Abstract Background: Selecting targeted therapies and assessing outcome in patients (pts) with castration-resistant prostate cancer (CRPC) are significant unmet medical needs. A proportion of CRPC remains dependent on androgen receptor (AR) activation. MDV3100, a second-generation AR antagonist optimized from a screen for nonsteroidal antiandrogens that retain activity in the setting of increased AR expression blocks nuclear translocation of AR and DNA binding and has no known agonist activity when AR is overexpressed. Methods: Pts with progressive CRPC were enrolled in sequential cohorts of 3–6 pts at 30, 60, 150, 240, 360, 480 and 600 mg/day. Once the safety of a dose was established, enrollment was expanded at doses >60 mg/day to include 12 chemotherapy-naïve and 12 post-chemotherapy pts per cohort. Antitumor effects were assessed using the Prostate Cancer Working Group (PCWG) 2 criteria reporting post-therapy PSA changes from baseline, and radiologic imaging for soft-tissue disease by RECIST and osseous disease on radionuclide bone scan as improved, progressed or no change. Circulating tumor cell (CTC) enumeration was performed with the FDA cleared analytically valid CellSearch assay (Veridex, LLC, Huntingdon Valley, PA) and reported as the number of cells per 7.5 ml of blood as previously described. Samples were collected at baseline, 4 weeks, and 12 weeks post-treatment and at progression of disease. Samples from outside Center were shipped overnight and processed in the CLIA certified MSKCC Clinical Chemistry laboratory. Results: A total of 140 pts were enrolled, of which 65 (46%) were chemotherapy-naïve, and 75 pts (54%) were in post-chemotherapy setting. Maximal PSA decline from baseline of ≥ 50% was achieved in 40/65 (62%) chemotherapy-naïve and 38/75 (51%) post-chemotherapy. At 12 weeks, radiographic control (no progression) was observed in 35/47 pts (74%) with evaluable soft tissue lesions per PCWG2 guidelines and 50/81 pts (62%) with bone lesions. At 600 mg/day, 2 of 3 pts had dose limiting toxicity (rash; seizure). Dose reductions due to fatigue were noted at 480 and 360 mg/day. The maximal tolerated dose was 240mg daily based on safety data. CTC counts were obtained from 128/140 (91%) pts enrolled; 16/60 pts chemotherapy-naive and 35/68 pts post-chemotherapy had ≥ 5 CTC/7.5 mL blood at baseline. CTC conversion from baseline CTC ≥ 5 (unfavorable) to CTC <5 (favorable) with treatment was noted in 75% (12/16) in chemotherapy naïve vs. 37% (13/35) in post-chemotherapy pts. In the post-chemotherapy group, patients with unfavorable CTC at baseline had a PSA decline ≥ 50% with treatment in 85% (11/13 patients) when they converted to favorable CTC counts vs. 32% (7/22) if CTC remained unfavorable (p=0.005, Fisher Exact). The PSA decline ≥ 50% was achieved in 16 of 30 (53%) pts when CTC remained favorable throughout treatment, and 1 of 3 (33%) pts with favorable baseline CTC but unfavorable CTC at follow up. In the chemotherapy naive group, CTC conversion from unfavorable to favorable was associated with PSA decline ≥ 50% in 8/12 (67%) pts compared to 1/4 (25%) in the group with a change from favorable to unfavorable, and 1/4 (25%) in the group with unfavorable CTC at baseline and follow up; when CTC stayed in the favorable category, the PSA decline ≥ 50% was achieved in 26/40 (65%) pts. Conclusions: MDV3100 demonstrated favorable efficacy assessed by prostate-specific antigen (PSA) decline and CTC enumeration. The efficacy comparable to that at higher doses and the better adverse event profile, led to the selection of 160 mg/day as the recommended dose for future studies. A prospective investigation of pre and post-therapy CTC counts and clinical outcomes in a phase III randomized, double-blind, placebo-controlled efficacy trial of MDV3100 has been recently initiated. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):CN02-03.