Prostate cancer became the leading cause of new cancer cases in the United States and the second leading cause of all cancer-related deaths in 2020, according to Cancer Statistics (1). Advanced prostate cancer has a poor prognosis and is difficult to treat. It can develop into castration-resistant prostate cancer (CRPC) within (1 to 2) years and readily progresses to metastatic CRPC (mCRPC) (2). In recent years, advances in therapeutic drugs have begun to change the treatment philosophy and strategies for this stage of the disease, although high-level evidence is still lacking. In addition, there are still controversies about how to rationally select and use different drugs to improve the overall treatment efficacy (3–7).

Abiraterone acetate is an enzyme inhibitor of CPY17 that inhibits the residual synthesis of androgen after androgen deprivation therapy and can be used for the treatment of mCRPC in patients who have previously received chemotherapy (8, 9). Enzalutamide (MDV3100), an oral drug targeting the androgen receptor signalling pathway, can competitively inhibit androgen receptor binding. Compared with antiandrogen drugs, such as bicalutamide, previously used in clinical therapy, enzalutamide has a 5- to 8-fold greater affinity for the androgen receptor (10, 11). When used for the treatment of advanced or metastatic hormone-sensitive prostate cancer, both abiraterone acetate and enzalutamide have been shown to reduce mortality and improve overall survival (OS) (12–14). It is necessary to maximize the efficacy of drug therapy for mCRPC and improve the OS of patients (6, 15, 16). Therefore, we conducted a meta-analysis to evaluate the efficacy and safety of these two drugs in patients with mCRPC.

We registered the protocol for this systematic review with PROSPERO (CRD42021226808). This report complies with the PRISMA meta-analysis extension statement. The trials covered in this article were registered on internationally recognized clinical trial registries, such as the North American Clinical Trial Registry (www.clinicaltrials.gov). In addition, only clinical data of patients are collected in this paper, without intervention in the treatment plan of patients, which will not bring physiological risks to patients, so there is no need for ethical review.

By calculating and combining HR, our results show that both enzalutamide and abiraterone improve patients’ OS compared with placebo. In addition, enzalutamide was more effective in improving TTPP and rPFS than abiraterone acetate and prednisone/prednisolone combination therapy. There was no significant difference in safety between the two drugs. This finding is significant because it provides substantial evidence for clinical studies in patients with metastatic prostate cancer.

Unlike the study by Wang et al (20) using OR to evaluate the effectiveness of two things, an advantage of this study is the use of combined HR to evaluate the efficacy of AR inhibitors. Compared to the median values for OS, rPFS, and TTPP, HR takes into account both time and queue size (21–23). However, when comparing security, we used RR to compare sAE because we could not obtain more details of the data. In addition, since there have been a large number of meta-analyses comparing abiraterone and enzalutamide, few meta-analyses have included all relevant randomized clinical trials (RCTSs) and directly combined the hazard ratio (HR) of overall survival (OS) and progression-free survival (PFS) for the two drugs. Additionally, in the absence of large RCTs for direct comparison, we used both subgroup analyses and indirect comparisons to evaluate the efficacy and safety of the drugs. Moreover, the study by Wang et al (20) contains multiple studies, but not every study has a sufficient sample size, and there may be regional bias. The clinical trials in this study have a large sample size and come from multiple regions, and the regional and ethnic bias may be smaller. Therefore, the conclusion of this study may be more convincing. First, in the subgroup analysis, there was some heterogeneity in the abiraterone group and the enzalutamide group, which may be caused by the different inclusion criteria of the experiment. In the study by Fizazi K et al (24), patients with mCRPC confirmed histologically or cytologically were eligible if they had previously received docetaxel and had received up to two previous chemotherapy treatments; however, in the study by Miller, K., et al (18), patients must not have received chemotherapy before. Second, another source of heterogeneity may be that the severity of disease and initial PSA levels were not exactly the same in the two studies. In addition, follow-up times were different between the two studies. All of these factors would result in heterogeneity within the abiraterone group. In the study by Devlin, N et al. (7, 25), on the other hand, patients had to have never received cytotoxic chemotherapy; however, in the study by Armstrong, A.J et al. (19, 26, 27), patients needed to be treated with one or two advanced chemotherapy regimens, and at least one regimen had to contain docetaxel. The intragroup heterogeneity of enzalutamide may be derived from this. However, the OS heterogeneity of the two subgroups was very small, so the heterogeneity could also be derived from statistics. The results of the meta-analysis were convincing despite the heterogeneity of the multiple aggregate results. For rPFS, the study by McCool.R et al. (28) uses the method of network analysis to evaluate. However, many comparisons are made through only one experiment, and the sample size of some experiments is too small, and there may be large heterogeneity among the subgroups of the network analysis. As the number of hypothesis tests increases, the rate of false positive errors will increase substantially, even if there is no difference in effect. For the comparison of two drugs, indirect comparison is a better choice. But the conclusion is similar. Enzalutamide is better than abiraterone in improving rPFS.

In addition, we comprehensively explored the safety of abiraterone and enzalutamide and found that AR inhibitors resulted in a higher overall incidence of AE, actually significantly reduced the incidence of high-grade AE, and similar rates of AE leading to death or withdrawal. All patients in the Karin and Kurt trials were assigned to the mandatory use of prednisone to avoid or mitigate adverse events associated with related mineral corticosteroids (18, 24, 29). In contrast, in the Andrew J and Nancy trials (19, 26, 27), enzalutamide did not require the use of prednisone. The safety of abiraterone and enzalutamide appears to be acceptable and manageable, as these AEs can be managed through appropriate medical monitoring. It is important to note that heterogeneity exists between studies, which may be due to differences and heterogeneity in the treatment of abiraterone and enzalutamide. Given the limitations of the study’s reliance on published study results rather than original individual patient data, some important baseline characteristics of patients, namely age, bone injury, visceral disease, ECOG performance status score, and GS, may also play a key role in this large heterogeneity. In addition, the incidence of high-level adverse events (sAEs) in the AR inhibitor group versus the control group was not adequately compared. To reduce potential bias, we only extracted data that strictly met our inclusion criteria, resulting in many AEs being excluded from the analysis. Therefore, in future clinical practice, AR inhibitors should be considered an effective and safe treatment option for patients with CRPC, although practitioners should pay particular attention to the AEs mentioned in our study, especially the high-level AEs. In addition, the use of uniform AE reporting standards for further in-depth data analysis is of great significance to the researchers who carried out the original study.

In the current study, it was significant and interesting for us to determine whether abiraterone acetate and enzalutamide obtained any different benefits in mCRPC through subgroup analysis. This will provide evidence for drug selection in clinical treatment. However, some questions remain unanswered: the most appropriate patient population, potential cross-resistance mechanisms, optimal sequential administration, and possible combination strategies.

There were limitations of this study, such as the limitation of the included studies to those published in English. The references of the included studies were not evaluated, which may have led to the omission of relevant studies. The use of random allocation was described, but the details of the random allocation and the allocation concealment were not included. When extracting the data, some studies did not directly report the effect size and the corresponding confidence interval; therefore, statistical methods were used to calculate the effect size based on the available information, which may have resulted in slightly different results. There are some differences between the results of direct and indirect comparisons; therefore, more prospective studies are needed to verify the findings.

In summary, the current evidence suggests that enzalutamide is not significantly different from abiraterone with regard to improving the OS of mCRPC patients, but it has a greater effect on TTPP and rPFS. The evidence from this study can be used when selecting a treatment option for mCRPC in clinical practice. Due to the lack of a direct comparison, the conclusions drawn from the results of the indirect comparison performed in this analysis need to be verified in high-quality prospective studies.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

HG and ZW conceived this study, conducted the searching, and drafted the manuscript. CC and BL participated in article screening and performed the statistical analysis. CC, BL, and YL checked the data. HG and ZW contributed to the design of this study and provided proposals for the manuscript. All authors contributed to the article and approved the submitted version.

This study was funded by Inner Mongolia Medical University and did not receive external funding.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.