Imatinib failure-free survival (IFS) in patients with localized gastrointestinal stromal tumors (GIST) treated with adjuvant imatinib (IM): The EORTC/AGITG/FSG/GEIS/ISG randomized controlled phase III trial.

Abstract

10500 Background: In 2004 we launched an open-label randomized trial with adjuvant IM for 2 yrs in localized, surgically resected, high/intermediate-risk GIST. Methods: Pts were randomized between 2 yrs of IM, 400 mg daily, and no further therapy after surgery. The primary end-point was OS, while RFS, RFI and toxicity were secondary end-points. Main eligibility criteria were: age >18 yrs, PS 0-2, localized CD117-positive GIST, intermediate or high risk according to the 2002 Consensus classification, R0 or R1 surgical margins, no previous medical therapy. Pts were stratified by risk, tumor site and margins. An accrual of 400 pts was planned, then escalated to 900. Given the prognostic improvement of advanced GIST pts, in 2009 the study IDMC authorized to change the primary end-point into IFS, whose failure was defined as the time when a different tyrosine kinase inhibitor (TKI) is started. We report on a planned interim analysis carried out after 115 events according to the new primary end-point, with a significance level of 1.5%. Hazard ratios (HR) and p-values were adjusted for stratification factors. Results: 908 pts were randomized between 2005 and 2008, 454 to IM and 454 to observation arm; 835 pts were eligible. 17% of pts treated with IM stopped early due to toxicity or refusal. With a median follow-up of 4.7 yrs, 5-yr IFS was 87% in the IM-arm vs 84% in the control arm (HR=0.80, 98.5% CI [0.51; 1.26], p=0.23); RFS was 84% vs 66% at 3 yrs, and 69% vs 63% at 5 yrs (p<0.001); 5-yr OS was 100% vs 99%. Among 528 pts with high-risk GIST by local pathology, 5-yr IFS was 79% vs 73% (p=0.11), and 77% vs 73% (p=0.44) amongst 336 high-risk GIST pts by centrally reviewed pathology. Conclusions: This study confirms that adjuvant IM has an overt impact on short-term freedom from relapse. In the high-risk subgroup, a non-statistically significant trend in favor of the adjuvant arm was observed in terms of IFS. This new end-point for the adjuvant setting, i.e. survival free from any failure of the first employed TKI, was designed to incorporate secondary resistance, i.e. the currently main factor adversely affecting prognosis of advanced GIST pts. Clinical trial information: NCT00103168.