We’ve already blogged about Judge Sweet’s decision last November in the Cymbalta litigation and why it was important. McDowell v. Eli Lilly, 2014 WL 5801604 (S.D.N.Y. Nov. 7, 2014). Its most important aspect was Judge Sweet’s holding that the label’s warning of the withdrawal risks associated with discontinuing Cymbalta was sufficient as a matter of law even though it didn’t include the clinical trial statistics that plaintiffs argued should have been included. The language of the warning was sufficient, so the statistics were unnecessary. Another important aspect of the decision was the nurse practitioner’s knowledge of the withdrawal risks of Cymbalta, which eviscerated plaintiff’s warnings causation case.
Judge Sweet had another opportunity to address these same issues just last week. The plaintiff had moved for reconsideration. See McDowell v. Eli Lilly, 2015 U.S. Dist. LEXIS 23445(S.D.N.Y. Feb. 26, 2015). Maybe he shouldn’t have. The court’s opinion reminds us of the ruling that Lieutenant Commander Galloway received from the bench after her follow-up “strenuous objection” in A Few Good Men. That ruling triggered this bit of sarcasm from Lieutenant Weinberg as the court was emptying: “I strenuously object?" Is that how it works? Hmm? ‘Objection.’ ‘Overruled. ‘Oh, no, no, no. I STRENUOUSLY object.’”
The first time around, plaintiff argued that the warning should have laid out the percentage of patients in the “Perahia” study – 44.3% – who had experienced withdrawal symptoms, rather than the descriptive language in the actual label and the list of identified withdrawal symptoms that had occurred in greater than or equal to 1% of the patient population. The study and the argument weren’t new issues for the court. Plaintiff used them in his opposition to summary judgment. This time he simply attached the study to a belatedly filed expert report.
Similar to Lieutenant Commander Galloway’s strenuous objection, this re-argument offered the court one more chance to announce that the label properly characterized the potential withdrawal risks without those statistics:
This was the express theory of McDowell's complaint . . . and the precise point made by McDowell's opposition to Eli Lilly's original motion: . . . The alleged discrepancies between the discussion of discontinuation symptoms in the label and the results of the 2005 Perahia study were considered and it was concluded that, as a matter of law, the label properly characterized the risks. . . . The report by a retained expert does not constitute "new evidence" meeting the high standard warranting reconsideration.
Id. at *7. In fact, the warnings section of the label was loaded with information on withdrawal risks:
Discontinuation: May result in symptoms, including dizziness, nausea, headache, fatigue, paresthesia, vomiting, irritability, nightmares, insomnia, diarrhea, anxiety, hyperhidrosis, and vertigo (5.6). . . .
If the decision has been made to discontinue treatment, medication should be tapered, as rapidly as is feasible, but with recognition that discontinuation can be associated with certain symptoms [see Dosage and Administration (2.4) and Warnings and Precautions (5.6) for descriptions of the risks of discontinuation of Cymbalta]. . . .
Discontinuation symptoms have been systematically evaluated in patients taking duloxetine. Following abrupt or tapered discontinuation in placebo-controlled clinical trials, the following symptoms occurred at a rate greater than or equal to 1% and at a significantly higher rate in duloxetine-treated patients compared to those discontinuing from placebo: dizziness, nausea, headache, fatigue, paresthesia, vomiting, irritability, nightmares, insomnia, diarrhea, anxiety, hyperhidrosis and vertigo.
During marketing of other SSRIs and SNRIs (serotonin and norepinephrine reuptake inhibitors), there have been spontaneous reports of adverse events occurring upon discontinuation of these drugs, particularly when abrupt, including the following: dysphoric mood, irritability, agitation, dizziness, sensory disturbances (e.g., paresthesias such as electric shock sensations), anxiety, confusion, headache, lethargy, emotional instability, insomnia, hypomania, tinnitus, and seizures. Although these events are generally self-limiting, some have been reported to be severe. Patients should be monitored for these symptoms when discontinuing treatment with Cymbalta. A gradual reduction in the dose rather than abrupt cessation is recommended whenever possible. If intolerable symptoms occur following a decrease in the dose or upon discontinuation of treatment, then resuming the previously prescribed dose may be considered. Subsequently, the physician may continue decreasing the dose but at a more gradual rate.
It’s not difficult to nit-pick even well-conceived warnings after the fact. We’ve seen plaintiffs’ lawyers do it effectively many times. Si it’s gratifying to see a court reject this approach. Twice.
Plaintiff’s reconsideration motion also lobbed a European label into the case. This is familiar tactic. Labeling in different countries varies. Different countries have different regulatory and medical systems. Plaintiffs almost always try to take advantage of these differences and, for the most part, are rejected by the courts. Trying such a complex issue is more likely to bog down the trial than provide a jury with material information. Judge Sweet rejected the use of the European label outright:
The mere existence of a differently structured and written European label does not establish that the U.S. label is insufficient, misleading, or legally inadequate, nor is foreign regulatory action even appropriate as a subject of expert testimony in pharmaceutical cases.
Id. at *12.
Finally, plaintiff’s reconsideration motion focused on warnings causation and the testimony of the prescriber. This too was one of the main focuses of the court’s summary judgment opinion. But the prescriber, a nurse practitioner, had already testified that she knew that Cymbalta carried a significant withdrawal risk and that she would still have prescribed Cymbalta even if the label included the statistics from the Perahia study. Id. at *14. That’s bad testimony for a plaintiff. In fact, it’s case-ending testimony. So in his reconsideration motion the plaintiff focused on a hypothetical question. Plaintiff’s attorney had asked the prescriber whether she would have continued to prescribe Cymbalta if she had been told that its withdrawal risk was as bad as or worse than that of a particular competitor drug – and the nurse testified that she might then have chosen a different drug. Id. The problem is that there was no support in the record for the hypothetical. Id. The court called it a false hypothetical. Id. When this type of questioning happens, the defense is forced to highlight the lack of support for the question either at the deposition or (more often) afterward in motion practice. A good court will then exclude the question. That’s what Judge Sweet did:
[T]he absence of any basis in the record for this hypothetical was sufficient reason to exclude it. Plaintiffs in pharmaceutical cases may ask doctors what they would have done in the face of a stronger warning . . . but that hypothetical stronger warning must have a factual basis.
Id. at *14-15.
The motion for reconsideration didn’t raise anything new other than an opportunity for the court to explain in more detail why judgment should be entered against plaintiff. The label’s sufficient warning remained sufficient, and the absent warnings causation remained absent. And the court denied plaintiff’s motion for reconsideration, upholding judgment for the defendant.