In 2017 the Supreme Court in Actavis v Eli Lily[1] established the doctrine of equivalents in UK patent law. Perhaps it would be more accurate to say that the Supreme Court reminded us that equivalents could infringe or brought the doctrine back to the UK after a 13-year period of exile since Kirin-Amgen in 2004[2], but it certainly seemed like a big deal at the time (and there were, I think, plenty of attorneys who had been practising long enough to see Kirin-Amgen come and go who thought it was a big deal in 2017 as well).
What the “doctrine of equivalents” says is that if even if a product clearly does not have a feature recited in the claim when properly interpreted, it might still infringe. How do you tell whether or not it does infringe? The Supreme Court set out three questions. Leaving aside for a moment the rider that these are said to be “guidelines, not strict rules”, those questions are:
(i) 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? (ii) 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? (iii) 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?
To get to a finding of infringement by equivalence, the answers have to be “yes”, “yes” and “no” in that order.
In Actavis itself, the Court found that a compound including pemetrexed dipotassium infringed the patent despite the dipotassium salt being clearly a different molecule from the disodium salt recited in the claim. It did the same thing, when used with vitamin B12 for cancer treatment, the skilled person would know that it did it in the same way, and there were no special reasons to restrict the scope of protection to the “literal[3]” meaning of the claim.
Part of the reason that Actavis caused such a shock at the time was that it seemed to give a lot of latitude to a patentee who, on one view, simply hadn’t put a lot of thought or a lot of care into drafting the patent in the first place. I remember wondering (perhaps slightly tongue-in-cheek) how I was going to be able to persuade anyone to pay a supposedly skilled attorney to draft their patent specification for them when the Supreme Court had basically said “don’t worry, if you put something in your claim that you end up regretting, we’ll just find it infringed on the doctrine of equivalents as long as the defendant is doing more or less a similar thing”.
Subsequent decisions only seemed to underline the power of this new-found path to patent infringement. In several cases[4] infringement was found on the basis that some feature of the claim simply did not matter to the inventive concept. For a long time even before Kirin-Amgen it was thought that, even if a feature could be interpreted broadly to include variants, it could not be ignored altogether – an integer of the claim could not be “struck out”. Now, it seems, it can. If a feature does not matter to the inventive concept, then not having that feature at all can be equivalent. As it says in the CIPA Guide to the Patents Acts[5]: “the reader should be in no doubts that Actavis appears to have effected a major change”.
One notable feature of the equivalents cases post-2017 was the lack of success defendants had in trying to argue for a “yes” answer to question three. Where questions one and two had been answered in the patentee’s favour, the alleged infringer would of course say “but look, the skilled person is going to read the patent and think that this feature is regarded as essential by the patentee”. The problem is that the obvious reason that the skilled person might well think that, which is that the patentee has put the feature in their main claim, is basically disregarded because it was the starting point for having to argue over equivalents in the first place. The description of a patent will generally be in conformity[6] with the claims so that features of the independent claims will be recited in mandatory terms whereas other features will be described as optional or preferable. But it seems the courts have been looking for something particularly special in the words of the patent to override the technical findings in questions one and two, and defendants to infringement claims have generally not been able to find that special something.
So the recent-ish decision in Shenzhen Carku Technology v The Noco Company[7] is interesting. This case is about jump starters for cars – portable batteries that you can put in your boot to make sure that you can still start your car even if you forget to turn the lights off while you are in the office catching up on Patents Court case law. The relevant “variant” used a relay to close an electrical connection instead of a power FET which is what was recited in the claim. Relays and FETs were, in relation to this invention, found to be common general knowledge alternative ways of achieving the same thing. They each have their own pros and cons but they achieve the same result in the same way. Questions one and two were answered in the patentee’s favour.
However, products using relays were found not to infringe based on question three. Why? Well, perhaps there was no single reason which did it on its own. The following factors however weighed into the balance:
- Both relays and power FETs were mentioned in the specification (albeit relays only in relation to the prior art), but only FETs were claimed.
- Broader, functional, language was present [8] in the specification – “switch” and “switching”.
- The claim itself uses broad functional language for some features (e.g. “sensor”) but “descends to physical detail in relation to the power FET”.
- The patentee clearly knew of both options, so it was significant that only one of them was mentioned in the claims.
So, what to think of all this? The patentee had limited their claim to FETs and with the benefit of 20-20 hindsight they probably shouldn’t have done. Between the lines of the judge’s reasons I think there is criticism of whoever drafted this patent – it would have been so easy for the patentee to claim something broad like a “switching device”, but they didn’t. It wasn’t a clever workaround that they couldn’t have anticipated – they even mentioned relays in the specification, but for some reason drafted a narrower claim. The ease with which someone could have done better in this case seems to have weighed on the judge’s mind.
Whether that really distinguishes previous cases in a principled and repeatable way may not be crystal clear. An appeal is pending. However, I think there are a couple points which can be made almost irrespective of whether the judge was right or wrong.
Six years on from Actavis, we have settled down a bit from the initial “anything is now possible” reaction of some. The cases have shown that the inquiries into what the inventive concept is, what result is achieved, how, and what the skilled person would think, are all perfectly workable. Of course, in many cases the facts are hotly contested and everything might turn on expert evidence, but the law can be applied in a logical way.
The third question may be more difficult. It is about what the skilled reader would think, and so on the face of it looks like going back to a question of objective construction. But it can’t just be about construction of the claim in the ordinary way because that has already been done and is the reason we are exercising our minds over a “variant” in the first place. The skilled person needs some reason to think that “nonetheless” the claim needs to be complied with strictly to be infringed. That reason may come from the patent specification and so there must still be scope for drafting one which you later regret. Words still matter, and hopefully you will be convinced that you still need to get a patent attorney who will draft your claim with care.
[1] [2017] UKSC 48 [2] Kirin-Amgen, Inc v Hoechst Marion Roussel Ltd [2004] UKHL 46 [3] The reference to “literal” meaning was a bit contentious in its own right for a while, although it seems to now have been mostly resolved that “literal meaning” just means the objective construction – what the skilled person would think the patentee was intending to claim. [4] You can read about a couple of them in Adrian Hocking’s article here: https://www.albright-ip.co.uk/2020/04/greater-protection-in-the-uk-for-patent-owners/ [5] 9th Ed, at 125.10 [6] Though the extent to which the EPO and UKIPO will insist on this in the course of prosecution is not entirely consistent and may be in a state of flux, something which might complicate future equivalents cases… [7] [2022] EWHC 2034 (Pat) [8] The judgment says the wording was “available” but this may disguise the point a little. “Power FETS” were in claim 1 from the outset – it was not an amendment made during prosecution. At the point of drafting the specification, the attorney doing so presumably had all the words in the English language “available”.
ASK AN ATTORNEY
Please select a valid form.
Author
