Apollo 14 and the Lessons from Fractures

Earlier this year I wrote a piece on the need to learn from the ERA-223 experience and pitched it in the context of the Apollo 13 moon mission, dubbed a “successful failure’ because it revealed a variety of problems in the space mission and created the opportunity to revise process based on continued close scrutiny of the data*. At ESMO 2018, the closer scrutiny of ERA-223 has indeed delivered some new information about our best practices in mCRPC.

To review, the ERA-223 study comparing abiraterone and prednisone (AA+P) plus placebo to AA+P+ Radium 223 in mCRPC was unblinded last November upon a recommendation of the Independent Data Monitoring Committee (IDMC) based on the detection of a trend for improved survival in the placebo arm ( HR=1.347, p=0.02) and a significantly higher rate of fractures in the combination arm than the placebo arm ( 102 vs 32 fractures, respectively). Now we see that, with continued analysis of the data, this overall survival (OS) difference is no longer statistically significant (HR =1.195, p=0.128), however, the fracture differences remain. However, based on this continued analysis we may no longer need to fear that we are doing damage to our patients by giving them this combination.

Perhaps most enlightening was new insight into fracture risk in mCRPC patients emerging from the deeper dive into these data.  Of the 134 fractures that occurred, 102 occurred in the combination arm, and 32 in the placebo arm, still a significant difference. However, the concern may lessen when you consider that the incidence of fractures in the 102 patients in the combination arm who were receiving a bone-targeted therapy (BTT) was only 20% vs 80% in those who were not. Further, the rate of fractures in the placebo arm for those receiving BTT was virtually identical at 25% in those receiving BTT versus 75% for those who were not.

So, two stories are emerging. One is that radium may increase the risk of fractures, but also that the combination of radium with BTT may eliminate that risk.  Here’s my perspective on some of the key takeaways from this study.

  1. Radium treatment may lead to osteoporotic fractures in areas not affected by the tumor. One really can’t dispute the numbers I quoted above, that the risk of fractures did go up. The possibility exists that radium may do some damage to the normal remodeling of non-cancer affected bone. Could this be due to damage to osteoclasts and osteoclasts in healthy bone? They, like cancer cells, likely reproduce in a healthy environment and damaging their DNA may induce a deleterious effect leading to demineralization. Consider also my previous blog about accelerated aging and cellular senescence in the setting of radiation (or radium) exposure. Perhaps in healthy tissue, the use of radium is inducing cell damage, cell senescence, and a more rapid demineralization? As we like to say, this is an area where more study is warranted.
  2. Bone-targeted therapy reduces fractures without regard to radium use. this is as clear as the data above. BTT use is associated with about a 75-80% reduction in the risk of fractures. The benefits of BTT were observed in both arms of the study  If you are targeting bone mets with Radium, please also target bone strength with zoledronic acid or denosumab!
  3. These data give us new conceptual insight into tumor compartments and the targets of therapy. In a bone metastasis, three biological compartments co-exist: First is the tumor cell itself, second is the healthy bone it seeks to invade and third is the interface of the two – the beachhead of tumor invasion if you will. Given this perspective, the use of BTT may target purely the host bone, whereas the radium 223 targets the interface of tumor and host. Thus, it is possible that in this randomized trial the fractures occurred in the radium treated patients without BTT use because the BTT was not present to prevent or reverse the damage done to the normal bone by the radium 223.
Then why didn’t the radium add benefit to the abiraterone? It may be that the relative contribution of Radium to bone protection was simply not enough of an incremental benefit – that the abiraterone and prednisone were effective enough in this patient population. One wonders if the study could have been more effective if the radium was added at the time of serologic progression in an effort to reduce the development of new mets, as opposed to being added at the beginning of the treatment. In the beginning, there is no beachhead to target because it is in retreat, at the time of serologic progression that interface between potentially growing tumor and bone becomes the battleground.

  1. BTT and Radium are qualitatively distinct. It has been stated previously that Radium is superior to BTT’s because its use led to a survival advantage in addition to a reduction in SRE’s. Many may have used this as a rationale for using it over BTTs, citing cost and adverse events as additional rationales. We now effectively see that the two classes of drugs are qualitatively different.
  1. When we hear that an interim study is negative, we need to be cautious but don’t walk away from the data. Dig deeper and keep analyzing. There may be some valuable information there. From that perspective, this study can be considered a successful failure.
Take home message, if you plan to use Radium, don’t forget about bone health – add a BTT.

*On February 5, 1971, the Apollo 14 mission landed safely on the moon. It was safe and, by lunar exploration standards, boring. With gratitude to the lessons learned from Apollo 13, I’m sure.

Written by: Charles Ryan, MD, B.J. Kennedy Chair in Clinical Medical Oncology, Director and Professor of Medicine, Division of Hematology, Oncology and Transplantation, University of Minnesota, Minneapolis, Minnesota

Published Date: November 12th, 2018

Further Related Content:
Watch: A Renewed Analysis of ERA 223 - Fred Saad