BACKGROUND The 5-year survival rate of patients with stomach cancer is usually around 20%. The clinico-pathological features that are presently used to assess patient prognosis are not sufficient to define gastric tumor behavior. Therefore, an accurate analysis of different biological characteristics of gastric cancer cells could allow the course of disease to be predicted and may help to improve treatment strategies. METHODS The prognostic values of DNA ploidy, proliferative activity and epidermal growth factor receptor (EGF-R) expression were studied in gastric tumors from a series of 63 patients. DNA ploidy and proliferative activity, evaluated in terms of DNA index (DI) and proliferative index (PI), respectively, were determined by flow cytometry on paraffin-embedded tumor tissues. EGF-R expression was detected by immunohistochemistry on paraffin-embedded tumor sections of the same specimens. The clinico-pathological and the biological parameters were then correlated, and the patients overall survival was calculated using a chi-square test and the Kaplan-Meier method. RESULTS DNA ploidy abnormal cell clones were found in 44% of cases (median DI = 1.4, range 1.04-2.5). Aneuploid tumors showed high PI more frequently than diploids (71% versus 36%, p = 0.01). The analysis of the expression of EGF-R revealed that 88% of aneuploid tumors were positive for receptor expression. On the contrary, diploid tumors showed the presence of EGF-R only in 56% of cases (p = 0.01). DI, PI, and EGF-R expression were not related to histological grade. Conversely, the three biological parameters were significantly correlated to clinical stage and tumor invasion. The Kaplan-Meier survival curves showed a 73% 5-year survival rate in patients with diploid tumors whereas only 33% of patients with aneuploid lesions had a good prognosis (p = 0.001). CONCLUSIONS We demonstrate that DNA ploidy, PI, and EGF-R expression are closely related to some pathological and clinical characteristics in gastric cancer. The close relationship between aneuploidy, EGF-R positive expression, node involvement, and tumor invasion suggests that these parameters may be indicators of high malignancy. Finally, the results also show that aneuploidy and EGF-R-positive expression are indicative of a worse prognosis in gastric cancer patients. The study of these parameters might allow a more accurate stratification of patients, so that a targeted therapeutic protocol may be defined.