A short history of recent implied preemption “impossibility” decisions: (1) In Wyeth v. Levine, 555 U.S. 555 (2009), impossibility preemption did not apply to innovator prescription drugs because simultaneous compliance with FDA and state tort law labeling obligations was possible due to the “changes being effected” (“CBE”) exception allowing updated warnings without prior FDA approval. Id. at 568-69. (2) In PLIVA v. Mensing, 131 S.Ct. 2567 (2011), the same “impossibility” principle, expressed pithily as a “hold[ing] that when a party cannot satisfy its state duties without the Federal Government’s special permission and assistance, which is dependent on the exercise of judgment by a federal agency, that party cannot independently satisfy those state duties for pre-emption purposes,” was applied affirmatively to preempt warning claims concerning generic drugs because a verbatim labeling requirement precluded unilateral CBE warning changes. Id. 2580-81. (3) In Mutual Pharmaceutical Co. v. Bartlett, 133 S.Ct. 2466 (2013), the Court extended impossibility preemption to (at least) generic design defect claims, recognizing that design changes affecting safety and effectiveness also required prior FDA approval, while recognizing that the same restrictions on “major changes” applied to the designs of all “drug[s] − whether generic or brand-name.” Id. at 2471.
What the implied preemption combination of Levine, Mensing, and Bartlett tells us about preemption due to impossibility is that state tort law can’t create a duty for a defendant to do something immediately that, under supreme federal law, requires the prior approval of a federal agency. That proposition is in no way limited to generic drugs, since implied preemption isn’t dependent on the language of any particular statute. We’ve pointed this out in a number of contexts, including “stop selling” claims , “major changes” in drug labeling, innovator drugs, warnings about off-label use. That’s not counting our posts on individual cases.
We’re now two years post-Mensing and a year post-Bartlett, so we thought we’d take stock and see to what extent defendants have had any success in applying the impossibility preemption principles those decisions recognized outside of the limited context of generic drugs. It turns out there has been some progress. Favorable decisions have been rendered in three general areas:
Warning claims against entities other than the NDA holder: Only the holder of the new drug application (“NDA”) has the right to change that drug’s labeling, whether by CBE or otherwise. Preemption has been recognized in warning claims asserted against defendants that could not institute a unilateral label change because they did not own the NDA.
One situation where preemption has prevailed is against defendants that sold their NDA to someone else before the drug that the plaintiff bought was made. The claims in In re Darvocet, Darvon, & Propoxyphene Products Liability Litigation, ___ F.3d ___, 2014 WL 2959271 (6th Cir. June 27, 2014), were about as strong as any plaintiff could hope to have. In Darvocet, although the former NDA holder sold out, it also became a contract manufacturer, and thus was still producing the drug. That didn’t matter, since the former innovator could no longer change its warnings:
[Plaintiff] ingested a product manufactured by [the former NDA-holder] . . ., after [it] had divested its NDAs. Therefore, her claims against [the former NDA holder] are preempted, failing for the same reasons that the Plaintiffs’ other claims against the Generic Manufacturers do. After the divestiture, [the former NDA holder] had no more power to change the label.
Id. at *19.
Similarly in In re Fosamax Litigation, 2012 WL 181411 (D.N.J. Jan. 17, 2012), plaintiffs’ claims against alleged distributors of both generic and innovator versions of the drug were dismissed because no distributor had the power to change drug labeling unilaterally. Whether innovator or generic, “a distributor of Fosamax . . . has no power to change Fosamax labeling. That power lies with the applicant who filed the New Drug Application (NDA) seeking approval to market Fosamax. Id. at *3 (citing FDA regulations).
As a result of the scheme set forth by the FDCA, [a distributor] has no authority to initiate a labeling change of Fosamax. That authority lies with the FDA and/or with [the NDA holder]. . . . [A] contractual relationship between [a distributor] and [the NDA holder] cannot change the fact that [it] is not the NDA holder. Consequently, [the distributor] has no power to unilaterally change Fosamax labeling. Because [the distributor] could not “independently do under federal law what state law requires of it,” the state law claims brought against it are preempted.
Id. at *4 (quoting Mensing). See also In re Yasmin & Yaz Drospirenone Marketing, Sales Practices & Products Liability Litigation, MDL No. 2100, 2014 WL 1632149, at *7 (S.D. Ill. April 24, 2014) (distributor of generic drugs had no power to change labeling unilaterally); Stevens v. Community Health Care, Inc., 2011 WL 6379298, at *1 (Mass. Super. Oct. 5, 2011) (same).
Black Box Warnings: FDA regulations explicitly require that all boxed warnings have prior FDA approval. 21 C.F.R. § 201.80(e); 21 C.F.R. § 201.57(c)(1). While this could be considered a “major” change, because of this additional regulatory support we consider those cases separately. Claims that the defendant “should have included an aggressive black-box warning” were held preempted by Mensing in Fulgenzi v. PLIVA, Inc., 711 F.3d 578, 584 (6th Cir. 2013), but that is a generic drug case. In Ray v. Allergan, Inc., 2012 WL 2120018 (E.D. Va. June 1, 2012), after deciding that, as a general proposition, Mensing did not undercut Levine as to innovator drugs, the court recognized that a black box warning was an exception:
[T]he federal regulations do control what must be in a black box warning and so any theory based on a black box warning as a requirement of state law would be preempted.
Id. at *7.
In Guenther v. Novartis Pharmaceutical Corp., 2013 WL 4648449 (M.D. Fla. Aug. 29, 2013), another non-generic case, like Ray, the court found Mensing/Bartlett generally “inapposite” to innovator warnings, but on the black box issue specifically, the plaintiffs failed to “dispute [defendant’s] contention that the FDA regulations preclude a manufacturer from adding a black box warning without preapproval. Accordingly, the Plaintiffs will not be permitted to argue at trial that [defendant] should have done so. Id. at *5. In In re Avandia Marketing, Sales Practices & Products Liability Litigation, 817 F. Supp.2d 535, 553 n.97 (E.D. Pa. 2011), the court noted its “agree[ment]” with the defense position that “unlike other revisions to augment warnings, a boxed warning can be added only with prior FDA approval.”
Major Label Changes: Not all changes to innovator drug labeling can be done through CBE. Quite a few changes – some directly impacting safety, some not – are considered “major” and require prior FDA approval. Several types of changes not falling within the scope of the CBE requirement have been held preempted in innovator cases:
Most notably, in Thompson v. Allergan USA, Inc., ___ F. Supp.2d ___, 2014 WL 308794 (E.D. Mo. Jan. 28, 2014), the plaintiff’s consumer-fraud based “overfilling” allegations were held preempted because the FDA approved the strength of the drug that the defendant was allowed to sell, and the defendant could not change it without going back to the FDA. If FDA pre-approval was required, a state-law claim based on that product attribute was preempted:
[I]f Defendants were unable, under federal law, to independently lower the volume in each vial of [their drug] to be in compliance with the state duties alleged by Plaintiff, Plaintiff’s state claims would be preempted. FDA regulations provide that once a drug, whether generic or brand-name, is approved, the manufacturer is prohibited from making any major changes to the “qualitative or quantitative formulation of the drug product, including active ingredients, or in the specifications provided in the approved application.”
The Court concludes that reducing the amount of medicine in each . . . vial is a major change requiring prior FDA approval. As noted above, FDA Guidelines from 2001 state that a “decrease in the fill volume” of a drug product “involves a change to the specifications and must be submitted in a prior approval supplement” for FDA approval. . . . [T]he Court agrees that a decrease in the fill volume of a drug product . . . involves a change to the specifications under the plain meaning of the statute.
Id. at *6 (citations and quotation marks omitted). Thus, since FDA pre-approval of strength changes for all drugs was necessary, any attack on the permissible dosage of an innovator drug was “therefore preempted under the Supremacy Clause.” Id.
Having proven its entitlement to preemption under Mensing/Bartlett, the defendant was not obligated to go further and provide additional regulatory evidence under the Levine “clear evidence” test:
Moreover, since neither Mensing or Bartlett required any further “clear evidence” beyond establishing the existence of an FDA pre-approval obligation, there was no need to speculate whether the FDA would have approved such a change. The requirement of pre-approval sufficed:
The Court does not find persuasive Plaintiff’s argument that . . . [their] claims are not preempted unless Defendants show by clear and convincing evidence that the FDA would have rejected such a change . . . . Applying this standard “would render conflict pre-emption largely meaningless because . . . [w]e can often imagine that a third party or the Federal Government might do something that makes it lawful for a private party to accomplish under federal law what state law requires of it.”
Id. (citing and quoting Mensing). See Guenther, 2013 WL 4648449, at *5 (distinguishing between actual change of strength (preempted) and changes of warnings relating to permissible dosage (not preempted).
Claims that the defendant should have changed the format of its label can also be preempted to the extent that such changes may not occur unilaterally. In Hill v Novartis Pharmaceuticals Corp., 944 F. Supp.2d 943 (E.D. Cal. 2013), the court “agree[d]” to “bar [plaintiff] from arguing that the FDA-approved label . . . should have placed the warning . . . in a different section of the label, or should have utilized a different font size or bolded text.” Although the CBE procedure could be used to modify the “contents” of a label, it did not extend to formatting:
FDA regulations for prescription drug labeling extend not only to content but to formatting. While the case law is clear that manufacturers may modify the contents of a brand-name drug label without FDA approval by adding to or strengthening the warnings, [plaintiff] has provided no authority − and the Court’s research reveals no authority − to suggest manufacturers may do the same with the label’s formatting. Accordingly, [defendant’s] motion is hereby granted to the extent it seeks to preclude [plaintiff] from arguing that the [innovator drug’s] labeling should have been formatted differently.
Id. at 957 (citations omitted).
As we’ve discussed in our major changes post, comparative claims (called “superiority” claims) require FDA pre-approval. FDA regulation specify that a certain level of scientific evidence (essentially statistically significant results from two controlled studies) must exist before a manufacturer may compare the safety (or effectiveness) of its drug to any other. 21 C.F.R. §§201.57(c)(2)(iii), 201.80(c)(3)(v). Thus, Guenther also indicated (but did not actually hold) that preemption would apply to claims that the defendant should have warned about the relative “safety” of its drug compared to alleged alternative medications in the absence of the requisite studies
[H]aving “indications” that one drug is safer than another is a far cry from having “adequate and well-controlled studies” reaching that conclusion, which the Plaintiffs appear to admit is the applicable standard. At this juncture, the Court does not know enough about the comparison(s) to be made or the evidence backing them to reach a final conclusion. But it appears that the Plaintiffs’ evidence would not satisfy the standard established by [FDA regulations].
2013 WL 4648449, at *6 (regulatory citations omitted).
Stop Selling Arguments: We haven’t seen any post-Bartlett preemption cases concerning arguments that an FDA-approved drug should simply not have been sold. Plaintiffs seem to be steering away. The state of Massachusetts, however, recently tried essentially the same thing by legislation and got shot down. As we discussed here, Zogenix, Inc. v. Patrick, 2014 WL 1454696, at *2 (D. Mass. Apr. 15, 2014), held that – regardless of Levine’s limitations on preemption − a state could not bar the sale of an FDA-approved innovator drug:
[D]efendants [the Commonwealth of Massachusetts] present no evidence or persuasive argument that [Levine’s] reasoning should control in this different context. Furthermore, Levine assumed the availability of the drug at issue and analyzed whether stronger state labeling requirements obstructed the FDA’s objectives. Here, the obstruction is clearer because the drug Massachusetts wants [the plaintiff manufacturer] to adopt . . . has not been approved by the FDA. To satisfy the Commonwealth, [plaintiff] would be required to return to the FDA and seek approval of a drug different from the one the FDA has already deemed safe. For the reasons described above, Zogenix is likely to prevail on the merits.
Id. at *2.
So by our count, we have seven cases that, in one way or another, apply the Mensing/Bartlett impossibility preemption rationale to some sort of innovator-drug-related claim. The prevailing argument is always (except for stop selling) some version of the plaintiff demanding something that cannot be done by the defendant without FDA preapproval.
This is still an area where the courts need some “getting used to,” and we hope that defendants will concentrate on claims (such as drastic design changes) where the need for FDA pre-approval is indisputable. Given that implied preemption is not limited by the language of any particular statutory scheme, plaintiffs should not prevail simply by screaming that “this isn’t a generic drug case.”