Reasons for an EBA referral on plausibility, post-published evidence and problem-solution

As recently reported here on IPKat, we are imminently expecting a referral to the Enlarged Board of Appeal (EBA) on the issue of using post-filed data to support plausibility for inventive step. The potential for a referral on this topic may have come as a surprise to some, given the established case law of the Board of Appeal on plausibility and post-filed evidence (Case Law of the Boards of Appeal, I-D-4.6). However, as highlighted in the preliminary decision of the Board of Appeal in the case (T 116/18), the Boards of Appeal are not fully harmonious on this issue. Part of the reason for the divergence of some Boards is the preference of the EPO to consider plausibility in the context of inventive step as opposed to under the umbrella of sufficiency (as is preferred in the UK courts). Particularly, for some Boards of Appeal, the requirement for the technical effect of a claimed invention to be rendered plausible in the application as filed, is incompatible with the problem solution approach for inventive step.  

Case background: Insecticides and synergy 

Not one of her favourite patent cases...

The patent (EP2484209) in T 116/18 relates to an insecticide comprising thiamethoxam and a compound having "formula 1a". The difference between claim 1 of the Main Request and the closest prior art was identified as the selection of thiamethoxam and formula 1a in combination. The purported technical effect of the selection was synergy between thiamethoxam and compounds falling under definition of formula 1a.

The patent included examples testing the interaction between thiamethoxam and various compounds of formula 1a. During appeal proceedings, the patentee provided post-published documents which contained further data showing synergy between thiamethoxam and compounds of formula 1a. The opponent argued that the effect of synergy between thiamethoxam and compounds of formula 1a had not been made plausible in the application as filed. As such, it was argued, the post-published documents should not be taken into account.

At appeal proceedings, the Board of Appeal concluded that the admittance or not of the post-published evidence was decisive for inventive step. 

Conflicting decisions on the admissibility of post-published evidence to support inventive step

The Board of Appeal in T 116/18 termed the requirement for a technical effect to be rendered plausible in the application as filed, as "ab-initio plausibility". Legal basis for this requirement, the Board noted, could be found in T 488/16 (r. 4.2), T 1329/04 (r. 12), and T 433/05 (r.28). 

The Board of Appeal, however, also noted conflicting decisions in which the criterion for ab-initio plausibility had not been applied. Most significantly, the Board of Appeal in T 2371/13 (3.3.10) found that a lack of ab initio plausibility was not a reason to disregard post-published evidence. The Board of Appeal in T 31/18 came to a similar view. For both of these Boards, the requirement for a technical effect to be plausibly demonstrated in application as filed was incompatible with the problem solution approach. This was because the closest prior art may not be known to the applicant at the time when the application as filed. Ab initio plausibility may thus require an applicant to provide evidence for all the technical features in the application as filed on which an inventive step argument could possibly be based. Given the potentially extensive number of technical features of an invention that the patentee may later have to rely on for inventive step, the Boards concluded that this could not be expected of the applicant. 

In view of the decisions in T 2371/13 and T 31/18 (as well as T 1422/12), the Board of Appeal concluded that there was divergent case law on the use of post-published evidence to support inventive step. 

The burden of proof for plausibility

In the preliminary opinion, the Board of Appeal identified a further question that needed to be addressed with regards to post-published evidence. Particularly, whether the burden of proof rested with the patentee or the opponent. In T 1329/04, for example, the Board of Appeal evaluated whether the patentee had demonstrated ab initio plausibility. In T 184/16, by contrast, the Board examined whether there were any reasons from the common general knowledge to assume that the alleged technical effect had not been obtained, i.e. "ab initio implausibility" (see also IPKat post and comments, on the related concepts of "negative" and "positive" plausibility). 

Furthermore, in seeking to demonstrate implausibility, the Board of Appeal in the present case asked, was it equitable to allow an opponent to rely on post-published evidence if the patentee was denied reliance on post-published evidence to support plausibility?

Questions to be answered

The Board of Appeal concluded that a number of questions on plausibility and post-filed evidence required a referral to the Enlarged Board of Appeal, summarised as follows: (1) Should post-filed data be disregarded if these data are the sole evidence for a technical effect. (2) If yes to (1), can post-filed data be taken into account if there is ab initio plausibility. (3) If yes to (1), can post-filed data be taken into account if there is ab initio implausibility. 

Inventive step or sufficiency?

One common confusion is whether plausibility is most properly considered in an analysis of sufficiency or inventive step. The case law on plausibility in the UK is primarily directed to the sufficiency requirement (Warner Lambert v Actavis (IPKat)), but has also been considered in the context of inventive step (Takeda v Roche (IPKat)). 

In the present case, the Board of Appeal considered it straightforward that plausibility should be considered in the context of inventive step (Article 56 EPC) as opposed to sufficiency (Article 100(b) EPC). As provided by G1/03 (r. 2.5.2), if the technical feature is included in the claim as a functional feature then it is a matter of sufficiency, whereas if the technical feature is not included in the claim, it is a matter of inventive step (see also the EPO Guidelines for Examination (F-III-12)). Given that the effect of synergy between the thiamethoxam and compounds of formula 1a was not included in the claim, the Board of Appeal concluded that this was a matter of inventive step.

The opponent in the case argued, however, that the arguments relating to plausibility under the context of sufficiency remained valid, given that technical effects that are not explicitly mentioned in the claims might be pertinent to other requirements of the EPC. However, sufficiency was not discussed at the hearing. Notably, in one of the Board of Appeal decisions cited as divergent to the approach requiring ab initio plausibility, it was noted that the issue of plausibility was one that should properly be considered under Article 83 EPC (Sufficiency) (T 2371/13). However, it remains unclear whether the expected referral, should it be deemed admissible by the EBA, would have any relevance to plausibility in the context of sufficiency (Case Law of the Boards of Appeal, II-C-7.2).

We now await the written decision of the Board of Appeal and confirmation of the referral. 

Further reading

Plausibility at the EPO

• Lost in translation: Are animal models predictive of a treatment effect?
• Second medical use dosage regimen claim successfully traverses both insufficiency and "obvious-to-try" attacks (T 0799/16)
• When does pre-clinical data plausibly support a therapeutic effect? (T 966/18)
• New referral expected to the EBA on the use of post-published data to support inventive step

Plausibility in the UK

• No pain, no gain: Plausibility in Warner-Lambert v Actavis
• Supreme Court unanimously dismisses Warner-Lambert's painful Lyrica appeal
• Takeda v Roche: "Is it plausible? Is it true?"