Photo of Eric Alexander

As we emerge from our fourth month-long trial in a little over two years, we are sure that we have missed some recent legal developments.  As good as this Blog is as a source for what is going on in drug and device product liability litigation, you do actually need to read it to absorb information.  With all due concern for Bexis’s feelings on the matter, reading the Blog every day can end up being a lower trial priority than witness prep sessions, sleep, and daily hygiene.  Whatever developments we missed, though, we do not think that Bartlett or Mensing are no longer good law or that the Yates decision has been consigned to the scrap heap of preemption history.  We also expect that Twombly and Iqbal, entwined as TwIqbal on these virtual pages, still apply to 12(b)(6) motions in federal court.  The decision in In re Tepezza Mkt’g, Sales Pracs., & Prod. Liab. Litig., MDL No. 3079, 2023 WL 7281665 (N.D. Ill. Nov. 3, 2023), gave us a moment of pause about these assumptions.  Then we realized it was just a bad decision about a common issue in an MDL.

We have been discussing the issues of what a design defect claim with a prescription drug looks like and whether it can survive preemption for a very long time.  When the Sixth Circuit issued the Yates decision that affirmed the dismissal of design defect claims as to the prescription contraceptive Ortho Evra, we took a fairly deep dive into the subject.  It was obvious to us that the teachings of Bartlett and Mensing, along with the interplay of state design defect law and the regulatory requirements for prescription drugs, dictate that a non-preempted design defect claim for a prescription drug is a nearly mythical beast.  In general, a design defect claim for a prescription drug requires proof that the design of the drug—essentially, its chemical composition—presented an unacceptable risk of injury, that plaintiff suffered that injury due to the design of the drug, and that plaintiff would not have been injured if she had used the properly designed version of the same drug instead.  Changing the chemical composition of a drug makes it a different chemical entity, which in turn changes its properties, risks, and benefits.  FDA approval of a different chemical entity to treat the same condition as the original drug cannot be assumed.  Adding on TwIqbal, the plaintiff has to plead sufficient facts to support a design issue with the drug—not merely the existence of some risk—that could have been fixed without sacrificing efficacy or approval (among other things).  Good luck with that.

Courts have sometimes divided the design defect preemption analysis into pre- and post-approval design claims.  The latter are clearly preempted—even the plaintiff in Tepezza conceded that.  Some pre-approval claims have been found to sidestep preemption, although we find the logic lacking.  In Tepezza, the plaintiff brought design defect and negligent design claims under the law of an unspecified state (Virginia or Illinois) for alleged “hearing loss and/or tinnitus”—yes, boilerplate on the case-specific injury—from the use (over one of two inconsistent alleged time periods) of an FDA-approved prescription biologic for an eye disease.  The alleged design defect was that the medication posed a risk of “hearing loss and/or tinnitus.”  And, no, we do not think that the allegation of a risk is the same thing as alleging facts that, if true, would establish the existence of a design defect, but that is the only allegation discussed in the Tepezza decision.  When the manufacturer moved to dismiss for failure to state a claim, the court proceeded without mentioning TwIqbal, any state law, or any FDA requirement.  The fuzziness of a preemption analysis untethered to actual factual allegations or state or federal requirements tends to favor the plaintiff.  And so it did in Tepezza.

Rather than looking at plaintiff’s factual allegations to see if they were sufficient to establish a design defect claim under the applicable state law and then comparing that to the federal requirements to see if it would be impossible for the defendant to comply with both independent of action from FDA, the Tepezza MDL court engaged in a more abstract analysis.  Although it started with a recitation of the Mensing principle that requires preemption unless the defendant manufacturer “could independently do under federal law what state law requires of it,” the decision quickly detoured to the much derided reasoning of the Holley case that a manufacturer was not “required to use the allegedly defective design in the first place.”  2023 WL 7281665, *2 (citations omitted).  By “derided,” we mean Holley netted seventh worst in 2019 and Gaetano, which followed it, nabbed fifth worst in 2021.  Based principally on Holley and Gaetano, the court held “to the extent that Horizon had a duty under state law to create a safer alternative design, Horizon could have satisfied that duty without coming into conflict with any federal requirement.”  Id.

Again without considering actual allegations about how the drug’s design could have been changed to make it non-defective, the Tepezza court buttressed its conclusion with three things.  First, without much of a survey, it concluded “[n]umerous district courts presented with this precise question have reached the same conclusion,” noting only two contrary decisions, Yates and another case within the sixth circuit, Fleming.  As can be seen from our prior posts, such as here, here, here, here, here, and here, the majority position actually favors preemption.  Second, the court looked to the Seventh Circuit’s decision in Kaiser, which we bashed here and hereKaiser concerned a class II medical device.  Since Lohr in 1996, design claims about such devices have not been preempted.  The facile conclusion that, “[e]ven though Kaiser involved a federal regulatory scheme for a medical device, its reasoning applies with equal force in this context” and that this reasoning supported the non-preemption of an ill-defined pre-approval design defect claim leaves us stunned.  We cannot recall seeing Lohr applied to a drug approved under an NDA or a biologic approved under a BLA.  We have seen pre-approval design claims against a biologic held to be preempted, though. 

Third, the court distinguished the treatment of the “stop selling rationale” in Bartlett with its application in Yates to pre-approval design defect claims.  In the Tepezza court’s view, there was a gap in the holding that “an actor seeking to satisfy both federal- and state-law obligations is not required to cease acting altogether in order to avoid liability” that allowed liability to be imposed on a failure to “have acted differently” rather than a failure to “have stopped acting.”  Id. at *3 (citations omitted; emphasis in original).  As we explained here, there is no duty to a user/consumer until the product is marketed or otherwise made available for use.  Indeed, whether a product is defective is typically measured at the time it leaves the defendant’s hands, not at some point during its development.  To sell the drug at issue in this case or its hypothetical non-defective variant, the manufacturer needed to obtain a BLA from FDA.  It cannot do that “independently” (see Mensing), so it would have to “cease acting altogether” in terms of trying to manufacture and sell the drug to avoid liability under state law (see Bartlett).  The only other option would be to sell the variant without a BLA, which would subject the manufacturer to all sorts of federal liability.  So, concluding plaintiff’s “claim is that Horizon should have submitted to the FDA a formulation that did not cause permanent hearing loss and tinnitus” and that claim was not preempted skips a few necessary steps.  Id.

The court punted on the issue of whether the drug, “as a biologic, is scientifically incapable of being redesigned,” noting it as a state law issue to be tested after discovery.  Of course, this is why we said above that the proper preemption analysis would have “look[ed] at plaintiff’s factual allegations to see if they were sufficient to establish a design defect claim under the applicable state law and then compar[ed] that to the federal requirements to see if it would be impossible for the defendant to comply with both independent of action from FDA.”  You do not need to wait for costly discovery to do that analysis.