Abiraterone versus bicalutamide in combination with gonadotropin releasing hormone antagonist therapy for high risk metastatic hormone sensitive prostate cancer.

Objectives To compare the ecacy of abiraterone with that of bicalutamide in combination with gonadotropin-releasing hormone antagonist treatment for high risk metastatic hormone-sensitive prostate cancer patients. Methods One hundred and forty-nine patients with high risk metastatic hormone-sensitive prostate cancer at our hospital and aliated hospitals between December 2013 and July 2020 were retrospectively identied. Fifty patients were administered abiraterone (1000mg/day) plus prednisolone (5mg/day) with gonadotropin-releasing hormone antagonist (degarelix) (group A) and 99 patients were administered bicalutamide (80mg/day) with gonadotropin-releasing hormone antagonist (group B). PSA- progression-free survival of group A was signicantly longer than that of group B. Abiraterone therapy and Gleason score were signicant independent factors for PSA-progression-free survival. By propensity score matching, total 56 matched patients were obtained. PSA-PFS (p<0.001) and OS (p=0.0071) of high risk mHNPC patients were signicantly longer in abiraterone group of matched patients. Abiraterone therapy and Gleason score were still shown to be signicant independent factors for PSA-PFS in matched patients. in with better than those of bicalutamide.


Introduction
Though prostate cancer today tends to be diagnosed at early stage due to the introduction of PSA screening, some patients with prostate cancer still present metastasis at the time of diagnosis in Japan 12 . Androgen deprivation therapy (ADT) as systemic therapy is generally accepted as a standard of care for patients with metastatic hormone-sensitive prostate cancer (mHSPC). In Japanese clinical practice guideline for prostate cancer, combined androgen blockade (CAB) therapy, which involves concurrent use of a gonadotropin-releasing hormone (GnRH) agonist and rst-generation antiandrogen such as bicalutamide is recommended as the standard rst-line therapy for metastatic prostate cancer. On the other hand, CAB therapy is not recommended in National Comprehensive Cancer Network guidelines and European Association of Urology guidelines 3 .
Recently, hormonal therapy using abiraterone, next-generation antiandrogen, was reported to improve overall survival (OS) and radiographic progression-free survival (PFS) in men with high risk mHSPC who exhibit at least two of the three following factors: 8 or more Gleason score, 3 or more bone lesions, and the presence of visceral metastasis 4 . Though the superiority of abiraterone plus prednisone with GnRH agonist to placebo with GnRH agonist for treatment of high risk mHSPC had been shown in the report, there are two concerns to apply this result into clinical practice in Japan. First, as stated above, most of metastatic prostate cancer patients receive CAB therapy as rst-line therapy in Japan 5 . Therefore, the effectiveness of abiraterone plus prednisone with GnRH analog should be compared with that of CAB therapy using bicalutamide. Second, GnRH antagonist, degarelix, which doesn't induce a transient rise in testosterone to aggravate the symptoms, instead of GnRH agonist are becoming major component of ADT especially for those who have a high metastatic burden in Japan 6 . However, there has been no reports to compare the e cacy of abiraterone with that of bicalutamide in combination with GnRH antagonist treatment for high risk mHSPC.
The aim of this study was to compare the e cacy of abiraterone with that of bicalutamide in combination with GnRH antagonist treatment for high risk mHSPC.

Patients and Treatments
We retrospectively identi ed 149 patients with high risk mHSPC at our hospital and a liated hospitals in KPUM-Prostate Cancer Study Group between December 2013 and July 2020. All patients have two or more following factors: 8 or more Gleason score, at least three bone lesions, and the presence of visceral metastasis. Fifty patients were administered abiraterone (1000mg/day) plus prednisolone (5mg/day) with GnRH antagonist (degarelix) (group A). Ninety-nine patients who were administered bicalutamide (80mg/day) with GnRH antagonist were categorized as group B.
Bone and visceral metastasis were assessed by bone scintigraphy and computed tomography (CT).
Extent of disease (EOD) score was measured by bone scintigraphy. This study was approved by the institutional review board of Kyoto Prefectural University of Medicine (ERB-C-1071-2) and each a liated hospitals and conducted in compliance with the Declaration of Helsinki. The institutional review board waived individual written informed consent due to the retrospective nature of this study.

Statistical Analysis
The chi-square test and Wilcoxon's rank sum test was used to compare the two groups as appropriate.
Kaplan-Meier analysis was used for estimation of the differences in time events between the two groups using the log-rank test. Cox's proportional hazard models were applied to investigate factors associated with progression free survival. Propensity score matching technique was used to adjust the clinical backgrounds between two groups. Statistical analyses were performed using SAS JMP, Version 14, and P<0.05 was taken to indicate statistical signi cance.

Results
Clinical background of the patients Table 1 shows clinical background of the patients. Fifty patients were administered abiraterone plus prednisolone with GnRH antagonist (degarelix) (group A). Ninety-nine patients were administered bicalutamide with GnRH antagonist (group B). There were no signi cant differences between group A and group B in terms of pretreatment PSA (p=0.554). The patients of group A was signi cantly younger than those of group B (p<0.001). Pretreatment ALP level, Gleason score and EOD score of group A were signi cantly higher than those of group B (p=0.0411, 0.0439 and 0.0402), indicating that patients of group A have more advanced and aggressive disease compared to those of group B. The signi cant shorter observation period of group A (p<0.001) was probably due to the late approval of abiraterone in Japan.
Prognostic factor of PSA progression free survival As shown in Figure S1, PSA-PFS of group A was signi cantly longer than that of group B. It has been reported that age, pretreatment PSA level, Gleason score, EOD score and pretreatment ALP level were prognostic factors associated with PSA control in mHNPC treatment 78 . Therefore, we performed Cox Logistic regression analysis to investigate factors associated with PSA-PFS in patients with high risk mHSPC using variables of those factors in addition to antiandrogen use (abiraterone or bicalutamide).
Abiraterone therapy and Gleason score were shown to be signi cant independent factors for PSA-PFS in high risk mHNPC treatment (Table. 2). However, OS of high risk mHNPC was not signi cantly different between two groups ( Figure S2).
The difference of effectiveness between abiraterone and bicalutamide for high risk mHNPC patients adjusted by propensity score matching technique.
Because several factors associated with PSA-PFS and OS were signi cantly different between two groups (group A and B) (Table1), we next used propensity score matching technique to adjust these clinical backgrounds between two groups to examine the difference for high risk mHNPC patients more precisely. Total 56 matched patients were obtained from 149 patients. As described in Table 3, clinical backgrounds were well adjusted between the two groups. PSA-PFS (p<0.001) and OS (p=0.0071) of high risk mHNPC patients were signi cantly longer in abiraterone group of matched patients ( Fig.1 and 2).
Abiraterone therapy and Gleason score were still shown to be signi cant independent factors for PSA-PFS in matched patients (Table 4).

Discussion
In this study we retrospectively compared the e cacy of abiraterone with that of bicalutamide in combination with GnRH antagonist treatment for high risk mHNPC and found that PSA-PFS and OS in patients who were treated with abiraterone were signi cantly better than those of bicalutamide.
In western countries, CAB therapy is rarely used as the standard rst-line therapy for metastatic prostate cancer, while in Japan it has been widely accepted since several previous reports had suggested that Japanese prostate cancer patients respond to CAB therapy more effectively than other races 9 . For example, it is reported that the adjusted prostate cancer-speci c mortality in Japanese prostate cancer patients who received CAB therapy is less than half of those in USA patients 10 . Several reasons such as genetic and dietary/environmental factors have been discussed to explain the discrepancy between countries 11 . Furthermore, the difference in typical dosage of bicalutamide (80mg/day in Japan vs 50mg in western countries) has also been accounted for the controversial results 12 . In fact, another report has also suggested that the dose of bicalutamide is associated with PSA response 13 . In another study, Mathew RC et al. reported that CAB therapy improved survival more signi cantly compared to LH-RH monotherapy in men with very-high-risk metastatic disease, but not with lower risk metastatic tumors, indicating that CAB therapy may be more effective especially in high risk metastatic disease than low risk metastatic disease 10 . One of the most important points in the current study is that we showed the superiority of abiraterone use compared to 80mg use of bicalutamide in Japan where these drugs are often selected for the treatment of patients with high risk mHSPC in the real clinical setting.
Bertrand et al suggested that GnRH antagonist, degarelix, may be more effective than GnRH agonist, leuprolide, as for PSA control 14 . Furthermore, Kashiwabara et al reported that GnRH antagonist was more effective than GnRH agonist for CAB treatment of bone metastatic prostate cancer with pretreatment PSA level ≥ 50ng/ml 6 . There are several hypotheses about the reason why GnRH antagonist treatment exhibits better outcome than GnRH agonist treatment. Previous report suggested that GnRH antagonist treatment decreases PSA levels at a faster pace than GnRH agonist 14 . This rapid effect may result in better tumor control for longer duration. From these reasons, GnRH antagonist, instead of GnRH agonist, are widely used for ADT in Japan. Currently, almost all patients who had been diagnosed as high risk mHSPC in our hospital also received GnRH antagonist instead of GnRH agonist. In the present study the PSA level decreased rapidly in both of agonist and antagonist groups and average PSA response rate after 3 months of treatment was 99% in the two groups (data not shown). Comparison of PSA-PFS and OS between GnRH agonist and antagonist in CAB therapy may be future work. GnRH receptor expression has been reported in various types of malignant cells including prostate cancer cells 1516 . It is reported that both GnRH agonist and antagonist decrease proliferation of prostate cancer cells 17 . The in vitro comparison of growth suppression effect between GnRH agonist and antagonist may lead to more profound understanding of differential effect between the two drugs. This is the rst report which compared the e cacy of abiraterone with bicalutamide, in combination with GnRH antagonist for high risk mHSPC patients in Japan. However, it is important to note that there were several limitations in the present study. The patient cohort was small, and there was a signi cant difference regarding patient background because of the retrospective nature of the study. Moreover, observation period was short especially in patients treated with abiraterone. Therefore, a further prospective study with a larger cohort for longer period is required.
In conclusion, we demonstrated that abiraterone use in combination with GnRH antagonist instead of bicalutamide may have advantages regarding OS as well as PSA-PFS in patients with high risk mHSPC.
The ndings from the study could provide useful information when physicians choose a treatment plan in patients with high risk mHSPC.