A Legal Tussle Over Adverse Event Disclosure
14 CommentsBy Ed Silverman // August 31st, 2010 // 8:09 am
Should drugmakers disclose all adverse event reports that may not show statistically significant evidence that a side effect is actually caused by a specific drug? The issue is now before the US Supreme Court involving a case brought by investors against Matrixx Initiatives, which was sued for allegedly concealing side effect reports that its Zicam cold med caused people to lose their sense of smell, known as anosmia.
Not surprisingly, BayBio, a trade group representing a goodly number of life sciences companies, has waded into the fracas with an amicus curiae, or friend-of-the-court brief, decrying the notion that drugmakers and biotechs should be forced to disclose adverse event reports that companies believe are irrelevant to the performance of their meds. “Laws requiring disclosure of anecdotal evidence can result in erroneous conclusions about a treatment’s safety and effectiveness,” BayBio argues.
UPDATE: PhRMA and BIO have jointly filed their own brief, in which they argue a “reasonable investor would not base investment decisions on statistically insignificant reports of adverse events, which are unlikely to threaten a drug’s sales” (read it here).
The underlying legal battle is complicated and intriguing. For their part, Matrixx shareholders are concerned Matrixx will succeed in obscuring the notion of reporting adverse events and, effectively, obtain a decision that could require drugmakers to report a tidal wave of adverse events, many of which may be irrelevant to a lawsuit. You can read all about it here and then digest the BayBio brief.
Nathan
The brief makes for an interesting read. Here’s a particularly interesting passage that is about halfway through:
“When doctors and consumers learn of reports of adverse events (particularly from the company itself), they understandably become reluctant to prescribe and use those drugs, even when there is no evidence from scientific studies to support any association between the drug and the adverse event.
Amicus’s members already have difficulty finding volunteers for clinical trials. Without these trials, the approval process for potentially life-saving drugs can be delayed. In “The Adverse Side Effects of Pharmaceutical
Litigation” 7 (2003) (in a poll of health care
professionals and patients, “[a] sizable number of physicians (43%) have avoided prescribing a particular drug that was appropriate for a patient because they were aware that it might be involved in product liability litigation”). And companies, rather
than face potentially staggering liability under United States securities laws, now have an incentive to prematurely disclose such reports, even if it is not scientifically responsible to do so.
For example, rather than wait for a study to be
carefully vetted, pharmaceutical companies Merck & Co. and Schering-Plough Corp. announced that preliminary results of a clinical study showed an increased risk of cancer in patients taking Vytorin. This announcement caused significant same-day declines in the prices of the two companies’
shares. A few months later, however, the researchers published the study’s full results, and the FDA concluded it was “unlikely” that Vytorin increases the risk of cancer. Vytorin remains on the market today.
In making the early disclosure, the companies
wanted to inform regulators and investors
immediately. But the companies’ decision to disclose the results of the study before it had been carefully vetted by outside experts was widely criticized. A Merck spokesman explained that the criticism “makes us feel like we’re damned if we do and we’re
damned if we don’t”
There was a significant concern that patients or doctors would make health decisions based on inconclusive or erroneous information.
Nathan
Woops — ignore the part about clinical trials. I accidently copied a footnote into the passage above.
Condor
I started writing my reply to just that portion, Nathan, and then you came along here — and helpfully highlighted it for all to see. Thanks!
Here is my general rejoinder (with graphics!):
BayBio uses the Merck-Schering-Plough SEAS cancer signal debacle to suggest that “premature” disclosure of “anecdotes” will not serve the investing, or prescription-consuming, public. The SEAS cancer signal was not “anecdotal”, in any sense of the word. The signal was observed during a clinical trial of the drug — not simply doctors or patients “phoning in” aftermarket adverse event reports. It does a disservice to the Court to imply otherwise. [We have previously discussed various proactive, premptive ways of solving SEAS-like crises.]
Moreover, and perhaps more importantly — as the below trading chart plainly indicates — Schering-Plough chose to delay its earnings call, and then chose to announce the SEAS signal in the middle of the NYSE trading session, rather than waiting until close, or seeking a temporary NYSE trading halt.
Either of these latter approaches would have allowed for time — time for calm, rational investors to react to the news, digest it, put it in perspective — and then react in a measured fashion. What actually transpired was a real time, largely unfounded, panic selling session.
In making this particular argument by analogy, BayBio is no “friend of the court” — it is, in fact, simply misleading the Supremes (click to enlarge) — note that it was the delay in Schering-Plough’s Q2 2009 earnings announcement that started the panic — not the “SEAS cancer signal news”, itself.
See it all, including the trading charts mentioned, at my site.
Namaste
Justice in MI
I assume the issue here is disclosure on the label, as opposed to sending Medwatch reports to FDA. In the latter contexts, all AEs are required to be reported, and explicitly without any speculation re: causality. That’s the only way we find out.
In the meantime, legal briefs are legal briefs, not policy studies. In general, studies consistently show that the “dangerous over-warning” argument is not supported by the facts. With rare exceptions, companies do _not_ defensively label, despite the common assumption (and logical surmise) that they do.
This was recurrently noted in internal FDA communications that came to light in connection with the Levine caswe.
Condor
No, JiM — not FDA — the argument here is that Matrixx owed the investing public a materially complete disclosure. A disclosure that its Zicam cold remedy was allegedly responsible for a permanent loss of sensation — the loss of the sense of smell — in an increasing number of people taking the cold-remedy.
So it is not an FDA disclosure issue — it is an SEC fraud on the stock market/investing public disclosure issue.
When these “loss of smell” AERs became known, Matrixx shares plunged almost 80 percent in two trading sessions. Investors sued. And here we are.
Just to keep it straight.
Namaste
Justice in MI
Thanks, Condor. Yeah, I asked the question before reading the brief….
Not surprising that much of it is warmed-over FDA preemption stuff though–FDA’s expertise at deciding what’s important and what isn’t, etc. etc.
When the conservatives take over, and they move again to dissolve FDA entirely (as in the mid-90s), we’ll probably be hearing less about the agency’s “expertise.”
Nathan
Condor, after reading your comments I’m really unsure of your position. Do you support the release of tox data that is not statistically significant or deemed non-relevant to the drug?
I’m on the fence about this one. I’m leaning towards “the more data, the better” — the public (and doctors) will quickly learn that interpritation of these adverse events is nontrivial. All drugs come with loads and loads of baggage. If the public wants to read about the minutia of epidemiology around a drug, let them go for it! I’m not convinced it will solve anything or really hurt anything.
Condor
Hi Nathan — I do support the release of such data.
Importantly, though, that is NOT what this Matrixx case is about — precisely speaking. This case is about whether — as a matter of federal securities law — the disclosure of AERs should be REQUIRED (not just permitted).
The most provacative version of this question would ask whether statistically-insignificant AERs must be immediately disclosed.
On the other side of the ledger, recall that Merck figures in this jurisprudence, as well [though curiously, BioBay makes no mention of this part of the history(!):
Several researchers have documented that Merck might have known as early as 2001 that a statistically significant increased heart attack risk was associated with the Vioxx arthritis drug. Of course, Merck waited until late 2004 to withdraw the drug from the market, and ended up paying $4.85 billion to settle injury claims.
So -- I think once the level of AERs approach statistical significance -- disclosure probably should be mandated by FDA (if not on label copy, at least in a "dear doctor" letter format). And, if the product is "material" -- a term of art, in SEC parlance, then the AERs ought to be disclosed in the company's SEC filings, as well.
[I'll likely write a follow-up post, at my site tomorrow, time permitting, on these later themes.]
Does that clarify things for you?
Namaste, to all of good will.
mike
there are multiple problems hidden in this discussion. One issue is OTC products that people over use / abuse. My friends and relatives take far more OTC products than they should and I try to get them to understand that they can cause healthby over using these substances. Did the person have a negative reaction from normal use or did they really over use the medication. We have this personal responsibility / knowledge issue. On the other side we have new perscription drugs. We do not have a lot of data when I product hits the market placee so as people use the product we will learn abut some side effects. If the products generate several complaints the difficult question becomes when is it significant. This is not a simple matter it is part of the learning curve. If a company knowingly holds back overwhelming data that is another issue.
Nathan
Condor writes: “So — I think once the level of AERs approach statistical significance — disclosure probably should be mandated by FDA”
That sounds great — but in practice this is useless. You either trust the company (and FDA) or you don’t. If you don’t, then ALL data has to be released and the interpritation of it must be done in the public sphere. “Approaching statistical significance” in one person’s mind is “completely insignificant” in someone else’s mind. Also, with AERs, I don’t think that there is a good way of measuring “statistical significance”. Rather, you are measuring correlations. Correlations are far more “fuzzy” than statistics.
In my mind it is “all or nothing”. Either decide to trust the company’s (or FDA’s) interpritation of the data, or open the entire process up and make *all* AERs subject to public disclosure.
Condor
Well, Nathan — there are rules on this, at FDA (this is not a new area at FDA; it is “new” at SEC per my earlier commentary) — and it is a difficult calculus, to be sure — but I am never fond of “black/white” ultimatums.
These companies make billions in profits by assessing risk to reward ratios — it’s what they do.
To tell me it can’t be done is to engage in sophisrty.
Try a little harder.
Namaste
Nathan
Try a little harder? Ha! Can you be any more trite? We throw billions and billions of dollars into R&D and fall flat on our face with no drug over 90% of the time. Even the drugs that do make it are loaded with unknowns that only come up after years and years of study, more clinical trials, and analysis after analysis of AERs. Even OTC drugs approved for decades are frequently flagged with new problems.
If “try a little harder” is your best answer, maybe you should get a job in this industry and come try it yourself! I say that ALL the data should be released publicly if you want to “try a little harder” — then fine, do it! Maybe you have a little spare time in the evenings to crunch through the numbers and come up with something useful that hundreds and hundreds of industry experts overlooked.
Sorry, I’ll get off my soapbox…
Condor
Actually Nathan — I am a 20 year industry vet — and I meant that you should “try harder” to frame persuasive arguments about why it cannot be done.
Just because you say so, does not make it true. Persuade us. Don’t just declare it so by fiat. That is what I mean when I tell you, personally, to try harder.
I’ll stop here, as we are likely too far apart to hold much more meaningful dialogue.
Namaste
JaT
PhRMA, are you kidding? So as not to scare off investors?
Consumers are paying to prove your new products - sometimes to their own detriment.
I don’t want to provide you with beneficial information after you have harmed me (sounds selfish doesn’t it?). You don’t want to provide me with information to allow me to determine if I want to take a chance (sounds non-disclosurish doesn’t it?). Let’s not pretend new drugs are not ~taking a chance~.
Nathan is right on a couple of points.
You either trust the drug maker and FDA or you don’t. I want the data. I believe consumers are paying for it in high drug costs and with their bodies.
He is also right that most people will not make use of the info. They trust that their doctors know what they are doing and will not cause them harm - giving little thought to the industry or the regulatory body placed to protect the public.
I had to chuckle a little reading mike’s comment (felt a collective cringe). Didn’t he get the memo? Shhh, man, if the public catches wind of how it works the industry is screwed.