Since the unfortunate decision in Wyeth v. Levine, 555 U.S. 555 (2009), successful applications of preemption against warning claims asserted in litigation involving innovator (branded) drugs have been depressingly rare. We were aware of only two decisions since Levine that found sufficient regulatory evidence to conclude that the Levine “clear evidence” standard – apparently unique to FDA-approved drugs – has been satisfied. Those cases were: Robinson v. McNeil Consumer Healthcare, 615 F.3d 861 (7th Cir. 2010) (an OTC case where the FDA had rejected a directly on point citizen’s petition), and Dobbs v. Wyeth Pharmaceuticals, 797 F. Supp.2d 1264 (W.D. Okla. 2011) (where the FDA to this day has held on several occasions that insufficient scientific evidence exists to justify a warning). Now there are three.
Recently, the court in In re Fosamax Products Liability Litigation, ___ F. Supp.2d ___, 2013 WL 3270387 (D.N.J. June 27, 2013), found the “clear evidence” standard satisfied with respect to certain post-2009 warnings – given the relevant FDA regulatory history. The regulatory facts in Fosamax are both a testament to the intensity of the FDA’s review of that product and a caution that preemption standard set (we would say erroneously) by the Court in Levine remains dauntingly high. Post-Levine preemption motions remain a matter of utmost care and discretion, and require (as is the case in Fosamax) a powerful regulatory case.
The preemption ruling in Fosamax was based on the following FDA-related facts relating to the alleged drug risk of atypical femur fracture (“AFF”):
-
- The use was on-label – the drug was approved for marketing for the treatment and prevention of osteoporosis in older women, and that’s what the plaintiff had it prescribed for. Fosamax, 2013 WL 3270387, at *1.
- The FDA had more information than the defendant – before the plaintiff’s injury, the Agency had contacted all manufacturers of this class of drugs, sought, and received, all of their information about reports of AFF. Id. at *1-2.
- The defendant fully complied with the FDA’s information request. Id. at *2.
- The FDA reviewed all this data, from the defendant and other manufacturers of the same class of drugs, and concluded that it “did not show an increase in” AFF among users of this type of drug. Id.
-
- The defendant submitted a proposed labeling revision to the FDA concerning AFF (denominated “low-energy femoral shaft fracture”), discussing the condition in considerable detail in two different sections of the labeling. Id.
- On April 15, 2009, the Agency informally rejected the bulk of the proposed revision – all but the mention of “reports” of AFF in the “adverse reactions” section. Id.
- Two days after the FDA’s rejection, the plaintiff broke her leg. Id.
- The next month, the FDA’s formal rejection issued. Id. at *3.
- A month after that rejection, the defendant submitted a changes being effected “(CBE”) labeling supplement that complied with the FDA’s disposition of the earlier proposal. Id.
- Almost a year (March, 2010) after the defendant’s initial proposal, the FDA issued a “Drug Safety Communication” about the class of drugs and AFF, which stated that no “clear connection” had been demonstrated. Id.
- After another six months (September, 2010) a medical “task force” reported an “association” between AFF and the class of drug, but not anything “proven to be causal.” Id.
- The FDA responded to the task force’s report with a statement also indicating that causation “is not clear,” and that the Agency was “considering label revisions.” Id.
- After another month (now, October, 2010), the FDA announced revised labeling for AFF for this class of drug. Causation was described as “not clear.” Id. at *4.
- Another FDA “not clear” causation statement followed on January, 2011. Id.
- The defendant’s current FDA-approved labeling states that “causality has not been established.” Id.
On the basis of this regulatory history, the court found the requisite “clear evidence” that the FDA would have rejected the plaintiff’s labeling position, which was that state law required an admission of causation prior to the plaintiff’s April, 2009 injury. In light of that evidence, the court held, under Levine, that the plaintiff’s warning action was preempted. [T]he evidence presented . . . establishes that the FDA would not have approved a label change to the Precautions section of the Fosamax label prior to [plaintiff’s] injury. . . . [C]lear evidence exists that the FDA would not have approved a label change . . . prior to [plaintiff’s] fracture because Defendant submitted a label change and the FDA rejected it, and the FDA never required Defendant to submit new language or change the label, which demonstrates that the FDA did not think that the label should have been changed at that time.
Fosamax, 2013 WL 3270387, at *7 (expert witness citations omitted).
The plaintiff quibbled with the FDA’s action, claiming that the Agency’s rejection was merely due to wording (“stress fracture”) and that a supposedly proper revision, calling the risk “AFF,” could have passed. Id. at *5-6. Trouble was, as the court pointed out, plaintiff had no evidence to back up this argument. Id. at *8. Indeed, the evidence was that “AFF,” as a medical phrase, had not even been invented when the plaintiff’s injury occurred. Id. Rather, the uncontested evidence was that the FDA rejected the proposed language because it viewed it as not supported by the relevant data. Id.
Nor would some other CBE notification have worked. A CBE, like an approved supplement seeking to do the same thing, “must be based on ‘reasonable evidence of’ an association between a hazard and the drug at issue.” Fosamax, 2013 WL 3270387, at *8 (quoting relevant regulation). Because it rejected the warning change using one avenue, the FDA would also have rejected the CBE, since the underlying requirements for scientific support were identical. “[S]ince the FDA rejected Defendant’s PAS, it would not have approved a CBE seeking to add the same language to the label that it just rejected in the PAS.” Id. That was the defendant’s “clear evidence,” and plaintiff came forward with nothing in response to refute it.
Finally, the plaintiff tried to argue that there had been fraud on the FDA. Again, there was a complete lack of evidence. Id. at *9. (Of course, we’d say that argument is also independently preempted under Buckman). Indeed, “as late as March 2010, the FDA did not see a clear connection” that would have supported a warning change. Id. (citation and quotation marks omitted).
In the end, the strength of the regulatory history won out. The defendant had, in fact, submitted an on-point warning to the FDA and the Agency had rejected it. That was enough:
[P]reemption is warranted in this case. Defendant submitted a PAS in 2008 seeking to change the Precautions section of the Fosamax label to include information on [the relevant risk], but the FDA rejected the PAS in May 2009, one month after [plaintiff’s injury]. This constitutes clear evidence that the FDA would not have approved a stronger warning prior to [that injury].
Fosamax, 2013 WL 3270387, at *9.
We agree. That certainly seems like “clear evidence” on the FDA’s position to us. However, we caution that the regulatory history the defendant could call upon in Fosamax is not suggestive of broadly available preemption. This particular defendant in fact made a submission to the FDA under the identical substantive scientific standard to the FDA, and the date of the FDA’s disapproval was so close in time to the plaintiff’s injury (a mere two days) that any suggestion that the FDA could have changed its mind was implausible. So while we’re pleased with this preemption win, the basic problem – the unique, peculiar, and undefined Levine “clear evidence” standard never imposed in any implied preemption case before or since – remains.