ESMO 2018: ERA 223 - A Phase 3 Trial of Ra-223 in Combination with Abiraterone Acetate & Prednisone for the Treatment of Asymptomatic or Mildly Symptomatic Chemotherapy-Naïve Patients with mCRPC

Munich, Germany ( As has been the trend in management of advanced solid malignancies, there has been increasing interest in combining proven therapies to identify synergistic effect – survival benefit greater than any one drug alone could provide. This is particularly useful if the mechanism of action of these combined agents targets different pathways.

In the setting of chemotherapy-naïve castration-resistant prostate cancer (CRPC), there are now numerous approved therapies that have changed the landscape of advanced prostate cancer management. Abiraterone acetate (+prednisone) (Abi+P/AAP) is an established oral androgen-axis targeted therapy which has been demonstrated to improve progression-free and overall survival in this patient population (COU-AA-301 and COU-AA-302).  Radium-223,  a targeted  alpha-emitter, approved for mCRPC with bone metastases has an OS  benefit (HR  0.70) and the ability to decrease symptomatic skeletal events (SSEs) (HR  0.66). Radium-223 has multiple methods of action, working not just on the tumor cells but also osteoblasts/osteoclast. It was approved in 2013 in Europe and USA based on the strength of the ALSYMPCA study results.

The combination of these two agents was chosen for a few reasons:

  1. AAP improves radiographic progression-free survival and OS in first-line treatment of mCRPC
  2. They do not have overlapping toxicity profiles
  3. Post-hoc analysis of an international, early-access, phase 3b open-label study suggested a survival benefit of Ra-223 + Abi/Enza over Ra-223 alone
In this trial, NCT02043678, the authors combined both these agents in a phase 3, double-blind, placebo-controlled randomized trial. The specific patient populations were asymptomatic or mildly symptomatic men with chemotherapy-naïve mCRPC and bone metastases (>2 bone metastases) in a 1:1 ratio of AAP + Ra-223 or AAP + PBO. AAP was dosed at 1000 mg daily + 5 mg prednisone/prednisolone twice daily. Ra-223 was dosed at 55 kBq/kg IV weekly for 6 weeks. As AAP is already established and a standard of care, it was included in both arms.

The addition of Ra-223 was tested specifically. As this affects bone metastases, the use of approved bone health agents (BHAs [bisphosphonates or denosumab]) were only allowed in patients receiving them at baseline. The primary endpoint was SSE-free survival (SSE-FS) – which included the use of external beam radiation therapy (EBRT) to relieve skeletal symptoms, new symptomatic pathologic fracture, spinal cord compression, or tumor-related orthopedic intervention. The study protocol is listed below from Dr. Smith’s slide set:

UroToday ESMO2018 ERA223 NCT020443678

In the study, 806 men were randomized (401 to AAP + Ra-223 arm and 405 to AAP + placebo arm).

  • The groups were evenly matched for baseline characteristics
  • 39% and 42% were receiving BHAs at baseline in the Ra-223 placebo arms, respectively
  • Median age 71
  • Primarily Caucasian (71%)
  • 58-58% were Gleason 8-10 at diagnosis
  • 65-66% had >5 bone metastases
  • Median PSA ~30
At the data-safety monitoring board evaluation in November 2017, due to an increased number of fractures and deaths noted in the Ra-223 arm, the study was unblinded early. They had not yet met their planned endpoint of 339 events. Importantly, all men had already completed study-specified Ra-223/PBO treatment prior to unblinding. Patients were continued to be followed from this point on.

In terms of outcomes, the primary endpoint was not met - median SSE-FS was 22.3 mo (95% CI 20.4−24.8) with AAP + Ra-223 and 26.0 mo (95% CI 21.8−28.3) with AAP + PBO, which translated to a 22% increased risk of skeletal events with Ra-223. (HR 1.122, 95% CI 0.917−1.374; p=0.2636). Fractures occurred in 29% and 11% of pts in the AAP + Ra-223 and AAP + PBO arms, respectively. An important assessment is that many of the differences were actually NOT pathologic fractures, but rather related to osteoporosis (37 in Ra-223 arm vs. 4 in the placebo arm).

In men receiving BHAs, 15% and 7% experienced a fracture in the AAP + Ra-223 and AAP + PBO arms, respectively, vs 37% and 15% without BHAs. This would seem to suggest, though not statistically assessed, that BHA’s helped reduce the deleterious effect of Ra-223.

As for overall survival, outcomes were not significantly different, probably primarily driven by the AAP - 30.7 mo (95% CI 25.8−not estimable) with Ra-223 and 33.3 mo (95% CI 30.2−41.1) with PBO (HR 1.195, 95% CI 0.950−1.505; p=0.1280).

Other secondary/exploratory endpoints were also not significantly different. These included: recurrence PFS, time to cytotoxic chemotherapy, time to opiate use for cancer pain, overall confirmed PSA response, time to PSA progression, overall confirmed ALP response, time to ALP progression, and time to deterioration in health-related quality of life.

In terms of adverse event profile, except for fractures, both groups were not statistically significant – a similar number of minor (Grade 1-2) and major (Grade 3-5) adverse events.

Invited Discussion

Following Dr. Matthew Smith’s presentation, Dr. Daniel Heinrich provided a brief discussion and highlighted the following points. 

His goals were simple:

1) Why did this study not meet its endpoints?
2) What implications does this have on clinical practice?

First, he went through and compared the placebo arm results to prior clinical trials to ensure that they matched appropriately – effectively ensuring that this trial got appropriate comparator arm results.

  • AAP+placebo arm OS results were compared to COU-AA-302 results - appropriate
    • OS in the placebo arm of this study: 33.3 months
    • OS in the COU-AA-302 study: 34.7 months
  • What about the primary endpoint of SSE-FS? This was not assessed in COU-AA-302
  • In COU-AA-301 (albeit a slightly worse disease stage, as these patients were treated with docetaxel already), the time to first skeletal event was actually 25 months
    • However, the authors set up this study to have an ~ 21-month SSE-FS in the placebo arm
    • Their placebo arm actually had an SSE-FS of 26 months, which is much more in line with the COU-AA-301 data
    • So, even if they had met their target improvement of 29 months, it would not have been significant
He next focused on the finding of increased fractures. He highlighted the basic concept that adding more agents naturally adds more toxicity. In these patients, we are adding both agents in the first 6 months of diagnosis – yet, as you see in the COU-AA-302 data below, the curves don’t even split (OS) until about 15-18 months. So, we may be adding the Ra-223 too early in treatment – adding side effects with no expected benefit.

UroToday ESMO2018 COU AA 302 Overal Survival

This is supported by the fact that most of the fractures in the ERA 223 study occurred in the first 6-12 months (from a slide that Dr. Smith did not present).

How does this impact management?

  1. AAP+Ra-223 should not be first-line therapy for mCRPC with bone metastases
  2. However, there is still a biologic rationale for therapeutic layering of Ra-223 once patients begin to progress on Abi/Enza (where the curves split in the COU-AA-302 trial)
  3. BHAs are underutilized and should be strongly recommended in addition to Ra-223

Presented by: Matthew R. Smith, MD, Ph.D., Director, Genitourinary Malignancies Program, Massachusetts General Hospital. Boston, US
Invited Discussant: Daniel Heinrich, Consultant, Medical Oncologist at Akershus University Hospital Lørenskog, Norway

Written by: Thenappan Chandrasekar, MD, Clinical Instructor, Thomas Jefferson University, twitter: @tchandra_uromd, @TjuUrology at the 2018 European Society for Medical Oncology Congress (#ESMO18), October 19-23,  2018, Munich Germany