Interpretation of G 2/21: Inventive step may be supported solely by post-published data (T 0116/18)

The minutes of oral proceedings have been published from the referring Board of Appeal case behind G 2/21 (T 0116/18). The minutes are brief but confirm the Board of Appeal's decision to acknowledge the inventive step of the claimed invention and to dismiss the appeal. The inventive step of the claimed invention was acknowledged based on the post-published data submitted by the Patentee. The minutes therefore support an interpretation of G 2/21 that permits inventive step to be based on a purported technical effect that is neither explicitly mentioned nor supported by data in the application as filed. 

G 2/21 case catch-up: The evidence standard for inventive step (it's not about plausibility)

At issue in the case referred to the Enlarged Board of Appeal (EBA) in G 2/21 were product claims to a new combination of known insecticides (EP 2484209). The application included data demonstrating the inventive, synergistic effect of the insecticide combo against two species of moth (T 0116/18).

At opposition, the Opponent submitted data showing that the insecticide combo did not in fact have a synergistic effect against one of the exemplified moth species (P. xylostella) at certain weight ratios. The Opponent argued that the purported technical effect of the claimed invention could thus not be supported. In response, the Patentee filed their own new data showing the efficacy of the insecticide combo against a third species of moth (C. suppressalis). The Patentee argued that the technical effect of the invention was the narrower effect of the insecticide combo against C. suppressalis. Importantly, the application as filed mentioned C. suppressalis but did not provide any efficacy data against this particular moth species for compounds falling under the scope of the claims.

Both the Patentee and the Opponent in T 0116/18 therefore wished to rely on post-published data for their inventive step argument. The Opponent wished to rely on post-published negative data, showing that a purported technical effect was not achieved. In response, the Patentee wished to rely on post-published data supporting an alternative inventive step argument for the claimed subject matter. 

How many moths needed for an invention? 

In its interlocutory decision, the Board of Appeal sought clarity from the EBA on whether post-published evidence may be taken into account for inventive step. Importantly, the Board of Appeal also came to interlocutory conclusions on inventive step for when the post-published was either ignored or taken into account. The Board of Appeal concluded that:

• If the data in the application as filed and the Opponent's post-published data were taken into account, inventive step (synergy against P. xylostella) could not be acknowledged (r. 12.4.9)

• If the data in the application as filed, and both the Opponent's and Patentee's post-published data were taken into account, inventive step (synergy against C. suppressallis) could be acknowledged (r. 12.5.1). 

The decision of the EBA in G 2/21

In G 2/21, the EBA decided that:

"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" (Headnote II, emphasis added)

The EBA did not give an opinion on how the principle that a technical effect is "encompassed and embodied" by the application as filed should be applied to the facts of the case in G 2/21. Crucial to the understanding of G 2/21 is thus how the Board of Appeal would implement the decision.

Minutes from Oral proceedings in T 0116/18

The EBA remitted the case in G 2/21 back to the Board of Appeal for a decision on inventive step. It was therefore left to the Board of Appeal to apply the "abstract principles" of G 2/21 to the facts of the case in question. As mentioned above, the Board of Appeal in T 0116/18 had already reached its own conclusions regarding inventive step for the different scenarios in which the Patentee's and/or Opponent's post-published data could be considered. According to the minutes of oral proceedings, the Board of Appeal found that these conclusions could not be revisited in view of the binding effect of the interlocutory decision. Further additional data from the Opponent, relating to C. suppressallis, could therefore not be taken into account (page 3, last paragraph). 

The minutes of oral proceedings do not record the details of the subsequent discussion over the interpretation of G 2/21. The minutes merely record the Board of Appeal's conclusion that the Patentee could "rely on the synergism against C. suppressallis" shown in their post-published evidence for technical effect. On this basis, the Board of Appeal found the claims inventive and dismissed the appeal. 

EPO and UK divergence on "plausibility"

The UK courts have also recently interpreted G 2/21. In Sandoz v BMS ([2023] EWCA Civ 472). Lord Justice Arnold interpreted G 2/21 as supporting an ab initio plausibility standard for both sufficiency and inventive step (i.e. that an invention and its technical effect must be rendered plausible in the application as filed in view of common general knowledge). However, the UK is increasingly out of step with the EPO on "plausibility". Most importantly, the UK continues to treat "plausibility" as almost a legal test in and of itself (IPKat), and consequently fails to make a distinction between the evidence standard for inventive step and sufficiency (IPKat). By contrast, the interlocutory decision in T 0116/18 confirmed the very well-established EPO case law that the sufficiency (Article 83 EPC) of a technical effect is not at issue, if the technical effect is not a claimed feature of the invention (G 1/03, r 2.5.2). In G 2/21, the EBA also dismissed "plausibility" as a catch-word for the evidence standard for a purported technical effect (G 2/21, r. 58), finding that it is not a legal test having basis in the EPC.

Final thoughts

We now await publication of the written decision from the Board of Appeal which will hopefully provide more guidance on how G 2/21 should be interpreted. However, we can already draw a few conclusions from the minutes:

First, by accepting the Patentee's post-hoc inventive step argument, it appears the Board of Appeal was comfortable finding inventive step based on a purported technical effect for which there was no direct data provided in the application as filed. This is a potentially important development in the case law, with implications across technical fields (IPKat). Technical effect in T 0116/18 was acknowledged with respect to a moth species, C. suppressalis, which was only mentioned in the context of data relating to insecticides that did not fall under the scope of the claim (Example 4). Therefore, it appears from the Board of Appeal decision in T 0116/18 that a purported technical effect does not need to be mentioned in the application as filed in order for it to be considered "encompassed and embodied" by technical teaching of the application as filed (G 2/21). Instead, a purported technical effect may be solely supported by post-published data. 

Second, the minutes of oral proceedings from T 0116/18 also confirm the apparent divergence of the UK and the EPO on the evidence standard for inventive step. It is difficult to align the interpretation of G 2/21 by Lord Justice Arnold in Sandoz v BMS with the Board of Appeal decision in T 0116/18 reported in the minutes of oral proceedings. As previously discussed, this discrepancy has at its heart the failure of the UK courts to distinguish between the evidence standard for inventive step and sufficiency (see Sandoz v BMS, para. 50 "subsequent data cannot be a substitute for sufficient disclosure in the specification") (IPKat). 

These issues and more will be discussed at CIPA Congress in Cambridge on 3 October. PatKat will be joining Ian Jones (GJE), James St.Ville KC (8 New Square) and Sarah J. Rodriguez (Finnegan), in a panel discussion on "plausibility" in all its guises, chaired by Greg Corcoran (ASML).

Further reading

• Plausibility demystified - a review of EPO case law before G 2/21 (Feb 2023)
• G 2/21: Is the technical effect embodied by the invention as originally disclosed? (March 2023)
• To encompass and embody: Applying the abstract principles of G 2/21 (March 2023)
• Applying G 2/21: Preliminary opinion from the referring Board of Appeal on post-filed evidence appeal (T 0116/18) (June 2023)
• UK divergence from the EPO on plausibility (Sandoz v BMS), Part 1: Is the "plausibility" test the same for both sufficiency and inventive step? (July 2023)
• UK divergence from the EPO on plausibility (Sandoz v BMS), Part 2: Interpretation of G 2/21 (July 2023)
• The relevance of G 2/21 to machine learning inventions (T 2803/18) (Aug 2023)