Actavis v Eli Lilly - what patent attorneys and examiners need to know

Actavis v Eli Lilly - what patent attorneys and examiners need to know

With the recent decision in Actavis v Eli Lilly Lord Neuberger has rethought, and rewritten, the rules of infringement. After 35 years of doctrinal development going back to the legendary Catnic case and seemingly settled in Kirin-Amgen in 2005, the Supreme Court has injected a nitro boost and taken the lead in establishing a test applicable across Europe. Litigators are going wild for it, but it is vital to look at the challenge that has been laid down to the patent makers, getters and even granters as it could have major ramifications for the day job.

The basics of the case

Eli Lilly were working on pemetrexed, which is conventionally used to treat tumours in its free acid form - which is toxic. After a very broad initial claim to use of a methylmalonic acid, and the usual added matter wranglings at the EPO, claims were granted to Eli Lilly directed to a specific salt, pemetrexed disodium - which is less toxic. Actavis used different pemetrexed salts with the same effect. "A fight ensued, which presumably wouldn't have happened if an amendment had been available to 'a pemetrexed salt' and it went all the way to the Supreme Court."

The decision 

It looks like Lord Neuberger felt that there was infringement, and that infringement would be found in other European jurisdictions. But, application of the UK’s current 'Protocol questions' (Reminder: is there an immaterial variant? Would that have been obvious at the publication date? Was strict compliance intended nonetheless?) would not deliver that result.

The old test 

The Catnic-to-Kirin-Amgen suite of cases tells us that you shouldn't do the following: 1) look for literal infringement; 2) if that doesn't catch the infringement, apply the Protocol questions. Instead, there is supposed to be a 'single cause of action'. You just interpret the claim and work out what it is intended to mean, kind of in vacuo and without knowledge of the infringement’s effects. That approach has been criticised in the past and probably isn't what anyone does, though they will pay lip service to it. 

The new test 

Here, Lord Neuberger has effectively said 'what about the Protocol to Art.69 EPC - which the Protocol questions are supposed to reflect? It is directed to determining the extent of protection, not to interpreting a claim, and it mentions equivalents!' He holds that shoehorning the equivalents issue via the 'immaterial variants' aspect into a single interpretation step can lead to errors. Instead, we should now interpret and if that doesn't settle it, apply a more equivalents-driven variant of the Protocol questions (step two above) to determine the extent of protection. We’re also allowed knowledge of the result achieved by the variant. 

The new test in this case – equivalents 

The facts in this case give a good illustration and, given that the revised approach reflects what many practitioners were probably inadvertently doing anyway, is unlikely to be dismissed as being fact or technology specific. First, the interpretation step. The Actavis product did not use pemetrexed disodium, but neat pemetrexed or different salts thereof. These do not infringe on a normal interpretation of the claim.

Second, the three revised Protocol questions. The first reformulated question is relatively unchanged and could be paraphrased as 'is it an equivalent?' The judge held that the Actavis products achieved substantially the same result in substantially the same way - so yes. The second Protocol question caused the most problems. Under the old approach there would not have been infringement because the unpredictability of substituting other salts means that it would not have been obvious at the date of the patent that the Actavis product was an immaterial variant. Neuberger disliked this, as it showed the artificiality of treating this purely as a matter of interpretation. 

He preferred a more pragmatic approach, stripped out the time limitation and effectively asked "can you tell from the patent, with knowledge at the date of infringement, whether it's an equivalent (ie. you don't have to guess if it works, and you can consider recent developments in the technology)?". Therefore, in the present case, the question wasn’t 'might it have been obvious that other salts are an immaterial variant?', but 'given that the other salts achieve the same result, would it be obvious that they did so in the same way?'. Based on the facts, his answer again was 'yes'. The third Protocal question of 'strict compliance' is still largely applicable, and could be expressed as 'did the patent nonetheless exclude the equivalent?'. Thankfully, this was considered from the perspective of the whole specification and again at the infringement date. Infringement was found. 

Relevance of the file history 

We'll look at what this means for patent attorneys below, but the change that has more immediate implications stems from the judge's consideration of the EP file history. The instruction is that we should now consider it, but 'sceptically', and only where it unambiguously resolves a point, or is in the public interest (e.g. where the patentee had expressly excluded an interpretation during prosecution). Having said that, Lord Neuberger used the file history for confirmatory purposes here; the original claims were broad; the permitted amendment to pemetrexed disodium was occasioned by intermediate generalisation objections, and so strict compliance isn't required.

This has implications both for drafting and prosecution (and although it doesn't affect the decision, practitioners may agree that the evidence given to Lord Neuberger on the operation of the EPO may have been a little sketchy), so patent examiners might be interested to read the relevant passages of the decision too. 

What does this mean for a patent attorney?

  1. 1) The test for infringement really is more pragmatic. You can expressly use knowledge of the invention and you can read the patent sensibly and assess whether there is an immaterial variant. In terms of certainty, there was always an argument that consulting the file history was a step too far. But, surely everyone already has a glance at it. Also, the most convincing argument against taking it into account appears to have been Lord Hoffman's statement, 'life is too short'! That was back in 2005, but accessing the file is now a matter of a couple of clicks, at least at the EPO. As long as we are not entering an era of pitched battles about what the content of a file history means, this seems a welcome development.
  2. Judging by the 'light touch' level of knowledge of what goes on at the EPO prosecution level, in this particular case, practitioner knowledge may be absolutely vital in analysing the file. We therefore may see more attorneys appearing as experts in this area.
  3. Less welcome is the impact on our daily practice before Patent Offices. What's done is done, and it's probably best not losing sleep about existing skeletons in closets. There is no suggestion that anything but the immediate file history is considered relevant (but of course someone is going to try and cite patent family members in other jurisdictions). Besides, we're usually so aware of the potential impact of statements here in, say, US equivalent proceedings that we're pretty careful anyway. Naturally, we should avoid statements expressly excluding certain interpretations - we have Lord Neuberger's clear guidance on that. Note, however, that unlike in many jurisdictions, the use of a limiting amendment did not restrict the scope of the claim. The judge looked at the original broad claim, concluded that strict restriction to a particular pemetrexed salt was not intended, and then extended the claim scope more broadly than the eventual allowed wording on that basis. However, the context and presentation of the amendment here were relevant. As it was required to overcome an intermediate generalisation objection (to the judge's apparent bemusement) there were no limiting statements in the accompanying submissions that might otherwise have caused problems. In essence, when making any amendment or statement to the office in the future, the attorney should imagine the judge reading it and drawing inferences about the patentee's intent.
  4. When it comes to drafting, the importance of dealing with potential added matter issues arises once again. Put in as much basis as you can, and consider a 'pyramid' of terminology from the broadest expression of a feature, through intermediate or alternative formulations, to the narrowest. The reality is that the perfect wording is usually only available in hindsight, as was undoubtedly the case here, and you could take comfort from the fact that in the end the court generously generalised the eventual term. Or you could consider how much cheaper it would have been to have been literally infringed... 


This is an important decision, and it is useful to consider its impact not just on litigators, but prosecutors, drafters and even examiners. Current best practice may not have changed too much, but every interaction with the patent office now carries just a little more weight, and the 'how would a judge read this' test needs to become second nature when formulating a response. At the broader level, it feels like a welcome clarification or confirmation of a practical application of the infringement determination tests, despite the extra pressure on patent attorneys. We will wait to see if this more 'pragmatic' approach develops not just in the UK but across Europe, and turns out to be as significant as it looks. 

Let us keep you up to date. If you’d like to receive communications from us, ranging from breaking news to technical updates, thought leadership to event invitations, please let us know.

Connect with us

Cookies improve the way our website works. By using this website you are agreeing to our use of cookies. For more information see our cookie policy I accept