Outcomes of patients (pts) with advanced gastro-intestinal stromal tumors (GIST) treated with multi-kinase inhibitors other than imatinib (IM) as first-line treatment.

11532 Background: IM is the standard first-line therapy in advanced GIST, with a median progression-free survival (PFS) of 30 months. Recent multi-kinase inhibitors (MKIs) such as nilotinib, dasatinib or masitinib have been tested as first-line therapies in phase II/III studies. This might theoretically result in increased PFS (by the addition of a new line of treatment), or in early emergence of resistance to approved MKIs. Methods: A retrospective chart review was performed in GIST pts who received investigational MKIs (in phase II/III trials) as first-line treatment, followed by IM as second line. Data on demographics, molecular profile, PFS, and overall survival (OS) were collected in two French referral centers. Results: Of 47 pts, (57% females), 22 (47%) had a KIT exon 11 mutation, one a KIT exon 9 mutation (2%), one a PDGFR D842V mutation (2%). Five patients were wild-type for KIT and PDGFR. The mutational status was unknown in 18 pts (38 %). From 2005 to 2011, 21 pts (45%) received masitinib, 18 (38%) received dasatinib and 8 pts (17%) received nilotinib. Median PFS on first-line treatment was 18.9 months [95%IC: 9.0-26.0]. Median time-to-failure (TTF) with IM was 19.7 months [95%IC: 14.8-53.4]. Median time to second relapse was 50.2 months [95%IC: 31.2-92.2]. Thirty-five patients (74.5%) were dead at the end of follow-up. The median OS from time of initial diagnosis was 5.9 years [95%IC: 4.5-8.2]. Conclusions: GIST pts who received MKIs other than IM as first-line treatment and IM as second-line had a time to second relapse longer than that observed historically with IM in first line. This suggests that using MKIs other than IM in first line does not decrease IM efficacy in second line. Further comparative studies are needed to confirm these findings, but this is encouraging to further develop studies with other MKIs in the first line setting.