Photo of Michelle Yeary

We all know hindsight is 20/20.  And, it’s easy.  There are dozens of television and radio programs that thrive on Monday morning quarterbacking.  There’s no risk in saying the coach should have called for a pass when you already know the run didn’t work.  It’s also dangerous because it’s easy.  People are often too quick to point out that you should have taken path B after everyone learns path A is full of potholes.  Pointing it out is one thing, holding you liable for it is another.

What does hindsight mean in the products liability arena? Unfortunately, it can be used to demand perfection.  It can be used to allow plaintiffs to proceed on what is essentially a stop-selling theory based on the fact that in hindsight later approved treatments were safer.  That’s what happened in Holley v. Gilead Sciences, Inc., 2019 WL 2077845 (N.D. Cal. May 10, 2019).

Gilead is the manufacturer of several drugs used to treat and/or prevent contraction of the AIDS virus.  Several contain as their active ingredient tenofovir disoproxil fumarate (“TDF”).  The TDF drugs were approved by the FDA between 2001 and 2012.  Certain other of Gilead’s AIDS medications contain tenofovir alafenamide fumarate (“TAF”) rather than TDF.  The TAF drugs were approved by the FDA in 2015 and 2016.  Plaintiffs allege that defendant failed to provide adequate warnings with the TDF drugs regarding the increased risk of kidney and bone damage.  Plaintiffs also claim that defendant should be liable for “its decision to develop drugs containing TDF rather than the safer compound [TAF].”  Id. at *1.  At the heart of this claim is an allegation that the defendant was aware of TAF and that TAF was safer before it started selling TDF.  Id. at *2.

Defendant moved to dismiss all counts on federal preemption grounds.  Plaintiffs’ primary design defect claim was that there was a safer alternative available at the time defendant sought FDA-approval for the TDF drugs.  Id. at *7.  So, plaintiffs’ argument is that defendant never should have brought TDF to market.  But never-start selling claims based on allegations that the defendant should never have submitted the product to the FDA should be preempted as disguised “stop-selling” claims.  See Yates v. Ortho-McNeil Pharmaceuticals, Inc., 808 F.3d 281, 300 (6th Cir. 2015); Utts v. Bristol-Myers Squibb Co., 226 F. Supp. 3d 166 (S.D.N.Y.  2016); and Brazil v. Janssen Research & Development LLC, 196 F. Supp.3d 1351, 1364 (N.D. Ga. 2016).  But the Holley court did not see it that way.  Where Holley misses the point, however, is its focus on some hypothetical pharmaceutical product rather than the one at issue in the case.  To even state a design defect claim, plaintiffs have to identify a defect in the product they used – the one that was sold and marketed.  The court is correct that there is no federal law that prevents a manufacturer from developing a drug different from the one at issue.  Id. at *9.  That would be silly.  The fact that it did so at a later date, however, can’t be used to second guess the FDA’s approval of the drug at issue.

The danger of drawing an artificial distinction between pre and post approval is even more clearly seen in plaintiff’s argument that one of its TDF products should have contained a lower dose of TDF.  That argument could be made of almost any drug.  And post-approval, the design claim is preempted by Mutual Pharm. Co. v. Bartlett, 133 S. Ct. 2466 (2013).  Pre-approval, it’s not?  We’ve argued time and again that a different product isn’t a safer alternative design.  Neither is it by-pass around preemption.

Holley doesn’t stop there, unfortunately.  To the contrary, it extends “pre-approval” claims to warnings as well as design.  To do that, it first had to discount the body of law that has interpreted Wyeth v. Levine as preempting failure to warn claims based on “information known to the FDA.” Id. at *11 (citations omitted).  That is why claims regarding the adequacy of a drug’s label at the time of approval are preempted.  That label was reviewed and approved by the FDA and the jury can’t be used to second guess that approval.  What a manufacturer can/should do with its label post-approval based on information not presented to the FDA is a different question.  So what is a pre-approval failure to warn claim anyway?  An allegation that the manufacturer should have submitted a different warning or different information to the FDA for approval?  That sounds an awful lot like fraud-on-the-FDA, which is Buckman preemption.  So, the fiction of calling plaintiffs’ claim a pre-approval failure to warn claim should not have saved it.

After allowing the pre-approval claims, the court did have to find that at least some of plaintiff’s post-approval warning claims were preempted due to lack of newly acquired evidence.  Id. at *11-12.  The “newly acquired information” requirement for CBEs wasn’t added until 2008.  So, for the older drugs, defendant could have changed the warning regardless of whether it had such new information; those claims aren’t preempted.  Id. at *12. Post-2008, plaintiffs point to different types of post-marketing information regarding the risks of TDF drugs.  But, they failed to plead whether and to what extent that information was provided to the FDA.  Therefore, the court could not determine what was “newly acquired evidence” that would support a CBE; those claims were preempted.  Id.  But dismissal is without prejudice as the court found there was a likelihood the deficiency could be cured.

Moving outside of preemption, defendant did make a few TwIqbal arguments.  First, warning causation.  Here the court said plaintiffs met the pleading standard by alleging that physicians would have “acted differently.”  Id. at *13.  So, an allegation that the prescriber would have monitored plaintiff more closely sufficed, as opposed to the prescriber would not have used the drug.  Second, consumer fraud.  Apparently, in this part of California, pleading of some consumer fraud claims doesn’t have to meet Rule 9(b).  Where plaintiff’s consumer fraud claim alleges “some fraud” and “some non-fraudulent conduct” – only the fraud is subject to Rule 9(b).  So, does that mean you can dismiss half a claim under a TwIqbal?  Here, the portion of the claim based on allegations that defendant acted “unfairly” instead of fraudulently weren’t subject to heightened scrutiny.  The court also didn’t dismiss plaintiffs’ omission based claims, but did dismiss without prejudice their misrepresentation claims.

A hindsight-oriented opinion on hindsight claims alleging that first-generation drugs should not have even been submitted to the FDA because later treatments were safer.  The logic is flawed because it’s based on Monday morning quarterbacking which should be left to sports commentators and ticked-off fans; not to judges and juries.