Prostate cancer (PC) is among the most common malignancies in men worldwide [1]. Androgen deprivation therapy (ADT) is the gold standard primary treatment for patients with recurrence after radical prostatectomy or radiotherapy, as well as for those with unresectable or metastatic PC [2]. However, most such patients eventually develop a castration-resistant phenotype.

Non-metastatic castration-resistant PC (nmCRPC) often progresses to metastatic CRPC (mCRPC), an incurable disease with poor prognosis [3]. Three previous phase III trials demonstrated significant oncological benefits of novel androgen receptor signaling inhibitors (ARSIs)—enzalutamide, darolutamide, and apalutamide—compared with placebo [4–6]. Although a real-world study reported increased use of novel ARSIs in metastatic hormone-sensitive PC (mHSPC) [7, 8], trends in novel ARSI use for nmCRPC have been sparsely documented [9–11]. Moreover, the safety and oncological benefits of novel ARSIs in real-world nmCRPC settings remain unclear.

The present study was performed to evaluate real-world trends in novel ARSI use and the safety and oncological benefits of these drugs in patients with nmCRPC.

This multicenter retrospective study included 318 consecutive patients with nmCRPC treated between June 2001 and December 2024 at 12 hospitals. nmCRPC was defined as a prostate-specific antigen (PSA) concentration of > 1 ng/mL, castrate testosterone concentration of < 50 ng/dL, and no metastatic lesions on conventional imaging (computed tomography or bone scintigraphy) [12]. The patients were categorized into two groups: those treated with any novel ARSI—enzalutamide, apalutamide, abiraterone acetate, or darolutamide—at any point during nmCRPC treatment (novel ARSI group) and those who were not (control group).

The following variables were analyzed: age at nmCRPC diagnosis, initial PSA concentration, biopsy Gleason score, clinical tumor (T) and node (N) stages at initial diagnosis, history of radical treatment, time of nmCRPC diagnosis, clinical N stage at nmCRPC diagnosis, and PSA doubling time (PSADT). The PSADT was calculated using the Sloan Kettering method, which requires at least three PSA concentrations of ≥ 0.2 ng/mL measured at least 1 month apart within 12 months before the nmCRPC diagnosis [13]. Adverse events (AEs) associated with novel ARSI therapy were assessed using the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0.

All patients in this study received ADT, which included bilateral orchiectomy or luteinizing hormone-releasing hormone agonists/antagonists throughout their PC treatment. In Japan, the clinical use of enzalutamide, abiraterone acetate, apalutamide, and darolutamide for nmCRPC has been available since May 2014, September 2014, March 2019, and January 2020, respectively. The choice of ARSIs for nmCRPC treatment was at the clinician’s discretion.

Of the 231 patients in the novel ARSI group, 137 were treated with first-line ARSIs. The most used ARSI as first-line therapy was darolutamide (n = 66, 48%), followed by enzalutamide (n = 43, 31%), abiraterone acetate (n = 17, 12%), and apalutamide (n = 11, 8.0%). Following the initial approval of novel ARSIs for nmCRPC treatment in Japan in 2014, their use as first-line therapy gradually increased (Fig. 1a). The rate of first-line novel ARSI use among patients diagnosed with nmCRPC between 2020 and 2024 was significantly higher than among patients diagnosed between 2014 and 2019 (68% vs. 33%, P < 0.001) (Fig. 1b).

By 2014, the rate of novel ARSI use in any line reached 69%, and it remained consistently above 60% thereafter (Fig. 1c). Among the 264 patients diagnosed with nmCRPC after 2014, 213 (81%) were treated with novel ARSIs in any line of nmCRPC treatment. However, the rates of novel ARSI use in any line were not significantly different between patients diagnosed with nmCRPC in 2020–2024 and those diagnosed in 2014–2019 (84% vs. 77%, P = 0.113) (Fig. 1d).

This study examined real-world trends in the use of novel ARSIs and their safety and oncological benefits in patients with nmCRPC in Japan. The findings revealed a gradual increase in first-line use of novel ARSIs after 2014, with 81% of patients diagnosed with nmCRPC after 2014 receiving novel ARSIs in any line of treatment. Furthermore, novel ARSIs contributed to significantly improved oncological outcomes while maintaining an acceptable safety profile. These results suggest that novel ARSIs have become a primary treatment strategy in the real-world setting, reflecting their efficacy and safety in the novel ARSI era.

Although three phase III trials demonstrated the significant survival benefits of novel ARSIs in patients with nmCRPC [6, 5, 4], their real-world utilization remains unclear. The present study showed that the rates of first-line novel ARSI use gradually increased after their initial approval for nmCRPC treatment in 2014, reaching 83% in 2022 (Fig. 1a), while rates of novel ARSI use in any line consistently exceeded 60% after 2014 (Fig. 1c). By contrast, several studies reported underutilization of novel ARSIs in nmCRPC. Swami et al. conducted a retrospective study in the United States, revealing that only 21% of patients received first-line novel ARSIs, even among those with a PSADT of ≤ 4 months [9]. Similarly, Wang et al. reported that only 6.4% of patients with nmCRPC in China received novel ARSIs based on electronic medical records [10]. Moreover, a retrospective observational study investigating real-world treatment patterns among veterans with nmCRPC demonstrated that only 13% of patients received novel ARSIs in first-line treatment [11]. Although the use of novel ARSIs gradually increased over time [11, 9, 14], its rate remained < 25% in 2018–2020 [9]. Because ARSI therapy is expensive, costing approximately $10,000 per month [15], differences in the rates of novel ARSI use between Japan and other countries might be influenced by health insurance systems. In Japan, the entire population is covered by a universal health insurance system with a maximum copayment of 10–30%. Additionally, a high-cost medical expense benefit system in Japan may further promote the use of novel ARSIs [16]. Beyond cost, patient characteristics, including age, frailty, and comorbidities, may influence treatment selection [8]. Further studies are needed to evaluate global trends in the use of novel ARSIs in patients with nmCRPC.

The present study demonstrated less frequent AEs and lower discontinuation rates (Fig. 2A) than those reported in three phase III trials [4, 6, 5], likely reflecting differences between clinical trial settings and real-world data. However, the rates observed in this study were consistent with another real-world study. A large retrospective cohort study in the United States found that among 870 patients with nmCRPC treated with novel ARSIs, 24.9% experienced any-grade of AEs, and 10–15% discontinued ARSI treatment because of AEs [17]. By contrast, other real-world studies reported higherAE rates, ranging from 52–57% [18, 19]. These discrepancies may result from differences in evaluated agents, treatment lines, and ARSI dosing. Overall, novel ARSIs appear to be safely used in nmCRPC, with acceptable AE and discontinuation rates.

Although previous studies focusing on patients with mHSPC or mCRPC demonstrated the real-world effects of novel ARSIs on oncological outcomes [20–22], data on their effects in nmCRPC in real-world settings remain sparse. Our earlier study, which included a relatively small cohort (n = 178), showed an improved MFS and OS in patients treated with novel ARSIs. The present study, with an expanded database of 318 patients and a median follow-up period of 46 months, confirmed the significant survival benefits of novel ARSIs (Table 2). Notably, while patients in the control groups of the three phase III trials received ADT alone [6, 5, 4], 92% of patients in the control group of this study were treated with vintage hormone therapies and/or chemotherapy. By contrast, a real-world study in Canada involving 233 patients with nmCRPC did not find a prolonged OS in patients treated with novel ARSIs [21]. These discrepancies highlight the need for further studies to evaluate the real-world effects of novel ARSIs on oncological outcomes in patients with nmCRPC.

This study had several limitations. First, its retrospective design precluded definitive conclusions because we were unable to control for selection bias and other unmeasured confounders. Second, the time from nmCRPC diagnosis to initiation of novel ARSI treatment varied among patients, which may have influenced the outcomes. Finally, the study did not compare the safety and efficacy of individual ARSIs.

Novel ARSIs have become a primary treatment strategy for nmCRPC in real-world settings, supported by their demonstrated safety and oncological benefits.