Assessment of adverse events All subjects were

questioned about adverse events (AEs) of treatment at each visit, and all adverse events reported were analyzed regardless of the investigators’ assessments of causality. The Medical Dictionary for Regulatory Activities (MedDRA, Version 8.1J) was used to categorize reported adverse events. Statistical analysis All the data analyses were performed by statisticians from Ono under the supervision and confirmation of data analyses by one of the authors (Ohashi, Y.). The intention-to-treat CT99021 (ITT) population comprised all CDK inhibitor patients who received at least one dose of study medication and who attended at least one follow-up visit for any observation Smad inhibitor of efficacies. The ITT population was used for all fracture and height analyses. Safety analyses population comprised all patients who received at least one dose of study medication

in either treatment group. A per-protocol (PP) approach was used as a primary approach to analyze the bone turnover markers because they can change rapidly by protocol violations, interruption of study therapy, or concurrent illness. The PP approach excluded protocol violators who took less than 75% study drug, who took prohibited medications during the course of the trial, or who violated the protocol in a significant manner as specified in the data analysis plan, and patients who took study drug for less than 12 months. This population included all patients in the ITT population, except those with a protocol deviation deemed to have a significant impact on the efficacy variables, i.e., major deviations regarding the inclusion/exclusion criteria, patients with insufficient compliance (<75% of the study medication), documentation of forbidden concomitant

medication that could bias the fracture results, and patients lacking an assessable baseline and follow-up for X-ray assessments for less than 12 months. The risk of patients with new morphometric vertebral fractures at 24 months, as the primary endpoint, was analyzed by testing the superiority of minodronate group to the placebo group by the time-to-event curves (Kaplan–Meier method), the event being the first new incident 4��8C vertebral fracture. The primary hypothesis was tested using an ITT analysis that was modified to include all subjects randomized, who had taken at least one dose of study drug, and attended at least one follow-up visit. A Cox regression model was used to estimate the relative risk of vertebral fracture and its 95% confidence interval in minodronate group and placebo group. Log-rank test was used to determine the superiority of the minodronate group to the placebo group. The power calculation was based on the predictive risk of vertebral fracture. For the study to achieve a power of 90% to detect the superiority, a sample size of 290 subjects per group was required.