FDA does recognize, however, the important public health and policy justification supporting dissemination of truthful and non-misleading [information] on unapproved uses of approved drugs and approved or cleared medical devices to healthcare professionals and healthcare entities. Once a drug or medical device has been approved or cleared by FDA, generally, healthcare professionals may lawfully use or prescribe that product for uses or treatment regimens that are not included in the product’s approved labeling. . . . These off-label uses or treatment regimens may be important and may even constitute a medically recognized standard of care. Accordingly, the public health may be advanced by healthcare professionals’ receipt of [information] on unapproved new uses of approved or cleared medical products that are truthful and not misleading.
FDA Guidance on Good Reprint Practices for the Distribution of Medical Journal Articles and Medical or Scientific Reference Publications on Unapproved New Uses of Approved Drugs and Approved or Cleared Medical Devices (available here).
We’ve called these restrictions unconstitutional because truthful scientific speech is protected by the First Amendment from governmental suppression on the basis of its content. We’ve proposed alternatives that would encourage more, not less, truthful speech about off-label uses while providing incentives to add new uses to FDA-approved drug labeling.
We’re thinking all these thoughts again, for a couple of reasons. The first reason is what appears to be (at least to the limited extent that we follow medical science) the most compelling off-label use situation since we began blogging more than two years ago. Over the last couple of weeks there’s been a rash of stories in the media (as this Google search demonstrates) that a drug called Zileuton – approved by the FDA for the treatment of asthma – might be effective in treating, or even preventing, chronic myeloid leukemia (“CML”), which is a common form of this disease. The coverage even reached northern Maine, where Bexis was on vacation.
CML is a serious condition – a killer. According to the National Cancer Institute:
The median survival is 4 to 6 years, with a range of less than 1 year to more than 10 years. Survival after development of an accelerated phase is usually less than 1 year and after blastic transformation is only a few months.
The same site says that there are about 5,000 new cases a year, with somewhat under 500 deaths, in the US alone.
We’ll leave the medicine and science to doctors and scientists, but insofar as we understand it, apparently the same gene (called “Alox5”) creates the chemicals that cause inflammation (why it's important to asthma) and also those that are essential to leukemia cancers. Thus, the researchers who discovered this dual genetic activity tried Zileuton – which inhibits the activity of this gene – on special mice that had CML (they were at the Jackson Laboratories, which is pretty much the mother ship of "knock-out" mice). They found that Zileuton worked better than the most effective currently approved drug for CML - at least in mice.
From what we gather, doctors who treat leukemia consider this development a least a potential medical breakthrough.
It’s also an off-label use. That’s why we’re intentionally not linking to the website (which we easily found) of the manufacturer of Zileuton. We're self-censoring because we don’t want to do anything that might get the manufacturer in trouble with the FDA.
We don’t have any independent knowledge about Zileuton, but if Wikipedia is to be believed, that drug's been around since 1997. Given how long drugs typically spend in development, this probably (we don’t know, we’re guessing) means that it was patented several years earlier. Drug patents typically run seventeen years, which means that Zileuton’s patent might well expire before clinical trials designed to confirm the recent research could be completed. With imminent generic competition, there's no incentive to spend the megabucks that such trials cost.
So the use of Zileuton to treat CML is likely to remain an off-label use, even if the initial research is borne out in clinical practice, and this drug turns out to be the best treatment (and perhaps even a prophylactic) for a pretty common form of cancer with a high death rate. That, in turn, means that it’s quite possible we’re looking at an eventual off-label use/standard of medical care treatment.
The off-label status of the use prohibits "promotion" of the use by the manufacturer. So the manufacturer of the Zileuton – which by statutory requirement receives all adverse drug incidence reports from doctors – cannot, for fear of illegal “promotion,” truthfully inform the public about the risks and benefits of Zileuton in treating CML. And didn’t somebody state recently that, “manufacturers have superior access to information about their drugs, especially in the postmarketing phase as new risks emerge.” Wyeth v. Levine, 129 S. Ct. 1187, 1202 (2009)? It seems we’ve heard that name somewhere.
For this reason, we think the FDA’s restrictions on the manufacturer truthfully telling the public what it learns about off-label uses are absurd – and dangerous. We’re dealing with a potential medical breakthrough involving a drug that’s immediately available for off-label use in life-or-death treatment decisions. It may be better than anything else available; it may not. Nobody knows the optimal dosage regimen for the treatment of CML. It may be different from the on-label treatment for asthma. But we expect that a lot of doctors treating possibly terminally ill CML patients are going to try to find out over the next few years.
And if something goes wrong with this off-label use, the manufacturer’s going to be sitting out there with a litigation target on its back, no matter what it does.
And that brings us to the second reason we’ve been thinking about off-label use recently.
There’s a new case, Riley v. Cordis Corp., ___ F. Supp.2d ___, 2009 WL 1606650 (D. Minn. June 5, 2009), that anybody interested in preemption and off-label use should read. While it’s a PMA device case, and thus a lot of the discussion involves Riegel v. Medtronic, Inc., 128 S. Ct. 999 (2008), enough of the discussion in Riley addresses implied preemption under Buckman Co. v. Plaintiffs’ Legal Committee, 531 U.S. 341 (2001), that the case can also be of use in defending off-label use allegations involving prescription drugs – like the Zileuton/CML situation just described.
Riley involved a drug-eluting stent – a nifty recent development that combines a stent (a tube of sorts that physically holds an artery or vein open) with the delivery of a drug that independently helps prevent occlusion of the affected vessel. Like any prescription drug or medical device, this product has risks. As the Riley court stated:
After implantation, the  stent slowly releases [a drug] to prevent the artery from being narrowed through restenosis (the build-up of new tissue). But the release of [the drug] also slows the normal healing process; specifically, it slows the beneficial growth of a thin, slippery layer of endothelial cells over the stent and arterial wall. Before this healing process is complete, there is an increased risk of blood-clot formation.
2009 WL 1606650, at *1. Plaintiff alleged that she got a clot, and that it caused a heart attack. Id.
Drug-eluting stents are considered by the FDA to be medical devices requiring PMA – and thus they're protected by preemption under Riegel. Given that a drug is also involved, presumably the FDA could have regulated such a combination product as a drug, but the Agency has chosen to apply its device rather than its drug regulatory scheme to this particular type of product.
This particular stent was approved for only a limited use – certain types of problems with arteries of a certain size. One thing plaintiff claimed was that the FDA had not approved the stent for was something called “direct stenting,” which means implanting the stent without first using a balloon catheter (that is, the type of device at issue in Riegel) to expand the diameter of the vessel prior to implantation. 2009 WL 1606650, at *2.
Five points to Gryffindor if you’ve guessed by now that the plaintiff’s surgery involved direct stenting.
So the plaintiff in Riley – looking for a way around Riegel preemption, played the off-label use card for all it was worth.
Fortunately, in Riley the off-label use card turned out to be the deuce of clubs.
Predictably, the plaintiff in Riley argued that preemption did not apply to “parallel” claims. 2009 WL 1606650, at *3. That’s where things start to get interesting. Employing analysis similar, but not identical, to that used in In re Medtronic, Inc. Sprint Fidelis Leads Products Liability Litigation, 592 F. Supp.2d 1147 (D. Minn. 2009), Riley applied both express and implied preemption principles to narrow this exception – although, oddly, the court never cited Sprint Fidelis (for our discussion of this aspect of Sprint Fidelis, see here).
To escape express preemption under Riegel, a “parallel” claim “must be premised on the breach of a state-law duty that is the same as a duty imposed under the FDCA (or one of its implementing regulations).” Riley, 2009 WL 1606650, at *3. Thus “the conduct that is alleged to give the plaintiff a right to recover under state law must be conduct that is forbidden by the FDCA.” Id. If the manufacturer’s conduct wasn’t prohibited, then to impose tort liability would impose the dreaded “additional or different” state requirement that the FDCA expressly preempts.
But that’s not enough. In Buckman Co. v. Plaintiffs’ Legal Committee, 531 U.S. 341 (2001), the Supreme Court also recognized that “enforcing the FDCA is the exclusive province of the federal government.” Riley, 2009 WL 1606650, at *3 (quoting Buckman).
Thus, a private litigant cannot sue a defendant for violating the FDCA. Similarly, a private litigant cannot bring a state-law claim against a defendant when the state-law claim is in substance (even if not in form) a claim for violating the FDCA-that is, when the state claim would not exist if the FDCA did not exist.Id. (not quoting Buckman).
Riley found only a “narrow gap” between the preemption Scylla of Riegel and the Charybdis of Buckman:
The plaintiff must be suing for conduct that violates the FDCA (or else his claim is expressly preempted [under Riegel]), but the plaintiff must not be suing because the conduct violates the FDCA (such a claim would be impliedly preempted under Buckman). For a state-law claim to survive, then, the claim must be premised on conduct that both (1) violates the FDCA and (2) would give rise to a recovery under state law even in the absence of the FDCA.
Id. at*4 (emphasis added).
Plaintiff Riley wasn’t able to thread that needle – and thus attacked off-label use instead. The first attack was pretty weak. He claimed that Riegel express preemption “only applies when the medical device is used in a manner that was reviewed and approved by the FDA.” Riley, 2009 WL 1606650, at *4. There was a slight problem with that argument. All right, a not-so-slight problem. Riegel itself involved an off-label use. Id. So plaintiff argued, in effect, ignore Riegel because the off-label use point was “not actually argue[d].” Id.
That didn’t work very well. The court in Riley responded with a long block quote from Buckman (including, we have to note, a cite to a law review article one of us wrote), and concluded that preemption in Buckman existed “in part because the Court did not want to deter off-label use of medical devices.” Riley, 2009 WL 1606650, at *5. In light of Buckman, plaintiff’s anti-off-label use argument was illogical:
Under [plaintiff’s] theory, though, a manufacturer of a medical device could scrupulously adhere to the FDA’s every command and meet every requirement. . .and nevertheless be sued under the tort law of any of the fifty states because a health-care provider, without the manufacturer’s consent or even knowledge, decided to put the device to an off-label use. Given what the Supreme Court said in Buckman about the off-label use of medical devices, it seems highly unlikely that the Court intended to create such a loophole in Riegel.
Id. We like this analysis especially because, as we alluded to earlier, Buckman was an implied preemption case, so the same rationale would also apply to prescription drugs – such as Zileuton being used off-label to treat (or prevent) CML.
The court also blew away the plaintiff’s next argument – that because the off-label use had not specifically been approved by the FDA, there were no applicable FDA “requirements” that could be a basis for preemption. Another slight (well, not-so-slight) problem: this argument was “inconsistent with the text of the statute." Riley, 2009 WL 1606650, at *5. That’s because, under the express terms of the FDCA (this textual argument wouldn’t apply to drugs) “the question is not whether there are federal requirements applicable to a particular use of a device; the question is whether there are federal requirements applicable ‘to the device.’” Id. (emphasis original).
Plaintiff also argued that, because there was an off-label use, the FDA couldn’t have known of the particular use, and therefore there was no preemption. The court countered that “nothing in Riegel even hints that whether a state-law claim is expressly preempted. . .turns on the nature or extent of the information made available to the FDA at the time it approved a device.” Riley, 2009 WL 1606650, at *6. Further, if the lack of information was allegedly deliberate, the plaintiff was alleging fraud on the FDA and was preempted under Buckman. Id.
Having thus crashed the last off-label use argument into the Riegel/Buckman brick wall, plaintiff in Riley seized upon the already mentioned combination nature of the product – claiming that because the stent was “coated with a drug,” Riegel was ousted by the implied preemption analysis of Wyeth v. Levine. Riley, 2009 WL 1606650, at *6. This argument, while creative, was nonsensical:
It makes no sense – indeed, it would probably be impossible – to pick apart the components of a medical device and apply different preemption analyses to different components. Because the FDA regulated the  stent as a medical device, the Court applies the express preemption analysis of §360k(a) to the entire device.
Plaintiff’s last general preemption argument – lashing out at the FDA as “a poorly run agency that is not able to ensure the safety of the medical devices it reviews” – also failed. The court declined to get into “policy,” holding that “[i]f [plaintiff] believes that [preemption] represents a poor policy choice, his argument is better addressed to Congress.” Riley, 2009 WL 1606650, at *7. Hear! Hear! If only the Supreme Court had been so wise in Levine. Tort suits brought under state law are not a proper vehicle for challenging the merits of federal policy.
Having rejected all of plaintiff’s general preemption arguments, the court went on to throw out each of the plaintiff’s specific claims. Some of these holdings raise additional off-label use points. Plaintiff claimed, for example, based upon an FDA regulation (21 C.F.R. §801.4), that any manufacturer that so much as knew about an off-label use is obligated to change its label to warn about it. Riley, 2009 WL 1606650, at *7-8. In this respect, the FDA’s restrictions on off-label promotion backfired on plaintiff – since only very limited promotion was permitted:
[F]ederal law at the time relevant to this action explicitly permitted manufacturers to promote off-label uses of their devices by disseminating certain types of information about those off-label uses-and to do so without having to provide instructions or warnings about those off-label uses. It seems highly unlikely that, at the same time, §801.4 was forcing manufacturers who knew of off-label uses of their devices but did not promote those off-label uses to provide instructions or warnings. It would make no sense to impose on manufacturers who were not promoting off-label uses of their devices a duty to instruct or warn, but to impose no such duty on manufacturers who were promoting off-label uses.
Id. at *8 (emphasis original). Thus, mere knowledge of an off-label use cannot, consistently with the FDA’s regulatory scheme, give rise to a duty to warn.
The only warning claim that might survive (but was inadequately pleaded) was that the defendant engaged in off-label promotion that was not only prohibited by the FDCA, but which was also inaccurate, and thus might support a traditional state-law inadequate warning claim. Riley, 2009 WL 1606650, at *10. Here’s where another of our favorite Supreme Court cases Bell Atlantic Co. v. Twombly, 550 U.S. 544 (2007), comes into play. To state a non-preempted claim of this sort, the plaintiff must be able to allege – with at least some “plausible” factual support:
- “that [defendant] affirmatively promoted the use of the  stent” for the off-label use plaintiff’s surgeon performed, which should “identify a particular oral or written communication” demonstrating such promotion;
- “that [defendant] was aware or should have been aware of the dangers inherent in those off-label uses and yet failed to warn of those dangers or give adequate instructions about those off-label uses”;
- “that implanting the  stent in a particular off-label way caused [plaintiff’s] heart attack”; and
- “that, had [defendant] adequately warned or instructed about this particular off-label use, [plaintiff’s] physician would not have implanted a  stent (or would have implanted it in a different way).”
A similar fate awaited for plaintiff’s fraud claims – except that, instead of the Twombly standard, the plaintiff was required to plead “with specificity” under Fed. R. Civ. P. 9(b). Riley, 2009 WL 1606650, at *12. Because the court concluded it was impossible for plaintiff to meet this stricter pleading standard, the fraud claims were dismissed with prejudice. Id. at *13.
Plaintiff alleged a raft of other claims in Riley. Only two merit any real attention. The court’s express warranty analysis was fairly standard among post-Riegel decisions: allegations based upon information approved by the FDA were preempted, but allegations based upon information (assuming there was any) not approved by the FDA were not preempted. Riley, 2009 WL 1606650, at *13-14. Typically, the express warranty allegations were “uninformative.” Id. at *13. Thus plaintiff was required to replead in accordance with Twombly. Id. at *15. Good luck. Our experience is that plaintiffs, forced to plead all the elements of express warranty usually can't do it – if they could have, they would have the first time around.
Finally, as to manufacturing defects, the plaintiff in Riley could not rely upon FDA inspections that took place after the stent in question was implanted. 2009 WL 1606650, at *15. This time plaintiff had pleaded too much, rather than too little. Pleading everything that had ever gone wrong during the device’s entire regulatory history was a no-no. Rather, the only allegations that might possibly survive were those that could be causal of the plaintiff’s injury:
[Plaintiff] does not clearly allege that any of [defendant’s] violations actually resulted in the manufacture of a single defective stent, much less that the particular stent that was implanted in [plaintiff] was defective. Obviously, [plaintiff] cannot sue [defendant] for negligently manufacturing stents that were implanted in other people.Id. at *15. Having seen numerous complaints that flagrantly violate this common-sense stricture, we’re quite pleased to see one get thrown out.
In Buckman the Supreme Court ruled that off-label use was both legal and medically proper. It can also be life saving, as the recently-discovered off-label treatment for leukemia could well turn out to illustrate. Thus, we’re pleased to find decisions like Riley that refuse to punish a manufacturer with additional liability simply because a physician chooses to use its drug or device off-label in treating a patient.