Treatment with sunitinib (Sutent), an oral, multitargeted tyrosine kinase inhibitor, improves clinical outcomes in patients with gastrointestinal stromal tumor (GIST) that has become resistant to imatinib (Gleevec), George D. Demetri, MD, director of the Center for Bone Oncology, Dana-Farber Cancer Institute, reported at the 2006 Gastrointestinal Cancers Symposium (abstract 8).
SAN FRANCISCO-Treatment with sunitinib (Sutent), an oral, multitargeted tyrosine kinase inhibitor, improves clinical outcomes in patients with gastrointestinal stromal tumor (GIST) that has become resistant to imatinib (Gleevec), George D. Demetri, MD, director of the Center for Bone Oncology, Dana-Farber Cancer Institute, reported at the 2006 Gastrointestinal Cancers Symposium (abstract 8). Relative to placebo, sunitinib was associated with a more than fourfold increase in the time to progression and a one-half reduction in the risk of death. During the GI meeting, the agent received FDA approval for use in GIST and renal cell cancer (see page 19).
Patients enrolled in the phase III trial were randomly assigned in a 2:1 ratio to receive sunitinib (207 patients) or placebo (105). Sunitinib was given in 6-week cycles, consisting of 4 weeks on therapy (with a starting dose of 50 mg once daily) and 2 weeks off. Patients continued their therapy as long as they derived clinical benefit. Patients in the placebo group who experienced a progression were allowed to cross over and receive sunitinib.
Some 79% of the patients had experienced disease progression after receiving imatinib for more than 6 months, and 17% had experienced progression on the drug sooner; the remaining 4% had not been able to tolerate imatinib therapy.
A planned interim analysis showed that patients in the sunitinib group had a median time to progression (the primary endpoint) of 27.3 weeks vs 6.4 weeks for placebo; this difference corresponded to a significant two-thirds reduction in risk of progression (hazard ratio, 0.33). These data prompted unblinding of treatment and switching of patients receiving placebo to sunitinib; the median duration of follow-up was about 7 months.
Patients in the sunitinib group also had a higher estimated 6-month survival (79% vs 57% for placebo); this difference corresponded to a significant one-half reduction in risk of death (hazard ratio, 0.49). Median survival has not been reached in either group.
The beneficial effect of sunitinib on time to progression was consistent across subgroups, including age, sex, and time between initial diagnosis and enrollment in the trial. In addition, the benefit was similar for patients whose prior imatinib therapy had lasted 6 months or less and those whose therapy had lasted longer, and across subgroups of patients who had received differing doses of imatinib.
By RECIST criteria, 24% of patients given sunitinib had either a partial response or stable disease lasting at least 22 weeks, compared with 2% for placebo.
The nine patients who were unable to tolerate imatinib generally had an excellent response (four partial responses and four with stable disease). These favorable findings suggest that sunitinib also works upfront in GIST, Dr. Demetri said.
"We have been very pleased with the tolerability profile [of sunitinib]," he said. Overall, about half of patients in each arm had adverse events, and the incidence of various types of events was generally similar in the two treatment groups. Among small differences between groups, the sunitinib group tended to have a somewhat higher incidence of diarrhea of any grade (40% vs 27%) and of grade 3 or higher (4% vs 0%), he said.
Positron emission tomography (PET) scans in patients with imatinib-resistant GIST showed a dramatic decrease in functional activity within tumors as early as 1 week after the initiation of sunitinib therapy, with complete quiescence evident in some patients, Dr. Demetri said. However, he noted, the tumors do not shrink substantially, which may render conventional approaches to assessing the effectiveness of cancer therapies less useful. "We are getting into a new era where our criteria for success may not necessarily be shrinkage of tumor but controlling the disease so that people live longer," he explained.