Background: Abi is a standard treatment (tx) in CN-pts with mCRPC. It is a potent and selective CYP 17 inhibitor. However, in many countries where abi has not been approved yet, keto is used as an alternative CYP 17 inhibitor. Preclinical data suggests that keto is a less specific and potent inhibitor of CYP 17. Clinical data (Peer et al, Prostate 2014) suggests that in docetaxel (D) refractory mCRPC, the outcome of abi tx may be superior to keto. However, there are limited clinical data comparing both agents in CN-pts with mCRPC. We aimed to compare the clinical effectiveness of abi vs keto in CN-pts with mCRPC, who were treated after the year 2004 (approval of D for the tx of mCRPC). Methods: Records from 72 CN-pts with mCRPC treated with abi in 5 Israeli centers were reviewed retrospectively, and matched by pre-tx risk category (favorable, intermediate, poor; Keizman, Oncologist 2012) to pts treated with keto 200 - 400 mg 3x day (international database, n = 156, from 4 centers across the US and Israel). We compared the PSA response (decrease ≥ 50% from baseline), biochemical and radiological progression free survival (PFS), and overall survival (OS) between the groups. PFS and OS were determined by Cox regression. Results: The groups were matched by pre-tx risk category (favorable, intermediate, poor; Keizman, Oncologist 2012), based on pretreatment NLR and PSA doubling time, and the prior response to a gonadotropin-releasing hormone agonist. The groups were balanced regarding age (72 abi vs 70 keto), time from primary tx to disease relapse, gleason score, pre-tx disease extent (limited-axial skeleton and/or nodal vs extensive- appendicular skeleton and/or visceral), and ECOG PS. In the groups of abi vs keto, PSA response was 75% vs 47% (OR 3.8, p = 0.04), median biochemical PFS 12 vs 6 months (HR 0.62, p = 0.03), median radiological PFS 16 vs 8 months (HR 0.54, p = 0.01), median OS not reached after a median follow-up time of 18 months vs 26 months, and tx interruption d/t adverse events 10% vs 22% (0R 0.65, p = 0.05). Conclusions: In CN-pts with mCRPC, the outcome of pts treated with abi may be superior to keto.