The risk of pre-grant description amendments (T 450/20)

The EPO Boards of Appeal have recently taken an uncompromising approach to claim interpretation. The Board of Appeal decision in T 450/20 is in line with the trend, finding that the description of a patent could not be mined for a definition limiting the meaning of the claims. However, in contrast to another recent decision (T 1924/20IPKat), the Board of Appeal in this latest case did find it permissible to use functional language from the description to construe the claims. Furthermore, in an easy to miss but critical passage, the Board also ruled out any role for the description of the patent application in claim interpretation. Instead, according to the Board of Appeal, only the description of the granted patent may be used for claim interpretation under Article 69(1) EPC, regardless of whether the description was adapted prior to grant. This is a worrying development, given the EPO's current bullish stance on the issue of description amendments, which currently necessitates patent applicants to make onerous description amendments prior to grant. 

How permanent is permanent (T 0450/20)

The decision in T 0450/20 related to EP 2254485, claiming a stent for unblocking blood vessels. The patent was upheld at Opposition. Claim 1 as granted claimed a device comprising inter alia a guide wire and a tapering mesh "permanently attached" to the wire. The feature of a tapering portion "permanently attached" to the guide wire was mentioned only once in the description and not further explicitly defined. 

The issue on appeal was whether the patent lacked novelty. The novelty of granted claim 1 depended on whether the device described in the prior art (US 2005/0209678) included an equivalent tapering portion "permanently attached" to a guide wire. The prior art device comprised a tapering portion that was firmly attached to a guide wire for the delivery of the device into the blood vessel of a patient. The prior art mentioned that the wire could be detached if necessary by applying electrical energy to a steel connecting element. However, detaching the device would be a complicated process requiring specialist equipment. The question before the Board of Appeal was primarily one of claim interpretation, particularly what did claim 1 mean by the term "permanently attached", and did it include the potentially detachable device of the prior art?

Using the description to interpret the claims

According to Article 69(1) EPC "[t]he extent of the protection conferred by a European patent or a European patent application shall be determined by the claims. Nevertheless, the description and drawings shall be used to interpret the claims." The most recent case law on claim interpretation from the Boards of Appeal has taken a hard line against using the description to interpret otherwise clear features in the claims. In both of the "catastrophic comma loss" cases (T 1473/19T 1127/16), the patents were found invalid for added matter in view of a single comma lost from the claims (IPKat). The Boards of Appeal in both decisions found that the claims would be clear to a skilled person without recourse to the description. As such, the description could not be used to interpret the claims and rescue the claims from the added matter trap. 

In contrast to the catastrophic comma loss cases, the granted claims of the patent in T 0450/20 were not considered clear per se. Particularly, given that the claims did not define the phrase "permanently attached", it was considered unclear what this feature actually encompassed. 

Purposive construction and using the description to interpret the claims

The description of the patent did not provide an explicit definition of the phrase "permanently attached". However, the description did indicate that the device of the invention may be removed from the patient's blood vessel after it had served its purpose. The Board of Appeal thus construed the term "permanently" in claim 1 to mean that the attachment between the mesh and the guide wire must last long enough to allow reliable extraction of the mesh using the wire. The Board of Appeal therefore considered it to be acceptable to use functional language provided in the description to interpret an unclear feature of the claim.

The Patentee argued that, instead of the Board of Appeal's broad functional definition, a narrower interpretation of the phrase "permanently attached" should be used in view of the description. Particularly, the Patentee argued that a skilled person would understand from the description that "permanent attachments" did not include attachments that were releasable. Prior to grant, the patent application had in fact been amended to explicitly state that an embodiment comprising a releasable attachment was "not in accordance with the invention". The Patentee thus argued that the claims should be interpreted in view of this amended description, whereby "a releasable attachment" could not be considered a permanent attachment. 

The Board of Appeal disagreed. Particularly, citing the second catastrophic comma loss case (T 1473/19IPKat, same Board of Appeal), the Board found that according to Article 69(1) EPC:

"only the claims determine the extent of protection. The description and the drawings have no such function, and must only be used to interpret the claims. This means that the description and the drawings can only be used for interpreting features which are already present in the claims, but not for adding further claim features or for replacing existing claim features by others". 

To the Board of Appeal, using the description to read-in a new limitation to a claim would not be in line with the primacy of the claims. The Patentee's recourse to the description to limit the meaning of "permanent" was therefore not accepted. 

The description of the patent application is irrelevant for claim interpretation

Interestingly, the Board of Appeal also rejected the Patentee's argument that the description of the application as filed may be used to interpret the claims. Before grant, the patent application in question specified an attachment that may be "releasable or permanent" (text-intended for grant). The Patentee argued that the intended meaning of a permanent attachment was therefore one that is not releasable. However, for the Board of Appeal, Article 69(1) EPC only permits the claims to be interpreted in view of the description of the granted patent, and not of the description of the patent application before grant (r. 2.16).

The Board of Appeal rejected the Patentee's argument that, because the prior art attachment was "releasable", it could therefore not be considered permanent. The purposive construction of the phrase "releasably attached" arrived at by the Board of Appeal in view of the functional language provided in the description was also found to encompass the attachment described in the prior art. Granted claim 1 was therefore found to lack novelty in view of the prior art. 

Final thoughts

There are a few interesting take home points from T 0450/20, relevant to both claim interpretation and description amendments. First, the Board of Appeal in this case found the description of a patent application irrelevant for claim interpretation under Article 69(1) EPC. According to the Board of Appeal in this case, only the description of the granted patent may be used.  Importantly however, following the recent controversial changes to the requirement to bring the description in line with the allowed claims before grant, the description of the patent may differ substantially from the description of the application as filed (IPKat). Therefore, if the approach taken in T 0450/20 is followed by other Boards, then the risk of making substantial amendments to the description prior to grant is significantly heighted. Making amendments prior to grant will not only run the risk of influencing doctrine of equivalents analysis in infringement cases, but may also adversely impact claim interpretation in any post-grant proceedings, including validity analysis at opposition. 

Second, there appears to be some divergence among the Boards of Appeal on what, if anything, may be taken from the description of a patent in order to interpret the claims. According to the latest decision T 0450/20, functional language in the description may be used to resolve a lack of clarity in the claims. This appears to contrast with another recent decision which found that functional language from the description could not be used in claim interpretation (T 1924/20IPKat). Unfortunately, diverge in practice from the Boards of Appeal on the role of the description in claim interpretation only serve to raise the stakes in the description amendment debate. 

Further reading

Claim interpretation

Description amendments

The risk of pre-grant description amendments (T 450/20) The risk of pre-grant description amendments (T 450/20) Reviewed by Rose Hughes on Monday, July 17, 2023 Rating: 5


  1. Comment part 1 of 2
    To me, it seems that there is a problem with the Board's interpretation of Art 69(1).

    The first sentence of Art 69(1) refers to both "a European patent" and "a European patent application". The second sentence of Art 69(1) merely refers to "the description and drawings" and does not draw a distinction between:
    - the description and drawings of "a European patent"; and
    - the description and drawings of "a European patent application".

    It is therefore unclear from the wording of Art 69(1) which description and drawings should be used to interpret which claims.

    However, using anything other than the description and drawings the European patent application as originally filed can cause serious problems. This is because relying upon an amended description to interpret the claims makes the interpretation of even original claim language a moving target.

    To illustrate, consider the hypothetical situation where:
    - original Claim 1 of a patent application referred to a stent comprising a "permanently attached or releasable" wire; and
    - original Claim 2 of that patent application referred solely to a stent comprising a "permanently attached" wire.

    How would Claim 2 of the patent application be interpreted?

    As "permanently attached" is an unclear term, it would presumably be interpreted in the light of the description and drawings of the application as originally filed.

    But what happens if original Claim 1 is amended by incorporating the limitation of original Claim 2, and the description and drawings are amended to bring them into line with the amended claims? Should the claims then be interpreted in the light of the amended description and drawings?

    In essence, the Board in T450/20 answered "yes" to this question. By doing so, they arrived at an interpretation of "permanently attached" that at least partially overlaps with the "releasable" alternative that was deleted from the claims. This then begs the following questions.

    1) Would the Board have afforded the same meaning to "permanently attached" if they had instead interpreted the claims in the light of the description and drawings of the application as originally filed?

    2) If the answer to question 1 is "no", does Claim 1 as amended (and as interpreted in the light of the amended description and drawings) cover more subject matter than original Claim 2?

    3) If the answer to question 2 is "yes", does Claim 1 as amended (and/or the amendments to the description) add subject matter relative to the content of the application as originally filed?

  2. Comment part 2 of 2
    In essence, the main problem here is that using the amended description and drawings to interpret the claims clashes with the basis upon which added matter is assessed, namely the original description and drawings (and claims).

    There is also the problem that the "description and drawings" can change during opposition proceedings, if the patent is upheld in amended form by the OD. This poses the following questions.

    A) Which "description and drawings" should a Board of Appeal use to interpret the claims: those of the patent as originally granted or those of the patent in the form upheld by the OD?

    B) Will the answer to question A depend upon whether the patentee has appealed and is pursuing the claims as originally granted (or at least broader claims than those upheld by the OD)?

    C) Will the "description and drawings" used to interpret the claims change if the Board only upholds narrower claims, and then insists upon the description being brought into line with those narrowed claims?

    According to the logic of the Board in T450/20, the answer to question C must be "yes". Indeed, if the patent will be republished in amended form, should not the description and drawings of the amended patent be used to interpret the claims from thereon?

    Of course, all of these problems would go away if the (unchanging) description and drawings of the application as originally filed were instead used to interpret the claims at all times. This would make a great deal of sense and be consistent with the EPO's practice on added matter. However, I cannot see a Board of Appeal agreeing with this view any time soon, not least because it would highlight the absurdity of the EPO insisting upon "strict" adaptation of the description to amended claims.

  3. Much as I agree with Proof that this approach is far messier than just leaving the description alone, T 0450/20 makes complete sense if you start from the position that the description must be adapted. If you were not supposed to use the amended description for claim interpretation then there would be no point in amending it. Since the EPO has decided you must amend the description, it follows that you must use the amended description to interpret the claims. That this makes a big mess when it comes to added matter is apparently ok, because... reasons?

  4. The position about adaptation of the description taken by some representatives, actually on both sides of The Channel, is well known. I do not wish to restart this decision. For me the topic is exhausted. My position is known.

    In my opinion, the fundamental question of law which should be decided by the EBA is whether recourse to the description under Art 69(1) and Art 1 of the Protocol has to be systematic or not, or only when dealing with Art 123(3) or in case of lack of clarity of the claims.

    In defending its position against the referral, the board referred to T 1473/19. In this respect it can be said that the board was at the same time judge and party.

    In a different blog, I expressed the view that the present board was going a trifle too far when claiming that permanently attached can also mean releasable. That in real life a physical connection can always be detached if an appropriate external force is applied is, in the present context, beside the point.

    A claim should be read with a will to understand and not misunderstand it. Here the interpretation by the board is really far fetched. Permanently attached means not releasable under normal conditions of use.

    The deletion of the various possibilities of use, and the limitation to the one of the two originally disclosed alternatives, i.e. to "permanently attached" could probably have helped to save the patent, without introducing any added matter.

    1. It seems to me to be far from certain that the Board misapplied the principles of claim interpretation in this case. That is, if we accept that the Board should use the (amended) description and drawings of the granted patent when applying Art 69(1) EPC, it does not seem unreasonable that the phrase "permanently attached" should be afforded a relatively broad interpretation.

      The problem as I see it is that the (pre-grant) amendments to the description removed some of the context of the original disclosure, which could well have cast a different light on the meaning of "permanently attached".

      In any event, there is little doubt that, under current EPO practice and case law, adapting the description to allowable claims has become a very complicated and risky procedure for applicants. As illustrated by T450/20, the risks for applicants include the possibility that certain amendments to the description might inadvertently change the (post-grant) interpretation of a term used in the claims ... possibly to a meaning that cannot be clearly and unambiguously derived from the disclosures of the application as originally filed.

  5. There is another point to add. That is, as they were well aware of the prior art found by the Board to anticipate Claim 1 as granted, the ED for the patent discussed in T450/20 clearly did not go through the process of interpreting that claim in accordance with the amended description.

    So is this how it is going to work from now on? An ED interprets a claim based upon an original description, and then, if that claim is amended, insists upon the filing of an amended description. The ED then does not revisit the issue of claim interpretation ... instead leaving it to an OD, Board of Appeal or court to decide whether the amendments to the description change the meaning of the claim (relative to the interpretation afforded by the ED prior to amendment of the description).

    This has the feeling of the EPO setting numerous, potentially unavoidable bear traps for applicants. Whether they mean to do this is irrelevant. This is because the only thing that will matter to applicants is what the EPO's practice actually achieves in practical terms.

  6. This is all fairly terrible for European patent law. Legal certainty, pragmatism, and good sense have all gone out of the window.

    More extensive amendments, disagreements between applicants and examiners about claim/description interpretation, and additional rounds of prosecution will all cause costs and delays in prosecution to rocket.

    On the plus side, more extensive amendments, disagreements between applicants and examiners about claim/description interpretation, and additional rounds of prosecution will all cause patent attorney fees for prosecution to rocket!

    1. When did the EPO ever shown good sense? How many other countries have the inescapable traps that the EPO has?


All comments must be moderated by a member of the IPKat team before they appear on the blog. Comments will not be allowed if the contravene the IPKat policy that readers' comments should not be obscene or defamatory; they should not consist of ad hominem attacks on members of the blog team or other comment-posters and they should make a constructive contribution to the discussion of the post on which they purport to comment.

It is also the IPKat policy that comments should not be made completely anonymously, and users should use a consistent name or pseudonym (which should not itself be defamatory or obscene, or that of another real person), either in the "identity" field, or at the beginning of the comment. Current practice is to, however, allow a limited number of comments that contravene this policy, provided that the comment has a high degree of relevance and the comment chain does not become too difficult to follow.

Learn more here:

Powered by Blogger.