Adjuvant capecitabine and oxaliplatin for gastric cancer: Results of the phase III CLASSIC trial.

Abstract

LBA4002 Background: Surgical resection is a recommended treatment for operable gastric cancer (GC) in general, despite high recurrence rates (40%–80%). Adjuvant chemotherapy aims to reduce recurrences; however, there is currently no universally accepted adjuvant regimen for GC. Methods: CLASSIC is a randomized, open-label, multicenter, international (South Korea, China, and Taiwan) study of XELOX (capecitabine 1000mg/m2 bid, d1–14, q3w and oxaliplatin 130mg/m2, d1, q3w x 8 cycles) vs observation, following D2 gastrectomy. Eligible patients were chemotherapy- and radiotherapy-naive, with stage II (T2N1, T1N2, T3N0), IIIa (T3N1, T2N2, T4N0), or IIIb (T3N2) GC resected within 6 weeks prior to randomization. The primary endpoint is 3-year disease-free survival (DFS). A sample size of 512 patients per arm was planned to observe the 385 DFS events required to provide 80% power at a 5% significance level for the hypothesized treatment effect (hazard ratio [HR] 0.75). The Independent Data Monitoring Committee recommended full evaluation and reporting of results following a positive pre-planned interim analysis at 266 events. Results: The XELOX and observation arms (ITT populations of 520 and 515 patients, respectively) were well balanced for baseline characteristics. The median duration of follow-up was 34.4 (16–51) months. XELOX-related grade 3/4 adverse events (AEs) occurred in 244/496 patients (49%) of the safety population. Neutropenia was the only AE observed in >10% of patients (21%, n=106/496). Serious XELOX-related grade 3/4 AEs occurred in 34/496 patients (7%). There were 62/496 (13%) and 80/476 (17%) deaths on study in the safety populations of XELOX and observation arms, respectively, mostly due to disease progression. Efficacy results in the ITT population are summarized below. Conclusions: This study demonstrates the superior efficacy of adjuvant XELOX vs observation alone following D2 gastrectomy. Although OS data are still immature, there is a trend towards superiority of XELOX. These data support the use of adjuvant XELOX for GC. [Table: see text]