GlaxoSmithKline LLC v. Teva Pharmaceuticals USA, Inc. (Fed. Cir. 2021)
In 2020, the Federal Circuit issued an odd decision in this ANDA case. The patent on the drug at issue (carvedilol) has expired, but GSK holds a patent on using the drug for treatment of congestive heart failure. Teva began selling the drug for other approved uses, such as hypertension and ensured that congestive heart failure was not part of its product label. Of course, its generic product is still prescribed for that purpose. AND, in 2011 the FDA required Teva to list congestive heart failure as one of the drug treatments — since Teva’s approval was based on GSK’s original new drug application the FDA required a label that was identical-in-content. The result — $234 million in lost profit damages for inducing infringement. That damage award included pre-2011 acts even with the label carveout since Teva had (accurately) described its product as the generic equivalent of GSK’s product. On appeal, the Federal Circuit affirmed (split decision).
At that point Teva petitioned for rehearing, and the original merits panel of Chief Judge Prost and Judges Newman and Moore have agreed to rehear the appeal on the merits on the following question:
Whether there is substantial evidence to support the jury’s verdict of induced infringement during the time period from January 8, 2008 through April 30, 2011. We find all other issues to be sufficiently briefed.
Although there is no evidence that Teva really encouraged folks to buy and use its product for an infringing purpose, I can just imagine Teva’s confused CEO “where did this extra $200 million come from, I thought the non-infringing market wasn’t this big…”
The Court has not asked for new briefs and will be holding oral arguments later this month. The case has already generated substantial amici support.
Update – I thought my $200m comment above was funny. But, the damage award here is not actually excess Teva revenue. Rather, it is GSK’s lost profits. Judge Prost wrote in her opinion that the Teva’s total revenue from its sales was $75 million — and the jury was told that 17.1% of the sales were for infringing uses. So, we’re talking here about $13 million in revenue to Teva.
Some on this thread seem to be drawing a distinction between an “express carve out” of a protected indication and an “omission” of a protected indication from a drug label, and then querying which would have been a better tactic for Teva to use. As far as I am aware, there is no general mechanism by which an ANDA applicant can expressly include in it’s label that its generic product is not approved for a protected indication (or by which FDA can require the ANDA applicant to do so). The closest provision I am aware of is 21 U.S.C. 355 (o), which states:
(o) Prompt approval of drugs when pediatric information is added to labeling
(1) General rule
A drug for which an application has been submitted or approved under subsection (b)(2) or (j) of section 355 of this title shall not be considered ineligible for approval under that section or misbranded under section 352 of this title on the basis that the labeling of the drug omits a pediatric indication or any other aspect of labeling pertaining to pediatric use when the omitted indication or other aspect is protected by patent, or by exclusivity under clause (iii) or (iv) of section 355(j)(5)(F) of this title, clause (iii) or (iv) of section 355(c)(3)(E) of this title, or section 360cc(a) of this title, or by an extension of such exclusivity under this section or section 355f of this title.
(2) Labeling
Notwithstanding clauses (iii) and (iv) of section 355(j)(5)(F) of this title, clauses (iii) and (iv) of section 355(c)(3)(E) of this title, or section 360cc of this title, the Secretary may require that the labeling of a drug approved pursuant to an application submitted under subsection (b)(2) or (j) of section 355 of this title that omits a pediatric indication or other aspect of labeling as described in paragraph (1) include—
(A) a statement that, because of marketing exclusivity for a manufacturer—
(i) the drug is not labeled for pediatric use; or
(ii) in the case of a drug for which there is an additional pediatric use not referred to in paragraph (1), the drug is not labeled for the pediatric use under paragraph (1) . . . .
Thus, FDA can require an ANDA or 505(b)(2) applicant to include an explicit “not labeled” statement in its label as a condition for approval of a product that omits a protected pediatric indication only. An example of such a statement is found in generic aripiprazole labels, which states “Additional pediatric use information is approved for Otsuka America Pharmaceutical, Inc.’s ABILIFY® (aripiprazole) product. However, due to Otsuka America Pharmaceutical, Inc.’s marketing exclusivity rights, this drug product is not labeled with that information.” Note that the statement does not explicitly indicate what the pediatric information is or what it covers (in the case of ABILIFY® it was Tourette’s syndrome, but a doctor would have do a bit of digging to find that out).
Just browsed the Fed. Cir. decision:
In 2011 the FDA required Teva to amend its carvedilol
label to be “identical in content to the approved [GSK
Coreg®] labeling (including the package insert and any
patient package insert and/or Medication Guide that may
be required).” Dist. Ct. Op. at 587. Teva amended its label
to include the indication for treatment of heart failure, as
required by the FDA. Dist. Ct. Op. at 587.
So no carve out…..
My understanding is that there was no carveout after 2011. The rehearing here is focused on the time period where the label included an express carveout.
Thanks for clarity. I would be nice to understand the timeline in the article. Teva’s ANDA filed with carve out. Teva launched at risk, correct? No injection during H-W trial? GSK wins at trail (including a win on inducement and seeks damages for lost profits? Teva argues the lost profits are too big because some of the sales were for a carved out indication? I am sorry I don’t follow, and don’t have the time to dig into the actual district court cases.
For clarity, Teva’s labeling never included an express carveout, such as “this product is not approved for the treatment of CHF.”
Better to say that before 2011 Teva’s labeling omitted CHF, and after 2011 it expressly included CHF.
Funny thing is: in district court Teva persuasively argued that, as a matter of fact, nobody actually reads their label. This allowed Teva to escape all liability even for the full label period. Because, you know, regardless of what the generic label says – if doctors don’t read it, then how can it “cause” them to infringe?
It doesn’t seem the Federal Circuit knew what to do with that argument, but it sure carried the day in district court.
Taken to its next logical step, that argument not only opens the lid, but rips it off and throws it far, far away from Pandora’s box of straight liability. Which doctor is going to step up and aver to any proscribing in the blind?
It’s half and half. For the first 4 years Teva sold its carvedilol with a skinny label. Then Teva broadened its label, and for the next 4 years sold its product with instructions for the infringing use.
To your point above though, I am curious if it IS a ‘legal equivalent’ between a Skinny Label and a label WITH an express carve-out.
Plus an adjunct and corollary view to your “don’t read anyway,” IF the fact situation were that an express carve-out were present, would your “don’t read anyway” serve to FLIP the result and reduce the defense of the express carve-out? (a goose and gander proposition)
Key to this decision is the facts. Upon approval, as Dennis’s article mentions, Teva’s label did indeed include the supposedly carved-out indication.: “FDA required Teva to list congestive heart failure as one of the drug treatments.” The result is that no indication was actually carved out! Teva tried to carve out the indication upon filing their ANDA, but the FDA required them to include this indication. So – no actual carve out!! I can’t stress this enough. It has been glossed over on other blog sites as well. People are describing this as a carve out label case when no indication was actually carved out!
For those that may be new to the “induced infringement” discussion, perhaps a link to the pertinent section of patent law (and a brief ‘state of the law’ in regards to that section) might be helpful.
(just a thought)
link to bfy.tw
vn, +1
Comments are closed.