Insights

BPCIA Info Exchange Is Failing To Streamline Patent Litigation

This article originally appeared in Law360.


For 10 years, a debate has swirled around the so-called patent dance created when the Biologics Price Competition and Innovation Act was enacted.

And with good reason — to the extent branded biologics and biosimilar companies have used the dance, it has not had a significant effect on the eventual number of patent assertions.

Although it was established in an attempt to streamline biosimilar drug applications, an in-depth review of the 30 litigations filed pursuant to the BPCIA shows that engagement with the patent dance has not led to noticeable differences in the number of patents asserted.

At most, the dance process has functioned to set the stage for subsequent settlement negotiations. Of course it is an open question whether the process itself materially advances the settlement process.

The BPCIA provides a regulatory framework for the applicant and reference product sponsor, or RPS, to exchange confidential information contained in the abbreviated biologics license application, or aBLA, in an effort to narrow the patents asserted in the first round of litigation. This exchange is commonly referred to as the patent dance.

Although it may not be possible to determine the value of the patent dance for a particular product before the application is made, our findings can provide some insight into potential strategies for both the applicant and the RPS.

Engagement in Patent Dance Doesn't Influence Number of Patents Asserted

Our key finding is that there is almost no difference in the average number of patents asserted at the beginning and end, or current stage, of litigation when comparing where parties did not engage in the patent dance with where the parties completed the dance. Indeed, a significant reduction in the number of asserted patents can occur even when the parties do not formally engage in the dance. For example, in the 2019 U.S. District Court for the District of New Jersey case Amgen Inc. v. Kashiv Biosciences LLC,[1] 17 patents initially were asserted, yet by the time the case settled only three patents remained in the case. That reduction occurred without any substantive ruling in the case indicating that negotiation in the ordinary course of litigation can achieve the same purpose as the formal patent dance process.

To date, 30 litigations have been filed under the BPCIA, involving eight distinct biologic products. Each biologic had exhausted its 12-year data exclusivity period prior to the commencement of the litigations. A majority of parties completed the patent dance and even more started the dance but did not finish. Less than a third had no engagement with the patent dance.

We categorized each of the 30 litigations in one of three ways. The parties: (1) engaged in the patent dance; (2) disputed taking part in the patent dance; or (3) voluntarily did not take part in the patent dance. We then reviewed the categories, noting the number of patents asserted in each litigation.

Completed the Patent Dance

Of the 30 litigations, 17 engaged in the patent dance. The parties in this category completed the patent dance, although 15 of the 17 parties disputed if the opposing party took the proper steps according to the BPCIA.

Generally, the dispute centered on an alleged deficiency of the copy of the aBLA sent, or allegations that the applicant did not send all of the requested additional information. But in all 17 cases, the parties ultimately engaged in final negotiations, with these results:

  • Three went from two patents to one at the cessation of litigation. 
  • Eleven remained at the same number of patents asserted throughout, ranging from one to 21.
  • One increased by one patent.
  • Two are in the preliminary stages, so it is not clear if the number of patents asserted will change.

The parties asserted approximately eight patents on average, while the average number of patents that remained at issue following the dance, or at issue at the conclusion of the litigation, is approximately seven.

Began but Did not Complete

In six of the 30 litigations, the parties got into a dispute during the patent dance, resulting in the parties not completing the dance. In these cases, the applicant sent the RPS its aBLA, thus beginning the patent dance, but at some point information exchange stops or the parties do not make it to or through the negotiation stage because of disputes.

Of those six matters:

  • One began with six patents and ended with one patent at issue.
  • One remained at two patents asserted.
  • One went from 24 to 25 patents asserted.
  • One went from 40 to 20 patents asserted.
  • One remained at 40 patents.
  • One went from 40 to 16.

On average, approximately 25 were asserted at the start, and the average number of patents at issue at the conclusion of the litigation was approximately 17.

Only one case with a disputed patent dance litigated questions related to the patent dance itself. In the 2018 case Janssen Biotech Inc. v. Celltrion Healthcare Co.,[2] the U.S. District Court for the District of Massachusetts ruled that Celltrion, the biosimilar applicant, did not engage in good faith negotiations, a required step in the patent dance.[3]

Given the ruling, Celltrion attempted to engage Janssen, the RPS, in the required negotiations. Janssen asserted that the patent dance does not mandate engagement in negotiations after litigation has begun, and thus it did not again engage in the process.

Did not Engage in the Patent Dance

In seven litigations, the parties did not engage in the patent dance. For cases that we identified in this category, the biosimilar applicant either affirmatively states that it is electing not to engage in the patent dance, affirmatively states that it is stopping the patent dance, or the applicant takes no steps to begin the patent dance, i.e., it never sends the RPS its aBLA — the first step in the patent dance process.

Of these seven cases:

  • One began litigation with one patent at issue and ended with two at issue.
  • One began with 17 and ended with three at issue.
  • One began with five and ended with two.
  • Four ended with the same number they started with: one, three, five and 24 patents.

In the litigations in which the parties did not engage in the patent dance, the average number of patents asserted was eight, and the average number of patents at issue at the conclusion of the litigation was approximately six.

The litigations with the most patents litigated are those in which the parties ended up disputing if the proper steps were being taken, resulting in the dance terminating. This scenario also produced the biggest difference in the number of patents initially asserted as opposed to the final number of patents asserted.

This could indicate a contentious relationship between the parties, or a specific strategy that prevented further streamlining of the litigation, regardless of whether the parties engaged in the patent dance.

Conclusion: The Failure of the Patent Dance

The patent dance is failing to streamline litigation. The process is not offering the biosimilar manufacturer any certainty about the cost of market entry, nor is it a valid mechanism for assessing risk at the outset. As one executive at a biosimilar manufacturer noted, the most important issue for him is:

providing certainty for the business as quickly as possible. The patent dance structure does not provide the same type of advance notice of relevant patents to biosimilars that the Orange Book process provides in small molecule context.

In contrast, the Hatch-Waxman Act Orange Book process for small molecule drugs operates more efficiently. In particular, by requiring the proactive listing of relevant patents, the Orange Book eliminates the need for any prolonged exchange of patent information as is required by the BPCIA.

Another pharmaceutical executive at an RPS company echoed this sentiment, noting that the patent dance is not a satisfactory tool for innovators to deal with the loss of exclusivity in a way that provides certainty. Instead, if valuable at all, the patent dance is at most a useful tool to help open lines of communication in a way that can provide a path to eventual settlement.

Reviewing the BPCIA litigations, a majority have settled; of the 14 settled cases, eight involved parties who engaged in the patent dance, two did not engage in the patent dance, and five disputed the patent dance, ultimately not completing it. Four out of four that disputed the patent dance still settled.

All litigations have so far been related to biologic products that had already passed their 12-year exclusivity period at the start of litigation. With newer biosimilars where the 12-year exclusivity period has not expired, there is not the same urgency to get to litigation, potentially making the patent dance more useful for the RPS.

At this point, the promise of the patent dance as a tool to help parties narrow the patents asserted is questionable. To fulfill that promise, the BPCIA needs a better mechanism for determining the patents at play. So far, relying on the parties to do this through an information exchange is not working, so both applicants and RPS innovators should carefully consider their options before entering into patent dance.


Eric Marandett is a partner and co-chair of the intellectual property litigation practice group at Choate Hall & Stewart LLP and Caila Heyison is a senior associate at the firm.

The opinions expressed are those of the author(s) and do not necessarily reflect the views of the firm, its clients or Portfolio Media Inc., or any of its or their respective affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.

[1] Amgen v. Kashiv, No. 18-3347 (D.N.J.).

[2] Janssen v. Celltrion, No. 17-11008 (D. Mass.).

[3] 42 USC 262(l)(4) and (5).