Prostate Cancer Resource Center
Systemic Management for Nonmetastatic Castration-resistant Prostate Cancer A Systematic Review and Network Meta-Analysis

American Journal of Clinical Oncology

Zefu Liu, MD, Tong Zhang, MD, Zikun Ma, MD, Shiyang Zheng, MD, Junying Chen, MD, Zhiming Wu, MD, Xia Zheng, BS, Xiangdong Li, MD, and Zhuowei Liu, MD, PhD



To indirectly compare the efficacy and safety of systemic therapies used for patients with nonmetastatic castration-resistant prostate cancer (nmCRPC).


The relevant randomized controlled trials were retrieved from PubMed and the Cochrane Library. Network meta-analyses were used to compare multiple drugs simultaneously for the outcomes of nmCRPC. Direct evidence in trials and indirect evidence across trials were combined by the network meta-analyses to estimate the treatment efficiency.


Eight studies were included in our research. For prostate-specific antigen progression-free survival, the rate of progression was significantly decreased following apalutamide, enzalutamide, bicalutamide+dutasteride, and bicalutamide treatment compared with placebo. Compared with placebo treatment, metastases-free survival was significantly increased in patients who received apalutamide (hazard ratio [HR]: 0.28, 95% confidence interval [CI]: 0.23-0.35), enzalutamide (HR: 0.29, 95% CI: 0.24-0.35), and darolutamide (HR: 0.42, 95% CI: 0.35-0.50). Direct comparison showed significant survival benefits in patients who received second-generation anti-androgen therapy (apalutamide, enzalutamide, and darolutamide: HR: 0.74, 95% CI: 0.61-0.91) compared with patients who received placebo. With respect to metastases-free survival, based on SUCRA analysis, there was 80% and 78% probability that apalutamide and enzalutamide were preferred treatment, while darolutamide was likely to be second-best choice. Compared with placebo, all agents were not associated with significantly higher likelihood of serious adverse events and grade 3 to 4 adverse events.


Our outcomes support equivalent efficacy and similar risk of adverse effects between apalutamide, enzalutamide, and darolutamide, supporting the use of these antiandrogen agents in high-risk of progression nmCRPC.

Radical prostatectomy and radical radiotherapy (RT) are considered curative treatment for patients with clinical localized prostate cancer, with 39.7% to 40% and 31.4% to 36% of patients undergoing the treatments, respectively, in the United States.1,2 However, >30% of patients undergoing radical prostatectomy will have aggressive pathologic features, including positive margin, extracapsular extension, and seminal vesicle invasion, and surgery alone cannot provide adequate locoregional oncological control.3,4 In the United States, adjuvant and salvage RT is administered to 7.8% to 17% and 8.4% of men, respectively, with adverse pathologic features.3,5

Androgen-deprivation therapy (ADT) has played an important role in prostate cancer treatment, and is a standard therapy patients with metastatic disease; moreover, it is the most common treatment after recurrence. However, 10% to 20% of patients eventually become castration-resistant within ~5 years. Treatment of castration-resistant prostate cancer (CRPC) depends on the pivotal presence and absence of metastasis. Among patients with nonmetastasis CRPC (nmCRPC), prostate-specific antigen (PSA) level ≥8 ng/mL, or PSA doubling time (PSADT) ≤10 months was associated with high risk of progression. Since 2010, sipuleucel-T immunotherapy has been considered the first new drug for metastatic CRPC (mCRPC) after receiving approval from the US Food and Drug Administration, and great progress has been made in mCRPC, such as with abiraterone, enzalutamide, cabazitaxel, and radium-233.6 However, ADT (ADT) remained the standard therapy for patients with nmCRPC pre-2018, as recommended by NCCN (National Comprehensive Cancer Network) guidelines, with first-line antiandrogen receptor drugs (flutamide, bicalutamide, nilutamide) for patients with high risk of progression, but without widespread use and with limited efficacy. However, in recent 2 years, the NCCN guideline recommend apalutamide, enzalutamide, and darolutamide for use in patients with nmCPRC based on the findings of the SPARTAN, PROSPER, and ARIMIS clinical trials (category 1), respectively.7 Patients who receive ADT subsequently develop resistance to ADT, driven in part by alteration of AR (androgen receptor) signaling (amplification, mutation, increased AR expression and splice variant expression, increased androgen biosynthesis).8 Apalutamide, enzalutamide, and darolutamide act as second-generation antiandrogens and can sustain their activity in the setting of increased AR expression.9

Unlike the poor prognosis of mCRPC, patients with nmCRPC have a relatively indolent natural history.10 Approximately 1 of 3 patients with nmCRPC develops bone metastasis within 2 years of diagnosis.11 However, some patients with nmCRPC do not receive systematic therapy until they progress to mCRPC. Therefore, quality of life is crucial for patients with nmCRPC, and there is an urgent, unmet need for effective treatment with fewer adverse events (AEs). According to preclinical study, darolutamide have characteristics of low penetration of blood-brain barrier and low binding affinity for γ-aminobutyric acid type A receptors, so, darolutamide might contribute o potential for relative fewer AEs. In addition to second-generation antiandrogen therapy, other target agents have also been tested in clinical trials. However, the majority of the control groups in the above-mentioned clinical trials are placebo. Therefore, there is no effective, direct evaluation between the various targeting agents in terms of AEs and efficacy. Accordingly, we conducted a systematic review of all clinical trials assessing systemic therapy for nmCRPC, and used network meta-analysis (NMA) to compare efficacy and safety outcomes indirectly.


Search Strategy

Relevant randomized controlled trials (RCTs) were retrieved from PubMed and the Cochrane Library (from their inception to December 2018) using the following MeSH terms: (CRPC OR “castration resistant prostate cancer” OR HRPC OR “hormone refractory prostate cancer”) AND (M0 OR “non metastasis” OR nonmetastatic OR “metastasis free”). The publication language was limited to English.

Selection Criteria

Eligible studies met the following inclusion criteria: (1) RCT; (2) all included patients were diagnosed with M0 prostate cancer with castration resistance; (3) described one of the following outcomes: overall survival (OS), metastases-free survival (MFS), BMFS, PSA progression, and AEs (fracture, grade [G] 3 to 4 events, serious events). The definition of PSA progression was according to Prostate Cancer Working Group 2 (PCWG2) criteria.12 MFS was defined as time from randomization to the distance metastases; OS was defined as time from randomization to death. Studies were excluded based on the following exclusion criteria: (1) duplicate reports; (2) nonclinical controlled trial, editorial, letter to the editor, review, meeting abstract; (3) non-English article. All investigators discussed any disagreements during the study selection, and consensus was reached through consultation.

Data Collection and Quality Appraisal

Two authors (Zefu L. and S.Z.) extracted data from the included articles independently, and disputes regarding the extraction were resolved by discussion and final consensus. The following data were extracted: first author, publication year, trial name ( identifier), sample size, investigational product, control drugs, baseline treatment, and the outcomes of interest mentioned above.

The quality of the included trials was assessed using the Cochrane Collaboration tool.13 The following items were used to assess the risk of bias in each trial: allocation (random sequence generation, allocation concealment), blinding (blinding of participants and personnel, blinding of outcome measurement), incomplete outcome data, selective reporting, and other potential bias. The results are presented in graphs constructed using Review Manager 5 (Cochrane Collaboration, Oxford, UK).

Statistical Analysis

NMA were used to compare multiple drugs simultaneously for the outcomes of nmCRPC. Direct evidence in trials and indirect evidence across trials were combined by NMA to estimate treatment efficiency.14 For the survival outcomes, we used Markov chain Monte Carlo methods in OpenBUGS 3.2.3 software15 to build a fixed-effect model Bayesian NMA as previously described.16 The estimated treatment effects are reported as hazard ratios (HRs) and 95% confidence intervals (95% CIs). Fixed-effects NMA of AEs were conducted in GeMTC 0.14.3 (, and the results are presented as odds ratios (OR) and 95% CI. Significance was defined as P<0.05 and 95% CI (according to whether the CI included the null value). The ranking probability of the treatments for each outcome was estimated using the surface under the cumulative ranking curve (SUCRA).17 Direct meta-analysis was conducted using STATA version 12.0 (Stata Corporation, College Station, TX).


Search Results and Characteristics and Risk of Bias of All Included Studies

We identified 395 references, and eventually included 8 studies in the present study.18–25Figure 1 presents the study flow; Table 1 shows the characteristics of the included studies. Intervention compared dutasteride (dual 5-alpha reductase inhibitor), enzalutamide, apalutamide, darolutamide, atrasentan and zibotentan (endothelin [ETA] receptor antagonist), denosumab (RANK antibody), bicalutamide, and placebo. Six clinical trials used placebo as the control; 2 used bicalutamide as the control (Table 1). All patients included in the studies received ADT. Figure 2 presents the risk of bias assessment results. All included studies were randomized, double-blind prospective clinical trials. Figure 2 shows that the quality of included RCTs was moderate to high. Considering allocation, all of the studies adopted random sequence generation, and 6 of 8 studies referenced adequate concealment measures for random alocation of intervention and placebo/control to patients with nmCRPC. Considering blinding, 6 of 8 papers reported that study participants or personnel were blinded to allocated intervention, and 7 of 8 studies described reasonable blinding of outcome investigators. Seven of 8 studies reported important outcomes with adequate follow-up time and thus had a low risk of incomplete data and reporting bias. No indication for other sources of bias was presented, so we judged all the included studies to be low risk for this domain. The study by Miller et al20 was terminated early, and present authors did not analyze the secondary end points. Therefore, that study was deemed potentially high risk for incomplete outcome data and unclear risk of selective reporting. Figures 3 to 5 show network plots of comparison for different end point events. There were 7, 5, and 5 studies for PSA progression (Fig. 3), MFS (Fig. 4), and OS end points (Fig. 5), respectively.

Patient Characteristics

Table 1 presents the ADT statuses before clinical trial recruitment. The history of receiving ADT, median PSA value, and PSADT baselines among the included studies were not balanced. Hussain et al,23 Smith et al,25 Smith et al,18 and Fizazi et al24 clearly stated that the inclusion criteria were only high risk of progression nmCRPC, whereas the other included studies did not specify the progression risk of included patients with nmCRPC.

PSA Progression

The SUCRA indicated that apalutamide, enzalutamide, darolutamide, bicalutamide+dutasteride, bicalutamide only, in order, showed improved possibility of PSA progression-free survival. The PSA progression rate was significantly decreased following treatment with apalutamide (HR: 0.06, 95% CI: 0.05-0.08), enzalutamide (HR: 0.07, 95% CI: 0.06-0.09), darolutamide (HR: 0.13, 95% CI: 0.11-0.16), bicalutamide+dutasteride (HR: 0.23, 95% CI: 0.12-0.45), and bicalutamide (HR: 0.37, 95% CI: 0.25-0.54) compared with placebo, whereas atrasentan did not prolong time to PSA progression (HR: 0.92, 95% CI: 0.75-1.12). Bicalutamide+dutasteride showed higher probability of PSA progression-free survival when compared with bicalutamide only (Fig. 3). On the basis of SUCRA analysis, there was 84%: Figure 3B and 80% probability that apalutamide and enzalutamide had the greatest PSA progression-free survival, respectively. Darolutamide had 63% probability being the second-best treatments.

MFS and OS

MFS (Fig. 4) was significantly increased in patients who received apalutamide (HR: 0.28, 95% CI: 0.23-0.35), enzalutamide (HR: 0.29, 95% CI: 0.24-0.35), and darolutamide (HR: 0.42, 95% CI: 0.35-0.50) as compared to placebo. With respect to MFS, based on SUCRA analysis, there was 80% and 78% probability that apalutamide and enzalutamide were preferred treatment, while darolutamidne was likely to be second-best choice. Although the significant advantage for MFS in apalutamude and enzalutamide did not translate into significant benefits for OS (Fig. 5), direct comparison showed significant survival benefits in patients who received second-generation antiandrogen therapy (apalutamide, enzalutamide, and darolutamide: HR: 0.74, 95% CI: 0.61-0.91) compared with patients who received placebo. Compared with placebo, patients who received atrasentan (HR: 0.91, 95% CI: 0.77-1.09) and zibotentan (HR: 0.89, 95% CI: 0.71-1.12) did not have significantly improved MFS. Compared with darolutamide, apalutamide and enzalutamide had similar effects on OS.

Adverse Events

All studies reported the rate of serious AEs and 7 studies reported the rate of G3 to G4 AEs; however, only 2 studies reported the incidence of fracture, and these indirect paired comparisons are summarized in Figure 6 and Supplemental Figure 1 (Supplemental Digital Content 1,, respectively. The network analysis results for AEs are presented as the SUCRA (Fig. 6) and OR with 95% CI (Supplemental Fig. 1, Supplemental Digital Content 1, Compared with placebo, all agents were not associated with significantly higher likelihood of serious AEs and G3 to G4 AEs. It is highly likely that apalutamide presented a relatively higher risk of fracture when compared with placebo (OR: 2.28, 95% CI: 0.79-6.91) and darolutamide (OR: 1.94, 95% CI: 0.40-9.54), although the difference did not achieve significance. About dizziness, darolutamide showed lower relative risk than apalutamide (OR: 0.74, 95% CI: 0.14-4.01) and enzalutamide (OR: 0.47, 95% CI: 0.09-2.56).


Patients with nmCRPC are comprised of various heterogenous groups. The NMA outcome supported the use of apalutamide, enzalutamide, and darolutamide in the setting of patients with high-risk of progression nmCRPC, when compared with placebo, with the probability of conferring the greatest benefits for MFS. We noted that the results of median MFS are similar between apalutamide (SPARTAN), enzalutamide (PROSPER), and darolutamide (ARAMIS), that is, 40.5, 36.6, and 40.4 months, respectively. However, in our NMA outcome, apalutamide and enzalutamide were associated with significantly improved MFS compared with darolutamide (Fig. 4F). Because of the limitation of relative short follow-up time, the survival data of apalutamide, enzalutamide, and darolutamide are still immature and median OS have not been reached for any of these trials. However, these 3 drugs (apalutamide, darolutamide, enzalutamide) were likely to be preferred treatment for improved OS when compared with placebo after performing direct meta-analysis (Fig. 5G). In addition, many patients receiving ADT with PSA progression have nmCRPC. Apalutamide, enzalutamide, and darolutamide could be options for patients with nmCPRC with high risk, as recommended by NCCN guidelines. However, in our study, we find that bicalutamide+dutasteride likely to be the better choice for PSA progression-free survival than bicalutamide only based on SUCRA analysis. Moreover, in the study conducted by Chu et al,21 the median of patients was only 4.4 and 4.5 ng/mL in the combined and control group, respectively. We believe that this result contributes to the treatment choice of patients experiencing PSA progression whose PSADT was <10 months, although no benefits of OS was observed.

As mentioned above, although great progress has been made in mCRPC treatment before 2018, the relative long life expectancy, that is, median OS of >3 to 5 years, in the observation arm of patients with nmCRPC has limited the development of treatments.18,24,26 Therefore, AEs and quality of life should be considered important factors in treatment planning. Although these 3 drugs (apalutamide, enzalutamide, darolutamide) are nonsteroidal AR antagonists, preclinical studies have suggested that darolutamide potentially has fewer fracture and central nervous system adverse effects than apalutamide and enzalutamide because of its low penetration of the blood-brain barrier.27 In our study, patients on darolutamide showed fewer and less fractures when compared with patients on apalutamide. Impressively, patients on enzalutamide showed a relative higher probability of dizziness and mental impairment disorders when compared with patients on apalutamide and darolutamide (Suplemental Fig. 1, Supplemental Digital Content 1, Importantly, ARAMIS did not exclude patients with history of seizures, in contrast to the SPARTAN and PROSPER trials. Enzalutamide safety was consistent in the PROSPER and STRIVE studies, which reported ≥G3 AEs of 31% and 36%, respectively. Moreover, several clinical trials have investigated the tolerability of enzalutamide in CRPC (PREVAIL,28 TERRAIN,29 AFFIRM30). A meta-analysis of the PREVAIL and AFFIRM trials indicated that enzalutamide presents the risk of neurological effects (RR=1.44, 95% CI 1.31–1.58) and psychiatric disorders (RR=1.43, 95% CI 1.21–1.69).31 However, in a phase IV trial of enzalutamide for patients with least 1 risk of seizures, no increase in seizures was observed.32 Although Saad et al33 demonstrated in post hoc analysis of the SPARTAN study that health-related quality of life was maintained after initiation of apalutamide, the possibility of fracture was greatly improved in patients receiving apalutamide compared with placebo.18 Comparable rates of G3 to G4 and serious AEs for all agents were observed in our study. The outcome of NMA of side effects contributes to selection of drugs for individuals.

Despite the prominent outcomes for apalutamide, enzalutamide, and darolutamide, some points should be addressed. To date, 3 clinical trials have been designed for prolonging the time to MFS in patients with high risk nmCRPC (SPARTAN,18 PROSPER,23 and the ARAMIS24). The indications for apalutamide, enzalutamide, and darolutamide were the precedents for the Food and Drug Administration approval of new prostate cancer drugs based on MFS benefits rather than the gold-standard OS benefits. MFS acting as an intermediate end point aimed at accelerating the approval of new drugs for patients with nmCRPC has been increasingly controversial. Meta-analysis based on individual patient data has validated the correlation of MFS as a strong surrogate of OS in patients with intermediate-risk and high-risk clinically localized prostate cancer.34 However, these data could not be applied to nmCRPC, because metastatic disease may be occulted due to insensitive conventional imaging.35 Therefore, the efficacy of the correlation as applied to patients with high-risk nmCRPC requires more studies.

Subsequently, abiraterone could be a candidate treatment for nmCRPC, and the IMAAGEN study has demonstrated its efficacy, with ≥90% PSA reduction in 59.8% of patients with nmCRPC.36

The 2018 American Urological Association (AUA) guideline amendment suggested that abiraterone could be an alternative therapy for high-risk nmCRPC (grade C) in addition to apalutamide and enzalutamide,37 whereas the NCCN guideline did not recommend it as an option.7 In particular, a recent phase III study demonstrated the noninferiority of low-dose abiraterone (250 mg) combined with low-fat food to standard-dose abiraterone (1000 mg) for mCRPC from the aspects of PSA response and progression-free survival.38 Treatment cost could be reduced remarkably if low abiraterone dosage were applied to nmCRPC.

In addition to antiandrogen therapy, other signaling pathway inhibitors, such as ETA receptor antagonist and RANK (TNF receptor superfamily member 11a) antibody, have been investigated in prostate cancer. Metastatic lesions were exclusively found in the bone in 86% of patients with prostate cancer.39 Zoledronic acid and denosumab were approved for patients with mCRPC for preventing skeletal-related events. However, the fact that zoledronic acid cannot prolong the time to bone metastasis in patients with nonmetastatic prostate cancer and potential renal injury limits its indication in the setting of patients with nmCRPC. Fizazi et al40 have shown that denosumab is better than zoledronic acid for preventing skeletal-related events in patients with CRPC; however, the equivalent effect of denosumab and placebo on OS also prevents their application in patients with nmCRPC, although there was significant improvement of BMFS (NCT00286091).25 When compared with the SPARTAN, PROSPER, and ARAMIS trials, NCT00286091 only included 48% to 52% patients with high risk of bone metastasis (PSA≥8 µg/L or PSADT ≤10 mo, or both), which might have contributed to the negative survival benefits. In the present study, the outcome of OS benefits did not support routine denosumab in patients with nmCRPC when indirectly compared with placebo. ET-mediated signaling plays an important role in prostate cancer progression and metastasis. Specifically, anti–ET-mediated signaling can inhibit prostate cancer growth, invasion, and angiogenesis.41 Zibotentan is a specific ETA receptor antagonist whose efficacy has been validated in phase II clinical trials of patients with mCRPC, improving survival benefits when compared with placebo (HR: 0.55, 95% CI: 0.41-0.73, P=0.008).42 However, the survival benefit was not maintained in phase III clinical trials of patients with nmCRPC20 and CRPC with only bone metastasis,43 which was consistent with the findings from another study involving atrasentan.19 In the present study, the use of zibotentan or atrasentan was not associated with possibility of improved MFS or OS.

Accompanied by increasing new therapies for distant metastases, successful local control of the primary tumor has become important. We noted that 30% to 45% of patients did not receive any radical treatment (radiotherapy or surgery) before diagnosis of nmCRPC.18,20,21,25 Emerging evidence from retrospective data suggest the improvement of clinical outcome for nmCRPC, with favorable 5-year results of local control rate (81% to 93%), CSS (65%), and OS (28% to 67%) after receiving RT.44 However, high-level prospective evidence is needed to determine the efficacy of local treatment.

This systematic review and NMA had several limitations. Given the nature of the indirect comparisons of NMA, it remains a surrogate for head-to-head clinical trials. Nonetheless, the present study outcome indicates that future clinical trials should consider apalutamide, enzalutamide, and darolutamide as standard treatment in the setting of patients with high-risk of progression nmCRPC. The present study addresses the survival benefits of different therapies for nmCRPC, such as antiandrogen, ETA inhibitor, and RANK antibody, but heterogeneities remain between the patients included in these clinical trials in terms of antiandrogen therapy history, risk of progression, and minimum serum PSA. The PROSPER, SPARTAN, and ARAMIS trials included patients with more aggressive features, median PSADT of only 3 to 4 months, nadir PSA≥2 ng/mL (PROSPER), and N1 patients (SPARTAN). In light of the relatively low sensitivity of traditional imaging techniques for detecting metastases, PSMA-PET CT could be a more sensitive imaging test in terms of metastatic sites.45,46 However, none of the included studies took this technique into consideration.


Our outcomes support equivalent efficacy and similar risk of AEs between apalutamide, enzalutamide, and darolutamide, supporting evidence for the use of these second-generation anti-androgen agents in patients with high risk of progression nmCRPC. ETA and RANK antibody do not confer survival benefits for nmCRPC. Our study may provide information on the standard treatment of high-risk of progression nmCRPC for the design of future clinical trials and contributes to selection of drugs for individuals.


1. Mahmood U, Levy LB, Nguyen PL, et al. Current clinical presentation and treatment of localized prostate cancer in the United States. J Urol. 2014;192:1650–1656.

2. Schymura MJ, Kahn AR, German RR, et al. Factors associated with initial treatment and survival for clinically localized prostate cancer: results from the CDC-NPCR Patterns of Care Study (PoC1). BMC Cancer. 2010;10:152–167.

3. Sineshaw HM, Gray PJ, Efstathiou JA, et al. Declining use of radiotherapy for adverse features after radical prostatectomy: results from the National Cancer Data Base. Eur Urol. 2015;68:768–774.

4. Leyh-Bannurah SR, Gazdovich S, Budaus L, et al. Populationbased external validation of the updated 2012 Partin Tables in contemporary North American prostate cancer patients. Prostate. 2017;77:105–113.

5. Morgan TM, Hawken SR, Ghani KR, et al. Variation in the use of postoperative radiotherapy among high-risk patients following radical prostatectomy. Prostate Cancer Prostatic Dis. 2016;19:216–221.

6. Sartor O, de Bono JS. Metastatic prostate cancer. N Engl J Med. 2018;378:645–657.

7. National Comprehensive Cancer Network prostate cancer (Version 4.2019 — August 19, 2019). 2019. Avaliable at: https://wwwnccnorg.

8. Scher HI, Sawyers CL. Biology of progressive, castration-resistant prostate cancer: directed therapies targeting the androgen-receptor signaling axis. J Clin Oncol. 2005;23:8253–8261.

9. Tran C, Ouk S, Clegg NJ, et al. Development of a secondgeneration antiandrogen for treatment of advanced prostate cancer. science. 2009;324:787–790.

10. Aly M, Hashim M, Heeg B, et al. Time-to-event outcomes in men with nonmetastatic castrate-resistant prostate cancer-A systematic literature review and pooling of individual participant data. Eur Urol Focus. 2019;5:788–798.

11. Smith MR, Kabbinavar F, Saad F, et al. Natural history of rising serum prostate-specific antigen in men with castrate nonmetastatic prostate cancer. J Clin Oncol. 2005;23:2918–2925.

12. Scher HI, Halabi S, Tannock I, et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol. 2008;26: 1148–1159.

13. Higgins JP, Altman DG, Gøtzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:889–893.

14. Mills EJ, Ioannidis JPA, Thorlund K, et al. How to use an article reporting a multiple treatment comparison meta-analysis. JAMA. 2012;308:1246–1253.

15. BUGS. OpenBUGS version 3.2.3. Avaliable at: www.openbugs. net/w/FrontPage.

16. Ades AE, Sculpher M, Sutton A, et al. Bayesian methods for evidence synthesis in cost-effectiveness analysis. Pharmacoeconomics. 2006;24:1–19.

17. Salanti G, Ades AE, Ioannidis JP. Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial. J Clin Epidemiol. 2011;64: 163–171.

18. Smith MR, Saad F, Chowdhury S, et al. Apalutamide treatment and metastasis-free survival in prostate cancer. N Engl J Med. 2018;378: 1408–1418.

19. Nelson JB, Love W, Chin JL, et al. Phase 3, randomized, controlled trial of atrasentan in patients with nonmetastatic, hormonerefractory prostate cancer. Cancer. 2008;113:2478–2487.

20. Miller K, Moul JW, Gleave M, et al. Phase III, randomized, placebo-controlled study of once-daily oral zibotentan (ZD4054) in patients with non-metastatic castration-resistant prostate cancer. Prostate Cancer Prostatic Dis. 2013;16:187–192.

21. Chu FM, Sartor O, Gomella L, et al. A randomised, doubleblind study comparing the addition of bicalutamide with or without dutasteride to GnRH analogue therapy in men with non-metastatic castrate-resistant prostate cancer. Eur J Cancer. 2015;51: 1555–1569.

22. Penson DF, Armstrong AJ, Concepcion R, et al. Enzalutamide versus bicalutamide in castration-resistant prostate cancer: the STRIVE trial. J Clin Oncol. 2016;34:2098–2106.

23. Hussain M, Fizazi K, Saad F, et al. Enzalutamide in men with nonmetastatic, castration-resistant prostate cancer. N Engl J Med. 2018;378:2465–2474.

24. Fizazi K, Shore N, Tammela TL, et al. Darolutamide in nonmetastatic, castration-resistant prostate cancer. N Engl J Med. 2019;380:1235–1246.

25. Smith MR, Saad F, Coleman R, et al. Denosumab and bonemetastasis-free survival in men with castration-resistant prostate cancer: results of a phase 3, randomised, placebo-controlled trial. Lancet. 2012;379:39–46.

26. Madan RA, Gulley JL, Schlom J, et al. Analysis of overall survival in patients with nonmetastatic castration-resistant prostate cancer treated with vaccine, nilutamide, and combination therapy. Clin Cancer Res. 2008;14:4526–4531.

27. Zurth C, Sandmann S, Trummel D, et al. Blood-brain barrier penetration of [14C]darolutamide compared with [14C]enzalutamide in rats using whole body autoradiography. J Clin Oncol. 2018;36:345–345.

28. Beer TM, Armstrong AJ, Rathkopf DE, et al. Enzalutamide in metastatic prostate cancer before chemotherapy. N Engl J Med. 2014;371:424–433.

29. Shore ND, Chowdhury S, Villers A, et al. Efficacy and safety of enzalutamide versus bicalutamide for patients with metastatic prostate cancer (TERRAIN): a randomised, double-blind, phase 2 study. Lancet Oncol. 2016;17:153–163.

30. Scher HI, Fizazi K, Saad F, et al. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med. 2012;367:1187–1197.

31. Gracia PR, Dearden L, Antoni L, et al. Meta-analysis of randomized clinical trials in metastatic castration resistant prostate cancer: comparison of hypertension, neurological and psychiatric adverse events on enzalutamide and abiraterone acetate plus prednisone treatment. Ann Oncol. 2016;27(suppl 6):738.

32. Slovin S, Clark W, Carles J, et al. Seizure rates in enzalutamidetreated men with metastatic castration-resistant prostate cancer and risk of seizure: the UPWARD study. JAMA Oncol. 2018;4:702–706.

33. Saad F, Cella D, Basch E, et al. Effect of apalutamide on healthrelated quality of life in patients with non-metastatic castrationresistant prostate cancer: an analysis of the SPARTAN randomised, placebo-controlled, phase 3 trial. Lancet Oncol. 2018;19: 1404–1416.

34. Xie W, Regan MM, Buyse M, et al. Metastasis-free survival is a strong surrogate of overall survival in localized prostate cancer. J Clin Oncol. 2017;35:3097–3104.

35. Yu EY, Miller K, Nelson J, et al. Detection of previously unidentified metastatic disease as a leading cause of screening failure in a phase III trial of zibotentan versus placebo in patients with nonmetastatic, castration resistant prostate cancer. J Urol. 2012;188:103–109.

36. Ryan CJ, Crawford ED, Shore ND, et al. The IMAAGEN study: effect of abiraterone acetate and prednisone on prostate specific antigen and radiographic disease progression in patients with nonmetastatic castration resistant prostate cancer. J Urol. 2018;200: 344–352.

37. Lowrance WT, Murad MH, Oh WK, et al. Castration-resistant prostate cancer: AUA guideline amendment 2018. J Urol. 2018;200: 1264–1272.

38. Szmulewitz RZ, Peer CJ, Ibraheem A, et al. Prospective international randomized phase II study of low-dose abiraterone with food versus standard dose abiraterone in castration-resistant prostate cancer. J Clin Oncol. 2018;36:1389–1395.

39. Hess KR, Varadhachary GR, Taylor SH, et al. Metastatic patterns in adenocarcinoma. Cancer. 2006;106:1624–1633.

40. Fizazi K, Carducci M, Smith M, et al. Denosumab versus zoledronic acid for treatment of bone metastases in men with castration-resistant prostate cancer: a randomised, double-blind study. Lancet. 2011;377:813–822.

41. Bagnato A, Loizidou M, Pflug BR, et al. Role of the endothelin axis and its antagonists in the treatment of cancer. Br J Pharmacol. 2011;163:220–233.

42. James ND, Caty A, Borre M, et al. Safety and efficacy of the specific endothelin-A receptor antagonist ZD4054 in patients with hormone-resistant prostate cancer and bone metastases who were pain free or mildly symptomatic: a double-blind, placebocontrolled, randomised, phase 2 trial. Eur Urol. 2009;55: 1112–1123.

43. Nelson JB, Fizazi K, Miller K, et al. Phase 3, randomized, placebocontrolled study of zibotentan (ZD4054) in patients with castrationresistant prostate cancer metastatic to bone. Cancer. 2012;118: 5709–5718.

44. Beauval JB, Loriot Y, Hennequin C, et al. Loco-regional treatment for castration-resistant prostate cancer: Is there any rationale? A critical review from the AFU-GETUG. Crit Rev Oncol Hematol. 2018;122:144–149.

45. Perera M, Papa N, Christidis D, et al. Sensitivity, specificity, and predictors of positive 68 Ga–prostate-specific membrane antigen positron emission tomography in advanced prostate cancer: a systematic review and meta-analysis. European Urology. 2016;70: 926–937.

46. Fendler WP, Weber M, Iravani A, et al. Prostate-specific membrane antigen ligand positron emission tomography in men with nonmetastatic castration-resistant prostate cancer. Clin Cancer Res. 2019;15:7448–7454.