Business for pharmaceutical and life sciences companies is rarely confined to one particular country. This is especially true in the European Union, which not only is a transnational economic and monetary union, but has also harmonised many regulatory and legal aspects of the pharma business. In today’s electronic age, when data, goods and services move effortlessly across borders, this phenomenon is compounded. It could even be said that the traditional legal concept of geographical territories and jurisdiction seems somewhat outdated.

Against this background, it is unsurprising that patent infringement rarely stops at geographical borders. More often, infringement takes place simultaneously in several jurisdictions – especially given that all online business activity is automatically cross-border.

Another important cross-border context arises when the infringement itself occurs across borders. Today’s large conglomerate pharmaceutical companies frequently divide work between their different national subsidiaries. Moreover, outsourcing to foreign contract manufacturing organisations is popular not only for Big Pharma, but also for start-up companies that cannot afford their own production facilities. Such contexts can give rise to complex patent infringement scenarios, in which not all steps of a patented method are performed within one country.

In contrast to these business realities, patents remain fundamentally territorial rights (ie, their legal effect and enforceability are usually confined to the borders of the country in which they are granted). This is also true of European patents, which – although they must be applied for and are examined centrally at the European Patent Office – disintegrate upon grant into a bundle of national patents. Unfortunately for patentees, these national patent rights must be enforced on a country-by-country basis. Thus, it seems, geographical borders have remained surprisingly resilient in the patent litigation context. In reality, enforcing a patent throughout Europe can be a complex and time-consuming enterprise.

Brussels I Regulation

International jurisdiction within the European Union is governed by the Brussels I Regulation, a revised version of which entered into force on January 10 2015 (EU Regulation 1215/2012), including an additional amendment stipulated in EU Regulation 542/2014, necessitated by the future Unified Patent Court (UPC). The new regulation replaces the previous EU Regulation 44/2001, which in turn was based on the Brussels Convention of September 27 1968.

The Brussels I Regulation is binding European law, directly applicable in all EU member states. It is to be interpreted independently (at the highest level by the European Court of Justice (ECJ)) and not according to national law.

Under the Brussels I Regulation, the claimant can choose between the following venues when bringing a patent infringement action:

  • the courts of the state where the defendant is domiciled (Article 4);
  • the courts of the state where the harmful event occurred or may occur (Article 7(2)); and
  • in the case of joint defendants, the courts of the state where one of the defendants is domiciled, provided that the claims against them are so closely connected that it is expedient to hear and determine them together in order to avoid the risk of irreconcilable judgments resulting from separate proceedings (Article 8(1)).

In addition, contractual prorogation of jurisdiction is possible (Article 25).

Regarding torts such as patent infringement, a defendant can be sued in the courts of the place where the harmful event occurred. The latter can principally be the place of the event giving rise to damages, as well as the place where the damage occurred. This would open the floodgates to excessive forum shopping and put claimants in an unjustifiably advantageous position. Unsurprisingly, the ECJ restricted such forum shopping in its landmark decision in Shevill (C-98/93), developing a mosaic principle regarding damages which has become known as the ‘Shevill doctrine’. Accordingly, proceedings in the courts of the place where the damage occurred can be brought only in respect of damage that actually occurred within that forum. If the claimant wants to concentrate all proceedings before a single court, it needs to address the courts of the place where the harmful event giving rise to liability occurred or the courts of the place where the defendant is domiciled.

In principle, the Brussels I Regulation allows a patentee to sue a defendant in the court of a single EU member state for infringement of a European patent where this has taken place in several EU member states. However, this flexibility is limited considerably by Article 24(4) of the regulation. With regard to the validity of registered IP rights such as patents, trademarks and designs, the country of registration (for European patents, the country of validation) has exclusive jurisdiction. Thus, as soon as defendants raise the invalidity of the patent-in-suit as a defence or file a counterclaim for nullification, cross-border jurisdiction is in limbo and the court must decline jurisdiction ex officio.

Only in the case of preliminary injunctions does the limitation set out in Article 24 of the Brussels I Regulation remain unaffected. This is due to Article 35, which allows for provisional (including protective) measures to be sought in a particular national court, even if – under other provisions of the Brussels I Regulation – the courts of another member state will have jurisdiction over the final case on the merits.

European cross-border litigation – a historic perspective

In its traditional sense, the term ‘cross-border’ was used by patent litigators to refer to cases in which – based on the Brussels I Regulation – a court in one EU member state decided on cross-border infringement of a European patent and granted injunctive relief against an infringer for multiple EU member states at the same time.

Since the 1990s, Dutch courts in particular had granted cross-border injunctions applicable in all contracting states designated by a European patent whenever infringement of the Dutch part of the European patent was claimed or the defendant was based in the Netherlands. Dutch courts had also accepted the inclusion in infringement proceedings of defendants which were not domiciled in the Netherlands and had not infringed the Dutch part of the patent if the main defendant was based in the Netherlands and the other defendants were part of a group of companies and acted on the basis of a common business policy (the ‘spider-in-the-web’ doctrine). For over a decade, a debate raged over whether cross-border injunctions were legally possible. Many national courts refused to examine cross-border infringement as soon as the validity of the patent was challenged. In addition, most national courts, apart from the Dutch courts, dismissed the spider-in-the-web doctrine, which was considered to be based on an extremely broad interpretation of Article 8 of the Brussels I Regulation.

The picture changed dramatically with two seminal ECJ decisions issued on July 13 2006: GAT v Luk (C-4/03) and Roche v Primus (C-539/03) effectively put an end to the Dutch practice of granting cross-border injunctions in cases on the merits.

In GAT v Luk the ECJ held that the exclusive jurisdiction stipulated in Article 24(4) of the Brussels I Regulation is not limited to proceedings that have already been initiated with the goal of declaring a patent invalid. Rather, the Article 24(4) exclusive jurisdiction provision kicks in fully as soon as foreign invalidity is invoked as a defence. The infringement court cannot consider, let alone rule, on that invalidity defence. Put more simply, this means that a court has no jurisdiction as to the infringement of the foreign parts of a European patent as soon as the validity of such parts are questioned or challenged by the defendant. Effectively, this has abolished cross-border decisions on the merits, as the invalidity of the patent in suit is nearly always argued by the alleged infringer as a defence against cross-border infringement claims.

In Roche v Primus the ECJ further limited cross-border litigation by adopting a narrow interpretation of Article 8(1) of the Brussels I Regulation (the joint defendant provision). Accordingly, claims relating to the infringement of various national parts of a European patent are not to be considered “so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings” as required by Article 8(1).

Realities of patent litigation today

The ECJ decisions in GAT v Luk and Roche v Primus represented a watershed and the number of cross-border patent litigation cases on the merits has been marginal ever since. Still, taking a look at the litigation statistics, it is clear that the vast majority of patent disputes in Europe are concentrated in just a few selected venues. According to statistics published in 2013, of the 1,500 to 2,000 patent infringement proceedings tried every year by European courts, 66% are brought in Germany, 15% in France, 14% in Italy, 3% in the United Kingdom and 2% in the Netherlands. As to the outcome of patent infringement proceedings initiated in Germany over the past years, on average roughly 40% of infringement actions resulted in an injunction, 40% were dismissed due to non-infringement and 20% were suspended pending the outcome of invalidity proceedings before the Federal Patent Court and the Federal Supreme Court. However, cross-border litigation cases make up only a very small percentage of these 1,500 to 2,000 patent infringement proceedings.

This concentration of cases in a handful of litigation venues cannot be explained by there being no need to sue in other countries. As explained above, in light of ECJ case law, cross-border verdicts are largely unavailable in patent cases today. Rather, the small number of popular venues is due to companies deciding to enforce their patents in only a few jurisdictions on the basis of a cost-benefit analysis. It is often sufficient to obtain an injunction in the most important economic markets (eg, Germany, France or the United Kingdom) or in a defendant’s home court in order to shut down the infringement effectively. The popularity of the German courts can also be explained by their high level of specialisation in patent matters, which leads to greater predictability for plaintiffs and quicker patent infringement proceedings (it takes about one year from filing the complaint to obtaining a first-instance decision in Germany).

Remaining application window for cross-border injunctions

The topic of cross-border injunctions regained momentum in 2012 following the ECJ’s acclaimed decision in Solvay v Honeywell (C-616/10), in which it tried several fundamental issues. First, the ECJ stated that a court in an EU member state can have proper jurisdiction over co-defendants from other member states in accordance with Article 8(1) of the Brussels I Regulation if the principal defendant is domiciled within the jurisdiction of that court and the unlawful acts of the defendants relate to the same product and the same national part of a European patent. The second issue was whether the defendant can prevent the issue of a cross-border preliminary injunction by raising an invalidity defence. This was denied by the ECJ, which held that the Brussels I Regulation does not rule out cross-border preliminary injunctions in such cases, as only a preliminary assessment of the validity would occur and this would not influence the later invalidity decision in the main proceedings.

In summary, two types of cross-border injunction still seem possible today:

  • A preliminary cross-border injunction is possible if defendants violate several national parts of the same European patent and the national requirements for an injunction are met.
  • In the main proceedings, a cross-border injunction will be possible only in the rare cases where a defendant raises no invalidity defence and the presiding court also has jurisdiction as the defendant’s home court.

In this context, the recent UK case of Actavis v Eli Lilly is notable (EWHC 1511 (Pat)). Here, a cross-border declaration of non-infringement in the United Kingdom, France, Italy and Spain was granted. Remarkably, the United Kingdom held that the Brussels I Regulation did not apply, because Eli Lilly had no domicile in the European Union. Moreover, jurisdiction over the foreign patents was also possible as Actavis had not asserted their invalidity. The regional court of Dusseldorf also assumed jurisdiction over this case. It did not consider the lis pendens provisions in Article 29(1) of the Brussels I Regulation applicable, as the parties in the UK and German proceedings were not identical.

A glimpse of the future – the UPC

In 2013, 25 EU member states (all member states except Croatia, Poland and Spain) signed the long-awaited agreement on a UPC.

The UPC will have exclusive jurisdiction over litigation involving all European patents (ie, not only the new European patent with unitary effect, but also all classic European patents). For a seven-year transitional period, patentees can opt out of the UPC’s exclusive jurisdiction for regular European bundle patents. If the UPC is chosen as the forum, it will decide infringement and validity centrally for the entire UPC territory. Thus, member states have conferred on the UPC the competence to issue true cross-border verdicts.

One of the steps necessary for the entry into force of the new judicial system – further revision of the Brussels I Regulation to accommodate the UPC (Regulation 542/2014) – has already been taken. The second requirement is ratification of the UPC agreement by at least 13 countries, including France, the United Kingdom and Germany.

Regulation 542/2014 added a new Article 71a to the Brussels I Regulation which expressly defines the UPC as a ‘court’ in the sense of the Brussels I Regulation. Whereas this in principle already achieved the requisite compliance of the Brussels I Regulation with the UPC agreement, the legislature did not stop there, but went further and conferred on the UPC an entirely new asset-based jurisdictional competence, which allows it also to exercise jurisdiction over defendants domiciled in non-EU member states (third states).

According to the new Article 71b(3), the UPC will be able to hear and rule on patent infringement disputes involving damages both inside and outside of the European Union. This will be possible where significant property belonging to the defendant is located in an EU member state and the dispute has a sufficient connection with such an EU member state (eg, because the claimant is domiciled there or the evidence relating to the dispute is available there).

For example, a Chinese company sued before the UPC for infringing a European patent that is in force in Germany, France, the United Kingdom, Switzerland and Turkey can also be held liable for damages that have accrued due to infringement of the European patent in Switzerland and Turkey, although these countries are not member states of the European Union or the UPC agreement. Especially for Swiss-based pharma companies, this new long-arm jurisdiction into Switzerland may have a substantial economic impact.

These changes to the Brussels I Regulation have considerably strengthened the position of patentees in the cross-border enforcement of their IP rights before the UPC, expanding it to a territory that is even broader than that of the European Union. In contrast to the UPC, national courts of EU member states regularly have no such extensive jurisdictional competencies in respect of third states. This advantage of the UPC jurisdiction is exciting and should be a further factor to help patentees decide whether to opt into the UPC’s exclusive jurisdiction.

Ralph Minderop

Arwed Burrichter

Natalie Kirchhofer

This article first appeared in IAM magazine. For further information please visit www.iam-magazine.com.