ASCO-STI-571 (imatinib mesylate, Gleevec), which received swift approval by the Food and Drug Administration for the treatment of chronic myelogenous leukemia (CML), is showing equally striking activity in patients with advanced gastrointestinal stromal tumors (GISTs), scientists reported at the plenary session of the 37th Annual Meeting of the American Society of Clinical Oncology.
ASCOSTI-571 (imatinib mesylate, Gleevec), which received swift approval by the Food and Drug Administration for the treatment of chronic myelogenous leukemia (CML), is showing equally striking activity in patients with advanced gastrointestinal stromal tumors (GISTs), scientists reported at the plenary session of the 37th Annual Meeting of the American Society of Clinical Oncology.
GIST is a newly recognized variant of soft tissue sarcoma, with approximately 5,000 cases diagnosed annually in the United States. The dominant cellular mutation in GIST is the c-kit oncogene. The protein product of c-kit (KIT) is a tyrosine kinase similar in structure to Bcr-Abl, the protein targeted by STI-571 in CML. STI-571 has been shown to inhibit KIT, providing the rationale for studies of the agent in GIST.
Charles D. Blanke, MD, director of gastrointestinal oncology, Oregon Health & Science University, Portland, reviewed the interim results of a phase II multicenter trial in which patients with unresectable or metastatic GIST were randomized to treatment with STI-571 at 400 mg or 600 mg daily for up to 24 months.
The trial enrolled 148 patients, with 145 currently evaluable for toxicity and 86 evaluable for efficacy with 3-month follow-up. The majority of patients had been treated previously, and fewer than 1% had responded.
Results showed that 50 patients (59%) had a partial response on STI-571: 50% on the 400-mg dose and 68% on the 600- mg dose. Stable disease was seen in 26%, Dr. Blanke said, for an overall tumor control rate of 85%. With median follow-up of 4.5 months, "no patient has progressed from a confirmed response," he said.
He noted that 86% of patients tested for c-kit expression had a mutation. "Patients with mutations had a higher chance of responding than those with wild-type c-kit," he said. For example, those with the most common mutation (at exon 11) had a partial response rate of 78% vs 29% in patients with no detectable mutation. Among patients with wild-type c-kit, almost 50% had disease progression.
"This is the first effective nonsurgical therapy for GIST, a cancer that historically has been resistant to chemotherapy and radiation therapy," Dr. Blanke said. Although follow-up is short, he said "these data are exciting because, historically, most GISTs have progressed within a median of 2 to 6 months."
Dr. Blanke noted that there was a trend for a higher response rate with the 600-mg dose. "The dose-intensity question should be definitively answered by the current phase III Intergroup GIST trial," he said.
Dr. Blanke said that the agent was, in general, well tolerated. One quarter of patients required dose reductions, delays, or drug discontinuation due to toxicity. There was no difference in toxicity rates between the 400 mg and 600 mg doses.
Severe to life-threatening drug-related toxicity occurred overall in 21% of patients. The most common grade 3-4 adverse events were GI bleeding (5%), liver toxicity (3%), edema (3%), neutropenia (3%), and infection (2%). The bleeding appeared to be tumor-related, Dr. Blanke said. Abdominal pain, rash, and diarrhea were seen rarely. No patient died from treatment-related causes.
A smaller, phase I dose-escalation trial conducted by the European Organization for Research and Treatment of Cancer (EORTC) also found that STI-571 resulted in clinical and radiological improvement in the majority of patients with GIST.
The collaborative study enrolled 40 patients with soft tissue sarcomas expressing KIT. Among 36 patients with GIST, only four (11%) progressed in the first 8 weeks of treatment. The 11% progression rate "surely represents an excellent response and exciting breakthrough," said Allan T. van Oosterom, MD, PhD, of UZ Gasthuisberg, Leuven, Belgium. Dr. van Oosterom is president of EORTC.
Results showed that 13 GIST patients (36%) had a partial response and 12 (33%) had an unconfirmed partial response (a 20% to 29% tumor reduction in the longest diameter). "Most of these patients have only been on treatment for 4 to 5 months, and a further increase in the confirmed partial response rate is to be expected," Dr. van Oosterom said. Disease was stable in seven GIST patients (19%) and one non-GIST patient. None of the 25 responders have relapsed to date.
The study results indicate that 400 mg twice a day is an acceptable dosage, "but it is not yet known whether it is superior to 400 mg daily," he said.
In his discussion, Charles Sawyers, MD, of the UCLA Jonsson Cancer Center, said that "it is hard to argue with the data we have just seen." However, he cautioned clinicians that STI-571 can have significant side effects. "All patients receiving STI-571 should be monitored closely for edema, bleeding, and cytopenias, especially in the early phase of treatment," he said.
Dr. Sawyers emphasized that the studies of STI-571 in CML and GIST have a short follow-up, and therefore patients may relapse. "We know that they can relapse in late-stage CML," he said. He pointed out that there are a number of ongoing combination trials of STI-571 with various agents in CML. "It is important that we continue to enroll patients in these trials despite the fact that the drug is FDA approved," he said.
Dr. Sawyers said that CML and GIST are ideal candidates for STI-571 because of the prominence of a single gene mutation as the cause of these cancers, and that it may be difficult to translate these successes in "single-hit diseases" to other, more heterogeneous cancers.
He noted, however, that a single oncogene can play a critical role even in a genetically complex cancer. Existing data, he said, show the primary importance of Bcr-Abl in CML blast crisis, despite the presence of numerous other oncogenic changes. "This is good news for the concept of applying targeted therapy to heterogeneous cancers," he said. "However, successful application of this concept to heterogeneous cancers will require measuring the drug targets in individual patients."