We like bright lines in the law. They streamline arguments for lawyers and, more important, they make it easier for non-lawyers to conduct their affairs with some degree of predictability. Rear-end a car and you’re liable, even if the other guy stopped short. Leave a sponge behind in a patient’s abdomen, and you and your operative team are on the hook. The thing speaks for itself, and the thing it speaks is very bad for your defense case. Utter a negative statement about someone in a courtroom, and you’re immune from libel liability. Say the same thing on the courthouse steps, and you might be in trouble. Sell a pharmaceutical product with a federally approved label, and when someone sues you for failure to warn you’re … uh oh. Thanks to the Supreme Court’s opinion in Wyeth v. Levine, you’re still in the soup if you could have added warnings via the Changes Being Effected (CBE) provision. You might not like that rule, but at least it’s clear, right? Wrong. Recent opinions have come out on opposite sides as to what would constitute “clear evidence” that the FDA would have rejected an additional warning, or even who gets to decide that ‘fact.’

But at least we know that the CBE provision is available only in certain circumstances, right? Thanks to the recent case of Blackburn v. Shire US, Inc., 2017 WL 1833524 (May 8, 2017), even that inquiry is muddy. The plaintiff in Blackburn alleged that he contracted kidney diseases from a drug known as LIALDA. Please forgive us for thinking that the failure to warn claim is a straight loser, since the LIALDA warned of precisely the injuries suffered by the plaintiff. That would be a nice example of a bright line rule. The defendant filed a motion to dismiss the entire First Amended Complaint. That motion to dismiss was strongly supported by the Wyeth v. Levine “clear evidence” standard. Or at least it should have been. The motion to dismiss was also strongly supported by the Pliva/Bartlett independence/impossibility principle. Or at least it should have been. The Blackburn court granted the motion to dismiss only in part, and it is hard to characterize the court’s reasoning as anything but disappointing.

The plaintiff argued that the label was defective because it merely recommended “periodic” evaluations for possible kidney damage. The plaintiff argued that “periodic” was too vague. According to the plaintiff, “periodic” might mean semi-monthly or annual. According to the plaintiff, the LIALDA label should have suggested evaluations every month for the first three months after therapy, then on a quarterly basis for at least one year. The court bought the plaintiff’s argument, which amounts to a duty to tell doctors how to practice medicine. Technicality and a belief in the vicissitudes of juror deliberations triumph over common sense. This ruling is further proof that no matter what a warning label says, a clever plaintiff lawyer and a cautious court can always conjure up an ‘improvement’ to a product label and thereby keep a weak case on the docket.

Even assuming that “every month for a while, then quarterly” would be an improvement, or even a distinction with a difference vs. “periodic,” the defendant had a pretty good argument in support of preemption. The CBE supplementation process applies only when “newly acquired information” becomes available and would support the label change that the plaintiff wants. Prior to the LIALDA label, there was already information regarding the sort of testing protocol argued for by the plaintiff. That’s not just how we defense hacks parse the record. The plaintiff admitted “that evidence of the renal toxicity of LIALDA and his proposed testing regimen existed prior to LIALDA’s FDA approval.” Blackburn, 2017 WL 1833524 at *4. That pre-existing information environment should support preemption. Shouldn’t it? No, said the court, because there was additional evidence about testing after the label went into effect. Such incremental evidence could have had a cumulative effect, and that effect could constitute “newly acquired information.” Ugh. How much more is enough to make a difference? Is mere repetition enough? Who knows? So we have a label that warned of the injury suffered, advised periodic testing, and even if it didn’t specify the period as much as the plaintiff would like, there was already pre-label information about such testing. Maybe the only bright line the court favors is one against preemption.

The plaintiff’s proposed change faced another preemption barrier. The change would seem to apply both to the “Full Prescribing Information” section in the label and the “Highlights” section. The regulations (21 C.F.R. §314.70(b)(2)(v)(C), to be specific) do not permit a drug manufacturer to change the Highlights section without FDA approval. That would be a “major change.” The plaintiff argued that the FDA might have approved a change in the Highlights section, but that sort of speculation is foreclosed by the Pliva/Bartlett independence/impossibility principle. If a manufacturer cannot take unilateral action, a lawsuit insisting on such action is preempted, and even this cautious court agreed that any change to the Highlights section would be preempted. The “periodic” language criticized by the plaintiff appeared in both the Highlights section as well as the body of the label. Is it possible to demand change in one without demanding the same for the other? The Blackburn court was not sure, but gave the plaintiff the benefit of the doubt. The plaintiff might be going forward on “shaky ground” (2017 WL 1833524 at *7), but the court permitted the plaintiff to proceed under the highly fictional notion that the term “periodic” in the Highlight section is okay so long as the body of the label supplies a “mere clarification” of what “periodic” means. So the elaboration of “periodic” to mean a particular testing regimen manages to be both crucial and perfunctory at the same time.

Even aside from preemption, the learned intermediary doctrine should have barred the plaintiff’s failure to warn claim. There was no allegation that the treating doctor would have prescribed a different treatment. It is not even clear that there was a specific, plausible allegation that the doctor would have done anything different. If preemption is Scylla and the learned intermediary doctrine is Charybdis, how did the plaintiff evade both? The court thought it sufficient that the plaintiff alleged that the treating doctor was a gastroenterologist, and that he probably missed the nephrology literature that expounded on the preferred testing regimen. Thus, the First Amended Complaint offered sheer speculation that the treating doctor would have implemented a different testing regimen if only the label had informed him of as much. But, did the treater read beyond the Highlights? That treating doctor’s deposition will turn out to be quite important, though the court’s roadmap promises a bumpy trip, filled with blind alleys and detours.

Not everything the Blackburn court did was dismal. The court dismissed the express warranty claim because the LIALDA label cannot be construed as “an express warranty of safeness.” Id. at *9. What the label did say was that LIALDA was safe to treat the illness at issue, and that the medicine presented a risk of precisely the injury suffered. How can there be a breach of warranty? At this point, we are nodding our head in agreement, quickly followed by shaking it in puzzlement over how this rationale does not equally apply to the failure to warn claim. The court also dismissed the fraud claims because there was no misstatement of a material fact. Rather, the plaintiff alleged only deficiencies with the label’s recommendation for testing. We certainly agree that the alleged deficiency is not material, but then wonder how that same immateriality permitted the plaintiff to circumvent both preemption and the learned intermediary doctrine.