Prognosis and safety of radium‐223 with concurrent abiraterone acetate or enzalutamide use for metastatic castration‐resistant prostate cancer: Real‐world data of Japanese patients

Abstract Objectives To evaluate the real‐world data on the efficacy and safety of a combination therapy with radium‐223 (Ra‐223) and second‐generation androgen‐receptor targeting agents (ARTAs), including abiraterone acetate (ABI) and enzalutamide (ENZ), among Japanese patients with bone metastatic castration‐resistant prostate cancer (CRPC). Patients and methods We retrospectively reviewed 79 patients with bone metastatic CRPC who were treated with Ra‐223. The number of patients with concurrent ARTA use was 24:17 receiving ABI and 7 receiving ENZ. We evaluated the overall survival (OS) according to ARTA use and compared the survival of patients treated with Ra‐223 with or without ARTA using multivariate analysis. Results The median survival in the entire cohort was 23.5 months. The patients receiving Ra‐223 combined with ARTA showed a tendency of better OS than patients treated with Ra‐223 alone, although no significant difference was observed (median OS, 26.5 vs 23.5 months; P = .115). A multivariate analysis showed that the extent of disease on bone scan (EOD) scores and pain at baseline were significant predictors of OS. The concurrent use of bone‐modifying agents (BMAs) was not significant for favorable OS (P = .050). However, the concurrent use of second‐generation ARTA was not a significant factor for OS. Regarding safety, a bone fracture occurred in only one (4.2%) of 24 patients treated with combined Ra‐223 and ARTA therapy. Conclusion Our real‐world data analysis suggested that Ra‐223 combined with a second‐generation ARTA is well tolerated in Japanese patients. The EOD score and pain at baseline are significant prognostic factors for OS, but the concurrent use of second‐generation ARTA has no influence on OS among men treated with Ra‐223. The concurrent use of BMA yields a marginally favorable OS.

As each of these drugs showed efficacy in monotherapy, their combination therapies were expected to show more survival benefit. However, in a randomized, double-blind, placebo-controlled, phase III ERA 223 trial, 10  Since then, this combination therapy is not being used.
However, certain studies have shown the effectiveness and safety of this combination therapy. [14][15][16] Thus, we investigated the real-world data on the efficacy and safety of a combination therapy of Ra-223 and a second-generation ARTA for mCRPC among Japanese patients who were treated before the EMA issued a warning.

| Patients
We retrospectively reviewed the data of 79 patients with mCRPC metastasized to bone and who were treated with Ra-223 between 2012 and 2017 at the Yokohama City University Medical Center and Kindai University Hospital. Some patients participated in the Japanese trial of Ra-223. All patients had histologically confirmed prostate adenocarcinoma at the initial diagnosis. The patients with distant metastases except bone and/or with regional lymph node metastases greater than 4 cm were excluded from this study. Ra-223 was administered at a dose of 55 kBq (1.49 microcurie)/kg intravenously every 4 weeks up to six cycles. In addition, the luteinizing hormone-releasing hormone (LHRH) agonist or antagonist was administrated to all patients. After Ra-223 failed, all patients continued to receive LHRH agonist or antagonist and were subsequently treated according to each physician's treatment strategy. For terminally ill patients, palliative therapy and pain control with morphine and palliative external-beam radiation were used, as appropriate.
Ra-223 combined with ARTA was used in 24 patients from the start of therapy; the ARTA included ENZ 160 mg/day (seven patients) and ABI 1,000 mg with prednisone 10 mg/day (17 patients) (combination group). The remaining 55 patients were treated with Ra-223 without ARTA (single group).

| Clinical assessments
The primary endpoint of this study was survival in the combination group compared with that in the single group. According to the ERA 223 results, we investigated the frequency of bone fracture only in the combination group because we were concerned about the increased frequency of bone fracture in the patients who received Ra-223 concurrently with ARTA. Therefore, the frequency of bone fracture in the single group could not be evaluated. The secondary endpoint was the frequency of bone fractures in the combination group.
Patient's disease state was clinically staged according to the 1997 TNM classification by computed tomography and bone scan using 99 mTc-methylene diphosphonate. The extent of bone metastases evaluated by the bone scan was classified according to the extent of disease on bone scan (EOD) score, as suggested by Soloway et al. 17 Eleven variables at baseline were assessed from the electronical medical records; these included patient age, time to CRPC from initial androgen-deprivation therapy, previous use of docetaxel, prostate-specific antigen (PSA) level, hemoglobin level, lactate dehydrogenase level, alkaline phosphatase (ALP) level, Gleason scores (GS), EOD scores, concurrent use of bone-modifying agents (BMAs), and pain at baseline. Denosumab 120 mg was administrated every 4 weeks and zoledronic acid 4 mg was administrated every 4 weeks to patients treated with BMAs. Pain at baseline wad defined as "patient-reported complaint" with or without painkiller use. In Japan, Ra-223 has been approved for not only mCRPC patients with pain, but also those without pain. GS was determined according to The 2005 International Society of Urologic Pathology Gleason grading. 18 Serum PSA levels were measured using the Elecsys PSA Assay (Roche Diagnostics, Basel, Switzerland). The occurrence of bone fracture after Ra-223 administration in the combination group was second-generation ARTA has no influence on OS among men treated with Ra-223.
The concurrent use of BMA yields a marginally favorable OS.

K E Y W O R D S
abiraterone acetate, castration-resistant prostate cancer, enzalutamide, prostate cancer, radium-223 evaluated by CT and X-ray radiography. The patients were followed up every 4 weeks in the outpatient setting.

| Statistical analysis
Differences in patient characteristics between the combination group and the single group were analyzed by Mann-Whitney U test and chi-square test. Overall survival (OS) was defined as the time between initiation of Ra-223 treatment to the final date of consultation or death. A Kaplan-Meier product-limit estimator was used to assess OS distribution. Log-rank tests were used to analyze differences in OS between the two groups. Univariate and multivariate Cox proportional hazard models were used to analyze prognostic factors of OS. Moreover, relative risks and 95% confidence intervals (95% CIs) were determined. All tests were two-sided; α = .05 was considered significant. All analyses were conducted using the IBM SPSS Statistics software for Windows, version 22 (IBM Corp., Armonk, NY, USA) and EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan). 19 Experimental procedures were conducted in accordance with the ethical standards of the Helsinki Declaration. This study was approved by the institutional review board of Yokohama City University Medical Center (B181000040). Informed consent to participate in the study was obtained from all subjects in an opt out manner.

| RE SULTS
The median observation period was 12.5 months and the number of the patients who were lost to follow-up was five.
The characteristics of patients in the combination group and single group are shown in Table 1  and ENZ is shown in Figure S1. The median OS was not reached (95% CI, 4.8 months to NA) and 26.5 (95% CI, 13.6-NA) months by patients concurrently treated with ENZ and ABI, respectively.
The Kaplan-Meier curve for OS according to EOD score and pain at baseline is shown in Figure 2A   There was not any specific adverse event in the combination group. osteoblasts from ABI, combined with the osteoclastogenic effects of androgen deprivation, glucocorticoids, and Ra-223 increased the bone fracture risk. 22 Cursano et al provided another possible reason to be the anabolic and antiresorptive effect of ABI on the bone. 23 As ABI inhibits osteoclastic function and promotes osteoblastic differentiation, 24 the abnormal deposition of radium-223 could accumulate in a healthy bone, and lead to cell and tissue damage. Another opinion to explain this result is the excessive dose of prednisone administered in the trial. The glucocorticoids are known to be associated with bone loss and fragility even at low doses. 25 In our study, both EOD classifications and pain at baseline were significant prognostic factors for the OS. Among the patients treated with Ra-223 for bone metastatic CRPC, the ALP levels at the baseline has been reported as a prognostic factor from various reports 30,14,31 and recently, the bone metastatic burden measured by the bone scan index (BSI) also has been reported to be of prognostic value. 32 The BSI is a more objective imaging biomarker rather than the EOD, a subjective measure, and is already used for its prognostic value in a prospective study for patients with mCRPC. 33 The pain at baseline has also been a significant prognostic factor for the OS among patients treated with Ra-223. 14  There are several limitations that we need to consider as well.

| D ISCUSS I ON
First, it was a retrospective study with a small number of patients and a short follow-up period. The baseline PSA and ALP levels were significantly higher in the single group than in the combination group.
To adjust the cofounding factor including the PSA and ALP levels, we performed a multivariate analysis and found that concurrent use of second-generation ARTA has no influence on OS among men treated with Ra-223. However, it cannot be denied that the differences in baseline variables between the two groups could influence our results because of the retrospective nature of the study. Second, the use of BMAs was decided by the attending physician and the timing of introduction and dosing interval differed between cases. Finally, our study included only Japanese patients; thus, our data cannot be generalized to the global population.
Our study showed that the concurrent use of second-generation ARTA with Ra-223 did not improve the OS compared with the use of Ra-223 alone among Japanese patients with bone metastatic CRPC, even though the combination group had better tendency of survival.
Bone fracture occurred in only one patient (4.2%) in the combination group; and thus, the combination therapy seemed well tolerated in our cohort.