Background: Given the historical low prevalence of measurable disease in mCRPC, phase II trials have employed suboptimal endpoints accounting for prostate specific antigen (PSA) and bone scan changes. Changes in bone scan and PSA have not always translated to survival improvement. Improved computerized tomography technology may be increasing the proportion of men with measurable disease, suggesting that measurable changes need to be reconsidered. We analyzed phase III trials of mCRPC to systematically quantitate the proportion of mCRPC with measurable disease. Methods: Data from the both arms of published phase III trials of mCRPC were eligible for analysis. Baseline characteristics were required including the number enrolled, setting (pre-docetaxel [D], D-based, post-D), proportion of patients with measurable disease and the year of completion of trial accrual. General Linear Model was used to evaluate the difference in measurable disease rate based on setting and year of completion of accrual. Results: Seventeen phase III trials totaling 17,609 men with mCRPC were evaluable; 5160 were pre-D, 7573 were D-based and 4876 were post-D. The trials completed accrual between 2002 and 2012. Ten trials used RECIST 1.0, 5 trials used RECIST 1.1 and 2 trials used other criteria. The overall proportion of men with measurable disease was 47.8%. The measurable disease rate (range) in trials in the pre-D setting was 40.5% (30.5-57), in the D-based setting was 51.9% (29.1-87.5), and in the post-D setting was 48.9% (38.8-56.6). There was no statistical difference in the proportion of men with measurable disease based on setting or year of completion of accrual. Conclusions: The proportion of men with measurable disease in phase III trials of mCRPC completing accrual between 2002 and 2012 was 47.8%% with no significant difference based on setting or year of completing accrual. Given these higher measurable disease rates compared to historical rates and recent demonstration of strong association of RECIST changes with OS in mCRPC, RECIST changes need to be considered as a co-primary endpoint in phase II trials to obtain a firm signal of efficacy before launching phase III trials.