Practical PI lessons at the UPC Court of Appeal: 8 things Abbott v Sinocare teaches us

Lessons from UPC_CoA_901/2025 based on the Order of 17 April 2026 and attendance at the oral hearing

Preliminary injunction (PI) practice at the UPC continues to develop at pace. For practitioners, the most valuable decisions are the ones that do more than announce outcomes and show the Court’s developing methodology: what it tests, what it assumes, and what it expects parties to prove within the short timeframe available. Abbott v Sinocare (UPC_CoA_901/2025) is one of those decisions. It is not just a case about a glucose-monitoring app. It is a case about how the Court of Appeal (CoA) is calibrating urgency, necessity and balancing of interests, and how it is integrating the EPO’s ‘technical character’ reasoning for excluded subject matter into the UPC’s own inventive step framework.

1. Urgency is evidence-driven: the clock starts when the patent owner could file with a realistic prospect of success

The order provides a clean statement of urgency principles built on earlier case law. The relevant period for ‘unreasonable delay’ runs from when the applicant became aware (or should have become aware) of the infringement to the extent it could file a PI application with a reasonable prospect of success. In practical terms, that means: when the applicant had (or should have had, with due diligence) the necessary facts and evidence to establish infringement in the front‑loaded manner required by the RoP.¹

That framing of the principles shaped the hearing in Luxembourg. The panel repeatedly tested what evidence was genuinely needed before filing, especially where the alleged infringing features were implemented in software. How long does it take to analyse an app? Is it enough to observe one instance of the relevant feature, or are days of usage required? The Court’s questions signal a willingness to interrogate the reality behind ‘investigation’ narratives. 

On the facts, the CoA accepted that the urgency requirement was met. It identified a decisive starting point (public availability of the user guide and products²), treated sample purchase, technical testing and market investigation as diligent preparation rather than delay, and decided that an eight‑week period, though ‘relatively long’³, was justified in these circumstances. 

The takeaway is not that eight weeks is a safe period. It is that urgency is best argued (and attacked) based on a timeline of evidentiary readiness: what could be known, when, and what was done promptly (and diligently) to obtain that knowledge. 

2. Necessity and balance of interests: the Court requires market reality, not basic assertions

The Court reiterates that PI proceedings are summary proceedings and are not a substitute for a full merits examination; they are justified only where proceedings on the merits cannot be awaited. Importantly, the ‘necessity’ requirement is not merely a formality; it is a core part of the merits of a PI application.⁴ The applicant must explain substantively why interim relief is needed now.

At the hearing, the panel’s questions were indicative of this approach. For example, how much market evidence must an applicant gather to support necessity and balance of interests? How, concretely, does a party prove price erosion, or its likelihood, rather than simply asserting it? The respondents argued that price erosion must be evidenced and is not automatic, especially where there are already players competing in the market. 

In the decision, the CoA recognised structural market effects as legitimate harm that may justify the necessity of a PI. The Court then applies this reasoning in a fact-sensitive way to the specific market in question (continuous glucose monitoring), including by giving weight to reimbursement market dynamics where tender outcomes and long contract cycles can make price recovery particularly difficult. The fact that other competitors were on the market already did not remove the risk of price erosion if the alleged infringer were to offer its products at substantially lower prices.

For PI applicants, the message is clear: necessity is a potential standalone ground for rejection, and its assessment should be evidence-based. It cannot be an afterthought or be treated as an assumption. For respondents arguing that the market is stable and price erosion speculative, expect to be pressed on what the evidence shows. 

3. Remedies: broad enough to be effective, disciplined enough to avoid opportunism

The Court’s analysis of PI scope demonstrates a pragmatic balancing exercise. It refused to extend the PI to a legacy product line that had been on the market since 2023 and against which the patentee had not acted, treating that history as a hard limit on the scope of provisional measures.⁵ At the same time, it confirmed that generally worded injunctions may be appropriate once infringement is held to exist in relation to one product, so that minor changes or re-labelling do not defeat enforcement.⁶

4. The cross-appeal admissibility point: academic in substance, important in mechanism

The CoA’s decision provides a procedural clarification that will affect how parties frame PI appeals. In this case, the respondents had prevailed at first instance, and the PI was rejected, but not on the basis of urgency. The first instance decision found that the applicant had acted with the necessary urgency, and the respondents filed a cross-appeal directed at this finding. However, the Court rejected this as inadmissible, stating that a cross-appeal under R. 237 of the Rules of Procedure (RoP) is only available to a party adversely affected by the decision’s result.⁷ It cannot be used merely to ‘fix’ reasoning in an otherwise favourable outcome. 

In this case, the distinction was largely academic. Urgency is a ‘necessary’ element of the Court’s PI analysis on appeal under R. 211.4 RoP ⁸,so the Court will assess it in any PI appeal as a matter of course. The same reasoning ought to apply to any decision on necessity and the balancing of interests where the PI is ultimately rejected. 

The practical issue arising from the decision is not whether non-determinative findings of the first instance can be argued against, but where those arguments should sit procedurally. The answer appears to be as part of the respondent’s defence on appeal, not as a ‘reasoning-only’ cross-appeal.

5. Validity in PI proceedings: no “shotgun” attacks - tell a substantiated “would” story

The Court’s most general procedural message on validity is its rejection of “shotgun”⁹ obviousness attacks. In PI proceedings, defendants must follow the UPC’s test for inventive step clearly and substantiate why the patent is more likely than not invalid. This includes an explanation of why the skilled person would combine references, and how the invention is obtained in an obvious manner. Simply filing numerous documents with isolated quotations is not enough. In this case, the Court refused to consider inventive step attacks that had not been fully articulated.¹⁰ The Court’s detailed reasoning in the decision focuses on only the specific combinations of documents raised in the appeal to the appropriate level of detail. 

That approach aligns with the Court’s emphasis on motivation/pointer. If the reason the skilled person would take the next step from a realistic starting point cannot be articulated clearly, the attack is unlikely to succeed. Parties should focus on selecting promising inventive step attacks and articulate these in full. 

6. Claim construction: functional coherence may prevail over literal construction for user interface features

The appeal turned on claim construction. The Court rejected The Hague Local Division’s narrow approach and focused on what the claim’s technical teaching achieves: a time correlation between monitored glucose levels and user-entered events. The order makes the point explicitly: event icons need not sit on or near a graph line or within a particular ‘panel’. They may be placed anywhere, provided time correlation is established so that the relationship is recognisable.¹¹

This is a helpful signal for disputes relating to user interfaces (UIs). Where the claim language does not mandate a specific screen geometry, the Court is prepared to read the claim in a way that preserves the underlying functional concept, rather than confining it to a particular layout. This reasoning may also be transposed into an equivalence argument, although that was not run in this case. 

7. Infringement: indirect infringement may be key in app + device ecosystems

The Court’s infringement analysis is a template for cases where the accused embodiment is an ecosystem. Because the defendants did not supply the mobile phones (receiver units), there was no direct infringement of the system claim on the Court’s analysis. The Court granted relief on the basis of indirect infringement instead: supplying essential means (hardware components and the app) intended to put the invention into effect in the UPC territory.¹²

For PI strategy, this matters. Where claims include third‑party components, applicants should plead indirect infringement early and cleanly, rather than treating it as a fallback.

8. Technical effect / excluded subject matter: EPO alignment inside a UPC inventive step framework

The decision is also notable for its treatment of alleged excluded subject matter and the technical character of an invention. The Court expressly endorses the EPO’s G 1/19 and COMVIK principles: a feature should not be excluded from inventive step analysis merely because it is non‑technical in isolation; interaction can make it contribute to technical character.¹³ Applying that approach, the Court treats the UI features as technical (or at least contributing to technical character) because they facilitate human‑machine interaction by enabling users to correlate events with glucose levels.¹⁴

What makes this especially interesting is how it integrates with the UPC’s inventive step analysis, which differs from the EPO problem-solution approach. In EPO practice, the G 1/19 and COMVIK principles are the gatekeeping step that determines which claim features ‘count’ at the problem–solution stage, following which the objective technical problem is formulated. The UPC does not adopt the EPO’s formal problem–solution method, but Abbott v Sinocare indicates that the G 1/19 / COMVIK filter can still be ‘plugged in’ upstream of the UPC’s holistic inventive‑step assessment to then reach a decision on inventive step. 

Conclusion: what to take into your next UPC PI case

Abbott v Sinocare is best read as a decision about procedure. It indicates that the Court of Appeal will: (i) treat urgency as a question of evidentiary readiness and diligence, not merely elapsed time; (ii) demand evidence on necessity and balance of interests, while recognising structural market-specific harms like price erosion; (iii) approach claim construction with an eye to functional technical teaching; and (iv) apply the EPO’s technical character approach to excluded subject matter within the UPC’s own inventive step framework.

For practitioners, the message is encouraging but exacting. Encouraging because the Court is prepared to grant powerful interim relief where the evidential and market case is properly made. Exacting because it will ask the difficult questions. How long did it really take, what did you really need to know, and what does the market evidence really show? In UPC PI practice, those questions are delineating between cases deserving of urgent enforcement and ordinary litigation being presented as an emergency.