Background: In patients (pts) with mCRPC receiving chemotherapy, measurable disease response by World Health Organization criteria is prognostic for overall survival (OS). We now explored the association of changes in measurable disease by Response Evaluation Criteria in Solid Tumors (RECIST) with OS. Methods: Data from the control arm (n = 612) of the VENICE trial receiving docetaxel, prednisone and placebo (DPP) were available. Data on baseline clinical variables and outcomes were obtained. Cox proportional hazards regression was used to evaluate the prognostic ability of RECIST changes adjusting for known factors for OS using a 90-day landmark analysis. The association was validated using 388 patients in the DPP arm of the MAINSAIL trial who had RECIST measurements prior to day 90 and survival beyond day 90. Results: 363 pts in VENICE had measurable lesions, of whom 296 were evaluable for landmark analysis. 28 (9.5%) had progressive disease (PD) prior to day 90, while 58 (19.6%) had unconfirmed partial response (PR). In a multivariable analysis, the hazard ratio (HR) for OS for pts with PR was 0.64 (95% CI 0.42 – 0.99, P = 0.045) compared to those without PR, and 1.78 (95% CI 1.07 – 2.95, P = 0.026) for those with PD compared to those without PD. PD remained significant (HR = 1.85, 95% CI 1.10-3.12, P = 0.020) after adjusting for PSA changes, but PR was not (P = 0.14). The association of PR (HR 0.51, 95% CI = 0.22-1.18, p = 0.12) and PD (HR = 3.51, 95% CI 1.92-6.43, p < 0.001) with OS was externally validated in 388 pts in MAINSAIL who had measurable disease. After adjusting for PSA changes, PD was associated with poor OS (HR = 2.36, 95% CI 1.11-5.04, p = 0.026), but PR was not (p = 0.15). Conclusions: In men with mCRPC receiving first-line docetaxel-based therapy, RECIST changes within 90 days are associated with OS. Given the frequent detection of measurable disease with current imaging and the unclear association of PSA and bone scan changes with outcomes in the setting of new agents, the accrual of patients with measurable tumors in phase II trials to assess RECIST changes may provide a more objective signal of efficacy of new agents.