Why Patentees Should Learn to Love Multidistrict ANDA Suits

A fresh focus on venue in recent decisions from the Supreme Court and the Federal Circuit Court of Appeals is making it less likely for cases involving multiple abbreviated new drug application (ANDA) filers to be brought in the same venue and then combined after filing.

Instead, multidistrict litigation is now emerging as an effective addition to the Hatch-Waxman litigation toolbox. While this trend has initially been greeted with trepidation by owners of branded drugs, a deep dive reveals that such suits can be an effective means of ensuring efficiency and consistency between cases – ultimately saving rights holders both time and money.

Overview of Hatch-Waxman litigation

To market a new drug, a sponsor submits a new drug application (NDA) to the to the Food and Drug Administration (FDA). In addition to information such as the proposed label, directions for use, and clinical data, the NDA must also list all patents covering the drug or its labelled methods of use – these are listed in a public register known as the Orange Book. Once the drug is approved, the FDA awards the NDA holder/brand-drug sponsor a certain period of marketing exclusivity depending upon the type of exclusivity sought (eg, five years for new chemical entity exclusivity (NCE)). This FDA exclusivity is different from any exclusivity provided by Orange Book patents that cover the licensed drug.

A company that wishes to market a generic version of an approved drug may file an ANDA. If the licensed drug has NCE exclusivity, then the FDA cannot accept an ANDA for a generic version of that drug for five years unless the ANDA submission includes a Paragraph IV certification to the patents listed in the Orange Book for the licensed drug. If the ANDA submission does include this, then the FDA will accept the ANDA submission one year before the end of the NCE exclusivity (NCE-1 date).

The Paragraph IV certification states that in the opinion of the ANDA filer – and to the best of its knowledge – the Orange Book patents listed for the licensed drug are invalid, unenforceable or will not be infringed by the generic drug product.

It is often the case that several different companies simultaneously file ANDAs with the FDA on the NCE-1 date. The first company to get its filing in wins the exclusive right to market the generic drug for 180 days, making this highly competitive.

Under the Hatch-Waxman Act, it is “an act of infringement to submit [an ANDA] for a drug claimed in a patent or the use of which is claimed in a patent… if the purpose of such submission is to obtain approval… to engage in the commercial manufacture, use or sale of a drug… claimed in a patent or the use of which is claimed in a patent before the expiration of such patent” (35 USC §271(e)(2)). 

If an ANDA filer includes a Paragraph IV certification in its FDA submission, it must also provide notice to the brand-drug sponsor, including a detailed statement of the factual and legal basis of its opinion that the patent is invalid or will not be infringed. If the brand-drug sponsor disagrees, it has 45 days in which to file an infringement suit against the ANDA filer. Once the litigation is initiated, FDA approval to market the generic drug is generally postponed for 30 months (commonly referred to as the ‘30-month stay’) unless the patent expires or is judged to be invalid or not infringed before that time.

Therefore, it is not uncommon to have multiple litigations against multiple ANDA filers initiated around the same time over the same patents and covering essentially the same products – generic versions of the licensed drug. Additionally, the remedy requested in each case is also the same – an injunction to prevent the ANDA filer from launching its generic product before the expiration of the patents. Brand-drug companies have historically tried to sue all ANDA filers in the same venue and then consolidate cases in order to avoid duplicative discovery and inconsistent rulings on claim construction.

The 30-month stay deadline is crucial in such cases. If the court’s validity and/or infringement determination does not take place before the expiration of the stay, then the ANDA filer could launch the product before these patent issues are resolved. This is referred to as launching “at risk” because there is a chance that marketing of the generic product could take place only for the court to then issue an injunction. However, once the ANDA product hits the market, the price of the licensed drug will begin to fall and insurance companies will start changing how they reimburse for it. Therefore, even if the ANDA product is enjoined after the at-risk launch, lasting damage can be inflicted on the brand-drug sponsor during that time.

Recent changes in venue rules for Hatch-Waxman actions

Section 1400(b) of Chapter 28 of the US Code provides that “[a]ny civil action for patent infringement may be brought in the judicial district where the defendant resides, or where the defendant has committed acts of infringement and has a regular and established place of business”. Both pathways to venue under this section – residence and location of infringement – have been the focus of considerable judicial attention in recent years.

Section 1400(b)’s predecessor statute was enacted in 1897 in response to confusion over whether patent cases were controlled by the same venue considerations that govern civil actions generally. The 1897 legislation resolved this confusion by “plac[ing] patent infringement cases in a class by themselves, outside the scope of general venue legislation” (Brunette Machine Works, Ltd v Kockum Industries, Inc, 406 US 706, 713 (1972)). Thus, a defendant’s ‘residence’ under Section 1400(b) is not defined by the general venue statute Section 1391; rather, a defendant corporation resides only in its state of incorporation for purposes of Section 1400(b) (Fourco Glass Co v Transmirra Prod Corp, 353 US 222, 229 (1957)).

Nevertheless, shortly after Section 1391 was amended in 1988, the Federal Circuit held that the newly added phrase “[f]or purposes of venue under this chapter” effectively imported Section 1391(c)’s definition of residency – (ie, that defendant corporations reside in any judicial district in which they are subject to personal jurisdiction when the action is commenced) into all statutes in the chapter, including Section 1400(b) (VE Holding Corp v Johnson Gas Appliance Co, 917 F 2d 1574, 1579-80 (Fed Cir 1990)). 

For decades following VE Holdings, venue in patent infringement actions was deemed proper wherever the defendant corporation was subject to personal jurisdiction. However, in 2017 the Supreme Court reversed that doctrine and reiterated that “a domestic corporation ‘resides’ only in its State of incorporation for purposes of the patent venue statute” (TC Heartland v Kraft Food Group Brands, 137 S Ct 1514, 1517 (2017)). This “dramatically changed the venue landscape in patent cases” (Valeant Pharms N Am LLC v Mylan Pharms Inc, 978 F 3d 1374, 1375 (Fed Cir 2020)).

The return to a narrow interpretation of residence increased the focus on Section 1400(b)’s second venue provision, allowing patent infringement actions to be brought “where the defendant has committed acts of infringement and has a regular and established place of business”. As applied to Hatch-Waxman litigation, the first major question was whether venue was proper only where a current infringement was taking place – namely, the submission of an ANDA, pursuant to Section 271(e)(2) of Chapter 35 of the US Code – or whether venue is also proper where future distribution of the generic in question was contemplated.

Shortly before the decision in TC Heartland, planned future acts of infringement had been deemed sufficient to justify personal jurisdiction and, thus, venue under Section 1400(b)’s residency provision (see Valeant, 978 F 3d at 1379 (discussing Acorda Therapeutics, Inc v Mylan Pharms Inc, 817 F 3d 755 (Fed Cir 2016)). However, in 2020 the Federal Circuit concluded that “the submission of the ANDA, and only the submission” constitutes an act of infringement in a Hatch-Waxman action (Valeant, 978 F 3d at 1381). Thus, venue is proper under the “acts of infringement” provision “only in those districts that are sufficiently related to the ANDA submission” (Id at 1384). 

The Valeant court recognised that the District of Maryland, where the FDA receives an ANDA, might satisfy this test for venue but declined to resolve that question or to determine “what all relevant acts involved in the preparation and submission of an ANDA might be” (Id at 1384 n8).  The Federal Circuit has revisited the issue only once since Valeant, clarifying that acts related to a submission but distinct from it – such as sending the mandatory Paragraph IV notice letter – are not acts of infringement for purposes of venue (Celgene Corp v Mylan Pharms Inc, 17 F 4th 1111, 1120-22 (Fed Cir 2021) (receipt of a Paragraph IV notice letter does not render venue in the recipient’s district proper)).

For the time being, it appears that venue in Hatch-Waxman cases is proper at least in the defendant corporation’s state of incorporation and in the District of Maryland, where the FDA receives the ANDA. It remains to be seen what other “relevant acts involved in the preparation and submission of an ANDA”, if any, will be deemed to support venue in additional districts.

Multidistrict litigation and its use in Hatch-Waxman litigation to date

As the Valeant court noted, the current limitations on venue in Hatch-Waxman litigation may require brand-drug sponsors “‘to file and maintain largely identical suits in multiple districts’ causing an increase in time and expense to resolve the cases and ‘result[ing] in inconsistent judgments’” (Valeant, 978 F 3d at 1383 (quoting Bristol-Myers Squibb Co v Mylan Pharms Inc, No CV 17-379-LPS, 2017 WL 3980155, at *12 n17 (D Del, 11 September 2017)). Although Valeant suggested that a solution to this issue may need to come from Congress, a partial solution – multidistrict litigation – already exists. 

Section 1407(a) of Chapter 28 of the US Code provides in relevant part that “[w]hen civil actions involving one or more common questions of fact are pending in different districts, such actions may be transferred to any district for coordinated or consolidated pretrial proceedings” (ie, centralised in a multidistrict litigation for pre-trial proceedings in a single district – if doing so “will be for the convenience of parties and witnesses and will promote the just and efficient conduct of such actions”). In such cases, any party may file a motion with the judicial panel on multidistrict litigation (JPML) seeking to centralise such cases in a specified district (the transferee court). 

Multidistrict litigations are particularly common in mass litigations such as product liability actions in which damages for individual injury are sought. They allow parties to numerous litigations pending in different districts to streamline discovery and pre-trial motion practices, conducting these activities once in a single district rather than repeatedly across the country. 

Historically, Hatch-Waxman litigation has rarely been centralised in a multidistrict litigation. Fewer than a dozen Hatch-Waxman multidistrict litigations were formed prior to TC Heartland, and all of those were between 2001 and 2012. However, those early cases emphatically underscored the propriety of centralisation in Hatch-Waxman litigation. The JPML granted every motion for centralisation of Hatch-Waxman actions filed before TC Heartland, even in cases in which only two actions were pending (see, for example, In re: Nebivolol (%2C040) Pat Litig, 867 F Supp 2d 1354 (US Jud Pan Mult Lit 2012)). In each case the plaintiff brand-drug sponsor filed the centralisation motion and in all but two (id and In re: Fenofibrate Pat Litig, 787 F Supp 2d 1352 (US Jud Pan Mult Lit 2011)), the JPML centralised the cases in the district that the plaintiff had requested.

No Hatch-Waxman multidistrict litigations were formed between 2013 and 2018. However, from 2019 through to 2021 seven centralisation motions concerning six brand drugs were filed. These motions display a pattern that reflects both the heritage of Hatch-Waxman multidistrict litigations and the clear impact of TC Heartland. Like earlier cases, all seven were filed by plaintiff brand-drug sponsors. Six were granted; the only exception involved a case in which the JPML had already created an earlier multidistrict litigation concerning the brand-drug in question and merely declined to form a second multidistrict litigation.

In each case, the JPML centralised the actions in the district requested by the plaintiff. However, there is more: in each of these recent centralisation motions, all but one action was pending in the District of Delaware. The lone exception in each case was an action pending against Mylan Pharmaceuticals Inc in the Northern District of West Virginia (ie, the defendant’s state of incorporation) (see, for example, defendant’s answer to plaintiffs’ first amended complaint and counterclaims at 3, Pfizer Inc v Mylan Pharms Inc, No 1:20-cv-00244-TSK (ND WVa, 15 January 2021). The plaintiff in each case requested that the actions be centralised in the District of Delaware, and the JPML obliged. 

In short, each of the recently created Hatch-Waxman multidistrict litigations had the effect, intended or not, of transferring a case from a district where venue was clearly proper to a district in which this may or may not have been the case.

Considerations for multidistrict litigation in the Hatch-Waxman context

The main takeaway from Hatch-Waxman multidistrict litigation practice to date is that motions to centralise are a viable option to get all the parties into one litigation in the same place. The JPML has consistently deemed centralisation appropriate in Hatch-Waxman cases due to the complexity and overlap of the issues involved, even when only two actions were pending. 

Parties seeking a multidistrict litigation should keep in mind that the venue in which an action is filed may not be the venue in which pre-trial proceedings will ultimately be conducted. Indeed, pre-trial proceedings may even be conducted in a district where the action could not have been filed due to recent changes in the venue rules. 

From the plaintiff’s perspective, centralisation serves three main purposes. First, the classic multidistrict litigation goal of avoiding duplicative discovery. Once all of the related Hatch-Waxman actions have been centralised, company discovery regarding the patent(s) can be greatly streamlined. Document collections and productions can be coordinated such that a single set of document requests serves all defendants in the various actions. Likewise, company witnesses need only prepare for and submit to one deposition rather than serial depositions by defendants in multiple pending actions.

A second benefit is the efficiency of holding only one Markman (claim construction) hearing so that the definitions of the terms of the patents are the same in each of the future final trials.  Claim construction is a legal determination by the court, so one court could conduct the hearing and render a decision on the construction of the claims that the other courts would use, similar to the practice of multidistrict litigation courts conducting Daubert hearings to make determinations on admissibility of scientific opinions that will inform the courts that will ultimately try the cases and deciding dispositive motions which cut across swaths of cases.

A third upside is the chance to conduct pre-trial proceedings in a venue that would not be proper under TC Heartland and recent Federal Circuit cases. To date, the JPML has centralised Hatch-Waxman actions almost exclusively in the district of the plaintiff’s choosing. That said, no defendant has yet argued that transfer to the plaintiff’s chosen district would result in pre-trial proceedings taking place in a district where venue is improper. Such an argument likely would not defeat centralisation, as “transfers under §1407 are not limited by general venue statutes” (Lexecon Inc v Milberg Weiss Bershad Hynes & Lerach, 523 US 26, 39 n2 (1998)). However, it remains to be seen whether such an argument might influence the JPML’s selection of the transferee court in close cases. 

Centralisation is less beneficial for Hatch-Waxman defendants. Discovery directed to defendants is unlikely to have been duplicative had the actions proceeded individually, so pooling litigation is less of an upside. Still, defendants can potentially benefit. First, they can share the responsibility and cost of conducting discovery against the plaintiff(s). No one defendant need depose all of the plaintiff’s key figures; rather, they can coordinate and share the load. Some defendants with less at stake may choose to take a back seat and let others with more exposure lead the litigation.

Second, some transferee courts allow broader discovery directed to the plaintiff than would be otherwise available. Broader requests may be necessary to address the various interests of the defendants in a multidistrict litigation and the proportionality analysis of Rule 26(b)(1) of the Federal Rules of Civil Procedure may favour deeper discovery than would be available in the individual actions.

Third, the transferee judge and/or case law governing the transferee court may be more favourable to the defendant’s position than those where the action was initially filed. In such cases, centralisation would be a welcome change of venue for the defendant.

Any party to an action may file a motion to centralise with the JPML, with a copy filed in the district court where the action is pending (28 USC §1407(c)(ii)). Although motions may be filed throughout a litigation, they are most likely to succeed if filed early enough that the benefits of centralised discovery can still be reaped in full. Seeking centralisation is particularly important in Hatch-Waxman cases where all of the cases need to be concluded within the 30-month stay. 

Moreover, when one of the actions being centralised has proceeded further than the others, the JPML sometimes chooses to centralise actions before the judge handling the more advanced action. Thus, the timing of a motion to centralise can impact not only the likelihood of centralization but also the selection of the transferee court. 

Looking ahead

Although Hatch-Waxman multidistrict litigations have been rare historically, their use has increased sharply since TC Heartland restricted venue in patent actions. With the Federal Circuit’s further narrowing of venue in Valeant and Celgene, litigants may continue to look to multidistrict litigations more and more as a means of managing litigation in which multiple defendants cannot be sued in a single district.