Long-term safety of regorafenib (REG) in advanced gastrointestinal stromal tumors (GIST): updated safety data of the phase 3 GRID trial

Abstract

Background: The GRID trial showed that REG improves progression‐free survival (primary endpoint; Da ta cutoff January 2012) versus placebo in patients ( pts) with advanced GIST after failure of imatinib and sunitinib (Lancet 2013; 381: 295). At progression, placebo pts could cross over to REG treatment and pts randomized to REG could continue. We present an updated safety analysis of pts in GRID who were treated with REG at any time (any REG; AR) and of the subgroup who had long‐term REG (LTR) treatment (>1 year). Methods: Of the 199 randomized pts (n = 133 REG; n = 66 placebo), 58 placebo pts crossed over to REG. At the time of this analysis (June 2015), a total of 190 pts (95%) were treated with AR and 75 (38%) had LTR. Starting dose was REG 160 mg once Da ily for the first 3 weeks of each 4‐week cycle. Results: The LTR group tended to have a better ECOG status (PS0: AR 57%/LTR 69%; PS1: AR 43%/LTR 31%). A similar proportion were ≥65 yrs of age (AR 32%; LTR 31%) and treated in the third‐line (56%; 61%) or fourth‐line and higher settings (44%; 39%). Median (range) treatment duration was 8.8 months (0.02‐ 48.4) for AR and 22.7 months (12.0 ‐ 48.4) for LTR. Median Da ily dose was 146.7 mg (AR) and 121.8 mg (LTR). All pts had an NCI‐CTCAE (v4.0) treatment‐emergent adverse event (TEAE), with the majority occurring in the first months of treatment and significantly lower rates in subsequent months. Most common REG‐related grade ≥3 TEAEs included (AR; LTR): hand‐foot skin reaction (20.5%; 28.0%); hypertension (24.2%; 34.7%); diarrhea (7.4%; 13.3%); fatigue (4.7%, 4.0%); and oral mucositis (2.1%, 2.7%). Although rates of treatment modifications due to REG‐related TEAEs were higher in the LTR group (66.8% AR; 82.7% LTR), discontinuation rates due to REG‐related TEAEs were similar (8.9% AR; 10.7% LTR), with the majority of events leading to discontinuation reported only once. There were no additional REG‐related deaths since the primary analysis. Conclusions: Patients who had LTR tended to have a better ECOG status. The updated safety profile of GIST pts treated with REG in GRID is consistent with the profile reported at primary study completion. For pts treated with REG >1 year, no unexpected safety findings were observed.