Prostate cancer (PCa) is one of the most common malignant tumors in men. With the continuous improvement in living standards and the aggravation of population aging, its incidence ranks second globally (1). Mortality ranks first in men, and fatality is lower than that in lung and colorectal cancers (2). In Asia, the incidence of prostate cancer is increasing every year (3). Some patients are in the middle and advanced stages of the disease when they are first diagnosed. As prostate cancer is an androgen-dependent tumor, androgen deprivation therapy (ADT) and inhibition of the androgen receptor signaling pathway are the main endocrinologic treatments for patients with metastatic prostate cancer (4). However, almost all patients eventually develop castration-resistant prostate cancer (CRPC) 14-30 months after ADT (5). Enzalutamide (Enz), abiraterone (Abi), and other new-generation anti-androgen drugs are more effective. However, most patients will still become resistant to new-generation anti-androgen drugs in the short term. For patients in this period, chemotherapy (such as docetaxel), molecular targeted drugs (such as sunitinib), and DNA damage repair-related poly ADP ribose polymerase inhibitors (such as olaparib) can only delay disease progression to a certain extent. There are currently no breakthrough treatments for CRPC (6, 7).

Bipolar androgen therapy (BAT) is a new treatment for patients with CRPC proposed by the Johns Hopkins University School of Medicine in recent years. The regulation of testosterone (T) between castration and supraphysiological levels suppresses cancer cell growth, thereby reducing prostate-specific antigen (PSA) and delaying castration-resistant prostate cancer progression (8). The opinion that SPA inhibits prostate cancer growth contradicts the mainstream treatment method of ADT. However, its therapeutic effect has often been verified both in vitro and in vivo in animal experiments (9–11). Clinical investigations have shown that BAT is effective (12–14). Thus, BAT is safer and more effective than cytotoxic drugs (15). BAT can restore some patients’ sensitivity to drugs, such as Abi and Enz, showing significant advantages in treating CRPC patients (16). However, there are few BAT clinical trials at present, and most are phase I and phase II clinical trials and a few case reports. There are no large phase III prospective randomized controlled trials on BAT. However, the mechanism of BAT has not yet been fully elucidated. The effectiveness and safety of BAT in clinical settings need to be further verified; there is no clear international consensus or guidance on how and when to use BAT.

Based on the above problems, we systematically analyzed the efficacy and safety of BAT in treating Abi- or Enz-resistant CRPC using a meta-analysis. We explored the best combination or sequence of BAT and current mainstream CRPC treatment drugs to provide reference and guidance for better clinical applications of BAT in the future.

Prostate cancer progression depends on androgens, therefore, reducing androgen levels or blocking the androgen receptor (AR) can inhibit prostate cancer (20). During ADT, prostate cancer cells can regulate AR activity through AR gene amplifications, mutations, and post-translational modifications to adapt to the chronic androgen deprivation environment. However, the castration level of androgens and overexpression of AR leads to increased sensitivity of CRPC cells to supraphysiological levels of androgens, further inhibiting DNA replications and inducing double-strand DNA breaks, inhibiting tumor cell growth, and promoting apoptosis. Based on the different responses of prostate cancer cells to different levels of androgens, BAT inhibits CRPC cells through the sudden rise and fall of androgen levels and rapid alternation and delays the disease progression in the CRPC stage. The results of our meta-analysis also confirmed the effectiveness and safety of BAT for CRPC patients who progressed after Abi or Enz, with a PSA50 rate of 27% and an ORR of 34%. BAT can also trigger the resensitization of CRPC patients to subsequent endocrine therapy; the PSA50 response rate was 67%, and the incidence of AEs (grade ≥3) after BAT treatment was 14%.

Xiong et al (8) proved the safety and effectiveness of this therapy in a previous meta-analysis, which played a guiding role in the promotion and application of BAT. However, the included clinical studies and the total number of people were not many; with the increase in BAT clinical trials, we updated the included BAT clinical studies. There have been a few case reports of BAT performed in individual clinical research centers. The results showed that the PSA50 response rate was significantly increased (67% vs. 54%), and the overall safety and effectiveness were better, further demonstrating the possibility, safety, and effectiveness of BAT in optimizing traditional endocrine therapy.

It is interesting that there are differences in BAT treatment efficacy in patients with previous Abi or Enz, therefore, we grouped the earlier BAT treatment studies with Abi or Enz treatment. We found that the PSA50 rate of patients with previous Abi was 25% (as shown in Figure 2B ), ORR was 25% (as shown in Figure 3B ), the PSA50 response rate was 53% (as shown in Figure 4B ). But in the Enz group, the PSA50 rate of patients was 30% (as shown in Figure 2C ), the ORR was 50% (as shown in Figure 3C ), and the PSA50 response rate increased to 70% (as shown in Figure 4C ) in patients continued to receive Enz after progression. According to the above results, re-sensitivity to novel hormonal therapy (NHA) can be achieved through BAT, thus prolonging the effectiveness of NHA. BAT resulted in a higher PSA50 and ORR in patients treated with Enz than in those treated with Abi. BAT is a better choice for patients and doctors and can be added to the NHA treatment base, in particular, BAT sequential Enz can further improve the curative effect.

In terms of safety, most of the AEs of BAT are grade 1-2, including fatigues, generalized pains, and lower extremity edema (8). Our results suggest that the incidence of AEs (grade≥3) was 14%. Compared to the incidence of AEs corresponding to Enz or Abi (21), BAT can significantly improve a patient’s quality of life. This shows that the BAT regimen can be reasonably applied in patients with advanced CRPC resistant to Abi or Enz.

Simultaneously, BAT can be used to predict patient prognosis. Denmeade et al (22) showed that a peak PSA level of less than 9 ng/ml after BAT treatment was associated with a longer duration of ADT response; PSA≥9 ng/ml (n=16) patients had a median progression-free survival of 20.6 months, while the median overall survival of patients with peak PSA <9 ng/ml was 79.6 months. In current clinical trials, it was also found that not all patients were sensitive to BAT treatment. It is important to identify suitable biomarkers to distinguish BAT responders from non-responders. With the current results, markers of response to BAT are still being explored (23), which is also the direction of our future research.

However, some studies included in this study (such as the case reports of Senji et al (18) and Liu et al (24) may have an impact on the results of this study due to the small number of cases.

The results of this meta-analysis indicate that BAT is a safe and effective treatment for patients who have progressed after Abi or Enz. BAT can trigger the resensitization of patients with CRPC to subsequent endocrine therapy and improve the overall survival of patients and their quality of life. However, biomarkers that can predict the efficacy of BAT still need to be identified.

The original contributions presented in the study are included in the article/ Supplementary Material. Further inquiries can be directed to the corresponding authors.

XY, SH: Protocol development, Data collection or management, Data analysis, Manuscript writing and editing. CY, XW: Manuscript writing and editing. JY, NG: Data collection and management. ZX, CX: Protocol development, Data analysis, Manuscript editing. All authors contributed to the article and approved the submitted version.