Photo of Bexis

We’re pleased to report on another preemption victory, Horne v. Novartis Pharmaceuticals, 2008 WL 1847077 (W.D. N.C. Apr. 23, 2008), this time involving the drug Lotensin, which (logically enough) is used to reduce hypertension/high blood pressure. While the opinion is only a “partial” dismissal, it’s our sense that the “partial” victory covers the really important claims.
Horne is an implied preemption prescription drug decision that squarely presents one of the categories of claims that the FDA believes is preempted – an “FDA already said no” situation. The plaintiff asserted that certain birth defects were caused by the use of Lotensin during the first trimester of pregnancy. At the relevant time (2003) the drug had a boxed warning against use during the second and third trimesters, but not the first.
Three years after the fact (2006) a new study suggested that there was also a risk of first trimester birth defects. The FDA looked at it, and told the manufacturer not to change the black box warning to add the first trimester. In a public health advisory, the FDA stated:

At this time, based on this one observational study, the FDA does not plan to change the pregnancy categories for ACE inhibitors.

Horne, slip op. at 8 (quoting FDA 6/7/06 Public Health Advisory).
That’s what gave rise to preemption. It was clear from the public health advisory that the FDA had specifically reviewed the birth defect/first trimester evidence and told the defendant not to change its label in 2006, after receipt of new information. A fortiori it would not have permitted a different warning earlier, before even that new information existed. This FDA decision brought the case within the preempted categories of the FDA’s 2006 Preemption Preamble, which the court relied upon. Slip op. at 22-24. The first trimester issue raised a direct conflict:

The text of the warning label as approved by the FDA indicates that the FDA affirmatively considered the medical and scientific proof available at the time and concluded that the risks of birth defects and fetal injury do not appear to result from the use of the drug during the first trimester. . . .
To hold the Defendant liable for failing to provide an additional warning to the effect that use of [the drug] during the first trimester poses risks of birth defect and fetal injury when the FDA has already determined that such risks do not appear to result from use of the drug in the first trimester would create a direct conflict between the requirements of federal law and the requirements of state law and would place the Defendant in an impossible situation whereby the Defendant could not comply with federal law and state law at the same time.

Slip op. at 29-30. The direct conflict thus required preemption, id. at 31 – as it must if federal authority over drug labeling is to be maintained.
The court allowed plaintiff to continue with her other, vaguer (“no specific factual allegations”), non-warning related claims past the pleadings stage. Slip op. at 32-35. Since those claims might not have been based on the allegation of failure to warn (although we think they probably were) the court allowed such claims to survive for the time being.
That kind of a split decision – even if the most important claims bit the dust – is one of the hazards of arguing preemption as a motion to dismiss. We don’t generally recommend that procedural posture (although it does make for a less expensive and time consuming defense), because even if granted, motions to dismiss make for lousy record evidence of conflict on appeal. But in Horne, the court was willing to take judicial notice of the FDA’s public health advisory. Slip op. at 14-15. The advisory certainly established the conflict starkly enough.
In this instance, the split decision probably doesn’t give the plaintiff very much to work with. Dismissal of the warning claims leaves the plaintiff without the claims that are ordinarily most important in prescription drug litigation. The testing claim will eventually follow suit, since there is no such thing as a stand-alone failure to test claim. Finally, it’s rather rare to find a manufacturing or design defect claim successfully asserted against a prescription drug – at least one that has but a single active ingredient (as Lotensin seems to have).
Horne is thus the kind of split decision we don’t really mind all that much.