How the doctrine of equivalents impacts patent protection in Europe

The 'doctrine of equivalents' can help patent owners achieve meaningful protection to stop products which differ immaterially to their protected original. Conversely, the application of the doctrine can make it harder for others to navigate patents with any certainty.

In Europe, the doctrine of equivalents derives from long-standing case law developed by national courts and, since 2001, the Protocol on the Implementation of Article 69 of the European Patent Convention (EPC). However, national courts have taken different approaches to the doctrine of equivalents, causing a degree of uncertainty for businesses. New guidelines recently introduced by the European Patent Office (EPO) also have potential to impact on claims of patent infringement on the basis of the doctrine.

The legal provisions

Read together, Article 69 of the EPC and the associated protocol confirm that the extent of the protection conferred by a European patent or a European patent application is to be determined by the patent's claims, though reference can be made to the description and drawings for the purpose of interpreting those claims in the event of any ambiguity.

The extent of the protection is not defined solely "by the strict, literal meaning of the wording used in the claims". The text requires "due account" to be "taken of any element which is equivalent to an element specified in the claims" when determining the extent of protection conferred by a European patent.

In essence, if a product or process is not substantially different from the elements of a patented invention, it can be considered to be infringing. This approach aims to prevent patents being circumvented by making small changes to the patented invention whilst reproducing its core and inventive elements. However, within this broad doctrine, courts and patent offices across Europe have diverged and adopted different approaches.

The UK position

The current legal test for determining whether there has been infringement of a patent on the basis of the doctrine of equivalents was set out in the UK by the Supreme Court in 2017 in the case of Actavis UK v Eli Lilly.

In his leading judgment, Lord Justice Neuberger, then president of the Supreme Court, said that infringement on the basis of the doctrine of equivalents will be determined if a three-stage test is met:

• the variant achieves substantially the same result in substantially the same way as the invention – i.e. the inventive concept revealed by the patent;
• it would 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, and;
• such a reader of the patent would not 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.

Lower courts have subsequently applied the three-stage Actavis test to the facts of the disputes before them in a variety of different cases, including a number in the context of life sciences patents. These have clarified several points but not fundamentally shifted the law as set out by the Supreme Court.

The Supreme Court's reference to 'inventive concept' in its ruling in the Actavis v Eli Lilly case confirmed that the focus of the infringement analysis should no longer be on the claim language, but should be directed to the 'inventive concept' of the claim when considering whether a variant would infringe. One key aspect of the assessment, therefore, requires the identification of the 'essence' or 'core' of the invention and as the subsequent case law has shown, this is not always an easy task. Recent decisions, including from the Illumina v MGI case, have further fuelled this fire, suggesting that a patent protects more than just the claimed invention and instead protects the 'inventive concept'.

What constitutes an 'inventive concept' is clearly fact specific, but its precise nature, including whether it amounts to the 'inventive concept' of a claim or specification as a whole, is fertile ground for future litigation. However, this has the potential to fundamentally alter the scope of a patent, and is a developing area of patent law.

Recognising the increase in patentees seeking to rely on the doctrine of equivalence in patent infringement proceedings in the UK, in a recent decision in the case of Facebook v Voxer, Lord Justice Birss confirmed that parties wishing to rely on this doctrine must say so up-front, and set out their position in their submissions at the start of the case.

In the Actavis v Eli Lilly case, the UK Supreme Court also introduced limited circumstances in which the court may refer to the patent prosecution history to determine whether the patentee had made any concessions during prosecution of the patent which might limit the equivalents that may be caught by the claim in question.

The Supreme Court said that the prosecution file history may be considered where either the point of construction is truly unclear but the contents of the prosecution file unambiguously resolve it, or it would be contrary to the public interest to ignore the contents of the file. In reality, in contrast to some other jurisdictions, so far the UK courts have so far been reluctant to consider the prosecution file history, and it certainly looks unlikely that the circumstances in which it may be relied on in UK patent litigation will be widened in the near future.

France

France has a long history of applying the doctrine of equivalents and the relevant test developed by settled case law leads to a very broad concept.

In France, the doctrine of equivalence can only be relied upon if the patent claims cover a "general means" and not a "particular means". In other words, the first question is to assess whether, even if the literal wording of the claims appears to be directed to a specific means or embodiment, the scope of protection is actually broader and extends to the function performed by such embodiment and thus extend to certain variants. That will be the case when the function performed by the claimed means is a new one. In such cases, the test for equivalence applies. Under such test a variant is considered equivalent if, despite having a different form, it performs the same function in order to achieve a similar result.

The French test for equivalence can be summarised as follows:

• Is the function performed by the claimed means a new one?
• If yes, does the variant perform the same function?
• If yes, does the variant achieve a similar result?

Courts have considered the questions of when a variant can be said to perform the same function and when it can be said to achieve a similar result.

In relation to the requirement that it performs the same function, a variant's function is considered to be the primary, direct and immediate, technical effect produced by the means for the relevant application(s).

In relation to the requirement that it achieves a similar result, a variant's result is considered to be the practical advantage provided by the means. Traditionally, case law considered that the result achieved with the variant needed not be identical to the original, just of the same nature, including of a lesser quality. More recent case law suggests that a stricter approach now prevails, with an identical result being required for infringement by equivalence to be characterised.

Unlike in other jurisdictions, the French test of equivalence is an objective one, and does not have subjective elements. It is therefore not relevant, for the purposes of determining infringement by reference to the doctrine of equivalents, whether the variant was obvious to the person skilled in the art at the priority date or whether the patentee intended to cover the variant.

However, judges can take into consideration declarations made during prosecution, or in foreign proceedings, and file history. That said, in a recent case between Eli Lilly and Fresenius Kabi, the Paris Judiciary Court said that the prosecution file is only an optional tool for judges when determining the scope of protection and made a distinction in this respect between different types of amendments to the claims made during examination of the patent, namely between:

• amendments made to distinguish the invention from prior art, such as to overcome an inventive step, which may be interpreted as limiting the scope of protection; and
• amendments made to overcome a formal issue, such as added matter, which should not affect the scope of protection.

Interestingly, in this case, the Paris Judiciary Court found that Eli Lilly's patent for pemetrexed was infringed literally by Fresenius Kabi's product, rather than on the basis of the doctrine of equivalents, despite the patent’s claims being directed to a specific embodiment. This finding was based on a broad interpretation of the claims in view of the description. The dispute between Eli Lilly and Fresenius Kabi has come before courts in other European jurisdictions too, but while those courts concluded that there was infringement of Eli Lilly’s patent, unlike the Paris court this was determined on the basis of the doctrine of equivalents. This demonstrates differences in national approaches which is a relevant consideration for clients that may consider raising such arguments in future disputes.

The French doctrine of equivalents can lead to a very broad scope of protection, very favourable to patentees. However, this is not without risks for them. As there is no bifurcation in France, courts will consider both validity and infringement at the same time. As a result, those defending an infringement claim may be able to raise a so-called squeeze argument when the scope of protection is so broad that it covers variants from the prior art, which would render the patent obvious and invalid.

There may also challenges in enforcing patent rights when relying on the doctrine of equivalents in the context of preliminary proceedings. In France, a preliminary injunction will only be granted if the patent is considered to be valid and infringed from a summary analysis, which requires that there are no serious arguments on either invalidity or non-infringement.

In the Saint-Gobain v Knauf case in 2017, the Paris Judiciary Court said that, as a matter of principle, preliminary injunction proceedings were not available when the patentee relied on infringement by equivalence, as this type of infringement would require an assessment which "goes beyond the prima facie case required for interim proceedings". However, more recently in the Eli Lilly v Zentiva case in 2021, the Paris Judiciary Court granted a preliminary injunction relying on both literal infringement and infringement by equivalence. Whilst the court did not take a position in principle, this may suggest that it now considers that preliminary injunctions would be available even if relying on the doctrine of equivalents.

Germany

In Germany the position is different still.

Infringement by equivalence is only possible if patent owners satisfy two separate legal tests.

The first test is comprised of three steps. In summary, they must demonstrate:

• that the variant solves the problem that the invention is designed to solve with means that objectively have the same effect;
• that the person skilled in the art, using the common general knowledge, would have realised at the priority date that the variant has the same effect, and;
• that the considerations which the skilled person takes into account for the variant in the light of the meaning of the invention were close enough to the considerations taken into account for the literal solution protected by the claims, such that the skilled person will consider the variant as a solution which is equivalent to the literal one.

The so-called ‘Formstein defence’ is a defence to infringement by equivalents, which originated from the so-called ‘Formstein’ judgment of the German Federal Court of Justice issued in 1986. This defence provides that if a patent is not infringed literally but would be under the doctrine of equivalents, the accused embodiment will nevertheless not be considered as infringing if the embodiment would not have been patentable over the prior art at the priority date of the patent.

The burden of proof is with the allegedly infringing party to show that the ‘Formstein’ defence applies. In essence, it needs to show that its variant was not patentable over the prior art at the priority date because it lacks novelty or does not provide an inventive step. The idea of the Formstein defence is to limit the scope of a patent so that embodiments which could not have been patented and monopolised in the first place will not become illegal. However, in the German bifurcated court system the infringement courts must in principle respect the fact that a patent has been granted. Therefore, the Formstein defence is in itself limited in that it is not permissible if applying the defence would have the consequence that the patent doesn't provide a patentable teaching and, in fact, has been granted incorrectly.

How does this play out in practice? Assuming a patent has been granted for a medicinal product with two features and a one-feature product has been known in prior art, and further assuming that the use of the one-feature product does not constitute a literal infringement of the two-feature patent but falls into such patent's scope of protection under the doctrine of equivalents: if a party manufacturing or selling the one-feature product can demonstrate that its product could not have been patented over the prior art at the priority date of the two-feature patent, it could defend against claims of infringement under the Formstein defence.

EPO examination guidelines

Often, patent applicants have to amend the original patent applications they file at patent offices in response to feedback obtained from the officials who examine their filing. Examination guidelines produced by the European Patent Office (EPO) which came into force on the 1 March address the topic of amendments and could have an influence on infringement by equivalents claims in the months and years ahead.

Part F-IV-4.3 (iii) of the new guidelines state that embodiments of the invention contained in the description which are no longer covered by the claims due to amendments made during examination must be deleted or, if they are useful for highlighting specific aspects of the amended claims, it must be "prominently stated" that they are not covered by the claims.

It is possible that these new requirements could impact the assessment of equivalents. This is because if an embodiment is explicitly said to be not covered then it may be more challenging to argue that a similar variant infringes by equivalence.

The prosecution file history is relevant in some jurisdictions to determining the scope of patent protection, though there are different practices in Europe as to the extent of which declarations or amendments made by the patentee during examination at the EPO can be used to interpret the scope of protection of the patent during national litigation.

What does this mean?

While the doctrine of equivalents is an established principle in some jurisdictions such as Germany and France, it is still a developing area of law in others. Certainly in the UK, the number of cases in which this allegation of patent infringement has been pleaded has seen a sharp increase since the Supreme Court’s ruling in the Actavis case, a trend which we expect to continue. Questions have been raised over whether the doctrine of equivalents, and consideration of the 'inventive concept', effectively broadens the scope of the patent beyond what is claimed by catching a wide range of variants. 

There are still open questions as to whether the Formstein defence is available in other jurisdictions.

In the UK, the Formstein defence was raised at the trial in Facebook v Voxer.  Lord Justice Birss found that the patent being asserted was not infringed literally or via the doctrine of equivalents, and that the patent was invalid on the ground of obviousness. He therefore did not need to consider the Formstein defence arguments but nevertheless did so. The judge said that if he did have to decide the matter, he would hold that the Formstein approach is the right one, so that the conclusion if the equivalent lacks novelty or is obvious is that the claim scope must be confined to its normal construction in that respect. This appears to open the door to the Formstein defence being applied in future cases in the UK. Given the increasing prevalence of the doctrine of equivalents in patent matters, it is expected that this will be considered further by the courts in due course.   

While the expansion of the doctrine of equivalents in some jurisdictions can be viewed as a positive development for innovators, it is perhaps less comfortable for generic and biosimilar manufacturers who may still be found to infringe a patent despite efforts to design around the originator product. That said, it is inevitable that there will be further legal developments in this space, particularly in the UK where the doctrine is in its infancy. Businesses should consider which jurisdictions and legal positions may be more favourable to their business model.

Before disputes are even contemplated, of more practical concern for companies is the need to clearly characterise and define the inventive concept of the patent application and extent of protection that may be offered. Put simply, the patent holder should know what their invention is.  For competitors, the ultimate focus when undertaking infringement analysis should therefore no longer be limited to the claim language but should also be directed to the inventive concept of the patent. This is relevant in all sectors, not just life sciences.