Photo of Stephen McConnell

Yesterday we commenced the dreary process of taking down the Christmas decorations. The German nutcrackers slid back into their cartons. Stockings marched from the mantle into plastic storage containers. We picked the ornaments off the tree, one by one, paying special attention to the souvenirs from this year’s trips (Newport mansions, the FBI tour, Comic-Con). After clearing out so many of these jolly gee-gaws, we spotted a present under the tree that had thus far gone unnoticed. This seems to happen every couple of years and never fails to make us feel simultaneously silly and grateful. Our carelessness somehow managed to extend the holiday.

The Utts v. Bristol-Myers Squibb Co., et al., 2016 WL 7429449 (S.D.N.Y. Dec. 23, 2016), is like a late-discovered Christmas present. Indeed, the court’s application of preemption in favor of a branded pharmaceutical is so strong, so compelling, and so useful, that it might have deserved spot on our 2016 top ten list.

The plaintiffs claimed that Eliquis, a drug used to reduce risks of stroke and embolism from a-fib, caused severe internal bleeding injuries. The complaint set forth the usual panoply of product liability actions under California law (the parties agreed that California law would apply) plus a claim under New York’s consumer protection law. The court was skeptical about that last bit, dismissed it, and afforded plaintiffs with leave to amend. The court was more than skeptical about the product liability claims, as we shall see momentarily. The court also did something very shrewd. It asked the parties to identify a motion to dismiss that might dispose of the case, and then held off initiation of discovery until resolution of that motion. Here, at last, is a court that has a proper sense of the burdens of discovery and the legal frailty of many plaintiff claims.

The motion to dismiss challenged most of the claims on preemption grounds. That is the issue of most interest to us. But the defendants also prevailed on some other grounds, which we will summarize briefly in our rush to celebrate the preemption ruling:

  • The court dismissed the manufacturing defect claim because, like most such claims, it utterly failed to allege any deviation from specs.
  • The court dismissed the strict liability design defect claim because California permits no such claim against prescription drugs.
  • The court concluded that the claims for breaches of express and implied warranties were deficient because the plaintiffs never identified precisely what those warranties were. The claims were dismissed (yay) but the plaintiffs were given leave to amend (boo).
  • The court dismissed the fraud claims to the extent the FDA was the alleged victim because Buckman preemption thwarted those claims. To the extent that the alleged fraud supposedly hoodwinked anyone else, the court dismissed for lack of particularity under Fed. R. Civ. P. 9(b), but … sigh … the plaintiffs were permitted to try again.

There is one other point worth mentioning before plunging into the preemption analysis. It seems to us that the Eliquis label had as strong a warning as one could want regarding the alleged injuries. It discussed the risk of internal bleeding, even “serious, potentially fatal, bleeding.” It disclosed that there was no established way to reverse the anticoagulant effect and that no specific antidote was available. Naturally, the plaintiffs argued that there were other things that should have been inserted into the label, but the examples supplied seem like pure fly-specking. To our plainly biased eyes, the label was plainly adequate. But the court declined to rule on the adequacy of the label. Maybe that is not such a big deal from a larger perspective, because not all labels will be as strong as the Eliquis label, but the court’s preemption analysis will be relevant for many, if not most, branded pharmaceutical cases going forward. Still, it is irksome how plaintiffs get away with so much gamesmanship when it comes to retrospective critiques of product labels. Wouldn’t it be wonderful if Congress passed a law requiring that plaintiffs first petition the FDA to change a label before they could institute a product liability action challenging the adequacy of a label? What would that do to litigation dockets around the country? But here we are greedily asking for another gift after pawing one we just received.

So let’s talk about the court’s marvelous explication of preemption principles.

The Utts court lays out a description of the Wyeth v. Levine, Pliva v. Mensing, and Mutual Pharmaceutical v. Bartlett suite that is as clear and useful as we have seen anywhere. In particular, the Utts court makes clear that there are limits to the scope of Levine‘s anti-preemption reach, and that the logic of Mensing and Bartlett is not limited to generic drugs. Here are the key principles the Utts court distilled from those SCOTUS cases:

  • Levine. The FDA’s premarket approval of a drug includes the exact text of the label. After approval, a manufacturer may make certain changes to the label via the changes being effected (CBE) process if the manufacturer finds newly acquired information implicating safety issues. Failure to warn claims based on newly acquired information are not preempted. But Levine “did not address whether a state law failure to warn claim addressed to the NDA process was preempted.” The Utts court concludes that such claims are preempted.
  • Mensing. Generic manufacturers are limited by the sameness requirement when it comes to product labels. That is, their labels must be the same as branded labels. Generic manufacturers cannot avail themselves of the CBE process. The plaintiffs suggested that the generic manufacturers could have proposed changes to the FDA, which might then work with the brand manufacturer to change the label. But SCOTUS refused to allow plaintiffs to play the what-if game. Instead, SCOTUS held that “when a party cannot satisfy its state duties without the Federal Government’s special permissions and assistance, which is dependent on the exercise of judgment by a federal agency, that party cannot independently satisfy those state duties for preemption purposes.” (Mensing, 564 U.S. 604, 623-24 (2011)). That independence principle is not limited to generic manufacturers.  We have been saying as much in this space for quite some time.  See here, for example.
  • Bartlett. The sameness requirement in Mensing applies to product design just as much as to product labels. Generic manufacturers lack the power to change product design without FDA action. (For that matter, branded manufactures cannot make major design changes without FDA action.) What, then, would the plaintiff in Bartlett, have the manufacture do? The plaintiff’s argument was that the defendant could have complied with both federal and state law simply by ceasing to sell the product. But in Bartlett, SCOTUS explicitly rejected the stop-selling rationale.

Reading these three cases together, the Utts court arrives at important conclusions. The lack of authority of generic manufacturers to change labels or designs – which was discussed in Mensing and Bartlett – “exists for brand name manufacturers at the time the NDA concludes.” Thus, if a claim “addresses newly acquired information and addresses a design or labeling change that a manufacturer may unilaterally make without FDA approval, then there may be no preemption of the state law claim.” But post-approval design changes that would require FDA approval are preempted (citing Yates v. Ortho-McNeil-Janssen Pharm. Inc., 808 F.3d 281, 298 (6th Cir. 2015) – a case that was, by the way, our favorite case in 2015, and was praised by us in several other posts, including here), and so are label changes if they are based on information known to FDA prior to approval of the label (citing In re Celexa & Lexapro Mktg. & Sales Practices Litig., 779 F.3d 34, 43 (1st Cir. 2015) – which was only our third favorite case in 2015).

How did all of this apply to the Utts case? Very nicely for the defense. The complaint focused on the results of a study that had concluded prior to NDA . This was not “newly acquired information” of a “different type or greater severity or frequency than previously included in submissions to FDA.” To be sure, the plaintiffs tried to point to newly acquired information, in the form of adverse event reports of severe hemorrhaging. Of course, there will always be adverse event reports post-approval. But the issue is whether such reports represent a different type or greater severity or frequency of injury as compared to what was revealed to FDA at the time of approval. The Utts court held that these adverse event reports constituted a “threadbare” allegation that failed to constitute newly-acquired preemption. Accordingly, the failure to warn claim was preempted. The plaintiffs have leave to amend, and we will be watching. This issue, in a manner similar to the overall adequacy of warning issue, seems ripe for plaintiff gamesmanship, and we fervently hope that courts will shut down the game.

The court also held that the negligent design claim was preempted. The plaintiffs argued that the defendants had a preapproval duty to submit a differently designed drug for FDA approval. But then the court would have to engage in “counterfactual’ speculation as to whether the FDA would have approved such a design and whether such a new design would have avoided the alleged injury. The Utts court declined to speculate. Moreover, the plaintiffs’ argument comes very close to replicating the stop-selling argument foreclosed in the Bartlett case. Thus, the Utts court concluded that the design defect was preempted. Even better, the Utts court dismissed the design defect claims with prejudice, with no leave to amend. Nothing the plaintiffs might say could alter the fact that the defendants simply had no ability to alter the composition of the drug without FDA approval.

Perhaps we are still under the influence of the season’s shiny lights and wassailing, but Utts makes us giddy. Its reasoning is unassailable. Its effect is to undermine many failure to warn claims. It might even doom most design defect claims (for devices as well as drugs), at least if courts apply the reasoning rigorously and force plaintiffs to stop playing vague shell games and actually specify their suggested design changes. Unwrapping Utts brought us as much joy as any sled, toy, or bacon-of-the-month membership.