The purpose of this study was to investigate the effect of long-term administration of G-CSF with regard to its impact on overall survival of patients with ovarian cancer. We report the results of a non-randomized trial on 64 patients with advanced ovarian cancer treated with 6 cycles of conventional chemotherapy. Chemotherapy comprised carboplatin 400 mg/m2 and epirubicin 70 mg/m2 on day 1 of each cycle and prednimustine 100 mg/m2 on days 3 to 7, every 28 days. Thirty-three patients received CEP chemotherapy with G-CSF support whereas 31 women received CEP chemotherapy alone. The schedule of G-CSF was 5 mg/kg/day subcutanously on days 8 to 21 of each cycle. The severity of reduction in white cells and neutrophil count was significantly different in the two treatment groups (p<0.05), with more toxicity in the non- G-CSF group. G-CSF users had a non significant 0.88-fold lower risk of dying from ovarian cancer (95% CI, 0.48-1.60, p=0.678). In a survival analysis using a Cox proportional hazards model, residual tumor remained as an independent prognostic factor. The increasing amount of residual tumor resulted in a 1.767-fold higher risk (95% CI, 1.23-2.53, p=0.002) of death secondary to the underlying disease. In conclusion, this trial has failed to demonstrate any negative impact on patients' overall survival for the additional use of G-CSF with platinum-based chemotherapy; our results were consistent with the beneficial effects of G-CSF treatment on cytotoxic chemotherapy-induced myelosuppression.