We do not mean the German Renaissance painter and thinker Albrecht Dürer. His work, while a poor cousin to that of some famous contemporaries to the south, remains as is. We mean the Supreme Court’s decision in Merck Sharp & Dohme Corp. v. Albrecht, 587 U.S. 299 (2019), which has been touted for the last five years as revitalizing preemption as a defense to failure to warn claims for branded prescription drugs. Albrecht stemmed from the Fosamax MDL created in 2011 to address plaintiffs’ pretty obviously preempted causes of action about the risk of atypical femoral fractures (“AFFs”) with a prescription osteoporosis medication. (We have written so many posts on preemption and other issues in the litigation that the summary below is necessarily incomplete. Feel free to click on the hyperlinks if you need a deeper dive.) After a number of prior preemption decisions—see here, here, and here—in 2017, the MDL court knocked out all of the design claims and the warning claims for cases claiming injury before September 14, 2010, and granted summary judgment to the branded manufacturer defendant in hundreds of cases. On the warnings part of it, the compelling regulatory record was enough to overcome the novel and fuzzy “clear evidence” standard articulated in Levine. Later that year, in In re Fosamax (Alendronate Sodium) Prods. Liab. Litig.852 F.3d 268 (3d Cir. 2017) (“Fosamax I”), the Third Circuit reversed the warnings part in a quite sloppy decision that was most notable for making preemption a jury question. See here for our detailed exposition on the decision and here for its place on our annual worst list.
In May 2019, the Supreme Court reversed the reversal and remanded the cases back to the MDL court. As only the second Supreme Court decision on preemption specifically in the branded drug context, Albrecht was definitely notable. For us, it had the unusual distinction of making both our best and worst lists for the year. The latter inclusion was largely because of Albrecht’s failure to undo the many mistakes in Levine, including the overemphasis on the actually limited availability of the Changes Being Effected (“CBE”) process for labeling changes that FDA can undo within weeks. Indeed, Albrecht’s discussion of the role of CBE gave something to each side without providing much actual guidance. 587 U.S. at 315 (“Of course, the FDA reviews CBE submissions and can reject label changes even after the manufacturer has made them. See §§314.70(c)(6), (7). And manufacturers cannot propose a change that is not based on reasonable evidence. §314.70(c)(6)(iii)(A). But in the interim, the CBE regulation permits changes, so a drug manufacturer will not ordinarily be able to show that there is an actual conflict between state and federal law such that it was impossible to comply with both.”).
Albrecht did undo, however, the Third Circuit’s determination that preemption is always a question of fact:
The complexity of the preceding discussion of the law helps to illustrate why we answer this question by concluding that the question is a legal one for the judge, not a jury. The question often involves the use of legal skills to determine whether agency disapproval fits facts that are not in dispute.
Id. at 316. It suggested that courts could “resolve subsidiary factual disputes that are part and parcel of the broader legal question” in deciding summary judgment. Id.at 317 (internal citation omitted). The standard to be applied could have been clearer had the Court jettisoned “clear evidence” completely:
We do not further define Wyeth’s use of the words “clear evidence” in terms of evidentiary standards, such as “preponderance of the evidence” or “clear and convincing evidence” and so forth, because . . . courts should treat the critical question not as a matter of fact for a jury but as a matter of law for the judge to decide. And where that is so, the judge must simply ask himself or herself whether the relevant federal and state laws “irreconcilably conflic[t].”
Id. at 315. Given this guidance and the history of the case, it seemed quite likely that the MDL court would again find preemption of the pre-September 14, 2010, warnings claims on remand. After another two years, that is what happened. Last week, however, a completely different panel of the Third Circuit reversed again, in a decision that that reads a bit like a judicial thumbing of the nose at the Supreme Court. In re Fosamax (Alendronate Sodium) Prods. Liab. Litig., No. 22-3412, – F.4th –, 2024 WL 4247311 (3d Cir. Sept. 20, 2024) (“Fosamax II”). Pending a reversal by the Third Circuit sitting en banc or by the Supreme Court, where does that leave the caselaw on preemption of warnings claims in cases like this?
Before we tackle that weighty question, please forgive us two detours. First, in these days of highly partisan politics, where many judicial decisions can be predicted by which presidents appointed the judges deciding the issue, preemption of product liability claims defies easy categorization. Four district judges have been in charge of the Fosamax MDL, two appointed by a president of each major party. The third, appointed by the second President Bush, granted summary judgment on both occasions discussed above. The first panel to reverse her had two circuit judges appointed by Democrats and one by a Republican. The second has two circuit judges appointed by Republicans and one by a Democrats. Albrecht was a unanimous decision, authored by a justice appointed by a Democrat on a Court with five justices appointed by Republicans. Even fierce advocates for states’ rights and limiting federal authority joined in a pro-preemption decision. Compare that to 2008 when three of the same nine had joined in a decidedly anti-preemption decision in Levine. (Two of the nine had been in the minority in Levine.) Not all issues break down along party or easy doctrinal lines. Resistance to the preemption of product liability claims sure does not.
Second, a variant of proximate cause is inherent in the conflict preemption focus on what FDA would do if presented with a proposed labeling change, as seen in Albrecht and many cases since Levine. If the defendant cannot independently take an action that will change the drug’s label in a way that will affect the prescription to a particular plaintiff claiming injury from the drug, then surely there can be no relationship between the defendant’s alleged failure to warn adequately and the plaintiff’s alleged injury. Proximate cause in the context of state failure to warn claims should narrow the focus in the case from minor and tangential issues to risk information that could actually change the decision to prescribe if set out in the label in a realistic way. If we assume that a physician’s prescription is based on the determination that it is in the plaintiff’s best interest to take a prescription drug for a diagnosed condition or need, the understanding of available options, and the choice of a drug she believes is safe and effective for that condition and appropriate for that patient, then there has to be a pretty high bar for which theoretical labeling changes are relevant to plaintiff’s state law warnings claim, having nothing to do with any consideration of preemption. To change the risk-benefit decision, there would need to be a material risk of a complication not in the label, a materially greater risk of an important complication already in the label, or something like a contraindication or precaution applicable to the plaintiff as a member of a subgroup of potential drug recipients (e.g., an extra risk in people with a certain blood test result). Again ignoring preemption, the information that arguably should have led to a labeling change would also have to have been available to the manufacturer before the plaintiff was prescribed the drug or before some critical duration threshold was crossed. That is because state laws on failure to warn always require proximate cause and never set technical requirements for how a drug should be labeled (divorced from claimed injury and proximate cause) or how FDA should be informed about risk information and possible attendant labeling changes. Remember, implied preemption is derived from the Supremacy Clause, so what state law requires independent of federal law (e.g., the FDCA) matters.
Fosamax II got it wrong again in part because it misunderstood this dynamic. In the second sentence of the decision, the panel stated that plaintiffs accused the defendant “of failing to comply with drug labeling requirements under state law.” 2024 WL 4247311, *1. Nope. There are no “drug labeling requirements under state law” and the decision did not cite any state laws at all. Although the term “state law” is used repeatedly, the only reference to specific states in Fosamax II comes from a list of amici. The actual allegation of the plaintiffs was that each was injured because his or her prescribing physician was not adequately warned about the risk of AFF when prescribing the drug to the plaintiff. The “because” part of the preceding statement brings in the impact on the prescriber’s decisions, as well as medical causation between the resultant use and the alleged injury. If state law liability is predicated on failing to take an action to deliver an adequate warning, then the proper conflict preemption inquiry is whether defendant could have taken an action that would have mattered to the plaintiff’s alleged injury while still complying with very real federal law requirements for drug labeling.
This was far from the only basic mistake in Fosamax II. For instance, it declares that Mensing was irrelevant to the issues on appeal because it involved a generic drug. Id. at n.34. Its resulting failure to consider the independence principle—something discussed in the Thomas concurrence in Albrecht that otherwise features prominently in Fosamax II—is pretty sloppy. Predictably, Fosamax II repeated the reliance in Fosamax I on the inapplicable presumption against preemption, even though there was no hint of such a presumption in Albrecht. Whereas the prior decision had cited only Levine on the purported presumption, which in turn relied solely on field preemption cases, Fosamax II relied exclusively on Bates v. Dow Agrosciences LLC, 544 U.S. 431, 449 (2005), a field preemption case that relied on Silkwood v. Kerr-McGee Corp., 464 U.S. 238, 251 (1984), another field preemption case. The Supreme Court’s decisions in Geier v. American Honda Motor Co., 529 U.S. 861, 872-873 (2000), which rejected a presumption against implied preemption and any “special burden” on the implied preemption defense, or Puerto Rico v. Franklin-California Tax-Free Trust, 579 U.S. 115 (2016), which rejected a presumption against express preemption. By this point, courts should understand that the presumption against preemption only applies to field preemption. We could go on with basic problems. Instead, we will note that the procedural history included the section heading “The Supreme Court Vacates our Fosamax I Decision.” 2024 WL 4247311, *11 (abnormal capitalization in original). Thus, Fosamax II reads more like an attempt to justify Fosamax I than to follow Albrecht in ruling on the current appeal.
Fosamax II first had to decide the standard of review to apply in evaluating the MDL court’s decision finding preemption again. It concluded that it would apply a “clearly erroneous” standard to factual findings and a de novo standard for legal issues. Id. at *16. As far as we can tell, however, Fosamax II did not always accord deference to the MDL court’s factual findings and, instead, substituted its own analysis of the facts on a critical issue. While Fosamax II did not disturb the findings that Merck did not withhold material information from FDA or provide misleading information, it disturbed the finding that FDA would not have accepted the change to the label that plaintiffs urged. Without doing our own de novo interpretation of the facts, which we have discussed in prior posts, it is clear the court overread the phrase from Albrecht about FDA not accepting “adding any and all warnings that would satisfy state law.” As we said above, the state law warning issue is never just a question of adding or changing a word in the label; the warning change has to convert the warning from inadequate to adequate in a way that would make a prescribing physician no longer prescribe the drug to her patient. In a situation where FDA rejects the proposed addition of language to the Precautions section, states information would be appropriate for the postmarketing experience portion of the Adverse Reactions section instead, and then requires much more substantial class labeling changes a month later, it is hard to imagine any interpretation that would support FDA previously accepting the sort of material labeling change that state law would have required. A misplaced presumption and imagined ambiguity in how FDA wrote its letter rejecting the proposed change should not change that.
Instead, Fosamax II interpreted the record alluded to above as supporting a mere possibility of a conflict rather than a mere possibility that some immaterial change to the label that would not have changed any physician’s prescribing decisions could have been accepted. As it explained in a meaty footnote,
As soon as one asks what the FDA would or would not do, one is confronted with figuring out just how much proof – regardless of whether a judge is making the assessment instead of a jury – is enough to persuade the decisionmaker of what that hypothetical future looks like. Thus, while the opinion in Albrecht declined to “further define Wyeth’s use of the words ‘clear evidence’ in terms of evidentiary standards, such as ‘preponderance of the evidence’ or ‘clear and convincing evidence’ and so forth,” id. at 315, it still asks courts to hold drug manufacturers to some standard of proof. It is not easy to get away from Wyeth’s statement, not disclaimed in Albrecht, that “clear evidence” is required. Wyeth, 555 U.S. at 571 (quoted in Albrecht, 587 U.S. at 313). As discussed, Albrecht defines “clear evidence” as “evidence that shows the court that the drug manufacturer fully informed the FDA of the justifications for the warning required by state law and that the FDA, in turn, informed the drug manufacturer that the FDA would not approve a change to the drug’s label to include that warning.” 587 U.S. at 303. That is the standard we are endeavoring to apply here.
Id. at n.28. Remember that, in Fosamax I, when preemption was supposed to be a question of fact, the same court had held “state-law failure-to-warn claim will only be preempted if a jury concludes it is highly probable that the FDA would not have approved a label change.” In addition, Fosamax II’s conclusion that Albrecht essentially blessed a “clear and convincing evidence” standard when it said that standard is irrelevant to the legal determination of “whether the relevant federal and state laws ‘irreconcilably conflic[t]’” is a stretch.
So is the court’s distillation of why it found the warnings claims not preempted:
To support the conclusion that there was pre-emption, the FDA, acting with the force of law, must have clearly rejected Merck’s label in a manner that made it evident that no label about atypical femoral fractures would have been appropriate at the time of Merck’s Prior Approval Supplement. That did not happen here. For that reason, Merck has not shown that the FDA would have rejected any and all labels that would have satisfied state law.
Id. at *26. In the more than one hundred collective years of experience of the Blog authors in defending product liability claims against prescription drug companies, we have never heard a plaintiff contend that a package insert would be adequate as long as it had some mention of the plaintiff’s alleged injury, no matter how placed or phrased. State law of the sort that Fosamax II does not discuss requires an adequate warning, not just any old mention. The inability for the drug manufacturer to rule out any possibility of any labeling change that mentions the injury at issue should not rule out conflict preemption. Indeed, at the end of the paragraph in Albrecht where Fosamax II got the “any and all warnings . . . that would satisfy state law” language—not “any and all labels” as misquoted above—the Supreme Court clarified that preemption of a warnings claim required a finding that “the FDA, in turn, informed the drug manufacturer that the FDA would not approve changing the drug’s label to include that warning,” specifically referring to “the warning required by state law.” 587 U.S. at 314 (emphasis added). As such, Fosamax II does not square with Albrecht, which is certainly controlling law.
With that, we return to our initial question. We expect that, until reversed by an en banc panel or the Supreme Court, in cases within the Third Circuit, winning warnings claims against the manufacturers of branded prescription drugs on conflict preemption will be harder. The regulatory record in most litigation will not be as favorable to conflict preemption as it is with Fosamax. However, it should be the case that, by wriggling around Albrecht, many plaintiffs will force themselves into having to prove proximate cause for failure to warn based on a proposed labeling change that prescribers will find inconsequential. Manufacturers prefer broad strokes of preemption, of course. Outside of the Third Circuit, it is hard to imagine that the ill-conceived Fosamax II will have much influence given that it basically tried to overrule the Supreme Court in Albrecht. Inherently anti-preemption judges will point to snippets of it to justify decisions not to preempt warnings claims they would have reached anyway. However, there are many decisions of circuit courts before and after Albrecht that find conflict preemption in branded prescription drug cases where the regulatory record is probably weaker than with Fosamax. So, much like a labeling change pursuant to a CBE, the effect of Fosamax II may be only temporary and limited in scope.