Neoadjuvant therapy for patients with locally advanced gastric carcinoma with etoposide, doxirubicin, and cisplatinum -: Closing results after 5 years of follow-up

BACKGROUND, The intent to curatively treat patients with gastric carcinoma is based on complete surgical resection of the primary turner and its lymphatic drainage. Postoperative adjuvant chemotherapy has failed to show a significant prognostic advantage for these patients. Preoperative chemotherapy, based on promising results in the treatment of patients with disease in primarily unresectable stages, is still being evaluated for those with locally advanced gastric carcinoma. Most published studies still lack adequate staging methods, and long term results of this treatment modality are not known at present. METHODS. In a Phase II study, a series of 42 patients with locally advanced gastric carcinoma (International Union Against Cancer Stages IIIA, IIIB, and nr) initially were staged with endoscopy, with endoscopic ultrasound to establish the clinical tumor classification, with computed tomography scans to rule out tumor infiltration of adjacent organs and to detect distant metastases, and with surgical laparoscopy to exclude occult peritoneal carcinomatosis. Three or four planned cycles of neoadjuvant chemotherapy with etoposide, doxorubicin, and cisplatinum were given prior to total gastrectomy. RESULTS. After a complete follow-up of at least 5 years, there was a median survival of 19.1 months for all patients. Only patients who underwent a complete surgical tumor resection appeared to have a survival benefit, with a median survival of 28.4 months. A superior survival rate was seen in patients who had a major clinical response to chemotherapy, with a median survival of 45 months. CONCLUSIONS. Phase III studies comparing results from patients who undergo neoadjuvant chemotherapy followed by surgery with results from patients who undergo surgery alone should stress the value of adequate pretherapeutic staging and must be accompanied by studies of potential methods for predicting tumor response. (C) 2001 American Cancer Society.