The study was designed as a phase II trial with a random assignement to a calibration BMN 673 research buy arm A and to an experimental arm B. The sample size for arm B was calculated according to the design described by A’Hern [32]. A sample size of 53 patients was considered sufficient to give a 90% probability of rejecting a baseline response rate of 35% with an exact 5% one-sided significance test when the true response rate was 55%. The drug regimen should have been

considered for further studies if at least 25 responses were observed. The calibration arm had the same sample size. No formal comparison was planned. The objective response rate have been reported with its 95% confidence interval. All patients enrolled were considered in the LEE011 order intention-to-treat population (ITT). This population have been evaluated for the efficacy analysis, which was performed also on evaluable patients. Subjects who assumed at least one dose of drug have been considered as denominator in the safety analysis. The time to event analysis was performed

according the Kaplan-Meier method. Results Patients Characteristics From March 2003 to November 2005, a total of 104 patients were enrolled from 4 oncologic centers of the GOIM (Gruppo Oncologico Italia Meridionale), with 54 patients randomized to arm A (EPI/VNB) and 50 patients to arm B (PLD/VNB). All randomized patients have been evaluated selleck compound for activity and toxicity according to ITT analysis. Patient characteristics are listed in Table 1. None of the patients

have received any chemotherapy of for advanced disease; 20 patients in arm A and 21 patients in arm B had received adjuvant chemotherapy, not including anthracyclines or vinka alcaloids; 35 and 30 patients had received previous adjuvant hormonal therapy, and 10 and 11 patients had received endocrine treatment for advanced disease in arm A and B, respectively. Median age was 63 and 61 years, 10 and 9 patients were premenopausal, 44 and 41 postmenopausal in arm A and B, respectively; dominant site of disease was soft tissue in 3 (5.6%) and 9 (18.0%), bone in 11 (20.4%) and 9 (18.0%), viscera in 40 (74.0%) and 32 (64.0%) patients in arm A and B, respectively. Hormonal receptors were positive (ER and/or PgR) in 39 and 32 patients, negative in 13 and 15 patients, unknown in 2 and 3 patients in the two arms, respectively. Her-2, retrospectively evaluated in 35 and 38 patients in arm A and B, was overexpressed or amplified in 8 patients in each arm (14.8% and 16%, respectively). The median number of chemotherapy cycles administered was 6 in both arms (range, 1 to 8 in both arms). Table 1 Patient and tumor characteristics Characteristics Arm A(EV) = 54 Arm B(PLD/V) = 50   No. % No.