Chandrajit P. Raut, MD
Patients with intermediate- or high-risk gastrointestinal stromal tumors (GIST) were less likely to experience recurrence after 5 years of treatment with imatinib (Gleevec), as determined by results from the phase II PERSIST-5 trial.
“Chronic imatinib was effective in preventing recurrences during treatment for patients with sensitive mutations,” said Chandrajit P. Raut, MD, director, surgical oncology, sarcoma, Dana-Farber Cancer Institute. He presented the findings at the 2017 ASCO Annual Meeting. “Five-year recurrence-free survival and overall survival rates were 90% and 95%, respectively.”
Raut said investigators measured recurrence-free survival (RFS) from treatment initiation to first incidence of recurrence or death from any cause, and found 9 RFS events. Five-year RFS was 90% (95% CI, 80-95). Of those events, 7 were considered “true” recurrences, and 6 occurred after discontinuation. Only 1 patient recurred while on treatment.
Raut said 6 of the 7 had high-risk disease. Time from discontinuation to relapse ranged from 7.4 months to 23.1 months.
Five-year overall survival (OS) was 95% (95% CI, 86-99), and estimated 8-year OS was 81% (62-91). There were 3 deaths in the study, including the patient who recurred while on treatment. Raut said that patient had a high-risk, gastric GIST with a PDGFRA
“The only patient who recurred on-drug had an insensitive mutation,” he said.
PERSIST-5 is a single-arm, open-label multicenter study. Patients (N = 91) were assigned to 400 mg of adjuvant imatinib at 21 medical centers in the United States. Patients underwent imaging every 4 months for the first 3 years of the study, then every 6 months for the final 2 years. Treatment continued for 5 years or until progression, relapse, or intolerance.
Adults with KIT
-positive primary GIST who had undergone macroscopically complete surgical resection within 12 weeks of the start of imatinib treatment were eligible. Patients also needed to have a “significant” risk for recurrence, defined as primary GIST ≥2 cm at any site with a mitotic count or a non-gastric primary GIST ≥5 cm, and an ECOG score of 0-1.
Eighty-five patients were screened for primary mutation. Of those, 73 had KIT
mutations, the most common being KIT
exon 11 (n = 57).
Median treatment duration was 55.7 months (range, 0.5-75.1) and median posttreatment follow-up was 19.6 months (range, 4.6-32.1). Duration of treatment for patients who experienced recurrence ranged from 0.5 months to 61.1 months.
Forty-six patients completed 5 years of treatment, and 45 discontinued early (49%). Twenty-one percent of patients discontinued by choice and 16% discontinued because of adverse events (AEs). Sixty-eight patients, including some who discontinued treatment, completed follow-up.
“For most of the reasons patients stopped—adverse events, progression, protocol deviation, lost to follow-up—most of those happened during the first 2 years of the study,” Raut said. “The most common reason patients stopped was that they just elected to come out of the study for various other reasons. That rate continued fairly steadily throughout the study. It wasn’t just confined to the first 2 years.”
He added that, other than patient choice, the reasons for discontinuation in PERSIST-5 were in-line with results from previous adjuvant imatinib trials.
“The take-home here is that, for patient choice reasons, the longer treatment was problematic and lead to discontinuation,” Raut said.
Thirty-six patients (40%) required dose reductions, mostly for AEs, and every patient experienced at least 1 AE. The most common side effects of any grade included nausea (71%) diarrhea (63%), fatigue (50%), muscle spasms (41%), vomiting (39%), and periorbital edema (34%).
However, most AEs were minor. The most common grade 3/4 AEs included abdominal pain (6%), hypophosphatasemia (4%), nausea (3%), and diarrhea (2%).
Raut CP, Espat NJ, Maki RG, et al. PERSIST-5: Five-year extended treatment with adjuvant imatinib for patients with intermediate/high risk primary gastrointestinal stromal tumor (GIST). J Clin Oncol. 35, 2017 (suppl; abstr 11009).