Background: Intracellular gemcitabine triphosphate (dFdCTP) levels can be optimised by administering gemcitabine at a fixed dose rate infusion. Patients and methods: Patients with chemonaive advanced non-small cell lung cancer (NSCLC) were randomised to receive gemcitabine at a fixed dose rate gemcitabine 750 mg/m2 over 75 min (arm A) or gemcitabine 1000 mg/m2 over 30 min (arm B) on days 1 and 8 every three week cycle. Carboplatin at AUC of 5 was administered in both treatment arms on day 1 of each cycle. End points were activity, tolerability and pharmacokinetics of plasma and intracellular gemcitabine. Results: 76 patients were randomised. Response rate was 34 % in arm A and 42% in arm B. Toxicity and quality of life scores were similar for both treatment arms. Mean plasma Cmax gemcitabine and mean dFdCTP AUC in arm A was 20.8 μM ± 17.2 μM and 35 079 ± 18 216 μM*min respectively and in arm B, 41.2 ± 13.9 μM and 32 249 ± 11 267 μM*min respectively. dFdCTP saturation was reached in Arm B but not in Arm A. Conclusion: The saturability of dFdCTP accumulation in Arm A suggests optimal delivery of gemcitabine is achieved using fixed rate infusion compared to 30-min infusion. Fixed dose rate gemcitabine is active and feasible, supporting the concept of fixed dosing rate of gemcitabine in advanced NSCLC. However, this entails a longer infusion time with associated higher costs involved.