Abstract Number: LB-248
Investigation of tumor cell activity of Abiraterone using in vitro model systems of androgen (A)-responsive prostate cancer
Abiraterone acetate (AA) has recently been approved for the treatment of metastatic castration-resistant prostate cancer. Clinical activity of AA has been demonstrated and is due to systemic reduction of A levels by inhibition of cytochrome P450 c17 (CYP17) activity. AA is converted to abiraterone (ABI) in vivo. We hypothesized that a direct effect of ABI on tumor cells also contributes to its antitumor activity. Utilizing LNCaP cells (human prostate cancer cell line [HPCCL]), the effects of ABI were compared in standard 2D and more complex 3D tumor cell cultures to explore tumor microenvironment factors (eg, stromal CYP17 activity).
Androgen receptor (AR) output was studied by Western blot, luciferase reporter assay, and Affymetrix® microarrays using ABI. By continued growth of LNCaP cells in defined calf serum, we have identified a culture-derived line that is sensitive to ABI inhibition in vitro, both in 2D and 3D cultures. These cells were used to study ABI inhibition of tumor cells in the presence or absence of ligand (dihydrotestosterone [DHT]) and with reference to other molecules targeting the A axis (MDV3100 and TOK-001).
ABI exposure resulted in a dose-dependent modulation of expression of A-dependent genes in Affymetrix microarray analysis of 2D and 3D cultures. Further analyses using 2D cultures only, at the level of AR-dependent protein modulation (eg, kallikreins [KLKs]), and an AR-driven luciferase reporter assay revealed that AR output was inhibited by concentrations of ABI as low as 10 nM (IC50 100-300 nM), comparable with inhibition by TOK-001 (IC50 30-100 nM). The potency of MDV3100 was greater (IC50 10-30 nM). Growth inhibition, as assessed by MTT assay, occurred at concentrations ~1 log higher for all compounds. To address the role of CYP17 inhibition, studies were conducted in the presence of DHT; IC50 for ABI increased (IC50 3-10 uM), but inhibition of AR output was still detected at the gene, reporter, and protein levels; similar results were seen with TOK-001. The shift in inhibition with MDV3100 was less (10-fold), but with apparent agonism at low nM concentrations. Similar data were also obtained in VCaP cells (another HPCCL), excluding a role for mutant AR in the observed response.
ABI has an inhibitory effect on AR output of prostate tumor cells at nM concentrations. Inhibition was partially overcome by addition of DHT, indicating that CYP17 inhibition and subsequent A reduction contribute to the intratumoral activity of AA. However, it appears that inhibition of AR output persists in the presence of DHT, implying that AA may exert additional anti-A effects in the prostate tumor cell. We are characterizing the LNCaP model further as a tool for rapid evaluation of: additional therapies targeting ARs, combinations of agents, and potential mechanisms of resistance to A-targeted therapies.
Abstract Number: LB-109
MDV3100, an androgen receptor signaling inhibitor, inhibits tumor growth in breast cancer preclinical models regardless of estrogen receptor status
Background: The androgen receptor (AR) is detected by immunohistochemistry in approximately 75% of all invasive breast cancer, with ~88% of estrogen receptor (ER) positive (ER+) tumors also expressing AR and ~20 to 30% of ER negative (ER-) tumors also retaining AR expression. Potent inhibition of AR activity could be a therapeutic strategy in AR+ ER+ breast cancer. MDV3100 is an androgen receptor signaling inhibitor (ARSI), which inhibits AR activity via three mechanisms: 1) inhibition of androgen binding to AR, 2) inhibition of AR nuclear translocation, and 3) inhibition of nuclear AR-DNA binding; and has demonstrated an overall survival benefit in men with post-docetaxel prostate cancer.
Methods: Two ER+/AR+ breast cancer cell lines, MCF7and BCK4 (recently derived from a pleurocentesis), were used to assess the proliferative effect of dihydrotestosterone (DHT) and estradiol (E2) in ovariectomized mice. MDV3100 was compared to bicalutamide and tamoxifen. Growth effects of MDV3100 on ER-/AR+ cells in tissue culture and in xenografts were also examined.
Results: Both bicalutamide and MDV3100 inhibited DHT-mediated proliferation of ER+/AR+ cell lines. Although MDV3100 binds AR very effectively and does not bind ER, it inhibited E2-mediated proliferation. MDV3100 also blocked E2 mediated upregulation of AR, PR, and SDF-1. MDV3100 inhibited E2-stimulated tumor growth of MCF7 mammary xenografts as effectively as tamoxifen. With ER-/AR+ cells (MDA-MB-453), MDV3100 reduced DHT-induced nuclear translocation, cell growth in tissue culture and tumor growth in mouse orthotopic xenografts.
Conclusions: MDV3100 blocked both DHT- and E2-mediated growth of breast cancer cells, whereas bicalutamide enhanced E2-mediated proliferation. MDV3100 may have unique therapeutic utility in patients with AR+ breast cancer, regardless of ER status.