O-0012 Prognostic Factors of Overall Survival (OS) in Previously Treated Advanced Gastric Cancer (AGC): Analysis of the Phase 3 Granite-1 Trial

Abstract

ABSTRACT Introduction Patients with AGC who progress after first-line chemotherapy have few treatment options and limited survival. Novel treatment options are critical to improving outcomes in this difficult-to-treat population. Ascertaining prognostic factors may improve outcomes by identifying patients most likely to benefit from treatment. Methods In GRANITE-1, patients aged ≥18 years with confirmed AGC who progressed after 1 or 2 previous systemic chemotherapy lines were randomized 2:1 to oral everolimus 10 mg/day plus best supportive care (BSC) or placebo plus BSC. Randomization was stratified by region (Asia versus rest of world [ROW]) and previous chemotherapy lines (1 versus 2). Study drug was continued until disease progression or unacceptable toxicity. Primary endpoint was OS. Secondary endpoints included progression-free survival (PFS) and safety. Potential prognostic factors for OS were investigated using a multivariate Cox regression model. The prognostic factors considered were ECOG performance status (PS), previous gastrectomy, liver involvement, primary tumor anatomic site, weight loss in previous 3 months, and diffuse/mixed adenocarcinoma per Lauren classification. Results A total of 656 patients were enrolled between July 2009 and December 2010 and received everolimus (n = 439) or placebo (n = 217). Baseline characteristics were well balanced between arms; median age was 62.0 years, 73.6% were men, 91.9% had ECOG PS ≤1, 50.2% had previous gastrectomy, 45.6% had liver involvement, 60.8% had distal tumors, 27.0% had diffuse/mixed adenocarcinoma, and 20.1% had weight loss >5% in the previous 3 months. Estimated median OS was 5.39 months with everolimus and 4.34 months with placebo (HR for OS, 0.90; 95% CI, 0.75-1.08; P = 0.1244). Subgroup analysis showed a trend for a reduced risk of death with everolimus in patients enrolled in ROW (HR, 0.85; 95% CI, 0.65-1.10) and those enrolled in ROW treated with 2 previous chemotherapy lines (HR, 0.74; 95% CI, 0.50-1.09). Previous gastrectomy, ECOG PS 1 or 2, liver involvement, and weight loss >5% in the previous 3 months had significant prognostic value for OS (Table). Median PFS was 1.68 months with everolimus and 1.41 months with placebo (HR for PFS, 0.66; 95% CI, 0.56-0.78; P Conclusion Everolimus did not significantly improve overall survival in AGC patients who received previous chemotherapy. A trend toward reduced risk of death was observed for patients enrolled in ROW and those enrolled in ROW who received 2 previous chemotherapy lines. The safety profile was consistent with that previously observed for everolimus. Previous gastrectomy, ECOG PS, liver involvement, and weight loss showed a prognostic effect on survival. Table 1 .