In 2012, prostate cancer was the fifth most commonly diagnosed malignancy in Korean men [1]. The incidence of prostate cancer is lower in Korea than in western countries [2]. However, a steady increase has been observed in the prevalence of prostate cancer in Korea during the last decade [3]. Progression of prostate cancer leads to metastasis, and most patients respond to androgen deprivation therapy (ADT). However, patients develop resistance to ADT, and this condition is known as metastatic castration-resistant prostate cancer (mCRPC). Metastatic CRPC is defined by progression of cancer despite the effective lowering of serum testosterone levels to castrate levels [4]. Docetaxel plus prednisone is the current standard of care for mCRPC in Korea, but this therapy has toxic effects and leads to the development of drug resistance [5]. Recent findings have shown that androgen receptor signaling plays an important role in CRPC, which suggests that inhibition of this pathway using non-cytotoxic therapies may confer a survival benefit. The results of the international, randomized, double-blind, phase 3, PREVAIL trial in chemotherapy-naïve men showed that enzalutamide improves overall survival (OS) [6].

Results of the PREVAIL showed that treatment with enzalutamide showed positive outcomes on all secondary endpoints, including the time until the first skeletal-related event (SRE), soft-tissue response rate, time until prostatespecific antigen (PSA) progression, and PSA response rate (≥50% decline), as well as patient-reported outcomes [6789]. Enzalutamide was approved in Korea in June 2013 for use in men with CRPC post-chemotherapy based on the results of the AFFIRM trial. In May 2015, the indication was expanded to include men with CRPC regardless of previous exposure to chemotherapy based on results of the PREVAIL trial [10]. This study aimed to evaluate the clinical results of enzalutamide in chemotherapy-naïve mCRPC patients using Korean, multicenter, real-world data and to compare them with the results of the PREVAIL study. In Korea, continuous ADT for mCRPC patients was prohibited until January 2017; subsequently, concurrent ADT was allowed. Thus, we were able to analyze the efficacy of enzalutamide and concurrent ADT in the treatment of patients with chemotherapy-naïve mCRPC. This study will support evidence to the current recommendations for continuing luteinizing hormone-releasing hormone (LHRH) therapy, which can improve the benefits of enzalutamide treatment in patients with chemotherapy-naïve mCRPC.

Demographic features and disease characteristics of patients are shown in Table 1 [610]. Initial results showed that 81.5% of the patients had Gleason score (GS) ≥8 and 33.3% of the patients had European Cooperative Oncology Group Performance Status (ECOG PS) 0. Eighty-nine patients (44.9%) received concurrent ADT. Median follow-up is 9.4 months (interquartile range, 3.9–13.5 months). The overall mortality rate was 12%. The Kaplan–Meier curves for OS and rPFS are shown in Fig. 1 [610]. Median OS, median rPFS, median time until chemotherapy, median time until first SRE was not reached until our analysis (Fig. 1, Table 2) [610]. Median time until PSA progression was 6 months (Table 2) [610]. After enzalutamide treatment, 35.2% of patients received more than one systemic chemotherapy (Table 3) [610]. The median OS was not achieved and 76.7% of patients were alive at 30 months. In a Kaplan–Meier curve with 95% confidence interval (CI), OS was 0.84 (95% CI, 0.76–0.91) at 12 months and 0.77 (95% CI, 0.67–0.87) at 24 months (Supplementary Fig. 1). The results of the univariate Cox proportional hazard analysis for OS showed that concurrent ADT, ECOG PS, and baseline PSA levels were significant predictors of OS (Table 4). Results of multivariate analysis showed that enzalutamide and concurrent ADT significantly improved the OS (hazard ratio, 0.346; 95% CI, 0.125–0.958). The median duration of enzalutamide treatment was 5 months, and the incidence of AEs of grade 3 or higher was 1.7% (Table 5) [610]. The Kaplan–Meier curves for OS, PSA PFS, and rPFS after concurrent ADT are shown in Fig. 2. The OS by 2 years was significantly improved with concurrent ADT (84.6% vs. 71.7%, p=0.015). In addition, patients receiving concurrent ADT showed a significant improvement in PSA PFS (8.0 vs. 4.6 months, p=0.008). However, rPFS at 2 years was not different in patients receiving concurrent ADT (71.8% vs. 68.4%, p=0.154). The waterfall plot of the change in PSA levels in patients receiving concurrent ADT is shown in Fig. 3. To compare Korean real-world data with the result of the PREVAIL study, we described data of PREVAIL study in Table 1, Table 2, Table 3, Table 5 and Fig. 1 [610].

The effects of enzalutamide observed in our study with asymptomatic or mildly symptomatic Korean chemotherapy-naïve patients with mCRPC were consistent with those observed in the study population of the PREVAIL trial [610].

However, the baseline disease characteristics of the subjects in our study were different from those of the subjects in the PREVAIL trial, which may be because of the differences in real clinical practice. A high percentage of Korean patients had a GS of at least 8, which indicates a high disease burden. Our finding was consistent with that observed in the PREVAIL trial in that 50.6% and 81.1% patients receiving enzalutamide among the overall population and in Korean subgroup, respectively, has a GS ≥8 [10]. Some differences were observed in the characteristics of the disease between the real-world Korean patients and the subjects in the PREVAIL trial; these differences may be related to differences in ethnicity. Compared to the overall study population of the PREVAIL trial, a larger proportion of Korean patients had poor PS. A lower proportion of patients in our study had ECOG PS=0 (33.3% vs. 67.0% in the PREVAIL study). In clinical practice, the physicians, patients, and relatives of chemotherapy-naïve patients with mCRPC with poor PS are reluctant to undergo chemotherapy. Therefore, they prefer treatment with enzalutamide despite its high cost before commencing chemotherapy. Thus, this finding may explain why our data showed that a high proportion of Korean patients had poor performance status. Korean patients receiving enzalutamide had a short median time until PSA progression, and poor baseline characteristics and PS may be attributed to this result. In addition to baseline characteristics, concurrent ADT affects the OS. Unlike the patients in the PREVAIL trial, a larger proportion of Korean patients (55%) had did not receive concurrent ADT with enzalutamide in the clinical setting because the Korean medical insurance system prohibited continuous ADT for mCRPC until January 2017. Our results showed that ADT administered concurrently with enzalutamide improved the oncologic outcomes of chemotherapy-naïve patients with mCRPC.

To date, no study has established evidence supporting continuation of LHRH therapy for CRPC [11]. Few studies have investigated the effects of termination of LHRH therapy in patients with CRPC and, therefore, the current European guidelines recommend medical castration [12]. The Korean medical insurance system prohibited continuous ADT for mCRPC because of lack of evidence until January 2017. Continuation of LHRH therapy may cause considerable side effects and unnecessary expenses [13]. Results from previous clinical trials indicate that castration offers only a marginal survival benefit [1415]. Taylor et al. [14] reported the results of a retrospective study, which showed a significant marginal survival benefit for patients continuing an LHRH agonist along with anthracycline, platinum, or ketoconazole treatment for mCRPC. However, a consensus remains to be achieved about the continuation of LHRH treatment. Analysis of 5 Southwest Oncology Group phase II chemotherapy trials conducted in the 1990s failed to show an obvious advantage of continuous ADT in terms of response to cytotoxic chemotherapy or OS in patients with CRPC [15]. The results of another retrospective study of 78 patients failed to show a survival benefit of docetaxel and concurrent ADT in patients with CRPC [16]. However, it could be an underpowered study because of the small number of patients. Lee et al. [16] found that serum testosterone levels did not recover to those observed in noncastrated individuals in most patients after withdrawal of ADT for the entire duration of chemotherapy. A similar phenomenon was observed in men with hormone-sensitive prostate cancer, in which approximately half of the patients who received ADT for a long-term remained castrated at 2.5 years after ADT cessation [17]. Results of previous studies indicate that serum testosterone levels remain low after long-term treatment with LHRH agonists [181920]. Therefore, cessation of LHRH therapy may have only little clinical relevance since serum testosterone levels remain low and may not affect proliferation of cancer cells. However, unlike serum testosterone levels, serum LH levels recover after several weeks to months after cessation of LHRH therapy [181920]. Cessation of LHRH therapy leads to rapid recovery of LH levels after a median of 58 days to 4.5 months, even after long-term medical castration [1819]. Apart from its action on the testes and adrenal glands, LH acts directly on prostate cancer cells via LH-specific receptors and increases the expression of several key steroidogenic enzymes [21]. Administration of abiraterone alone increases the LH [22]. To date, the SPARE trial (NCT02077634) is the only trial that provides further insights into the complex hormonal changes that occur after treatment with abiraterone with or without concurrent LHRH therapy [23]. The SPARE trial is investigating the role of continuation of LHRH therapy when starting treatment with abiraterone in patients with asymptomatic or mild symptomatic chemotherapy-naïve CPRC. To date, no study has investigated the efficacy of enzalutamide and concurrent ADT in the treatment of chemotherapy-naïve mCPRC. Therefore, the results of our trial provide important evidence based on unique data.

This study is subject to the usual limitations inherent with retrospective design and data collection. The follow-up period of this study was relatively short. Further, some missing values are another weakness of the data. The rate of AEs in our study was lower than those in the PREVAIL trial, which may be because of the retrospective study design. The advantages of our study are that we analyzed real-world data from multiple institutions. Furthermore, the primary endpoint was OS, which is relatively objective and less confounded even in a retrospective study. Thus, our results provide novel data about the efficacy of enzalutamide and concurrent ADT for the treatment of men with chemotherapy-naïve mCRPC due to the previous medical environment in Korea, in which continuous ADT therapy for mCRPC patients was prohibited.

A higher proportion of patients in our cohort study showed a GS ≥8 and poor ECOG performance status than those in the PREVAIL study, but the efficacy results in our study were similar to those in the PREVAIL study. Despite the limitations of retrospective studies, the safety was also acceptable. Enzalutamide is effective and safe for chemotherapy-naïve patients with mCRPC in the real-world setting. Further, compared to enzalutamide alone, concurrent administration of ADT with enzalutamide significantly prolonged the survival. Thus, we recommend concurrent administration of ADT with enzalutamide for the treatment of chemotherapy-naïve patients with mCRPC.