151 Background: ENZAMET randomized 1125 participants with mHSPC to compare enzalutamide (ENZA) versus a standard non-steroidal anti-androgen (NSAA) and demonstrated superior progression-free survival and overall survival (PFS and OS) with ENZA. Radiographic progression in the absence of prior/concurrent PSA progression (rProg1st) is an emerging biomarker of poor clinical outcomes. We sought to determine the frequency of rProg1st, and correlate the impact of enzalutamide on transitions between disease states for the ENZAMET cohort. Methods: The ENZAMET dataset was analyzed using a multi-state Cox proportional hazards regression model that partitioned the clinical experience of participants (pts) into 4 states: (1) Evt-Free (event-free), (2) rProg1st (radiologic progression recorded without prior/concurrent evidence of confirmed PSA progression), (3) OtherProg (All Other type of clinical progression events (PSA and treatment switch, excluding death), (4) Death. Results: Radiographic progression was recorded in 388/1125 (34%) pts. Radiographic progression without confirmed prior/concurrent PSA progression per protocol was recorded in 114/1125 (10%) with similar proportions in those assigned ENZA 55/114 (48%) vs NSAA 59/114 (52%). Baseline characteristics of the 114 pts with rProg1st were similar to the other pts in ENZAMET, and were similar in the ENZA and NSAA groups (Table 1). Compared with NSAA, ENZA delayed both rProg1st (HR 0.66, 95%CI: 0.46 to 0.96, p=0.03) and OtherProg (HR 0.37, 95%CI: 0.31 to 0.44, p<0.001). 5-year OS rates were 24% (95%CI: 18-34) in the rProg1st group versus 42% (95%CI: 38-47) in the OtherProg group. Of those who had not progressed (495/1125) with a median follow-up of 68 months, 8% had died of causes other than prostate cancer. As previously reported ENZA prolonged overall survival in the whole trial cohort (N=1125, HR 0·70, 95%CI 0·58 to 0·84, p<0·0001). Conclusions: Participants who had radiographic progression without prior/concurrent PSA progression had worse overall survival whether assigned NSAA or enzalutamide. Enzalutamide reduced the hazards for, and delayed the times to, rProg1st and OtherProg. There were no baseline characteristics that helped identify these pts upfront; we plan molecular biological analyses to help identify this unique group earlier. Clinical trial information: NCT02446405 . [Table: see text]