The State of Bone Health in mCRPC: ERA-223 Results Revealed

At the 2018 meeting of the European Society of Medical Oncology (ESMO), Matthew Smith, MD, PhD, presented the highly anticipated results of ERA-223, a phase III trial of abiraterone acetate prednisone (AAP) combined with radium-223 or placebo in men with asymptomatic or minimally symptomatic metastatic castration-resistant prostate cancer (mCRPC)1. Radium-223 is the alpha- emitting radiopharmaceutical that achieved wide regulatory approval since 2013 for the treatment of men with symptomatic bone metastatic CRPC based on the ALSYMPCA study, which demonstrated a statistically significant improvement in overall survival 2. Based on an interim safety analysis of ERA-223 in late 2017, a press release from the sponsor warned of an increased risk for fractures and lower overall survival associated with the concomitant use of AAP and radium-223 versus AAP and placebo. 

Since then we have seen two very different responses to this data from the FDA and EMA. The FDA (along with regulatory authorities from Japan and Canada) issued a new warning advising against the concomitant use of radium-223 with AAP, without changing the indication, while the EMA dramatically limited its label to only patients with mCRPC progressing after at least two prior lines of systemic therapy for mCRPC (other than LHRH analogues), or ineligible for any available systemic mCRPC treatment.  So what did the data show?

Consistent with last year’s press release, the initial management of mCRPC patients with AAP and radium-223 did significantly increase the risk of subsequent fracture, approximately 2.5 fold over AAP and placebo (26% vs 10%).  Importantly, patients who were on bone strengthening agents (BSA) like denosumab or bisphosphonates experienced roughly half the rate of fractures on both arms (15% and 7%, respectively). However, there was no statistically or clinically meaningful difference in overall survival between the two groups with a hazard ratio (HR) of 1.195 (95% confidence interval 0.95 -1.505; p-value 0.128). There were also no other significant differences in the safety profile reported, or differences in secondary endpoints including radiographic or PSA progression. Interestingly, the symptomatic skeletal event-free survival (the primary endpoint) was not statistically different between the two arms (HR 1.12), certainly not 2.5 times worse. So how do we interpret these results?

First, these fractures are real. But these mostly asymptomatic fractures have a different clinical implication than skeletal events which tend to be symptomatic and require surgery and/or radiation therapy. Could asymptomatic fractures reflect the overall health of patients or be a sign of frailty? More work needs to be done, but one thing we can say is that these events are happening at higher rates in patients who are not on a BSA.

Second, we should be alarmed at the low use of BSA (roughly 40% overall) at enrolment in this bone mCRPC population. Zoledronic acid and denosumab are both indicated for these patients and are likely underutilized early in our management of these patients, myself included.

Personally, ERA-223 doesn’t change my use of radium-223. I have never used them together in untreated mCRPC. I will still use radium-223 to improve survival in our bone metastatic CRPC patients, but I will be sure to start a BSA beforehand barring other risks. More importantly, ERA-223 is a wakeup call to proactively protect the bone health of our patients given the increased risks associated with our newer life-prolonging agents.

As for the label changes to radium-223 based on this study, I believe the FDA got it right.  

Written by: Daniel J. George, MD, Medical Oncologist, Professor of Medicine, Professor of Surgery, Duke Cancer Institute, Durham, North Carolina

Published Date: November 6th, 2018


1. Smith M, et al. ESMO 2018
2. Parker C, et al. N Engl J Med. 2013;369:213–23.

Further Related Content:
Watch: ERA 223: Presentation Summary from ESMO 2018 - Fred Saad
email news signup