Summary:Preliminary randomized studies have failed to show a survival benefit of high-dose chemotherapy with alkylators in advanced breast cancer. Idarubicin is an active agent in breast cancer and is suitable for dose escalation. We designed a dose finding study with escalating high-dose idarubicin (HD-Ida) followed by fixed high-dose thiotepa+melphalan (HD-TM) with peripheral blood progenitor cells (PBPC) in MBC patients with stable disease or in partial response after six courses of induction chemotherapy with gemcitabine 1000 mg/m 2 days 1 and 4, epirubicin 90 mg/m 2 day 1, taxol 175 mg/m 2 day 1 (GET). Aims of the study were to identify the maximum tolerated dose (MTD) of idarubicin, to evaluate the cardiac safety and activity of HD-Ida and HD-TM after GET and to study the pharmacokinetic profile of idarubicin and idarubicinol. A total of 14 patients were treated. Idarubicin was administered as a 48 h continuous i.v. infusion at the following dose levels: 40 mg/m 2 (three patients), 50 mg/m 2 (three patients), 60 mg/m 2 (five patients) and 70 mg/m 2 (three patients). Mucositis was the dose-limiting toxicity and the MTD was 60 mg/m 2 . C max of Idarubicin and idarubicinol were 7.772.0 and 26.379.7 ng/ml at 40 mg/ m 2 and increased to 14.8+3.0 and 47.4+12.6 ng/ml at 70 mg/m 2 . AUCt 0-264 of idarubicin and idarubicinol increased from 423.27111.6 and 25817606 hng/ml at 40 mg/m 2 to 732.87140.2 and 459071258 hng/ml at 70 mg/m 2 . Conversion rates after HD-Ida and HD-TM were 28.6 and 38.5%, respectively. No episodes of cardiac toxicity were observed. We conclude that HD-Ida followed by HD-TM is feasible and devoid of cardiac toxicity. Moreover, the activity of HD-Ida after a epirubicin-containing regimen suggests incomplete crossresistance between the two drugs. High-dose chemotherapy (HDC) with autologous hemopoietic support has been used as intensification/consolidation treatment in advanced breast cancer responsive to standard chemotherapy. However, despite encouraging preclinical data and phase II results, preliminary data from phase III randomized trials have failed to show any survival benefit. 1-4 Most of the randomized trials have used four to six courses of induction chemotherapy followed by late intensification with a single course of HDC. High-dose regimens usually have included alkylating agents because of their steep dose-response curve, non-cell cycle specificity and broad clinical activity. 5 Attempts to improve these disappointing results include the development of more active induction regimens, the upfront use of HDC and sequential administration of high-dose drugs or regimens.Our group has shown that the combination of gemcitabine+epirubicin+paclitaxel (GET) is very active with an overall response rate of 92% and a complete response rate of 31%; 6 the administration of high-dose thiotepa followed by high-dose melphalan as consolidation of GET induces an improvement in quality of response in 44% of patients. 7 On the basis of the promising activity and acceptable toxicity observed in this study...