Everything seemed to go right for the defense in Anderson v. Abbott Labs., 2012 U.S. Dist. LEXIS 141585 (N.D. Tex. Sept. 30, 2012). The court applied Buckman preemption, it applied TwIqbal, it dismissed the case early, and it didn’t allow an amended complaint. Everything was good in Texas. It’s an easy decision to blog about.
But it also provides a basis to remember why states like Texas have chosen to defer to FDA decisions on labeling and safety. It gives manufacturers greater clarity on the process that they must follow when developing and marketing drugs. It promotes predictability and with that efficiency. Manufacturers can rely on the process and decisions of the FDA, their primary regulator and an expert in the field, rather than engaging in the almost hopeless task of predicting what multiple future juries might decide. In the end, this can lower cost and help speed newer medicines to market.
The particular drug at issue in Anderson was Humira, a TNF blocker developed by Abbott and approved by the FDA to treat rheumatoid arthritis and other autoimmune conditions. Id. at *2. The case was filed by parents of a child whose doctor prescribed Humira for the off-label treatment of psoriasis. They claimed that Humira caused the child to develop pediatric leukemia. Id. Just after the doctor prescribed Humira, the FDA issued an “Early Communication” announcing that it was investigating a possible association between Humira (and other drugs in its class) and pediatric leukemia. When the FDA finished its investigation, it required a warning in the label for Humira and other drugs in its class as to an increased risk of adolescent leukemia. Id. at *3. While the investigation was still ongoing and before the FDA required a label change, however, the child in the Anderson case developed pediatric leukemia.
You can easily see how these facts might make an attractive plaintiff’s case in front of a jury. There was concern about a potential association between the drug and the disease, but the label’s warning section didn’t address it until after the child developed the disease. To say the least, it would be difficult to convince a jury to make a complicated policy decision to defer to the process and decisions of the FDA, rather than to act upon their understandable sympathies for the child.
That’s where statutes like Texas’s come in. They don’t let such difficult policy decisions get to a sympathetic jury. The Texas statute says that an FDA-approved warning is adequate as a matter of law – in other words, it’s not a fact decision for the jury – unless certain limited exceptions apply. Tex. Civ. Prac. & Rem. Code §82.007(a).
The Anderson plaintiffs relied on the exception that applies if a manufacturer withholds from the FDA material information that it requires. Id. §82.007(b)(1). That sounds like something that would be preempted under Buckman. And it is. The Anderson court said so, citing the 5th Circuit’s opinion in Lofton v. McNeil Consumer & Specialty Pharms., 672 F.3d. 372 (5th Cir. 2012), which we blogged about here and which held that this exception was preempted under Buckman:
With regard to Plaintiffs’ claims for strict liability, negligence, and breach of warranty under section 82.007(b)(1), the court concludes that these claims are preempted based on the court’s reasoning in Lofton. Accordingly, Plaintiffs have failed to state claims upon which relief can be granted under section 82.007(b)(1), and Abbott is entitled to dismissal of all of Plaintiff’s claims under this section.
Anderson, 2012 U.S. Dist. LEXIS 141585, at *10. This is FDA deference in about the purest form that you’re going to get it in litigation. The FDA was monitoring this class of drugs, including this very issue, and approved a label regarding its safety and efficacy, and Texas wasn’t going to let that regulatory process be shaken by varying decisions of different juries.
Now, as can be expected, the plaintiffs tried to get around the Texas statute. They argued that the FDA no longer approved the label for Humira once it issued its “Early Communication.” But an FDA “Early Communication” isn’t a final labeling decision. By its very name, it’s early. The “Early Communication” involved in Anderson said itself that the “FDA will communicate the conclusions and any resulting recommendations to the public after it completes its evaluation.” (Emphasis added.) The court understood this perfectly:
The court further rejects Plaintiffs’ alternative argument that section 82.007(b)(1) is inapplicable because Abbott’s Humira warning did not include the most current FDA warning information. Plaintiffs have cited no authority, and the court is aware of none, to support their argument that the Early Communication affected the FDA’s prior approval of Humira warnings or that Abbott was required to revise its Humira label warning prior to August 9, 2009, the date the FDA required warnings for all TNF blocker drugs to include information about the increased risk of lymphoma, adolescent leukemia, and other malignancies in children and adolescents. Moreover, if as Plaintiffs contend, that Abbott was required as a result of the FDA’s Early Communication to warn of the potential effects of TNF blockers to children and adolescents, there would have been no reason for the FDA to subsequently mandate such warnings in August 2009. Accordingly, this argument does not save Plaintiffs’ claims from the applicability of section 82.007(a) and presumption that the warning in the FDA-approved Humira label was adequate as a matter of law.
Id. at *10-11.
This FDA history, in fact, highlights that the FDA was actively evaluating this association and proper labeling of Humira and other drugs in its class. Jury decisions would serve only to second-guess that process, undermine it, making it harder for manufacturers to rely on it.
The plaintiffs tried yet another avenue though. When the court asked the parties to provide supplemental briefing on Buckman preemption, the plaintiffs submitted a supplement brief that didn’t address Buckman at all. Instead, plaintiffs for the first time said that they were relying on a different exception to the Texas statute’s presumption of adequacy, one that applies when there’s been “off-label” promotion. But that exception requires fairly particular facts:
The claimant may rebut the presumption in Subsection (a) as to each defendant by establishing that . . . (A) the defendant recommended, promoted, or advertised the pharmaceutical product for an indication not approved by the United States Food and Drug Administration; (B) the product was used as recommended, promoted, or advertised; and (C) the claimant’s injury was causally related to the recommended, promoted, or advertised use of the product.
Abbott argued, and we agree, that this section requires the manufacturer to have promoted the drug off-label to the prescribing doctor and that the promotion caused the doctor to prescribe it. Id. at *13. Plaintiff argued that it had made such allegations, pointing to general “information and belief” allegations that Abbot had engaged in such promotion with the prescriber. Id. at *14.
Enter TwIqbal. The court found these allegations to be boilerplate, and insufficient under TwIqbal:
The court concludes that Plaintiffs’ allegations in this regard are insufficient to withstand dismissal under Rule 12(b)(6). As previously noted, “[w]hile a complaint need not contain detailed factual allegations, it must set forth “more than labels and conclusions, and a formulaic recitation of the elements of a cause of action will not do.” Twombly, 550 U.S. at 555 (citation omitted). Plaintiffs’ allegations are nothing more than a formulaic recitation of the first two elements under section 82.007(b)(3), and Plaintiffs allege no facts whatsoever in support of the third causation element. Accordingly, the court determines that Plaintiffs have failed to state a claim under section 82.007(b)(3), and Abbott is entitled to dismissal of Plaintiffs’ claims under this section.
Id. at *14-15.
The court dismissed the complaint with prejudice, allowing plaintiffs no opportunity to amend. In fact, they didn’t even ask for one. So the court assumed that this was the best that the plaintiffs could do.
There’s lots of helpful stuff in this opinion. When everything goes so well for the defense, however, it’s useful to step back and consider why it should be that way. And this case, more than many, provides a fairly pure example of a legislature’s policy decision to keep juries from meddling with FDA decisions actually working.