Anyone who has checked our post-Levine innovator drug & vaccine cheat sheet lately has no doubt noticed our two most recent entries, Gentile v. Biogen Idec, Inc., 2016 WL 4128159 (Mass. Super. July 25, 2016), and Christison v. Biogen Idec Inc., No. 2:11-CV-01140-DN-DBP, slip op. (D. Utah Aug. 5, 2016). With respect to preemption and innovator drug warnings, these cases provide further support to an emerging, common-sense bright line in the otherwise all-too-murky world of “clear evidence” – that a warning change rejected by the FDA for lack of scientific evidence must be “clear evidence” that this change would have also been rejected at any earlier date. The logic is inescapable that, if there was insufficient scientific evidence at moment X, there is no more, and usually less, evidence on the same issue at any time before X.
Looking at our cheat sheet, the first case to so hold appears to be In re Fosamax (Alendronate Sodium) Products Liability Litigation, 951 F. Supp.2d 695 (D.N.J. 2013). Fosamax involved the FDA’s partial rejection of a prior approval supplement after the date of the plaintiff’s injury. Id. at 703 (FDA rejection occurred “approximately one month after” plaintiff’s injury). The label change failed because “the data that FDA has reviewed have not shown a clear connection” between the drug and the risk at issue. Id. at 699.
[C]lear evidence exists that the FDA would not have approved a label change to the Precautions section of the [drug] label prior to [plaintiff’s] fracture because Defendant submitted a label change and the FDA rejected it, and the FDA never required Defendant to submit new language or change the label, which demonstrates that the FDA did not think that the label should have been changed at that time.
Id. at 703-04. See In re Fosamax Alendronate Sodium Products Liability Litigation, 2014 WL 1266994, at *11 (D.N.J. March 26, 2014) (applying this ruling “to those Plaintiffs’ whose injuries occurred prior to [the FDA rejection date], without allowing additional discovery”).
Similar decisions – based on FDA insufficient-evidence rejections of label changes that occurred after the plaintiff’s prescription/treatment/injury – have also occurred in two Depakote cases: In re Depakote, 87 F. Supp.3d 916 (S.D. Ill. 2015), and Rheinfrank v. Abbott Laboratories, Inc., 119 F. Supp.3d 749 (S.D. Ohio), reconsideration denied, 137 F. Supp.3d 1035 (S.D. Ohio 2015).
In light of the fact that the FDA rejected the [particular risk] warning in 2006 because it did not find that the available scientific evidence at that time supported the addition of such a warning, it is highly unlikely that the available scientific evidence, seven years prior to that date in 1999, would have supported the addition of such a warning.
Depakote, 87 F. Supp.3d at 922 (emphasis original).
Preemption is warranted because there is clear evidence the FDA would not have approved a change to the [drug] label adding a [particular risk] warning prior to [plaintiff’s] injury. The Court finds the FDA’s February 2006 decision that [these] warnings “should not be incorporated into [the drug’s] labeling” and the FDA’s 2008 belief that “the data do not provide sufficient evidence to support  labeling changes at this time” constitute “clear evidence” that when confronted by the issue in 2003, the FDA would have rejected an attempt to add a [the same risk] warning.
Rheinfrank, 119 F. Supp.3d at 766. Cf. In re Incretin-Based Therapies Products Liability Litigation, 142 F. Supp. 3d 1108, 1124-27 (S.D. Cal. 2015) (multiple, recent FDA insufficient evidence conclusions supported “clear evidence” preemption regardless of individual plaintiff prescription or injury dates); In re Byetta Cases, 2015 WL 7184655, at *13-14 (Cal. Super. Nov. 13, 2015) (same).
Several months ago, we discussed another case drawing the same line, Cerveny v. Aventis, Inc., ___ F. Supp.3d ___, 2016 WL 1065826 (D. Utah March 16, 2016). In Cerveny, two FDA citizen’s petition denials for “fail[ure] to provide reasonable evidence to demonstrate” a causal connection to the alleged risk doomed product liability claims based on a prescription that was made prior to those denials:
[W]here a citizen petition post-dates the alleged injury and addresses the failure-to-warn claim proffered by a plaintiff, a citizen petition denial can be evidence to support a manufacturer’s preemption defense. . . . In this case, the FDA heard and rejected the Plaintiffs’ theory embodied in [the] citizen petitions, which post-dates [plaintiff’s drug] use. . . . If the FDA concluded in 2009 and 2012 that [the drug] . . . does not pose a risk of [what plaintiff suffered], the Court cannot say the FDA would have approved a contrary warning prior to 1992. . . . The FDA’s denial of the Plaintiffs’ theories embodied in [the] citizen petitions is clear evidence that the FDA would not have permitted [defendant] to strengthen [the drug’s] label prior to 1992.
Id. at *9-11 (numerous citations omitted)
The Gentile and Christison Tysabri cases follow this line of precedent – and more. In these cases, the link between the drug in question and the rare, fatal condition that the plaintiffs alleged was undisputed – and quite graphically warned about (more on that after we finish with preemption). However, the drug was very useful against other, more common, and likewise potentially fatal conditions, and at the time(s) in question, there was essentially nothing predictive of who would suffer the rare risk. Gentile, 2016 WL 4128159, at *1-3; Christison, slip op. at 6, 9-15. In both cases, the plaintiffs focused on the potential value of certain antibody tests, but both of them used the drug prior to FDA approval of those tests. In 2006 FDA determined that “no consensus” existed on whether those tests were “clinically relevant.” Gentile, 2016 WL 4128159, at *2-3; Christison, slip op. at 15. In 2010, the FDA reached the same conclusion – “that there is currently sufficient information to support the clinical utility of the . . . antibody assay in determining the risk.” Gentile, 2016 WL 4128159, at *3; Christison, slip op. at 47. Ultimately, after additional study and submissions, the FDA approved an antibody-based test for risk susceptibility in 2012. Gentile, 2016 WL 4128159, at *3; Christison, slip op. at 15-16.
In both Gentile and Christison, the relevant use of the drug occurred prior to one or both of the FDA’s 2006 and 2010 insufficient evidence-based refusals to approve the antibody tests. Gentile, 2016 WL 4128159, at *2-3 (used drug between January, 2007 and September, 2009); Christison, slip op. at 18-20 (used drug between February, 2007 and June, 2009). Given that, even after the last prescriptions, the FDA had considered and refused to approve the tests at issue due to insufficient scientific support, both cases concluded that “clear evidence” existed to support preemption of warning claims asserting that the tests should have been recommended in the drug’s labeling.
In the present case, the FDA’s decision to reject the proposed modifications regarding risks associated with providing [the drug] to patients with the  antibody demonstrates that prior to [plaintiff’s decedent’s] diagnosis, defendants could not have strengthened their warnings to comply with state law without violating the FDA’s decision. . . . Here, FDA rejected defendants’ proposed change in the warning not because of the language used, but because the supporting data was not yet sufficiently persuasive. For this reason, summary judgment is further appropriate, at least in part, because plaintiff’s claims, insofar as they are based on defendants’ allegedly inadequate warnings about the risks of administering [the drug] to patients who are found to have [the] antibodies, are preempted by federal law.
Gentile, 2016 WL 4128159, at *8-9 (citation omitted).
The Companies are correct that if [the data] were insufficient to persuade FDA in 2010 of the predictive value of antibody testing . . ., there clearly would not have been sufficient data more than two years earlier. Accordingly, there is “clear evidence” that the FDA would not have approved a change to the [drug] label prior to [plaintiff’s decedent’s] death, and the causes of action that [plaintiff] raises in this matter − negligence; negligent failure to warn; and negligent misrepresentation − which are based in state law, are preempted by federal requirements mandating FDA approval of changes to a drug’s warning label.
Christison, slip op. at 48 (citations, quotation marks, and footnote omitted).
One of the biggest problems with the Levine “clear evidence” test is that it is so bound up in the facts of any individual case that precedent has been of limited value, and courts have reached inconsistent results on the same facts. Gentile and Christison demonstrate that, at least on one front, it is now possible to draw a bright line in post-Levine “clear evidence” preemption cases – If the FDA finds insufficient evidence to support a label change on a given date, then logically any warning claim based on prescriptions that occurred before that FDA action will be preempted, because even less scientific evidence than what the FDA found wanting will necessarily exist during an earlier time period. As we’ve just discussed, this principle can be applied to a variety of branded (and, indeed, OTC) drugs.
But there’s more – considerably more. The warnings and special prescription protocol that surround the use of Tysabri are detailed and extensive (far longer than we would quote here), as befits the magnitude of the risk (an almost always fatal condition) involved. They are as impressive as the warnings and related special precautions that attend use of Accutane. Accutane cases have produced quite a few favorable decisions concerning the adequacy of challenged warnings. See, e.g., Felix v. Hoffmann-LaRoche, Inc., 540 So.2d 102, 105 (Fla. 1989); Banner v. Hoffmann-La Roche Inc., 891 A.2d 1229, 1239 (N.J. Super. App. Div. 2006); In re Accutane Products Liability Litigation, 2012 WL 3194954, at *5 (M.D. Fla. July 24, 2012); Gerber v. Hoffmann-La Roche Inc., 392 F. Supp.2d 907, 915-18 (S.D. Tex. 2005). The same thing occurred here. The courts in both Gentile and Christison held that the warnings at issue were adequate as a matter of law:
Because [the condition] may lead to death or severe disability, the warnings should have been commensurate with a high level of risk. The black box warning in effect when [the prescriber] first prescribed [the drug] to [the decedent] explicitly warned against the precise risk . . . that [the decedent] ultimately suffered, and fully disclosed the serious consequences of that disease. . . . [W]hen read as a whole, the warnings unmistakably conveyed the seriousness of [the risk] and its association with [the drug’s] treatment.
Gentile, 2016 WL 4128159, at *6 (applying New York law) 2016 WL 4128159 (citations omitted).
The Companies are correct that the labeling is adequate as a matter of law. . . . the Companies provided an adequate warning because the [the drug’s] warning (1) was designed so it could reasonably be expected to catch the attention of the consumer; (2) was comprehensible and gave a fair indication of the specific risks involved with [the drug]; and (3) was of an intensity justified by the magnitude of the risk.
Christison, slip op. at 40-41.
And more: The plaintiff’s expert testimony was deficient with respect to the adequacy of the warnings. In Gentile, because the learned intermediary rule applied, the plaintiff needed a physician to testify to the inadequacy of warnings to the medical profession. 2016 WL 4128159, at *7 (plaintiff’s expert “is not a medical doctor, does not have the qualifications necessary to opine on matters of treating patients, and does not assert that he has the experience necessary to interpret warnings from the perspective of a prescribing physician”). In Christison, the plaintiff failed to offer any warning expert at all.
[Plaintiff] is incorrect. A lay jury is not able to call on “their own life experiences” to determine the adequacy of a drug label. Where the injury involves obscure medical factors which are beyond an ordinary lay person’s knowledge, necessitating speculation in making a finding, there must be expert testimony that the defendant’s conduct probably caused the injury. . . . A lay jury would not be able to determine by calling upon “their own life experiences” and would be speculating in making findings about the adequacy of the [drug’s] warning label, which warned extensively about [the risk].
Slip op. at 36-37 (footnote omitted). See Id. at 37 (plaintiff must have expert testimony “on what ‘a more adequate label’ would contain”) (footnote omitted).
And in Gentile still more – enough that, in our opinion, Gentile is the best trial court opinion ever to emerge from a Massachusetts trial court in a drug or medical device case. Specifically:
- Warning causation: The plaintiff’s attempted nit-picking of the warnings in Gentile also failed because the prescribing surgeon’s decision “would not have changed” with the additional information. 2016 WL 4128159, at *7.
- Good Samaritan liability: “[P]laintiff cannot prove defendants voluntarily assumed a duty to warn [the decedent] by providing her with information directly, and not through her doctor.” Id. at *8.
- Design defect/comment k: “New York does not recognize a design defect theory of liability for prescription medicines” because a drug with adequate warnings is “not defective.” Id.
- Mensing-style preemption for innovator drugs: All warning claims against the distributor defendant were preempted because it was impossible for the distributor to alter drug warnings because it was not the holder of the FDA new drug application. Id. at *9 (another company “is the holder of the original, approved application for [the drug]. Consequently, [the distributor] could not have sought modifications of the label”).
Innovator drug preemption (2 kinds), warning adequacy as a matter of law, learned intermediary rule-based expert requirements, warning causation, design defect, and assumed duty – that’s a bunch of significant rulings. We’ll certainly be watching out for similar cases coming down the pike in the future, but given the thankfully rare nature of the risk involved, there probably won’t be that many.