This post is from the non-Reed Smith side of the blog.
We haven’t talked about the district court decisions in Dolin v. GlaxoSmithKline LLC, because in our opinion there simply hasn’t been anything good to talk about. Plaintiff sued the manufacturer of the brand drug Paxil arguing that the brand manufacturer should be liable for the death of her husband who took a generic version of the drug manufactured by a different company. Plaintiff alleged that the brand manufacturer should have amended the warning that accompanied its drug to include the risk of suicidality in adults even though the FDA had expressly rejected such a warning change several times. So, the fact that the defendant’s innovator liability and federal preemption based summary judgment motions were denied left us frankly baffled at how two district courts got it wrong on both counts.
But, now we finally have something good to talk about. The Seventh Circuit just reversed the verdict for the plaintiff in this case overturning the district court’s preemption decision. In Dolin v. GlaxoSmithKline LLC, ___ F.3d ___, 2018 WL 4001208, slip op. (7th Cir. Aug. 22, 2018), the appellate court found overwhelmingly clear evidence that the FDA rejected the very warning proposed by plaintiff, applied Mensing to a brand manufacturer, and also found no evidence of newly acquired information to support a unilateral label change. That’s definitely worth talking about.
Before we get to all those great decisions, we note that the Seventh Circuit opted not to reach the innovator liability question finding that the “evidence of federal preemption is decisive.” Id. at p.25. However, the court did note that the issue had not yet been decided by the Illinois courts, and therefore any ruling would have to be “a prediction of state law under Erie.” Id. We point this out as one of problems with the district court decisions was that in the face of an undecided state law question, they over-reached in creating innovator liability where such a claim did not exist. It was an improper expansion of state law. Federal Erie predictions are supposed to be conservative, not radical as was the case here. As discussed in our “Innovator Liability at 100” post, every other court of appeals to consider innovator liability on its merits, has rejected it. That’s seven courts of appeals (4th, 5th, 6th, 8th, 9th, 10th, & 11th), construing the law of 23 states — including Illinois. See In re Darvocet, Darvon, & Propoxyphene Products Liability Litigation, 756 F.3d 917, 944-45 (6th Cir. 2014).
Turning to the dispositive preemption question, defendant argued that that the FDA would not have allowed it to include the warning sought by plaintiff and therefore plaintiff’s state law tort claim conflicted with federal law and was preempted. Under the standard announced in Wyeth v. Levine, 555 U.S. 555 (2009), a state law claim based on labeling is not preempted “if the manufacturer could have added the warning unilaterally under the [Changes Being Effected (“CBE”)] regulation.” Dolin, at p.15. Under the CBE regulation, a manufacturer can change its label without advance FDA permission if the manufacturer has “newly acquired information.” Id. at p.4. But Levine also held that “there could be preemption if the manufacturer met the stringent standard of proving that there was clear evidence the FDA would have rejected the proposed change in the drug’s label.” Id. at p.16. So, the issue for the court was whether or not defendant could have changed the drug’s label using the CBE regulation.
Before getting to the substance, we’ll start with another question the court decided not to decide – whether preemption under Wyeth v. Levine, 555 U.S. 555 (2009) is a question of fact or law. The court acknowledged the recently created split in the circuits created by the Third Circuit’s decision in In re Fosamax Products Liab. Litig., 852 F.3d 268 (3rd Cir. 2017) (our worst case of the year for 2017), on which, thankfully, the Supreme Court has granted certiorari to review. But even the Third Circuit decision left a window open. One in which Dolin certainly fit. “[W]hen no reasonable jury applying the clear-evidence standard could conclude that the FDA would have approved a label change, then the manufacturer will be entitled to judgement as a matter of law.” Dolin, slip op. at p. 18 (quoting In re Fosamax).
More specifically, the Seventh Circuit held that “no reasonable jury could find that the FDA would have approved an adult-suicidality warning for Paxil under the CBE regulation between 2007 and [decedent’s] suicide in 2010.” Id. Why was it such an open and shut case? About one-third of the opinion is taken up with setting out the extensive regulatory history demonstrating that both GSK and the FDA thoroughly examined the issue and the FDA completely rejected the addition of an adult-suicidality warning.
- June 1991 – based on supplemental analysis of data related to suicide the FDA determined there was no signal for additional risk of suicide, id. at p.7;
- September 1991 – an independent committee convened by the FDA “unanimously agreed that there is no credible evidence of a causal link,” id.;
- January 2004 – FDA concluded based on review of multiple data sets that there was no increased risk of suicide, id. at p.8;
- 2004 – the FDA requires a black box warning regarding an association between SSRIs (the class of drugs to which Paxil belongs) and suicide in pediatric patients but not adults, id.;
- April 2006 – GSK unilaterally changed its label under CBE regulations to include the risk of suicide in adults, id. at p.9;
- November 2006 – the FDA completed a meta-analysis that led it to conclude “the net effect appears to be neutral on” adult suicidality, id. at p.10;
- 2007 – the FDA orders class-wide labeling to state that “studies did not show an increase in the risk of suicidality with antidepressants . . . in adults beyond the age of 24,” id. at p.11;
- On at least 4 separate occasions in 2007, GSK asked the FDA whether it could retain the warning it added via CBE in 2006 and the FDA said no. Id. at p.11-13.
The court found this evidence “undisputed,” id. at p. 18, and held “[i]t is hard to imagine clearer evidence that . . .the FDA would not have approved a change.” Id. at p.19. It compared this regulatory history to the one the Supreme Court considered in Levine and found the Paxil evidence filled all the “evidentiary gaps” which led to the Levine non-preemption ruling. The risk of suicidality was given “more than passing attention” by the defendant and the FDA. Defendant provided the FDA with re-analyzed data in 2006. Defendant unilaterally changed the label to add a warning of the risk at issue. And, the FDA, more than once, rejected the warning. Id. at p.19-20.
Plaintiff made two arguments in response. First she argued that the FDA only rejected the defendant’s proposed warning because defendant “proposed adding it to the wrong spot on the label.” To which the court responded:
Plaintiff asks us to believe that the FDA – after deciding against an adult-suicidality warning based on its own analysis – rejected [defendant’s] warning only because GSK proposed putting it in the wrong place. That is unreasonable.
Id. at p.21. The court called that “unreasonable.” We’d probably have found a stronger adjective.
Plaintiff’s second argument is more important because it emphasizes what Levine preemption is at its core. Plaintiff argued that the defendant could have asked for a formal meeting with the FDA and therefore defendant lacks clear evidence that the FDA would have rejected the warning after such a meeting. Id. This argument misses the mark because “[s]tate laws requiring a label change are preempted unless the manufacturer could unilaterally add the new warning under the CBE regulation.” Id. In other words, if you need FDA approval there is conflict preemption.
This is where Pliva, Inc. v. Mensing, 564 U.S. 604 (2011), comes in and we of course laud its use in brand drug cases. Mensing held that claims against generic drug manufacturers were preempted because generic manufacturers are not permitted to unilaterally change a drug’s label. Because generic manufacturers cannot independently comply with their state law duties — changing their label would require “special permission or assistance” from the FDA — claims against them are preempted. Equate that to plaintiff’s argument in Dolin:
The preemption analysis asks only whether [defendant] could have added the adult-suicidality warning through the CBE regulation, not whether [defendant] might have been able to persuade the FDA to change its mind in a formal meeting – and certainly not whether [defendant] could have persuaded the FDA after already asking four times to include the warning and being told no four times.
Id. at p.22.
That takes care of the preemption issue up until 2007, when the FDA repeatedly rejected defendant’s CBE warning language. But plaintiff’s husband did not take the drug until 2010. So the remaining question is whether between 2007 and 2010, defendant acquired any “new” information that would have permitted a CBE label change during that period. Newly acquired information is defined as “data, analyses, or other information not previously submitted to the Agency.” 21 C.F.R. § 314.3. Plaintiff argued that in the data defendant submitted to the FDA, it “improperly attributed suicides that occurred in the wash-out phase of the drug tests as occurring on the placebo.” Dolin, slip op. at p.23. But the evidence showed that the FDA was aware that wash-out events were included and that defendant re-submitted the data to the FDA excluding the wash-out phase. Id. at p.24. Plaintiff also pointed to an article published in 2011 but the article was based on a 2006 analysis. So there was no evidence that the analysis was either new or not previously submitted to the FDA. Id. The court, therefore, concluded that plaintiff had offered no evidence that defendant acquired any new information after 2007 that would have supported a CBE label change.
All this was strong preemption evidence indeed. We note that Dolin was written by Judge Hamilton, who is no friend of preemption generally. See Bausch v. Stryker Corp., 630 F.3d 546 (7th Cir. 2010), also authored by Judge Hamilton.
We were always optimistic that the district court’s decisions in Dolin and the verdict would not stand and the Seventh Circuit did not disappoint (well, maybe they could have tossed innovator liability too, but a nod to Erie doesn’t hurt).