Facts of the Case. Andrea and George Perry's doctor prescribed Elidel, an anti-eczema drug, for their two-year-old son, Andreas, who later contracted lymphoma. The FDA approved Elidel in December 2001. Even at that time, however, concerns had been raised about a possible link between "calcineurin inhibitors", the class of drug to which Elidel belonged, and cancer. The labeling language that Novartis proposed to the FDA was general in nature, and did not warn of the risk of cancer.

Because of its concerns about Elidel, the FDA imposed significant restrictions on the drug: it was allowed only for short-term and intermittent treatment of eczema, and only when traditional topical remedies had failed.

In 2003, six months after the drug was prescribed to Andreas, an FDA advisory committee recommended that Novartis add a "black box" warning against the use of Elidel by children younger than two. In March 2005, the FDA ordered Novartis to add such a warning, but Novartis did not do so in January 2006.

The Perrys filed suit against Novartis, arguing that the company had promoted Elidel as safe for use even though it knew that the drug could cause cancer.

The Pre-Emption Argument. Novartis argued that the preamble to an FDA FInal Rule, "Requirements on Content and Format of Labeling for Labeling for Human Prescription Drug and Biological Products," 71 Fed. Reg. 3922 (January 24, 2006), pre-empted failure-to-warn claims.

The preamble states that the following claims would be pre-empted:

(1) failing to put in the Highlights or otherwise emphasize information that appears elsewhere in the labeling; (2) failing to include in an advertisement information that appears in the labeling; (3) failing to include contraindications or warnings that are not supported by evidence that meets FDA standards; (4) failing to include a statement in labeling or in advertising that had been proposed to the FDA if that statement was not required by FDA at the time plaintiff claims the sponsor had an obligation to warn; (5) failing to include in labeling or in advertising a statement that the FDA has prohibited in labeling or advertising; and (6) making statements that the FDA approved for inclusion in the drug's label (unless the manufacturer withheld material information relating to the statement).

Novartis contended that category (4) of the preamble applied to this case. The FDA submitted an amicus curiae brief in which it, too, argued that the Perrys' case was barred by pre-emption.

The Plaintiffs' Arguments Against Pre-Emption. In their brief, Perry's attorneys argued that the Supreme Court's general presumption against pre-emption and, in addition, that several factors further weakened the Novartis's argument. First, there was no evidence that Congress intended to pre-empt suits such that brought by the Perrys. Second, The Food, Drug and Cosmetic Act expressly barred pre-emption except in cases of "direct and positive conflict" (in fact, Congress was well aware of state-court suits when it passed the Act in 1938). Third, there was no "direct and positive conflict" between the Perrys' failure-to-warn claim and the FDA labeling regulation. On that point, the Perrys' attorneys pointed out that FDA rules expressly permit a drug manufacturer to upgrade label warnings simply by notifying the agency of the changes; and noted that a number of state and federal courts had cited those rules in support of their conclusion that failure-to-warn claims were not pre-empted.

Perry's attorneys also argued against the FDA's position that its rules are an upper limit, not a minimum standard, governing the information that must be given to doctors and patients. They noted that the FDA had reversed course: as recently as 1998, it took the position that its rules established minimum standards. They went on to argue that the case did not fall with category (4) of the preamble because the FDA had never rejected stronger warnings for Elidel but instead was slow in responding to the advisory committee's call to do so; and that Novartis knew, at the time that Elidel was prescribed for Andreas, that the drug increased the risk of cancer, and thus the company could have upgraded its warning without prior FDA approval. They also distinguished Colacicco v. Apotex, 432 F. Supp. 2d 514 (E.D. Pa. 2006), which dismissed an suit against the manufacturer of generic Paxil on pre-emption grounds. They pointed out that while Elidel was a brand-name drug, Colacicco involved an anti-depressant that belong to a family of drugs for which the FDA had on a number of occasions rejected stronger warnings.

Finally, the Perrys' attorneys attacked the preamble head-on. They characterized it as an expression of opinion, not a statement of the law, since the agency had not asked for comments on the issue of pre-emption before adopting the rule. Furthermore, they argued that since the pre-emption was even less tenable because the FDA had recently changed position: in the past, it took the stance that its rules did not have pre-emptive effect.

The Bottom Line. A lot has been written about so-called activist judges, but as Perry and other drug-liability cases demonstrate, we now have a federal agency that is attempting to behave like an activist court, as well as like defense counsel in the same proceeding.

Update on last month's blog entry. Supporters of regulatory-taking measures went 1-for-6 in the November elections. Arizona's takings initiative was approved by that state's voters but similar measures failed in California, Idaho, and Washington. Regulatory-takings measures were ruled off the ballot by the courts in Montana (on account of fraud in obtaining petition signatures) and Nevada (because combining eminent-domain and regulatory-takings measures violated the single-subject requirement of the state constitution).