Bone scan index can be a useful biomarker of survival outcomes in patients with metastatic castration‐resistant prostate cancer treated with radium‐223

Abstract

Background

Bone metastasis of prostate cancer is associated with pain and reduced overall survival (OS). Radium‐223, which is expected to reduce bone pain and prolong OS, was recently approved in Japan.

Aim

The aim of this study was to investigate the usefulness of the bone scan index by comparing the outcomes and factors according to response in Japanese patients treated with radium‐223.

Methods and results

Twenty patients receiving radium‐223 were divided into two groups according to whether they did or did not achieve a bone scan index decrease of at least one point (beneficial and non‐beneficial groups, respectively). The clinical characteristics at baseline and after three and six treatment cycles were compared using χ² tests and Student's t‐tests or Mann–Whitney U tests, and survival was estimated and compared using the Kaplan–Meier method and log‐rank test, respectively. Fourteen (70%) and six patients (30%) were categorized into the beneficial and non‐beneficial groups, respectively. Patients in the former group were significantly more likely to have a higher Eastern Cooperative Oncology Group performance status score and receive a greater number of radium‐223 injections (P < 0.05). Furthermore, patients in the beneficial group had a significantly longer OS (P < 0.05). Regarding safety, one and three patients in the beneficial and non‐beneficial groups, respectively, prematurely discontinued radium‐223 because of an increased prostate‐specific antigen level, decreased hemoglobin level, or femoral fracture.

Conclusion

Radium‐223 appears generally safe in this population. Patients with good bone scan index response have better performance status, receive more injections of radium‐223, and achieve OS prolongation. Bone scan index is a useful biomarker of survival outcomes and can be a valuable assessment tool in patients with metastatic castration‐resistant prostate cancer who are treated with radium‐223.

1. INTRODUCTION

In Japan, the incidence of prostate cancer (PC) is increasing, and it is expected to account for 14% of all malignant tumors in Japanese men in 2018. 1 Although PC has a relatively good prognosis overall, the 5‐year survival rate of metastatic PC is only 49.1%.1 Bone metastasis of PC is of particular concern, given its associations with increased pain and reduced overall survival (OS). PC that progresses despite a decrease in testosterone levels due to castration is defined as castration‐resistant PC (CRPC), and nearly half of affected patients develop significant bone pain.2

In May 2013, radium‐223 (Ra‐223; Xofigo®; Bayer HealthCare Pharmaceuticals Inc., Hanover, NJ, USA) was approved by the US Food and Drug Administration to improve OS and delay symptomatic skeletal events in patients with metastatic CRPC (mCRPC).2 Ra‐223, an alpha‐emitting isotope of the alkaline earth metal radium, produces double‐stranded DNA breaks that are difficult to repair.3 To date, most clinical data regarding Ra‐223 were obtained from the phase III ALSYMPLA trial of patients with CRPC with symptomatic bone metastasis and no visceral metastasis.4 In that study, patients treated with Ra‐223 had a median survival duration 3.6 months longer than that of patients who received placebo.4 Additionally, Ra‐223 extended the median time to the first symptomatic skeletal event.4

In March 2016, Japan approved the use of Ra‐223 to reduce the bone pain associated with metastatic PC and prolong OS. Although one study reported the good efficacy and tolerability of Ra‐223 in Japanese patients,5 little is known about the outcomes of this fairly new treatment in the Japanese population. In addition, no useful biomarker of survival outcomes in patients with mCRPC who are treated with Ra‐223 has been reported. To address these research gaps, we investigated the outcomes of Ra‐223 in a cohort of Japanese patients with metastatic PC and compared various parameters between those who experienced a decrease in the bone scan index (BSI), which is thought to be a marker of efficacy, and those who did not.

2. MATERIALS AND METHODS

The study was reviewed and approved by the institutional review board and patient consent was obtained. We retrospectively identified 20 patients with CRPC who initiated Ra‐223 therapy at our hospital between October 2017 and November 2018. Patients were administered a dosing regimen comprising six injections of 50 kBq Ra‐223 per kg body weight at 4‐week intervals. None had previously been treated with docetaxel or cabazitaxel.

In all patients, we evaluated the artificial neural network‐based BSI, a marker of the extent of bone metastasis, and the prostate‐specific antigen (PSA) level at the initiation of Ra‐223 treatment and after three and six injection cycles. We calculated the BSI values using software developed against Japanese training databases from nine institutions (BONENAVI, version 2, BN2, n = 1532; FUJIFILM RI Pharma Co., Tokyo, Japan). All patients underwent blood collection during every treatment cycle and computed tomography and bone scintigraphy scans every three cycles. All adverse events were determined using the Common Terminology Criteria for Adverse Events v. 4.0, Grade 3.6

All patients were classified into one of two groups depending on the BSI outcomes to investigate its usefulness. Patients in the beneficial group experienced a decrease in the BSI of at least one point, whereas those in the non‐beneficial group did not experience any decrease in the BSI. These groups were compared with respect to baseline characteristics. Regarding safety, we investigated the reasons recorded for Ra‐223 cessation in each patient group.

In the statistical analysis, comparisons of baseline characteristics between patients in the beneficial and non‐beneficial groups were assessed using the χ² test for categorical variables and Student's t‐test or the Mann–Whitney U test for continuous variables. Categorical variables are presented as numbers and percentages, and continuous variables are presented as means ± standard deviations or medians (interquartile ranges). Survival durations were estimated using the Kaplan–Meier method and compared using the log‐rank test. Hazard ratios with 95% confidence intervals (CIs) were also determined. In all analyses, a P value <0.05 was considered statistically significant. All analyses were performed using JMP 14.0 software (SAS Institute, Inc., Cary, NC, USA).

3. RESULTS

The characteristics of the 20 included patients are described in Table 1. Fourteen patients (70%) exhibited a decrease in BSI during Ra‐223 treatment and were included in the beneficial group, whereas six patients (30%) did not exhibit a decrease in BSI and were included in the non‐beneficial group. Within the beneficial group, seven patients (50%) also experienced a beneficial decrease in PSA levels.

Table 1.

Patient characteristics at baseline and Ra‐223 treatment initiation by group

Subgroup	Beneficial (N = 14)	Non‐beneficial (N = 6)	P value
Baseline characteristic
Age (years)	73 ± 7	72 ± 13	0.893
Body mass index (kg/m2)	22.9 ± 2.9	19.9 ± 4.0	0.067
PSA at the time of diagnosis (μg/L)	988.0 ± 1879.4	872.3 ± 1257.9	0.893
Cancer‐positive core rate (%)	68 ± 27	71 ± 24	0.807
Gleason score	9 (7–9)	9 (8–9)	0.862
Time from diagnosis to the beginning of Ra‐223 (months)	37 ± 39	34 ± 15	0.893
Time from diagnosis to CRPC (months)	30 ± 36	24 ± 19	0.611
Time from CRPC to the beginning of Ra‐223 (months)	6 ± 7	10 ± 13	0.356
ECOG performance ‐statues score	0	0.5 (0–‐1)	0.002
Bone pain (%)	3 (21.4)	4 (66.7)	0.052
Previous use of enzalutamide or abiraterone (%)	8 (57.1)	2 (33.3)	0.329
Previous use of denosumab (%)	2 (14.3)	2 (33.3)	0.329
Extent of skeletal disease score	3 (3–4)	3 (2.8–‐4)	0.944
Bone scan index	1.72 ± 1.49	2.11 ± 2.03	0.636
PSA (μg/L)	18.6 ± 13.7	26.2 ± 24.1	0.38
Neutrophil (/μL)	5483 ± 2128	4413 ± 1364	0.275
Hemoglobin ((g/dL)	12.2 ± 1.9	11.7 ± 1.3	0.545
Platelet (×104μL)	23.0 ± 7.8	22.8 ± 7.6	0.95
ALP (U/L)	399 ± 265	649 ± 588	0.197
LDH (IU/L)	191 ± 51	195 ± 49	0.885
Number of total Ra‐223 injections	6	4.5 (3–‐6)	0.01

Ra‐223, radium‐223; PSA, prostate‐specific antigen; CRPC, castration‐resistant prostate cancer; ECOG, Eastern Cooperative Oncology Group; ALP, alkaline phosphatase; LDH, lactate dehydrogenase

The Eastern Cooperative Oncology Group (ECOG) performance status (PS) score was significantly lower in the beneficial group than in the non‐beneficial group, as all patients in the former group had a score of 0 (0.002). Patients in the beneficial group also received a significantly greater number of Ra‐223 injections than those in the non‐beneficial group (0.01).

The reasons for premature discontinuation of Ra‐223 therapy are listed in Table 2. Four patients (20%) prematurely discontinued Ra‐223, including one patient (7.1%) in the beneficial group and three (50%) in the non‐beneficial group. The reasons cited for treatment discontinuation included increased PSA level, decreased hemoglobin (Hb) level, and a fall resulting in a fractured femur. The only drug‐related adverse event was a decreased Hb level, which occurred in one patient.

Table 2.

Reasons for stopping Ra‐223 treatment and number of patients per group

Reason	Beneficial	Non‐beneficial
Increased PSA level	1	1
Decreased Hb level	0	1
Fractured femur	0	1

PSA, prostate‐specific antigen; Hb, hemoglobin

Figure 1 presents a Kaplan–Meier curve of OS. The log rank test confirmed that the OS duration was significantly longer in the beneficial group than in the non‐beneficial group (P < 0.05; hazard ratio, 0.21; 95% CI: 0.045–0.95).

Figure 1.

Kaplan–Meier curves of overall survival by group. The beneficial and non‐beneficial groups included patients who respectively did or did not exhibit a decrease in the bone scan index value after treatment with radium‐223. The P‐value, hazard ratio, and 95% confidence interval were determined using the log‐rank test

4. DISCUSSION

In this study of Ra‐223 treatment for CRPC, we classified patients into two groups based on their BSI response, and determined that patients who benefitted from treatment (i.e., exhibited a decrease in BSI) had a better baseline ECOG PS, received a greater number of Ra‐233 doses, and had longer OS. This shows that the BSI can be an indicator of survival in patients who received Ra‐223. We selected the BSI as an indicator of treatment response based on the existing literature. For example, the Prostate Cancer Clinical Trials Working Group 3 suggested the use of the BSI as a quantitative measurement of bone metastasis,7 and this index was proven to be more valuable than the extent of skeletal disease (EOD) score for the assessment of prognosis and responses in patients with PC.8 Furthermore, patients with higher automated BSI levels tended to have worse disease‐specific survival than those with lower levels.9 In addition, the BSI was also reported to be useful in determining the outcomes of Ra‐223 treatment, although only a few studies have examined this. Among patients who received Ra‐223 in a previous study, those with a BSI of >5 had a significantly shorter median OS than those with a BSI of ≤5, and the former group also had a higher incidence of hematologic toxicity.10 These results suggest that the BSI is a useful biomarker of survival outcomes and hematologic toxicity, and thus could be a valuable assessment tool in patients with mCRPC who are treated with Ra‐223.10 In our study, only one patient experienced hematologic toxicity and stopped Ra‐223. Hence, we cannot state that the BSI is a biomarker of hematologic toxicity, but it is a useful biomarker of survival outcomes and could be a valuable assessment tool. We note that the bone flare phenomenon may confound the interpretation of early BSI changes during therapy10; to overcome this issue, we evaluated the BSI before treatment and after 3 and 6 months of treatment and used a decrease in the BSI of at least one point over time to indicate a beneficial effect.

Regarding safety outcomes, only one patient at our institution stopped Ra‐223 therapy because of hematopoiesis, as previously mentioned. However, this patient also exhibited an excessive increase in PSA, suggesting disease progression. A pre‐existing impairment in hematopoiesis, particularly thrombocytopenia and anemia, prior to Ra‐223 therapy has been identified as an important risk factor for a worse treatment outcome.11 However, only two patients had hematopoiesis prior to therapy: one patient had anemia (Hb <10 g/dL) and one had thrombocytopenia (platelet count <13.5 × 10⁴ μL). The former patient discontinued Ra‐223 after four cycles because the PSA increased dramatically; however, it was beneficial because the BSI decreased. The other patient received six cycles of Ra‐223, which was beneficial as assessed by a decline in the PSA and BSI. Therefore, we could not determine the significance of this factor on the effect of Ra‐223. In another study conducted to identify significant baseline predictors of grade 2–4 hematologic toxicities related to Ra‐223 treatment,12 the EOD score (<6 bone metastases), PSA level, and prior docetaxel use were also identified as risk factors. However, we note that, among our patients, the EOD score and baseline PSA level were not significantly higher in the beneficial group than in the non‐beneficial group. Furthermore, none of our patients previously received docetaxel, which may have also contributed to the low frequency of hematologic toxicities. Hence, we could not identify that the BSI is a biomarker of hematologic toxicity in patients with mCRPC who are treated with Ra‐223.

An increasing PSA level during Ra‐223 treatment may indicate true disease progression in the form of a new visceral metastasis or flare reaction.13 In our study, two patients exhibited dramatic increases in their PSA levels, but these were attributed to the progression of bone metastases, as no visceral metastasis was identified. Although one of these patients exhibited a slight decrease in the BSI, his PSA level increased from 9.4 to 71.4 ng/dL. Accordingly, we decided to stop Ra‐223 and initiate docetaxel as soon as possible. As decisions regarding whether to stop Ra‐223 are difficult, it is important to determine whether an increase in PSA reflects true disease progression or a flare reaction.

Finally, we observed that the beneficial group received a significantly greater number of Ra‐223 doses than the non‐beneficial group. In a previous study, the completion of at least five cycles of Ra‐223 therapy was associated with a lower alkaline phosphatase (ALP) level, higher Hb level, and longer median OS.14 In our study, although the two groups did not differ significantly in terms of ALP and Hb levels, the beneficial group experienced a longer OS, consistent with the previous study.

We used the BONENAVI version 2 software, which is a computer‐assisted diagnosis system with an increased sensitivity to detect metastases. The device is also associated with a reduced interobserver variation in planar whole‐body bone scans.15 BONENAVI version 2 uses databases from nine institutions, whereas version 1 uses databases from a single institution. Therefore, BONENAVI version 2 significantly improved its identification capabilities from version 1.16

As a limitation of this study, the small sample size from a single‐center presents some challenges. We hope that more patients enter our study as our goal is to further investigate Ra‐223 in more detail to determine the relationship between the BSI and hematologic toxicity.

In conclusion, Ra‐223 appears to be safe for the treatment of CRPC in Japanese patients, with few adverse events. Our results suggest that the BSI is a useful biomarker of survival outcomes and can be a valuable assessment tool in patients with mCRPC who are treated with Ra‐223.

CONFLICT OF INTEREST

The authors have no conflicts of interest to declare.

AUTHORS' CONTRIBUTION

All authors had full access to the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Conceptualization, K.H.; Methodology, K.H.; Investigation, M.N.; Formal Analysis, M.N.; Resources, R.U.; Writing ‐ Original Draft, M.N.; Writing ‐ Review & Editing, M.N.; Visualization, R.U.; Supervision, R.U.; Funding Acquisition, K.H.

Naito M, Ukai R, Hashimoto K. Bone scan index can be a useful biomarker of survival outcomes in patients with metastatic castration‐resistant prostate cancer treated with radium‐223. Cancer Reports. 2019;2:e1203. 10.1002/cnr2.1203

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are not publicly available due to privacy or ethical restrictions, but the data are available from the corresponding author upon reasonable request.