Wednesday 12 July 2017 was probably a pretty ordinary day for most UK citizens. The sun shone across large parts of the British Isles as Andy Murray was defeated in the quarter-finals of the Men’s Singles at Wimbledon. However, to most UK patent lawyers, this day will be etched forever in their minds as the day that the law on the construction and infringement of patents changed beyond all recognition with the ruling of the Supreme Court in Actavis v Eli Lilly.
It will not be long until the second anniversary of the Actavis decision is reached. This short article will look at some of the important rulings in the period since the decision and address some of the key issues that remain outstanding.
In the 13 years between 2004 and 2017, patent practitioners had understood that the English patents courts fulfilled the obligations prescribed by article 69 of the European Patent Convention (EPC) and its Protocol (as revised) by addressing a single question. This question, laid down by Lord Hoffmann in the Kirin-Amgen case was quite wordy but in essence simple to understand: “What would the skilled person have understood the patentee to be using language of the claim to mean?” This question, sometimes referred to as the “bedrock” question, did not permit an acontextual, literal approach to the claims but rather required a purposive interpretation based not only on the specification of the patent but additionally, and sometimes more importantly, on the context of the teaching of the patent against the backdrop of the common general knowledge. Nevertheless, the claim was king and if the skilled person would have understood the patentee to have limited its claim in some way, it was not the job of the court to look beyond the claims to ascertain what the patentee could have claimed or to examine whether alternative products and processes, which achieved the same outcome as the patented invention, were equivalent. As Jacob LJ commented in 2009 when applying Kirin-Amgen: “It follows that there is no general doctrine of equivalents.” Although they were never officially abandoned, the so-called Improver questions, prescribed by Hoffmann J (as a first instance judge) in the case of that name in 1990 and faithfully adhered to by many patents judges in the 1990s and early 2000s, by and large fell into disuse.
The facts of the Actavis case are well known. In essence Lilly had a patent that claimed the use of the disodium salt of pemetrexed together with vitamin B12 for the treatment of cancer. Actavis wanted to sell non-disodium salts of pemetrexed, such as the dipotassium salt to be used with vitamin B12 for the treatment of cancer. At first instance and on appeal, very experienced, specialist English patent judges considered there could be no direct infringement – the claim was to pemetrexed disodium and the skilled person would be in no doubt that the dipotassium salt was a different thing.
The UK Supreme Court took a different view. Giving the leading judgment, Lord Neuberger held that the Kirin-Amgen approach did not truly implement the requirements of article 69 EPC and that extensive revisions to the UK approach were required. Boiled down to its basic components, the approach prescribed in Actavis was as follows:
Does the variant infringe any of the claims as a matter of normal interpretation?
If not, does the variant nevertheless infringe because it varies from the invention in a way or ways which is or are immaterial? This is to be determined by asking three questions:
• 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, ie, 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?
English law operates according to a quite rigid system of precedent. So it is not open to lower courts to ignore the ruling of the Supreme Court in Actavis. Indeed, they must apply it unless and until the decision is overturned or amended by a subsequent decision of the Supreme Court.
In the 18 months or so since July 2017, the English patents courts have considered the Actavis ruling one way or another about a dozen times. Probably the most important ruling on the issues was the decision of a strong Court of Appeal in Icescape v Iceworld. Having found the patent invalid in this case, the comments of the Court were strictly obiter. However, the judgment makes it absolutely clear that Actavis represents a completely new approach to the question of the construction and infringement of patents under UK law. As Lord Kitchin commented in this case: “It is, in my view, clear that [the Actavis] approach is markedly different from that which the courts in this country have adopted since Catnic.”
The Court of Appeal case in Icescape and the following cases provide some guidance as to how to apply this “markedly different” approach, but this is ever evolving as more cases grapple with these issues. It seems generally accepted that the first stage, considering whether the variant falls within the claims on a “normal interpretation”, involves a purposive construction. However, this appears not be the same purposive construction as applied before the Actavis case, which has been criticised as it conflates the two issues – the normal construction and equivalents, when each should be considered separately. Instead, in Regen Lab v Estar, HHJ Hacon suggests that “the normal construction of the claim is confined to interpreting the words of the claim in the context of the specification”.
When considering variants and applying the second part of the test the focus is now on the problem underlying the invention and the patent’s inventive core. Such an approach can involve ignoring claim language, even in characterising portions of the claim, if such integers do not play a role in this inventive core. So if, for example, a claim is to a process with four steps, and the inventive concept (or the “clever bit”) is in the fourth step, the focus of the equivalence questions will be on that step. To take a further example, if a patentee has invented a car that runs on ethanoic acid, and claims a blue car that runs on ethanoic acid, then a red car running on ethanoic acid may well infringe if the skilled person would regard the colour of the vehicle as irrelevant to the inventive concept of the patent. Even if the specification stated “the car of my patent is blue”, this might not rule out the red car from infringing. Thus, the ruling of the Court of Appeal in STEP v Emson is probably unsafe to rely on with confidence these days.
The following decisions have provided important guidance on some points of detail.
Early in 2018, the Court of Appeal was tasked with looking at claims containing numerical ranges in Jushi v OCV. The judgment of the Court makes clear that numerical range integers are not to be treated any differently from other sorts of integers. In particular, they are not to be regarded as bookends incapable of flexibility both using normal interpretation and on equivalence. This was underlined in Regen v Estar where HHJ Hacon held that a solution of buffered sodium citrate solution at a concentration of 0.136M was equivalent to a claim which stated that the concentration should be 0.10M.
The third Actavis equivalence question is often the most challenging when considering infringement by equivalence. It might be possible to find out by speaking to a suitably qualified expert whether the variant achieves substantially the same result in the same way as the invention and whether this would have been obvious at the priority date knowing that the variant does this. In reality, given Lord Neuberger’s tweak to it, the second question will rarely be answered in the negative. But how does one assess whether the skilled person would 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? How is the skilled person supposed to reach into the mind of the patentee to ascertain his or her intentions?
Clearly, the fact that language of the claim does not on any sensible reading cover the variant will not lead to a negative response to this question; if this were the case then variants would never infringe. Rather, as has been made clear by the decision of HHJ Hacon in Regen v Estar, the focus with this question is asking whether the patent integer corresponding to the variant part is an essential part of the inventive concept because, if so, the patentee may well have intended strict compliance with this part of the claim. Here, much like the other two steps in assessing equivalence, it is the “the clever bit” of the patent that is key and not the overall product or process.
Back in 2010, in Occlutech v AGA the Court of Appeal held that where a patentee had used the word “clamps” in its claim but had described other methods of holding components together in the specification, the patentee would be taken to have included a deliberate limitation in its claims. Whether this principle, which is applied in the German courts, could still hold sway in the UK remains unclear. In L’Oréal v RN Ventures Henry Carr J held that it was unnecessary to decide whether this doctrine of deliberate selection should be applied to equivalents in UK law. However, he did go on to find obiter that there would be no infringement under equivalence due to a negative answer to the third Actavis question, as the patentee had clearly considered the potentially infringing variant but deliberately chosen not to claim it.
One issue that has not been directly addressed by the courts post Actavis is the impact on validity. One question which arose in the minds of many practitioners as soon as the Actavis decision was handed down was whether a claim could be invalidated by a piece of prior art that did not anticipate or render obvious the claim on a normal/purposive analysis, but which would if equivalents to the prior art were permitted to factor into the assessment. This issue has not been fully resolved although in Generics v Yeda, Arnold J opined that it was not sufficient for anticipation if the prior art disclosed subject matter that would infringe the claim applying the doctrine of equivalents. Arnold J did not, however, address the issue of inventive step in his analysis in this case. Other judges, such as Birss J in Liqwd Inc v L’Oreal, seem to take the view that the wider claim scope for infringement may also need to be taken into account for validity.
Related to this is whether the UK does or should embrace a doctrine of similar scope to the German Formstein defence. In essence the Formstein defence provides that if an accused product would fall within the scope of a claim under the doctrine of equivalents, but the equivalent would have lacked novelty or been obvious in light of the prior art at the priority date, then the accused product is deemed to fall outside the scope of the claim. Thus, in a similar way to the ensnarement principle in the US, the scope of the claim can be shaped by the prior art. In Technetix v Teleste, HHJ Hacon considered this issue, obiter, and concluded that he could not rule out the possibility that the UK Supreme Court may in due course introduce a Formstein defence into the law. The judge opined that if UK law embraced a Formstein defence, Teleste would have been able to take advantage of it.
In Actavis, Lord Neuberger emphasised that the UK courts should be cautious in adopting a doctrine of file wrapper estoppel. He said that recourse should be had to the prosecution file only in two circumstances: “where (i) the point at issue is truly unclear if one confines oneself to the specification and claims of the patent, and the contents of the file unambiguously resolve the point, or (ii) it would be contrary to the public interest for the contents of the file to be ignored.” The lower courts have followed this approach. In L’Oréal v RN Ventures Henry Carr J held:
It should be emphasised that reference to the prosecution history is the exception, and not the rule . . . Parties should think carefully in future before incurring additional costs in arguing about the prosecution history.
Nearly two years on from the decision, the dust is beginning to settle with respect to the new principles laid down in Actavis. Although they are mostly obiter, rulings from the lower courts have helped to clarify a few issues not fully explored by the Supreme Court. As many readers will know, Kitchin LJ was promoted to the Supreme Court in autumn 2018. Sooner or later, he will be given the opportunity to review and perhaps revise Actavis. It is obviously not possible to say what Lord Kitchin and the other judges hearing such a case would decide. However, a distinct possibility is that a middle way between the extremes of Kirin-Amgen and Actavis will be found. It seems to the authors that if a broad doctrine of equivalence is to be embraced, suitable checks and balances should be adopted, including some kind of Formstein defence.