This article was first posted on the September- October issue of the Patent Lawyer Magazine.
In July 2017, just as many patent practitioners in the UK were thinking about trying to enjoy a little summer sun, the Supreme Court caused something of a splash by handing down its judgment in Actavis v Eli Lilly1. This judgment introduced a doctrine of equivalents into UK law – something which until then, many had thought the reserve of our continental (and transatlantic) cousins.
Two years on, now that summer is with us again, the authors have collated a few bite-size updates on how the case law has developed since the Supreme Court’s judgment, picking out a few tasty morsels and identifying a few areas where further clarification is needed.
Setting out the picnic blanket – what is the test?
Lord Neuberger explained in Actavis v Eli Lilly that when assessing patent infringement, there are two issues to consider: first, does the variant infringe any of the claims as a matter of “normal interpretation” (issue 1); and, if not, does the variant nonetheless infringe because it varies from the invention in a way or ways which is or are immaterial (issue 2)? If the answer to either issue is “yes”, there is infringement; otherwise, there is not. Whereas the previous approach under Improver v Remington2 was used as a tool to construe the claim, the new approach is to construe the claim and then determine whether, notwithstanding the fact that the variant might not fall within any sensible construction of the claim, the scope of the claim nonetheless covers it.
The three questions that act as guidelines for assessing issue 2, (the “Actavis questions”) are:
- Notwithstanding that it is not within the literal meaning of the relevant claim(s) of the patent, does the variant achieve substantially the same result in substantially the same way as the invention, i.e. the inventive concept revealed by the patent?
- Would it be obvious to the person skilled in the art, reading the patent at the priority date, but knowing that the variant achieves substantially the same result as the invention, that it does so in substantially the same way as the invention?
- Would such a reader of the patent have concluded that the patentee nonetheless intended that strict compliance with the literal meaning of the relevant claim(s) of the patent was an essential requirement of the invention?
Following the Supreme Court’s judgment, there has been significant debate over what Lord Neuberger meant by “normal interpretation” (issue 1). The confusion arises from the fact that when referring back to issue 1, instead of referring to “normal” interpretation, he referred instead to the “literal meaning”. Some Patents Court judges have held that issue 1 must be directed to the purposive interpretation of the claim (see, for example, Arnold J in Mylan v Yeda3 and Henry Carr J in Illumina v Premaitha4). However, as noted by Richard Meade QC (sitting as a Deputy Judge) in Fisher & Paykel v ResMed5 it was at least arguable that the reference to the “literal meaning” in the Actavis questions implied a narrower construction for the purposes of issue 1.
When this issue came before the Court of Appeal in Icescape v Ice-World6, Lord Kitchin (giving the leading judgment) had “no doubt” that issue 1 involves purposive interpretation, notwithstanding Lord Neuberger’s reference to the literal meaning in issue 2. This means that previous case law on the principles of purposive construction (excluding any issues of equivalents) appears still to apply. However, it does beg the question as to whether, in assessing “normal”, i.e. purposive construction, one is permitted to ask the old Improver v Remington (or “Protocol”) questions.
Circulate the menu – what is the invention?
The term “invention” is used in all three Actavis questions. The first Actavis question purports to include a definition and states “the invention, i.e. the inventive concept revealed by the patent”. However, there appears to be uncertainty about exactly how to identify the “invention” and how this relates to the claims.
In Fisher & Paykel v ResMed, Richard Meade QC observed that identifying the inventive concept “may not be an easy exercise (and it was often difficult under Improver).” In that case, Richard Meade QC suggested that he “should pay attention to the two-part form of the claim, and assume that the invention relates importantly but not exclusively to the characterising part of the claim.” Interestingly, in contrast, in Icescape v Ice-World the inventive concept was not found in the characterising portion of the claim.
In Regen v Estar7, HHJ Hacon opined at  on the importance for the purposes of the third Actavis question (in contrast to the third Improver question) of differentiating between the invention as a whole, as opposed to the inventive concept. He states that “the distinction between the invention as a whole and the inventive concept matters. The invention is that which is claimed, see s.125(1) of the Patents Act 1977. I take the inventive concept or core of the invention to be the new technical insight conveyed by the invention – the clever bit – as would be perceived by the skilled person. This will be assessed by reference to the specification and the evidence.” He concludes that the third Actavis question “requires the court to consider whether the relevant integer, that corresponding to the alleged equivalent, would have been regarded by the skilled person as an essential part of the inventive concept.” It appears therefore that, despite the wording of the Actavis questions, it is the inventive concept of each claim that must be considered (when read in light of the specification).
In BDI Holding v Argent Energy8, HHJ Hacon summarised the various contexts in which the Court has considered the formulation of an inventive concept. The judge noted that in Actavis v Eli Lilly, Lord Neuberger held that the “inventive concept” is synonymous with the “inventive core” of the claim, and is to be ascertained by focussing on the problem underlying the invention. HHJ Hacon also noted that in Pozzoli v BDMO9, in the context of the structured approach to assessing obviousness, Jacob LJ stated that the inventive concept of the claim in question must be considered, rather than deriving the inventive concept from the specification as a whole. HHJ Hacon went on to opine that the two approaches do not differ by much (if at all). In BDI Holding v Argent Energy the judge went on to identify the inventive concept of the claim by reading the specification through the eyes of the skilled person. Following this approach, the cited prior art has no role to play in determining the inventive concept.
In Marflow v Cassellie10, HHJ Hacon considered the Defendant’s argument that the claimed method achieved a particular advantage, and as the variant did not achieve this advantage, the variant did not achieve the same result as the inventive concept. The judge disagreed that the alleged advantage was relevant to the inventive concept. However, this leaves open the question, noted by the judge (at ), of whether “an advantage consequent upon the use of an inventive concept is invariably a ‘result’ achieved by the inventive concept” and “if so whether the variant must achieve all the advantages of the inventive concept”.
Being inevitably very fact sensitive, this will be surely be a ripe area for debate in future cases.
Did anyone chill the bubbly – the relevant date of the assessment
In Eli Lilly v Genentech11, Arnold J considered the requirement for the variant to achieve substantially the same result in substantially the same way (Actavis question (i)). The judge considered this question based on “current evidence” (see ). This suggests that the relevant date for assessing whether the alleged infringement achieves substantially the same result in substantially the same way is the date of infringement. However, this in itself raises a yet further interesting question: in a fast moving art, what date is considered when infringement continues over a period of years before the action is tried?
The relevant technical field in Eli Lilly v Genentech had moved on considerably since the priority date of the patent, in these circumstances, Actavis question (i) is considered with the benefit of understanding how the infringement works. This is consistent with the approach to Actavis question (ii), that the skilled person is taken to know that the variant achieves substantially the same result as the invention.
Surveying the spread, or, assessing the alleged infringement as a whole
In Regen v Estar, the judge decided that where multiple integers of a claim are not infringed on a normal construction of the claim, the correct approach is to consider whether the variant as a whole is equivalent to the claim, rather than considering each of the integers separately.
This approach is consistent with focussing the analysis on whether the alleged infringement achieves the inventive concept of the claim (as discussed above).
How large is a slice of quiche? Numerical ranges
Also in Regen v Estar, the issue of numerical ranges was considered. The claimed method specified the use of 0.10 M sodium citrate and the question was whether the use of 0.136 M sodium citrate would infringe the patent. HHJ Hacon found that the molarity of the sodium citrate was not essential to the inventive concept, as there was not a sufficiently clear indication in the specification or the CGK that the skilled person intended strict compliance was required. He therefore found that there was infringement.
The judge also provided an interesting comment on the way the numerical limit is set out. While the patentee’s choice of 0.10M (rather than 0.1M) was relevant to the normal interpretation of the claim, it was not relevant to the question of equivalents. This again emphasises the clear focus, as referred to by Lord Neuberger at  of Actavis v Eli Lilly, on “the inventive core”, or the “inventive concept” of the patent.
Who brought cheese and pineapple? The Court remains skeptical of the prosecution history
Reference to the prosecution history has been made in several cases following Actavis v Eli Lilly. In Icescape v Ice-World, Icescape failed to garner much interest from the Court in relation to Ice-World’s response to an objection raised by the EPO. In Illumina v Premaitha, Premaitha was similarly unsuccessful in its attempts to rely on a limitation which had been added to the claim by the patentee during opposition to avoid objections of insufficiency. There, the judge held that the limitation did not exclude the type of method practised in Premaitha’s allegedly infringing product. In L’Oreal v RN Ventures12, the judge held that there was no obligation on the patentee to correct any misunderstanding of the Examiner about the scope of the claims. The Defendant also sought unsuccessfully to rely on the prosecution history in Regen v Estar.
The judgments in these cases indicate that, whilst the Supreme Court’s judgment in Actavis v Eli Lilly expressly leaves open the possibility of referring to the prosecution history, the Patents Court has maintained its skepticism. That is not to say it will never be relevant, but likely only in limited circumstances.
Like strawberries at Wimbledon, is the three-stage test mandatory?
Whereas Lord Neuberger’s judgment appeared to give guidance on the issue of assessing infringement by equivalents, it appears to have been adopted by the Court of Appeal in Icescape v Ice-World as a mandatory test for assessing equivalents. This is indicated by the language at  and, in particular,  of Icescape v Ice-World. Lord Kitchin held that, in order to assess whether the variant differed in ways that were immaterial, one “must” address the Actavis questions.
Room for more? The validity gap and the status of a Formstein defense
Perhaps the most interesting issue is whether there is a role for equivalents in assessing validity, given that this was not discussed in Actavis v Eli Lilly.
Speaking at the University College London Institute of Brand and Innovation Law (“UCL IBIL”) event in November 2017 (a few months after he had handed down the judgment in Actavis v Eli Lilly), Lord Neuberger expressed his view on the issue. He suggested that where an aspect of an alleged infringement might otherwise be covered by equivalents, if that feature was in the prior art then it would not be covered by the claim.
It is not uncommon for parties sued for infringement and challenging the validity of a patent to allege that if the patent covers their product or process, it also lacks novelty or inventive step over the prior art and they are entitled to a defense to infringement. This is commonly referred to as a “Gillette defense” following the precedent for the approach, Gillette v Anglo-American13. The case law on the Gillette defense has also come into focus in the context of declaratory relief – there has been a recent slew of cases in which parties have requested declarations that their own product or process would have been obvious or not novel at the priority date (an “Arrow declaration”).
Unlike the UK, where the Gillette defense is not a stand-alone defense to infringement, the German courts have established the Formstein defense (from Case X ZR 28/25 Formstein GRUR 1986, 803). Judge Meier-Beck of the Bundesgerichtshof (the German Supreme Court) commented at the UCL IBIL event mentioned above that it “almost never” happens that an alleged infringer argues that they have a Formstein defense. However, in Technetix v Teleste14, having found the patent invalid, HHJ Hacon found (obiter) that the Defendant would have been entitled to a Formstein defense, if such a defense existed in English law. This was on the basis that the Defendant’s product would infringe under the Actavis questions, but not the normal interpretation of the claim, and that based on the expert evidence Teleste’s product would not be inventive over the common general knowledge.
In Technetix v Teleste, HHJ Hacon noted the principle of ensnarement which exists in the US (that equivalents cannot be applied to also ensnare the prior art) and commented that the existence of a similar defense under Dutch law indicates that it can work in a non-bifurcated system. Similarly, in Liqwd v L’Oreal 15, Birss J opined at  that he could see room for arguing that for validity purposes some account ought to be taken of the wider scope which is now allowed for the purposes of infringement.
It therefore remains unclear whether a non-inventive modification of the prior art could be held to infringe a patent following Actavis v Eli Lilly. Although, as HHJ Hacon commented in Technetix, and in line with Lord Neuberger’s comments at the UCL IBIL event, it would be “surprising” if the Supreme Court intended to abandon the principle in Gillette without specifically mentioning it.
In relation to a different aspect of patent validity, in Conversant v Huawei & ZTE16, the Defendants argued that the Claimant’s position on equivalents provided a squeeze on added matter. Arnold J disagreed with this and stated that infringement by equivalents cannot give rise to an added matter objection, since it is an argument about scope of protection and not disclosure.
Anyone for cake? The future
As Floyd LJ noted in Icescape v Ice-World “a number of issues remain unresolved”. Many patent practitioners have strong views on what the answers to these issues should be, but only time will whether any of them are correct. The authors await further developments with intrigue, and quite possibly a slice of cake