Abstract Background Although modest improvements in the survival of patients with non-small cell lung cancer (NSCLC) can be achieved with cisplatin-based chemotherapy (CT), its value is disputed in the geriatric setting. In this study, we evaluate the feasibility of vinorelbine/cisplatin CT for elderly NSCLC patients. Methods In this pilot phase I/II trial, all patients received CT with vinorelbine 25 mg/m 2 , on day 1 and 8, and cisplatin on day 1, in 28 days-cycles. After stratification for age (up to 75 years), younger patients were sequentially allocated to moderate cisplatin doses (80 mg/m 2 or 90 mg/m 2 ), and older patients were allocated to lower cisplatin doses (60 mg/m 2 or 70 mg/m 2 ). We recruited patients aged over 70 years with newly diagnosed NSCLC, clinical stage III or IV, Karnofsky performance status ≥ 70%, normal serum creatinine, peripheral neuropathy ≤ grade 1, and no prior cancer therapy. Results Analysis was by intention to treat. Main toxicities (grade 3–4) was as follows: neutropenia, 20%; anemia, 11%; and thrombocytopenia, 2%; alopecia, 55%; fatigue, 11%; and peripheral neurotoxicity, 2%. No grade 3–4 emesis or renal toxicity occurred. Global median time to progression (TTP) and overall survival (OS) were 27.0 (95% CI: 10.1 to 43.7) weeks and 30.1 (95% CI: 24.4 to 35.8) weeks; 1- and 2-year survival rates were 36.3% and 13.2%, respectively. Overall response rate was 50.0% (95% CI: 35.4% to 64.5%), with 1 complete response; no difference on response rate was noticed according to cisplatin dose. Median overall survival was 30.1 weeks, with 1- and 2-year survival rates of 36.3% and 13.2%, respectively. Conclusion Age does not preclude assessment on the role of cisplatin-vinorelbine CT for elderly NSCLC patients with good performance status and adequate bodily functions.