians treat patients with GIST. KIT mutational testing is now available at OHSU and testing for the newly identified PDGFRA mutations will be available soon.

"Mutational testing can be helpful in confirming the diagnosis of GIST and in defining the prognosis for patients who need Gleevec therapy," said Christopher Corless, M.D., associate professor of pathology in the OHSU School of Medicine, and co-investigator in the GIST research.

"As Gleevec is used more and more in combination with surgery, we believe that testing for mutations in GIST will be important in deciding whether Gleevec therapy should be used before and/or after surgery for GIST," Heinrich said.