The UPC did not end parallel litigation: it turned it into a weapon

The Unified Patent Court was supposed, so the story went, to make parallel patent proceedings in Europe disappear. Three years on, the conclusion is inescapable: the UPC has not eliminated parallel litigation, it has turned it into an instrument. Opposition before the EPO, residual national actions, non-participating territories: fragmentation is not a flaw of the system, it is its architecture.

For in-house counsel, the consequence is direct: there is no such thing as the right forum in the abstract, only sequences of actions ordered in time. A party that treats the UPC, the EPO and national courts as separate arenas is no longer merely making an analytical mistake: it is exposing itself strategically in its multi-jurisdictional patent litigation.

The myth of the automatic stay

One received idea persists: that a pending EPO opposition entails a stay of the UPC action. Nothing is further from the truth. The Court treats the stay as a discretionary case management tool, not as a mechanical consequence of the parallel procedure. Invoking a "rapid decision expected" from the Office is not enough; real procedural maturity must be shown: a developed written phase, an imminent hearing, tangible proximity to a decision.

The strategic consequence cuts both ways. The proprietor knows its infringement action can advance despite the opposition; the opponent knows that filing an opposition buys no time, unless it builds precisely the maturity record the Court demands. The EPO's calendar thus becomes a piece on the board, not a shelter.

Bifurcation: exploitable timelines

Within the Court itself,Article 33(3) UPCA and Rule 37.4 of the Rules of Procedure allow a division seised of infringement and of a counterclaim for revocation to refer validity to the central division while retaining infringement. A stay is not the default regime: the infringement proceedings continue unless the patent is highly likely to be invalid.

This bifurcation creates distinct, and therefore exploitable, procedural timelines. A proprietor may seek an infringement decision before validity is resolved; a defendant may, conversely, try to align the calendars. Mastering the tempo is a skill in its own right, on a par with the merits.

Claim amendments transform the dispute

Parallel litigation is not played out only between forums: it is also played out on the title itself. Rule 30 of the Rules of Procedure requires a patentee defending its patent in amended form to demonstrate that the claims comply with the requirements of the EPC and to articulate infringement on that new basis; Rule 32 allows the defendant to adapt its non-infringement arguments in response. Each amendment shifts the ground of validity and of infringement simultaneously.

Litigation before the UPC therefore no longer follows a linear sequence, validity then infringement: the layers of argument intertwine, and the party that masters their interaction controls the case. The appellate case law embraces this complexity: divergent outcomes between the UPC and the EPO are not regarded as inherently irreconcilable, as the NanoString v 10x Genomics litigation showed. Uncertainty is not an accident of the system, it is one of its constitutive features.

Jurisdictional competition on the move

Predictions of the decline of national courts were premature. The UPC's caseload grows steadily: 274 cases in February 2024, 538 in October 2024, 635 in January 2025, and the official database listed more than 2,200 published decisions and orders by late summer 2026. But national courts have not disarmed, they have adapted. The Landgericht München I order of 25 September 2025 (corrected on 8 October 2025) is the starkest illustration: protection extending to 22 foreign jurisdictions, grounded in the development introduced by the Court of Justice in BSH Hausgeräte v Electrolux (C-339/22, 25 February 2025), in an assumed break with the earlier practice of refusing cross-border protection as soon as validity was challenged.

Competition does not disappear: it moves and intensifies. forum choice, including between UPC divisions, between the UPC and national judges, and between Europe and third-country courts, remains the first decision in any case. Those who have maintained an opt-out deliberately retain, for that matter, access to the national forum alone.

No universal forum

Expansion has its limits. The UPC has confirmed that it cannot revoke designations of non-participating states, such as the United Kingdom: there is no universal European patent forum. But the border has become asymmetrical: since Kodak v Fujifilm (Court of Appeal, 2 June 2026), the Court holds that Article 34 UPCA does not cap its jurisdiction and that it may hear infringement claims concerning non-UPC designations applying the foreign law, validity then being examined only between the parties. Revocation remains territorial; infringement travels.

The system's real architecture thus rests on organised division rather than unification: theUPCA created an additional arena, a powerful one, but one arena among others. The proprietor does not simply seek an injunction, it chooses where and when to apply pressure; the defendant does not merely invoke the EPO, it decides whether that argument delays, weakens or reframes the dispute. And the financial stakes of these sequences can now also be read in damages determination proceedings, where the figure obtained in one arena becomes a reference for all the others.

In practical terms, this reading dictates a discipline: mapping the arenas that are open and openable before any move, auditing the portfolio's opt-outs, monitoring opposition calendars, and preparing the defensive tools, protective letters and stay files, before the opponent chooses the battleground. Parallel litigation is not something to be endured: it is something to be planned, in both directions.

Key takeaways

• A stay of UPC proceedings pending an EPO opposition is discretionary: it requires proof of real procedural maturity, never the mere pendency of the opposition.
• Article 33(3) UPCA and Rule 37.4 allow validity/infringement bifurcation: the diverging timelines can be exploited by both sides.
• Rules 30 and 32 make claim amendment an engine of transformation of the dispute, on validity and infringement alike.
• Divergent UPC/EPO outcomes are not deemed irreconcilable: uncertainty is constitutive of the system.
• National courts are adapting: the Landgericht München I extended protection to 22 foreign jurisdictions after the BSH judgment (C-339/22).
• No universal forum: revocation remains territorial, but since Kodak v Fujifilm infringement of non-UPC designations can be adjudicated in Luxembourg, Paris or Munich.

Frequently asked questions

Does an EPO opposition stay the UPC action?

No, not automatically. A stay falls within the Court's discretion, and requires real proximity to the Office's decision: a completed written phase, an imminent hearing. The mere pendency of the opposition, even accelerated, is not enough.

Can the UPC separate validity and infringement?

Yes. Where a counterclaim for revocation is raised, the division seised may refer validity to the central division and retain infringement (Article 33(3) UPCA, Rule 37.4). The infringement proceedings stop only if the patent appears highly likely to be invalid.

What happens if the EPO revokes the patent after a UPC decision?

Revocation by the EPO takes effect retroactively: the patent is deemed never to have had effect, which deprives measures based upon it of their foundation. This is precisely why aligning the UPC and EPO calendars is a first-order strategic issue.

Is the United Kingdom entirely beyond the UPC's reach?

No. The UPC cannot revoke a UK designation, but it can, under the case law flowing from BSH and Kodak v Fujifilm, rule on infringement of that designation applying English law, validity being assessed only between the parties.

Dhenne Avocats designs and coordinates these litigation sequences between the UPC, the EPO and national courts, for claimants and defendants alike, as part of its patent litigation. Talk to us.

This article is an original and substantially updated adaptation of an analysis by Matthieu Dhenne first published on Kluwer Patent Blog on April 7, 2026: The UPC Didn't Kill Parallel Litigation, It Weaponized It.