Prostate cancer cells move from their primary site of origin, interact with a distant microenvironment, grow, and thereby cause death. It had heretofore not been possible to selectively inhibit cancer cell motility. Our group has recently shown that inhibition of intracellular activation of Raf1 with the small-molecule therapeutic KBU2046 permits, for the first time, selective inhibition of cell motility. We hypothesized that simultaneous disruption of multiple distinct functions that drive progression of prostate cancer to induce death would result in advanced disease control. Using a murine orthotopic implantation model of human prostate cancer metastasis, we demonstrate that combined treatment with KBU2046 and docetaxel retains docetaxel's antitumor action, but provides improved inhibition of metastasis, compared with monotherapy. KBU2046 does not interfere with hormone therapy, inclusive of enzalutamide-mediated inhibition of androgen receptor (AR) function and cell growth inhibition, and inclusive of the ability of castration to inhibit LNCaP-AR cell outgrowth in mice. Cell movement is necessary for osteoclast-mediated bone degradation. KBU2046 inhibits Raf1 and its downstream activation of MEK1/2 and ERK1/2 in osteoclasts, inhibiting cytoskeleton rearrangement, resorptive cavity formation, and bone destruction in vitro, with improved effects observed when the bone microenvironment is chemically modified by pretreatment with zoledronic acid. Using a murine cardiac injection model of human prostate cancer bone destruction quantified by CT, KBU2046 plus zoledronic exhibit improved inhibitory efficacy, compared with monotherapy. The combined disruption of pathways that drive cell movement, interaction with bone, and growth constitutes a multifunctional targeting strategy that provides advanced disease control.

This article is featured in Highlights of This Issue, p. 1

The abnormal movement of cancer cells throughout the body, their ability to interact with distant organ sites, and to grow constitute separate cellular functions whose integration results in end-organ destruction and ultimately death. The underlying biological processes that drive these functions are collectively known as the hallmarks of cancer (1). Hallmarks related to growth and survival are proximally responsible for the development and outgrowth of primary and metastatic tumors, while the hallmark termed invasion and metastasis drives the movement of cancer throughout the body and mediates its interaction with distant vital organs. Increased cell motility and resultant metastasis is the process by which the vast majority of solid organ cancers cause death (2, 3). Considering all forms of cancer, metastasis is responsible for up to 90% of mortality (4), and for prostate cancer it is responsible for essentially 100% (5, 6).

An approach that coordinately engages targets of pathways that drive distinct functions that act together to sustain cancer at the systemic level represents a rational therapeutic strategy. Cell motility drives invasion and metastasis (7). Historically, there has been an inability to selectively target processes related to cell motility (8). A central aspect to this longstanding roadblock relates to the fact that a wide array of gene products have been shown to regulate cell motility, but essentially all lack specificity (8). Recent advances by our group have identified a novel regulatory mechanism, demonstrating that inhibition of Raf1 activation inhibits cell motility and resultant metastasis across several comprehensive and complementary in vitro and in vivo models (9). These advances by us were achieved by our designed synthesis of the small-molecule therapeutic, KBU2046, that operates through a heretofore unknown mechanism ultimately inhibiting activation of Raf1, as previously described by us (9). In brief, it binds in a cleft formed when HSP90β interfaces with its co-chaperone CDC37. It thereby acts as a selective HSP90 activity modulator (SHAM) agent, selectively modulating the activity of client proteins that regulate cell motility, with inhibition of Raf1 activation being of central importance. Classical HSP90 inhibitors bind to HSP90, inhibit its chaperone cycle, inclusive of ATP hydrolysis, and broadly disrupt client protein function resulting in protein degradation. KBU2046 is distinctly different: it does not bind HSP90 in isolation, binds to a cleft formed at the interface of where HSP90 binds to the co-chaperone CDC37, and thereby will only bind to the HSP90/CDC37 heterocomplex. This results in precision modulation of chaperone function, resulting in decreased binding of Raf1 to the HSP90/CDC37 heterocomplex, decreased phosphorylation of ser338 on Raf1′s activation motif, and decreased Raf1 kinase activity. Furthermore, Raf1 was shown to modulate cell motility, and to mediate the pharmacologic action of KBU2046 on inhibition of cell motility. These advances now permit selective inhibition of cell motility via a Raf1-targeting strategy.

We have a long standing ability to target cancer cell growth and viability through a relatively diverse array of targeted strategies (10). Conversely, our ability to manipulate the interaction between cancer and distant organ sites, while feasible, has in general been less fruitful. Specifically, in the case of bone, bisphosphonates have been shown to physically bind bone mineral (i.e., calcium–phosphate, in the form of hydroxyapatite), to modify its structure and to inhibit its destruction by metastatic cancer (11). However, bisphosphonates are not considered highly effective, and their use is not without toxicity. Consider prostate cancer, bone metastases are present in over 90% of those with metastatic disease, and in the majority of cases, bone is the only site of metastasis (2). Bone metastases dominates the clinical management of these patients, causing pain, bone fracture, spinal cord and nerve root compression, bone marrow replacement, hypercalcemia, and they serve as a major reservoir of tumor burden. Together these adverse clinical consequences of bone metastasis constitute major contributors to morbidity and mortality (12). Against this backdrop of bone destruction, the bisphosphonate, zoledronic acid (ZA), has been shown to decrease adverse events due to bone destruction in men with metastatic prostate cancer, and its administration is considered standard practice (11). However, ZA only delays bone destruction, and the absolute difference in incidence compared with controls is only 11%.

We hypothesized that simultaneous disruption of pathways driving distinct functions that sustain cancer systemically would result in advanced disease control. In this study, we demonstrate that this represents a viable approach that should be investigated further. Specifically, studies demonstrate the increased control of systemic disease when Raf1 and microtubules are targeted, providing for combined disruption of motility and growth regulatory pathways. Furthermore, they demonstrate increased control of systemic disease when Raf1 and the bone microenvironment are targeted, providing for combined disruption of pathways regulating motility and honing to bone. We also went on to demonstrate for the first time the therapeutic relevance of inhibiting cell motility in osteoclasts, through inhibition of Raf1 activation, leading to disruption of actin reorganization and inhibition of bone-destructive resorptive cavity formation. Together, these findings support further investigations of a multi-function targeting paradigm and one involving inhibition of cell motility in particular.

Cell culture and reagents

The following prostate cancer cell lines, PC3, LNCaP, and VCaP were obtained from ATCC, the constitutively active luciferase–expressing PC3-luc cell line was obtained from PerkinElmer, the original characteristics of PC3-M cells have been previously described by us (13). The constitutively active luciferase–expressing PC3-M-luc cell line was established by transducing the parental cell line with pGL4.50 luciferase reporter (Promega), and culturing under hygromycin selection for stable integration. The LNCaP/AR-luc cell line was kindly provided by Charles Sawyers (Memorial Sloan Kettering Cancer Center, New York, NY; ref. 14).

The osteoclast precursor mouse macrophage cell line, RAW 264.7, was obtained from ATCC. All cells were cultured as described previously (9, 13, 15) and were maintained at 37°C in a humidified atmosphere of 5% carbon dioxide with biweekly media changes. All cell lines were drawn from stored stock cells, and replenished on a standardized periodic basis and were routinely monitored for Mycoplasma (PlasmoTest, InvivoGen). Cells were authenticated as follows: they were acquired from the originator of that line, grown under quarantine conditions, expanded and stored as primary stocks, and not used until following conditions were met: Mycoplasma negative; through morphologic examination; growth characteristics; hormone responsiveness; or lack thereof. KBU2046 was synthesized as described previously (9). Enzalutamide (#S1250), docetaxel (#S1148), and ZA (#S1314) were purchased from Selleckchem, and reconstituted per the manufacturer's recommendations.

Cytotoxicity assays and colony formation assays

To assess the impact of cytotoxic chemotherapy on prostate cancer growth kinetics 7-day soft agar colony formation assays and 3-day MTT cell growth inhibition assays were performed as described by us (16). MTT assays were in replicates of N = 3, and were repeated, soft agar assays were in replicates of N = 2 and are presented as mean number of colonies expressed as percent of untreated controls.

Reverse transcription and qPCR analysis

RNA was isolated and reverse transcription quantitative PCR (qRT-PCR) performed as previously described by us (17). Resultant data were analyzed using the 2−ΔΔCt method (18), normalized to GAPDH, for the following primer/probe sets (ABI) prostate-specific antigen (PSA) (Hs02576345_m1) and GAPDH (Hs99999905_m1). Assays were performed in triplicate, and were repeated.

Enzalutamide-mediated cell growth inhibition assays

Enzalutamide-mediated growth inhibition was assessed using Sulforhodamine B (SRB) growth inhibition assays, were performed as described previously (19). In brief, LNCaP and VCaP cells were preincubated in charcoal-striped medium for 48 hours and then seeded (1.9 × 104) in 96-well plates. Cells were stimulated with 1.0 nmol/L R1881 and treated for 72 hours with 10 μmol/L KBU2046 or vehicle control in the presence or absence of 10 μmol/L enzalutamide, and subsequently SRB assays were performed. Assays were in replicates of N = 6, and were repeated.

Subcellular localization of androgen receptor

LNCaP and VCaP cells were preincubated in the charcoal-striped medium containing 10 μmol/L KBU2046 or vehicle control for 72 hours. Cells were then stimulated with 1.0 nmol/L R1881 and treated for 24 hours with 10 μmol/L KBU2046 or vehicle control in the presence or absence of 10 μmol/L enzalutamide. The nuclear and cytosolic subcellular fractions of the cells were prepared using the NE-PER Nuclear and Cytoplasmic Extraction Kit (Pierce) according to the manufacturer's instructions. Protein was quantified by BCA Assay (Thermo Fisher Scientific), the per manufacturer's instructions.

Western blot analysis

Western blots were performed as described by us (9, 16). All Western blots were repeated at least once. Antibodies recognizing androgen receptor (AR, #5153S), Lamin B1 (#13435), and α-Tubulin (#3873) phospho-c-RAF (ser338) (#9427), c-Raf (#53745), MEK 1/2 (#12671), phospho-MEK 1/2 (#9121), ERK 1/2 (#4695), phospho-ERK 1/2 (#4370), GAPDH (#2118), antimouse IgG-HRP linked secondary (#7076), and anti-rabbit IgG-HRP linked secondary (#7074) antibodies were purchased from Cell Signaling Technology. Pierce ECL Western blotting substrate (#32106) and SuperSignal West Femto maximum sensitivity substrate (#34096) were purchased from Thermo Fisher Scientific. All primary antibodies were used at a dilution of 1:1,000 and corresponding secondary antibodies used at a dilution of 1:5,000.

Immunofluorescence

Raw 264.7 cells were grown in differentiation medium (described below), with or without KBU2046, for 4 days on a Chamber Slide (Nunc Lab-Tek). Chambers were washed once with PBS, followed by 10-minute fixation in cold 10% neutral-buffered formalin (VWR) and a post-fixation wash with PBS. Permeabilization was achieved with incubation in 0.1% Triton X-100 (Sigma) for 5 minutes, and then blocked overnight at 4°C with 1% BSA (Sigma) containing 10% goat serum (Thermo Fisher Scientific), 0.3 mol/L glycine (Sigma), and 0.1% PBS-Tween20. The cells were then incubated for 2 hours at 4°C in Dylight Phallodin (Cell Signaling technology, #12935) antibody diluted 1:40 in 1 × PBS with 1% BSA and 0.3% Triton X-100. Subsequently, the slide was washed once with PBS, a coverslip mounted with ProLong Gold Antifade Reagent with DAPI (Cell Signaling Technology, #8961), and imaged on a Confocal Microscope (Nikon/Yokogawa Spinning Disc).

Bone degradation assays

Low passage (p ≤ 4) RAW 264.7 cells were suspended in osteoclast differentiation media [MEM-alpha (Gibco) with 10% FBS, 1% antibiotic-antimycotic, and 50 ng/mL RANKL (EMD Millipore, R0525)] and plated in 96-well Osteo Surface Assay plates (Corning) at 5,000 cells per well. For wells receiving ZA treatment, the osteosurface was precoated for 30 minutes at 25°C with the indicated concentration of ZA suspended in sterile water, followed by a gentle rinsing in sterile water to remove free compound prior to plating of cells. Cells were treated with 10 μmol/L KBU2046 or control for 6 days with the media and treatment refreshed on day 4. On day 6, the wells were washed with PBS, and cellular material removed by a 5-minute incubation in a 10% bleach solution followed by a through rinsing with ultrapure water. To visualize the resultant osteoclast-mediated bone degradation, the intact mineralized matrix was stained following the manufacture's recommendations (Corning) utilizing a modified Von Kossa staining protocol. The resultant stained surfaces were imaged and subsequently evaluated using the “BoneJ” plugin of ImageJ. All treatments were performed in replicate of N = 3, and were repeated.

ApoTox-Glo assay

The ApoTox-Glo Triplex Assay (Promega) was performed to assess cell viability, cytotoxicity, and caspase-3/7 activation, per the manufacturer's instructions. In brief, 5000 RAW 264.7 cells were plated in 96-well assay plates and cultured in the osteoclast differentiation media for 4 days, at which point the media was replenished with or without KBU2046 and cell responses measured after 24, 48, and 72 hours following the manufacture's protocol.

Animal models of systemic effects and metastasis

All animal studies adhered to the NIH Guide for the Care and Use of Laboratory Animals, were treated under Institutional Animal Care and Use Committee–approved protocols by Oregon Health and Sciences University, complied with all federal, state, and local ethical regulations, and their design and implementation followed sanctioned guidelines (20). Animals were housed in barrier (for immunocompromised mice) facilities, with a 12-hour light/dark cycle and given soy-free food and water ad libitum.

Animal study sample size determination

Sample sizes were determined using the sample size estimation formula for differences in means with power set 80%, two-sided α = 0.05, and a prespecified effect size of 30%.

Prostate cancer subcutaneous implantation

A total of 2.5 × 105 PC3-M cells in sterile PBS were implanted into the right flank of 6- to 7-week-old male athymic nude mice (Charles River Laboratories) and tumor growth measurements were performed twice a week as previously described by us (21). Treatment via oral gavage 5 days per week with 80 mg/kg KBU2046 or vehicle control (sesame oil), as well as weekly intraperitoneal injections of docetaxel (0 and 20 mg/kg) began 18 days post implantation when tumors reached our enrollment criteria of approximately 200–300 mm3. Experimental groups were randomly assigned to cages prior to the initiation of the study. Tumor measurements were obtained in a blinded fashion, and tumor volumes were calculated with the following formula: length × (width)2 × 0.5.

Prostate cancer orthotopic implantation

Orthotopic implantation of 2.5 × 105 PC3M-luc into 7- to 8-week-old male athymic mice was performed as previously described by us (9, 22). Daily oral gavage treatment with 150 mg/kg KBU2046 or was initiated 3 days prior to implantation and continued throughout the duration of the study. Weekly intraperitoneal injections with 7.5 mg/kg docetaxel or control were administered beginning 1 week post implantation. Primary tumor out growth was monitored via weekly in vivo imaging system (IVIS imaging). At the completion of the study the primary tumor was resected and weighed and lungs were resected and snap frozen. To determine the human prostate cancer metastatic burden in each mouse lung, the snap frozen tissue was pulverized, DNA isolated, and Alu-sequencing performed as described previously (23). The following custom TaqMan primer probe set was utilized: forward primer (101 F) GGTGAAACCCCGTCTCTACT, reverse primer (206 R) GGTTCAAGCGATTCT CCTGC, and hydrolysis probe (144RH) CGCCCGGCTAATTTTTGTAT. The 144RH probe was labeled with a 5′ fluorescent reporter (6-FAM) and a 3′ fluorescent quencher (Black Hole Quencher). The relative content of human DNA was calculated by using the comparative CT method, CT (sample) – CT (negative control). All samples were run in triplicate and were repeated.

Castration-resistant prostate cancer subcutaneous implantation

Six- to 7-week-old male athymic mice received trans-scrotal bilateral orchiectomy with a single incision in the midline of scrotums under general anesthesia and then allowed to recover for 10 days. Following the recovery period, mice received a subcutaneous implantation of 2.0 × 106 LNCaP/AR-luc into the right flank as previously described by us (21). Treatment with 80 mg/kg KBU2046 or control was administered via oral gavage 5 times weekly starting 4 weeks post-implantation, prior to the emergence of the castration-resistant phenotype, and continued for 12 weeks. Experimental groups were randomized prior to the initiation of treatment. Twice weekly tumor measurements were performed and AR activity monitored biweekly through IVIS imaging.

Castration-resistant prostate cancer orthotopic implantation

Castrations of athymic mice were performed as described above, and following a 2-week recovery period, orthotopic implantations of 5.0 × 105 LNCaP/AR-luc were performed as previously described by us (9, 22). Biweekly IVIS imaging was performed to assess AR activity and treatment with 80 mg/kg KBU2046 or control via oral gavage 5 times weekly began 12 weeks post-implantation, as the castration-resistant phenotype began to emerge, and continued for duration of 4 weeks. Experimental groups were randomly assigned to cages prior to the initiation of the study.

Prostate cancer intracardiac injection

Intracardiac injection of 4.0 × 105 PC3-luc cells into the left ventricle of 7- to 8-week-old athymic mice under ultrasound guidance was performed as previously described by us (9). Weekly intraperitoneal injections of 100 μg/kg ZA or vehicle control (PBS) began 7 days prior to the intracardiac injection and continued throughout the duration of the study, whereas daily oral gavage treatment with 150 mg/kg KBU2046 or control began 3 days prior to the intracardiac injection. All treatments were randomly assigned to cages prior to the initiation of the study. IVIS imaging was performed 30 minutes post intracardiac injection to confirm systemic distribution of cells and conducted weekly starting 7 days post injection, continuing for a period of 4 weeks. Animals were excluded from the analysis if 30-minute post-injection IVIS imaging revealed a focal signal only in the chest indicating a failed injection where cells were not distributed into circulation. CT radiographic imaging of the entire cohort was performed with an Inveon X-ray μCT Scanner (Inveon, Siemens) at the completion of the study. Resultant images were analyzed in a blinded fashion using Inveon Research Workplace 4.2 visualization/analysis software and ImageJ. Specifically, to assess metastasis-associated bone destruction, images were exported from the Inveon workplace in the Digital Imaging and Communications in Medicine format for analysis in ImageJ. Each imported image was adjusted to a common brightness/contrast threshold and the “Stack → Crop (3D)” and “volume viewer” plugins were utilized to perform multiplanar reconstruction (MPR) of regions of interest (ROI, mandible and femurs). Utilizing the MPR of the bilateral femurs, a blinded operator obtained step-section serial images of trabecular bone at both the proximal and distal ends of each femur. Subsequently, the “ROI manager” plugin was utilized to specify fixed ROIs in each step-section images, and from which measurement bone density were obtained. Bone disruption of the mandible was assessed in a blinded fashion utilizing the MPR of the mandible. Here, the loss of bone in the periodontal cavity was calculated by measuring the distance between the edge of molar root and the wall of periodontal cavity at both coronal and transverse planes.

Statistical analysis

For all experiments unless otherwise stated, comparisons between two groups were evaluated with the two-sided Student t test. Some experiments, as denoted, used Fisher exact test using a significance threshold of P ≤ 0.05. In such instances, the mean value of the control group was used as the threshold to assign the categorical nature (i.e., >/ = versus < the mean value) of individual outcomes within different treatment groups.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Raf1 targeting does not interfere with inhibition of growth mediated by targeting microtubule function

Stabilization of microtubules, by the tubulin-binding agents docetaxel and cabazitaxel, are potent inhibitors of the growth of androgen-independent prostate cancer (24), and both have been shown to prolong life in the clinical setting (25, 26). Combined targeting of Raf1, using KBU2046, and microtubules, using either docetaxel or cabazitaxel, did not affect the ability of microtubule targeting to inhibit the growth of androgen-independent PC3-M cells in vitro (Fig. 1A and B). A similar lack of interference of growth inhibition was also observed when KBU2046 was coadministered in the context of agents that act through different mechanisms and upon different targets, and was demonstrated in both PC3-M human prostate cancer and MDA-MB-231 human breast cancer cells (Supplementary Fig. S1).

To evaluate whether Raf1 inhibition would inhibit the efficacy of microtubule-targeting agents in vivo, mice were subcutaneously implanted with PC3-M cells and treated with docetaxel ± KBU2046 (Fig. 1C). Compared with control mice, docetaxel significantly inhibited tumor growth (P < 0.01). However, KBU2046 did not significantly affect growth when added to control or docetaxel-treated mice (P = 0.264 and 0.926, respectively). These findings demonstrate that Raf1 inhibition, mediated by KBU2046, does not inhibit growth inhibition, mediated by targeting microtubules.

Combined targeting of Raf1 and microtubule function has improved efficacy in systemic models of disease

Unregulated cell growth and unregulated cell movement constitute two elemental processes of cancer. A therapeutic strategy wherein both of these processes are targeted is rationally based. We sought to examine whether combined Raf1 and microtubule targeting would allow for sustained inhibition of tumor growth kinetics achieved by targeting microtubule function, while providing for improved antimetastatic efficacy, as compared with either approach alone. The PC3-M orthotopic implantation murine model allows for the quantification of primary tumor growth, the assessment of distant metastasis to the lungs, and has previously been used to measure the effects of therapy and genetic alterations on metastasis (9, 16, 27).

To assess the efficacy of a spectrum of docetaxel concentrations, mice received orthotopic implants of PC3-M-luc cells, were treated with 0, 10, or 20 mg/kg docetaxel, and tumor growth monitored with weekly IVIS imaging (Supplementary Fig. S2). Docetaxel significantly inhibited growth compared with controls in a dose-dependent manner. Subsequently, mice with orthotopic implants of PC3-M-luc cells were treated with docetaxel, KBU2046, docetaxel+KBU2046, or vehicle, as denoted, and effects on primary tumor growth and lung metastasis quantified (Fig. 2). Prior studies demonstrated that it is possible to detect as few as 10 human cells per mouse organ, that is, kidney, using human Alu element–specific qPCR (23). We demonstrated that there is a linear relationship between the number of PC3-M-luc cells present in mouse lung and qPCR-based detection of human Alu elements (Pearson R2 = 0.96; Supplementary Fig. S3). Using human Alu element–specific qPCR to quantify PC3-M-luc metastasis to lungs, we demonstrated that docetaxel, KBU2046, and docetaxel+KBU2046 decreased metastasis by 27% (P < 0.05), 20% (P < 0.05), and 48% (P < 0.001), respectively, as compared with control mice. Of note, the combination of docetaxel+KBU2046 achieved a significantly greater decrease in metastasis than that observed with either docetaxel (P < 0.05) or KBU2046 (P < 0.01) alone (Fig. 2A). Based upon the effect of each treatment alone, the calculated combined effect of docetaxel+KBU2046 is 42% reduction, giving an experimental/calculated ratio of 1.1, or additive-to-slightly-synergistic.

Evaluation of prostate tumor growth by weekly IVIS imaging demonstrated exponential growth kinetics in cohorts not receiving docetaxel, and slower growth in cohorts receiving docetaxel (Fig. 2B). Measurement of actual tumor weight demonstrated that docetaxel and docetaxel+KBU2046 decreased tumor weight by 49% (Fisher exact test, P < 0.05) and 37% (Fisher exact test, P < 0.05), respectively, compared with control (Fig. 2C). KBU2046 by itself or combined with docetaxel had no significant effect, nor did it significantly impair docetaxel efficacy. These findings demonstrate that combined Raf1 and microtubule targeting allows for dual targeting of cell growth and cell movement, resulting in sustained control of primary tumor growth and enhanced control of metastatic disease progression.

Raf1 inhibition does not interfere with hormone therapy

Targeted inhibition of the androgen axis effectively inhibits prostate cancer cell growth, and constitutes the mainstay of the treatment strategy for metastatic prostate cancer (28). It is therefore important to ensure Raf1 inhibition does not interfere with it. Initial hormone treatment involves decreasing testicular androgen production, and is termed androgen deprivation therapy (ADT). Disease progression is inevitable, resulting in castrate-resistant prostate cancer, and is commonly treated with an AR antagonist, such as enzalutamide (29). To evaluate the effect Raf1 inhibition upon these paradigm investigations began with the androgen-sensitive LNCaP and VCaP human prostate cancer cells, which have been well-characterized for their response to hormonal-targeted therapies (14, 30).

Cells were cultured in media depleted of androgen and after a washout period of 72 hours were treated with the synthetic androgen R1881, KBU2046, enzalutamide, vehicle (control), and combinations, as denoted. The resultant effects upon induction of the androgen-responsive gene, PSA were measured (Fig. 3A and B). In both cell lines, R1881 significantly increased PSA expression and enzalutamide decreased it. In LNCaP cells, KBU2046 had no significant effect. In VCaP cells, KBU2046 significantly decreased PSA expression in R1881-treated cells by 36% (P < 0.05). While it also significantly increased expression in control cells (P < 0.05), absolute baseline levels were very low and the relative increase was only 14%. Evaluating effects on cell growth, enzalutamide significantly inhibited it compared with controls (P < 0.05) in both LNCaP and VCaP cells, whereas the addition of KBU2046 to either control or enzalutamide-treated cells had no significant effect (Fig. 3C and D).

Several considerations warranted a deeper examination of the effect of Raf1 inhibition on AR function. In VCaP cells, KBU2046 significantly inhibited R1881-mediated PSA gene expression, consistent with inhibition of AR function. In addition, AR is a client protein of HSP90, AR function requires HSP90-mediated chaperone action, and direct inhibitors of HSP90 decrease client protein expression and inhibit AR function (31). KBU2046 is not a direct inhibitor of HSP90, and as such does not globally decrease client protein expression. As a SHAM acting agent, it selectively modulates client proteins that regulate cell motility. However, the effect of KBU2046 on AR expression and function had not been previously explored. After ligand binding, functional AR translocates from the cytoplasm to the nucleus, and nuclear localization therefore provides a measure of functional AR. By treating LNCaP and VCaP cells with R1881, KBU2046, and/or enzalutamide, and measuring AR expression by Western blot in resultant nuclear and cytoplasmic cellular fractions, we demonstrated that KBU2046 did not affect AR expression or nuclear localization in response to R1881, nor its inhibition by enzalutamide (Fig. 3E).

To complete our assessments of the androgen-related paradigm, the effect of KBU2046 on resistance to hormone therapy was evaluated. The outgrowth of LNCaP/AR-luc cells in castrate mice evaluates the transition to a castration-resistant phenotype (14). Tumor outgrowth is detectable approximately 8 weeks post implantation, and growth kinetics are typically monitored until 16 weeks post implantation. Here, two treatment approaches were investigated, the effects of beginning treatment early at 4 weeks, or late at 12 weeks, and were performed in mice receiving subcutaneous or orthotopic implants, respectively (Fig. 3F and G). KBU2046 did not alter tumor growth in either model. Taken together, these findings support the notion that Raf1 inhibition does not affect androgen signaling, nor does it inhibit the efficacy of ADT or of AR antagonist therapy.

Combining Raf1 inhibition with modulation of the bone microenvironment has improved efficacy

Metastasis to the bone occurs in over 90% of people with metastatic prostate cancer (2). Upon arrival to bone, prostate cancer cells modify the environment in a manner conducive to their outgrowth engaging what is known as the “vicious cycle” (32–34). Specifically, prostate cancer cells secrete IL11 and parathyroid hormone-related protein, stimulating osteoblasts to secrete receptor activator of nuclear factor kappa-B ligand (RANKL), RANKL induces osteoclasts to degrade bone, the resultant degradation of bone matrix releases insulin growth factor and TGFβ, which act to sustain prostate cancer cell growth. We previously demonstrated that KBU2046 inhibits cell motility by inhibiting activation of Raf1, specifically, by inhibiting phosphorylation of its Ser338 activation motif (9). A central function of Raf1 relates to regulating reorganization of the actin cytoskeleton, which in turn affects how cells interact with and move through the microenvironment (35). Cytoskeleton reorganization is required for osteoclasts to successively move across and to bind bone matrix, doing so by arranging their actin cytoskeleton to form a low pH resorptive cavity in which degradation of bone mineral occurs (36). These facts led us to consider that osteoclast-mediated bone degradation was dependent upon cytoskeleton reorganization, and therefore might be inhibited by KBU2046.

To examine this paradigm, we first evaluated whether KBU2046 would inhibit Raf1 phosphorylation in osteoclasts. After treatment with RANKL, RAW 267.4 cells differentiate into mature osteoclasts. Mature osteoclasts are multinucleated, express tartrate-resistant acid phosphatase (TRAP), and have the capacity to form actin rings that create an occluded cavity in which bone is degraded. KBU2046 inhibits phosphorylation of Raf1′s activation motif in RANKL-treated RAW 267.4 osteoclast cells, and does so in a time-dependent manner (Fig. 4A). In osteoclasts, Raf1 is known to phosphorylate MEK1/2, in turn phosphorylating ERK1/2 (37), with the latter acting to stimulate osteoclast differentiation and bone resorptive activity (38, 39). The functional relevance of KBU2046-mediated inhibition of Raf1 phosphorylation is shown by demonstrating inhibition of MEK1/2 and ERK1/2 phosphorylation. Osteoclasts in Fig. 4A were not cultured on bone mineral, demonstrating that KBU2046-mediated inhibition of Raf1 phosphorylation is not dependent upon the presence of bone mineral. We show in Supplementary Fig. S4 that similar effects are observed when RAW 267.4 cells are grown in the presence of bone mineral.

Treatment of RAW 267.4 cells with KBU2046 induces several structural and functional changes. It reduces RANKL-mediated formation of multinucleated cells, and those that do form have lower numbers of nuclei (see 40× panels, Fig. 4B). The characteristic prominent actin ring (white arrow) surrounding the resorptive cavity (yellow arrow) is readily apparent in control cells treated with RANKL ligand, but is lacking in KBU2046-treated cells (100×, panels). Consistent with it inhibiting maturation of RAW 267.4 cells, TRAP staining is decreased in KBU2046-treated cells (Fig. 4C). In the absence of RANKL, undifferentiated RAW 267.4 cells exhibit a phenotype characterized by isolated cells with long thin cytoplasmic extensions (Fig. 4B). However, treatment with KBU2046 markedly reduces cellular extensions, demonstrating that its ability to induce effects on RAW 267.4 cells is not RANKL dependent.

To determine whether KBU2046 inhibits osteoclast-mediated bone degradation, we used Corning Osteo Assay Surface plates, which are coated with calcium-phosphate–based bone mineral. This platform supports osteoclast growth, allowing for the quantitative measurement of bone degradation, and has previously been used to measure bisphosphonate-mediated inhibition of bone degradation (40). RAW 267.4 cells were grown on Osteo Assay plates in the presence of RANKL, treated with KBU2046 at 1 μmol/L, 10 μmol/L, or with vehicle (control), and effects on bone destruction quantified after 6 days. In this manner we demonstrated that when 10 μmol/L KBU2046 is present, it significantly increased bone surface area by 28% ± 1.4% (mean ± SEM) compared with control (P < 0.05; Fig. 4D). To further substantiate the role of Raf1, we demonstrate in Supplementary Fig. S5 that each of the Raf1-specific inhibitors, ZM336372, GW5074, and NVP-BHG712, inhibit osteoclast-mediated bone degradation, as well as RANKL-induced osteoclast maturation. The latter is demonstrated by decreased TRAP staining, decreased formation of multinucleated cells, and inhibition of resorptive cavity formation. Together, these findings demonstrate that we can inhibit degradation of bone matrix through an innovative strategy targeting Raf1.

We next examined targeting of the bone microenvironment coupled to targeting cell motility. Bisphosphonates modify the bone microenvironment by chemically binding to calcium-phosphate bone mineral, are then taken up by osteoclasts after they digest the bound mineral, and once inside the cell, they inhibit osteoclast-mediated bone degradation (41). The bisphosphonate, ZA, is used clinically to decrease bone fractures in men with metastatic prostate cancer (11). ZA is administered to humans by injection, whereupon it binds to bone mineral, while the remainder is rapidly cleared from the body, and that bound to bone mineral is responsible for inhibition of osteoclast activity and resultant decrease in bone fractures in men with metastatic prostate cancer (42). We therefore focused our investigations upon ZA. Furthermore, we emulated the pharmacologic situation in humans by pretreating Osteo Assay plates with ZA, washing nonbound ZA away prior to adding RAW 267.4 cells, then treated, or not, with KBU2046, and measured resultant effects upon bone degradation. ZA was evaluated at 2, 10, and 20 μmol/L, demonstrating a concentration-dependent increase in bone surface area, all significant compared with control (P < 0.05; Fig. 5A and B). KBU2046 similarly significantly increased bone surface area (P < 0.05). Importantly, the combination of ZA+KBU2046 yielded significantly improved efficacy compared with either agent alone (P < 0.05), did so with all ZA concentrations tested, and with the combination of 20 μmol/L ZA+KBU2046, bone surface area increased to 193% ± 1.6% of control (mean ± SEM). Based upon the effect of each treatment alone, the calculated combined effects of ZA+KBU2046 are 170%, 190%, and 210% compared with untreated controls for 2, 10, and 20 μmol/L ZA, respectively, giving experimental/calculated ratios of 0.91, 1.0, and 0.92, respectively. These findings approach pure pharmacologic additivity.

We next examined the effect of ZA and KBU2046 on osteoclast cell viability, apoptosis, and cytotoxicity. RAW 267.4 cells were treated with RANKL for the duration of the experiment, at day 4 they were then treated with ZA, KBU2046, the combination, or vehicle (for controls), and effect on cell viability, apoptosis, and cytotoxicity measured by Triplex assay at 24, 48, and 72 hours after treatment (Fig. 5C–E). During the 7-day time course of this experiment, RANKL-treated RAW 267.4 cells underwent maturation to terminally differentiated osteoclasts. Findings in control cells are consistent with this process, demonstrating successive decreases in cell viability, accompanied by increases in cellular cytotoxicity and apoptosis with time. Findings with KBU2046 are consistent with it delaying the maturation process, and include significant increases in cell viability and decreases in cytotoxicity at 72 hours, compared with control (P < 0.05). There is a notable significant increase in apoptosis in KBU2046-treated cells at 72 hours compared with controls (P < 0.05), possibly reflecting prior findings by other investigators linking suppression of Raf1 signaling to an increase in apoptotic activity (43). The combination of ZA+KBU2046 significantly increased apoptosis and cytotoxicity at 48 hours, but there was little-to-no effect on viability at 72 hours.

To evaluate the effectiveness of this strategy in vivo, we used the human PC3-luc prostate cancer intracardiac injection model (9, 44). In this model, mice develop widespread metastasis to the bone, with those to the mandible being the most prominent and symptomatic. Mice were treated with either ZA, KBU2046, ZA+KBU2046, or vehicle (control), and at the end of the experiment CT images of the bones were obtained and metastasis to the mandible and femur quantified (Fig. 5F–H). Weekly IVIS imaging demonstrated progressive tumor growth over the 5 weeks of the experiment in the mandible, femur, and whole body, across all cohorts of mice (Supplementary Fig. S6). Representative CT images demonstrate the typically large destructive lesions in the mandible (Fig. 5F), and these were quantified by direct measurement (Fig. 5G). Metastasis to the femur were smaller, led to generalized loss of bone, with representative images demonstrating decreased density of trabecular bone and thinning of cortical bone (Fig. 5F), and were quantified by measurement of density (Fig. 5H). In the mandible, KBU2046 inhibited bone destruction by 33% ± 4.2% (mean ± SEM; P < 0.05) compared with control. While ZA also decreased bone destruction, this effect was not significant. Importantly, ZA did not impair KBU2046 efficacy. In the femur, bone density was significantly (P < 0.05) improved to 156% ± 2.7% and 122% ± 2.4% of untreated control mice by ZA and KBU2046, respectively. Furthermore, the combination of ZA+KBU2046 significantly (P < 0.05) increased density to 176% ± 3.7% as compared with either treatment alone. Based upon the effect of each treatment alone, the calculated combined effect of ZA+KBU2046 on the femur is 192% of control, giving an experimental/calculated ratio of 0.92, which approaches pharmacologic additivity for effects on the femur. Taken together, these findings demonstrate that combined targeting of cell motility and the bone microenvironment exert desirable therapeutic effects across different anatomic sites that exceed the effects achieved with either agent alone, resulting in improved therapeutic efficacy, an effect observed across both in vitro and in vivo model systems.

Increased cancer cell motility is a characteristic of cancer, which represents a clinically relevant driver of morbidity and mortality (2–6), and constitutes an important function that can now be effectively and selectively inhibited by targeting activation of Raf1 (9). As such, it is centrally important to consider it as one of several abnormal functions. In this regard, two criteria become apparent. First, this new found therapeutic capability cannot mitigate the efficacy of agents that act upon targets with established clinical efficacy. Second, it underscores the strategic importance of treating cancer in a manner that targets multiple functions. Such an approach constitutes multifunctional therapy. This concept should not be confused with multi-modality therapy. The latter uses multiple therapies directed at different targets and/or pathways, but which relate to a similar set of cellular functions, such as growth and or viability.

We herein demonstrate for the first time that antimotility therapy, through KBU2046-mediated inhibition of Raf1 activation, can be deployed with agents that act upon several different pathways relevant to cancer progression at the systemic level, and that have established efficacy in humans. Specifically, we demonstrate sustained efficacy in conjunction with inhibition of cell growth, through taxane-mediated inhibition of microtubule function, and do so in cell culture–based studies, as well as in two murine models, one utilizing subcutaneous and one utilizing orthotopic implantation. In the case of ZA, an agent that modifies the bone microenvironment, we also demonstrate that antimotility therapy does not interfere with its activity. In fact, we demonstrate that it enhances it. Here too, experiments spanned cell culture- and human murine xenograft–based models of bone destruction. In addition, we demonstrated that antimotility therapy does not interfere with therapies that target the androgen axis. Here too, studies involved cell culture models, as well as murine models of subcutaneous and orthotopic implantation. Studies were performed in the context of ADT and androgen receptor antagonist therapy, examining the effects upon cell and tumor growth, androgen-responsive gene activation, and AR activation.

After demonstrating that antimotility therapy did not inhibit the efficacy of established therapy, we went on to demonstrate its applicability in the context of multifunctional therapy.

Two notions support the rationale for pursuing a multi-functional treatment strategy. The first relates to the requirement for different functions to sustain cancer progression at the systemic level. The current example fits this, where motility, interaction with a distant organ microenvironment, and cell growth are all necessary functions for the development of clinically relevant metastasis. The second notion relates to evidence that therapeutic pressure can shift cancer dependence from one hallmark-related function to another, with antiangiogenic therapy fostering increased cell invasion serving as an example (45, 46).

In this study, we specifically demonstrated that the combination of docetaxel+KBU2046 not only retained the growth inhibitory efficacy of docetaxel but also exhibited additive-to-slightly-synergistic efficacy at inhibiting metastasis. Such a strategic approach would be applicable to existing clinical scenarios where docetaxel is used to treat prostate cancer, especially in the newly diagnosed metastatic setting, where the combination would serve to decrease formation of secondary metastasis. Because robust preclinical systemic models of the formation of metastasis as a function of targeting the androgen axis are lacking, we have not demonstrated that KBU2046 would provide additive antimetastatic efficacy when combined with hormone therapy. However, existing data supports that the use of KBU2046 along with hormone therapy would further decrease the development of metastasis. Specifically, we have previously demonstrated that KBU2046 will inhibit the motility of androgen-responsive prostate cancer cell lines (9). Furthermore, KBU2046 did not interfere with AR antagonist therapy mediated by enzalutamide, a widely used agent shown to decrease metastasis in humans (47).

We also demonstrated that antimotility therapy can enhance the efficacy of bone microenvironment targeting agents such as ZA. Osteoclasts modify the bone microenvironment in manner that requires them to interface with the bone mineral surface and to reorganize their actin cytoskeleton to form a chamber whose function is to degrade bone. Given that Raf1 is a known regulator of this process (37–39), we hypothesized that KBU2046 mediated inhibition of Raf1 phosphorylation on its activation motif would inhibit actin reorganization and as a result inhibit bone degradation. We demonstrated that this was the case, and did so evaluating bone degradation in vitro and with animal models of bone destruction. These findings represent a new mechanism and new strategic approach to inhibiting osteoclast function and bone destruction, and have wide spread implications.

Given that KBU2046 affects osteoclast motility and that ZA effects the bone microenvironment, that is, separate but complementary mechanisms, we hypothesized that together they would have at least additive therapeutic efficacy at inhibiting osteoclast-mediated bone destruction. Our findings directly supported this concept. Specifically, our in vitro model system of osteoclast-mediated bone destruction demonstrated additive effects of the two agents over a spectrum of ZA concentrations. A combined targeting approach was also supported by our findings with the murine model of human prostate cancer–mediated bone destruction, where we observed two forms of improved therapeutic efficacy. In the femur, the combination of both agents gave additional efficacy when compared with either agent alone. The other form of improved efficacy relates to anatomic site. In the mandible, KBU2046 exhibited high efficacy, whereas ZA exhibited no significant efficacy. In the femur, ZA was highly efficacious, and while KBU2046 was also efficacious, it was much less than ZA. When ZA and KBU2046 are combined, they exert desirable therapeutic effects across different anatomic sites that exceed the effects achieved with either agent alone, resulting in improved therapeutic efficacy. It will be important in future studies to understand the mechanistic underpinnings of these anatomic differences in agent efficacy.

In summary, we demonstrated that inhibition of cancer cell motility can effectively be combined with other targeted functional therapies for cancer. Such an approach addresses the paradigm of cancer arising from a relatively wide set of dysfunctions that together constitute characteristics of cancer. Such an approach is associated with improved efficacy across several clinically relevant models, and together supports the translation of this approach into human studies.

R. Bergan has ownership interest (including patents) in Third Coast Therapeutics, patents (US Patent 8,481,760, US Patent 8,742,141, US Patent 9,839,625, and US Patent 10,231,949). No potential conflicts of interest were disclosed by the other authors.

Conception and design: L. Zhang, A. Pattanayak, R. Gordon, R. Bergan

Development of methodology: L. Zhang, A. Pattanayak, W. Li, R. Gordon

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L. Zhang, A. Pattanayak, W. Li, R. Gordon

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L. Zhang, W. Li, R. Gordon, R. Bergan

Writing, review, and/or revision of the manuscript: L. Zhang, A. Pattanayak, H.-K. Ko, R. Gordon, R. Bergan

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): L. Zhang, W. Li, G. Fowler, R. Gordon

Study supervision: R. Bergan

This study was supported with funding to R. Bergan by the United States Veterans Administration (IBX002842A) and the United States Department of Defense (W81XWH-15-1-0527). The authors would like to thank William Packwood for his assistance in performing the high-resolution CT imaging.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Hanahan
D
,
Weinberg
RA
. 
Hallmarks of cancer: the next generation
.
Cell
2011
;
144
:
646
74
.
2.
Minn
AJ
,
Massague
J
. 
Invasion and metastasis
. In:
DeVita
VT
,
Lawrence
TS
,
Rosenberg
SA
,
editors
.
CANCER: principals and practice of oncology
.
New York, NY
:
Lippincott Wiliams and Wilkins
; 
2008
. p.
135
46
.
3.
Seyfried
TN
,
Huysentruyt
LC
. 
On the origin of cancer metastasis
.
Crit Rev Oncog
2013
;
18
:
43
73
.
4.
Gupta
GP
,
Massague
J
. 
Cancer metastasis: building a framework
.
Cell
2006
;
127
:
679
95
.
5.
Norgaard
M
,
Jensen
AO
,
Jacobsen
JB
,
Cetin
K
,
Fryzek
JP
,
Sorensen
HT
. 
Skeletal related events, bone metastasis and survival of prostate cancer: a population based cohort study in Denmark (1999 to 2007)
.
J Urol
2010
;
184
:
162
7
.
6.
Pound
CR
,
Partin
AW
,
Eisenberger
MA
,
Chan
DW
,
Pearson
JD
,
Walsh
PC
. 
Natural history of progression after PSA elevation following radical prostatectomy
.
JAMA
1999
;
281
:
1591
7
.
7.
Talmadge
JE
,
Fidler
IJ
. 
AACR centennial series: the biology of cancer metastasis: historical perspective
.
Cancer Res
2010
;
70
:
5649
69
.
8.
Krishna
SN
,
Bergan
RC
. 
Therapeutic modulation of prostate cancer metastasis
.
Future Med Chem
2014
;
6
:
223
39
.
9.
Xu
L
,
Gordon
R
,
Farmer
R
,
Pattanayak
A
,
Binkowski
A
,
Huang
X
, et al
Precision therapeutic targeting of human cancer cell motility
.
Nat Commun
2018
;
9
:
2454
.
10.
Masood
I
,
Kiani
MH
,
Ahmad
M
,
Masood
MI
,
Sadaquat
H
. 
Major contributions towards finding a cure for cancer through chemotherapy: a historical review
.
Tumori
2016
;
102
:
6
17
.
11.
Saad
F
,
Gleason
DM
,
Murray
R
,
Tchekmedyian
S
,
Venner
P
,
Lacombe
L
, et al
Long-term efficacy of zoledronic acid for the prevention of skeletal complications in patients with metastatic hormone-refractory prostate cancer
.
J Natl Cancer Inst
2004
;
96
:
879
82
.
12.
Costa
L
,
Badia
X
,
Chow
E
,
Lipton
A
,
Wardley
A
. 
Impact of skeletal complications on patients' quality of life, mobility, and functional independence
.
Support Care Cancer
2008
;
16
:
879
89
.
13.
Liu
YQ
,
Kyle
E
,
Patel
S
,
Housseau
F
,
Hakim
F
,
Lieberman
R
, et al
Prostate cancer chemoprevention agents exhibit selective activity against early stage prostate cancer cells
.
Prost Cancer Prost Dis
2001
;
4
:
81
91
.
14.
Tran
C
,
Ouk
S
,
Clegg
NJ
,
Chen
Y
,
Watson
PA
,
Arora
V
, et al
Development of a second-generation antiandrogen for treatment of advanced prostate cancer
.
Science
2009
;
324
:
787
90
.
15.
Korenchuk
S
,
Lehr
JE
,
L
MC
,
Lee
YG
,
Whitney
S
,
Vessella
R
, et al
VCaP, a cell-based model system of human prostate cancer
.
In Vivo
2001
;
15
:
163
8
.
16.
Pavese
JM
,
Bergan
RC
. 
Circulating tumor cells exhibit a biologically aggressive cancer phenotype accompanied by selective resistance to chemotherapy
.
Cancer Lett
2014
;
352
:
179
86
.
17.
Ding
Y
,
Xu
L
,
Jovanovic
BD
,
Helenowski
IB
,
Kelly
DL
,
Catalona
WJ
, et al
The methodology used to measure differential gene expression affects the outcome
.
J Biomol Tech
2007
;
18
:
321
30
.
18.
Livak
KJ
,
Schmittgen
TD
. 
Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method
.
Methods
2001
;
25
:
402
8
.
19.
Vichai
V
,
Kirtikara
K
. 
Sulforhodamine B colorimetric assay for cytotoxicity screening
.
Nat Protoc
2006
;
1
:
1112
6
.
20.
Hollingshead
MG
. 
Antitumor efficacy testing in rodents
.
J Natl Cancer Inst
2008
;
100
:
1500
10
.
21.
Gordon
RR
,
Wu
M
,
Huang
CY
,
Harris
WP
,
Sim
HG
,
Lucas
JM
, et al
Chemotherapy-induced monoamine oxidase expression in prostate carcinoma functions as a cytoprotective resistance enzyme and associates with clinical outcomes
.
PLoS One
2014
;
9
:
e104271
.
22.
Pavese
J
,
Ogden
IM
,
Bergan
RC
. 
An orthotopic murine model of human prostate cancer metastasis
.
J Vis Exp
2013
;
79
:
e50873
.
23.
Funakoshi
K
,
Bagheri
M
,
Zhou
M
,
Suzuki
R
,
Abe
H
,
Akashi
H
. 
Highly sensitive and specific Alu-based quantification of human cells among rodent cells
.
Sci Rep
2017
;
7
:
13202
.
24.
Lyseng-Williamson
KA
,
Fenton
C
. 
Docetaxel: a review of its use in metastatic breast cancer
.
Drugs
2005
;
65
:
2513
31
.
25.
Tannock
IF
,
de Wit
R
,
Berry
WR
,
Horti
J
,
Pluzanska
A
,
Chi
KN
, et al
Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer
.
N Engl J Med
2004
;
351
:
1502
12
.
26.
de Bono
JS
,
Oudard
S
,
Ozguroglu
M
,
Hansen
S
,
Machiels
JP
,
Kocak
I
, et al
Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial
.
Lancet
2010
;
376
:
1147
54
.
27.
Lakshman
M
,
Huang
X
,
Ananthanarayanan
V
,
Jovanovic
B
,
Liu
Y
,
Craft
CS
, et al
Endoglin suppresses human prostate cancer metastasis
.
Clin Exp Metastasis
2011
;
28
:
39
53
.
28.
Gomella
LG
,
Singh
J
,
Costas
L
,
Trabulsi
EJ
. 
Hormone therapy in the management of prostate cancer: evidence-based approaches
.
Ther Adv Urol
2010
;
2
:
171
81
.
29.
Beer
TM
,
Armstrong
AJ
,
Rathkopf
DE
,
Loriot
Y
,
Sternberg
CN
,
Higano
CS
, et al
Enzalutamide in metastatic prostate cancer before chemotherapy
.
N Engl J Med
2014
;
371
:
424
33
.
30.
Furr
BJ
. 
The development of Casodex (bicalutamide): preclinical studies
.
Eur Urol
1996
;
29
:
83
95
.
31.
Fang
Y
,
Fliss
AE
,
Robins
DM
,
Caplan
AJ
. 
Hsp90 regulates androgen receptor hormone binding affinity in vivo
.
J Biol Chem
1996
;
271
:
28697
702
.
32.
Guise
TA
. 
The vicious cycle of bone metastases
.
J Musculoskelet Neuronal Interact
2002
;
2
:
570
2
.
33.
Guise
TA
,
Mundy
GR
. 
Cancer and bone
.
Endocr Rev
1998
;
19
:
18
54
.
34.
Logothetis
C
,
Morris
MJ
,
Den
R
,
Coleman
RE
. 
Current perspectives on bone metastases in castrate-resistant prostate cancer
.
Cancer Metastasis Rev
2018
;
37
:
189
96
.
35.
Yotova
I
,
Quan
P
,
Gaba
A
,
Leditznig
N
,
Pateisky
P
,
Kurz
C
, et al
Raf-1 levels determine the migration rate of primary endometrial stromal cells of patients with endometriosis
.
J Cell Mol Med
2012
;
16
:
2127
39
.
36.
Florencio-Silva
R
,
Sasso
GR
,
Sasso-Cerri
E
,
Simoes
MJ
,
Cerri
PS
. 
Biology of bone tissue: structure, function, and factors that influence bone cells
.
Biomed Res Int
2015
;
2015
:
421746
.
37.
Bradley
EW
,
Ruan
MM
,
Vrable
A
,
Oursler
MJ
. 
Pathway crosstalk between Ras/Raf and PI3K in promotion of M-CSF-induced MEK/ERK-mediated osteoclast survival
.
J Cell Biochem
2008
;
104
:
1439
51
.
38.
He
Y
,
Staser
K
,
Rhodes
SD
,
Liu
Y
,
Wu
X
,
Park
SJ
, et al
Erk1 positively regulates osteoclast differentiation and bone resorptive activity
.
PLoS One
2011
;
6
:
e24780
.
39.
Nakamura
H
,
Hirata
A
,
Tsuji
T
,
Yamamoto
T
. 
Role of osteoclast extracellular signal-regulated kinase (ERK) in cell survival and maintenance of cell polarity
.
J Bone Miner Res
2003
;
18
:
1198
205
.
40.
Abe
K
,
Yoshimura
Y
,
Deyama
Y
,
Kikuiri
T
,
Hasegawa
T
,
Tei
K
, et al
Effects of bisphosphonates on osteoclastogenesis in RAW264.7 cells
.
Int J Mol Med
2012
;
29
:
1007
15
.
41.
Polascik
TJ
,
Mouraviev
V
. 
Zoledronic acid in the management of metastatic bone disease
.
Ther Clin Risk Manag
2008
;
4
:
261
8
.
42.
Cremers
S
,
Papapoulos
S
. 
Pharmacology of bisphosphonates
.
Bone
2011
;
49
:
42
9
.
43.
Alejandro
EU
,
Johnson
JD
. 
Inhibition of Raf-1 alters multiple downstream pathways to induce pancreatic beta-cell apoptosis
.
J Biol Chem
2008
;
283
:
2407
17
.
44.
Chu
K
,
Cheng
CJ
,
Ye
X
,
Lee
YC
,
Zurita
AJ
,
Chen
DT
, et al
Cadherin-11 promotes the metastasis of prostate cancer cells to bone
.
Mol Cancer Res
2008
;
6
:
1259
67
.
45.
Ebos
JM
,
Lee
CR
,
Kerbel
RS
. 
Tumor and host-mediated pathways of resistance and disease progression in response to antiangiogenic therapy
.
Clin Cancer Res
2009
;
15
:
5020
5
.
46.
Ellis
LM
,
Reardon
DA
. 
Cancer: the nuances of therapy
.
Nature
2009
;
458
:
290
2
.
47.
Hussain
M
,
Fizazi
K
,
Saad
F
,
Rathenborg
P
,
Shore
N
,
Ferreira
U
, et al
Enzalutamide in men with nonmetastatic, castration-resistant prostate cancer
.
N Engl J Med
2018
;
378
:
2465
74
.