Supreme Court Preview: Will “Skinny Labels” Get a Weight Check?

by Dennis Crouch

The Supreme Court will soon be asked to weigh in on the the skinny-label debate — particularly the question of how much a generic drug manufacturers can say about their products without inducing patent infringement? The case is Hikma v. Amarin.

It is very common for a drug to follow the following innovation-patent pathway:

  1. First the composition is discovered and patented along with a particular therapeutic use.
  2. Later a more effective treatment regimen is discovered and patented.

Once the first set of patents expire, generic manufactures should be permitted to begin marketing the drug — except for uses still patented.  But, this gets tricky.  We know that there will be a very $trong incentive for insurance companies, doctors, and patients to use the cheaper generic drug for the still-patented treatment regime.  The question is whether the generic manufacture can be blamed for inducing this infringement.

Congress delt with this issue somewhat with its “skinny label” legislation codified in Section 505(j)(2)(A)(viii) of the Federal Food, Drug, and Cosmetic Act.  The law provides a pathway for generic drug approval when only some of the FDA-approved uses (indications) of a drug are patent-protected.  This section, often called the “section viii statement,” allows generic manufacturers to file what’s known as a “carve-out” – essentially telling the FDA they will not seek approval for the patented uses of the drug. This creates a “skinny” version of the drug’s label that omits the protected indications. As the Supreme Court explained in Caraco Pharmaceutical Laboratories v. Novo Nordisk, 566 U.S. 399 (2012), this provision helps “facilitate the approval of generic drugs as soon as patents allow.”

The process of creating a skinny label begins when a generic manufacturer files an Abbreviated New Drug Application (ANDA). While the ANDA must generally show that the generic’s proposed labeling is the same as the brand name drug’s labeling (21 U.S.C. § 355(j)(2)(A)(v)), section viii provides the exception – allowing the generic to file a statement that it is not seeking approval for the patented methods of use. The FDA then reviews this proposed carved-out labeling to ensure it still provides adequate information for safe use of the drug for the remaining, non-patented indications.  The resulting skinny label must comply with FDA labeling requirements while carefully avoiding any explicit or implicit suggestions about the patented uses.  However, everyone knows its the same drug.

The Hikma case follows the similar pattern. Vascepa (icosapent ethyl) was initially approved to treat severe hypertriglyceridemia (SH) – a use no longer patent-protected. Later, Amarin obtained approval and patents for using the drug to reduce cardiovascular (CV) risk. Hikma launched a generic version with a skinny label covering only the SH indication. However, Hikma’s press releases referred to their product as the “generic version of Vascepa” and cited Vascepa’s total sales figures – which were predominantly from the patented CV use.

The Federal Circuit, reversing a district court dismissal, held that Amarin had plausibly alleged induced infringement based on the totality of Hikma’s communications. The court emphasized several key factors:

  1. Though Hikma’s label alone does not induce infringement, it contained elements (like clinical study descriptions and the removal of certain warning language) that could suggest CV use when combined with other messaging.
  2. Hikma’s press releases called their product the “generic version” of Vascepa while noting Vascepa was indicated “in part” for SH treatment – potentially implying other uses.
  3. The cited sales figures predominantly reflected CV use revenues, which could signal to physicians the intended broader use.

In seeking Supreme Court review, Hikma is likely to present two fundamental questions:

  1. Active vs. Passive Inducement: Can induced infringement liability exist without any statement explicitly encouraging the patented use? Hikma argues the Federal Circuit’s ruling effectively creates “passive inducement” liability, contrary to the Patent Act’s express requirement for “active” inducement under 35 U.S.C. § 271(b).
  2. “Generic Version” Safe Harbor: Should referring to a product as a “generic version” of a branded drug and quoting its sales figures create infringement liability when the manufacturer has properly carved out patented uses from its label?

The case highlights several tensions in Federal Circuit precedent — particularly the Federal Circuit’s 2021 decision in GSK v. Teva. In that case, the Federal Circuit upheld a jury verdict finding that Teva induced infringement of GSK’s patent on using carvedilol to treat congestive heart failure, even though Teva had carved that use out of its label. The court focused on Teva’s marketing materials and press releases that referred to its product as a “generic carvedilol” and promoted it as “AB-rated,” finding this evidence could support the jury’s conclusion that Teva encouraged doctors to prescribe its generic for the patented use. In its opinion, the Federal Circuit distinguished GSK because Hikma’s press releases, unlike those in GSK, did not include AB-rating disclaimers limiting therapeutic equivalence to approved uses.

The case’s current status: Hikma’s petition for writ of certiorari is due January 15, 2025, but the company has requested a 30 day extension to February 14. The court tends to liberally grant these requests.  Charles Klein of Winston & Strawn is representing Hikma in the case.

 = = =

I mentioned Caraco v. Novo above.  I wanted to circle back to that case just for a moment.  The court’s unanimous 2012 decision focused on “patent use codes” that brand-name manufacturers submit to the FDA. The Court, in an opinion by Justice Kagan, held that generic manufacturers can use the Hatch-Waxman Act’s counterclaim provision to force corrections of use codes that inaccurately describe the scope of a patent. The case arose because Novo had submitted an overly broad use code suggesting its patent covered all three approved uses of its diabetes drug repaglinide, when in fact the patent only covered one use. This prevented Caraco from getting approval for a “skinny label” that would have carved out the patented use while marketing the drug for two unpatented uses.  The court based its reasoning on statutory interpretation of the FDA Act.  Justice Sotomayor wrote separately to emphasize that while helpful, the counterclaim provision was an imperfect solution that could still lead to delays in generic approval, suggesting that Congress or the FDA needed to do more to address overly broad use codes.