The description of a patent should "always" be used to interpret the claims (DexCom vs. Abbott, UPC_CFI_230/2023)
The Paris Local Division of the Unified Patent Court (UPC) first substantive decision on patent validity in DexCom v Abbott (UPC_CFI_230/2023), provides some interesting commentary on how much the description should be taken into account when interpreting the claims. Claim interpretation is the subject of the highly anticipated referral to the EPO's Enlarged Board of Appeal, G1/24 (IPKat). In DexCom v Abbott, the Paris Local Division followed the approach taken in previous UPC decisions, namely finding that the description should always be consulted when interpreting the claims.
Case Background
The case involved DexCom, a US company specialising in continuous glucose monitoring systems, and Abbott, a global healthcare group. DexCom's European patent EP3435866 related to an analyte monitoring system for measuring glucose levels in diabetes patients.
The patent claimed a system comprising a sensor configured to take analyte measurements, a sensor electronics unit, and a display device. A key feature of the system was the use of two different communication protocols, a Bluetooth/BLE for transmitting a first portion of measurement data, and NFC/RFID for transmitting a second portion in response to a data request command.
DexCom alleged infringement by Abbott's FreeStyle Libre 2 products. Abbott challenged the validity of the patent through a counterclaim for revocation, whilst the main validity dispute focused on whether the claimed dual-protocol system was obvious over the prior art.
 |
| Overlapping portions |
The patent in question had previously been opposed, and the opposition rejected with on appeal filed.
Claim interpretation: The description and drawings must always be considered
The Local Division first addressed the issue of claim interpretation, adopting the standard set by the UPC Court of Appeal that "the patent claim is not only the starting point, but the decisive basis for determining the protective scope of the European patent" (para. 4.2).
The Local Division adopted the claim interpretation standard set by two previous UPC Court of Appeal decisions, establishing five key principles for balancing patent protection with legal certainty. As noted by the Local Division, the Court of Appeal had previously found that: "The patent claim is not only the starting point, but the decisive basis for determining the protective scope of the European patent." (para. 4.2). Following this lead, the Local Division agreed that interpretation of the claims also "does not depend solely on the strict, literal meaning of the wording used" as "the description and the drawings must always be used as explanatory aids" (paragraph 4.2, quoting UPC_CoA_335/2023). In other words, for the Local Division it is necessary to examine both the claim language and the description in order to interpret the claims. The Local Division also emphasises that the principles of claim interpretation apply uniformly to infringement and validity assessments.
Applying these principles to the case in question, the Local Division addressed the specific disputes over claim interpretation, particularly regarding the meaning of what constituted "first" and "second" portions of the measurement data. Abbott had argued during EPO opposition proceedings that Dexcom had adopted a narrow interpretation of "portion" requiring: (1) two distinct portions of data using different protocols, and (2) neither portion could contain all the data. Abbott claimed Dexcom was now taking a broader view in litigation because Abbott's Freestyle Libre 2 transmitted "the same analyte measurement data" using both protocols, resulting in "total overlap" (para. 4.6).
Applying its principles of claim interpretation, the Division looked to the description to understand the terms used in the claims. The Division agreed with Abbott that "portions of data are not comparable to parts of a book or portions of a cake" (para. 4.7). In contrast to cake, the Local Division found that while "portions" of data could overlap and contain some of the same data, they also could not be identical. For the Local Division, this would contradict both the usual distinction between "first" and "second" portions and the description's reference to "a portion of the analyte data... and another portion." The Local Division's interpretation therefore relied on both the ordinary meaning of "portion" and the distinction made in the description of the patent between "a portion" and "all" data.
Inventive step
On inventive step, the Local Division found that the prior art disclosed most features of the claimed system, with the only difference being that it did not expressly specify using NFC/RFID as the second protocol. Critically, the Division's definition of portions as being able to overlap but not be identical, meant that the prior art was interpreted as possessing this feature. The Local Division instead determined the core inventive concept as being assessed whether using NFC as the second protocol was obvious.
The Local Division concluded the choice of NFC would have been an obvious to a skilled person, noting that the prior art "expressly discloses that the device 120 already uses NFC for the request command initiating the transmission of the second portion, meaning that the user has already brought the device 120 into close proximity with the device 102. It would therefore be obvious for the skilled person to continue using the same protocol, namely the second near-field communication protocol NFC, to transmit the data and achieve the effects commonly ascribed to this protocol." (para. 23.7)
The Local Division was unconvinced by DexCom's arguments regarding technical difficulties or synergistic effects. It found that selecting NFC as the second protocol would have been a routine choice given its well-known advantages of low power consumption and security due to limited range.
The Local Division therefore revoked the patent in its entirety across all designated UPC states, also dismissing DexCom's infringement claims and auxiliary requests to amend the patent. DexCom was ordered to bear the costs of the proceedings.
Final thoughts
The Paris Division in UPC_CFI_230/2023 follows the lead of other divisions and the Court of Appeal of the UPC in finding that the claims of a patent should be understood in the context of the patent as a whole. In UPC_CoA_335/2023, for example, Court of Appeal similarly found that the description "must always be used as explanatory aids for the interpretation of the patent claim and not only to resolve any ambiguities in the patent claim" (IPKat). By contrast, EBA has indicated in its preliminary opinion that it believes the description "can" be referred to for the purposes of claim interpretation (IPKat). It will be interesting to see how and if any divergence emerges between the EBA and the UPC on this highly contentious issue, and whether the EBA takes the decisions of the UPC into account when tackling the questions of G1/24.
Further reading
- EBA's preliminary opinion in G1/24: "the description and the figures can be referred to in the course of claim interpretation" (Feb 2025)
- UPC Munich Local Division takes a novel approach to claim interpretation (SES vs Hanshow, UPC-CFI-292/2023) (Jan 2024)
- First substantive decision of the UPC Court of Appeal overturns preliminary injunction in NanoString v 10x Genomics (UPC_CoA_335/2023) (March 2024)
Reviewed by Dr Rose Hughes
on
Friday, February 21, 2025
Rating:
Follow the IPKat on LinkedIn 
Posts
Well if the description should always be used to interpret the claims, then it would be pretty risky not to at least consider adapting your description so that it is consistent with any amendments you made to those claims during examination. That's whether the EPO asks you to amend the description or not. Otherwise your claims might be interpreted differently from what you intended with the amendments, and be unpatentable as a result. Maybe you won't need to do it every time, but you'll probably need to do it a lot of time.
ReplyDeleteThat is an interesting thought. Still, I am not persuaded that it is a logical conclusion.
DeleteAlthough the description is used to interpret the wording of the claims, the claims remain "not only the starting point, but the decisive basis for determining the protective scope".
Limitations added to the claims during prosecution will have a meaning. That meaning cannot be ignored simply because the description might include broader embodiments (which lack the additional limitations).
On the other hand, tinkering with the description risks (inadvertently) changing meanings of terms that can be derived from the text of the original description. On balance, it therefore seems far more sensible to just stick with the original description text, which is the approach adopted by most other patent offices, including the US PTO.
Note the guardrail against a narrowing interpretation based on the description, laid out in the recent decision LD Düsseldorf 355/2023 28 January 2025 Fuji v Kodak :
DeleteHeadnote 2.
"The terms used in a claim should normally be given their broadest technically sensible meaning in the context of the claim in which they appear. Art. 69 EPC and its Protocol do not provide a justification for excluding what is literally covered by the terms of the claims by a narrowing claim construction based on the description or the drawings. A narrowing interpretation of the claims which deviates from the broader general understanding of the terms used therein by a skilled person can only be permitted if there are convincing reasons based on the circumstances of the individual case in question."
The ”broadest technically sensible meaning” (well settled in the EPO case law) is the claim interpretation rule which reflects the primacy of the claims. It precludes reading into a claim a feature only disclosed in the description.
I agree with the first "anonymous". I think they are just saying that you need to look at it on a case-by-case basis - there's rarely a one-size fits all approach when it comes to patents. Sometimes a description amendment might be useful, sometimes it might not. Obviously tinkering with the meanings of terms in the description is risky, but no one seems to be suggesting doing that. I'm also not sure the US approach of leaving the description untouched is necessarily the right model for Europe - in the US, they've developed their long line of case law to take into account the amendments and arguments that were made in exam (file wrapper estoppel and all that). That's an alternative in some ways to just amending the description and ending up with a self-contained patent that makes sense as a standalone document.
ReplyDeleteThe point about file wrapper estoppel is a good one. Such a concept is not embedded in patent law in Europe, and so it is difficult to make direct comparisons to practice in the US. On the other hand, looking at patent law across the world, it is clear that the description amendment practice in Europe is the exception and not the rule, and there are plenty of jurisdictions that do not amend descriptions but yet do not have a cast-iron file wrapper estoppel rule.
DeleteAssuming that file wrapper estoppel is off the table, it is interesting to contemplate what post-grant practice in Europe would look like if we stopped adapting descriptions to claims as allowed. As the first anonymous pointed out, if the claims are afforded a broader interpretation than might be expected based upon amendments made during prosecution, this represents a double-edged sword for the patent proprietor. On the one hand, the greater claim breadth provides an advantage for the purpose of capturing alleged infringements. On the other hand, the greater breadth is a liability when it comes to demonstrating patentability over the disclosures of the prior art.
Of course, none of these potential (dis)advantages would be likely to arise in a scenario where an identical approach claim interpretation is adopted both pre- and post-grant. It therefore seems that stopping the description adaptation practice would only cause problems for third parties in cases where (a) the patent office adopts a different claim interpretation practice to the national courts and (b) the national courts bifurcate infringement and validity proceedings.
Whilst those in Germany might be inclined to view (a) and (b) as facts of life, I think that we need to start thinking more broadly, and to question whether such practices are in fact obstacles to the establishment of a coherent and just patent system.
By the way, I also think that we need to remember that all non-claimed embodiments (or subject matter) in the description provide context for the invention. If the claims will be interpreted in the light of the description, then deleting or changing such context risks inadvertently changing the meaning of terms used in the claims.
For historical reasons, the fathers of the EPC have decided that the EPO would not be competent, besides in opposition procedure for post-grant procedures and especially for infringement.
DeleteIt is thus quasi inevitable that there, has been and will be, a different approach in claim interpretation between pre- and post-grant procedures. Just look how the German Federal court assesses added matter in a granted patent.
Looking at the EPC, I can certainly agree that it carves out a role for the EPO that is distinct from that played by the national courts. However, you will have to help me to understand why this fact alone makes it "quasi inevitable" that the EPO and the national courts will differ in their claim interpretation approaches.
DeleteAs I understand it, determining claim scope is a necessary but not sufficient step for determining infringement. That is, merely establishing the scope of a claim does not tell you whether the allegedly infringing acts are, for example, exempted from infringement, infringing by way of equivalence or infringing indirectly (i.e. by inducement).
From my perspective, this means that there is no rule of law or logic which dictates that the separate step of determining claim scope cannot be conducted by the national courts and by the EPO using an identical approach. Indeed, I cannot see anything in the EPC which obliges the EPO to interpret claims by using a different approach to that adopted by the national courts.
I can certainly understand why the EPO might be inclined to favour an approach that is less time-consuming than that adopted by the national courts. But that does not mean that such a "rough and ready" approach to claim interpretation has any kind of sound basis in law.
@ “Anonymous”
DeleteYou might not see anything in the EPC which obliges the EPO to interpret claims by using a different approach to that adopted by the national courts. I have a more differentiated view on this.
In my opinion interpreting claims in a post grant infringement is different from interpreting when it comes to validity at the EPO. In infringement, the scope of protection may be different as this has been enshrined in Art 69(1) and Art 1 of the Protocol.
When looking at the very coherent line of case law of the EBA relating to “directly and unambiguously derivable”, cf. e.g. G 2/98 and G 2/10, it is indeed inevitable that what can be derived during infringement cannot be the same as when applying Art 69(1) and Art 1 of the Protocol.
Some national courts, e.g. the German Federal Court claims to follow the EBA’s case law, but does exactly the contrary when it comes to Art 123(2). There are enough decisions of the boards dismissing the various remedies proposed by the BGH to avoid revoking a patent in case of added matter.
Improver and pemetrexed are two good examples of how the claims are interpreted by the EPO and by national courts.
This has nothing to do with any "rough and ready" approach to claim interpretation. The sound basis in law is the EPC and its interpretation by the EBA. In G 2/88, it held that Art 69(1) and Art 1 of the Protocol should be applied. In G 2/12, it came to the opposite conclusion.
Let’s thus wait and see what will be said by the EBA in G 1/24.
Sure, let's see what happens in G 1/24.
DeleteI have my doubts about whether the EBA will be bold enough to disrupt the current status quo. Nevertheless, it would be a shame if they passed up on what is a golden opportunity to bring some fully joined up thinking to this area of the law.
My reference to "rough and ready" claim interpretation perhaps needs some further explanation.
Firstly, there can be no doubt that, by applying Art 69 and its Protocol, the national courts are interpreting the claims precisely in accordance with the EPC. On the other hand, and as illustrated by G 1/24, there are doubts over whether the same could be said for the EPO.
Secondly, under current practice, the EPO and the national courts will sometimes arrive at different interpretations of the same claims of the same patent. This is because, in contrast to the national courts, the EPO does not fully consider context provided by the description and drawings whereas national courts.
Thus:
potentially questionable legal basis + "non-contextual" analysis = "rough and ready" claim interpretation.
This comment is not intended to cast aspersions on EPO examiners, who I am sure are very diligent in their claim interpretation analyses. It is merely to point out that the EPO's current approach to claim interpretation deliberately excludes a wealth of information (from the description and drawings) that must be considered by the national courts. It also reflects the fact that divergences between EPO and national court claim interpretations needlessly create problems for patent proprietors and the public alike.
One last thing: did you have any thoughts on whether there is any reason why determining claim scope (as opposed to assessments based upon a claim scope so determined, i.e. assessments of infringement or patentability) cannot be conducted by the national courts and by the EPO using an identical approach? I am not sure whether there is any established EPO case law which directly addresses this point.
That the CoA states that the description should be taken into account when interpreting the claims does not come as a surprise. In view of its chairperson, who comes from the German Federal Court, where this is standard practice, known under “Auslegung”, something else would not have been expected. It is thus not surprising that the CD Section Paris took the same position.
ReplyDeleteThe opposition had indeed been rejected and no appeal was filed. The opponent at the EPO was defendant 2 in the UPC case. In view of the litigation pending in Germany, the intervener at the EPO was defendant 8 in the UPC case,
The part of the reasoning on added matter is similar to that of the OD.
Before the OD, lack of inventive step was argued over US 015/0205947= D14op=D1upc +common general knowledge or, US 2015/0038818=D2op=D2upc + common general knowledge as well as US 2011/0213225=D7op=D3upc + common general knowledge. A further line of attack was over US2014/266785=D4op + D2op=D2upc or + D7op=D3upc or + D14op=D1upc.
The UPC decided that claim 1 as granted lacked IS over US 015/0205947= D14op=D1upc +common general knowledge.
As the OD rejected the opposition, none of the 27 auxiliary requests filed in opposition by the UPC claimant were discussed at the EPO. The two auxiliary requests filed before the UPC have not been checked against the 27 auxiliary requests filed at the EPO. Those 27 auxiliary requests could be considered as having been admissibly raised, and would most probably have to de used in a possible appeal.
Knowing that 70% of decisions of ODs are set aside by the boards, and that after appeal, only 15% of patents survive an opposition, but 50% are revoked, it could have been worth for the opponent and the intervener to appeal the OD’s decision which was so to say set aside by the UPC CD.
It can be expected that an appeal will be filed by the claimant, and only after the appeal we will know the fate of EP3435866.
Whatever G 1/24 brings, it is to be hoped that leaving a different possible interpretation of claimed features in the description, especially a more limited one, compared to that in the claim, should not any longer be tolerated at the EPO.
It is also to be hoped that in G 1/24, the EBA will follow what it said in G 2/12 and does not abolish the notion of “directly and unambiguously derivable” set out in G 2/10 and G 2/98.
I have a question. I fear it is a stupid question but I am buoyed by the thought that "There are no stupid questions". I look forward to a quick, clean and clear answer from a fellow reader.
ReplyDeleteWhen the court decreed that when construing a claim having validity date X, "the description" which provides the obligatory "support" for that claim cannot be ignored, was it talking about A or B, namely a) the description as filed on date X or b) the description in the B-publication? In its judgement, did it make it directly and unambiguously clear which of those two descriptions it was talking about? Or is the answer no more than "derivable" from what is implicit in the text of the judgement?
That is a very good question. What do you think the answer might be? Can you see anything in the EPC which supports either of the options that you mention?
DeletePersonally, I think that option a) is the only sensible / logical choice. But that is mainly because I believe that option b) would give rise to far more questions than it answers, and would likely also produce some nonsensical results.
The answer? For me, BOTH the A and the B publication must be taken into account. The legislator deliberately excluded as a ground of invalidity a lack of clarity. The court has to do the best it can with the claims given to it by the EPO. The best way to arrive at the proper interpretation of the claim is to look at both the description as filed and the description of the granted patent. Focus on the prosecution amendments and milk them for what they tell us about (in the words of the UK Supreme Court) what the patent applicant was using the words of the issued claim to mean. As judges in the USA like to warn "Be careful what you ask for".
DeleteWhat is the alternative? File wrapper estoppal? Yet more extrinsic evidence? No Thank You. Remember what Robin Jacob used to counsel "We can learn a lot from the Americans: observe closely what they do, and then make sure not to make the same mistakes".
The UPC is feeling its way forward. Perhaps its ambiguity is deliberate. You know, like Donald Trump, say something outrageous (Riviera luxury hotels along the Gaza waterfront) see how everybody reacts and on the basis of the feedback make ever more robust decisions. One calls it "judicial economy" right?
I guess that we shall just have to agree to differ.
DeleteFor me, the best way to interpret a claim is to look at both the specification as filed and the claim of the granted patent.
For reasons outlined in my 23 Feb comment, I think that the practice of adapting the description to allowed claims is problematic on the grounds that, by changing description text that provides context for the claimed invention, it can (inadvertently) change the meaning afforded to terms used in the claims. However, even if we assume that a description has been "perfectly" adapted to the allowed claims, what could we possibly learn from the amended description that is not derivable from the claims upon which the amendments to the description were based?
As I understand it, file wrapper estoppel is a way to ensure that patentees cannot disown, or argue contrary to, the reasons why they made certain claim amendments during prosecution. We do things differently in Europe, as national courts prefer to reach their own conclusions regarding the subject matter encompassed by the wording of the claims as granted.
The way that we do things in Europe means that both the patentee and the patent office might be surprised to learn that their pre-grant interpretation of the wording of a claim is different to a national court's preferred (post-grant) interpretation. In such scenarios, file-wrapper estoppel might, at least on some occasions, prevent the national court from settling upon a divergent interpretation that they might otherwise prefer (based upon their consideration of various intrinsic and extrinsic factors). But how might amendments to the description achieve the same effect? Only by changing (relative to the description as filed) the interpretation of terms used in the claims. In other words, only by adding new matter.
Is that really a coherent, workable and just solution?