On 19 June the Court of Appeal gave judgment in Jushi v OCV  EWCA Civ 1416, an appeal from the Intellectual Property Enterprise Court (IPEC). The case concerned a patent for glass fibre strands for use in fibre-reinforced composite materials. The main issue on the appeal was the validity, or otherwise, of a claim which included numerical limits. There were several ranges in the claim, the most relevant one being a requirement that the ratio of calcium oxide to magnesium oxide be "≤ 2, preferably ≥ 1.3".
In the wake of last year's Supreme Court judgment in Actavis v Eli Lilly  UKSC 48, one issue that has remained unresolved is whether the scope of protection of a claim for infringement, including by equivalents, remains the same as the claim's scope validity. In Mylan  EWHC 2629 (Pat) Arnold J held obiter that the previous symmetry of infringement and validity no longer held. However, the issue has not been addressed in detail since.
Separately, following Actavis, many commentators have discussed how the principle of equivalents might be applied in the context of a claim including numerical limits. In particular, in considering the three-part test provided by the Supreme Court for guidance on how to assess whether an alleged infringement is equivalent to what is claimed, numerical limits seem liable to lead to intense arguments on two fronts. First, whether an alleged infringement slightly outside the range expressed in the claim is achieving the same result and obviously doing so in the same way. This arises because, in most cases, numerical limits in patent claims do not represent a 'hard-stop' or 'cliff-edge' at which the relevant advantage or function ceases. Second, whether the use of numerical limits in the claim may indicate that the patentee intended that strict compliance with the words of the claim was an essential requirement of the invention. These two arguments would tend to work in opposite directions in a particular case, one expanding and the other limiting the scope of protection depending on the parties' respective needs in the litigation.
It might also have been thought at one time that numerical limits would present an interesting case for the assessment of the "normal interpretation" of a patent's claim, since some observers considered that the Supreme Court's "normal interpretation" was to be equated with "literal meaning". What could be easier to interpret literally than a number? However, since the Actavis decision the Patents Court, with apparent concurrence by the Court of Appeal, has chosen to apply "purposive construction" to the issue of "normal interpretation". Following this approach numerical limits present no special easy case and must be interpreted according to the usual principles, albeit applied in a particular way, as had been discussed by the Court of Appeal not long before Actavis in Convatec  EWCA Civ 607 and Napp  EWCA Civ 1053. In both those cases, the Court held that the numerical ranges in those patents should be understood as referring to "whole number rounding", i.e. ±0.5 from the number stated. Thus in Convatech a limit of "between 1% and 25%" was construed as extending from 0.5-25.5% and similarly in Napp a limit of "10 to 15%" from 9.5-15.5%.
In Jushi, the claimant sought to revoke the patent in part on the basis that it was anticipated by a piece of prior art, referred to as Neely, discussed in the patent's specification. One of the examples from Neely was reproduced in the patent's description which described favourable mechanical properties of the glass fibres of the invention compared with those of Neely. The Neely fibres had a calcium oxide to magnesium oxide ratio of 2.14. Thus, if the patent's limitation of "≤ 2" were to be interpreted using whole number rounding 2.14 would fall within it. In this case, in particular because of what the Court of Appeal considered to be an apparent intention by the patentee to contrast the composition of Neely with that of the invention, the Court interpreted the numerical ranges in the claim as "exact, and not meant to be broadened by whole number rounding". This narrow construction, combined with other factors, led the Court to reject Jushi's anticipation argument.
What is apparent, however, is that the Court of Appeal did not discuss any arguments about equivalents. It may be that Jushi felt unable to run any such argument on the facts, or that it felt precluded from doing so in the context of a validity argument. Whatever the reasons, a case which, at least from the outside, would appear to have offered an opportunity for debate about the application of the principles of equivalence contains no reference to the Actavis decision at all. Whether this is merely a reflection of the parties' choice of arguments or whether it reflects, at least in part, a desire by the Court of Appeal to minimise the change in approach to the majority of patent cases notwithstanding Actavis is not an easy question to answer.