Kevin Lomangino is the managing editor of HealthNewsReview.org. He tweets as @KLomangino.
Update: Scroll down for comments from Larry Husten, a veteran journalist who covers cardiovascular medicine, about the issues raised in this post.
No sooner had I pressed the “publish” button on this post complaining about Amgen’s data-free promotion of a “landmark” study on its cholesterol-lowering drug Repatha, when here comes Bayer touting the “overwhelming efficacy” of its blood-thinner rivaroxaban (Xarelto) – again without providing a shred of supporting evidence.
According to the release:
Following a planned interim analysis conducted by the independent Data Monitoring Committee (DMC), the DMC recommended to stop the trial early as the primary MACE endpoint has reached its prespecified criteria for superiority. Owing to the magnitude of effect and the confirmation of the existing safety profile of rivaroxaban, Bayer, Janssen and the Population Health Research Institute (PHRI) will offer rivaroxaban to study participants in an open-label extension trial.
I again feel compelled to point out that this is marketing–not useful information. Neither Bayer nor the study investigators should make claims about “overwhelming efficacy” of Xarelto without providing the data to back up those assertions.
All we get is a vague assurance that the data will be presented at “an upcoming medical meeting in 2017.”
It took me only a few moments of poking around on Twitter to find an independent expert voice who was critical of Bayer’s release and skeptical of the “overwhelming efficacy” claims.
Why is this the new way to break trials? Much hype but no data to support. EBM pearl: if need to randomize >25K pts, effect size likely tiny https://t.co/NKWVR28Vf0
— Anil Makam, MD (@AnilMakam) February 9, 2017
Makam, an internist and assistant professor at UT Southwestern Medical Center, is pointing out that the trial is huge by almost any standard. And such an enormous study is likely capable of finding an effect that is statistically significant but clinically nothing to write home about. A much smaller and cheaper study likely could have established a meaningful benefit with fewer patients. So there’s good reason to wonder how big the benefit actually is. Which begs the question: Why pre-hype the study before we have those specifics?
The question of safety also is paramount. Bleeding is always a concern with blood-thinning medications, as noted in this story about the announcement by journalist Larry Husten on the Cardiobrief blog. The release says the data offers “confirmation of the existing safety profile.” Will independent experts agree?
These are just my off-the-cuff observations – an attempt to slow down the hype train before it gathers too much steam. We’ve assigned the release to be systematically reviewed by our team of expert reviewers and we’ll share their assessment when it’s complete.
I wrote to Larry Husten, a veteran journalist who covers cardiovascular medicine, for his take on how companies should release potentially market-moving results such as Bayer’s COMPASS study or Amgen’s FOURIER study (which I discussed separately in this post) and how journalists should cover such announcements.
For background, let’s recall that companies must meet regulatory requirements to notify investors when studies such as this meet their primary endpoint. These companies also want to preserve the juicy details from the study for presentation at a high-profile meeting such at the American College of Cardiology conference. But if an announcement must be made, should companies err on the side of transparency and release the full results when they claim that their study has succeeded and that the drug has “overwhelming efficacy”? Here’s what Husten had to say:
“I’ve been wrestling with this general issue for many years. It’s a perfect example of competing rights and obligations. I am convinced that there is no one perfect solution, so the best we can do is reach some sort of reasonable compromise with which most parties can exist.
I don’t think it’s feasible, in the current environment, to expect the company (or the trial investigators) to release all the study data, as you suggest. I’m well aware of the limitations of meeting- and publication peer review, but this would lead to the release of entirely unvetted data.
But I completely agree with your idea that companies should absolutely not be allowed to include marketing hype in these situations. These news releases should be pro forma and neutral. In general, a press release about a trial stopped for positive findings should only state that the primary endpoint was met and that no safety issues were found. If a trial turns up any safety issues, especially if the drug is already on the market, more information may be necessary.”
Husten also noted that journalists working on deadline will have a tough time providing context for surprising announcements such as Bayer’s news release about the Xarelto COMPASS trial. He said that he was fortunate to have already been working on a story about Amgen’s Repatha trial when the news about the study broke. He knew that the study was scheduled for presentation at the ACC meeting next month and already had interviews lined up. That’s why his coverage of the Repatha study was more detailed than his Xarelto story, he says
“In fact, I even speculated on Twitter that Amgen might release the top line results [on Repatha] in conjunction with their quarterly results. So my story was nearly ready to go when the press release came out, and I was able to publish shortly afterwards a really detailed story based on several days of interviews and research.
By contrast, I had no early warning about COMPASS and it was not on anyone else’s radar, as far as I can tell. So in that case I wrote a brief, relatively superficial story in order to just get the news out. This is a good example of how press releases can drive the news cycle. Most organizations can’t do a researched, detailed news story in a situation like this.”
I sympathize with journalists who may feel obligated to cover such announcements for their investor or clinical audiences. Although we criticized a Reuters story for not questioning Amgen’s optimistic framing on Repatha, we understand that such stories must be written. The question is whether they can be framed more cautiously and conservatively, and include at least a few lines about what is not known and what might be too good to be true.