VEGFR signaling pathway the enzastaurin arm. By using a log rank test, the study had at least 60% power to achieve statistical significance at a 1 sided level of.20, assuming at least 50 PFS events at the final analysis and a true PFS hazard ratio of the enzastaurin arm to the placebo arm of.73.17 PFS and OS were compared between the 2 treatment arms using the log rank test at the 1 sided significance level of.20. In addition, Kaplan Meier estimations18 were performed on the observed distributions of PFS and OS, and Kaplan Meier curves were generated. The Cox regression model19 was used to estimate treatment group differences adjusted for prognostic factors. The incidences of selected AEs from the 2 treatment arms were compared using the Fishers exact test at a 1 sided significance level of.20. In addition, an interim analysis was conducted to assess HSP the safety parameters of LV5FU2 plus bevacizumab with or without enzastaurin after 15 patients had been randomized and completed at least 1 cycle of study treatment. A post hoc power analysis was also done to estimate the probability of the studys success if the primary efficacy analysis had been performed as originally planned, that is, after achieving 118 PFS events.
From March 7, 2008 to January 27, 2010, 123 patients were entered into the study. Of the 123 patients, 117 patients were randomly assigned to receive enzastaurin or placebo. Approximately 23 of the patients were male, with a median age of 64 years. Greater than 90% of patients aromatase presented with stage IV disease. Drug Administration A total of 115 patients were treated, with 57 patients in the enzastaurin arm and 58 patients in the placebo arm completing at least 1 cycle of treatment. The median number of cycles received was 9 in the enzastaurin arm and 10 in the placebo arm. A total of 27 patients in the enzastaurin arm completed at least 10 cycles of treatment, as did 33 patients in the placebo arm. At least 1 dose reduction occurred because of AEs in 2 patients in the enzastaurin arm and 2 patients in the placebo arm. Eighteen patients in the enzastaurin arm and 15 patients in the placebo arm had at least 1 dose omission because of an AE. AEs resulting in a dose omission that occurred in 1 patient in either treatment arm included diarrhea, dizziness, mucosal inflammation, fatigue, and vomiting. Efficacy In the 117 patients randomized, there were 73 PFS events abiraterone for the primary efficacy analysis. The median PFS from randomization was 5.8 months in the enzastaurin arm and 8.1 months in the placebo arm.
Censoring rates were 32.8% and 42.4% in the enzastaurin and placebo arms, respectively. No significant differences in treatment effect were found when PFS from regimen randomization was analyzed by subgroup. The median PFS from the start of first line therapy was 8.9 months in the enzastaurin arm and 11.3 months in the placebo arm. The median OS from randomization was not calculable because of high censoring rates of 77.6% and 91.5% for enzastaurin and placebo, respectively, 41 patients in the enzastaurin arm and 39 patients in the placebo arm had discontinued study treatment at the time ofTable 2 summarizes laboratory and nonlaboratory grade 3 and 4 AEs possibly related to study drug. Most AEs occurring during treatment within either arm were grade 1 or 2.