On Wednesday, Jan. 6, Forbes ran an article detailing a 7% drop in the price of Regeneron’s (REGN) stock price. The company attributed that decline to investors obtaining data from the FDA Adverse Events Reporting System (FAERS) showing eight suicide cases from patient’s taking Regeneron’s new PCSK9 inhibitor Praluent (alirocumab).
The issue, of course, was that the people who obtained those data knew nothing about FAERS or its myriad problems with case duplications, incomplete records, or how to value reports submitted with multiple drugs and/or from dubious sources.
All these Regeneron investors saw was eight suicide cases and started dumping the stock. Unfortunately for them (and all of the other Regeneron shareholders), there were NOT eight suicide cases from patients taking Praluent. There was one, and even that case report was dubious. And while not completely surprising, the article featured quotes by “status quo” adherers that cling to the misguided, and antiquated (to even leading industry journals ) view that FAERS is of no value:
“This is a classic overreaction to what is not a reliable source of information.”
“These kinds of spontaneous reports don’t help us very much.”
Regeneron’s problem wasn’t a result of bad FAERS data, but bad data curation and analytics. Two very different things.
Data’s value are only realized through the analytics created to manage and interpret those data. In Praluent’s case the difference was in how the data were read and by whom. The investors neglected to remove duplicate case reports within FAERS, a known problem and one which we’ve corrected for in our Adverse Events Explorer subscription site. Advera Health clients knew immediately that there was only one case of reported suicide and that it did not raise to the level of being an “Active Signal” based on our analytic review of the cases. Is a completed suicide (or suicidal ideation) serious? Absolutely, but it needs to be examined in the context of rate, disproportionality, and many other factors.
As the Forbes article rightly points out, even this one case was somewhat questionable due to the patient taking multiple other medications, including anti-depressants. And since the report came from a consumer directly and not a healthcare provider with knowledge of the case, assuming some causality is difficult.
Is raw FAERS data unreliable? Yes. Is properly curated and managed FAERS data an important adjunct to findings from clinical trials? Absolutely. Among other things, these very same FAERS data were able to dis-prove another rumor floating around before the holidays that this same class of drugs, the PCSK9 Inhibitors, were causing brain infections. You can see our analysis on that subject here.
The article concludes,
“The lesson: A little data can be a dangerous thing. Often, what looks like information in medicine is just noise. But, from a stock perspective, things get more complicated.”
No, the lesson here is that data management and curation matters. Data is a resource – it’s neither dangerous or benign. How you manage that resource and what you do with it is what makes it useful, useless, or “dangerous”.
While FAERS naysayers will likely never come around, many within the healthcare system already have and are actively using the data, analytics, and insights we provide to help lower costs, improve patient safety, and obtain a real-world view of emerging drug safety issues.
So the lesson of the day – before you start dumping stock or rallying behind one - is make sure you bet on the right data curation, management, and analytics and not just the data.
To learn more about how Advera Health’s analytics are being used to improve patient outcomes, lower healthcare costs and make better informed decisions, request a free trial of our Adverse Events Explorer application.
Brian Overstreet, President