Introduction

The Prostate Cancer Working Group (PCWG1 and PCWG2) assembled consensus recommendations for prostate cancer clinical disease states with the goal of standardizing criteria not only for patient enrollment into clinical trials but also for clinically relevant endpoints. The current paradigm of clinical disease states recommended by PCWG2 includes 5 clinical subtypes in prostate cancer based on location of disease: first, locally progressing tumors and no metastatic disease, second, rising prostate-specific antigen (PSA) and no detectable metastatic disease, third, nodal spread and no visceral disease, fourth, bone disease, and fifth, visceral disease (liver and lung). Although these categories are based on location of metastatic sites and are recorded for clinical trial purposes, it is frequently the case that patients are stratified more simply by the lack of detectable metastases versus the presence of detectable metastases, regardless of whether in lymph node, bone, or viscera. Additionally, there has been no consensus prognostic difference between nodal and bone disease or, perhaps more importantly, no clear directives regarding treatment based on site of metastasis.

As previously mentioned, there are limitations to the current criteria. For example, the nonmetastatic, biochemically recurrent niche is a misnomer because it can be assumed that the sources of rising PSA are microscopic metastases falling below the limits of detection of standard imaging modalities. Thus, this clinical classification does not so much define the biology of this disease state as it reflects the limitations of current imaging technology, although there is usefulness in prognosis and treatment decision making. Similarly, there is increasing evidence that castration-resistant prostate cancer (CRPC), defined by an increasing PSA level in the setting of castrate levels of serum testosterone, remains reliant on previously unrecognized intracrine tumor androgens. This is further validated as more potent androgen axis manipulations such as abiraterone, enzalutamide (previously known as MDV3100), and others have been proved to slow prostate cancer disease progression and improve overall survival in patients with CRPC. Nevertheless, this designation remains clinically useful in clarifying treatment options. Similarly, the distinction between the pre-docetaxel and post-docetaxel settings that are currently used should be recognized as an artifice created by the field, albeit a necessary one, for the purpose of clinical trials and the drug approval process, instead of being of biologic significance. However, as the authors discuss, these terms may be rapidly evolving along with the landscape of the field.

Nonmetastatic (biochemically recurrent) castration-sensitive prostate cancer

Nonmetastatic castration-sensitive disease is found in patients observed to have a rising PSA, including those who have received definitive localized therapy such as surgery and/or radiation. When following prostatectomy, one commonly accepted definition of biochemical recurrence is a PSA level of 0.2 ng/mL (or greater with a second confirmatory value) although there are other definitions for biochemical failure after prostatectomy. When following external beam radiation, the definition of biochemical recurrence is a PSA at least 2 ng/mL higher than the nadir PSA, as dictated by the the RTO-ASTRO Phoenix consensus conference (distinct from the American Society for Radiation Oncology criteria). The PCWG2 criteria for progression of disease and clinical trial eligibility include a sequence of increasing PSA values at a minimum of 1-week intervals and a 2.0 ng/mL minimum starting value. If possible, PSA doubling time (PSAdt) can be estimated if 3 or more values are available at 4 or more weeks apart. If criteria are met, it is standard to assess for metastatic involvement by radiographic studies, most commonly radionuclide bone and computed tomography scanning. When no metastases are seen on imaging, the patient is considered to have nonmetastatic disease, although, to reiterate, it is assumed that the source of increasing PSA levels is microscopic metastases.

The prognosis of patients with castration-sensitive prostate cancer (CSPC) can be estimated with several different measures, including the Gleason score, lymph node involvement, time to PSA recurrence, and PSAdt. Several other PSA measures have been examined, such as PSA velocity and PSA density. PSAdt has emerged as a useful measurement that can be used to define prognosis. After prostatectomy, a PSAdt of greater than 10 months is correlated with a longer time to the development of distant metastases and to survival. Regardless of initial therapy, if PSAdt is less than 3 months, the outcome is universally poor.

The mainstay of therapy in CSPC is androgen-deprivation therapy (ADT), typically accomplished through orchiectomy or by treatment with either a gonadotropin-releasing hormone agonist or antagonist. Although ADT universally suppresses PSA after initiation, the long-term benefits of starting ADT in this early setting remain unclear, with little data to support early introduction. Consequently, any potential benefits of introduction of ADT in this setting must be balanced with concerns for short- and long-term side effects of treatment, including bone health, metabolic effects, body morphology changes, sexual side effects, and hot flashes. As a consequence, the challenge remains to distinguish the patients who may benefit from earlier introduction of ADT from those who may not.

A means of minimizing the long-term side effects and improving quality of life for men with nonmetastatic prostate cancer is the use of intermittent compared with continuous ADT. This has been most recently studied in the phase III randomized trial comparing intermittent versus continuous androgen suppression for patients with PSA progression after radical therapy (NCIC CTG PR.7/SWOG JPR.7/CTSU JPR.7/UK Intercontinental Trial CRUKE/01/013), in which nearly 1400 men with increasing PSA levels were randomized to receive intermittent androgen suppression (IAD) or continuous androgen deprivation (CAD) after primary or salvage radiotherapy, where IAD was given in 8-month treatment intervals with a threshold to resume therapy when PSA level was greater than 10 ng/dL. Preliminary results were reported at the American Society of Clinical Oncology (ASCO) 2011 annual meeting and demonstrated noninferiority of IAD with respect to overall survival. Moreover, the IAD arm had a reduction in hot flashes and improvements in quality of life measures, including less fatigue and sexual side effects compared with the CAD arm. As a result, IAD is an accepted standard of care option for men with nonmetastatic castration-sensitive disease.

Bone mineral loss and resultant osteoporotic fracture are long-term morbidities of ADT and additional threats to bone health in a disease that preferentially metastasizes to the bone in its later stages. An estimate of long-term bone effects of ADT comes from the placebo control arms of the trials of toremifene and denosumab in patients with nonmetastatic CSPC. Patients receiving CAD and randomized to placebo had reductions in bone mineral density of between 1% and 5% at the lumbar spine, total hip, femoral neck, and distal third of radius at 24 to 36 months. In contrast, in a study of 56 men receiving IAD, the long-term bone mineral density was only slightly lower than baseline after a median follow-up of 5.5 years. The use of bone-protective agents such as zolendronic acid and denosumab are discussed further in another article in this issue by Morgans and colleagues.

An additional benefit of IAD in the setting of biochemical recurrence is the ability to glean additional prognostic information. For example, after 9 months of induction ADT, the duration of the first off-cycle interval as defined by time to reach a specific PSA threshold can prognosticate for long-term outcomes such as castration-resistance or prostate cancer mortality. More recent evidence suggests that a longer time to PSA increase during the first off-cycle of IAD and after testosterone recovery (as defined as >50 ng/dL) may be associated with longer time to castration resistance.

The current challenge is to further prognosticate men with nonmetastatic, biochemically recurrent castration-sensitive disease—to discriminate between patients who will benefit from IAD and those with better prognosis who would be better served with close observation. Additionally, novel agents currently approved for later stages of disease are being investigated in earlier stages, such as sipuleucel-T. As more trials in this setting are planned, it is increasingly important to select patients with more aggressive disease features and subsequent event rates that will be adequate to design appropriate clinical trials with novel agents. Alternatively, the ability to accurately identify those with indolent disease allows us to more confidently direct these patients toward clinical trials studying nutraceuticals and less toxic agents, such as sulforaphane, which may offer lower risk of long-term morbidity.

Nonmetastatic (biochemically recurrent) castration-sensitive prostate cancer

Nonmetastatic castration-sensitive disease is found in patients observed to have a rising PSA, including those who have received definitive localized therapy such as surgery and/or radiation. When following prostatectomy, one commonly accepted definition of biochemical recurrence is a PSA level of 0.2 ng/mL (or greater with a second confirmatory value) although there are other definitions for biochemical failure after prostatectomy. When following external beam radiation, the definition of biochemical recurrence is a PSA at least 2 ng/mL higher than the nadir PSA, as dictated by the the RTO-ASTRO Phoenix consensus conference (distinct from the American Society for Radiation Oncology criteria). The PCWG2 criteria for progression of disease and clinical trial eligibility include a sequence of increasing PSA values at a minimum of 1-week intervals and a 2.0 ng/mL minimum starting value. If possible, PSA doubling time (PSAdt) can be estimated if 3 or more values are available at 4 or more weeks apart. If criteria are met, it is standard to assess for metastatic involvement by radiographic studies, most commonly radionuclide bone and computed tomography scanning. When no metastases are seen on imaging, the patient is considered to have nonmetastatic disease, although, to reiterate, it is assumed that the source of increasing PSA levels is microscopic metastases.

The prognosis of patients with castration-sensitive prostate cancer (CSPC) can be estimated with several different measures, including the Gleason score, lymph node involvement, time to PSA recurrence, and PSAdt. Several other PSA measures have been examined, such as PSA velocity and PSA density. PSAdt has emerged as a useful measurement that can be used to define prognosis. After prostatectomy, a PSAdt of greater than 10 months is correlated with a longer time to the development of distant metastases and to survival. Regardless of initial therapy, if PSAdt is less than 3 months, the outcome is universally poor.

The mainstay of therapy in CSPC is androgen-deprivation therapy (ADT), typically accomplished through orchiectomy or by treatment with either a gonadotropin-releasing hormone agonist or antagonist. Although ADT universally suppresses PSA after initiation, the long-term benefits of starting ADT in this early setting remain unclear, with little data to support early introduction. Consequently, any potential benefits of introduction of ADT in this setting must be balanced with concerns for short- and long-term side effects of treatment, including bone health, metabolic effects, body morphology changes, sexual side effects, and hot flashes. As a consequence, the challenge remains to distinguish the patients who may benefit from earlier introduction of ADT from those who may not.

A means of minimizing the long-term side effects and improving quality of life for men with nonmetastatic prostate cancer is the use of intermittent compared with continuous ADT. This has been most recently studied in the phase III randomized trial comparing intermittent versus continuous androgen suppression for patients with PSA progression after radical therapy (NCIC CTG PR.7/SWOG JPR.7/CTSU JPR.7/UK Intercontinental Trial CRUKE/01/013), in which nearly 1400 men with increasing PSA levels were randomized to receive intermittent androgen suppression (IAD) or continuous androgen deprivation (CAD) after primary or salvage radiotherapy, where IAD was given in 8-month treatment intervals with a threshold to resume therapy when PSA level was greater than 10 ng/dL. Preliminary results were reported at the American Society of Clinical Oncology (ASCO) 2011 annual meeting and demonstrated noninferiority of IAD with respect to overall survival. Moreover, the IAD arm had a reduction in hot flashes and improvements in quality of life measures, including less fatigue and sexual side effects compared with the CAD arm. As a result, IAD is an accepted standard of care option for men with nonmetastatic castration-sensitive disease.

Bone mineral loss and resultant osteoporotic fracture are long-term morbidities of ADT and additional threats to bone health in a disease that preferentially metastasizes to the bone in its later stages. An estimate of long-term bone effects of ADT comes from the placebo control arms of the trials of toremifene and denosumab in patients with nonmetastatic CSPC. Patients receiving CAD and randomized to placebo had reductions in bone mineral density of between 1% and 5% at the lumbar spine, total hip, femoral neck, and distal third of radius at 24 to 36 months. In contrast, in a study of 56 men receiving IAD, the long-term bone mineral density was only slightly lower than baseline after a median follow-up of 5.5 years. The use of bone-protective agents such as zolendronic acid and denosumab are discussed further in another article in this issue by Morgans and colleagues.

An additional benefit of IAD in the setting of biochemical recurrence is the ability to glean additional prognostic information. For example, after 9 months of induction ADT, the duration of the first off-cycle interval as defined by time to reach a specific PSA threshold can prognosticate for long-term outcomes such as castration-resistance or prostate cancer mortality. More recent evidence suggests that a longer time to PSA increase during the first off-cycle of IAD and after testosterone recovery (as defined as >50 ng/dL) may be associated with longer time to castration resistance.

The current challenge is to further prognosticate men with nonmetastatic, biochemically recurrent castration-sensitive disease—to discriminate between patients who will benefit from IAD and those with better prognosis who would be better served with close observation. Additionally, novel agents currently approved for later stages of disease are being investigated in earlier stages, such as sipuleucel-T. As more trials in this setting are planned, it is increasingly important to select patients with more aggressive disease features and subsequent event rates that will be adequate to design appropriate clinical trials with novel agents. Alternatively, the ability to accurately identify those with indolent disease allows us to more confidently direct these patients toward clinical trials studying nutraceuticals and less toxic agents, such as sulforaphane, which may offer lower risk of long-term morbidity.

Metastatic CSPC

Metastatic CSPC (mCSPC), as the name suggests, is defined by the presence of detectable metastases, either nodal, visceral, or bone lesions, and by responsiveness to ADT. ADT is typically achieved by surgical orchiectomy or administration of a gonadotropin-releasing hormone agonist or antagonist, but either can also be combined with an oral antiandrogen for a combined androgen blockade approach. Whether this combined approach is superior has been the subject of several meta-analyses, which were in agreement in demonstrating minor differences in favor of the combined approach, although all recommend a balanced discussion with patients about risks and benefits. A natural question is whether combined androgen blockade could be more successful in the current era of more potent hormonal agents such as abiraterone and/or enzalutamide.

Although the hallmark of therapy is ADT, here, too, there was controversy about whether the optimal treatment schedule should be intermittent or continuous. The SWOG 9346 trial recently answered this question by randomizing men with mCSPC to continuous or intermittent ADT. The results from this study were presented at the plenary session of the ASCO 2012 annual meeting and revealed that the primary endpoint of noninferiority was not met when comparing IAD with CAD. In other words, this study does not provide support for IAD in this disease state, although quality of life was superior for patients on IAD.

Several earlier prognostic analyses have been reported from the Southwest Oncology Group clinical trial SWOG 9346, including one reporting that the PSA level after 7 months of ADT was prognostic for survival. Specifically, patients whose PSA level decreased to less than 0.2 ng/mL had a median survival of 75 months. Patients whose PSA level decreased to between 0.2 and 4 ng/mL had a median survival of 44 months, and those whose PSA level never decreased to less than 4 ng/mL had a median survival of 13 months. Another analysis from SWOG 9346 showed a correlation between PSA progression (as defined by an increase of ≥25% greater than the nadir and an absolute increase of at least 2 or 5 ng/mL) and overall survival in more than 1000 men with castration-sensitive disease who were included in the analysis.

These biomarker results from SWOG 9346 have already proved to be useful in differentiating patients by different prognoses within the overly general classification of mCSPC. For instance, the post-ADT initiation 7 months PSA endpoint of PSA of less than 0.2 ng/mL is being used as a “go or no go” decision point in randomized phase II SWOG 0925 trial, testing combined ADT versus combined ADT and cixutumumab (IMC-A12), a human IgG1 monoclonal antibody that targets the insulin-like growth factor I receptor. Hints from preclinical studies provide support that this agent may offer greater benefit specifically in the castration-sensitive environment, validating a surrogate endpoint before launching a larger, randomized phase III trial may prove rational.

Other areas of active investigation within this disease classification include the ongoing CHAARTED (Chemohormonal Therapy versus Androgen Ablation Randomized for Extensive Disease in Prostate Cancer) study, which is designed to determine whether treatment with a combination of chemohormonal therapy, thus an early introduction of docetaxel, offers improved survival compared with ADT alone when tumors are still castration sensitive. Efforts are ongoing to move newer hormonal agents such as abiraterone (anticipated phase III trial) and the upcoming SWOG 1216 trial with orteronel (TAK 700) to this earlier disease state.

Nonmetastatic CRPC

This disease state is defined by a sequence of increasing PSA levels at a minimum of 1-week intervals with a 2.0 ng/mL minimum starting value without detectable metastases by standard imaging modalities, commonly radionuclide bone and computed tomography scanning. As in the castration-sensitive setting, the distinction between metastatic and nonmetastatic in the castration-resistant setting is largely reflective of the limits of detection rather than of biology, because it is presumed that PSA is produced by microscopic metastases. Nevertheless, this classification based on PCWG2 is used to stratify patients for treatment and clinical trials because of the lack of more robust and informative criteria.

Attempts to improve treatment approaches in this stage of prostate cancer have been disappointing and largely unsuccessful because of 2 main issues: (1) the heterogeneity of patient disease progression to a detectable metastatic lesion and (2) the lack of effective therapeutic interventions and thus infrequent surveillance and limited capture of patients while they are in this “premetastatic state.” Regardless, these trials have provided useful data to help prognosticate and further future clinical trial planning and design.

One trial seeking to determine whether zolendronic acid could delay/prevent development of bone metastasis was unsuccessful because of the unanticipated low number of events. Fortunately, the authors characterized the natural history of nonmetastatic CRPC by evaluating events in the placebo arm of the trial. At 2-year follow-up, the median time to first bone metastases and overall survival were not reached, emphasizing that the most of these men had more indolent disease than anticipated. However, PSAdt was found to be prognostic for time to discovery of bone metastasis, and this information has been helpful in guiding the design of subsequent clinical trials for patients with this clinical disease state. For example, more than half of patients with a PSAdt of less than 6.3 months developed bone metastases or died 1.5 years after randomization. The randomized control trial of atrasentan versus placebo illustrates the important point that invariable disease progression leads to high rates of attrition and underpowered statistical analyses. Moreover, a major consequence of infrequent restaging studies is a high screen-fail rate for studies involving this patient population—up to a third—due to identification of metastatic sites after enrollment.

The development of bone metastases is a major morbidity of late-stage prostate cancer; thus, agents to delay or prevent bone metastases have been the aim of several clinical trials that are discussed separately. However, a remaining issue is when and how frequently bone scans are obtained to evaluate for metastases. In the nonmetastatic castration-resistant state, there is no consensus guideline on the standard PSA threshold above which bone scans are recommended. There is a clear need to balance the benefits of early detection and intervention with the risks of increased cost and potential morbidity.

Optimal management for patients in this clinical state remains an open question. In clinical practice, the addition of secondary hormonal manipulations such as antiandrogens, ketoconazole, and estrogenic therapies have long been used in efforts to impede the androgen axis, although none have been tested in a randomized clinical trial with a primary survival endpoint. Future trials with newer agents in this population hold promise in changing the standard of care and improving survival for patients whose disease burden is limited to microscopic castration-resistant metastases.

Metastatic CRPC

The final disease state is that of metastatic CRPC (mCRPC), which is defined by a sequence of increasing PSA values at a minimum of 1-week intervals with a 2.0 ng/mL starting value, serum testosterone less than 50 ng/dL, and the presence of detectable metastases. The greatest recent advances have been made in the treatment of patients reaching this stage of disease. As is true with other cancers, clinical trials of new agents have focused first on patients with metastatic disease because demonstrating clinical benefit is more achievable within a feasible timeline as a result of their limited prognosis. However, the prognosis for patients with mCRPC is a moving target, largely because of the US Food and Drug Administration (FDA) approval of docetaxel in 2004, followed more recently by sipuleucel-T, cabazitaxel, and abiraterone, with many new promising agents in the clinical trial pipeline. One of the greatest challenges to optimal implementation of these many treatments is parsing out biologically distinct subgroups within the still heterogeneous group of patients who have mCRPC and matching the best treatment with disease biology. Within mCRPC, there is an urgent need for better prognostic, predictive, and response biomarkers, to assist with triaging appropriate drugs and timely and accurate response evaluation. These biomarkers are ever more critical in the era of a rapidly growing armamentarium that includes biologically and molecularly targeted agents.

Currently, the prognosis for patients with mCRPC can be estimated through the use of several measures, including an earlier predictive model incorporating lactate dehydrogenase, PSA, alkaline phosphatase, Gleason score, Eastern Cooperative Oncology Group performance status, hemoglobin, and visceral disease involvement to estimate prognosis. More recently, an updated nomogram has incorporated additional independent prognostic factors, including pretreatment PSAdt. Other prognostic models follow the response to docetaxel because of a significant association observed between tumor response and median overall survival. Additionally, a 30% or greater decline in PSA levels has been associated with a radiographic response. Pain from bone metastases has also been studied in men with mCRPC, in whom worse pain interference scores have been associated with an increased risk of death, and many trials now use pain measures for eligibility criterion and as major trial endpoints. Pain and quality of life measures must continue to be evaluated using validated assessment measures.

Circulating tumor cells (CTCs) have garnered considerable interest as biomarkers of prognosis in disease. Baseline CTCs have been shown to be associated with survival and have been proposed as an intermediate endpoint for survival. CTCs are currently under investigation in multiple large, phase III trials to further validate their utility with in hope of eventually using changes in CTCs as a meaningful surrogate endpoint for survival. Ultimately, the power of CTCs may extend beyond simple enumeration to molecular characterization. The hope is that mechanisms of tumor sensitivity and resistance to specific therapies may be elucidated with a simple tube of blood with CTCs.

None of these measures—response to docetaxel, pain interference scores, or CTCs—are yet routinely used to define distinct clinical disease states with respect to clinical practice or enrollment in clinical trials. Currently, patients with mCRPC are further divided based on whether they are asymptomatic or symptomatic (which can include, but is not limited to, pain) and whether they are have not yet undergone chemotherapy or have received chemotherapy (mainly referring to docetaxel). These latter 2 distinctions are arbitrary and simplistic and, importantly, do not reflect the biology of disease but instead the practical necessity of a minimal criteria for subdividing disease states to prioritize the growing number of treatment options.

The current armamentarium of therapeutics for mCRPC consists of several major classes: androgen ablative agents, immunotherapy, cytotoxic chemotherapy, and bone-targeted agents, including radioisotopes. In addition, novel molecularly targeted agents and antiangiogenic agents are in phase II/III testing, and results are eagerly anticipated. Each of these are discussed in further detail in other articles in this issue.

Historically, the cytotoxic chemotherapeutics have included mitoxantrone and estramustine. However, the landscape changed in 2004 when docetaxel was approved for mCRPC by the FDA following results of 2 phase III, multi-institution, randomized trials. Cabazitaxel, another taxane, was approved for treatment of patients who received prior docetaxel in 2010 following the phase III trial of cabazitaxel plus prednisone compared to mitoxantrone plus prednisone in hormone refractory metastatic prostate cancer (TROPIC). Cabazitaxel is now being directly tested as a first-line cytotoxic directly against docetaxel.

Sipuleucel-T is a first-in-class, autologous dendritic cell therapeutic vaccine designed to prime the T-cell response against prostatic acid phosphatase, a prostate cancer antigen. The phase III trial of Sipuleucel-T active cellular immunotherapy treatment of metastatic prostate cancer after failing hormone therapy (IMPACT) demonstrated that at median follow-up of 34 months, patients in the sipuleucel-T arm had a 4.1-month improvement in median survival, leading to its FDA approval in 2010. Of note, a difference in radiographic progression-free survival by an independent review board was not observed, highlighting the limitations of our surrogate endpoints, particularly with newer biologically driven therapies. Other immunotherapies are also being investigated for mCRPC, including ipilimumab, a monoclonal antibody recently FDA approved for metastatic melanoma, and Prostvac-VF, a therapeutic PSA-targeted poxviral vaccine. Phase III studies are under way to further test these immunotherapies.

Considerable excitement surrounds the development of several new agents aimed at further blocking androgen receptor signaling at the level of the prostate cancer tumor cells—either by inhibiting androgen synthesis or by blocking the androgen receptor. Abiraterone was the first agent to be FDA approved and is part of the first group, blocking synthesis of androgen through inhibition of the steroidogenic enzymes 17,20-lyase and 17α-hydroxylase. The phase III trial of abiraterone acetate in castration-resistant prostate cancer previously treated with docetaxel-based chemotherapy (Cougar 301) was performed in patients who had been previously treated with docetaxel, and results showed that abiraterone with prednisone increased overall survival compared with placebo with prednisone. The follow-up phase III trial of abiraterone acetate in chemotherapy-naive patients with metastatic castration-resistant prostate cancer (Cougar 302) was conducted to determine benefit in asymptomatic and minimally symptomatic patients with mCRPC before chemotherapy. The Independent Data Monitoring Committee unanimously recommended unblinding based on a planned interim analysis of progression-free survival, overall survival, and clinical benefit, which was reported at the ASCO 2012 annual meeting. At 43-month follow-up, neither median progression-free or overall survival in the abiraterone arm had been reached, but the hazard ratio for progression-free survival was reported as 0.43 (95% confidence interval, 0.35–0.52), and for overall survival, it was 0.75 (95% confidence interval, 0.61–0.93) in favor of abiraterone. Another newer hormonal agent garnering much excitement is enzalutamide (MDV3100), which has been tested in the phase III trial of oral MDV3100 in patients with progressive castration-resistant prostate cancer previously treated with docetaxel-based chemotherapy (AFFIRM) and demonstrated a median overall survival of 18.4 months for enzalutamide versus 13.6 months in the placebo arm. In addition, results are anticipated in the phase III trial of MDV3100 in chemotherapy-naive patients with progressive metastatic prostate cancer who have failed androgen deprivation therapy (PREVAIL). Close on the heels are several other agents aimed at blocking androgen synthesis including orteronel (TAK700) and ARN 509.

Bone-targeted agents such as zoledronic acid and, more recently, denosumab are already in widespread clinical use to prevent skeletal-related events for patients with bone metastases. Joining the arsenal of bone-targeting agents is ²²³ Ra-chloride, which is an alpha emitter that has a limited range compared with beta or gamma emitters, thus minimizing bone marrow toxicity. In the phase III trial of alpharadin in the treatment of patients with symptomatic hormone refractory prostate cancer with skeletal metastases (ALSYMPCA), of 922 patients with prostate cancer bone metastases, ²²³ Ra significantly improved overall survival compared with placebo (median 14.0 vs 11.2 months) and demonstrated a low incidence of myelosuppression.

Several molecularly targeted agents including tyrosine kinase inhibitors such as dasatinib and, more recently, cabozantinib are in later-stage clinical trials. In addition, the antisense molecules OGX-011 (targeting clusterin) and OGX-427 (targeting heat shock protein 27) and antiangiogenic agent tasquinimod are also the subjects of intense study in the field.