Photo of John Sullivan
As we flagged for you on Friday, the New Jersey Supreme Court recently issued an opinion addressing preemption and the statute of limitations in a medical device claim.  Cornett v. Johnson & Johnson, 2012 N.J. LEXIS 831 (Aug. 9, 2012).  To say the least, it had interesting rulings and raised a number of issues to discuss. 
The case involved a drug-eluting stent that was implanted in the plaintiff for an off-label indication, and the plaintiff later suffered a thrombosis and died.  Id. at *9-11.  The plaintiff’s estate made a number of state law claims alleging improper warnings, manufacturing and design defects, improper off-label promotion and the like.  Id.  The statute of limitations and preemption issues eventually made their way to the New Jersey Supreme Court. 
Statute of Limitations
On the statute of limitations defense, the Court held that the plaintiff’s claims were barred by Kentucky’s one-year statute of limitations.  The opinion has a detailed choice of law analysis (between Kentucky and New Jersey law) that we won’t address here.  What was more interesting to us was the Court’s ruling that a plaintiff exercising due diligence should have known by December 2006 that the stent may have caused Mr. Cornett’s thrombosis.  December 2006 is when an Advisory Committee assembled by the FDA made findings related to plaintiffs’ claims, including that thrombosis from drug-eluting stents (like the defendants’) was a problem, the problem was greater when the stent was used off-label (as with the plaintiff), and drug therapy after the device is implanted should last for an extended period of time.  Id. at *28. 
With this ruling, the New Jersey Supreme Court has placed significant importance on Advisory Committee findings when considering when the statute of limitations starts to run in a product liability case.  Often the devices and drugs involved in product liability litigation have been the subject of FDA Advisory Committee meetings, and often those meetings address issues related to plaintiffs’ claims.  Now, in New Jersey, their findings may be a significant tool for weeding out time-barred claims. 
Preemption
The Cornett complaint was a consolidated complaint with over 40 other plaintiffs whose claims were not time barred.  So the Court was able to go beyond the statute of limitations to address preemption under the Medical Device Act (“MDA”) amendments to the FDCA. 
First, some general propositions about medical device preemption: Section 360k(a) of the MDA states that it preempts state law claims that address the safety and efficacy of a device and impose requirements that are “different from or in addition to” FDA regulation.  In Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996), the US Supreme Court held that §360k(a) preemption does not apply if the FDA cleared the device under §510k.  (You may recall that we just recommended some light summer reading that reexamines whether Lohr’s reasoning should still apply.)  In Riegel v. Medtronic, Inc., 552 U.S. 312 (2008), the Court held that §360k(a) preemption does apply if the device was approved under the FDA’s more rigorous Premarket Approval (PMA) process.  Finally, in Buckman v. Plaintiffs’ Legal Comm., 531 U.S. 341 (2001), the Court held that – in situations in which §360k(a) express preemption doesn’t apply, such as with a device cleared under §510k – FDA regulations nonetheless implicitly preempt state law claims, in particular fraud-on-the-FDA claims, that improperly impact the balance struck by the FDA’s regulatory scheme. 
As for the Cornett case, it does not involve a device cleared under §510k.  The drug-eluting stent was approved under the PMA process.  This implicates §360k(a)’s express preemption under RiegelLohr is irrelevant, as likely is Buckman and its implied preemption given that express preemption is already in play.
In its preemption analysis, the Cornett Court considered two claims, failure to warn and breach of an express warranty.  While the Court found significant portions of each of these claims preempted, it allowed limited parts to go forward. 
Plaintiffs’ Failure to Warn Claim
On the failure to warn claim, the Court held as preempted plaintiffs’ claim that the label lacked necessary information on the fact that comparative studies were not conducted on the stent and that drug therapy should be used for an extended period of time after the stent is implanted.  This was an easy one.  The claim goes to the heart of the FDA’s approval process and the Court saw through it:
This claim is nothing more than a challenge to the adequacy of the information required by the FDA during the PMA process and the label approved by the agency.  This failure to warn claim falls within the PLA rebuttable presumption [we’ll get back to this] and the Riegel express preemption rule.  We affirm its dismissal.
Id. at *45. 
The Court also properly dismissed plaintiffs’ failure to warn claim based on an allegation that “defendants obtained FDA approval for the device only after submitting fraudulent representations to or withholding material information from the FDA,” citing Buckman.  We agree that this claim is properly preempted.  But we also note that this isn’t a 510k case in which §360k(a) express preemption doesn’t apply.  It’s a PMA case, and §360k(a) express preemption does apply under Riegel.  That is likely all that’s necessary, not implied preemption under Buckman
Our real concern, though, is the Court’s decision not to preempt plaintiffs’ claim that the defendants (1) withheld from doctors and the public the fact that the label’s instructions for post-implementation drug therapy were not part of the PMA process and (2) made misrepresentations to doctors and the public that the stent was not pro-thrombotic.  Id. at *47-48.  The Court reasoned that this claim was not preempted because it “overcomes the PLA’s rebuttable presumption of adequacy.”  Id. at *48.  In other words, the complaint’s allegations, for pleading purposes, were enough to overcome the NJ Product Liability Act’s presumption that an FDA-approved label was adequate. 
But that doesn’t seem to be the issue.  Before the PLA’s presumption should be considered, the determination first should be made as to whether PLA can even be applied – that is, whether it is preempted.  And a claim of failure to warn doctors and the public is preempted under Riegel.  While the PLA and its presumption might eventually be relevant in a prescription drug case where there generally is no express preemption, this is a medical device case – one that involves the PMA process and Riegel’s express preemption.  What is and is not involved in the PMA process, what is said about that to doctors and the public, and what is said about post-implementation drug therapy or the dangers and safety of the product (here, it allegedly being pro-thrombotic) are all in the wheelhouse of the FDA’s regulatory scheme. 
For instance, the FDA-approved label, according to the Court, already addressed post-implementation drug therapy.  Id. at *12.  Requiring further disclosure to doctors and the public about what the FDA considered or didn’t in granting that approval is something for the FDA to determine.  The fact that is has not done so does not act to allow the plaintiffs’ claim.  It only underscores that plaintiffs’ claim is preempted as imposing requirements different from or in addition to what the FDA required.
The Court next addressed plaintiffs’ failure to warn “concerning off-label uses . . . based on a failure to warn the medical community and the general public of adverse information gathered about the device and that instructions for post-implantation therapy were not part of the PMA process.”  Id. at *48.  The Court held that, “[t]o the extent [this] claim is founded on promotion by defendants of off-label uses of the device beyond the [FDA’s] safe harbor [for off-label promotion], the claim is not preempted.”  In other words, if the defendant’s off-label statements do not fall under a safe-harbor created by the FDA for off-label statements, then it is not preempted.
We also question this decision.  The FDA has had a lot to say about off-label promotion.  The Cornett decision cites some of it.  FDA regulations require manufacturers to warn about off-label uses in certain circumstances.  Id. at *35.  There is a safe-harbor for off-label promotion that was previously codified in the United States Code and has since been adopted by the FDA in a guidance document.  It  says that the Secretary of HHS won’t consider a manufacturer’s dissemination of information on unapproved uses to be off-label promotion if certain conditions are met.  Id. at *33-35.  And, as described above, the FDA convened an Advisory Committee to address off-label use and post-implementation drug therapy for the very stent at issue in this litigation.  There is no doubt that the FDA regulates off-label promotion and was actively doing so for this device.  Even the Riegel case, which established express preemption, involved off-label use.  552 U.S. at 320.
And we don’t see how the plaintiffs alleging off-label promotion “beyond the safe harbor” somehow exempts plaintiffs’ failure to warn claim from §360k(a)’s express preemption.  Falling within the safe harbor simply means that the FDA itself will not consider the statements to be evidence of off-label promotion.  It doesn’t mean that statements outside the safe harbor are off-label promotion.  It just means that the FDA considers them evidence of off-label promotion.  The ultimate determination is for the FDA, not private litigants.  And the United States, not private litigants, enforce the FDCA and FDA regulations.  (Setting that aside, it’s not clear how, in Cornett, falling outside the safe harbor matches up with a failure to warn claim.) 
Now, plaintiffs may argue that by invoking the safe harbor they are making a parallel violation claim, which is arguably permissible under Riegel.  But that doesn’t seem to work either.  The FDA hasn’t determined that the alleged off-label promotion in Cornett violated any FDA requirements.  And, failure to satisfy the safe harbor doesn’t create a violation either.  So plaintiffs aren’t really seeking to enforce an FDA requirement.  They can’t be making a parallel violation claim.  In short, they are seeking to enforce a requirement “different from or in addition to” FDA requirements. 
Plaintiffs’ Breach of Express Warranty Claim
Plaintiffs alleged that the defendants breached express warranties that were created by the labeling and packaging.  The Court quickly dismissed this claim as preempted under Riegel:
Plaintiffs contend that the MDA does not preempt an express warranty claim based on the information contained in FDA approved product labels and packaging inserts.  We disagree. . . . [I]n order to succeed on the breach of express warranty claim, plaintiffs must show that the label provides inaccurate or insufficient information in spite of FDA approval following the rigorous PMA process. Success on this state law claim would inevitably impose greater requirements than those already established by the MDA.  This claim is, therefore, preempted.
Id. at *51-53. 
The Court, however, also held that an express warranty claim based on statements voluntarily made to third parties and that “deviated from the labeling and instructions” are not preempted.  Id. at *53.  This seems, if we understand it correctly, to be an attempt to again make a parallel violation claim.  But it doesn’t seem to be a fit, at least not based on the information in the opinion.  A manufacturer can make statements that are not in the labeling.  Now, it may be that some of those would violate FDA regulations.  But that doesn’t mean that any deviation from the labeling is a violation.  And without any particulars, it’s not clear that plaintiffs properly stated a parallel violation claim for breach of express warranty. 
                              *                   *                 *                 *                             

All this said, there is still hope for preemption of the claims that remain after the Court’s decision.  The Court more than once qualified its decision by saying that it was being indulgent to plaintiffs’ allegations on a motion to dismiss [in state court, by the way, where TwIqbal doesn’t apply].  Discovery may reveal further information establishing that plaintiffs’ remaining claims are, as we discussed above, preempted.  Id. at *49-50, 53.  We think that, come summary judgment time, Defendants will have a good chance of establishing that the remaining claims are in fact preempted too.