Predictors of overall and recurrence-free survival after neoadjuvant chemotherapy for gastroesophageal adenocarcinoma: Pooled analysis of individual patient data (IPD) from randomized controlled trials (RCTs)

Abstract

BackgroundNeoadjuvant chemotherapy improves prognosis of patients with locally advanced gastroesophageal adenocarcinoma. The aim of this study was to identify predictors for postoperative survival following neoadjuvant therapy. These could be useful in deciding about postoperative continuation of chemotherapy. MethodsThis meta-analysis used IPD from RCTs comparing neoadjuvant chemotherapy with surgery alone for gastroesophageal adenocarcinoma. Trials providing IPD on age, sex, performance status, pT/N stage, resection status, overall and recurrence-free survival were included. Survival was calculated in the entire study population and subgroups stratified by supposed predictors and compared using the log-rank test. Multivariable Cox models were used to identify independent survival predictors. ResultsFour RCTs providing IPD from 553 patients fulfilled the inclusion criteria. (y)pT and (y)pN stage and resection status strongly predicted postoperative survival both after neoadjuvant therapy and surgery alone. Patients with R1 resection after neoadjuvant therapy survived longer than those with R1 resection after surgery alone. Patients with stage pN0 after surgery alone had better prognosis than those with ypN0 after neoadjuvant therapy. Patients with stage ypT3/4 after neoadjuvant therapy survived longer than those with stage pT3/4 after surgery alone. Multivariable regression identified resection status and (y)pN stage as predictors of survival in both groups. (y)pT stage predicted survival only after surgery alone. ConclusionAfter neoadjuvant therapy for gastroesophageal adenocarcinoma, survival is determined by the same factors as after surgery alone. However, ypT stage is not an independent predictor. These results can facilitate the decision about postoperative continuation of chemotherapy in pretreated patients.