Background: Orteronel is an investigational, non-steroidal selective inhibitor of 17,20-lyase; a key enzyme in the production of steroidal hormones. Methods: Eligible men with metastatic castration-resistant prostate cancer (mCRPC) had progressive disease (PD; radiographic or prostate-specific antigen), castrate levels of testosterone, and had received more than or equal to 360 mg/m²docetaxel within the prior 6 months. Prior orteronel, abiraterone, or ketoconazole was not permitted. Patients were randomized 2:1 to continuous 28-day cycles of oral orteronel 400 mg BID + prednisone 5 mg BID, or placebo + prednisone without regard to food. Primary endpoint: overall survival (OS); other key endpoints: radiographical progression-free survival (rPFS), 50% or more PSA decrease at 12 weeks, pain response at 12 weeks and safety (NCT01193257). Results: One thousand ninety nine patients were randomized.The study was terminated for failing to meet its primary endpoint: median overall survival (OS) was 17.0 months (95% CI 15.2, 19.9) in patients receiving orteronel versus 15.2 months (95% CI 13.5, 16.9) in those receiving placebo (HR: 0.886 [95% CI: 0.739, 1.062]; P=0.1898). Substantial regional differences were seen in OS benefit: median OS (orteronel vs. placebo) was 20.9 vs. 16.9 mo (HR: 0.889) in North America (n=112), 18.3 vs. 17.8 mo (HR: 1.048) in Europe (n=590), and 15.3 vs 10.1 mo (HR: 0.709) in the rest of the world (n=397). In the overall population, rPFS was significantly improved in the orteronel arm, with a median of 8.3 months vs. 5.7 months in the placebo arm (HR: 0.76 [95% CI: 0.653, 0.885]; P=0.00038). Drug-related adverse events (AEs; any grade) included (orteronel/placebo) nausea (30/16%), vomiting (23/8%), fatigue (17/11%), and diarrhea (16/9%); grade 3 or higher drug-related AEs included increased lipase (12/less than 1%), increased amylase (8/less than 1%), and fatigue (3/3%). Additional sub-analyses looking at potential factors that affected OS will also be reported. Conclusions: Whileorteronel + prednisone did not show a statistically significant overall OS improvement versus placebo + prednisone, rPFS findings and striking regional OS differences suggest that orteronel has clinically meaningful activity. Clinical trial information: NCT01193257.