The other day we found ourselves in the unusual position of being sent copies of the same document by both sides of the litigation.  Both Linda Maichl, at Ulmer & Berne, who represents generic drug manufacturers, and Justinian Lane, who hosts a plaintiff-side blog, sent us copies of the Solicitor General’s brief for the United States as amicus curiae in the Supreme Court Pliva/Mensing case.  Not only that, the guys over at the FDA Law Blog sent us a link to their post.  That brief was filed the other day in connection with the defendant generic drug manufacturers’ petition for certiorari.  The issue, of course, is preemption – specifically implied preemption as applied to generic drugs.
We’ve blogged about Mensing before. We covered the Supreme Court’s request that the SG’s office present the government’s views.  We opined on the the Eighth Circuit’s opinion, Mensing v. Wyeth, Inc., 588 F.3d 603 (8th Cir. 2009).  That opinion had the unusual distinction of simultaneously making both of our top-ten lists for 2009 – worst and best.  Okay, so it was honorable mention for the latter, but you get it – we’ve been following this case.
Maybe that’s why they both decided to send it to us.  Or maybe it’s that we represent pioneer drug manufacturers (also called “name-brand” or “listed” manufacturers).  Since we do that, we normally can’t represent the generics.  Generic and pioneer/name brand/listed manufacturers sue the bejeezus out of each other constantly on any number of non-products issues.  So to the extent we have a dog in any generic preemption, hunt it’s on a long, long leash.
In any event, we thank both of them for thinking of us.
Before we look at the substance of the SG’s Mensing brief, we want to set out, so there’s no misunderstanding, where we stand on the issue of generic preemption.  We thought it was a pretty good argument before Wyeth v. Levine, 129 S. Ct. 1187 (2009), and we said so more than once.
Our main problem with generic preemption isn’t legal, but practical.  That is, if generic preemption became the rule, then pro-plaintiff courts would be tempted to stretch the substantive law in ways that might give new remedies to plaintiffs against our, non-manufacturer clients in cases where the plaintiff only took a generic product.  Don’t laugh, and don’t call us paranoid.  That’s exactly what in fact happened in the execrable (and fortunately not widely followed) decision in Conte v. Wyeth, Inc., 85 Cal. Rptr.3d 299 (Cal. App. 2008).  The court accepted preemption, and promptly engineered some new, bizarre version of “fraud” to hold a name brand manufacturer liable to a plaintiff that never used its product (which had been driven off the market by generic competition.  We’re sufficiently ticked off about Conte-type cases that we have a scorecard devoted solely to this subject.
After Levine, we can’t say that we’re surprised that generic preemption has been on a losing streak.  We’ve thought that the generic argument shares the same Achilles heel with Levine – that on a crucial point it relies upon a Bush-era FDA Federal Register notice.  But again, maybe it’s just that misery loves company.  Heck, if we’re losing cases like Wimbush v. Wyeth, 619 F.3d 632 (6th Cir. 2010); Mason v. SmithKline Beecham Corp., 596 F.3d 387 (7th Cir. 2010), and Baumgardner v. Wyeth Pharmaceuticals, 2010 WL 3431671 (E.D. Pa. Aug. 31, 2010), then the generics should be taking their lumps in preemption cases, too.
And, frankly, they have been.
Bottom line?  Ideologically, we’d like just about all product liability preemption arguments to win.  Preemption is the strongest defense there is, since the merits of the underlying claim are irrelevant. When it wins, preemption can shut down an entire mass tort (see Sprint Fidelis) with one fell swoop.  But we’d like our clients’ preemption arguments to win most of all.  If they’re not, then we don’t want our clients left holding the liability bag when they didn’t sell the product and thus didn’t make any profit from it.
Back to Mensing. We frankly weren’t expecting anything very good out of the current SG’s office on prescription drug preemption. After all, this is the same administration that unleashed an administrative search and destroy mission against preemption after only a few months in office.  We were lucky enough to get a pro-preemption brief in Bruesewitz – and that involved vaccines, the ultimate “public good” product.
So our expectations were so low that we were pleasantly by what we found in the Mensing SG amicus brief (hereafter simply “Mensing br.”). The SG found a way to come out, bottom line, against review and therefore in support of the lower court’s decision rejecting preemption, but at least there are a few worthwhile tidbits for the defense side to chew on.  The SG even stated that “federal law may circumscribe” a plaintiff’s liability theories, even if it does “not outright bar” them.  Mensing br. at 10.  That alone was interesting enough for us to read further.
For one thing, the plaintiffs like to portray the FDA as completely dependent on the manufacturers.  The Mensing brief stated, however:

In reviewing an NDA, FDA considers evidence submitted by the applicant, and other relevant scientific information, to determine whether the proposed labeling is accurate, truthful, not misleading, and adequate. Thus, FDA’s approval of an NDA includes approval of the proposed drug labeling.

Mensing br. at 4. A Bush SG could have written that.
The generics got a stronger confirmation that that their labeling is supposed to be the “same” as the branded labeling than we would have expected.  Mensing br. at 5.  Conversely, the SG stated that, as to new information, that generics have the same responsibility to keep current as a brand name manufacturer:

NDA and ANDA holders must keep records of clinical experiences and ensure that their products remain safe and effective as labeled. In particular, implementing regulations provide that a manufacturer must record and report certain adverse events to FDA.

Mensing br. at 5-6.
Representing the brand name side, we’re comfortable with this FDA reporting obligation.  Our clients do it all the time.  It may be a problem, however, for the generic defendant in Bartlett, since there was some question in that litigation whether it was paying enough attention.
The SG argued that there was no “categorical” preemption of warning claims based upon this duty to keep current with evolving science, and also because the a generic manufacturer “could have asked FDA to coordinate appropriate [Dear Doctor] letters (or, by extension, to take other action with respect to labeling).”  Mensing br. at 11.
But that’s as far as things went in the plaintiff’s direction.  On the defense theory of “impossibility” preemption, the SG actually agreed with the defendant that generic manufacturers cannot modify their labeling without FDA pre-approval through the changes being effected (“CBE”) process or through the prior approval supplement (“PAS”).  Id. (“the district court correctly concluded that the CBE process was unavailable to petitioners”).  According to the government:

The holder of an approved ANDA is not free to change its approved labeling at will.  At the time of the events in this case:  (1) an ANDA holder in petitioners’ position could not unilaterally change its approved labeling under the CBE process; (2) the PAS process was not expressly available to any manufacturer to change approved labeling to add or strengthen a warning; (3) ANDA holders were nonetheless required to provide FDA with new information about risks, and FDA would have acted on such information if appropriate; and (4) an ANDA holder unilaterally sending [dear doctor] letters of the kind respondent seems to envision could have resulted in misbranding the drug.

Mensing br. at 12-13 (regulatory citations omitted) (emphasis added).  Unlike the Supreme Court in Levine, the SG stuck with the Bush-era FDA in concluding that unilateral warning “strengthening” isn’t available in the generic context:

In light of the substantive limitations on generic labeling, FDA has consistently taken the position that an [generic manufacturer] may not unilaterally change its approved labeling. In promulgating its final rule implementing labeling requirements for [generics], FDA . . . explain[ed] that the [generic] product’s labeling must be the same as the listed drug product’s labeling because the listed drug product is the basis for [abbreviated] approval. FDA stated that an [generic manufacturer] wishing to add a warning to approved labeling should furnish adequate supporting information to FDA, which would then determine whether the labeling for all products should be modified.

Id. at 14 (regulatory citations and jargon omitted).  That’s more than the generics have been winning in court lately.  A similar discussion confirms that PAS, as well, isn’t intended for unilateral changes to generic labels.  Id. at 14-15.
The SG agreed with the defendants that the Eighth Circuit in Mensing got it wrong when it held that manufacturers could do essentially whatever they pleased through Dear Doctor letters.  Not so, said the SG, since Dear Doctor letters are “labeling”:

[Generic manufacturers] do not customarily send [Dear Doctor] letters without coordinating with FDA. . . . [Generic manufacturers] holders also operate under a regulatory constraint. . . . [B]ecause [Dear Doctor] letters are “labeling” . . . FDA may withdraw approval of an [generic drug] if the labeling of the drug . . . is misleading in any particular. Depending on its content, a [Dear Doctor] letter from an [generic manufacturer] could inaccurately imply therapeutic differences . . . and therefore be misleading.

Mensing br. at 17 (citations, quotation marks, and regulatory jargon omitted).
Thus a claim for failing to send a Dear Doctor letter could be preempted.  “State law may not impose liability on an [generic manufacturer] for failing to send such a letter unilaterally.”  Id. at 18.  “[A]bsent FDA’s assent, petitioners could not lawfully have disseminated their product with the sort of warning respondent seems to propose.”  Id. at 19.  In other circumstances, these quotes might come in handy.
Frankly, that’s quite a bit more than we thought we would get out of the government on preemption.  So the plaintiffs are essentially left with what’s behind door three – submit updated risk information to the FDA and see what the agency does.

[T]he court of appeals appeared to anticipate that the parties could litigate on remand the question of what action FDA would have taken in response to a hypothetical warning proposal from petitioners. The court concluded that petitioners would bear the burden of showing the likelihood of FDA inaction. That allocation – which ultimately turns on litigation considerations, not an interpretation of the FDCA or FDA’s regulations – is reasonable.

Mensing br. at 15-16.  In future cases, if the FDA doesn’t act, a defendant might, under the government’s view, actually win a preemption motion, depending of course, on the facts.  Thus while the government has kept the defendant out of the end zone, it didn’t try to remove the goal posts altogether.
Then there’s this footnote, which appears applicable to FDA activity generally, and not just to generic drugs:

A fully informed, actual decision by FDA that a particular warning would be inconsistent with the FDCA or FDA’s regulations would presumably preempt a state law claim predicated on the necessity of such a warning.

Mensing br. at 19 n.9.  We’ve said before that when reading government briefs, it’s a very good idea to read the footnotes.  They’re often more interesting than the text.
The generics did get hammered, however, on their purposes and objectives preemption argument that there’s a conflict between tort liability and congressional intent that generic drugs be “low-cost.”  “That argument is wrong. The Hatch-Waxman Amendments do not pursue the objective of low-cost generic drugs without limitation.”  Mensing br. at 20.  Ouch.  That’s a pretty categorical rejection.  Under the defendant’s argument:

Individuals harmed by inadequately labeled generic drugs would have no remedy, while individuals who took the same drug with the same labeling in its brand-name form would have a state tort remedy.

Id. at 21. We have to admit that, stated that way, this argument does sound a little weird.
Finally, there’s something else we like – because it might help our clients, too – because the government believes that failure-to-remove-from-the-market claims present entirely different preemption considerations:

[I]mposing on a generic manufacturer a state law duty not to market its product without developing for itself knowledge as comprehensive as FDA’s or the NDA holder’s could pose preemption questions different from the ones respondent’s complaint raises.

Id. at 22. We’ve already commented that plaintiffs played bait and switch with the Supreme Court in Wyeth v. Levine, 129 S. Ct. 1187 (2009), denying that they were attacking FDA-approved uses before the Court (the so-called “failure to contraindicate” claim) but then turning around and arguing (successfully) that Levine defeats preemption of such claims.  The FDA isn’t likely to look kindly on tort claims that attack its approval decisions, and seek to impose liability because a defendant shouldn’t have marketed an approved drug at all.  This Parthian shot by the government suggests that, if Wimbush were appealed and Supreme Court asked the government for its views, we might just have a chance.