UPC appeals after three years: Early indications on appeal success rates

The UPC has now been running for three years. With first instance proceedings typically taking around one year and appeals taking a similar period, now is perhaps the first opportunity to, cautiously, look at a sufficiently mature body of Court of Appeal decisions and seek to identify patterns.

Appeals in the UPC, as for most courts, are to a material extent self-selecting. Parties are less likely to appeal clear cut decisions, while cases with a realistic prospect of success are more likely to proceed. Even though the UPC requires no permission to appeal final decisions, the fact that appeals at the UPC do not automatically have a suspensory effect (as in the EPO, for instance) or offer material procedural advantage, would reduce incentives for parties to file more ‘ambitious’ appeals. Moreover, additional disagreement between first instance and appeal is also unsurprising at this stage, while the UPC remains a relatively young court and its case law is still settling. Against that backdrop, an overturn rate of just over half for both substantive appeals and appeals on the grant of a preliminary injunction (56% or five out of nine for substantive appeals; 56% or 9 out of 16 for appeals on preliminary injunctions) feels broadly in line with what one would expect.

Within both categories of appeals, decisions are being overturned for a range of reasons. Some notable developments are as follows:

• On provisional injunction appeals, the Court of Appeal has generally conducted a detailed substantive review of the questions of novelty and infringement, mirroring what seems to have become common at first instance as well. It is not at all unusual for judgments to include 20-30 pages (or more) on the assessment of validity and infringement.  However, the consideration of urgency has also been a source of reversal on appeal.  Given that suspensive effect is not generally granted for appeals of provisional injunctions, the value of an appeal over-turning a provisional injunction seems rather limited since in the meantime there is time for the first instance substantive decision to be either already available or expected soon.
• For substantive appeals, again there is a range of reasons why the appeal outcome varied from first instance. Sometimes (as also in the first successful PI appeal of Nanostring v 10x Genomics) a difference in claim construction leads to a difference in conclusion on either validity or infringement.
• The Kodak v Fujifilm appeal was particularly interesting as much of the substantive analysis on appeal differed little from that at first instance, but the Court of Appeal recognised a prior user right in respect of Germany that the first instance had not considered substantiated, while in respect of the UK the defendants were considered not the parties responsible for the UK acts.
• On the patentee-friendliness of the Court of Appeal, it is interesting that despite over half the appeals succeeding, the overall rate of success rate of patentees is unchanged following appeals. In five of the nine first instance decisions that were appealed, there had been a finding that a patent was valid and infringed. Following appeal, there was found to be a valid and infringed patent in five of the nine cases. This would suggest that overall the Court of Appeal is no more pro- or anti-patentee than the first instance courts.