Response to neoadjuvant chemotherapy best predicts survival after curative resection of gastric cancer

OBJECTIVE: In Western populations, long-term survival rates after curative resection of gastric cancer remain extremely poor. The lack of effective adjuvant therapy has prompted the evaluation of neoadjuvant approaches. Since 1988, we have conducted three separate phase II trials using neoadjuvant chemotherapy to treat patients with potentially resectable gastric cancer. The present study was conducted to evaluate whether response to neoadjuvant chemotherapy is predictive of survival in patients with resectable gastric cancer. METHODS: Eighty-three patients with pathologically confirmed gastric adenocarcinoma were treated with neoadjuvant chemotherapy before planned surgical resection. Response was assessed by upper gastrointestinal series, endoscopy, computed tomography scan, and pathologic examination. RESULTS: For the three phase II trials, clinical response rates ranged from 24% to 38%. Three patients (4%) had a complete pathologic response. Sixty-one patients (73%) underwent a curative resection. Median follow-up was 26 months. Univariate analysis revealed T stage, number of positive nodes, and response to chemotherapy to be significant predictors of overall survival. However, on multivariate analysis, response to chemotherapy was found to be the only independent prognostic factor. CONCLUSIONS: Response to neoadjuvant chemotherapy is the single most important predictor of overall survival after neoadjuvant chemotherapy for gastric cancer. These findings support further evaluation of neoadjuvant approaches in the treatment of this disease.