We have seen the news that FDA is preparing a proposed rule to change the regulations to permit generic drug manufacturers (ANDA holders) to update their labels based on post-approval information without waiting for the branded/NDA drug’s label to change. This seems like a fairly obvious response to Mensing’s preemption of most generic drug labeling claims and the call to a legislative solution by various judges and commentators.
It is for another day whether FDA can actually offer a regulation that both undoes Mensing and makes enough sense within the statutory scheme that can it survive inevitable challenges. In the meantime, we return to the gap in Mensing that some plaintiffs exploit.
As in many of the decisions on generic preemption and innovator liability over the last several years, Garza v. Wyeth LLC, No. 2:12-CV-00198, 2013 WL 329356 & 3293704 (S.D. Tex. June 28, 2013), involves allegations of “abnormal muscle movements” with long-term use of the generic prescription drug metoclopramide. Like many of the decisions we have griped about, the plaintiff in Garza was permitted multiple amendments of her complaint, including one after the first set of motions to dismiss had been argued. Those motions resulted in Garza v. Wyeth LLC, 2013 WL 878586 (S.D. Tex. March 7, 2013), where warning claims—the only type of claim permitted with a prescription drug under Texas law—against two generic manufacturers were preempted “as there are no allegations that their labels were inconsistent with the FDA-approved, name-brand drug labels.” Id. at *4. The claims against two other generic manufacturers were not preempted because the current complaint alleged that they had failed to update their labels to match the branded labels. Id. The court also held that, to get past a motion to dismiss, the complaint need not allege the fraud-on-the-FDA exception to the statutory rebuttable presumption of adequacy for FDA-approved labels was met. Id. at **4-5. This followed from the fact that the complaint did not allege that the labels at issue had been approved; rather, it alleged that the same two generic manufacturers that ended up on the wrong side of the preemption decision had “failed to adhere to FDA label requirements.” Setting aside whether plaintiffs would ever agree that approval of a drug’s label meant that the manufacturer had adhered to “FDA label requirements,” the March decision set the stage for the two in June.
The two remaining generic manufacturer defendants filed a second motion to dismiss and one of them filed a motion for summary judgment. Actually, according to the court, all the generic manufacturer defendants, even the two dismissed in March, filed the second motion to dismiss. We do not know why. We also do not know the status of the branded defendants—although we suspect they are gone given the jurisdiction—whether the plaintiff really took metoclopramide made by four different manufacturers, or why the court considered the second motion to dismiss of the defendant to which it granted summary judgment that same day. Oops, there goes the suspense, but bear with us. This time around, the generic defendants offered three arguments in their motion to dismiss: 1) the non-preempted failure to update claim was not recognized under Texas law, 2) the plaintiff lacked standing and the court subject matter jurisdiction for the claims being asserted, and 3) a re-hash of the 82.007 presumption argument from before.
The court again pointed to plaintiff’s allegations that the generic defendants had “failed to update their labels to match the 2003, 2004, and 2009 label changes approved by the FDA”—allegations that were necessary to get past the first motion to dismiss. 2013 WL 3293956, *2. First, defendants contended that plaintiff really claimed that “all labels were inadequate, at least through 2009, including labels containing the 2003 and 2004 FDA-approved updates to the brand-name labels.” Id. at *3. The court rejected this, holding that plaintiff had asserted “a viable product liability claim under Texas law, which is not preempted by federal drug labeling laws under Mensing,” because plaintiffs alleged that defendants had notice of a need to update their labels because of
“FDA-approved labeling changes to the brand-name drug in 2004 and 2009 specifically warning of this risk” of long-term use. Id. We do not think the court quite grasped the argument here. Plaintiff clearly alleged the inadequacy of the branded labels the generic labels had to match, so there was no non-preempted (non-conflicting) failure-to-warn claim being asserted. Ignoring, for now, the statutory presumption of adequacy, a legitimate failure to warn claim under Texas law could have either asserted that the generic labels needed to match the branded labels or that the generic labels needed to go beyond the branded labels in some way. The former claim would not have been not preempted and the latter would have been. But the former claim was not being asserted made and the latter was. So, the plaintiff was really asserting a preempted claim. The court did not see it that way, though, maybe because the argument was labeled as a state law challenge.
Second, the defendants contended that plaintiff was really trying to bring an action under the FDCA, which has no private right of action, rendering plaintiff without standing and the court without jurisdiction. The court brushed this aside with the conclusion that “Nowhere does the Fourth Amended Complaint assert a violation of the FDCA.” Id. We found this surprising because the denial of part of the first motion to dismiss was based on plaintiff’s allegation that defendants “failed to adhere to FDA label requirements” and put out generic drugs with labels that were “inconsistent with the FDA-approved, name-brand drug labels.” 2013 WL 878586, **4-5. That sounds like asserting a violation of the FDCA to us. So does the basis for the court’s denial of the third argument (in the second motion to dismiss), which can be found six sentences later: “Plaintiff[’s] theory of liability is that Generic Defendants failed to provide FDA-approved warnings and information.” 2013 WL 3293956, *4. Such an allegation may be a basis for concluding that the statutory presumption is not triggered by the pleadings, but it certainly is an allegation that defendants should be liable because they violated the FDCA, specifically 21 U.S.C. 352.
This is all the more confusing because the court granted summary judgment the same day to one of the two defendants whose motion was just (or about to be) denied. Even though plaintiff alleged that the defendant failed to update its label to match the approved branded labels—and presumably echoed its allegation in opposing two motions to dismiss—when defendant came forward with evidence that its label did match during the period of plaintiff’s use of the drug, plaintiff did “not contest the admissibility of Defendant’s summary judgment evidence, dispute the accuracy of Defendant’s statement of undisputed material facts, or present any adverse evidence demonstrating the existence of a genuine dispute as to the material facts of this case.” 2013 WL 3293704, *1. Instead, plaintiff argued that matching the approved branded labels was not enough and defendant should have sent Dear Doctor letters. Id. at *3. The court rightly, and quickly, rejected this argument based on the Fifth Circuit decision in Morris v. Pliva, Inc., 713 F.3d 774, 777 (5th Cir. 2013). Id. So, the claim really was preempted when considered on summary judgment, which is clearly the right result.
Two things gall us here, though. The first is that, for all the citation of TwIqbal standards in the orders on the motions to dismiss and the opportunities afforded plaintiff to tweak with her complaint, plaintiff was allowed to play fast and loose with what conduct she alleged created liability for the defendants. When it suited her, her claims were for failure to comply with the FDCA and FDA regulations. Or they were not.
Sometimes she contended the generic labels were inadequate simply because they did not match the approved branded labels. Sometimes she contended matching was not enough. This should not be allowed to happen. Second, at some point while she was opposing the second motion to dismiss—or perhaps well before that—plaintiff realized that her allegations were false and the undisputed evidence was that one of the remaining defendants had a matching label. Rule 11, for example, requires that “the factual contentions [in pleadings and briefs] have evidentiary support or, if specifically so identified, will likely have evidentiary support after a reasonable opportunity for further investigation or discovery.” This extends to “later advocating” the positions taken pleadings and briefs. We wonder if this court or another faced with similar facts will impose sanctions against the plaintiff’s counsel under Rule 11 or attorney’s fees under Rule 54(d). We understand the pleading standards and that not every grant of summary judgment for the defendant in a drug and device case means that the plaintiff lacked a basis to bring the suit or even oppose summary judgment.
The gap in Mensing for failing to update the generic label to match the approved branded label, however, apparently creates an incentive for plaintiffs to allege a specific fact pattern to avoid preemption. We do not know the facts for all of these cases, except as we have gleaned and retained them from the many decisions we have seen, but we expect that lawyers who bring these cases over and over do. If they persist in boilerplate allegations against a bunch of defendants even after they know their allegations are not well-founded, then maybe some court will get their attention.