Seventeen patients with small cell lung cancer entered a phase II trial testing the feasibility of adding high dose epirubicin (100-120 mg/m2, day 1) in combination with etoposide (60-80 mg/m2, days 1-5) and cisplatin (70 mg/m2, day 1) courses repeated every three weeks. Complete responders received thoracic (40 Gy) and prophylactic cranial (30 Gy) irradiation. Sixteen patients were evaluable for response and toxicity. Myelosuppression was the dose-limiting side effect. Neutropenic fever was observed in eight patients (53%) and stomatitis in six (40%). No patient had a greater than 14% decline in the cardiac ejection fraction. Strict adherence to the dose-schedule designed was impossible as doses were trimmed and delayed in 30% of instances. The overall objective response rate was 81% (95% confidence limits, 62% to 100%), in limited disease there were complete remissions in 57%. With a 16 months median follow-up, overall median survival was 13 months. This study was unable to prove the feasibility of epirubicin escalation when added to etoposide-cisplatin combination, hampering the dose-intensification Norton-Simon model test.