Combination treatment in metastatic castration-resistant prostate cancer: can we safely boost efficacy by adding radium-223?

ABSTRACT

ERA 223, a phase III, international, multicenter, double-blind study published in Lancet Oncology, was the first randomized controlled trial to investigate combined radium-223 (Ra-223) and abiraterone acetate plus prednisone or prednisolone (AAP) therapy. The data from ERA 223 demonstrated no increase in efficacy for this combination over AAP alone, and instead identified a significant safety concern due to the higher risk of fracture in the co-treatment group. The surprising results of this trial likely stem from the compounding osteoporotic effects of the different treatments, particularly the addition of prednisone, and supplementing therapy regimens with osteoprotective agents may aid in mitigating this safety risk.

KEYWORDS:

• Abiraterone
• radium-223
• mCRPC
• fracture risk
• prednisone
• denosumab

Since 2004, multiple monotherapy options for metastatic castration-resistant prostate cancer (mCRPC) have been approved.¹ Despite the overall survival benefits of these agents alone,^2–7 efforts to further improve clinical outcomes for mCRPC through combinatorial approaches have consistently failed. One possible pairing of treatments is abiraterone acetate plus prednisone or prednisolone (AAP) and radium-223 (Ra-223). Abiraterone inhibits cytochrome P450 c17 (CYP17), obstructing intratumoral androgen synthesis and further reducing circulating androgens beyond standard androgen deprivation therapy.² Standard AAP is administered as 1000 mg of abiraterone acetate once daily by mouth, with either 5 mg of oral prednisone or prednisolone twice daily to reduce upstream steroid accumulation, cortisol suppression, and mineralocorticoid excess.^8–10 Ra-223 is an alpha emitter that preferentially causes double-stranded breaks in the DNA of bone metastases and induces cytotoxicity in these areas,³ a notable function as over 90% of patients with mCRPC develop bone lesions.⁶ Their distinctive mechanisms of action suggest a possible benefit to their coupled use, with abiraterone limiting additional tumor growth through androgen depletion¹¹ and Ra-223 selectively initiating cell death in areas of bone metastasis.³

A 2016 open-label, single-arm phase 3b trial first suggested the benefits of this combination when they observed an improvement in overall survival for patients with mCRPC receiving both Ra-223 and abiraterone compared to those receiving Ra-223 alone.¹² Furthermore, two different phase II studies have since demonstrated the safety of administering Ra-223 in conjunction with hormonal therapies, such as abiraterone or enzalutamide.^13,14 These initial investigations found that patients presenting with fewer symptoms, less advanced disease and/or fewer prior cancer treatments experienced the best response to combined therapy with abiraterone and Ra-223.^12,13 A recent study published in Lancet Oncology by Smith et al. (ERA 223) focused on these patients with asymptomatic or mildly symptomatic disease not previously treated with docetaxel or AAP, and this trial is the first randomized, controlled study to examine the safety and efficacy of Ra-223 with AAP.¹⁵

ERA 223 was an international, multicenter, placebo-controlled, double-blind, randomized, phase III trial that assigned 806 patients (1:1) to receive AAP (standard dosing) with six doses (55 kBq/kg) of IV Ra-223 administered every 4 weeks or AAP alone. Patients were eligible for enrollment if they had histologically confirmed and progressive mCRPC with ≥ 2 bone metastases. Enrolled patients were also asymptomatic or mildly symptomatic with an Eastern Cooperative Oncology Group (ECOG) performance score of 0 or 1 and life expectancy ≥ 6 months. Exclusion criteria included any confirmed visceral or brain metastases and any of the following prior treatments: AAP, mCRPC chemotherapy, or hemibody external radiotherapy. Of note, patients were not excluded for prior treatment with other anti-androgen therapies, such as enzalutamide or bicalutamide, or for prior external radiotherapies so long as they had sufficient bone marrow function. Symptomatic skeletal event-free survival (SSE-FS) was the primary endpoint of ERA 223, and the secondary endpoints included overall survival, radiological progression-free survival, time to opiate use for cancer pain, and time to cytotoxic chemotherapy.¹⁵

ERA 223 found that co-treatment with AAP and Ra-223 did not improve SSE-FS when compared to AAP monotherapy (22.3 months vs. 26 months, respectively; HR: 1.122, 95% CI: 0.917 to 1.374, p = .2636). Furthermore, there was no significant difference between the treatment arms for any of the chosen secondary end points, including overall survival (30.7 months vs. 33.3 months, AAP + Ra-223 vs. AAP, respectively). The co-treatment group had a higher occurrence of total bone fractures compared to those receiving AAP alone (112 [29%] vs. 45 [11%], respectively), particularly in the first 12 months (95 of 112 [85%] vs. 26 of 45 [58%], respectively). Of the patients who had fractures confirmed via independent assessments, those receiving Ra-223 experienced a higher rate of osteoporotic fractures. However, patients in the co-treatment group also had a lower likelihood of spinal cord compression (10 [5%] vs. 19 [10%]) and fewer instances of external beam radiotherapy (73 [37%] vs. 80 [42%]), though these findings were not statistically significant. The prevalence of adverse events was otherwise comparable between groups, with serious events occurring in 160 patients (41%) vs. 155 patients (39%) in the AAP + Ra-223 and AAP alone arms, respectively. The high number of fractures and deaths on study led to premature unblinding of ERA 223.¹⁵

Overall, ERA 223 demonstrated no increase in efficacy for the combined use of AAP and Ra-223 and instead raised a significant safety concern observed by the first interim analysis.¹⁵ Though a lack of increased efficacy for the combination treatment could explain these data, the unexpected increase in bone fragility provides a clearer explanation.¹⁶ In the AAP + Ra-223 treatment group, most of the fractures developed in non-metastatic sites (60 of 76 [79%]) and were not directly influenced by disease progression.^15,16 Ra-223 monotherapy has been shown to limit osteoblastic activity via reduction of alkaline phosphatase levels, however, this effect was not associated with an increased risk of fracture.³ While Ra-223 targets areas with elevated levels of osteoblastic activity,¹⁷ abiraterone stimulates osteoblastic differentiation and bone matrix deposition.¹⁸ Concurrent treatment with abiraterone could lead to Ra-223 deposits outside of metastatic sites, increasing damage to bone and leading to an increased fracture risk.¹⁷ Furthermore, continuous glucocorticoid administration can result in both diminished bone formation and increased bone resorption, leading to synergistic imbalances in bone remodeling when combined with Ra-223 and abiraterone.¹⁶ Administering Ra-223, abiraterone, and prednisone in patients already at a higher risk of fracture from uninterrupted androgen deprivation therapy ultimately surpassed a threshold for sufficient bone health management,¹⁶ resulting in the unacceptable adverse event profile of this regimen.

The unfavorable data from the ERA 223 ultimately led to an urgent safety letter, which encouraged the use of osteoprotective agents (OPAs, e.g. denosumab, zoledronic acid) in combination with Ra-223.¹⁹ The use of OPAs in prostate cancer is not a new idea. In fact, a phase III study has examined the relative efficacies of denosumab, a RANKL inhibitor, and zoledronic acid, a bisphosphonate, in patients with prostate cancer, finding that denosumab was superior for the prevention of SREs.²⁰ Distinct mechanisms of action help explain the differences in efficacy for denosumab and zoledronic acid; denosumab reduces osteoclast activity via feedback inhibition of highly expressed RANKL,²¹ whereas zoledronic acid inhibits the mevalonate pathway by inducing apoptosis in osteoclasts.²² Widespread use of denosumab is limited to patients at high risk for SREs because of the insignificant benefits to cancer-specific and overall survival, and the risk of jaw osteonecrosis with prolonged use.²³ Nonetheless, a sub-group analysis of ALSYMPCA, the pivotal phase III trial that demonstrated the overall survival benefit of Ra-223 in patients with mCRPC metastasized to bone, established that denosumab increases the time to symptomatic SREs when administered with Ra-223.²³ The previously mentioned early access, phase 3b trial also suggested increases in overall survival for Ra-223-treated patients receiving denosumab.¹²

Phase III trials investigating OPA administration in conjunction with Ra-223 and anti-hormonal therapy are currently ongoing. Approximately 40% of patients in each treatment arm of ERA 223 took a OPA concurrently (i.e. denosumab or a bisphosphonate),¹⁵ and OPA usage in either trial arm was associated with a lower risk of fracture.¹⁶ There was still no significant benefit to combined Ra-223 and AAP treatment over AAP alone, however, when symptomatic skeletal event-free survival was estimated with (HR: 0.932, 95% CI 0.666–1.306) and without concurrent OPA treatment (HR: 1.252, 95% CI 0.971–1.615).¹⁵ Preliminary findings from the PEACE III trial, a multicenter, placebo-controlled, open-label, randomized, phase III study examining co-treatment with Ra-223 and enzalutamide, demonstrated similar results to ERA 223, with patients in the co-treatment group experiencing a higher cumulative risk of fracture at 12 months than those receiving enzalutamide alone (37.4% vs. 12.4%).¹⁹ Amendments to the PEACE III trial were implemented in response to the ERA 223 study, requiring patients in both treatment arms to receive continuous administration of denosumab or bisphosphonate beginning ≥ 6 weeks before the first dose of Ra-223.¹⁹ As a result, the cumulative risk of fracture at 12 months dropped to 0% in both treatment groups.¹⁹

While the results of the PEACE III study have not been fully validated, it is striking that only one fracture has been observed since mandating OPA administration.¹⁹ The decrease in fracture risk was not as dramatic in the ERA 223 trial, and many of the patients who sustained a fracture while receiving Ra-223 and AAP (21%) or AAP alone (24%) were also taking an OPA.¹⁵ The summation of these data further suggests the significance of glucocorticoid administration in the ERA 223 trial, and thus, Ra-223 with enzalutamide might prove to be the safer combination. Validation of the efficacy results for the PEACE III trial will be important for discerning whether co-treatment with Ra-223 and anti-androgens is a feasible treatment option in the future. Data from a phase III trial investigating docetaxel with Ra-223 (NCT03574571) will provide further insight into treatment combinations for mCRPC, particularly the impact of glucocorticoids (co-administered with docetaxel) when used with Ra-223 in the absence of abiraterone.

The ERA 223 trial provides a cautionary tale as the role of both Ra-223 and OPAs evolve in the management of patients with mCRPC. The lessons learned from the ERA 223 trial will continue to impact ongoing studies, potentially leading to more optimized treatment strategies in the future.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Additional information

Funding

National Institutes of Health.