RESUMO
BACKGROUND: Although postoperative adjuvant chemotherapy with S-1, an oral fluoropyrimidine, has become a standard of care for gastric cancer in Japan, nonresponders may suffer from the cost and adverse reactions without clinical benefit. This multicenter exploratory phase II trial was conducted to see whether a chemosensitivity test, the collagen gel droplet embedded culture drug sensitivity test (CD-DST), can adequately select patients for chemotherapy. METHODS: The CD-DST using four different concentrations of 5-fluorouracil was conducted with resected specimens from preregistered patients who underwent gastrectomy with D2 or more extensive lymphadenectomy. Patients who were histopathologically confirmed to have stage II or greater disease without distant metastasis were eligible for final enrollment. All patients underwent protocol-specified adjuvant chemotherapy with S-1. Three-year relapse-free survival was compared between patients determined as sensitive by the CD-DST (responders) and those deemed insensitive (nonresponders). Appropriate cutoff values for in vitro growth inhibition were defined when the hazard ratio for relapse in responders and the log-rank P values were at their minimum. RESULTS: Of the 311 patients enrolled, 14 were ineligible and 27 failed to start the protocol treatment. The CD-DST failed in 64 other patients, and survival analyses were conducted with the remaining 206 patients (39 stage II disease, 155 stage III disease, and 12 stage IV disease). The outcome of patients who were determined to be responders was significantly superior to that of nonresponders regardless of the 5-fluorouracil concentrations, although no differences in clinicopathologic characteristics were observed between the two groups, except for age. CONCLUSIONS: The CD-DST identified those who benefit from adjuvant chemotherapy. It deserves further evaluation in the setting of a prospective randomized trial. ClinicalTrials.gov identifier: NCT00287755.