Reduction of Initial Dose of Enzalutamide does not Decrease the Incidence and Severity of Adverse Events in Castration- Resistant Prostate Cancer

Shunsuke Tsuzuki Tokyo Jikeikai Ika Daigaku Fuzoku Byoin Honin https://orcid.org/0000-0001-6164-4684 Shotaro Nakanishi Ryukyu Daigaku Mitsuyoshi Tamaki Naha City Hospital Takuma Oshiro Naha City Hospital Jun Miki Jikei University School of Medicine: Tokyo Jikeikai Ika Daigaku Hiroki Yamada Jikei University School of Medicine: Tokyo Jikeikai Ika Daigaku Tatsuya Shimomura Jikei University School of Medicine: Tokyo Jikeikai Ika Daigaku Takahiro Kimura (  tkimura@jikei.ac.jp ) Tokyo Jikeikai Ika Daigaku Fuzoku Byoin Honin Nozomu Furuta Jikei University School of Medicine: Tokyo Jikeikai Ika Daigaku Seiichi Saito Ryukyu Daigaku Shin Egawa Tokyo Jikeikai Ika Daigaku Fuzoku Byoin Honin


Background
Prostate cancer (PC) is the most common type of cancer among elderly men worldwide (1). Overall, 7%-15% of PC patients are initially diagnosed with metastatic PC and started with androgen deprivation therapy (ADT) (2). In addition, 20%-50% of non-metastatic PC patients treated with surgery or radiation experience biochemical recurrence (BCR) and are started with ADT (3). Finally, these patients receiving ADT develop castration-resistant prostate cancer (CRPC) within a couple of years.
Enzalutamide, a next-generation androgen receptor (AR)-targeted agent, was approved for CRPC based on the results of phase 3 double-blind, randomized trials (PREVAIL and AFFIRM) (4,5). In these trials, enzalutamide signi cantly prolonged the survival of patients compared to placebo in, both, pre-and post-chemotherapy settings. On the other hand, a wide variety of adverse events (AEs), such as fatigue, decreased appetite, and hypertension, were reported. In both trials, > 90% of CRPC patients treated with enzalutamide experienced AEs of varying severity. A similar tendency was observed in the subgroup analysis of a Japanese cohort in the PREVAIL trial. In detail, approximately 95% of patients treated with enzalutamide had AEs of varying severity, with decreased appetite, weight loss, and fatigue being the commonly experienced AEs (6).
In real-world practice, patients are sometimes required to undergo dose reduction or discontinuation of enzalutamide because of severe AEs (7). A retrospective study reported the predictive factors for the occurrence of AE, which were as follows: older age, worse performance status (PS), and initially started with a full dose of enzalutamide (8). To minimize the occurrence of AEs and prolong the treatment period of enzalutamide, the initial starting dose is sometimes reduced for patients with such risk factors (i.e. older age and worse PS). Recently, inferiority of low-dose of enzalutamide in late-elderly patients over standard dose of enzalutamide has reported by retrospective study with 59 metastatic CRPC patients (9).
However, there was no clear evidence whether the initial dose reduction of enzalutamide results in a longer treatment duration, thereby leading to the improvement of survival in these patients. Therefore, we aimed to investigate whether the initial dose of enzalutamide affects the incidence of AEs and oncological outcome in patients with CRPC.

Methods
This study was approved by the Ethics Committee of both The Jikei University School of Medicine ((30-390(9411)) and University of the Ryukyus, Graduate School of Medicine.

Patients and study design
We retrospectively reviewed the clinical records of CRPC patients who were treated with enzalutamide from June 2014 to December 2018 at The Jikei University Hospitals and University of the Ryukyus, Graduate School of Medicine and its a liated institution. Patients were excluded from this study based on the following criteria: 1) lack of clinical information, 2) short follow-up periods (< 1 month) and 3) presence of prior treatment with abiraterone acetate. Eventually, 233 patients were enrolled for this study. The initial dose of enzalutamide was decided based on the discretion of the attending clinicians. Patients were classi ed to whom introduced with full dose of enzalutamide or whom with reduced dose and the e cacy and safety of enzalutamide treatment were compared. The treatment sequence for CRPC was basically decided according to the guidelines (10). CRPC was de ned according to the guidelines of the Prostate Cancer Clinical Trials Working Group 2 (PCWG2) (11). Prostate speci c-antigen (PSA) progressions after enzalutamide treatment were de ned according to the PCWG2 criteria (11). In this study, the discontinuation of enzalutamide treatment either temporally or permanently due to the occurrence of any AEs was used as the categorical variable in the Cox regression analyses.

Statistical analysis
The patients' characteristics between the full and reduced dose groups shown in Tables 1 and 2 were compared using the chi-square test and t-test. Propensity score for selecting the initial full dose of enzalutamide or the reduced dose of enzalutamide was calculated, thereafter 1:3 ratios of k-neighbor Propensity score matching (PSM) was performed to reduce the variance of the baseline characteristics in the full and reduced dose groups. The association of the initial dose of enzalutamide with PSA progression-free survival (PFS) and overall survival (OS) rates were estimated using the Kaplan-Meier method, and the log-rank test was used to compare the survival outcome. Univariate and multivariate Cox regression analyses were performed to identify the independent factors for PFS and OS. All data were analyzed using STATA 14 (Stata Corp., College Station, TX). Differences were considered signi cant if the two-sided p-values were < 0.05.  When stratifying the patients to full or reduced dose groups, the patients in the reduced dose group were signi cantly older and had worse PS than those in the full dose group ( Table 1).
The patients' baseline characteristics after PSM are shown in Table 2. After PSM, among the 124 patients, 31 and 93 patients received a reduced and a full dose of enzalutamide, respectively. The baseline characteristics were not signi cantly different between the full and reduced dose groups.
In the PSM cohort, the PSA reduction rates after enzalutamide treatment are shown in Table 3 and Fig. 1. The PSA response rate after enzalutamide treatment was signi cantly better in the full dose group (-87.4%) than in the reduced dose group (-66.3%). The proportion of patients with PSA decline of > 90% was signi cantly higher in the full dose group than in the reduced dose group (46.2% and 25.8%, respectively, p-value = 0.03). The median time to PSA nadir was 2 months, which was not signi cantly different between the two groups. There was no signi cant difference in the duration of enzalutamide treatment between the full (median: 8months) and the reduced dose group (median 7months).  Fig. 2. There was no signi cant difference in PSA PFS and OS between the full and reduced dose groups. In detail, the median time to PSA progression in the full and reduced dose groups was 8months and 6 months, respectively (p-value = 0.10) ( Table 3).
The AE pro le of the patients is presented in Tables 4 and 5. The incidence rates of AEs of any grade and of grades 3-5 were 30.2% and 8.9%, respectively, in the entire cohort. Fatigue and appetite loss were the most frequent AEs observed in this cohort. The incidence of AE of any grade was not signi cantly different between the full and reduced dose groups (34.4% and 22.6%, respectively, p-value = 0.24).   Finally, the univariate and multivariate Cox regression analyses were conducted to investigate the independent prognostic factors for PSA PFS and OS. The results are summarized in Table 6. We found that the initial dose of enzalutamide was not a predictive factor for PFS and OS. In the multivariate analyses, time to CRPC within 12 months (hazards ratio [HR]: 2.55) was the independent predictive factor for PSA progression. Lower PSA response rate (HR 1.00) was a signi cant predictive factor for worse OS.

Discussion
In the present study, we assessed the in uence of the initial starting dose of enzalutamide on the survival outcome and incidence of AEs in CRPC patients. The results indicated that the rate of patients with PSA decline of > 90% after enzalutamide treatment was lower in the reduced dose group than in the full dose group, despite the fact that the reduction of the initial dose was not signi cantly associated with worse PFS and OS in the multivariate analyses. On the other hand, the incidence and severity of AEs were not signi cantly different between the full and reduced dose groups.
A dose escalation study of enzalutamide (30-360 mg per day) was conducted in a previous preliminary and phase 1-2 trials, which found the dose-dependent increase of enzalutamide in the plasma (12,13). In fact, the dose dependency of enzalutamide in the percentage change of PSA from baseline was analyzed in the phase 1-2 study (13). The proportion of patients with PSA decline of > 50% was lower in the patients received 60 mg per day of enzalutamide than those received higher doses (> 150 mg per day) (13). Likewise, the dose dependency of enzalutamide was observed on the incidence of AE in the phase 1-2 study. Patients administered ≤ 150 mg per day of enzalutamide did not complain of fatigue or required discontinuation of treatment due to AEs (13). Thus, the standard dose of enzalutamide was set as 160 mg per day.
However, at the start of its use in clinical practice, there were a number of patients with severe AEs after receiving 160 mg per day of enzalutamide in our study, even though the incidence was not higher than that of the previous phase 3 study (14).
To minimize the incidence and severity of AEs and prolong the treatment period, some clinicians introduce enzalutamide with a reduced initial starting dose, even though reducing the dose might lead to insu cient treatment e cacy. The results of this study indicated that reducing the initial dose of enzalutamide might not decrease the incidence of AEs and might impair the PSA response.
The incidence of grade 3-5 AEs was generally very low in this study (9.7% and 6.5% in the full and reduced dose groups, respectively). Nevertheless, 8.1% of the patients preferred to discontinue the enzalutamide treatment due to any AEs.
Regarding the median time to discontinuation of the enzalutamide treatment, no statistical difference was found between the full and reduced dose groups (7.0 and 6.5 months, respectively; p-value = 0.48), suggesting that the initial dose reduction of enzalutamide might not prolong its treatment duration.
In this study, the rate of patients with PSA decline of > 90% was signi cantly higher in the full dose group than in the reduced dose group. The PSA response after enzalutamide was reported to be a predictor for radiographic PFS and OS of CRPC (15).
The e cacy of enzalutamide might be impaired by reducing the initial dose, even though the OS and PFS were not signi cantly different possibly because of limited number of patients and events analyzed in this study. Thus, we considered that introducing the enzalutamide with a full dose might be more effective for disease control. Dose reduction or temporary discontinuation of enzalutamide should be considered if patients develop any AEs.
In this study, the multivariate analysis revealed that the time to CRPC within 12months was the independent predictive factor for worse PSA PFS, and lower PSA response to enzalutamide treatment was an independent predictive factor for worse OS. These ndings were consistent with those of previous reports (5,16,17), and suggested that enzalutamide might show strong effects if it was administrated at the early phase of CRPC treatment.
This study has several limitations. First, although we performed PSM, our research was a retrospective study with a small sample size; thus, a selection bias might have been introduced. Second, AEs were observed only in 31.5% of the patients, which might be due to the retrospective nature of this study. However, we were able to analyze the data of patients with AEs of grade 3 or higher. Third, the short follow-up period might have affected the results of the survival outcomes. Finally, due to the small number of patients analyzed, we did not divide the CRPC patients according to the history of prior chemotherapy. Despite these limitations, the present work is an original study that investigated the relation of the initial dose of enzalutamide to the survival outcomes and incidence of AEs. We believe that our data will useful in daily clinical practice.
Further external validation cohort and prospective study are warrant in the future.

Conclusions
Starting the enzalutamide treatment with a reduced dose did not signi cantly decrease the incidence rate of AEs and the PSA response rate after enzalutamide treatment was even lesser, thus, the full dose of enzalutamide might be the rst choice. However, we should consider dose reduction or temporary discontinuation of treatment if the patients develop any Medicine. As the analyses only used de-identi ed data, no additional approval from patients was required.

Consent for publication Not applicable
Availability of data and material The de-identi ed datasets used during the current study are available from the corresponding author on reasonable request.