15169 Background: Currently, there is no established second-line chemotherapy for pts with advanced gastric cancer who failed to respond or progressed after a first-line chemotherapy. Many of these pts still have a good performance status or have symptoms to be palliated at the time of first-line failure and are candidates for second-line chemotherapy. However, phase II trials demonstrate divergent results about pts more likely to benefit from second-line chemotherapy. We retrospectively analyzed the influence of various clinicopathologic factors on the survival of pts treated with second-line chemotherapy. Methods: Analysis is based on the data of 169 pts consecutively treated at 3 oncology department with a second-line chemotherapy. Univariate and multivariate analyses were performed to determine prognostic factors of survival. The variable used for analysis were: sex, age, ECOG performance status, a weight loss > 5 Kg in the last month; site of primary tumor, histopathology; hemoglobin, serum albumin, and CEA levels, number and site of metastatic disease, response to and time-to-progression (TTP) with the first- line chemotherapy. Results: The variables predictive of better survival were: ECOG PS 0–1 (p<0.001), no weight loss (p=0.001), hemoglobin level > 10 g/dl (p=0.01), CEA level <50 U/ml (p<0.02), number of metastatic sites = 2 (p=0.002), TTP of the first-line chemotherapy > 4 months (p=0.008). Peritoneal carcinomatosis was predictive of poor survival only when associated with one or more signs or symptoms as vomiting, anorexia, abdominal pain, ascites(p=0.03). Four factors were independently associated with better overall survival: ECOG PS 0–1 (p=0.002; HR 0.46; CI 95%, 0.29–0.75), CEA level <50 U/ml (p=0.008; HR 0.54; CI 95%, 0.35–0.85), one or two metastatic sites of disease (p=0.01; HR 0.58; CI 95%, 0.39–0.88), and TTP of the first-line chemotherapy > 4 months (p=0.02; HR 0.66; CI 95%, 0.45–0.95). Conclusions: In the absence of data deriving from randomized, controlled, clinical trials, this analysis suggests that some clinical factors may help clinicians to better select groups of pts with gastric cancer more likely to benefit from a second-line chemotherapy. No significant financial relationships to disclose.