Photo of Michelle Yeary

It can sometimes be difficult for us here at the DDL Blog to address “mixed bag” cases. We are quite clear that we are a defense side blog. Love us or hate us – we don’t pull punches. We hoard and covet preemption and learned intermediary wins and treat each one like the gem that it is. We collect and pile up class action denials, no duty cases, and Lone Pine orders. We have scorecards, and cheat sheets and surveys all designed to celebrate defense victories and assist in creating more defense victories. So, when faced with a case that only goes half way, we are left a bit deflated. It’s sort of like seeing a movie where you say the special effects were amazing, but the story – not so much. You can’t trash the movie because there was something decent. But you also aren’t giving it two thumbs up (you know, the standard movie rating before it became based on tomatoes). If a decision is all bad, we trounce it. If a decision is all good, we praise it. If it’s somewhere in the middle, we talk about it.

That about sums up our take on McWilliams v. Novartis AG, 2018 U.S. Dist. LEXIS 113862 (S.D. Fla. Jul. 9, 2018). It is a failure to warn case concerning a drug used to treat leukemia. Id. at *2. Plaintiff suffered a stroke after using the drug to treat his chronic myeloid leukemia for a little over 2 years. Id. at *3. Defendant moved for summary judgment on the grounds of preemption, adequacy of the warning, lack of proximate causation, and no punitive damages under the law of New Jersey. The court denied the first three and granted the motion on punitive damages.

As to preemption, defendant moved that the case should be dismissed based on impossibility preemption arguing that there was clear evidence that the FDA would have rejected a warning on the risk of stroke if defendant had proposed one. In support of its argument, defendant noted multiple occasions on which the issue of arterial and vascular occlusive events was raised with the FDA before the date of plaintiff’s stroke:

  • In 2011, Defendant proposed adding peripheral artery occlusive disease to the Medication Guide and the Adverse Reaction section of the label and the FDA rejected the proposal.
  • Twice in 2013 before plaintiff’s stroke, Defendant told the FDA of labeling revisions required by Canada regarding cerebrovascular events.
  • Later in 2013, the FDA required another drug in the same class to add a boxed warning that included language that “similar rates of serious vascular events have not been observed in several other drugs of this class.”

Id. at *5-6. We view that as a pretty well-documented history of FDA regulatory rejection or inaction related directly to the risk at issue. Perhaps most compelling to us was the fact that when the FDA added a “serious vascular events” warning to another product it specifically said defendant’s product (as one of the others in the class) didn’t have the same risk profile. That feels like clear evidence that the FDA would have rejected that warning if proposed by defendant. The court reasoned that just because the FDA changed one label doesn’t mean that it would not have change another. Id. at *11-12. But that misses the language the FDA used distinguishing the drug that required the warning and the others that did not.

Unfortunately for the defendant in this case, the court compared its regulatory history to the histories set forth in the cases cited where clear evidence was found and concluded that the others had more extensive histories. But is that the correct analysis? Perhaps the evidence of rejection has been stronger in other cases, but that doesn’t make the evidence in the current case any less clear.

Moving on to the actual warning itself, defendant’s first argument was that the risk of stroke was not known or knowable during the time period that plaintiff took the drug. Defendant argued that there were no article or published reports of patients experiencing strokes while taking the drug until 2 adverse events were noted in 2013. Id. at *16. Plaintiffs on the other hand argue that Canadian regulators provided defendant of information regarding an association between the drug and atherosclerotic disease in 2012. That’s not the same thing as strokes, but it was enough for the court to find a disputed issue of fact. Id. at *16-17.

Defendant also challenged proximate cause arguing that plaintiff could not prove that his prescriber would not have prescribed or would have stopped the prescription if he received a different warning. Id. at *17. As evidence of this, defendant relied on the prescriber’s testimony that he continues to prescribe the drug today. Id. Plaintiff countered with the prescriber’s testimony that when he prescribes today he does so “with careful warning.” The court concluded that because the prescriber has now changed his warning to patients, causation remains in dispute. Id. at *19.

That leaves plaintiff’s punitive damages claim which turned on choice of law. Plaintiff is a Florida resident, was prescribed the drug in Florida, and suffered his stroke in Florida. Defendant is a New Jersey corporation. Florida allows claims for punitive damages in prescription drug cases. New Jersey does not. See N.J.S.A 2A:58C-5 (no punitive damages for drugs and devices that are approved by the FDA). There’s a clear conflict. Florida uses the significant relationship test to resolve conflicts. McWilliams, at *23. Which means the court has to decide which state has the most significant relationship to the particular issue. Id.

We are all familiar with the assumption in most personal injury cases that the law of the place of injury applies. And that used to be the beginning, middle, and end of the inquiry. Until about a decade ago, the argument that defendant’s domicile should control for punitive damages didn’t have much support.  It is in defendant’s home state that the conduct that allegedly serves as the basis for punitive damages takes place. While the drug was marketed to doctors in Florida, “the alleged sales and marketing strategies originated from [defendant’s] New Jersey headquarters, and plaintiffs have not identified any way in which those strategies were implemented differently in Florida than any other state.” Id. at *26. According to the Restatement (Second) of Conflict of Laws, when the issue under consideration is about deterrence or punishment, the place where the conduct took place may have the dominant interest. Id. at *25. Since punitive damages are “designed to deter and punish” the defendant rather than compensate the plaintiff that means New Jersey has the more significant relationship and its law applies.

In this case that means no punitive damages. But as we’ve mentioned before, one of the reasons we don’t spend much time on choice of law is the rulings are double-edged and can easily be turned to apply the law of states we don’t like. As we said at the outset, this is a mixed bag case so we’ll take from it what we can get. Good news for New Jersey drug and device companies sued in other states.