乌鸦传媒

MUNICH -- Men with bone-predominant metastatic castration-resistant prostate cancer (mCRPC) had worse outcomes when radium-223 (Xofigo) was added to abiraterone (Zytiga) and prednisone, according to the randomized ERA-223 trial reported here.

The study ended early after an interim analysis showed that the three-drug combination was associated with worse survival and a substantially higher fracture rate (26% versus 10% in placebo group), reported Matthew Smith, MD, PhD, of Massachusetts General Hospital Cancer Center in Boston.

On average, survival without a symptomatic skeletal event (SSE), the primary endpoint, was about 4 months less among those who received radium-223. Overall survival also appeared worse in this group.

And an analysis of treatment-emergent adverse events (TEAEs) showed almost three times as many fractures in the radium-223 groups, as reported at the European Society for Medical Oncology.

"Based on data from this study, the use of radium-223 in combination with abiraterone is not recommended," said Smith. "Prescribing information for radium-223 has been revised, based on these findings."

Use of bone-health agents substantially reduced the risk of fracture in both treatment arms, he noted.

Invited discussant Daniel Heinrich, MD, of Akershus University Hospital in Lørenskog, Norway, suggested that the trial "never had a chance of being positive." He pointed out that the assumed 21-month SSE-free survival in the control arm was 4 to 5 months lower than the combination of abiraterone and prednisone achieved in a prior placebo-controlled trial. In the ERA-223 trial reported by Smith, the abiraterone-prednisone combination led to a median SSE-free survival of 26.0 months, consistent with the prior results.

The ERA-223 trial also assumed a 39% improvement in SSE-free survival in the radium-223 arm versus the control arm, or a median SSE-free survival of 29.2 months.

"Even if the experimental arm had provided 29.2 months, or 8.2 months extra, the trial statistically would have been negative, because the assumption of 21 months was too low," said Heinrich.

Eleni Efstathiou, MD, PhD, of MD Anderson Cancer Center in Houston, said the study emphasized the importance of bone-directed therapy in the management of mCRPC. She noted that 60% of SSE events occurred at sites without metastases, and that incidence was reduced by 60% in patients who received bone-targeted therapy.

Radium-223 received FDA approval primarily on the basis of data from the placebo-controlled phase III ALSYMPCA trial. The study showed a 3.6-month improvement in overall survival (OS) and almost a 6-month improvement in time to SSE in men with mCRPC and bone metastases treated with radium-223 and prednisone.

The decision to pair radium-223 with abiraterone was based on several lines of evidence, said Smith. Abiraterone improved both progression-free survival (PFS) and OS as first-line therapy for mCRPC; radium-223 and abiraterone have non-overlapping toxicity; and a of a phase III trial suggested a survival benefit with the addition of radium-223 to abiraterone or enzalutamide (Xtandi).

Eligible patients had asymptomatic or mildly symptomatic bone-predominant (at least two bone metastases) mCRPC, no prior chemotherapy for CRPC, and no known brain or visceral metastases. They were randomized to abiraterone and prednisone/prednisolone plus placebo or radium-223. Use of bone health agents was allowed only for patients who were already on the agents at enrollment.

The primary endpoint was SSE-free survival, defined as freedom from use of external-beam radiotherapy to relieve skeletal symptoms, new symptomatic pathologic bone fractures, spinal-cord compression, or tumor-related orthopedic surgical intervention.

Data analysis included 806 patients, and unblinding occurred after an interim analysis showed that more deaths and fractures had occurred in the radium-223 group. All patients had completed study-specified treatment at that point, Smith said. Study procedures and treatment continued, and the protocol was amended to allow bone-targeted therapy in all patients.

The data showed a median SSE-free survival of 22.3 months in patients randomized to radium-223 and 26.0 months in the placebo group (HR 1.12, 95% CI 0.92-1.37, P=0.263). Death prior to SSE and other components of the primary endpoint occurred in a similar proportion of patients in each group, except for pathologic fracture, which occurred in twice as many patients in the radium-223 arm. Median overall survival was 30.7 months with radium-223 and 33.3 months with placebo (HR 1.20, 95% CI 0.95-1.50, P=0.128).

Analysis of TEAEs showed that 103 patients in the radium-223 group had a fracture, as compared with 38 in the placebo group. Independent review of patients with fractures showed that 76 patients in the radium-223 group had at least one fracture versus 23 in placebo group. All subcategories of fractures occurred more often in the radium-223 arm:

• Pathologic: 19 versus 6
• Traumatic: 27 versus 13
• Osteoporotic: 37 versus 4

As Efstathiou pointed out, patients on bone health agents at baseline had a substantially lower incidence of fractures (11% versus 29%).

Comment