Recently, several drugs have been introduced for the first-line treatment of chemotherapy-naive metastatic castration-resistant prostate cancer (mCRPC), but few studies have compared treatment outcomes directly. This indirect comparison among 10 clinical trials (
Prostate cancer (PCa) is one of the most commonly diagnosed solid organ malignancies in the United States (US) and the third leading cause of cancer death among American men [
Androgen-deprivation therapy is the first treatment strategy used for advanced PCa, but this treatment only slows progression [
Although many clinical trials have investigated first-line treatments for mCRPC, few have compared these drugs directly. Therefore, it is difficult for urologists to decide which drug to use first. Hence, this indirect comparison among clinical trials was performed to assist clinicians as well as PCa researchers planning treatment trials.
The protocol for this indirect comparison was registered at PROSPERO (42017069009). Details of the protocol can be accessed at
Only randomized controlled trials (RCTs) and phase-3 clinical trials (CTs) comparing any of the following six drugs, docetaxel, cabazitaxel, mitoxantrone, abiraterone, enzalutamide, and sipuleucel-T, as initial treatment for mCRPC in adult males (18 years or older) were included. Clinical trials that focused on treatment of patients after failed docetaxel therapy or chemotherapy were excluded.
PubMed, Web of Science, Cochrane Collaboration, and ClinicalTrials.gov were searched to identify relevant studies up to June 29, 2017. Reference lists were also searched for related articles. Titles and abstracts were first used to screen articles, and then full-text reviews were used for the final decision. Quality and bias of publication were assessed using the Cochrane Risk of Bias tool. Two reviewers (H. F. Z. and J. L. C.) completed this process, and all disagreements were settled by a senior author (Y. Q. F.).
Data were first gathered from the results of ClinicalTrials.gov and then updated by the relevant articles. The complete or most recent research report was used when several studies involved the same population.
Primary outcome was overall survival (OS) and secondary outcomes were prostate-specific antigen (PSA) response and adverse events (AEs). In addition, progression-free survival (PFS) or radiographic PFS (rPFS), time to tumor progression (TTP), PSA-progression free survival (PSA-PFS), and health-related quality of life (HRQL) were used in subgroup analyses. Detailed definitions of these outcomes can be found in our online protocol.
Considering the goals of this research, only trials that used the dosages recommended by the EAU were included. A lower dose of cabazitaxel (20 mg/m2) was included because efficacy was similar to a higher dose (25 mg/m2) but with lower toxicity in one study [
Hazard ratio (HR) was used to assess most outcomes (OS, PFS, rPFS, TTP, PSA-PFS, and HRQL) because it provides time-to-event information with adjustment for confounders. If the HR could not be obtained from the research directly, it was estimated using the method described by Tierney et al. [
Indirect comparisons were conducted using WinBUGS version 1.4.3 (MRC Biostatistics Unit, Cambridge, UK). Normal prior distributions, noninformative uniform, and 3 different sets of starting values were used to fit the model. In order to obtain the posterior distributions of model parameters, 150,000 iterations (50,000 per chain) were yielded. For each chain, 20,000 burn-ins and a thinning interval of 10 were used. Efficacies of anti-mCRPC drugs were ranked by calculating the HR or RR compared to placebo, docetaxel, or mitoxantrone. Other outcomes were ranked by the same method.
Review Manager 5.3 (Cochrane Collaboration, Oxford, UK) was used to present the results of indirect comparisons and to conduct a traditional pairwise meta-analysis. A
A total of 2533 potentially relevant articles were identified after the initial database search. After excluding 2510 articles not meeting inclusion criteria, 23 full-text articles which described 10 CTs, including 3 for docetaxel [
Flow diagram of study identification, inclusion, and exclusion.
Table
Characteristics of clinical trials included in this study.
Clinical trail | Design | Sources of patients | Follow-up (median, month) | Intervention | Control | Number of |
Age (median, years |
Gleason score (median) | PSA level (median, ng/ml) | ECOG PS (range) |
---|---|---|---|---|---|---|---|---|---|---|
D9902B |
Phase 3 | 75 centers in the United States and Canada | 34.1 | Sipuleucel-T (3 complete doses/2 weeks) | APC placebo (3 complete doses/2 weeks) | 341:171 | 72:70 | NR | NR | 0 to 1 |
|
||||||||||
D9901 |
Phase 3 | 19 centers in the United States | 36 | Sipuleucel-T (3 complete doses/2 weeks) | APC placebo (3 complete doses/2 weeks) | 82:45 | 73:71 | 7:7 | 46.0:47.9 | 0 to 1 |
|
||||||||||
D9902A |
Phase 3 | 24 clinical trial sites | 36 | Sipuleucel-T (3 complete doses/2 weeks) | APC placebo (3 complete doses/2 weeks) | 65:33 | 70:71 | NR | 61.3:44.0 | 0 to 1 |
|
||||||||||
PREVAIL |
Phase 3 | 207 sites in 22 countries | 31 | Enzalutamide (160 mg/d) | Placebo | 872:845 | 72:71 | NR | 54.1:44.2 | 0 to 1 |
|
||||||||||
COU-AA-302 |
Phase 3 | 12 countries | 49.2 | Abiraterone (1000 mg/d) plus prednisone (10 mg/d) | Prednisone (10 mg/d) | 546:542 | 71:70 | NR | 42:37.7 | 0 to 1 |
|
||||||||||
TAX 327 Tannock et al. | Phase 3 | 24 countries | 20.8 | Docetaxel (75 mg/m2)/3 weeks plus prednisone (10 mg/d) | Mitoxantrone (12 mg/m2)/3 weeks plus prednisone (10 mg/d) | 335:337 | 68:68 | NR | 114:123 | NR |
|
||||||||||
Shen et al. | RCT | Single center | 16.8 | Docetaxel (75 mg/m2)/3 weeks plus prednisone (10 mg/d) | Mitoxantrone (12 mg/m2)/3 weeks plus prednisone (10 mg/d) | 30:31 | 64:66 | NR | 63.45:100.86 | NR |
|
||||||||||
Zhou et al. |
RCT | 15 centers in China | 60 | Docetaxel (75 mg/m2)/3 weeks plus prednisone (10 mg/d) | Mitoxantrone (12 mg/m2)/3 weeks plus prednisone (10 mg/d) | 113:115 | 70.7:70.8 | 8:8 | 70.9:100 | NR |
|
||||||||||
FIRSTANA |
RCT | 159 centers in 25 countries | 51 | Cabazitaxel (20 mg/m2)/3 weeks plus prednisone (10 mg/d) | Docetaxel (75 mg/m2)/3 weeks plus prednisone (10 mg/d) | 389:391 | NR | NR | NR | 0 to 2 |
|
||||||||||
WILLIAM BERRY | Phase 3 | Multicenter in the United States | 21.8 | Mitoxantrone (12 mg/m2)/3 weeks plus prednisone (10 mg/d) | Prednisone (10 mg/d) | 56:63 | 70:74 | NR | 56.7:71.0 | 0 to 2 |
PSA: prostate specific antigen; ECOG PS: Eastern Cooperative Oncology Group Performance Status; NR: not reported; RCT: randomized controlled trial.
Network of indirect comparisons. The size of the nodes indicates the number of patients (listed under the nodes) and line width the number of trials comparing each pair of treatments (listed under the lines).
Results of the OS comparisons, including HRs and ranks, are presented in Figure
Overall survival and PSA response for the included comparisons. The hazard ratio (HR) is used to express differences in overall survival (column treatment versus row treatment), with HR < 1 favoring column treatment. For PSA response, risk ratio (RR) is used (row treatment versus column treatment) with RR < 1 favoring row treatment. Text in red indicates a significant difference (
Enzalutamide demonstrated a higher PSA response rate compared to both mitoxantrone (RR 0.03, 95% CI 0.00 to 0.90) and placebo (RR 0.01, 95% CI 0.00 to 0.11), while no obvious differences were found among other comparisons (Figure
The top 10 most frequent AEs among all CTs (Figure
Top 10 most frequent adverse events (EAs) among all clinical trials. Fatigue ranks first, back pain ranks second, and vomiting ranks last. Risk ratio (RR) is used to express the difference in AEs, with RR < 1 favoring the treatment.
Docetaxel was associated with the highest OS (HR 0.74, 95% CI 0.64 to 0.85), PFS (HR 0.50, 95% CI 0.32 to 0.79), and PSA response rate (RR 0.49, 95% CI 0.11 to 1.76) among the three chemotherapy drugs when compared to mitoxantrone. However, docetaxel was also associated with the most AEs among chemotherapy drugs (RR 0.64, 95% CI 0.15 to 2.71) when compared to mitoxantrone (Figure
Pooled analysis of the chemotherapy subgroup (mitoxantrone, docetaxel, and cabazitaxel). Hazard ratio (HR) is used to indicate differences in overall survival (OS) and progression-free survival (PFS), while risk ratio (RR) is used for PSA response and adverse events (AEs). An HR or RR < 1 favors that treatment group. First rank indicates superior outcome (excluding AEs). PSA: prostate-specific antigen.
Sipuleucel-T showed the highest probability (59.4%) of being the most efficacious for OS improvement among the three nonchemotherapy drugs (HR 0.74, 95% CI 0.61 to 0.89) compared to placebo, while enzalutamide yielded the best PFS (HR 0.32, 95% CI 0.28 to 0.37), the best PSA response (RR 0.01, 95% CI 0.00 to 3.32), and the fewest AEs (RR 0.38, 95% CI 0.02 to 6.87) when compared to placebo (Figure
Pooled analysis of the nonchemotherapy subgroup (abiraterone, enzalutamide, and sipuleucel-T). Hazard ratio is used to indicate differences in overall survival (OS) and progression-free survival, while risk ratio (RR) is used for PSA response and adverse events. An HR or RR < 1 favors that treatment group. First rank means superior outcome (excluding AEs). PSA: prostate-specific antigen.
This indirect comparison of first-line treatments for chemotherapy-naive mCRPC across 10 CTs (4870 patients) suggests that docetaxel has the greatest potential efficacy as indicated by OS, while cabazitaxel shows no obvious difference in efficacy but causes fewer AEs than docetaxel. Of nonchemotherapy drugs included for comparison, enzalutamide shows the highest probability for superior OS and PFS as well as fewest AEs. Therefore, docetaxel is recommended as the first-line chemotherapy and enzalutamide as the first-line nonchemotherapy treatment for advance prostate cancer.
Based on this indirect comparison of multiple chemotherapy and nonchemotherapy drugs, chemotherapy appears to be the best choice for initial treatment of mCRPC, although there was no significant difference among first-line treatments. Docetaxel and cabazitaxel both bind to and stabilize tubulin, inhibiting microtubule depolymerization and resulting in tumor cell cycle arrest and apoptosis [
Docetaxel 75 mg/m2 every three weeks combined with prednisone 5 mg twice daily is one first-line treatment recommended by the EAU, but serious side effects are a substantial issue with this regimen [
Prostate-specific antigen response is commonly used in CTs as an efficacy measure for mCRPC response, but the clinical significance of the PSA response is unclear [
While prolonging life is the primary aim of cancer therapy, treatment decisions must also account for quality of life. Most mCRPC patients have no noticeable tumor-related symptoms initially and might not be suitable for chemotherapy. Therefore, new drugs or nonchemotherapy drugs with better side effects profiles are recommended. In fact, nonchemotherapy drugs such as abiraterone, enzalutamide, and sipuleucel-T are recommended by the EAU as first-line treatments for mCRPC [
The results of our indirect comparison of the AR-related drugs abiraterone and enzalutamide are consistent with previous indirect comparisons [
Results of this indirect comparison of nonchemotherapy drugs are consistent with the latest clinical trials presented at the 2017 American Society of Clinical Oncology (ASCO). A randomized phase-2 cross-over study by Kim N. Chi comparing abiraterone plus placebo versus enzalutamide found no notable difference in time to PSA progression or time to tumor progression. Further, the efficacy of abiraterone was confirmed by several important clinical trials, including LATITUDE and STAMPEDE. With the assistance of androgen-deprivation therapy, abiraterone was found to improve the OS of locally advanced as well as metastatic PCa patients [
Therapy for mCRPC is a systematic process, and all treatments will ultimately fail [
The main limitation of this study is that all comparisons were across CTs, so the evidence level can be regarded as equivalent to a retrospective study. Second, several methodological features of these CTs varied (Table
In conclusion, docetaxel is likely to be the most effective drug for the chemotherapy of mCRPC and so is recommend as first-line treatment, while enzalutamide is the recommended first-line nonchemotherapeutic drug. Further clinical trials are required to confirm these results and to establish the optimal order of drug administration for mCRPC.
The authors declare no conflicts of interest.
Haofeng Zheng, Jialiang Chen, and Wenhan Qiu contributed equally to this paper.
This study was funded by the Science and Technology Planning Project of Guangzhou (Grant no. 201704020052), the Science and Technology Project of Guangdong (Grant no. 2013B021800084), and the Fundamental Research Funds for the Central Universities (Grant no. 14YKPY25).
Supplemental Table 1. The protocol adhered strictly to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement. Supplemental Figure 1. Quality and bias assessment of the included clinical trials. Quality and bias of publications of included articles were assessed using the Cochrane Risk of Bias tool. Supplemental Figure 2. Serious adverse events among all the clinical trials. RR is used for adverse events. RR lower than 1 favors treatment. RR is the risk ratio. Supplemental Figure 3. Pooled analysis of abiraterone and enzalutamide. HRs are used for overall survival and radiographic progression-free survival, PSA-progression free survival, and health-related quality of life. HRs lower than 1 favor treatment group. PSA is the prostate-specific antigen and HRs are the hazard ratios.