Objective: Cisplatin-gemcitabine (PG) and cisplatin-etoposide (PE) combinations are active regimens for non-small cell lung cancer (NSCLC). The present study aimed to compare PG with PE in the treatment of patients with stage IIIB and IV NSCLC. Methodology: We conducted a prospective, multicentre trial. A total of 166 patients were enrolled into the study and received either gemcitabine (1000 mg/m2) on days 1, 8 and 15 plus cisplatin (80 mg/m2) on day 2 every 4 weeks, or etoposide (100 mg/m2) on days 1, 2 and 3 plus cisplatin (80 mg/m2) on day 1 every 3 weeks. Results: The overall response rate was superior in the PG group (54.8%vs 39.0%, P = 0.045). There was no significant difference in survival between the two groups, with respective median and 1-year survival of 38 weeks and 33.3% for the PG group, and 34 weeks and 23.2% for the PE group. There was also no statistical difference for time to progression between the two groups. Neutropenia and thrombocytopenia were seen more frequently in the PG group (grade 3 neutropenia, 33.3%vs 15.9%, P = 0.012; grade 3 thrombocytopenia, 27.4%vs 3.7%, P < 0.001 and grade 4 thrombocytopenia, 10.7%vs 1.2%, P = 0.018). Conclusion: PG is an active chemotherapy regimen and has a better response rate than PE in advanced NSCLC, although there was no difference in time to progression and overall survival. A higher incidence of haematological toxicity was seen with PG than with PE.