Ronellenfitsch, Ulrich; Schwarzbach, Matthias; Hofheinz, Ralf; Kienle, Peter; Kieser, Meinhard; Slanger, Tracy E; Burmeister, Bryan; Kelsen, David; Niedzwiecki, Donna; Schuhmacher, Christoph; Urba, Susan; van de Velde, Cornelis; Walsh, Thomas N; Ychou, Marc; Jensen, Katrin
Preoperative chemo(radio)therapy versus primary surgery for gastroesophageal adenocarcinoma: systematic review with meta-analysis combining individual patient and aggregate data.
The prognosis of patients with gastroesophageal adenocarcinoma is poor. There is conflicting evidence regarding effects of preoperative chemotherapy on survival and other outcomes.We conducted a meta-analysis with aggregate and individual patient data (IPD) to assess the effect of preoperative chemotherapy for gastroesophageal adenocarcinoma on survival and other outcomes. Two independent reviewers identified eligible randomised controlled trials (RCTs) comparing chemotherapy+/-radiotherapy followed by surgery with surgery alone for gastroesophageal adenocarcinoma. IPD was solicited from all trials. Meta-analyses were performed using the two stage method.We identified 14 RCTs (2422 patients). For eight RCTs (1049 patients; 43.3%) we obtained IPD. Preoperative chemotherapy was associated with longer overall survival (hazard ratio [HR] 0.81; 95% confidence interval [CI] 0.73-0.89; p<0.0001). There were larger treatment effects in tumours of the gastroesophageal junction and for chemoradiotherapy compared to chemotherapy, but the tests for subgroup differences were not statistically significant. Preoperative chemotherapy was associated with longer disease-free survival, higher likelihood of R0 resection and more favourable post-treatment tumour stage, but not perioperative complications.Preoperative chemotherapy for locoregional gastroesophageal adenocarcinoma increases survival compared to surgery alone. It should be offered to all eligible patients. There appear to be larger survival advantages in tumours of the gastroesophageal junction and for chemoradiotherapy, but these findings require prospective confirmation.