Origins of the Theory that the “Patent Dance” Might Not Be Required: Omission of Process Patents

I recently wrote a preliminary reaction to the Supreme Court’s Sandoz v. Amgen decision on Health Affairs Blog.  This was the Court’s first foray into the Biologics Price Competition and Innovation Act (BPCIA), which created a pathway for licensure of biosimilar biologics.  As my essay for Health Affairs notes, the issues presented by this litigation are not entirely resolved.  There is fodder for a great deal of further discussion.

Today I am starting with the Court’s ruling that Amgen was not entitled to a federal injunction ordering Sandoz to share its marketing application and manufacturing information with Amgen.  The topic here is the origin of the argument that Sandoz made to the Court — the omission of manufacturing process patents from the declaratory judgment provision.

First some background, which unfortunately makes this a long post.

The Biologics Price Competition and Innovation Act

At a high level, the BPCIA, which was enacted as part of the Affordable Care Act in 2010, did two things, through amendments to the Public Health Service Act (PHSA) and the Patent Act.

  • First, it authorized FDA to approve abbreviated (thus, smaller and less expensive) applications for biologics. An abbreviated application cites an earlier application that FDA has approved and proposes a replica of the product in question.  The first product is called the “reference product.”  The replica is called a “biosimilar,” and it is roughly analogous to a generic drug.  The company that proposes a biosimilar does not have to reproduce all of the safety and effectiveness research performed by the innovator (called the “reference product sponsor” in the statute), which is why the marketing application is “abbreviated.”
  • Second, the statute created a mechanism for addressing patent disputes between the two companies before the biosimilar company launches its product on the market. Because the patent infringement litigation can at least start before product launch, it is sometimes referred to as “premarket” patent litigation.

Overview of the patent resolution mechanism

The mechanism for resolution of patent issues boils down to this: a private process in which the two companies identify the relevant patents, followed by immediate litigation on a subset of those patents, and separately an opportunity for litigation on the rest of the patents in the final 180 days before the biosimilar company launches its product.

The mechanism is described in nine paragraphs of section 351(l) of the PHSA.  (The Court refers to this provision by its section number in the U.S. Code, which is section 262 of title 42.)  The entire process is sometimes called the “patent dance.”  Paragraphs (2) and (9) were at issue in the Supreme Court case.

  • Paragraph (2) says that within 20 days after FDA accepts its application, a biosimilar company must provide its application and information about its manufacturing process to the innovator.
  • Paragraph (3) prescribes a process through which the two companies generate a master list of relevant patents. These are patents that either company believes could reasonably be asserted by the reference product sponsor if someone made or sold the biosimilar without its permission.
  • Paragraphs (4) and (5) prescribe a process through which the two companies identify a subset of those patents for “immediate” litigation. I’ll call those the “first phase” patents.
  • Paragraph (6) gives the innovator 30 days to bring suit on the first phase patents after the subset has been finalized.
  • Paragraph (7) prescribes a process for adding new patents to the master list after the subset has been created.
  • Paragraph (8) requires the biosimilar company to provide notice to the innovator no later than 180 days before commercial launch. Once the innovator receives this notice, it may seek a preliminary injunction prohibiting manufacture or sale of the biosimilar until a court rules on patent validity and infringement with respect to the rest of the patents in the master list. I will call those the “second phase” patents.

Paragraph (9) is entitled “Limitation on Declaratory Judgment Action,” and it lays out three scenarios and explains their consequences.

  • (A): if the biosimilar applicant provides its application and manufacturing information under paragraph (2), then neither company can bring a declaratory judgment action regarding the second phase patents until the notice of commercial launch.
  • (B): if the biosimilar applicant fails to complete an action required under paragraph (3), (5), (6), (7), or (8), then the innovator can bring a declaratory judgment action for the first phase patents, but the biosimilar company cannot. B is entitled “subsequent failure to act,” and most people — including the Supreme Court (Slip Op. at 7) — assume this provision refers to situations where the applicant previously provided its application and manufacturing information.  (It doesn’t actually say this.)
  • (C): if the biosimilar company fails to provide its application and manufacturing information under paragraph (2), then the innovator can bring a declaratory judgment action with respect to any patent that claims the product or a method of using the product (remember that in this scenario, no master list was created), but the biosimilar company cannot.

The theory that participation might be optional

During the legislative negotiation process, which I have written about in detail, the notion that the BPCIA should have a premarket patent resolution mechanism was taken for granted.  Premarket patent litigation for generic drugs had been around for more than 20 years.  Only one proposal (the Inslee bill in 2007) omitted patent provisions, and it went nowhere.  In every other serious proposal, from the first bill in September 2006 to the time of enactment in March 2010, there was a scheme for resolution (or at least commencement) of litigation over patents prior to biosimilar market entry.  Of course, the specifics varied, and the schemes were always different from the Hatch-Waxman scheme.

To my knowledge, the first public mention of the theory that biosimilar companies might be able to “opt out” of the process described in section 351(l) appeared in a May 2010 article authored by Jim Czaban and two colleagues.  On page 8, he included one paragraph, pointing out the declaratory judgment provision in (9)(C) and suggesting that the application disclosure requirement “may not be as mandatory as it would seem.”

The omission of manufacturing process patents

Czaban offered this suggestion because manufacturing process patents appear to be carved out of (9)(C).

The Hatch-Waxman premarket litigation framework applies only to patents that claim the reference drug or a method of using the drug.  (See section 505(b)(1) of the FDCA, here, which identifies the patents that an innovator must list.)  It does not apply to patents that claim any other method or process (such as a manufacturing process), so for instance you will not find those patents in the Orange Book.

The BPCIA was not limited in this way, and that was intentional.  During the legislative process, many people argued that manufacturing process patents could be more important than other types of patents, for biological product innovators.

As a result, most of the premarket patent resolution provisions in the BPCIA apply to any patent that the innovator reasonably believes could be asserted.  This is not true of the provision that the Court interpreted in Sandoz v. Amgen.

Paragraph (9)(C) states that if the biosimilar company fails to provide its application and manufacturing information, the innovator — but not the biosimilar company — may bring a declaratory judgment action with respect to “any patent that claims the biological product or a use of the biological product.”

It was this disparity that prompted Czaban to wonder whether application disclosure was really mandatory.  Perhaps, he wrote, the point of this omission was to make it possible for biosimilar companies to avoid premarket litigation of manufacturing process patents.  The idea was: if the company provided its application, manufacturing process patents would be part of the master list and subject to premarket litigation.  If the company did not provide its application, then the innovator would have the declaratory judgment option, but only with respect to product and use patents.  A biosimilar company wanting to avoid litigation of the manufacturing process patents, he reasoned, might choose not to provide its application.

Are process patents really excluded?

Yes, but probably not in the way that the May 2010 article suggested.

To understand this, you need to read the artificial act of infringement added by the BPCIA to the Patent Act.

What is an artificial act of infringement?  Making and testing a drug or biologic in order to submit a marketing application is not patent infringement, which means ordinarily a federal court would not hear a patent infringement case before FDA approval and product launch.  Premarket patent litigation for FDA regulated products is made possible by an “artificial” act of infringement: submitting an application with the intent to market before patent expiry.

Congress crafted the artificial act of infringement in 1984, as part of the Hatch-Waxman Amendments.  Section 271(e)(2) of the Patent Act made it an act of infringement to submit a generic drug application if the purpose was to gain approval prior to expiry of a patent claiming the innovator’s drug or a method of using the innovator’s drug.  This provision doesn’t include manufacturing process patents.  When Congress enacted the generic animal drug statute in 1988, it just added the equivalent of “or submitting a generic new animal drug application.”  And when Congress passed the BPCIA, it added the equivalent of “or submitting a biosimilar application.”

When you read these two statutes – 35 U.S.C. § 271(e)(2)(C) and 42 U.S.C. § 262(l)(9)(C) – together, this is what you find:

  1. When the companies have followed the information exchange process and generated a master list of relevant patents, section 271(e)(2) says that submission of the biosimilar application is an artificial act of infringement, if the goal is to market before expiry of a product patent or use patent. Then (e)(2)(C)(i) tells us which patents are artificially infringed: any patent in the master list.  This includes manufacturing process patents.
  2. When the biosimilar applicant did not provide its marketing application, section 271(e)(2) still says that submission of the biosimilar application is the artificial act of infringement, if the goal is to market before expiry of a product patent or use patent. Then (e)(2)(C)(ii) tells us which patents are artificially infringed: any patent that could have been identified by the innovator for the master list.  This unquestionably includes manufacturing process patents.  The problem is that paragraph (9)(C) in the PHSA – the “remedy” described by the Court, and the provision on which the May 2010 article focused – seems to contemplate suit only on a patent claiming the product or a use.
  3. If the only patents at issue are manufacturing process patents, it seems there is no possibility of premarket patent litigation at all. This is true even if the companies have followed the information exchange process and generated a master list and a subset list.  The artificial act of infringement requires a product or use patent.  (See the U.S. brief for this point.)

Bottom line

Manufacturing process patents appear to be excluded from premarket patent litigation if they are the only patents implicated.  But – although the origin of the theory that the “patent dance” might be optional was the supposed exclusion of manufacturing process patents from premarket patent litigation when the biosimilar company fails to provide its marketing application, in truth it is not obvious that manufacturing process patents are excluded in this situation.  To be sure, the “remedy” provision references only product and use patents.  But as the Federal Circuit pointed out below (see note 3 of that opinion, on page 13), the patent statute unquestionably creates an artificial act of infringement with respect to manufacturing process patents.  If the biosimilar company has not provided its marketing application, it may be hard for the innovator to determine whether there is a basis for suit, but the court of appeals wrote unambiguously that the innovator can assert a manufacturing process patent in this situation.

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s