Non-metastatic castration resistant prostate cancer (M0CRPC) is a heterogenous disease state affecting an estimated 100,000 men in the United States. Development of more sensitive modalities for detection of metastasis has altered the landscape of advanced prostate cancer, but M0CRPC has remained a condition that previously lacked FDA-approved treatment. The emerging data on new generation Androgen Receptor (AR) pathway inhibitors should address this gap in the management of such patients.
We reviewed and summarized the current literature for the definition, diagnosis and treatment of M0CRPC. We highlight the results of recent Phase III trials that show significant impact on the outcomes of M0CRPC.
Androgen deprivation therapy remains the foundation of therapy for M0CRPC. Recently published Phase III trials provided data on improved progression free survival when ADT is augmented with newer AR pathway inhibitors. The SPARTAN trial showed that metastasis-free survival (MFS) for patients treated with apalutamide plus ADT is 40.5 months compared to 16.2 months for patients who received standard ADT plus placebo, a 72% reduction in the risk of distant metastasis or death in apalutamide plus ADT compared to ADT plus placebo. The PROSPER trial demonstrated that MFS for patients treated with enzalutamide plus ADT was 36.6 months compared to 14.7 months for patients who received standard ADT only indicating a 71% reduction in the risk of developing metastatic CRPC or death compared to ADT alone. The ARAMIS trial on darolutamide, another AR pathway inhibitor, is also ongoing, and can potentially be another fitting option for M0CRPC.
The recent Phase III trials SPARTAN and PROPSER demonstrate effective treatment options for the M0CRPC disease state that has historically lacked treatment from high level evidence. In particular, an FDA-approved treatment, Apalutamide, can finally be offered for M0CRPC patients. The newer AR pathway inhibitors should provide a basis for further investigation into treatments for M0CRPC.
In the United States, prostate cancer is the most commonly diagnosed cancer, the third leading cause of cancer related death, and worldwide is the fifth leading cause of cancer related death in males [1, 2]. Prostate specific antigen (PSA) screening, although controversial, has aided in the early detection and diagnosis of prostate cancer [3, 4]. Once diagnosed with prostate cancer, most patients will undergo attempted curative therapy in the form of radiation or surgery, and then undergo monitoring for biochemical recurrence with serum PSA [5, 6]. Approximately 30% of patients who undergo definitive treatment will have recurrence . While the 5-year median survival rate for localized prostate cancer approaches 100%, the ideal treatment of progressing and metastatic disease has remained a subject of investigation .
Upon development of disease recurrence, the standard of care is continuation of androgen deprivation therapy (ADT) . However, patients will still eventually develop a castration resistant state despite castrate levels of testosterone (less than 50 ng/mL), with 10–20% occurring in the first 5 years from initiation of ADT . This disease state is known as castration resistant prostate cancer (CRPC) and is further categorized as metastatic (mCRPC) or non-metastatic (M0CRPC). mCRPC is a lethal disease state with several series reporting a median survival of less than 3 years . Development of resistance is attributed to, among others, alterations in androgen receptor (AR) levels, mutations and function, and prostate cancer intracrine androgen synthesis [9, 11, 12]. The fact that despite castrate levels of serum testosterone, there remains a higher level of intra-tumoral androgens in CRPC compared to hormone-naïve cases, points towards the central role that the AR axis plays in CRPC development [13, 14]. It is with this knowledge that development of treatment for CRPC focuses on the AR axis.
M0CRPC is defined by the National Comprehensive Cancer Network (NCCN) as progression of disease in the absence of radiographic evidence of metastatic disease . The American Urology Association (AUA) defines this patient as Index Patient 1, asymptomatic, non-metastatic CRPC. A more specific definition is provided by the Prostate Cancer Working Group (PCWG) 3; minimum PSA level of 1.0 ng/mL, a rising PSA that is at least 2 ng/mL higher than the nadir PSA with this rise being at least 25% over the nadir PSA, castrate levels of testosterone ( < 50 ng/mL), and without radiographic evidence of metastases . M0CRPC is viewed as a disease state with variable course, with pre-treatment PSA level and PSADT as useful measures of disease aggressiveness. Those with PSA ≥ 8 ng/ml or PSA doubling time PSADT ≤ 10 months are considered to be at high risk for rapid progression, and may benefit from earlier initiation of ADT, while observation may be employed for those with lower PSA levels and PSADT > 10 months .
An estimated 100,000 men are living with M0CRPC in the United States, with the annual incidence estimated to be 50,000–60,000. Of these men, 34% are expected to progress to mCRPC annually, and nearly 60% progress within 5 years . Without treatment, large phase III trials have shown the median bone metastasis-free survival ranges 25–30 months, with up to 70% of men remaining bone metastases free at 2 years [19, 20]. Prior to progression, patients could benefit from targeted therapy to delay development of metastases . Disease progression impacts patients in many ways, primarily with urinary symptoms, sexual dysfunction, and activity restriction. But often overlooked is the emotional impact that is commonly observed, including frustration or anxiety over the diagnosis, symptoms, and treatment .
To evaluate for metastatic disease, Prostate Cancer Radiographic Assessments for Detection of Advanced Recurrence (RADAR) group guidelines recommend that patients with M0CRPC should be evaluated with conventional imaging when the serum PSA exceeds 2 ng/mL . In the absence of radiographic evidence, patients should be reassessed when serum PSA exceeds 5 ng/mL and again if the serum PSA value doubles . Imaging modalities recommended by the RADAR group for initial assessment include 99mTc bone scintigraphy and abdomen/pelvis/chest CT. Other modalities for consideration include MRI, and 18F-sodium fluoride positron emission tomography (NaF PET). With the introduction of more sensitive imaging modalities, including prostate-specific membrane antigen (PSMA) PET, NaF PET, and 11C-choline PET/CT, the M0CRPC landscape may continue to evolve and measurable metastatic disease can be detected earlier [22,23,24,25,26]. Advancements in imaging, screening, and upcoming therapeutic agents maybe reduce the incidence of M0CRPC .
Treatment for M0CRPC
Previous treatment strategies used for M0CRPC were observation using PSADT with serial imaging until metastasis is documented, or some form of hormonal agent or medications that did not have definite evidence for survival benefit for those with shorter PSADT. Due to lack of guidelines and consensus, enrollment into clinical trials was encouraged.
First generation anti-androgens and hormonal manipulation
Continuing ADT is the most commonly employed strategy for M0CRPC patients. A retrospective review reported a 2- to 6-month median survival advantage in patients with CRPC who had undergone orchiectomy compared with patients who when castration resistance developed, discontinued their LHRH agonists . There is evidence that the AR may independently be oncogenic, despite maximum androgen blockade, and should remain a target for modifying the progression of the disease . Adding an antiandrogen to an existing ADT regimen is an option, with limited survival benefit in only a third of patients and potential for increased rates of side effects .
Paradoxically, antiandrogen withdrawal can also produce a temporary PSA level decrease in a third of those with prior exposure to maximum androgen blockade (MAB). It has been reported that long-term use of antiandrogens can lead to AR modulation, which converts antiandrogens to AR agonists . Interchanging to different GnRH analogues may increase the time to PSA progression, but the clinical impact of this is not proven [30, 31]. Some reports point to the utility of GnRH antagonists in delaying the time to progression, but there is not enough data to support this strategy [7, 32].
Since bone metastasis is frequently the first site of metastasis in prostate cancer patients, bone-targeted agents including clodronate, zoledronic acid, and denosumab have been investigated for use in M0CRPC. So far, none of these agents have shown improvement in overall survival (OS) and thus are not FDA approved for this indication [17, 33]. The lack of durable effect, along with adverse effects has made bone-targeting agents unappealing for treating M0CRPC patients. Two potent bone-targeted agents, endothelin-A receptor antagonsits, atrasentan and zibotentan, were studied to determine their efficacy in preventing development of bony metastases in M0CRPC patients. Trials for both fail to show survival benefit [34, 35].
Use of poxvirus-based PSA vaccine PSA-TRICOM, nilutamide, or a combination in patients with M0CRPC have been reported to improve median survival, but had very limited sample size . Sipuleucel-T, an immunotherapeutic agent designed to elicit a T-cell-mediated response against the prostatic acid phosphatase antigen in PCa cells, has been investigated preliminarily for M0CRPC, and demonstrated a trend of increasing PSADT. The sample size was limited and further investigation has not been pursued. Bevacizumab (Avastin), a humanized monoclonal antibody against vascular endothelial growth factor A (VEGF-A), was also explored for M0CRPC treatment. The authors concluded that bevacizumab therapy had minimal impact on the disease course .
The use of docetaxel has likewise been investigated. In a retrospective study of 98 CRPC patients, 46 patients had M0CRPC, and 52 had distant metastases. As expected, the OS was significantly longer in the M0CRPC group than in the mCRPC group (52.2 months vs. not reached, respectively, P = 0.006). On multivariate analysis, docetaxel use in M0CRPC was a significant predictor for improved OS (P = 0.019). The authors noted the need for prospective studies to further define the role of docetaxel in M0CRPC patients .
Secondary hormonal therapies
Abiraterone acetate is an irreversible CYP17 inhibitor targeting androgen biosynthesis in the testicles, adrenal glands, and within the tumor. Based on increased radiographic PFS and overall survival abiraterone is indicated in mCRPC but not for M0CRPC [40, 41]. The IMAAGEN trial was a phase II, multicenter study that evaluated PSA responses to abiraterone acetate in 131 men with M0CRPC with a PSA ≥ 10 ng/mL or a PSADT of ≤ 10 months. The primary endpoint of the study was PSA response at 6 months. Results appeared promising, in that 87% of patients exhibited a PSA decline of ≥ 50% and 60% of patients had a PSA decline of ≥ 90% at 6 months. Median time to PSA progression was 28.7 months and median time to radiographic progression had not been reached. Toxicities reported were comparable to safety data presented in phase III studies evaluating its efficacy in mCRPC patients[42, 43]. However, no phase III studies are ongoing to further investigate this.
Another potent androgen synthesis inhibitor, TAK-700 (Orteronel) was investigated for M0CRPC . Orteronel preferentially inhibits CYP 17, 20 lyase and precludes the concurrent use of hydrocortisone/prednisone/dexamethasone. It also has fewer fluctuations in mineralocorticoids often seen with abiraterone acetate or ketoconazole. In a phase II study of 38 patients with M0CRPC treated with TAK-700, at 3 months, 16% achieved PSA ≤ 0.2 ng/mL, 76% achieved ≥ 50% decrease, 32% achieved ≥ 90% PSA reduction, and median time to PSA progression was 14.8 months. Currently, no phase III trials are available for this agent.
More contemporary investigations of AR pathway inhibiting agents are showing more promise for M0CRPC treatment. Enzalutamide (Xtandi) is a potent AR ligand-binding domain and signaling inhibitor, which has been approved for use in the mCRPC setting demonstrating prolongation of progression free survival (PFS) and OS [45, 46]. The STRIVE trial was a phase II, double-blinded, randomized controlled trial evaluating the addition of enzalutamide 160 mg versus bicalutamide 50 mg in 396 men with M0 or mCRPC. In this trial, 139 M0CRPC were included, and enzalutamide was shown to reduce the risk of progression or death by 76% compared with bicalutamide (hazard ratio [HR], 0.24; 95% CI, 0.18–0.32; P < 0.001). Median PFS was 19.4 months with enzalutamide versus 5.7 months with the addition of bicalutamide. Enzalutamide also demonstrated significant improvements in time to PSA progression (HR, 0.19; 95% CI, 0.14–0.26; P < 0.001), and proportion of patients with a ≥ 50% PSA response (81% vs. 31%; P < 0.001) . However, the study was deemed underpowered for survival analysis due to small sample size and relatively short follow up, but laid groundwork for further investigation on the role of enzalutamide for M0CRPC.
The PROSPER Trial (NCT02003924) is a double-blind phase III trial, which evaluated enzalutamide in combination with ADT compared to ADT only in 1,401 prostate cancer patients with M0CRPC, with a PSA doubling time ≤ 10 month and PSA ≥ 2 ng/mL at screening . These 1,401 patients were randomized 2:1 to enzalutamide (N = 933, 160 mg once daily) or placebo (N = 468). Median duration of treatment for enzalutamide and placebo was 18.4 and 11.1 months, respectively and median follow up was 22 months. The primary endpoint was metastatic-free survival (MFS) defined as time from randomization to radiographic progression or death within 112 days of treatment discontinuation. The secondary endpoints included time to PSA progression, time to first use of new antineoplastic therapy, OS, safety, and quality of life (QoL). At interim analysis, MFS for patients treated with enzalutamide plus ADT was 36.6 months compared to 14.7 months for patients who received standard ADT only (HR, 0.29; P < 0.0001), with enzalutamide plus ADT resulting in a 71% reduction in the risk of developing metastatic CRPC or death compared to ADT alone. Enzalutamide treatment prolonged both time to first antineoplastic treatment (39.6 vs. 17.7 months; HR, 0.21; P < 0.0001) by 22 months and time to PSA progression (37.2 vs. 3.9 months, HR, 0.07; P < 0.0001), which is a 93% reduction in PSA progression. Because the MFS, time to PSA progression and time to first use of new antineoplastic therapy were all significantly improved at this interim analysis, these results were unblinded and presented. Final OS analysis will be conducted later since the median OS had not yet been reached in either treatment arm at this interim analysis. The current OS results demonstrated a non-significant trend favoring enzalutamide plus ADT over ADT alone (HR, 0.80; P = 0.1519) with a risk reduction of 20%. It is anticipated the OS curve will separate more at the time of final analysis, though crossover treatment following the unblinding can potentially confound OS results and interpretation.
Apalutamide is a potent AR antagonist that selectively binds the ligand-binding domain of AR and blocks AR nuclear translocation or binding to androgen response elements, and can antagonize AR-mediated signaling in AR overexpressing human CRPC cell lines . In a multicenter phase II study, 51 M0CRPC patients with a high risk for progression (PSA ≥ 8 ng/ml or PSA doubling time PSADT ≤ 10 months) received 240 mg/d apalutamide while continuing on ADT. The primary end point was 12-wk PSA response (PCWG 2 criteria), with 89% of patients had ≥50% PSA decline at 12 wk. Secondary end points included safety, and (MFS). Overall, median time to PSA progression was 24.0 months (95% confidence interval [CI], 16.3 mo–not reached [NR]) and median metastasis-free survival was NR (95% CI, 33.4 mo–NR). However, most of the patients discontinued study treatment (n = 33) due to disease progression (n = 11 [22%]) or AEs (n = 9 [18%]). The results of this study paved the way to further investigation on the utility of Apalutamide in M0CRPC setting .
The SPARTAN trial is a double-blind phase III (NCT01946204) trial of apalutamide (Erleada) combined with ADT versus placebo plus ADT in M0CRPC patients who are high risk for developing metastases . The SPARTAN trial randomized 1,207 patients 2:1 to apalutamide plus ADT (N = 806, 240 mg once daily) versus placebo plus ADT (N = 401). The primary endpoint of the SPARTAN trial was MFS, and the secondary endpoints included OS, PFS, time to symptomatic progression, time to cytotoxic chemotherapy, safety, and quality of life (QoL) measures. There were also additional exploratory endpoints, such as second PFS (PFS2) that measured the PFS after patients received second treatment at physician’s choice including open-label abiraterone (Zytiga, Janssen). 48% of patients who discontinued treatment of apalutamide (145/314) received either abiraterone or enzalutamide in subsequent treatment, while 68% (189/279) who discontinued treatment in the placebo arm received either abiraterone or enzalutamide. The MFS for patients treated with apalutamide plus ADT was 40.5 months compared to 16.2 months for patients who received standard ADT plus placebo (HR, 0.28; P < 0.0001). This represents a 72% reduction in the risk of distant metastasis or death with apalutamide plus ADT compared to ADT plus placebo. At the interim analysis, the median OS was not reached in the apalutamide arm, while it was 39.0 months in the placebo control arm (HR, 0.7; P = 0.07). There was no statistically significant difference between the arms, but a 30% risk reduction of death in the apalutamide group. For patients who received subsequent therapies, the PFS2 from the apalutamide arm was not reached, while it was 39.0 months in the placebo arm (HR, 0.49; P < 0.0001), translating to a 51% risk reduction of secondary progression, while also reducing risk of of symptomatic progression by 55%. Looking at SPARTAN Forest plots, patients benefited whether their PSA doubling time was < 6 months or > 6 months. Thus, the FDA did not include PSA doubling time as a prescribing requirement .
Both SPARTAN and PROSPER trials utilized MFS as a primary endpoint. Using MFS allows for more expeditious completion of trials and investigations of therapies. While a long follow-up time leading to the OS is still the ultimate endpoint, MFS has been shown as a predictor for men with biochemically recurrent prostate cancer, and more recently, the the ICECaP Working group has also established MFS as a strong surrogate for OS in localized cancer [52, 53].
Rates of AEs were noted to be similar in both the SPARTAN and PROSPER trials. (Tables 1 and 2) In the PROSPER trial, safety findings were consistent with previous reports of enzalutamide treatment. Any-grade AE rates were 87% and 77% for enzalutamide and placebo, respectively (grade ≥ 3, 31 vs 23%). The Grade 3 or higher adverse events was 31% in patients treated with enzalutamide plus ADT compared to 23% in patients treated with ADT alone. The most common side effects in the treatment group included fatigue (33%), hot flush (13%), hypertension (12%), and nausea (11%). The PROSPER trial noted cardiovascular side effects are of particular concerns for senior patients (aged ≥ 75 years) who had baseline history of cardiovascular diseases. In total 5% of patients treated with enzalutamide experience major adverse cardiac events compared to just 3 % in the placebo group. AEs were a primary reason for treatment discontinuation in 10% vs. 8% in the enzalutamide and placebo arms, respectively . Therefore caution should be given on considering enzalutamide in this subset of patients.
Comparatively, the SPARTAN trial noted any grade AEs in 96.5 and 93.2% for apalutamide and placebo, respectively. Grade 3/4 AE rates in the apalutamide and placebo arms were 45% and 34%, respectively; baseline health-related QoL was maintained in both treatment arms. Rates of treatment discontinuation were 10 and 6.3%, respectively [50, 51]. The most common AEs included fatigue (30.4%) and hypertension (24.8%). Worth noting are some peculiar AEs with Apalutamide. Rash, described as macular or maculo-papular, was reported in a significant number of patients (23.8%). While over 80% were self-limiting, patients must be informed of this potential dermatologic AE. Hypothyroidism was likewise observed in 8.1% of patients, and those with thyroid disorders should be warned of this risk, and consulting endocrinology maybe warranted if considering use of this agent. There was also note of increased risk for falls (15.6%) and fractures (11.7%) in the study, and therefore attention to bone health is necessary. A baseline screening for osteoporosis and consideration for initiation of bisphosphanate treatment may be worthwhile. For both enzalutamide or apalutamide, seizure history or conditions predisposing to seizure should be noted prior to initiating treatment given that both agents report seizures as an AE.
Another trial deemed to have potential is the Phase III ARAMIS trial, an efficacy and safety study of darolutamide (BAY1841788 [ODM-201]) in men with high-risk M0CRPC. This trial compares darolutamide with placebo in the M0CRPC population defined as PSADT < 10 months, PSA > 2 ng/dL. Darolutamide has a mechanism of action similar to the other AR pathway inhibitors, but in addition, has the ability to inhibit some mutant forms of the AR . Darolutamide theoretically may have a lower seizure risk than enzalutamide or apalutamide since it does not cross the bloodbrain barrier. Darolutamide’s primary outcome data collection will be completed in April 2018 and final results potentially reported in 2020.
Overall, the M0CRPC disease space has historically been an unmet need in the treatment of prostate cancer. Continued ADT with or without second line hormonal manipulation has been the standard of care. The lack of treatment based on prospective evidence brought clinical uncertainty for urologists in caring for M0CRPC patients. More importantly, patients were put at a very uncomfortable position of having to receive treatments for a disease state that was not fully understood and therefore had no defined therapy. The results of the two recent trials finally allows for an evidence-based management for M0CRPC patients, with the SPARTAN trial in particular paving the way to an FDA-approved treatment. However, some issues that arise regarding these new therapeutic advancements are worth mentioning. First, given an evidence-based treatment for M0CRPC, placebo-controlled trials may now be considered unethical. Also, it is not definitively known whether earlier treatment at the M0CRPC point is beneficial compared with waiting for therapy at mCRPC, although the suggestion based from current data is yes. This is because the PFS2 in the SPARTAN trial had a 51% benefit and most patients at progression received treatment with abiraterone. While not powered to make this conclusion, it is likely that the benefit exists. The question also arises as to how long the M0CRPC disease space will exist, or how many true “M0” patients with CRPC there really are, in light of new, more sensitive imaging modalities. The PROSPER and SPARTAN trials used standard nuclear imaging modalities to rule out metastasis. Use of more sensitive imaging techniques can decrease the number of patients diagnosed with M0 disease. However, from a clinical standpoint, if there is benefit for the patient to receive newly approved M0 therapies, some of which are not available for metastatic CRPC yet, then it makes sense not to perform more sensitive imaging. Finally, urologists will be increasingly at a place to diagnose M0CRPC in their practice and should become facile with the definitions, diagnosis and treatment of these patients.
Chritopher P. Evans has received research funding, honorarium, consulting and speaking relationships with Medivation, Janssen, Astellas and Sanofi in the past.
About this article
Prostate Cancer and Prostatic Diseases (2018)