Gnrh Antagonists Have Direct Inhibitory Effects on Castration-Resistant Prostate Cancer Via Intracrine Androgen and AR-V7 Expression Vito Cucchiara1, Joy C

Published OnlineFirst July 24, 2019; DOI: 10.1158/1535-7163.MCT-18-1337

Small Molecule Therapeutics Molecular Cancer Therapeutics GnRH Antagonists Have Direct Inhibitory Effects On Castration-Resistant Prostate Cancer Via Intracrine Androgen and AR-V7 Expression Vito Cucchiara1, Joy C. Yang1, Chengfei Liu1, Hans H. Adomat2, Emma S. Tomlinson Guns2, Martin E. Gleave2, Allen C. Gao1,3, and Christopher P. Evans1,3

Abstract

Hormone therapy is currently the mainstay in the manage- degarelix, leuprolide, or buffer alone for 4 weeks. Leuprolide ment of locally advanced and metastatic prostate cancer. slightly suppressed tumor growth compared with the vehicle Degarelix (Firmagon), a gonadotropin-releasing hormone control group (P > 0.05). Tumors in degarelix-treated mice (GnRH) receptor antagonist differs from luteinizing hor- were 67% of those in the leuprolide-treatment group but mone-releasing hormone (LHRH) agonists by avoiding "tes- 170% larger than in surgically castrated ones. Measurements tosterone flare" and lower follicle-stimulating hormone (FSH) of intratumoral steroids in serum, tumor samples, or treated levels. The direct effect of degarelix and leuprolide on human cell pellets by LC/MS confirmed that degarelix better decreased prostate cancer cells was evaluated. In LNCaP, C4-2BMDVR, the levels of testosterone and steroidogenesis pathway inter- and CWR22Rv1 cells, degarelix significantly reduced cell via- mediates, comparable to surgical castration, whereas leupro- bility compared with the controls (P 0.01). Leuprolide was lide had no inhibitory effect. Collectively, our results suggested stimulatory in the same cell lines. In C4-2B MDVR cells, a selective mechanism of action of degarelix against androgen degarelix alone or combined with abiraterone or enzaluta- steroidogenesis and AR-variants. This study provides addition- mide reduced the AR-V7 protein expression compared with al molecular insights regarding the mechanism of degarelix the control group. SCID mice bearing VCaP xenograft tumors compared with GnRH agonist therapy, which may have clin- were divided into 4 groups and treated with surgical castration, ical implications.

Introduction one surge") are the most commonly used agents. GnRH agonists, after a desensitization of the GnRH receptor response, determine a Prostate cancer is the most common tumor and the second reduction in luteinizing hormone (LH), follicle-stimulating hor- cause of cancer death in men (1). For advanced and metastatic mone (FSH), and testosterone production (2, 3). GnRH antago- prostate cancer, androgen deprivation therapy (ADT) using lutei- nists, like degarelix, are also an approved form of ADT (4). GnRH nizing hormone-releasing hormone (LHRH) agonists or the antagonists, by blocking GnRH receptors, produce a more rapid gonadotropin-releasing hormone (GnRH) receptor antagonist, suppression of testosterone without testosterone surge or micro- alone or in combination with radiotherapy, is considered the best surge (4). Presently, international prostate cancer guidelines treatment option (2, 3). A new generation of androgen receptor recommend the use of either GnRH agonists or antagonists as signaling inhibitors, such as abiraterone or enzalutamide, have treatment options for ADT in patients with prostate cancer (2, 3). been approved as treatment options in castration-resistant pros- Although some studies have suggested differences in efficacy and tate cancer (CRPC; refs. 2, 3). Different ADT strategies have been disease-related outcomes (musculoskeletal and urinary events) tested to achieve castration levels of testosterone (<50 ng/dL). To with degarelix compared with LHRH agonists (4), a systematic date, long-acting depot formulations of GnRH agonists (in com- review of the literature did not support the superiority of antago- bination with antiandrogens for 4 weeks to avoid the "testoster- nists over agonists (5). All the available phase III studies included in the analysis have treatment bias, short-term follow-up and 1Department of Urologic Surgery, University of California at Davis, Sacramento, heterogeneous populations (5). California. 2Vancouver Prostate Centre, University of British Columbia, GnRH agonists and antagonists both induce castrate levels of Vancouver, British Columbia, Canada. 3UC Davis Comprehensive Cancer Center, testosterone by altering the intracellular signaling of pituitary University of California at Davis, Sacramento, California. cells, but several attempts have been made to elucidate the Note: Supplementary data for this article are available at Molecular Cancer effect of these agents on other cells that express GNRH Therapeutics Online (http://mct.aacrjournals.org/). receptor (GnRH-R; refs. 6, 7). These studies revealed that extra- V. Cucchiara and J.C. Yang contributed equally to this article. pituitary tissues are affected by compounds directed toward GnRH-R (8–11). In prostate cells, GnRH-R manipulation may Corresponding Author: Christopher P. Evans, University of California at Davis, influence several biological processes such as cell growth, apo- 4860 Y St., Suite 3500, Sacramento, CA 95817. Phone: 916-734-7520; Fax: 916- – 734-8904; E-mail: [email protected] ptosis, angiogenesis, and cell adhesion (8 11). We hypothesized that, unlike agonists, GnRH antagonists may Mol Cancer Ther 2019;18:1811–21 have a direct mechanism of action on prostate cancer cells growth, doi: 10.1158/1535-7163.MCT-18-1337 by affecting the AR signaling pathway. Specifically, we investigat- 2019 American Association for Cancer Research. ed the role of GnRH agonists and antagonists in castration

www.aacrjournals.org 1811

Cucchiara et al.

sensitive and castration resistant prostate cancer cell lines and group was treated with 10 nmol/L of R1881, one cotransfected xenograft models and their possible interaction with AR and AR with an AR-V7 expression vector, the last with DMSO as control. splice variants (AR-Vs such as AR-V7). It is already know that All were treated with 10 to 20 mmol/L of leuprolide or degarelix. aberrant AR signaling and AR-Vs are able to promote the devel- Cells were harvested 48 hours after transfection and transactiva- opment of CRPC and may drive drug resistance (8–11). Preclin- tion was examined by the dual-luciferase assay (Promega). ical and clinical trials have described a direct correlation between AR-V7 expression, one of the most intensively studied AR-variant, Proliferation assay and resistance to enzalutamide and abiraterone. Moreover, AR-V7 LNCaP, C4-2B, CWR22Rv1, VCaP, and C4-2B MDVR cells were detection has been independently associated with poor prognosis seeded on 12-well plates at a density of 0.5 105 cells/well in in CRPC (12, 13). AR-V7 has been proposed as a prognostic media containing 10% FBS. When treated with leuprolide marker of PSA response, progression-free survival (PFS), and (20 mmol/L) and degarelix (20 mmol/L) alone or in combination overall survival (OS) among CRPC patients treated with AR- with enzalutamide (20 mmol/L) and abiraterone acetate targeted agents (abiraterone and enzalutamide) or chemothera- (5 mmol/L), cells were maintained in complete medium and py (14, 15), but there are no studies that have investigated the harvested after 3 or 6 days of treatment for cell counting. A total relationship between the type of ADT (GnRH agonist or antag- of 20 mmol/L of leuprolide and degarelix are equivalent to 24.2 onist) and the expression of AR-V7. and 32.6 mg/kg if using density of water 1 g/mL. In this study, we showed that different prostate cancer cell lines are sensitive to the antiproliferative effect of the GnRH antagonist Real-time qRT-PCR degarelix. Furthermore, the use of degarelix, alone or in combi- Total RNAs was extracted from LNCaP, C4-2B, CWR22Rv1, nation with enzalutamide or abiraterone, affected the expression VcaP, and C4-2B MDVR cells using the Qiagen Rneasy Kit. qPCR of AR-V7. In particular, we observed a reduction of AR-V7 at both analysis was performed in SsoFast EvaGreen Supermix (Bio-Rad) the protein and transcription levels. These insights suggest extra- with specific primers for AR-FL, AR-V7, or PSA and analyzed with pituitary activity of GnRH-R in prostate cancer tumors and may Bio-Rad CFX96 Real-Time PCR system (Bio-Rad). Each reaction have implications regarding resistance to second generation AR was normalized by co-amplification of b-action and triplicate pathway inhibitors. runs. The experiments were repeated 2 to 3 times for statistical analysis. Primers used for real-time-PCR were: AR-full length: 50-AAG Materials and Methods CCA GAG CTG TGC AGA TGA, 30-TGT CCT GCA GCC ACT GGT Reagents and cell culture TC; AR-V7: 50-AAC AGA AGT ACC TGT GCG CC, 30-TCA GGG TCT LNCaP, VCaP, and CWR22Rv1 cells were obtained from the GGT CAT TTT GA; actin: 50-AGA ACT GGC CCT TCT TGG AGG, 30- ATCC. All experiments with cell lines were performed within GTT TTT ATG TTC CTC TAT GGG; PSA: 5-GAT GAA ACA GGC TGT 6 months of receipt from ATCC or resuscitation after cryopreser- GCC G, 30-CCT CAC AGC TAC CCA CTG CA; GnRH-R subtype 1: vation. The cells were maintained in RPMI1640 supplemented 50-CAC CCT GAC ACG GGT CCT; 30-TTT ACT GGG TCT GAC AAC with 10% FBS, 100 units/mL penicillin, and 0.1 mg/mL strepto- C; GnRH-R subtype 2: 50-GTT TCT CTC CAG GCC ACC AT, 30-CAT mycin. VCaP cells were maintained in DMEM supplemented with CAG TGT CCG ACA TGC GA. 10% FBS, 100 units/mL penicillin, and 0.1 mg/mL streptomycin. C4-2B MDVR (C4-2B enzalutamide resistant; ref. 16) cells were In vivo tumorigenesis assay maintained in 20 mmol/L enzalutamide containing medium. All Animal studies were performed based on the protocols of the cells were maintained at 37C in a humidified incubator with 5% Institutional Animal Care and Use Committee of the University of carbon dioxide. California at Davis (Sacramento, CA). Male SCID mice were maintained in pressurized, ventilated cages with standard rodent Western blot analysis chow and water and a 12-hour light/dark cycle. VCaP cells Total protein was extracted from cultured cells and/or xenograft (2 106 cells) mixed with an equal amount of Matrigel (1:1) tumors and the concentrations were estimated using the BCA were injected subcutaneously into the flanks of 4-week-old male Protein Assay Reagent (Pierce). Equal amounts of denatured SCID mice. Tumor volumes were measured twice a week with protein samples were loaded on a 10% SDS-PAGE. After electro- calipers and tumor volumes were calculated according to the phoresis, proteins were transferred to Immobilon PVDF mem- following formula: 1/2 (length width2). After tumor size brane. Immunoblotting was done by incubating membranes with reached 100 mm3 mice were treated as follows: 3 mice: surgical the primary antibodies overnight at 4C with the indicated castration; 3 mice: vehicle control (5% Tween 80 and 5% ethanol primary antibodies GnRHR-2 (PA5-67876; Thermo Fisher Scien- in PBS), 9 mice: degarelix (35 mg/kg, s.c.), 9 mice: leuprolide tific), AR (N-20; SCBT), AR-V7 (Percision lab), b-actin (Sigma). (6 mg/kg, s.c.). Degarelix and leuprolide human loading doses are b-Actin was used as loading control. Following by 1 hour of 80 and 3.75 mg monthly, respectively, equivalent to 1.33 and secondary antibody incubation, immunoreactive proteins were 0.625 mg/kg for 60 kg adults. We used approximately 26 and visualized with SuperSignal West Pico CL (Pierce) coupled with 96 times higher doses of degarelix and leuprolide in mice, X-ray film exposure. respectively plus weekly dosing to compensate for the fast metabolic rate of small laboratory animals. Blood samples were Luciferase assay collected 10 days after castration and serum testosterone levels LNCaP C4-2B cells were seeded in 24-well plates in regular FBS were measured by an EIA Assay Kit (Cayman). Animals were medium or CSS medium, transfected with 0.2 mg of PSA-Luc sacrificed at the end of 6 weeks from tumor implant. Tumors were (promoter region, 63 0bp) with the internal control pTK-RL using collected for further analysis. IHC staining for Ki67 was per- Lipofectamine 2000 (Invitrogen). For those in CSS media, one formed to visualize the proliferative activity of tumors in each

1812 Mol Cancer Ther; 18(10) October 2019 Molecular Cancer Therapeutics

GnRH Antagonist Inhibits Intracrine Androgen and AR-V7

treatment group, using monoclonal antibody, anti-Ki-67 (SP6; induced AR transcriptional activation by 30%; however, none of 1;200 dilution; Thermo Fisher Scientiﬁc). the inhibition registered a signiﬁcant difference (Fig. 1E).

LC/MS-MS (17) Cell viability of prostate cancer cell lines treated with degarelix Tissue extracts were prepared by transferring a xenograft sample and leuprolide to a preweighed 2 mL screw cap vial containing 10 to 20 zirconia/ The direct effect of degarelix on human prostate cancer cell silica beads (2.3 mm; BioSpec) and tissue mass noted (20 to growth was evaluated. In LNCaP (androgen sensitive) and C4-2B 60 mg). Following addition of 100 mL water, homogenization was MDVR (enzalutamide resistant with induced AR-V7 through carried out with a Precellys homogenizer (4 cycles, 6,000 rpm, 20 long-term exposure) cells, degarelix significantly reduced the seconds each). Internal standard (IS, deuterated T & DHT) was cell viability assayed by WST-1 compared with the control group added and samples extracted twice by 30 minutes vortexing with (P 0.01) but not in CWR22Rv1 (castration resistant with 1 mL 60/40 hexane/ethyl acetate (hex/EtOAc). Extracted steroids truncated AR variants) cells after 6 days of treatment (Fig. 2A). were dried (CentriVap) and reconstituted in 50 mL of 50 mmol/L Conversely, leuprolide, an LHRH agonist, had a stimulatory hydroxylamine/50% methanol, incubated 1 hour at 60C and the effect in the same cell lines, significantly promoted cell growth resulting oximes analyzed using a Waters Aquity UPLC Separa- (P 0.01). tions Module coupled to a Waters Quattro Premier XE Mass Based on these data, the expression of AR, AR-V7, and PSA Spectrometer. Separations were carried out with a 2.1 100 mm transcripts in C4-2B MDVR cell line treated with leuprolide and BEH 1.7 mmol/L C18 column, mobile phase water (A) and 0.1% degarelix for 48 hours were evaluated by qRT-PCR (Fig. 2B). formic acid in acetonitrile (B; gradient: 0.2 minutes, 25% B; 8 Degarelix showed a significant reduction in the transcription minutes, 70% B; 9 minutes, 100% B; 12 minutes 100% B; 12.2 levels of AR-V7 and PSA (P 0.05). Western blot analysis of the minutes, 25% B; 14 minutes run length). All data were collected in same cell line (C4-2B MDVR), treated at 2 different concentra- ESþ by multireaction monitoring (mrm) with instrument para- tions (10 and 20 mmol/L) of leuprolide and degarelix, confirmed meters optimized for the m/z's and corresponding fragments of this finding that degarelix at 20 mmol/L reduced AR-V7 at protein the oxime-steroids. Data processing were done with Quanlynx level compared with the control and leuprolide groups (Fig. 2C). (Waters) and exported to Excel for additional normalization to When same treatments were applied to Rv1 and VCaP cells, both weights and volumes as required. Serum extracts (50 mL) were harboring high levels of AR variants, AR-V7 level also reduced extracted similar to above, omitting homogenizing and with a comparable to that in C4-2B MDVR cells. And yet, leuprolide single 1.5 mL volume of hex/EtOAc. Tissue culture samples were either enhanced AR-V7 marginally in Rv1 cells or maintained it prepared similar to tissues except that a slurry was generated from around the same in VCaP cells (Supplementary Fig. S1A and S1B). pellets by freeze/thaw cycles (3) and vortexing prior to transfer of 100 mL to extraction tubes. Leuprolide and degarelix combined with AR pathway inhibitor therapies Statistical analysis We then examined the benefit of combinations of ARSI with Data are shown as the mean SD. All were from at least 3 these GnRH modulators in growth inhibition of enzalutamide independent experiments and subjected to unpaired Student t resistant C4-2B MDVR and CWR22Rv1 cells with the trypan-blue tests and 1-way ANOVA for comparison of means. P 0.05 was cell counting method. With cell proliferation not affected by considered statistically significant. enzalutamide, 20 mmol/L of leuprolide alone or in combination did not affect the cell growth to a statistical significance. However, 20 mmol/L of degarelix alone significantly decreased proliferation Results by 75% and slightly further reduced viable cell counts when Expression of GnRH receptors and androgen receptors in combined with enzalutamide (Fig. 2D). The MDVR cells were prostate cancer cell lines cross-resistant to abiraterone. Nevertheless, degarelix combined Considering the mechanism of action of GnRHR agonist and with abiraterone showed benefit with significant inhibition antagonist in vitro, we measured the expression of GNRHR type 1 compared with control in both MDVR and Rv1 cells. Western and 2, AR full length (AR-FL), and its variant V7 (AR-V7) by blot analysis of treated C4-2B MDVR lysates (Fig. 2E) revealed qRT-PCR. GnRH-R type 1 mRNA levels were undetectable in 5 that, although the protein levels of AR-FL remained similar, prostate cancer cell lines (LNCaP, C4-2B, C4-2B MDVR, VCaP, AR-V7 levels were significantly reduced in degarelix-treated and CWR22Rv1). On the contrary, GnRH-R type 2 mRNAs were samples, alone (40% of the control) or in combinations with present in the cell lines investigated at different expression levels enzalutamide or abiraterone (26% of control). Conversely, and compared with LNCaP as the reference (Fig. 1A). Western leuprolide slightly increases AR-V7 protein level when used blotting analysis showed the presence of GnRH-R2 in all alone, but significantly when combined with enzalutamide or AR-positive cells and even in PC-3 cells (Fig. 1B). We focused on abiraterone. the 4 AR bearing PC lines and measured their levels of AR-FL and To scrutinize whether this reduction of AR-V7 by degarelix AR-V7 side-by-side to establish the comparison and predict their treatments was via downregulation of its transcript or protein response to GnRHR modulators. It was validated that only VCaP degradation, we performed qRT-PCR assays of C4-2B MDVR cells and CWR22Rv1 cells have readily detectable AR-V7 (Fig. 1C treated with leuprolide, degarelix, and enzalutamide, alone or and D). We then used C4-2B cells transiently transfected with in combination (Fig. 2F and G). Indeed, quantification of the the PSA-Luc plasmid to examine the action of GnRHR modula- mRNA levels of AF-FL, AR-V7, and PSA (Fig. 2G) further tors. Luciferase activity stimulated by synthetic androgen R1881 confirms that degarelix targets AR-V7 specifically at the transcript was slightly decreased by both leuprolide and degarelix, but much level whereas leuprolide shows stimulation when combined with less than by enzalutamide. These 2 reagents also inhibited AR-V7 enzalutamide.

www.aacrjournals.org Mol Cancer Ther; 18(10) October 2019 1813

Cucchiara et al.

Figure 1. Survey of the levels of GnRHR type 2, AR-FL, and AR-V7 molecules in prostate cancer lines and response to leuprolide and degarelix in AR transactivation activity. A, Comparison of GnRHR2 messages in various prostate cancer lines. B, Western blot analysis of GnRHR2 in prostate cancer lines. C, Quantitation of AR-FL and AR-V7 in prostate cancer lines. D, Western blot analysis of AR-V7 and AR-FL in prostate cancer lines. E, PSA-Luc assays of C4-2B cells in CS medium, stimulated by R1881 or AR-V7 in response to leuprolide (leup) or degarelix (deg) treatments. Error bars represent SE.

In vitro analysis of androgens and androgen precursors levels in respectively). The same result was observed for the other intra- prostate cancer cell lines cellular androgen, DHEA. As for DHT, the effect of degarelix was In parallel to the in vivo study, we treated 2 CRPC (C4-2B and only observed in VCaP cells (P 0.05). VCaP) lines with leuprolide and degarelix in androgen-deprived conditions to investigate the concentrations of androgens and Inhibition of growth of VCaP xenografts by GnRH agonist and androgen precursors under GnRH agonist and antagonist treat- antagonists ments. We measured steroids in these cell pellets using a validated The common concept how GnRHR agonist or antagonist works LC/MS-MS assay (Fig. 3). When maintained in CS medium, the is through regulating the pituitary gland of males to suppress basal level of testosterone detected in these cells was very low. testicular androgen synthesis. Although we observed direct effect Degarelix further decreases testosterone signiﬁcantly in both lines of degarelix on PC cells in vitro, we sought to test both drugs in vivo whereas leuprolide displays no difference (P 0.01 and 0.05, to demonstrate potential differences in treating prostate tumors. It

1814 Mol Cancer Ther; 18(10) October 2019 Molecular Cancer Therapeutics

GnRH Antagonist Inhibits Intracrine Androgen and AR-V7

Figure 2. A, Viable counts of LNCaP, C4-2B MDVR, and CWR22Rv1 cells after 6 days of treatments with 20 mmol/L of leup or deg. B, Levels of AR-FL, AR-V7, and AR reporter gene PSA after 6 days of treatments with leuprolide and degarelix. C, Western blot analyses of AR-FL and AR-V7 of C4-2B MDVR cells treated with leuprolide or degarelix. D, Cell counts of C4-2B MDVR and CWR22Rv1 cells after 6 days of treatments with leup, deg, AA, enza alone or in combinations with AA or enza. E, Western blot analyses of AR-V7 and AR-FL in C4-2B MDVR cells treated with single or combined agents. F and G, Quantitation of AR-FL, AR-V7, and PSA in C4-2B MDVR cells with respective treatments. Error bars represent SE. , P 0.05; , P 0.01. has been demonstrated that both GnRHR agonist and antagonist tumor growth compared with the vehicle control group (P > are functional in mouse studies (18, 19). We chose VCaP cells 0.05). However, tumors in the degarelix-treated group were half because of their CRPC characteristics and yet still demonstrating a the size of those from control group (P 0.05) and 67% of those response to castration. SCID mice bearing VCaP xenograft in leuprolide group (Fig. 4B). There were no differences in mouse tumors (average volume of 136 mm3) were treated with degarelix body weights among the treatment groups. After 4 weeks of (35 mg/Kg weekly), leuprolide (6 mg/kg weekly), surgical treatment, we measured testosterone levels in the mouse xeno- castration, or vehicle for 4 weeks (Fig. 4A). Surgical castration grafts. Because of a considerable variability especially in the effectively controlled the tumor growth, with slight relapse control group, testosterone levels did not show any signiﬁcant 2 weeks after the surgery. Leuprolide only slightly suppressed differences among degarelix, leuprolide, and surgical castration

www.aacrjournals.org Mol Cancer Ther; 18(10) October 2019 1815

Cucchiara et al.

Figure 3. LC/MS measurements of steroids in C4-2B and VCaP cells treated with respective single or double agents in CS medium for 3 to 6 days. Representative data of testosterone, DHT, and DHEA were shown. Error bars represent SE. , P 0.05; , P 0.01.

groups (Fig. 4C). All 3 treatments effectively decreased serum disease, they ultimately decrease serum testosterone levels to testosterone levels. Analysis of tumor lysates from representative castrate and are considered a standard of care (20). GnRH tumor samples from each group showed that AR-V7 was below antagonists result in castrate testosterone levels without testos- detection in control and with light exposure in degarelix but terone flare (21). Arguably, some data have suggested degarelix to strongly enhanced in both castration and Leuprolide groups have better disease control and, in particular, a superior PSA (Fig. 4D). A comprehensive blot with more tumors samples under progression-free survival (22) and a more favorable control of longer exposure was shown in Supplementary Fig. S2. This is in serum alkaline phosphatase (23). A recent pooled analysis of data agreement with the in vitro studies that degarelix alone down- from 5 randomized trials of degarelix versus GnRH agonists regulates AR-V7. To estimate growth fraction, IHC staining for showed higher overall survival during the first year of treatment anti-Ki-67 antibody was used to determine the percentage of Ki- for men receiving degarelix (4). Data also showed that, in patients 67-positive tumor cells on slide-mounted, paraffin-embedded with a history of cardiovascular disease, there was a significantly tumor sections (Fig. 4Eand F). Surgical castration and degarelix lower risk (>50%) of a subsequent cardiovascular event or death groups were associated with significantly lower Ki-67 positive over 1 year of treatment with degarelix versus GnRH agonists (24). cells compared with the leuprolide or control groups (P 0.05), Despite these data, a recent metanalysis did not find any supe- demonstrating a greater reduction of cell proliferation after sur- riority of a drug over another. One important factor in these trials gical castration or degarelix treatment. is the short follow up period (12-months; ref. 25). OurstudysoughttoidentifyanydirecteffectofGnRH Tissue androgen and androgen precursor analysis antagonists on prostate cancer tumors. Although it has been We assessed whether mean intratumoral testosterone androgen previously described that prostate and other periferical tissues precursors levels differed among the four treatment arms in SCID express GnRH-R (26, 27), we needed to confirm these results mice bearing VCaP xenograft tumors. There are significant differ- to investigate the direct effects of degarelix on prostate cancer ences in tumor testosterone levels between control and surgical cell lines. We corroborated that all the human prostate cancer castration (6.345 vs. 0.189 ng/g tumor, P 0.001) and control cell lines express GnRH receptor type II, but not GnRH receptor and degarelix groups (6.345 vs. 1.059 ng/g tumor, P 0.001), as type I. well as degarelix versus Leuprolide (1.059 vs. 6.812 ng/g tumor, Finding GnRH-R in LNCaP, VCaP, CWR22-Rv1, C4-2B, C4-2B P 0.001; Fig. 5). Most of other intermediates in steroidogenesis MDVR cells led to their use in our studies. Indeed, prostate cancer followed the same pattern as testosterone with significant differ- cells respond to both agonist and antagonist in cell proliferation ence between control and castration, and control and degarelix but ambivalently. The stimulatory effect of leuprolide on cell groups. In serum samples, testosterone levels are much higher in growth is not predicted. Upon binding to GnRH-R on cell surface, the leuprolide group where degarelix treatment provides com- leuprolide might turn on some unknown signaling pathway to parative reduction as surgical castration (Fig. 6). Levels of 3 produce a short-term enhancement. Inhibition of tumor growth intermediates, androstenedione, pregnan-3,20-dione and 5preg- by leuprolide was moderate but more in agreement of its efficacy. nan-3-ol-20-one also reflect that of testosterone in response to Both leuprolide and degarelix affect AR transactivation activity. treatments with significant changes by castration and degarelix. AR-driven PSA-Luciferase activity changes upon leuprolide and degarelix treatments when induced by exogenous AR-V7 in C4-2B cells. The magnitude of change did not register statistical signif- Discussion icance probably due to the overexpression of AR-V7 through ADT is a treatment option for locally-advanced and metastatic transfection. prostate cancer (2, 3). Although GnRH agonists are associated Cell proliferation assays showed a lower cell viability after with an initial testosterone surge in patients with metastatic degarelix exposure compared with leuprolide treatment. This

1816 Mol Cancer Ther; 18(10) October 2019 Molecular Cancer Therapeutics

GnRH Antagonist Inhibits Intracrine Androgen and AR-V7

Figure 4. In vivo study with the VCaP xenograft model. A, Tumor progression shown by tumor volumes over time under control, castration, leuprolide, or degarelix treatments. B, Comparison of tumor volume from treatment groups at the termination of study. C, Testosterone levels by EIA after 4 weeks of treatment. D, Western blot analyses of AR-V7 in tumor lysates from representative groups. E, Representative Ki67 IHC staining micrograms and counts of Ki67-positive cells from respective treatment groups. F, Quantitation of Ki67-positive cells from all 4 treatment groups. Error bars represent SE. , P 0.05. suggests a direct antitumorigenic role of degarelix on prostate proinﬂammatory cytokines (28, 29). In our experiments, we did cancer cell growth and it is consistent with the previous report that not ﬁnd that the GnRH agonist leuprolide inhibits prostate cells degarelix determines the activation of caspase 3/7 in prostate viability, even when tested at higher concentrations. cancer cell lines (6). Moreover, another GnRH antagonist, cetror- We have repeated observed the effect of degarelix on the level of elix, revealed a direct inhibitory effect on prostate cell line growth AR-V7, both at mRNA and protein levels. The occurrence of AR with a mechanism involving cell-cycle arrest and a change in variants is mostly through gene rearrangement and RNA

www.aacrjournals.org Mol Cancer Ther; 18(10) October 2019 1817

Cucchiara et al.

Figure 5. LC/MS measurements of steroids: testosterone, DHEA, pregnenolone, androsterone, androstenedione, and 5-pregnan-3-ol-20-one from tumors in all 4 treatment groups. Error bars represent SE. , P 0.05; , P 0.001.

splicing (30–32). ADT with prolonged GnRH-R agonist or antag- studies are consistent with the trend in proliferation assays. onist induces the emergence of AR-V7 (31). Indeed, both cell and Degarelix alone signiﬁcantly reduced the intracellular levels of tumor lysates from the leuprolide groups reveal upregulation of testosterone compared with leuprolide. Degarelix also reduces the AR-V7, mimicking what have been observed in some patients levels of most of the intermediates in steroidogenesis, comparable with CRPC after long-term ADT (12, 33, 34). However, in our to those detected in the surgical castration group. The canonical hands, both in vitro and in vivo treatments of degarelix reduced effect of degarelix on testosterone is through pituitary cells. AR-V7. It is not clear whether degarelix has any effect on Although it may be the rationale how tumor inhibition by molecules regulating splicing such as U2AF65, ASF/SF2, degarelix was accomplished in the xenograft model, there is no JMJD1A, HoxB13, or hnRNPA1 (31, 35–37). With its higher collaboration of pituitary tissue in tissue ﬂasks. Inhibition of impact on the protein level, it is possible that degarelix causes intracrine steroidogenesis via the backdoor pathway is a plausible some protein degradation through the ubiquitin proteasome explanation. It is not known whether degarelix imposes any system (38, 39). The exact mechanisms behind this down- inhibitory effect on the steroid biosynthetic enzymes such as regulation remain to be investigated. CYP17A1, HSD3B2/3, AKR1C1/2/3, etc. We observed an As intratumoral androgens are known to drive CRPC (40, 41), additive effect on cell proliferation and reduction in AR-V7 we measured androgen levels using LC/MS-MS (17) in prostate when using abiraterone that targets CYP17A1 in combination cancer cell lines and xenograft tumor samples. Overall, the mea- with degarelix. Therefore, CYP17A1 is unlikely the target. surements using LC/MS-MS from both the in vitro and in vivo Knockdown of AKR1C3 with shRNA restores sensitivity to

1818 Mol Cancer Ther; 18(10) October 2019 Molecular Cancer Therapeutics

GnRH Antagonist Inhibits Intracrine Androgen and AR-V7

Figure 6. LC/MS measurements of steroids: testosterone, androsterone, androstenedione, progestorone, pregnan-3,20-dione, and 5-pregnan-3-ol-20-one from blood samples collected from all 4 treatment groups. Error bars represent SE. , P 0.05; , P 0.01; , P 0.005. enzalutamide in C4-2B MDVR cells (16). Inhibition of AKR1C3 Disclosure of Potential Conﬂicts of Interest with shRNA or the small molecule inhibitor directly down- C.P. Evans reports receiving a commercial research grant from Ferring regulates AR-V7. Whether degarelix targets intracrine androgen Pharmaceuticals. J.C. Yang reports receiving a commercial research grant from ﬂ synthesis and AR-V7 directly through AKR1C3 remains to be Ferring Pharmaceuticals. No potential con icts of interest were disclosed by the other authors. explored. Above all, this study provides insight regarding a direct Authors' Contributions anti-prostate cancer tumor effect of degarelix. In our hands, Conception and design: V. Cucchiara, C. Liu, C.P. Evans the GnRHR antagonist degarelix inhibits CRPC cell growth Development of methodology: V. Cucchiara, J.C. Yang, C. Liu, H.H. Adomat, in vitro and tumor progression in vivo possibly through down- E.S.T. Guns, C.P. Evans regulation AR-V7, superior to the receptor agonist. It may Acquisition of data (provided animals, acquired and managed patients, suggest a biological rationale to consider antagonists rather provided facilities, etc.): V. Cucchiara, J.C. Yang, C. Liu, H.H. Adomat, than agonists especially in combination with ARSI drugs to E.S.T. Guns, M.E. Gleave prolong the onset of CRPC. However, this is hypothesis Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): V. Cucchiara, J.C. Yang, C. Liu, H.H. Adomat, generating and demands further investigation into the interplay M.E. Gleave, C.P. Evans between androgen deprivation therapies and resistance Writing, review, and/or revision of the manuscript: V. Cucchiara, J.C. Yang, mechanisms in CRPC. C. Liu, H.H. Adomat, M.E. Gleave, A.C. Gao, C.P. Evans

www.aacrjournals.org Mol Cancer Ther; 18(10) October 2019 1819

Cucchiara et al.

Administrative, technical, or material support (i.e., reporting or organizing The costs of publication of this article were defrayed in part by the data, constructing databases): J.C. Yang, C. Liu, A.C. Gao payment of page charges. This article must therefore be hereby marked Study supervision: J.C. Yang, C. Liu, C.P. Evans advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Acknowledgments This work was supported in part by the grant DOD PC150040P1 and Ferring Received December 12, 2018; revised March 28, 2019; accepted July 23, 2019; Pharmaceuticals to C.P. Evans. published ﬁrst July 24, 2019.

References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin 18. de Pinieux G, Legrier ME, Poirson-Bichat F, Courty Y, Bras-Goncalves R, 2016;66:7–30. Dutrillaux AM, et al. Clinical and experimental progression of a new model 2. Cornford P, Bellmunt J, Bolla M, Briers E, De Santis M, Gross T, et al. EAU- of human prostate cancer and therapeutic approach. Am J Pathol 2001; ESTRO-SIOG guidelines on prostate cancer. part II: treatment of relapsing, 159:753–64. metastatic, and castration-resistant prostate cancer. Eur Urol 2017;71: 19. Ogawa T, Dobrinski I, Avarbock MR, Brinster RL. Leuprolide, a gonado- 630–42. tropin-releasing hormone agonist, enhances colonization after spermato- 3. Mohler JL, Armstrong AJ, Bahnson RR, D'Amico AV, Davis BJ, Eastham JA, gonial transplantation into mouse testes. Tissue Cell 1998;30:583–8. et al. Prostate Cancer, Version 1.2016. J Natl Compr Canc Netw 2016;14: 20. Thompson IM. Flare associated with LHRH-agonist therapy. Rev Urol 19–30. 2001;3:S10–4. 4. Klotz L, Miller K, Crawford ED, Shore N, Tombal B, Karup C, et al. Disease 21. Klotz L, Boccon-Gibod L, Shore ND, Andreou C, Persson BE, Cantor P, et al. control outcomes from analysis of pooled individual patient data from five The efficacy and safety of degarelix: a 12-month, comparative, randomized, comparative randomised clinical trials of degarelix versus luteinising open-label, parallel-group phase III study in patients with prostate cancer. hormone-releasing hormone agonists. Eur Urol 2014;66:1101–8. BJU Int 2008;102:1531–8. 5. Salciccia S, Gentilucci A, Cattarino S, Sciarra A. GNRH-agonist or antag- 22. Tombal B, Miller K, Boccon-Gibod L, Schroder F, Shore N, Crawford ED, onist in the treatment of prostate cancer: a comparision based on onco- et al. Additional analysis of the secondary end point of biochemical logical results. Urologia 2016;83:173–8. recurrence rate in a phase 3 trial (CS21) comparing degarelix 80 mg versus 6. Sakai M, Martinez-Arguelles DB, Patterson NH, Chaurand P, Papadopou- leuprolide in prostate cancer patients segmented by baseline character- los V. In search of the molecular mechanisms mediating the inhibitory istics. Eur Urol 2010;57:836–42. effect of the GnRH antagonist degarelix on human prostate cell growth. 23. Schroder FH, Tombal B, Miller K, Boccon-Gibod L, Shore ND, Crawford PLoS One 2015;10:e0120670. ED, et al. Changes in alkaline phosphatase levels in patients with prostate 7. Dondi D, Moretti RM, Montagnani Marelli M, Pratesi G, Polizzi D, Milani cancer receiving degarelix or leuprolide: results from a 12-month, com- M, et al. Growth-inhibitory effects of luteinizing hormone-releasing hor- parative, phase III study. BJU Int 2010;106:182–7. mone (LHRH) agonists on xenografts of the DU 145 human androgen- 24. Albertsen PC, Klotz L, Tombal B, Grady J, Olesen TK, Nilsson J. Cardio- independent prostate cancer cell line in nude mice. Int J Cancer 1998;76: vascular morbidity associated with gonadotropin releasing hormone ago- 506–11. nists and an antagonist. Eur Urol 2014;65:565–73. 8. Park MK, Kanaho Y, Enomoto M. Regulation of the cell proliferation and 25. Sciarra A, Fasulo A, Ciardi A, Petrangeli E, Gentilucci A, Maggi M, et al. A migration as extra-pituitary functions of GnRH. Gen Comp Endocrinol meta-analysis and systematic review of randomized controlled trials with 2013;181:259–64. degarelix versus gonadotropin-releasing hormone agonists for advanced 9. Chatzaki E, Bax CM, Eidne KA, Anderson L, Grudzinskas JG, Gallagher CJ. prostate cancer. Medicine (Baltimore) 2016;95:e3845. The expression of gonadotropin-releasing hormone and its receptor in 26. Limonta P, Manea M.Gonadotropin-releasing hormone receptors as endometrial cancer, and its relevance as an autocrine growth factor. molecular therapeutic targets in prostate cancer: current options and Cancer Res 1996;56:2059–65. emerging strategies. Cancer Treat Rev 2013;39:647–63. 10. McArdle CA, Franklin J, Green L, Hislop JN. The gonadotrophin-releasing 27. Rozsa B, Nadji M, Schally AV, Dezso B, Flasko T, Toth G, et al. Receptors for hormone receptor: signalling, cycling and desensitisation. Arch Physiol luteinizing hormone-releasing hormone (LHRH) in benign prostatic Biochem 2002;110:113–22. hyperplasia (BPH) as potential molecular targets for therapy with LHRH 11. Kraus S, Naor Z, Seger R. Intracellular signaling pathways mediated by the antagonist cetrorelix. Prostate 2011;71:445–52. gonadotropin-releasing hormone (GnRH) receptor. Arch Med Res 2001; 28. Rick FG, Schally AV, Block NL, Halmos G, Perez R, Fernandez JB, et al. 32:499–509. LHRH antagonist Cetrorelix reduces prostate size and gene expression of 12. Antonarakis ES, Lu C, Wang H, Luber B, Nakazawa M, Roeser JC, et al. AR- proinflammatory cytokines and growth factors in a rat model of benign V7 and resistance to enzalutamide and abiraterone in prostate cancer. prostatic hyperplasia. Prostate 2011;71:736–47. N Engl J Med 2014;371:1028–38. 29. Yang D, Hou T, Yang X, Ma Y, Wang L, Li B. Mechanisms of prostate atrophy 13. Welti J, Rodrigues DN, Sharp A, Sun S, Lorente D, Riisnaes R, et al. after LHRH antagonist cetrorelix injection: an experimental study in a rat Analytical validation and clinical qualification of a new immunohisto- model of benign prostatic hyperplasia. J Huazhong Univ Sci Technolog chemical assay for androgen receptor splice variant-7 protein expression in Med Sci 2012;32:389–95. metastatic castration-resistant prostate cancer. Eur Urol 2016;70:599–608. 30. Li Y, Alsagabi M, Fan D, Bova GS, Tewfik AH, Dehm SM. Intragenic 14. Antonarakis ES, Lu C, Luber B, Wang H, Chen Y, Nakazawa M, et al. rearrangement and altered RNA splicing of the androgen receptor in a Androgen receptor splice variant 7 and efficacy of taxane chemotherapy in cell-based model of prostate cancer progression. Cancer Res 2011;71: patients with metastatic castration-resistant prostate cancer. JAMA Oncol 2108–17. 2015;1:582–91. 31. Liu LL, Xie N, Sun S, Plymate S, Mostaghel E, Dong X. Mechanisms of the 15. Onstenk W, Sieuwerts AM, Kraan J, Van M, Nieuweboer AJ, Mathijssen RH, androgen receptor splicing in prostate cancer cells. Oncogene 2014;33: et al. Efficacy of cabazitaxel in castration-resistant prostate cancer is 3140–50. independent of the presence of AR-V7 in circulating tumor cells. 32. Nyquist MD, Li Y, Hwang TH, Manlove LS, Vessella RL, Silverstein KA, Eur Urol 2015;68:939–45. et al. TALEN-engineered AR gene rearrangements reveal endocrine 16. Liu C, Lou W, Zhu Y, Yang JC, Nadiminty N, Gaikwad NW, et al. Intracrine uncoupling of androgen receptor in prostate cancer. Proc Natl Acad Androgens and AKR1C3 activation confer resistance to enzalutamide in Sci U S A 2013;110:17492–7. prostate cancer. Cancer Res 2015;75:1413–22. 33. Dehm SM, Schmidt LJ, Heemers HV, Vessella RL, Tindall DJ. Splicing of a 17. Adomat HH, Bains OS, Lubieniecka JM, Gleave ME, Guns ES, Grigliatti TA, novel androgen receptor exon generates a constitutively active androgen et al. Validation of a sequential extraction and liquid chromatography- receptor that mediates prostate cancer therapy resistance. Cancer Res 2008; tandem mass spectrometric method for determination of dihydrotestos- 68:5469–77. terone, androstanediol and androstanediol-glucuronide in prostate tis- 34. Hu R, Dunn TA, Wei S, Isharwal S, Veltri RW, Humphreys E, et al. sues. J Chromatogr B Analyt Technol Biomed Life Sci 2012;902:84–95. Ligand-independent androgen receptor variants derived from splicing

1820 Mol Cancer Ther; 18(10) October 2019 Molecular Cancer Therapeutics

GnRH Antagonist Inhibits Intracrine Androgen and AR-V7

of cryptic exons signify hormone-refractory prostate cancer. Cancer Res receptor targeted therapy in advanced prostate cancer. Nat Commun 2009;69:16–22. 2018;9:4700. 35. Chen Z, Wu D, Thomas-Ahner JM, Lu C, Zhao P, Zhang Q, et al. Diverse 39. Moses MA, Kim YS, Rivera-Marquez GM, Oshima N, Watson MJ, Beebe KE, AR-V7 cistromes in castration-resistant prostate cancer are governed by et al. Targeting the Hsp40/Hsp70 chaperone axis as a novel strategy to treat HoxB13. Proc Natl Acad Sci U S A 2018;115:6810–5. castration-resistant prostate cancer. Cancer Res 2018;78:4022–35. 36. Fan L, Zhang F, Xu S, Cui X, Hussain A, Fazli L, et al. Histone demethylase 40. Locke JA, Guns ES, Lubik AA, Adomat HH, Hendy SC, Wood CA, et al. JMJD1A promotes alternative splicing of AR variant 7 (AR-V7) in prostate Androgen levels increase by intratumoral de novo steroidogenesis during cancer cells. Proc Natl Acad Sci U S A 2018;115:E4584–E93. progression of castration-resistant prostate cancer. Cancer Res 2008;68: 37. Tummala R, Lou W, Gao AC, Nadiminty N. Quercetin Targets hnRNPA1 to 6407–15. overcome enzalutamide resistance in prostate cancer cells. Mol Cancer 41. Mostaghel EA, Marck BT, Plymate SR, Vessella RL, Balk S, Matsumoto AM, Ther 2017;16:2770–9. et al. Resistance to CYP17A1 inhibition with abiraterone in castration- 38.LiuC,LouW,YangJC,LiuL,ArmstrongCM,LombardAP,etal. resistant prostate cancer: induction of steroidogenesis and androgen Proteostasis by STUB1/HSP70 complex controls sensitivity to androgen receptor splice variants. Clin Cancer Res 2011;17:5913–25.

www.aacrjournals.org Mol Cancer Ther; 18(10) October 2019 1821

GnRH Antagonists Have Direct Inhibitory Effects On Castration-Resistant Prostate Cancer Via Intracrine Androgen and AR-V7 Expression

Vito Cucchiara, Joy C. Yang, Chengfei Liu, et al.

Mol Cancer Ther 2019;18:1811-1821. Published OnlineFirst July 24, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-18-1337

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2019/07/24/1535-7163.MCT-18-1337.DC1

Cited articles This article cites 41 articles, 12 of which you can access for free at: http://mct.aacrjournals.org/content/18/10/1811.full#ref-list-1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/18/10/1811. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.