Evolution of patent case law: UPC Court of Appeal overturns the first preliminary injunction

Everyone knows that patent litigation is both expensive and lengthy, right?

Well, maybe not any longer, if early indications from the Unified Patent Court (UPC) are to be believed.

The newly established UPC came into being on 01 June 2023, less than 10 months ago, and yet we have already had our first substantive decision from the Court of Appeal (CoA).

One of the primary aims of the UPC was to make the process of patent litigation quicker (and cheaper) with the court setting itself admirable (if perhaps optimistic) goals of hearing cases within 12 months and having a 1-day trial.

In the latest update of case figures (correct to 29th February 2024), the court of first instance had received 274 cases – 96 infringement actions, 129 counterclaims for revocation (against 39 of the infringement actions); 23 provisional measures; 24 revocation actions; and 2 “others”.

Although it is still early days, it would appear that the 12-month goal may be achievable in the short to medium term, at least. Indeed, the Court of Appeal has been issuing Orders since October 2023, the most recent of these being the highly anticipated first substantive Order in the NanoString Technologies v 10x Genomics case.  

Background to the case

The case in hand relates to patent EP4108782, owned by Harvard and exclusively licensed to 10x Genomics, covering methods and compositions for detecting analytes (e.g. DNA) in cell and tissue samples. 10x Genomics enforced the patent against NanoString, alleging that NanoString’s molecular imaging devices (CosMX SMI) and detection probes infringe the patent, both directly and indirectly.

The initial case was heard by the Munich Local Division (LD) during which 10x Genomics requested a preliminary injunction against NanoString. The LD was sufficiently satisfied that the patent was novel and inventive and provisionally considered the patent to be infringed. The preliminary injunction was granted in what was another first for the Court, citing that both the urgency and irreparable harm requirements were met in the present case.

Appeal

The UPC CoA reassessed the LD’s decision on validity of the patent on appeal and, as would be expected, the issue of claim interpretation was central. The patent in question provides a solution to the problem of testing biological samples for the presence of multiple different molecules simultaneously (rather than running multiple tests). In applying Art 69 EPC (which dictates the extent of protection conferred by the granted patent), the CoA considered that the claims are the decisive basis for determining the scope of protection but that the description and drawings should be used as an aid.

In the present case, the granted claims specify that the method is for analysing “cell and tissue samples” and the description contains examples processing cell and tissue samples where nucleic acids and proteins are isolated from the sample. Both the LD and the CoA construed the sample to be structurally recognisable as a cell or tissue. Furthermore, the CoA determined that the patent claims differentiate between the sample and the “plurality or nucleic and protein analytes” and that the examples in the description (which isolated the analytes from the samples) did not broaden the definition of samples. In its reasoning, the CoA states that the description’s definition that an analyte “can be a component of a whole cell, a tissue or a body fluid, a cell or tissue extract, a fractionated lysate thereof or a substantially purified molecule” was irrelevant in construing the claim.

The CoA disagreed with the LD’s construction of steps c) and d) of claim 1 which describe at step c), incubating analytes with detection reagents to allow the reagents to bind, and at step d), detecting the reagents. The LD had required the reagents to remain bound to the analytes during step d) but the CoA did not consider this a requirement, nor did the claim preclude cycling of incubation and detection.

To this point, 10x Genomics’ argument that cycling would be considered impractical by the skilled person was found unconvincing by the CoA due to the time scales for incubation ranging from 30 seconds to 48 hours. Given the range, the CoA considered that the skilled person would not rule out multiple incubations.

In view of the different interpretations of the claims, the CoA was not convinced that the validity of the patent was established to such a degree that it justified the preliminary injunction. Although the CoA did agree with the LD that the claims are novel in view of the prior art (Goransson et al 2009), the CoA’s consideration that the cycling steps in the prior art could fall within the scope of the claim changed the question of inventive step. In doing so, the inventive step question became whether the prior art method could be applied directly to cell and tissue samples.

The CoA found it would be obvious for the skilled person to modify Goransson’s method to use tissue and cell samples, particularly in view of the prior art referring to detection methods carried out on cells in situ. The link established, the CoA considered that there would be no reason why a skilled person would not apply the prior art to the in situ environment in a cell or tissue sample. The CoA was then not convinced by the patentee’s argument that the skilled person would need to overcome significant difficulties to achieve that.

The CoA also mentioned that the patent does not provide guidance on how to apply the method to in situ methods thereby leading to a lack of sufficiency.

In concluding, the CoA stated: “it is more likely than not that the patent at issue will prove to be invalid in proceedings on the merits due to a lack of inventive step, there is no sufficient basis for the issuance of a preliminary injunction".

10x Genomics was ordered to pay costs of €7 million. In the midst of this battle, NanoString Technologies has filed for chapter 11 bankruptcy in the USA at the beginning of February.

It is interesting to note that there are also EPO opposition proceedings ongoing in relation to the patent in question (EP 4108782). Watch this space for how that goes.