Meta-analysis of preoperative chemotherapy (CTX) versus primary surgery for locoregionally advanced adenocarcinoma of the stomach, gastroesophageal junction, and lower esophagus (GE adenocarcinoma).

4022 Background: Patients (pts) with locally advanced GE adenocarcinoma have a poor outcome. We performed a systematic review with meta-analysis to assess if preoperative CTX compared to primary surgery improves survival. Methods: The review followed Cochrane methodology. Using a prespecified strategy, we searched relevant databases for randomized controlled trials comparing preoperative CTX (with or without radiotherapy) with primary surgery for locally advanced GE adenocarcinoma. Two independent reviewers selected eligible trials, extracted data and assessed quality. We performed random effects model meta-analysis with overall survival as primary endpoint. Hazard ratios (HRs) based on individual patient data (IPD) were calculated. If IPD were unavailable, HRs or binary mortality data were extracted from publications. Subgroup analyses for “pure” CTX, radiochemotherapy (RCTX) and tumor sites were prespecified. Results: 12 studies with 2, 128 pts met inclusion criteria. Preoperative CTX improved survival (HR = 0.81; 95% CI 0.73-0.90, p < 0.0001). Subgroup analyses for “pure” CTX and RCTX (the latter group including only trials with esophageal or GE junction tumors) yielded HRs of 0.82 (95% CI 0.73-0.91, p = 0.0002) and 0.75 (95% CI 0.52-1.08, p = 0.12). Survival varied according to tumor site (subgroup data not available from all trials) with an HR of 0.80 (95% CI 0.58-1.12, p = 0.2) for esophagus, 0.71 (95% CI 0.55-0.91, p = 0.007) for GE junction and 0.94 (95% CI 0.82-1.06, p = 0.31) for stomach. Conclusions: Preoperative CTX significantly improves survival of pts with locoregionally advanced GE adenocarcinoma and should be considered standard of care. An effect can be observed regardless of whether CTX is combined with radiotherapy or not. However, randomized controlled trials have assessed RCTX only for esophageal and GE junction tumors, and the power of the subgroup analysis was too low to reach significance. In subgroup analyses for tumor site, a significant effect on survival was found only for GE junction tumors, and not for tumors of the esophagus and stomach, but the power of the latter two analyses was low. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Novartis