Key decisions in European patent disputes from the UPC – July edition

We cover some recent key decisions in European patent disputes, this time looking at developments in the Unified Patent Court (UPC)

Part 1 – Abbott v Sibionics - Added Matter considered as a ground for invalidity in UPC PI proceedings

Relevance of the decision

The UPC hears the validity of a patent in “normal” infringement actions (after a counterclaim filed by the alleged infringer), in central revocation actions, as well as in preliminary injunction proceedings (PI) proceedings. While the first substantive revocation action has only issued very recently (see below), there have been already a few PI decisions assessing the validity of the respective patent-in-suit. In the decision Abbott v Sibionics (UPC_CFI_131/2024, in the following “the decision”), the Local Division The Hague issued its first PI decision and found that the presence of added matter led to the rejection of the request for a PI, without assessing all the other arguments raised by the Defendants, including lack of urgency and lack of inventive step. Importantly, to come to this conclusion, the LD The Hague applied the so called “gold standard” of the European Patent Office, ie it followed EPO principles.
General Principles

According to Art. 65 (2) UPCA, presence of added matter is a ground for revoking a patent at the UPC. According to the ruling of the Court of Appeal of the UPC in 10x Genomics, for granting a PI, the Court must be convinced that there is a sufficient degree of certainty that the patent is valid. Such a sufficient degree of certainty is lacking if the court considers it on the balance of probabilities to be more likely than not that the patent is not valid (see item 2 and 2.1 of the decision). For acknowledging that the subject-matter of a claim is based on the original application, according to the gold standard of the EPO, it is required that such subject-matter is directly and unambiguously disclosed in the original application.

The reasons/decision

The present case relates to devices to devices for continuous glucose measurements. The underlying patent EP 3831283B1 was granted with claims which were not based on the claims as filed, but have been formulated on the basis of various parts of the original application, including a claim-like clause (clause 32) and various figures and the corresponding figure legends and descriptions in the general application. In such a case, the LD The Hague found that “a careful assessment is necessary” (see item 4.1 of the decision). The LD based this assessment on the EPO gold standard, because the parties relied on the case law of the EPO, but it also made clear that applying the gold standard is a matter of legal certainty for third parties and is necessary to ensure that patent proprietors do not benefit from an unwarranted advantage. The LD then carefully analysed the various passage of the description and found that claim 1 is the result of an unallowable generalization, because at least one specific feature, the presence of an elastomeric seal in the recess of the base portion of the enclosure of the claimed device, had been omitted in the claims as granted, although it was disclosed together with the other features of the claims in the original application. No other possible reasons for added matter or lack of patentability were, therefore, assessed by the court (see item 9 of the decision).

Take Home Message

This case is important for the future strategy in UPC cases for several reasons, including the following:

• Patentees need to be careful in UPC PI proceedings, because the UPC may thoroughly analyse the patentability of the claims at stake, including the assessment of formal issues like added matter.
• This distinguishes the UPC or at least the LD The Hague from the German courts, where added matter normally is not assessed in PI proceedings. Since the Dutch courts have a different tradition, this case is another proof that the “couleur locale” is important at least in the first period of the UPC and needs to be taken into account when establishing a UPC strategy. This applies both for plaintiffs and defendants.
• The UPC is willing to adopt the principles established by the EPO, but as other cases including the case discussed below show, that this is not automatically the case. Rather, the parties have to convince the court to (or not to) do so.

A Simmons & Simmons team is representing Sibionics in this case.

Part 2 – First substantive revocation outcome in the UPC’s Central Division

Significance of the decision

Revocation actions in the UPC must be brought in the Central Division, which, as of very recently, allocates such jurisdiction across 3 locations in Milan, Paris and Munich, which between them have the jurisdiction previously allocated to London. Allocation is made according to IPC patent classifications (and further criteria depending on the number of patents and/or classifications per patent):

• Milan: (A) Human necessities (e.g. pharmaceutical and medical device industries, agriculture, food tobacco, home appliances, clothing, wood & furniture, fire-fighting, sorts & gaming), without supplementary protection certificates.
• Paris: (B) Performing operations, transporting (D) Textiles, paper (E) Fixed constructions (G) Physics (H) Electricity, supplementary protection certificates
• Munich: (C) Chemistry, metallurgy, without supplementary protection certificates (F) Mechanical engineering, lighting, heating, weapons, blasting.

The UPC’s first revocation action was brought in the Munich Central Division by Sanofi against one of Amgen’s patents for anti-PSCK9 antibodies. Consistent with its jurisdiction both on Day 1 and now, this was allocated to Munich as (C) is the first IPC classification listed (out of the patent’s multiple classifications across (A) and (C)).

Initially, this case was of significant interest procedurally, as it was filed just ahead of Amgen’s corresponding infringement action under the same patent, which was also brought in Munich; but at the Local Division.

That race to file first and the subsequent interim disputes between the parties inferred a strong preference by Sanofi to have validity heard by the Central, and not the Local Division. That preference won out, including with the consequence that validity and infringement became bifurcated between the Central and Local Divisions.

As it turned out, the Central Division trial happened first and was an enjoined trial of both Sanofi’s original revocation action and Regeneron’s subsequent counterclaim (transferred from the Local Division). So the interest in this case soon turned to the substantive issues, and how the UPC would consider the multiple grounds of invalidity. The conduct of trial was also of interest, taking place in only 1 day, using slides/demonstratives exchanged in advance and without the cross examination of any of the multiple expert witnesses.

Furthermore, this case concerned a divisional patent in a patent family in which other members had been heavily litigated in global patent litigation, spanning multiple courts and examination authorities. Hence, this case was not only highly anticipated to see how the UPC would approach invalidity, but because of the insight it might provide into how the UPC would compare to other jurisdictions.

The reasons/decision

Both in the decision’s useful headnotes and the otherwise clearly and concisely drafted reasons, a number of key issues standout:

1. Construction – a growing body of UPC caselaw is adopting an approach closely based on Article 69 EPC and the Protocol of Interpretation. While this was a validity only decision, this would be expected to the same for infringement. There is a clear role for a patent’s specification, which is said may be the patent’s own “lexicon”. In this case, this resonates with the US decision finding Amgen’s claims to be incredibly broad and covering a “universe” of antibodies. By contrast, it suggests an arguably greater prominence of the specification in the UPC compared to the EPO; albeit interestingly timed versus the ongoing debate and referral now bound for the Enlarged Board in G1/24.

2. Priority – under this heading, the UPC followed the EPO’s playbook, requiring parity via a directly and unambiguously derivable “same invention”. Following the position under Abbott v Sibio (discussed above), this suggests that on more disclosure- or formality-type issues, the UPC plans to align strongly with the EPO ‘gold standard’ approaches. In this case, the priority attacks were rejected, thereby also neutralising novelty attacks based on intervening disclosures between priority and filing dates.

3. Inventive step (general) – most notably, following the course taken by the UPC Court of Appeal in overturning the preliminary injunction awarded to 10x Genomics against Nanostring on the same ground, the Central Division took a more holistic view of inventive step. Thus, the Problem-And-Solution test was not applied. In particular, the UPC did not identify any "closest prior art" and the Central Division held it was sufficient to start from prior art cited by the claimant if it was a "realistic starting point" (even if was not the most promising).

4. Obvious “next step” – Having adopted the general approach above for inventive step, the Central Division, again aligning with the UPC Court of Appeal in 10x Genomics, focussed on the presence of motivation (which it held was generated by the prior art) and thereafter, what it was obvious for the skilled team to do next. Once created, in this case by prior art that did not disclose antibodies, the motivation to pursue further, uninventive work appears to propel the skilled team with significant momentum because the decision (like 10x Genomics) refers to the need for “serious doubts” to dampen it. The UPC emphasised that the next steps need not “inevitably” reach the claim. While heavily fact dependent – as inventive step always is – it remains to be seen how this approach will develop at the UPC, in particular any nuances in the degree of motivation and how many steps it will drive if they comprise only uninventive or “routine work”. On the latter, the UPC appears to be closer to EPO approach in considering that antibody development for a known target (by 2007) does not justify an inventive step without something special. This creates the impression that, if an obviousness attack gains suitable momentum, the burden practically switches to the patentee to stop it. In this case, Amgen’s arguments based on complexities in the PCSK9 field were rejected.

5. Arbitrary claim features – as well as finding the development of PCSK9 antibodies to comprise obvious next steps, it was also an important part of the UPC’s reasoning that certain of the claim features were not relevant for inventive step – in particular the binding or “epitope” integer. In this respect, the UPC did not address the breadth of the claim or plausibility (and indeed, nor did the UPC go on to consider the insufficiency arguments at all). Instead, the decision focusses on the lack of link between the epitope and the function (i.e. the reduction of PCSK9 binding to its native target (LDRL)). Hence, the UPC dismisses the feature as arbitrary, in particular as a consequence of its broad construction (discussed above in terms of the UPC’s approach). Finally, while the medical use feature of the claim was not rendered arbitrary, the decision’s final remark is the lack of any argument that all that would be needed is the testing of candidate antibodies in a mouse model (i.e. those being further routine steps the skilled team could take).

Next steps

The decision merits close examination and the above points represent a summary of only some of the points. It seems highly likely that Amgen will appeal. And it has now been formally ordered that Amgen’s Local Division infringement action has been stayed in the meantime, cancelling its scheduled hearing dates for interim conference (this month) and trial (in October). The latter followed and was based on the agreement between all parties, which the UPC duly ordered based on its general case management powers and the likelihood of the patent being revoked in the now bifurcated proceedings. There is also the matter of Amgen being liable for the attorney fees of €1,375,000, understood to be due for each case (i.e. the revocation claim and the revocation counterclaim).

There are also parallel EPO opposition proceedings by Sanofi and Regeneron in which Amgen has only recently filed its Reply, and in which this UPC decision has also now been lodged. The EPO has confirmed this case will be accelerated due to the UPC proceedings. Notwithstanding the differences at the EPO based on Amgen’s reply, it will be interesting to see whether the OD reaches the same outcome as the UPC, including both if the OD agrees that the UPC has correctly applied the EPO’s gold standard approach for priority and if it considers the application of its now alternative problem-and-solution approach reaches the same outcome.