Our services are centred around intellectual property that can be registered. We protect innovation, design, and branding across all sectors of industry, and at all stages in the supply chain.

For each IP right we offer services covering strategic advice, pre-registration searches, registrations and renewals, oppositions and dispute resolution. We handle work throughout the world, working with local colleagues in over 100 countries.


Our attorneys specialise in one or more sectors of industry, which enables them to provide quality advice with a commercial focus.

Our patent specialists have detailed understanding of the background technology, which ensures that your patent applications are prepared with the correct scope, reducing the likelihood of challenges from third parties and objections from the patent office.

They also advise whether other forms of protection would be more appropriate. Our brand specialists work with brand managers for leading brands and their advice is commercially focussed making sure that you get the best value from your budget.

An overview of Technical Board of Appeal decisions since G2/21: where are we with post-published evidence?


At the end of November the Board of Appeal (3.3.02) handed down its decision on T 116/18. As a reminder, T 116/18 was the case that led to G2/21; the referral to the Enlarged Board of Appeal on the standard required to rely on post-published evidence to support inventive step (the “plausibility” decision). A summary of that decision can be found here.

This decision was not the first case to refer to G 2/21, but it is the first to analyse it in detail and it is also the first decision to essentially create a new test which appears to diverge significantly from other interpretations of G 2/21.

In this article we first look at T 116/18 and then compare it to the approach taken by the Technical Boards of Appeal in other decisions relying on G 2/21.

What we have found is that T 116/18 appears to be an outlier in how G 2/21 is being interpreted, suggesting that we are still far from having clarity on when post-published evidence may be relied on.

T 116/18

In short, the Board of Appeal (the “TBA”) concluded that post-published evidence may be relied on to show a technical effect in support of an inventive step as long as the skilled person would have no “legitimate reason to doubt that the effect could be achieved”.

For those of you who have been following T 116/18, this phrase will be familiar to you. It was used by the Enlarged Board of Appeal (the “EBA”) in paragraph 7 of G 2/21, when they referred to the line of case law relating to ab initio implausibility:

“(7) The referring board discussed a second line of case law which requires that post-published evidence must always be taken into account if the purported technical effect is not implausible (type II, called “ab initio implausibility” by the referring board, point 13.5 of the Reasons for the referring decision). In accordance with this line of case law, post-published evidence can only be disregarded if the skilled person would have had legitimate reasons to doubt that the purported technical effect would have been achieved on the filing date of the patent”

There has been much debate about how the guidance provided by the Enlarged Board of Appeal in G 2/21 would be interpreted by the Boards of Appeal but one thing appeared to be clear from their judgement … they did not conclude that the test for allowing post-filed data was one of “ab initio implausibility.  So how did the referring Board arrive at a standard that required a “legitimate reason to doubt that the effect could be achieved” to prevent the use of post-published evidence?

What did G 2/21 actually say?

The EBA set out a two-step test for assessing whether post-filed data may be relied on to support an alleged technical effect for the purpose of assessing inventive step. 

It concluded that the relevant standard requires an assessment of what the skilled person, with the common general knowledge in mind, would understand, at the filing date of the application as originally filed, as the technical teaching of the claimed invention. This “relevant standard” is considered to be met if the skilled person would derive the technical effect as being:

  1. encompassed by the technical teaching of the claimed invention; and
  2. embodied by the same originally disclosed invention

Whilst most commentators agreed that that this new test was ambiguous, it was generally accepted that that it was not intended to be a re-wording of the “ab initio implausibility” standard. 

How did T 116/18 interpret point 1?

T 116/18 is long but there is very little “working” showing how the Board arrived at the need for there to be “legitimate reasons to doubt that the purported technical effect would have been achieved”.

In 11.3.2, the TBA acknowledge that the EBA “did not refer to any of the plausibility standards identified by the board in its referring decision … [o]n the contrary, it used new legal terminology that had not been applied so far in the context of inventive step”.

They then put forward two possible reasons for this, suggesting that either the EBA considered all previous plausibility standards to be wrong and wanted them to be replaced with the new test or, in the alternative, the EBA considered all three requirements to reflect the same requirement and wanted to condense this requirement by that defined in the new test.

The second of these reasons seems odd. If the purpose of the referral was that the existing plausibility standards could lead to very different outcomes, it seems counterintuitive to then suggest that they may all “reflect the same requirement”.

In 11.8, the TBA suggested that the EBA’s view in G 2/21 “seeks to prevent speculative inventions” and that “the broader the application as filed, the more likely it is that the invention defined in it was speculative from the outset”.  Here, they appear to be ruling out “armchair inventions” but the TBA then goes on to suggest that this requires an assessment of the requirements of the new test to be made “based on the broadest technical teaching of the application as filed … with regard to the claimed subject matter”.

Based on this, the TBA conclude that for point 1 (“the technical effect encompassed by the technical teaching of the claimed invention”) to be met, “the purported technical effect … need only be conceptually comprised by the broadest technical teaching of the application as filed”.  The only example for what “conceptually comprised” might mean is if the skilled person “recognises that said effect is necessarily relevant to the claimed subject-matter”.  Such a standard is both broad and vague and is the first indication of how the TBA ended up where they did with their decision.  It also does not appear to be dissuading “armchair” inventions which are, by definition, conceptual.

In the absence of teaching away, what effect cannot be said to be “necessarily relevant” to a claimed invention?

How did T 116/18 interpret point 2?

When it comes to point 2 of the G 2/21 test, the TBA do not waste any time coming to an interpretation of how the technical effect may be “embodied by the same originally disclosed invention”

In a single paragraph (11.11) they conclude that this just requires asking the question “would the skilled person … have legitimate reason to doubt that the technical teaching at issue … is an embodiment of the originally disclosed invention?”.  In adopting this phrase, the Board revert to a definition of the ab initio implausibility standard that they provided in their referral to the EBA without providing any reasoning of how they got there and despite acknowledging that the Enlarged Board of Appeal “did not refer to any of the plausibility standards identified … in [the] referring decision”

Other aspects of T 116/18

T 116/18 then discusses various arguments put forward by the opponent/appellant relating to whether an application must contain data relevant to the purported technical effect. Such a discussion appears irrelevant to the purpose of G 2/21 which relates to scenarios where no (relevant) data is present. Even under the previous plausibility standards, there were scenarios where data in the application was not necessary to rely on post filed data.

In paragraph 11.13.1, the Board addresses the question of whether a positive statement regarding the purported technical effect is required in the application as filed. Here, they state that such a requirement would “correspond to, or at least come very close to” the very strict gold standard of direct and unambiguous disclosure developed for assessing added matter. 

Once again, this does not appear to have even been an issue prior to the G2/21 referral so it is unclear why the Board felt a need to address this point here.

Taking all of the above into account, it seems that even prior to the referral to the EBA, the Board in T 116/18 was tending towards the ab initio implausibility standard.  

There was already evidence of this in 12.5.1 of the Interlocutory Decision of T 116/18 from 11 October 2021. When considering whether the (potentially admitted) post-published evidence (D21) provided by the Patentee shows that there is an inventive step across the whole scope of the claims, the TBA stated that “in the absence of any counter-evidence from the appellant, the board sees no reason not to acknowledge a synergistic effect against C. suppressalis also for other insecticide compositions covered by claim 1”.  This was in response to argument that D21 was not valid proof of a synergistic effect against C. suppressalis for all insecticide compositions covered by claim 1, because formula Ia of the claim covered over 10 million compounds.

The Board’s approach is confirmed at 17.4.1 of T 116/18. When considering whether to admit D21 into proceedings for the assessment of inventive step, the Board refers to Example 3-a of the patent as demonstrating a synergistic insecticide effect. Example 3-a relates to only one species of insect and uses a compound according to claim 1 in combination with a compound which is structurally very similar to (but not the same as) the second compound of claim 1.

In reference to Example 3-a, the Board reasoned that the skilled person would have no legitimate reason to doubt that the synergistic effect would be maintained based on an example that related to a different compound to that claim showing efficacy against a single insect species.

It seems the Board were heavily influenced by, or at least chose to rely heavily on, the EBA’s use of the words “derivable”/”derive” in their decision. The Board felt that the EBA intentionally used this term to distinguish over the gold standard used for assessing added matter of “directly and unambiguously derivable”.

However, whilst choosing to not follow a gold standard added matter test may be considered a reasonable position to take, it still does not explain how the Board ended up at the other end of the spectrum, requiring that there had to be a legitimate reason to doubt that the technical teaching is an embodiment of the originally disclosed invention. 

Whilst it looks like in G 2/21 the EBA sought to find a compromise between the “gold standard” disclosure of a technical effect and ab initio implausibility, the Board in T 116/18 clearly did not.

How has G 2/21 been interpreted in other cases?

T 116/18 is just one of many decisions over the past nine months that has looked to G 2/21 for guidance.

To give some perspective on how this EBA decision is being adopted, we have looked a selection of other cases – from a variety of Technical Boards of Appeal – that have relied on G 2/21.

T 2735/19

The TBA stated that:

“The Board agrees with the respondent that, due to the direct action of FTD on DNA, TAS-102 could be expected to have a general effect on tumours and not to be limited to a single cancer type. Therefore, the consideration of the post-published evidence in document D11 confirming this effect on additional tumour types was in line with the principles established in G 2/21…”.

The Board felt that the disclosure of the action of FTD (an anti-tumour drug) meant that the technical effect was encompassed by the technical teaching in the originally disclosed invention. In essence, they applied an ab initio plausibility test that was met by the reference to the action of FTD.

T 0258/21

The TBA decided that the technical effect substantiated by post-published evidence was not to be taken into account for the assessment of inventive step, because it was considered that the effect was “neither contemplated nor even suggested in the original application”.

Once again, the Board applied the standard of ab initio plausibility. At no point did they see a need to assess whether there was teaching in the prior art, or the original application, that provided a legitimate reason to doubt the alleged technical effect. The original application’s failure to contemplate or suggest the technical effect was sufficient for post-published evidence to not be considered. As such, the application was found to lack inventive step.  

T 0681/21

For the main request in this appeal, the patentee argued that the technical effect of the claimed invention was improved softness and in particular a synergy due to the combination of silicone with a particular polymer (CPP).

The Board noted that “it is not in dispute that the application as filed does not relate to a synergistic effect arising from the combination of a silicone with CPP or any other component. Also the fact that the application as filed (page 2, lines 7-8) indicates the CPP to be a preferred cationic polymer without explaining the reason for this preference cannot foreshadow that the claimed combination would provide any type of synergism…Therefore, it follows from the above reasons that the alleged synergistic effect would not have been considered by the skilled person as being encompassed by the technical teaching of the application as filed and has to be disregarded… it can only be concluded that improved softening has not been convincingly proven across the entire scope of claim 1”. Thus the Main Request was considered to lack an inventive step.

The Board used the same standard when assessing whether post-published evidence could be relied on for auxiliary request 4. The Proprietor submitted that anionic silicone in combination with CPP provided unexpectedly better softening. The Board allowed the Proprietor to rely on such a technical effect, commenting that “skilled person reading the application as originally filed and having the common general knowledge in mind would derive therefrom as a technical teaching that the addressed improved silicone softness is especially obtained by using a combination with the components indicated as preferred, such as a CPP and an anionic silicone”.

This case has many similarities with T 116/18. An originally broad combination claim relying on post-published evidence to support a synergistic technical effect. Had the Board considered the relevant test for considering post-published evidence as being no legitimate reason to doubt the synergistic effect, it is difficult to see how they would have found against the Main Request. The fact that they did not consider the technical effect to be encompassed by the broad technical teaching of the application as filed shows that they were clearly applying an ab initio plausibility standard – i.e. the complete opposite to the Board in T 116/18.  

As it turns out, even when the post-published evidence was considered for auxiliary request 4, the Board found the claims to be obvious.

T 0728/21

The Proprietor provided post-published evidence showing optimisation of the dissolution of a claimed tablet composition, wherein the tablet comprised either 40% or 50% solid dispersion, along with other claimed compounds.

The Board commented that “in view of the consistent increase in dissolution from the decrease in the amount of the solid dispersion from 50% to 40% the Board considers it reasonable to assume that the same optimization… is achieved with the 34.1% of the solid dispersion as defined in claim 1 of the main request”.

The Board went on to state that “the application as originally filed (WO 2011/019413) explicitly addressed the dissolution of tablets comprising a solid dispersion as an aspect of the disclosed invention (see paragraph [0031]) and specifically described the claimed tablet composition as an embodiment of the disclosed invention. The effect of the optimization of the dissolution associated with the specific tablet composition defined in claim 1 of the main request may therefore in accordance with the principles established in G 2/21 be taken into account for the assessment of inventive step”.  

Once again, the language used indicates a standard of ab initio plausibility achieved by the fact that the technical effect of improved dissolution was explicitly disclosed in the application as filed.

T 0873/21

The Applicant relied on post-published evidence to demonstrate that the claimed combination provided improved insulin sensitivity, and in particular a synergistic interaction between claimed compounds A and B.

The Applicant argued that the therapeutic effect of the claimed combination would have been expected at the priority date because both compounds were each known at the priority date as useful for the claimed therapeutic application. The Applicant further argued that it was generally described in the original application that combination therapy according to the invention leads to an improved effect in terms of insulin sensitivity where monotherapy with either of A or B is insufficient.

The Board agreed that this general description in the original application was sufficient to derive the therapeutic synergistic effect that was substantiated in the post-published evidence (D16).  The data of D16 merely provided a quantification of the obtained improvement in insulin sensitivity that was already described in the original application.

Interestingly, as in T 0681/21, the Board considered the preferred embodiment to be relevant stating that “the synergistic effect relied on by the Applicant was encompassed by the technical teaching of the original application … and was embodied by the claimed combination since it was clearly the preferred combination in the original application”.    

In line with the appeal decisions already discussed above, the Board again used an ab initio plausibility standard for assessing admissibility of post-published evidence.

T 1045/21

When the patentee attempted to establish an inventive technical effect over the prior art, the Board noted that “[a]ccording to established case law, if comparative tests are chosen to demonstrate an inventive step on the basis of an improved effect over a claimed area, the nature of the comparison with the closest state of the art must be such that the alleged advantage or effect is convincingly shown to have its origin in the distinguishing feature of the invention compared with the closest state of the art. It rests with the proprietor to properly demonstrate that the purported advantages of the claimed invention have successfully been achieved (see G 2/21, point 26 or the reasons)”.

In this case the Board’s interpretation of G2/21 put the burden on the Proprietor to establish that the described technical effect is achieved in the application as filed.

This is arguably an even higher test than that of ab initio plausibility in that it requires that the technical effect is “convincingly shown” and “properly demonstrate[d]” in order for post-published evidence to be relied on.

T 2465/19

The Board overturned the Examining Division’s decision that the claims lack inventive step. The ED had concluded that the original description was silent about an advantage or beneficial/surprising effect of the distinguishing feature of the claimed invention, and thus the feature was not important and represented an arbitrary modification.

The Board commented that “the technical effect of an invention over the closest prior art need not be explicitly stated in the application, as long as it is derivable from the original application”. The Board stated this despite the Applicant not relying on post-published evidence to support the inventiveness of the claims.

Thus the BOA clearly follow G 2/21 in their Decision.

T 1891/21

This case concerned claims directed to an electrode containing manganese, wherein the number of manganese atoms compared with the total number of manganese atoms and iron atoms is >50% and <100%. The proposed technical effect was a higher initial coulombic efficiency and the patentee submitted post-filed evidence showing that the increase in coulombic efficiency is greater when the relative number of manganese atoms is high.

The Board commented that all of the examples related to manganese containing materials with a high number of manganese atoms and the patent discloses that a high relative number of manganese atoms of >50% and <100% leads to improved initial coulombic efficiency.

Based on this the Board concluded that the post-published data “supports the indicated general teaching in the patent that a high proportion of manganese contributes to an improved initial coulombic efficiency”. The Board also noted that “furthermore, no counter-evidence is available”.

Perhaps unsurprisingly, the Board concluded that the indicated technical effect can be derived from the application as filed.

Whilst this case provides yet another example of the Boards of Appeal applying an ab initio plausibility standard to consider post-published evidence it is interesting for the fact that they also referred to no counter-evidence being available. 

The notable difference between this decision and T 116/18 is that here the Board required some indication of the technical effect in the patent but still suggested that may not be enough if counter-evidence is available. In contrast, T 116/18 suggests that even in the absence of a general teaching of the technical effect in the original application, only counter-evidence (or something providing “legitimate reason to doubt that the technical teaching at issue”) can prevent a patentee relying on post-published evidence.

T 0885/21

Finally, the claims of the Patent were directed to antibody conjugates having the alleged technical effect of providing optimised characteristics such as homogeneity.

When considering whether to admit post-published evidence showing these advantageous effects, the Board stated that “the patent substantiates the suitability of the defined conjugates for therapeutic use and indicates that the claimed invention is aimed at optimized conjugates for therapeutic application in terms of inter alia homogeneity, pharmacokinetics, stability, toxicity and efficacy….Taking account of the disclosure of the invention in the patent the Board considers that the effects described in document D51 are encompassed by the technical teaching and embodied by the disclosed invention and may in accordance with the principles confirmed in G 2/21 (see section 94) be relied upon by the proprietor for inventive step”.

Here again, the Board applied a standard of ab initio plausibility.  They required at least an indication towards the technical effect in the original application as opposed to assessing only whether there was “no legitimate reason to doubt” the presence of the technical effect.

Concluding remarks

This summary only covers a selection of the decisions that have cited G 2/21 but it shows that T 116/18 presently appears to be an outlier in the way the Enlarged Board of Appeal’s decision is being interpreted.

Unfortunately, what is clear is that the original fears of practitioners when G 2/21 was published have come true; there is no aligned approach on the test for assessing whether post-published evidence may be relied upon to support an alleged technical effect.

This partition of interpretation of G 2/21 is also reflected in across EPC national courts. For example, in Sandoz v BMS, Lord Justice Arnold of the UK Court of Appeal (the “CoA”) adopted an approach of ab initio plausibility, and commented on G 2/21 that “the core question for inventive step is whether the technical effect relied upon by the applicant or patentee was derivable from the application at the filing date in the mind of the skilled person with the CGK”. The UK CoA found the Patent invalid for lack of sufficiency and inventive step. Using the same facts, the Dutch CoA found the Patent to be valid, applying a less strict standard for assessing the admissibility of evidence in support of inventive step.

If you are looking for some form of guidance then the general trend does not appear to be aligned with T 16/118. As such, wherever possible (and in line with the majority of other jurisdictions around the world) an application should include at least some pointer to the technical effect that will be relied upon for supporting the presence of an inventive step.

This article is for general information only. Its content is not a statement of the law on any subject and does not constitute advice. Please contact Reddie & Grose LLP for advice before taking any action in reliance on it.

Saved Staff
Staff member

Remove all

Call +44 (0)20 7242 0901
Call +44 (0)1223 360 350
Call +49 (0) 89 206054 267
Call +(00) 31 70 800 2162
This field is for validation purposes and should be left unchanged.