OBJECTIVE Bendamustine is an alkylating agent with high efficacy in non-Hodgkin's lymphoma and multiple myeloma. Even in solid tumours, monotherapy with bendamustine has resulted in subjective remissions and has been associated with a low rate of side effects. The current dose-finding study was designed to determine the maximum tolerated dose (MTD) of combined carboplatin/bendamustine in previously untreated patients with extensive-stage small cell lung cancer (SCLC). METHODS Carboplatin was administered as a 1-hour infusion on day 1 at increasing dose levels, and bendamustine was administered as a short infusion on days 1 and 2 at increasing dose levels (80-120 mg/m(2)). The regimen was administered every 3 weeks. Four dose levels were planned, starting with 80 mg/m(2) bendamustine and carboplatin area under the curve (AUC) 5 (dose level I). The other dose levels were 100 mg/m(2) bendamustine and carboplatin AUC 5 (dose level II), 100 mg/m(2) bendamustine and carboplatin AUC 6 (dose level III), and 120 mg/m(2) bendamustine and carboplatin AUC 6 (dose level IV). A minimum of three patients were enrolled at each dose level. RESULTS Dose-limiting toxicities, which included fatigue, infection and tachyarrhythmia, were observed at dose level III. The recommended dose for phase II studies was therefore established at dose level II. The majority of haematological and non-haematological toxicities observed were only mild (grade 1 or 2) in patients at dose levels I and II. None of the patients developed severe alopecia. Objective responses were observed in eight of the ten patients involved in this trial. CONCLUSIONS Because of its acceptable toxicity and favourable preliminary antitumour efficacy, the combination of carboplatin and bendamustine appears to be a potentially useful chemotherapeutic option in patients with extensive SCLC.