In this study, we compared treatment effects between patients who received ADT plus ARSI and those who received ADT alone or ADT plus bicalutamide by adjusting for background factors using propensity score matching. The time to CRPC was significantly longer with ADT plus ARSI than with ADT alone or ADT plus bicalutamide; however, no significant differences were observed in OS. In addition, no significant difference was observed when analyzing the high- and low-volume subgroups and high- and low-Gleason grades subgroups separately. These findings suggest that adding ARSI to ADT prolongs the time to CRPC; however, it does not consistently affect OS. Interestingly, these results are not consistent with those of a globally conducted randomized control trial (RCT) that demonstrated the efficacy of ARSI added to ADT in treating patients with castration-sensitive metastatic prostate cancer [4,5,6].

The results of retrospective studies conducted to determine the efficacy of ARSIs in Japanese patients with castration-sensitive metastatic prostate cancer were reported after 2021. Patients with LATITUDE high-risk disease (meeting at least two of the following three criteria: (i) Gleason score ≥ 8, (ii) presence of ≥ 3 lesions on bone scan, and (iii) presence of measurable visceral lesions) receiving ADT plus abiraterone were compared with those previously treated with ADT plus bicalutamide. Two studies did not report a significant difference in OS [10, 11]. A study that focused only on CHAARTED high-volume disease reported a significantly prolonged OS in patients treated with ADT plus abiraterone or docetaxel [12]. Another study indicated that treatment with ADT plus abiraterone resulted in prolonged OS only in patients with Gleason grade 5 primary lesions [13]. In the largest retrospective multi-institutional study in Japan, named the J-ROCK study, OS was significantly longer in patients receiving any ARSI added to ADT than in those receiving bicalutamide added to ADT; however, the patient backgrounds were not matched [14]. Another study suggested that patients treated with ADT and apalutamide may have better OS than patients treated with bicalutamide alone [15].

These retrospective studies indicate that the efficacy of ARSI in combination with ADT in Japanese patients with castration-sensitive metastatic prostate cancer is controversial. In our retrospective multi-institutional study, OS was not significantly different between the ARSI and ADT groups. An important factor that could explain the lack of observed differences is the approach to treatment initiation. In the ADT group, PSA levels were regularly monitored and ARSI treatment was initiated at the first indication of any increase in PSA. In the global RCT, ARSI treatment was delayed until radiological progression occurred in the ADT group to evaluate radiological progression-free survival. Approximately 90% of patients started ARSI treatment only in PSA progression without radiological progression in ADT group, although the accurate number of patients was not examined in this study. This practice may have mitigated the potential benefits of initiating ADT plus ARSI from the outset, thereby influencing study outcomes.

Global RCTs conducted to determine the efficacy of abiraterone (LATITUDE), apalutamide (TITAN), and enzalutamide (ARCHES) included Japanese patients and reported on the results of subgroup analyses in these patients. All the studies clearly revealed the efficacy of ARSIs in improving OS in all patient subgroups [4,5,6]. However, no significant improvement in OS was observed in the Japanese subgroup analyses [16,17,18]. One reason for this discrepancy could be the inclusion of small number of patients in the Japanese cohort. Another reason is that the OS in the ADT group in the Japanese cohort was better than that in the global cohort. Therefore, the addition of ARSI might not have prolonged the OS further. Previous reports have indicated that patients with metastatic prostate cancer in Japan have better prognoses than those of patients in the United States [7]. The efficacy of ADT for treating prostate cancer is higher in Asians than in Caucasians [19]. Despite advances in treatment strategies, the question regarding whether Japanese patients, who are known to have a better prognosis with ADT, benefit equally from the addition of ARSI to treatment regimens remains unanswered. This gap not only highlights the need for further investigations but also underscores the potential for gaining novel insights into personalized cancer treatment strategies.

As shown in Fig. 2, the median OS tended to be shorter in the ARSI group than in the ADT group (48 months and 64 months, p = 0.42), although the time to CRPC was significantly longer in the ARSI group than in the ADT group. It was suggested that the death from causes other than prostate cancer was higher in the ARSI group. Therefore, the cancer-specific survival was evaluated. Although the median cancer-specific survival time was not reached in both ARSI group and ADT group, the Kaplan–Meier curves show similar trends with those of OS (Supplemental Fig. 1). These results indicated that most of the death in the ARSI group was caused by prostate cancer. The reasons for the discrepancies between time to CRPC and OS might be the differences in the follow-up duration between the groups. The follow-up duration was not enough for evaluating the median OS in the ARSI group.

To explore the characteristics of Japanese patients with metastatic prostate cancer in whom the addition of ARSI was effective, OS was compared in the subgroups based on metastatic volume and Gleason grade between the ARSI and ADT groups after propensity score matching. No significant differences were observed between the high- and low-volume subgroups or between the high and low Gleason grade subgroups. However, the Kaplan–Meier survival curves indicated that the survival rate tended to be better at the short-term follow-up after treatment initiation. Therefore, we compared the 1- and 2-year survival rates. The 1-year survival rate was 6% higher in the high-volume group than in the low-volume group (11.1% vs. 5.1%) and 10% higher in the high Gleason grade group than in the low Gleason grade group (12.2 vs. 2.2%). These results indicate that the addition of ARSI may suppress the short-term risk of death in high-risk subgroups of Japanese patients with metastatic prostate cancer, although the difference was not significant in log-rank test.

This study has several limitations. Because of the retrospective multi-institutional nature of the study, treatment selection, including ADT or ADT plus ARSI and the type of ARSI (abiraterone, apalutamide, or enzalutamide), for each patient differed among institutes and was based on the physician’s preferences. Due to the observational nature of this study, biases originating from unmeasured covariates could not be completely eliminated. In comparing the differences between the treatment groups, especially in the subgroup analyses, the sample size was insufficient to draw definitive conclusions. Moreover, the follow-up duration is not enough to compare the OS in this study. To see the efficacy of ARSI to prolong the prognosis of metastatic prostate cancer patients, further study is needed with longer follow-up duration. However, this is the first study to show the efficacy of adding ARSI to the treatment regimen in subgroups of Japanese patients with metastatic prostate cancer based on metastatic volume and Gleason grade. The results may help in deciding between starting with ADT and adding ARSI at the time of PSA increase or starting with ADT plus ARSI as the first-line therapy for metastatic prostate cancer in Japan.