T 655/24 - credibility test for improved technical effect applied to determine inventive step for a single amino acid change following G 2/21
Summary
Board 3.3.04 in T 655/24 applied a credibility test that is more familiar under consideration of sufficiency. The Board found that assessment of inventive step can only take an improved effect into account if that improved effect is credible from the disclosure in the application as filed. Of note, the Board explicitly notes that this is different from the approach taken by Board 3.3.02 in T 1989/19, T 2716/19 and T 840/22.
Inventive step was acknowledged on the basis that even though the claimed Fc mutations were known:
- the skilled person could not have had a reasonable expectation that the advantageous features of these variants would be maintained when they were combined in a single Fc region and
- the advantageous features were credible from the disclosure in the application as filed.
The decision highlights important drafting and practice points for users of the EPO system, particularly in the antibody field.
Background
The claims of the Patentee’s main request related to a human IgG1 Fc region defined functionally (by plasma clearance rate and by not binding to any Fcγ receptors) as well as structurally by a combination of amino acid residues (the ‘FEA variant’). It was undisputed that the Examples of application as filed discussed the particular advantages of the FEA variant (Reasons 22 of the decision). It was also established that closest prior art documents (D10 or D21) disclosed an FE variant. The difference between the FE variant and the claimed variant was that the latter includes an additional single amino acid change, namely modification of D265A (Reasons 55 of the decision).
Key issue: inventive step and G 2/21 (post-filed data)
The key issue in this decision was whether the claimed subject-matter involved an inventive step over the closest prior art.
The patentee, referring to Figure 3A in the application as filed as well as additional post-published data, considered that the claimed combination of amino acid residues resulted in reduced CD69 expression, which was an improvement compared to the FE variant and showed that the FEA variant Fc region induced less effector function.
To determine whether the post-published data could be taken into account, the Board applied Enlarged Board of Appeal decision G 2/21:
"A patent applicant or proprietor may rely upon a technical effect for inventive step if the skilled person, having the common general knowledge in mind, and based on the application as originally filed, would derive said effect as being encompassed by the technical teaching and embodied by the same originally disclosed invention." (emphasis added)
Applying these principles, the Board did not consider that an improvement of an effect would meet the test of G 2/21 merely because the effect itself, but not the improvement, was shown to be achieved in the application as filed (Reasons 58 of the decision). In this context, the Board took the position that post-published evidence could only be taken into account when the improved technical effect is already credible from the disclosure in the application as filed itself.
In their reasoning, the Board (3.3.04 in this case, chaired by Chakravarty) explicitly acknowledged that this is a “different approach” to that taken in T 1989/19, T 2716/19, and T 840/22 (all originating from Board 3.3.02 and all chaired by Müller). The Board further appeared to criticise these three decisions for not giving any explanation for why they depart from “the basic legal principle that an invention must have been made at the effective date of the patent or patent application”.
Those earlier decisions had reasoned that once a technical effect itself was derivable from the application as filed, an improvement of that effect was "implicitly derivable" and could therefore be established by post-published data. For instance, in T 1989/19, the Board explicitly stated “[i]n the Board's view, once the above-mentioned criterion of the derivability of a technical effect is met, this applies equally to the improvement of this effect”. T 2716/19 reasoned similarly that an improvement in yield was "encompassed by the technical teaching" where the application as filed focused on the relevant reactants and the skilled person would have "immediately recognised" an improvement in yield as a "fundamental objective." T 840/22 relied expressly on T 1989/19 to hold that where the derivable effect was corrosion resistance, the patentee could rely on an improvement in that effect over a specific prior art document.
The Board pointed out that the approach in these earlier decisions would allow an applicant to rely solely on post-published evidence to establish inventive step by filing an application disclosing an effect that was already known from the prior art and subsequently invoking an improvement of said effect as "implicitly derivable" from the application as filed. The Board said that this “directly contradicts the "first to file" principle enshrined in the EPC” and “goes against the principle that patents should not be granted for subject-matter whose inventive contribution is identified only after their effective date”.
Reviewing the data in the application as filed, the Board considered that the improved effect of decreased CD69 release of the FEA variant compared to the FE variant was credibly disclosed in the application as filed. Therefore, the objective technical problem was formulated as the provision of a protein having an improved Fc region and the Board decided that the skilled person would not have had a reasonable expectation of success that the combination of the FE variant with D265A would result in an improvement. Given this evidence in the application as filed it, was “not strictly necessary to consult the evidence in documents D25 and D26”, but the Board nevertheless noted that their disclosure “supports” the improved effect shown in the application as filed (Reasons 78 of the decision).
Interestingly, the Board went on to say that even if aiming to provide an alternative Fc variant which had similar functionality as the known and tested variants (D265A or FES), the skilled person could not have had a reasonable expectation that the advantageous features, i.e. reduced or no effector function, of these variants would be maintained when they were combined in a single Fc region. Therefore, despite the detailed discussion of G 2/21 and the Board’s interpretation that the improved effect must be credible, the Board made the point that defining the problem less ambitiously led to the same outcome.
Inventive step was acknowledged whichever way the technical problem was defined, and the case was remitted to the opposition division with the order to maintain the patent on the basis of the main request (originally filed as AR6).
Comment
This decision is helpful for applicants when formulating a prosecution strategy since it confirms that the EPO may award an inventive step based on the unpredictable function of a combination of amino acid mutations even where this only differs from the prior art by one amino acid.
The fact that the technical effect in this case was considered under the ground of inventive step in this decision goes back to a different decision of the Enlarged Board of Appeal decision; G 1/03. This decision primarily concerned undisclosed disclaimers, but also held the following:
“If a claim comprises non-working embodiments, this may have different consequences… Either there is a large number of conceivable alternatives and the specification contains sufficient information on the relevant criteria for finding appropriate alternatives over the claimed range with reasonable effort. If this is the case, the inclusion of non-working embodiments is of no harm…
If this is not the case and there is lack of reproducibility of the claimed invention, this may become relevant under the requirements of inventive step or sufficiency of disclosure. If an effect is expressed in a claim, there is lack of sufficient disclosure. Otherwise, ie if the effect is not expressed in a claim but is part of the problem to be solved, there is a problem of inventive step…”
Therefore, whether the technical effect is recited in the claim will determine whether the effect falls to be considered under inventive step or sufficiency. This distinction of inventive step versus sufficiency is important because the scope of reliance on post published evidence generally is much narrower under sufficiency of disclosure compared to the situation under inventive step. This was confirmed by the Enlarged Board in G 2/21 where they noted that attaining the claimed therapeutic effect is a functional technical feature characterising a medical use claim and that in order to meet the requirement that the disclosure of the invention be sufficiently clear and complete for it to be carried out by the person skilled in the art. The Enlarged Board said:
“the proof of a claimed therapeutic effect has to be provided in the application as filed, in particular if, in the absence of experimental data in the application as filed, it would not be credible to the skilled person that the therapeutic effect is achieved. A lack in this respect cannot be remedied by post-published evidence." (G 2/21, Reasons 77).
The approach in T 655/24 thus appears to muddy the water. A “credibility” test is much more familiar under sufficiency but here it was applied under inventive step as a means for determining compliance with the test in G 2/21. Specifically the G 2/21 requirement for the improved effect to be “encompassed by the technical teaching and embodied by the same originally disclosed invention” was determined by reference to whether the improved effect had been “credibly disclosed” in the application as filed.
The Board’s departure in this case from the approach in T 1989/19, T 2716/19, and T 840/22 did not lead to a different outcome – the claimed subject-matter was still considered to be inventive. Moreover, the decision was curiously only marked with internal distribution code “D” meaning that it was not further communicated among the EPO staff. This code is determined by the presiding Board and typically is reserved for decisions containing no new points of law or being of no particular legal significance. It is perhaps worth noting that the earlier, and specifically criticised, T 1989/19 had been distributed to the Chairmen of the Chambers (distribution code “C”).
Therefore, the Board chose to depart from the approach in existing decisions whilst also not widely circulating this fact. It remains to be seen whether the approach of Board 3.3.04 in T 655/24 is followed in other cases by the same or other Boards and whether this conflict may ultimately require resolution by another referral to the Enlarged Board. Applicants may be disappointed to see a decision which, given the disagreement with three earlier decisions, creates legal uncertainty with regard to the allowability of reliance on post-filed data. It also remains to be seen how the case law develops in this area, particularly where comparisons with prior art documents uncovered later (e.g. during the search) are required to clearly demonstrate the advantages of the invention over the prior art.
Given the decision at hand, filing strategies built on demonstrating an improvement solely through post-published experiments now carry a risk before Board 3.3.04. As far as possible, applicants should give careful thought at the drafting stage as to what the closest prior art might be and should take particular care to include comparative data in their applications as filed that credibly demonstrate any improvement over the prior art they may later wish to rely upon for inventive step. Including well-designed comparative experiments at the filing stage, even if preliminary, is advisable.
Suitable claim language defining demonstrated technical effects as functional features, as well as language combining such functional features with structural features should also be drafted.
This is particularly applicable to the antibody field where inventive step is often argued based on an improvement.
Conclusion
In short, a single amino acid change can confer an inventive step, but applicants should be in possession of the resulting technical effect at the time of filing in order to provide the greatest chance of success during prosecution.
