The FDA approval of 5 new therapeutic agents within a 4-year time frame, all of which potentially improve survival for patients with metastatic castration-resistant prostate cancer, is unprecedented in oncology. The diversity of the mechanisms of action of these agents is compelling, including the first and only therapeutic cancer vaccine, the first approved use of an alpha emitter, 2 novel next generation hormonal therapy agents, and a second-generation taxane. However, with this therapeutic bounty comes new challenges in the optimal use of these new and expensive agents in a disease well known for its heterogeneity.
Am J Manag Care. 2014;20:S282-S289For decades following the seminal work of Huggins and Hodges,1 who recognized that testosterone suppression via surgical castration was a highly effective, although noncurative, management option for metastatic prostate cancer, the management of advanced prostate cancer stagnated. A high degree of nihilism was associated with the utility of therapeutics following disease progression in what was historically termed androgen-independent or hormonerefractory disease.2
The first hint of optimism followed the publication of a large phase 3 trial of the cytotoxic agent docetaxel. The randomized trial TAX 327 compared docetaxel plus daily prednisone to the contemporary standard of care, mitoxantrone plus prednisone.3 The study included both weekly and every-3-week docetaxel therapy arms. A survival advantage (median 18.9 months vs 16.4 months) was reported for docetaxel every 3 weeks plus prednisone (but not weekly docetaxel plus prednisone) over mitoxantrone plus prednisone, for a 24% reduction in the hazard of death (P = .009). This trial and a somewhat similar US Intergroup study4 were the first studies to demonstrate a survival benefit with chemotherapy in advanced prostate cancer. With FDA approval, docetaxel administered every 3 weeks became standard therapy for metastatic castration-resistant prostate cancer (mCRPC).
The 6 years following FDA approval of docetaxel was relatively dormant in terms of improvements in patient outcomes. However, during this time, our understanding of the molecular biology of the disease, particularly of the androgen receptor (AR), rapidly increased.5
Metastatic Prostate Cancer: Initial Management
Today, once metastatic disease is documented, typically in the context of a rising prostate-specific antigen (PSA) level after definitive local therapy (eg, radiotherapy/surgery), patients exhibiting evidence of metastases on standard technetium-99m methylene diphosphonate (99mTc-MDP) bone scintigraphy or abdominal/pelvis computed tomography (CT) scans are evaluated for primary androgen-deprivation therapy (ADT). Options for androgen ablation in patients with advanced prostate cancer include bilateral orchiectomy, therapy with luteinizing hormone-releasing hormone (LHRH) analogues and antagonists, and combined androgen blockade (typically a LHRH analogue plus a first-generation anti-androgen). Although orchiectomy remains the historical gold standard for androgen ablation, LHRH therapy is equivalent therapeutically; most US patients are managed with medical therapy, in part because of the psychological implications of surgical castration.6 In men, 5% to 10% of circulating testosterone originates from the conversion of adrenal steroid precursors. Nonsteroidal anti-androgens act at the level of the AR to inhibit the stimulatory effects of testosterone. The role of combined androgen blockade remains somewhat controversial despite numerous randomized studies. Meta-analysis of these published trials provides evidence of a very modest improvement in survival, with some added toxicity and significant expense.7 Many patients who are started on LHRH agonists will receive a 30-day course of a first-generation anti-androgen to lessen the potential of a testosterone flare.8
The side effect profile of ADT includes hot flashes, loss of libido, and an increased risk of osteoporosis, metabolic syndrome, and cardiovascular morbidity.9,10 Thus, the concept of intermittent ADT arose. LHRH agonist therapy is “held” to allow recovery of testosterone; re-administration of therapy begins before clinically apparent symptoms appear. The largest randomized trial to test this management approach in patients with metastatic disease was the US Intergroup study, SWOG 9346.11 In this trial, men with metastatic prostate cancer who received combined androgen blockade for 8 months and achieved a nadir PSA less than 4 ng/mL were randomized to continuous or intermittent therapy with variable time off ADT as determined by a threshold value of serum PSA. The results of the SWOG trial were inconclusive: overall survival of patients in the intermittent arm was noninferior to survival of those treated with continuous ADT.11
Following the results from the US Intergroup trial E 3805, also referred to as CHAARTED (chemo-hormonal therapy vs androgen ablation randomized trial for extensive disease in prostate cancer), a major shift in the management paradigm of a subset of patients with metastatic prostate cancer began to occur.12 Men with newly diagnosed metastatic prostate cancer were stratified, according to the extent of their metastatic disease, into low- and high-volume groups, with the latter defined as having visceral metastases and/or at least 4 bone lesions (including 1 or more bone lesions outside of the vertebral column or pelvis). Over a 6-year period, 790 men were randomized to receive combined androgen blockade with or without 6 cycles of docetaxel administered every 3 weeks without prednisone. At a median follow-up of 29 months, median overall survival improved 13.6 months (hazard ratio [HR] = 0.61; P = .0003). Among patients with high-volume disease, those who received docetaxel had a median overall survival of 49.2 months compared with 32.2 months with ADT alone (HR = 0.60; P = .0006). Among patients with low-volume disease, median survival end points for both groups have not yet been reached. This striking improvement in overall survival for high-volume, advanceddisease patients will clearly change the management paradigm with a much earlier introduction of docetaxel than has been the historic precedent.
Managed Care Implications: Initial Management of Metastatic Disease
The relatively large number of patients diagnosed yearly with metastatic prostate cancer and their need for lifetime therapy have significant implications on initial therapeutic management decisions. Within the managed care environment, at least 4 issues warrant discussion:
1. LHRH agonists versus bilateral orchiectomy
2. Combined androgen blockade versus monotherapy
3. LHRH agonists versus antagonists
4. Potential utility of monitoring serum testosterone levels to guide LHRH agonist dosing
Medical Versus Surgical Castration
Randomized trials have demonstrated the comparability of LHRH agonists and bilateral orchiectomy in the management of metastatic prostate cancer.6 The development of effective medical methods of castration has led to a major shift away from bilateral orchiectomy as the primary testosterone suppression modality; most men prefer to avoid the psychological impact of surgical castration.13 However, compelling evidence suggests the comparative cost-effectiveness of surgical castration over current methods of medical castration. Is it possible to achieve cost savings without compromising patient outcomes by “rediscovering” the potential role of orchiectomy?14,15 A potential approach that obviates some of the psychological impact of standard surgical castration is subcapsular orchiectomy, a surgical method that avoids an empty scrotum without lessening oncolytic effectiveness.16 This approach has been tested in a small randomized trial and found to be comparable to standard bilateral orchiectomy.17
Combined Androgen Blockade Versus Monotherapy
In the 1990s, one of the more important therapeutic aims of metastatic prostate cancer treatment was to minimize as much circulating testosterone as possible. About 10% of circulating testosterone derives from adrenal sources in men. Thus, the addition of a first-generation anti-androgen to standard medical or surgical castration to minimize the impact of additional testosterone production was considered “complete” or “maximal” androgen blockade. A full description of the trials testing this concept is beyond the scope of this review; however, a meta-analysis of 27 randomized studies demonstrated that 5-year survival was improved by 2% to 3%.7 Whether this modest survival benefit has been superseded by the addition of docetaxel in high-volume mCRPC patients or by the impact of much more potent next-generation AR pathway drugs is unknown. Many patients today are given only a 4-week course of a first-generation antiandrogen to prevent LHRH agonists from inducing a testosterone flare. No current method can determine patients who are most likely to benefit from combined androgen blockade. In the 2007 update of the American Society of Clinical Oncology Practice Guidelines for ADT in Prostate Cancer, the expert panel simply stated that “combined androgen blockade should be considered.” 18
LHRH Agonists Versus Antagonists
LHRH agonists work primarily by competitively blocking luteinizing hormone (LH), whereas LHRH antagonists block both LH and follicle-stimulating hormone (FSH). In contrast to LHRH agonists, LHRH antagonists work more rapidly to suppress testosterone, do not have the potential of a testosterone flare, and lead to lower levels of serum testosterone over time.19 From a clinical perspective, however, the lack of a testosterone flare with LHRH antagonist use provides minimal clinical advantage given the ease of blocking the flare with routine, short-term use of a first-generation anti-androgen. The larger and more critical question is whether the use of LHRH antagonists provides improved patient outcomes (ie, survival). To date, although intriguing clinical evidence suggests possible better outcomes, no compelling prospective evidence exists to support this hypothesis. Thus, the higher cost of LHRH antagonists currently appears unsupportable in managed care environments.20,21
Testosterone Level Dosing of LHRH Agonists
A variety of depot formulations of LHRH agonists are in broad clinical use, with most patients receiving 3- or 6-month injections. The utility of monitoring serum testosterone as a guide to determine dosing intervals of LHRH agonists is not a new concept: Oefelein and colleagues22 reported potential economic savings with this approach in the late 1990s. Pathak and colleagues23 at Kaiser Permanente prospectively evaluated 42 patients receiving initial therapy with a 3-month depot LHRH agonist. At the end of 18 months of follow-up, they found that the median dosing interval required to maintain castration levels (≤50 ng/dL) of testosterone was 6 months (range, 5-12 months).
Castration-Resistant Metastatic Prostate Cancer
Despite the unequivocal utility of ADT for men with metastatic prostate cancer, the therapy is not curative, and essentially all men ultimately manifest disease progression, historically referred to as androgen-independent or hormone-refractory disease. Recent advances in the biology of prostate cancer have codified the importance of the AR pathway and its role in the pathogenesis and throughout the disease spectrum, now referred to as the castration-resistant state.24-26 Rapid progress in understanding the biology of advanced prostate cancer has translated over the last few years into unprecedented development and regulatory approval of novel agents for use in men with mCRPC.27-31
Given the rapid introduction of novel agents into the therapeutic armamentarium and the well-recognized heterogeneity of prostate cancer, it is not surprising that the optimal therapeutic paradigm for patients with mCRPC remains undefined. The broad use of docetaxel early in the management paradigm further complicates the decision-making process.
Currently, initial therapy is most commonly administered when patients with mCRPC are asymptomatic, yet have radiographic evidence of bone and/or nodal metastases, typically with a rising PSA and with minimal or no symptoms. Options typically include sipuleucel-T or one of the novel AR-pathway agents, abiraterone or enzalutamide. As the disease evolves, subsequent therapy decisions depend on the patient’s clinical status (symptomatic or asymptomatic), prior therapies received and the quality of response to those agents, and the presence or absence of radiographic evidence of disease progression. Although the utility of docetaxel has been defined for more than a decade, the optimal timing of its use has varied. The availability of newer therapeutic agents, including cabazitaxel and radium-223, add to the complexity of the decision-making process.
Sipuleucel-T is the only FDA-approved therapeutic vaccine in oncology following FDA approval based upon demonstration of a survival benefit in those patients receiving the therapeutic vaccine in the phase III Impact study.30 Utilization of Sipuleucel-T has been sporadic throughout the United States for a variety of reasons, including the delay in production capacity after FDA approval, its poorly understood mechanism of action, and its lack of objective antitumor activity in most patients. Although nearly 20% of patients in the phase 3 trial that led to its regulatory approval received sipuleucel-T following chemotherapy, there is reasonably compelling evidence that the optimal time to administer sipuleucel-T is early in the mCRPC state.30,32 In a post hoc analysis of the phase 3 IMPACT study of sipuleucel- T, Schellhammer and colleagues32 evaluated a range of clinical factors and assessed their association with overall survival. They divided patients’ baseline PSA values into quartiles, demonstrating that patients in the lowest PSA quartile (<22.1 ng/mL) had prolonged median survival (13 vs 2.8 months) compared with those with the highest PSA quartile (>134 ng/mL), providing reasonable clinical evidence for earlier use of the drug.
Although originally described in the early 1990s, abiraterone’s broad clinical development as the first of the next-generation AR-pathway inhibitors is relatively recent. Early phase 2 studies with abiraterone clearly demonstrated significant antitumor activity, and its side effect profile significantly improved when it was given in conjunction with low doses of glucocorticoids to ameliorate the mineralocorticoid side effects associated with its mechanism of action.33
In the randomized trial that would lead to FDA approval of abiraterone, 1195 patients with mCRPC with disease progression following docetaxel were randomized to abiraterone-prednisone or placebo-prednisone. Of the patients entering the study, 70% had previously received 1 previous chemotherapeutic regimen and 30% had undergone 2 previous regimens. Compared with patients receiving placebo-prednisone, those getting active treatment with abiraterone had improved overall survival (14.8 months vs 10.9 months). Secondary end points, including progression-free survival, pain palliation, and PSA response rate were also improved for patients on the abiraterone treatment arm. Therapy was well tolerated; anticipated mineralocorticoid side effects were more prevalent in the abiraterone arm.27
In the pre-docetaxel setting, 1088 men who were asymptomatic or minimally symptomatic and had bone and lymph node (but not visceral) mCRPC were randomized to abiraterone-prednisone or placebo-prednisone regimens with co-primary end points of radiographic progression-free survival (rPFS) and overall survival. At a median follow-up of 22.2 months, abiraterone demonstrated a statistically and clinically significant improvement in rPFS from 8.3 months to l6.5 months (HR = 0.53; P <.001).34 Although there was a trend for increased overall survival for abiraterone, the trial did not meet prespecified statistical significance. An updated analysis of overall survival at a median survival of 27.1 months again trended toward favoring abiraterone (35.3 months vs 30.1 months in the placebo arm; HR = 0.79; P = .015).35 The data from this trial led the FDA to broaden the label of abiraterone to include men across the mCRPC spectrum.
Enzalutamide is a second-generation AR inhibitor that acts by inducing a conformational change in the AR that inhibits nuclear translocation and DNA binding.36 A large phase 1/2 trial in men with mCRPC provided robust evidence of both significant antitumor activity and a favorable toxicity profile.37
The initial phase 3 trial of enzalutamide was conducted in 1199 men with mCRPC with disease progression following docetaxel-based chemotherapy. Patients were randomized in a 2 to 1 ratio to receive enzalutamide or placebo. A planned interim analysis demonstrated a 37% reduction in the risk of death in the enzalutamide arm as compared with placebo (P <.001). The median OS was 18.4 months compared with 13.6 months in favor of enzalutamide. Treatment with enzalutamide was also superior in all secondary end points, including rPFS, time to PSA progression, and quality of life.29
The recently published Prevail trial randomized 1717 chemotherapy-naïve mCRPC patients to enzalutamide or placebo. This trial was halted at the interim analysis based on a statistically significant improvement in both overall survival and rPFS.38 In 2012, the FDA approved enzalutamide for use in patients with mCRPC post docetaxel; the recently published Prevail trial led the FDA to expand the indication for enzalutamide to all settings of mCRPC, akin to indications for abiraterone.
Docetaxel-based regimens were evaluated in 2 landmark randomized phase 3 studies that compared these regimens to mitoxantrone, a cytotoxic agent that had demonstrated a palliative benefit without impact on survival.39 Study TAX 327 was an industry-sponsored phase 3 trial comparing weekly and every-3-week docetaxel regimens to standard mitoxantrone plus prednisone.3 SWOG 9916 compared docetaxel every 3 weeks plus estramustine to mitoxantrone and prednisone, also demonstrating a survival advantage for the docetaxel-containing arm (median 18 months vs 16 months, 20% reduction in the hazard of death; P = .01).4 As mentioned previously, these 2 studies were the first to show a survival benefit with chemotherapy in advanced prostate cancer and, with the subsequent FDA approval, established docetaxel administered every 3 weeks as standard therapy for mCRPC. Long-term followup of TAX-327 patients confirmed the improvement of survival in those receiving docetaxel plus prednisone every 3 weeks. Almost 19% of patients in the docetaxel arm were alive at more than 3 years of follow-up.40
Of note, despite the modest survival benefit seen with docetaxel, this agent has a very high level of clinical activity. It is not uncommon for patients with symptomatic mCRPC to begin to improve following a single dose of therapy. Additionally in the years since docetaxel’s FDA approval, 9 randomized trials comparing docetaxelprednisone to docetaxel and an investigational agent (DN-101, GVAX, bevacizumab, VEGF-Trap, atrasentan, ZD 4054, dasatinib, lenalidomide, OGX-011) have been reported, all failing to demonstrate improvement over this standard therapy.41
Cabazitaxel is a semisynthetic taxane that in preclinical evaluation demonstrated activity in both docetaxelsensitive and docetaxel-resistant cancers. The efficacy and safety of cabazitaxel in combination with prednisone were evaluated in a phase 3 trial of 755 patients with mCRPC who had previously received docetaxel. Patients were randomized to receive daily prednisone with either cabazitaxel or mitoxantrone every 3 weeks for a maximum of 10 cycles. The median survival time of patients in the cabazitaxel group was 15.1 months compared with 12.7 months for patients in the mitoxantrone group. The most frequent grade 3 or 4 toxicity was neutropenia with an increased rate of neutropenia and febrile neutropenia in the cabazitaxel arm of 7.5% and 1.3%, respectively.28 Although cabazitaxel is frequently administered with growth-factor support at the 25 mg/m2 dose, an ongoing phase 3 trial is comparing cabazitaxel at 20 mg/m2 to the FDA-approved 25 mg/m2 dose and may provide evidence of utility of the drug at a lower dose without the need for routine growth factor use (NCT01308580).
Radium-223 is an alpha-particle—emitting radioisotope that appears to exert cytotoxic action by inducing damaging double-strand breaks in the DNA of adjacent cells. As a calcium mimetic, radium-223 targets bone, forming complexes with the bone mineral hydroxyapatite, and preferentially targets areas of high bone turnover.42 Efficacy and safety in early clinical trials led to a randomized, phase 2 double-blind multicenter study of 122 patients with CRPC and bone metastases demonstrating safety and a degree of antitumor activity.43
These results paved the way for a large randomized, double-blind, placebo-controlled study that randomized 921 patients in a 2-to-1 ratio to radium-223 administered every 4 weeks for 6 cycles plus best supportive care versus best supportive care alone. The study was intended to enroll patients whose disease had progressed following docetaxel. However, the study permitted entry to patients who either refused docetaxel or were deemed by their clinicians to be unfit for docetaxel. Patients who received radium-223 had median overall survival of 14.9 months, compared with 11.3 months for patients who received placebo—a 30% decrease in deaths (P <.001). Of note, patients treated with radium-223 experienced a significant delay in the median time to first symptomatic skeletal event (15.6 vs 9.8 months; P <.001). Therapy was generally well tolerated with acceptable hematologic toxicity. Although post hoc analysis of the study suggests that radium-223 treated patients had improvements in pain control, this end point was not prospectively assessed.31 Of note, slightly more than 40% of the patients enrolled in the study were docetaxel naïve. In 2013, following review of this trial, the FDA approved radium-223 with a broad mCRPC label, thus making no distinction between chemotherapy-treated or -naïve patients.
Therapeutic Decision Making in mCRPC
The rapid FDA approval of 5 new therapeutic agents over the course a 3-year span is at the same time unprecedented and extremely challenging. The explosion of new agents without the “infrastructure” clinical trials to provide insight into optimal scheduling/sequencing has led to a period of “on the job” assessment to determine the most rational management approach.
In the early phase of the development of next-generation AR-targeted therapies (lyase inhibitors and secondgeneration anti-androgens), there was hope that because of the divergent mechanism of actions of these agents, sequential use or combinations of these agents would provide significant therapeutic benefit. However, what has become clinically apparent is that these agents have a significant degree of cross-resistance. Although definitive insights into the degree of cross-resistance will require prospectively conducted studies, both broad clinical experience and small retrospective series provide ample evidence that the response rate to either abiraterone or enzalutamide following the use of the other is decreased; in some reports, rather significantly.44,45
In this same timeframe, a number of theoretical mechanisms have been proposed to explain both potential de novo and acquired resistance to abiraterone and enzalutamide, which may in time allow for better decision making for individual patients regarding choice of agent and/or sequence.46,47
Although management approaches to individual patients will vary based on their specific circumstances, the initial management of mCRPC patients who are asymptomatic or minimally symptomatic will typically consist of abiraterone plus prednisone, with or without prior treatment with sipuleucel-T. When progression occurs, more symptomatic patients will receive docetaxel, or in some settings (bone-predominant disease), radium- 223, while asymptomatic or minimally symptomatic patients may receive enzalutamide. Patients with symptomatic progression following docetaxel may receive either radium-223 or cabazitaxel.
Other Bone-Targeted Agents
Until recent evidence demonstrated the impact of radium- 223 on survival, the utility of the available approved bone-targeting agents in treating mCRPC (zolendronic acid and denosumab) was limited to decreasing skeletalrelated events, defined as pathologic fractures, spinal cord compression, the need for surgery or radiation for a symptomatic bone lesion, or hypercalcemia of malignancy.48-50 In a large phase 3 head-to-head comparison trial, 1904 men with mCRPC were randomized to receive a standard dose and schedule of denosumab or zolendronic acid. Patients treated with denosumab had a statistically improved time to first skeletal event compared with patients receiving zolendronic acid.51
Managed Care Implications: Management of mCRPC
Given the rapidity of introduction of 5 new agents, significant heterogeneity in patient management should come as no surprise. Although the costs associated with this evolving therapeutic paradigm are significant, potential areas where management can be optimized with potential cost savings and without impact on patient outcomes exist.
Next-Generation Therapeutics and Symptomatic Skeletal Events Symptomatic skeletal events such as pathological bone fracture, spinal cord compression, orthopedic surgical intervention, or radiation to bone are identified clinically and are different from asymptomatic radiologically detected events. The former are now accepted as more clinically relevant end points.52 Since the FDA approval of zolendronic acid, and subsequently denosumab, a high percentage of patients managed with mCRPC receive these agents on a chronic basis with the incumbent economic costs and some, albeit infrequent, toxicity.
Among the secondary end points tested in the phase 3 trials of abiraterone, enzalutamide, and radium-223, was the impact of these agents on skeletal-related events. As noted above, radium-223 treated patients had statistically and clinically meaningful delay in time to first symptomatic skeletal event. In the post-docetaxel phase 3 mCRPC trial of abiraterone-prednisone, patients treated with abiraterone had a significantly longer median time to occurrence of first skeletal-related event compared with patients treated with prednisone-placebo.53 Similar findings were reported in the enzalutamide phase 3 trial in post-docetaxel mCRPC in which enzalutamide-treated patients had a statistically significant delay in time to first skeletal event.54 Thus, among the questions to be addressed in the near term are whether the routine use of first-generation bone targeting agents should remain a standard of care for patients with bone metastases from mCRPC beyond that achieved by these more potent agents.
Impact of Cross-Resistance With Next-Generation AR-Pathway Inhibitors
Among the more clinically relevant questions (at least in the United States, where both abiraterone and enzalutamide are broadly indicated in mCRPC treatment) is, which drug first? In the long term, absent additional evidence to inform the most appropriate selection of these agents, physician familiarity, the prednisone requirement issue, and others will likely be dwarfed by issues related to cost, both for the patient and healthcare system, and will drive decision making—especially given the anticipated differences in costs due to the more limited patent life for abiraterone. Ultimately, however, improvements in our understanding of resistance mechanisms may increasingly provide clinicians with data to better inform therapeutic choices. Ongoing trials—including a phase 3 trial comparing abiraterone-prednisone plus enzalutamide with enzalutamide (NCT01949337)—will provide some prospective evidence regarding management.
Few other disease states have seen the translation of new insights in disease biology into paradigm-shifting therapeutics in such a brief timeframe as has advanced prostate cancer. Changes in our rapidly shifting healthcare environment add pressure for providing both state-of-theart care and recognition of the increasing costs of care. Appropriate use of all of the recently approved agents will require prospective testing of sequence and combinations with an eye on both improving patient management and controlling costs. The ability to develop and gain regulatory approval for additional new agents will potentially become even more challenging given what appears to be movement away from the use of pre and post-docetaxel as regulatory time points for new drug approval. Despite these challenges, there are opportunities to meld clinical and translational questions with studies designed to inform practice with regard to optimal, cost-effective care.Author affiliation: Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, Cleveland Clinic Lerner College of Medicine, Cleveland, OH, and Case Comprehensive Cancer Center, Cleveland, OH.
Funding source: The activity was supported by educational grants from AbbVie Inc, Astellas Scientific and Medical Affairs, Inc, and Janssen Biotech, Inc., and administered by Janssen Scientific Affairs, LLC, and sanofi-aventis US.
Author disclosure: Dr Dreicer reports serving as a consultant for Astellas Pharma, Janssen Pharmaceuticals, Inc, Medivation, Inc, Merck, Millennium Pharmaceuticals, Inc, and Roche. Dr Dreicer also reports receipt of honoraria from Astellas Pharma.
Authorship information: Concept and design; drafting of the manuscript; and critical revision of the manuscript for important intellectual content.
Address correspondence to: E-mail: Dreicer@ccf.org.