(A) ARN-509 structure. (B) Representative competitive binding-assay vs ¹⁸F-FDHT (LNCaP/AR(cs)). IC₅₀-values: 11.5.nM (FDHT), 16nM (ARN-509), 21.4 nM (MDV3100), 160nM (Bic (bicalutamide)) (error-bars:SD, n=3). Inset: IC₅₀-values (mean ± SEM) from five replicate experiments. (C) qRT-PCR analysis of AR-regulated genes (normalized to GAPDH). LNCaP/AR(cs) cultured (5% CSS) 3d with/without 1nM R1881 and DMSO (Veh) or drug (1,3,10μM). (D) VCaP cultured (5%CSS) for 2d, then treated for 7d. Left-panel:agonist-mode. Right-panel:antagonist-mode with 30pM R1881. Viable cells (CellTiter-GLO) plotted as %vehicle-control (DMSO) (n=3, mean±SEM).

(A) Representative confocal microscopic images (scale-bars:10μm) of LNCaP with AR-EYFP cultured (5% CSS) and treated with DMSO (Veh) or drug. Nuclear:cytoplasmic fluorescence-intensity of individual cells was quantified (n=3, mean±SEM). (B) ChIP of AR in LNCaP/AR(cs) cultured (5%CSS) and treated 1h with/without 1nM R1881 and either DMSO (Veh) or 10μM anti-androgen. RT-PCR: PSA, TMPRSS2 enhancers (mean±SD, n=2). (C) Activation of an androgen-regulated 4XARE-Luc reporter gene by wtAR and VP16-AR in LNCaP/AR-luc or Hep-G2 cells treated for 40–48h with/without 100 pM R1881 and either DMSO (Veh), or anti-androgen. Luciferase assay conducted on cell lysates: light-units relative to vehicle-control (n=3, mean±SEM).

(A) Luciferase-imaging of castrate male mice harboring LNCaP/AR-luc tumors pre-treatment (d0) and post-treatment (d17), normalized to tumor volume. (B) Castrate male mice bearing LNCaP/AR(cs) tumors (mean tumor-volume 200mm³) treated by daily-gavage with vehicle, or drug. %-change in individual tumor volume (9–10 tumors/treatment-group) after 28d. Unpaired t-test:Vehicle vs ARN-509: p=0.0001; Vehicle vs Bic: p=0.004; ARN-509 vs Bic: p=0.002. Plasma-concentrations measured 24h post-dose on d28 (mean±SEM: Bic 30.74±3.68; ARN-509 9.69±1.58). (C) Castrate male mice bearing LNCaP/AR tumors >100mm³ treated 5d by daily-gavage with vehicle or 10 mg/kg/day anti-androgen. Ki67 IHC performed on tumor-tissue resected 2h after final dose. # positively-stained cells calculated out of 300 cells counted in each of 5 random views (n=3 tumors/treatment group), and expressed relative to vehicle-controls (error:SD; Veh vs ARN-509 and Bic vs ARN-509: p>0.05 (1-way ANOVA with Bonferroni’s post-test)). (D) Castrate male mice bearing LNCaP/AR tumors treated as in (C) for 25d. TUNEL-staining performed on tumor-tissue resected after final dose. # positively-stained cells calculated, out of 300 cells counted in each of 5 random views (n=3 tumors/treatment-group), expressed relative to vehicle-controls (error:SD).

A–C, Castrate male mice bearing LNCaP/AR(cs) tumors treated by oral daily-gavage with vehicle, ARN-509 or MDV3100. Unpaired t-test used for statistical comparisons. (A) 1,10 mg/kg dose: %change in individual tumor volume (n=7–8/treatment-group) after 28d-treatment. p<0.05: drug vs vehicle. (B) 10, 30 mg/kg dose: %change in individual tumor volume (n=8/treatment-group) after 42d-treatment. p<0.05: drug vs vehicle. (C) 30, 100 mg/kg dose: %change in individual tumor volume (n=19–20 tumors/treatment-group) after 28d-treatment. * p<0.05 drug vs vehicle or drug vs drug.

A–C, Intact male beagle dogs treated by oral-gavage for 28d with vehicle (n=5) or ARN-509 10 mg/kg/day (n=4). Organs resected 24h after final-dose. Mean plasma-concentration of ARN-509 measured at sacrifice, 24h after final dose, was 17.5 μg/mL. (A) Prostate and epididymis weights (mean±SEM). p<0.05:Veh vs ARN-509 in prostate (2-way ANOVA, with Bonferroni post-test). Comparison for epididymis is insignificant. (B) H&E staining of transverse prostate-sections (upper-panels) and longitudinal epididymis sections (lower-panels) at low-power (100x, scale bar:200μm) and high-power (600x, scale-bar:20μm). Upper-panels: Fully-mature, active acini in vehicle-treated prostate containing intraluminal secretions (arrowhead), lined with folds of a single-layer of columnar epithelial cells with abundant cytoplasm. ARN-509-treated prostates: bilayer of cuboidal epithelial cells (arrow), small lumens with rare eosinophilic secretory material (arrowhead). Lower-panels: Vehicle-treated epididymis: tall columnar pseudo-stratified epithelium with lumens containing abundant spermatozoa (arrowhead). ARN-509-treated epididymis: cuboidal epithelium, lumens with little-to-no spermatozoa, containing multi-nucleated round cells suggestive of sloughed degenerate germ cells (arrowhead). (C) Dog prostates stained with antibody to Ki67 (200x, scale-bar:100μm) or TUNEL (400x, scale-bar:50μm). Ki67-staining (arrows) and TUNEL (arrowheads). No TUNEL-staining is evident in ARN-509-treated prostates. Results quantified as mean percentage (± SEM) of positive cells from representative areas from each prostate (n=5 for vehicle, n=4 for ARN-509). p<0.05 for TUNEL staining of Veh vs ARN-509 (1-way ANOVA with Bonferroni’s multiple comparison post-test). (D) Intact male mice bearing LNCaP/AR(cs) tumors, treated for 28d by oral daily-gavage with vehicle, ARN-509 or MDV3100 (10, 30 mg/kg). %change in individual tumor volume (n=5–8/treatment-group) after 28d-treatment.. * p<0.05 for each treatment relative to vehicle (unpaired t-test).