En mars 2023, la Grande Chambre de recours de l’Office européen des brevets (OEB) a défini, dans sa décision G2/21, deux critères déterminant l’admissibilité des données post-dépôt dans l’analyse de l’activité inventive et de la suffisance de description. Depuis, une trentaine de décisions ont été rendues sur la recevabilité de telles données, qu’elles aient été acceptées ou écartées. Cet article présente les décisions les plus significatives, afin de mieux comprendre comment les Chambres de recours ont interprété les critères de G2/21 selon les domaines techniques concernés et les particularités de chaque affaire.
Les données expérimentales jouent un rôle essentiel pour établir l’activité inventive et la suffisance de description d’une invention. Elles permettent de démontrer, tant au regard de la personne du métier que des examinateurs, que l’invention revendiquée constitue bien une « solution technique à un problème technique » — répondant ainsi simultanément à deux des quatre critères fondamentaux de brevetabilité.
Various criteria for the admissibility of post-filing data have been applied in the past: some were highly unfavorable to the patent proprietor (requiring the original application to explicitly contain the information supporting the newly alleged effect and/or the successful resolution of the technical problem), while others were far more lenient—for example, the “ab initio implausibility” criterion: if the skilled person, the examiner, or the technical judge had no reason to doubt that the effect could be achieved upon reading the application, then the data could be accepted, even if the application or the prior art did not suggest this effect. Finally, some examiners or technical judges focused on whether it was “plausible,” in light of the application’s teaching and the skilled person’s general knowledge, that the newly alleged technical effect could be achieved by the claimed invention (the “ab initio plausibility” criterion); if so, the new data could be taken into account.
In March 2023, the EPO’s Enlarged Board of Appeal issued its Decision G2/21 to clarify the conditions under which post-filing data may be admissible when assessing inventive step and sufficiency of disclosure. The decision notably emphasized that the key issue is what the skilled person would have understood from the overall teaching of the application, considering their general knowledge (points 71 and 72 of G2/21). The Enlarged Board also formulated a test to evaluate the admissibility of post-filing data. According to this test, such data should only be considered in patentability assessments if the skilled person, based on their general knowledge and the content of the application as filed, would have regarded the effect in question as being “encompassed” in the technical teaching of the application as filed (criterion 1), and as “embodied” in the invention as originally disclosed (criterion 2) (point 94 of the decision as published in French). Thus, the first criterion requires analyzing the entirety of the application’s content (including passages not describing the invention), while the second focuses more precisely on the invention as originally disclosed (often the claimed invention).
Though presented as essential for determining the admissibility of post-filing documents, these two criteria required clarification. For example, the relevant date for assessing the “common general knowledge” available to the skilled person was not specified in Decision G2/21. Furthermore, it did not indicate whether the technical teaching of the application should be interpreted strictly (i.e., limited to results and statements explicitly disclosed in the application), or whether it could be generalized—and if so, to what extent. Lastly, the crucial term “embodied” chosen by the Enlarged Board appeared somewhat opaque (translated in G2/21 as “incarnated” and “forming part of,” and rendered as “implemented” in certain decisions). The Enlarged Board itself referred to these concepts as “abstract” and ultimately acknowledged that the interpretation of these criteria “may indeed, to some extent, be influenced by the technical field of the claimed invention” and by “the circumstances of each case” (point 95 of G2/21).
T116/18
These clarifications were provided by Board of Appeal 3.3.02 in decision T116/18, which concluded the case that gave rise to the referral to the Enlarged Board and thus to Decision G2/21.
In its opinion issued on 28 July 2023, Board of Appeal 3.3.02 clarified the following points:
- The relevant date at which the skilled person must assess the content of the patent application is its filing date; accordingly, their general knowledge must, in principle, be limited to what existed on that date (point 11.2), and not extended to knowledge available at the time of the dispute.
- The technical teaching of the application must be understood in the broadest sense, as anything that can be derived from the application; this does not mean limiting the assessment to a direct and unambiguous disclosure of the relevant information, an approach reserved for the evaluation under Article 123(2) EPC (points 11.8 and 11.13).
- The technical effect to be considered in the inventive step analysis does not need to be explicitly described or supported by data; it must be “conceptually encompassed” in the teaching of the application taken in its broadest sense (criterion 1) and “necessarily implemented” by the claimed subject-matter (criterion 2) (point 11.10).
- To assess whether the technical effect satisfies criterion 2 of G2/21, the key question is whether the skilled person, with the general knowledge available at the filing date and relying on the content of the application as filed, would have had a legitimate reason to doubt that the alleged technical effect could be achieved by the claimed invention (point 11.14). The burden of establishing such doubt often lies with the opponent (point 17.4.2).
This notion of the “absence of legitimate doubt,” used to satisfy criterion 2 of G2/21, has been adopted by other Boards of Appeal, primarily in the assessment of sufficiency of disclosure (see, for example, T728/21, T0025/20, T0853/22, and T0197/22).
Since the publication of G2/21, numerous decisions have addressed whether to admit or reject post-filing data. We have reviewed them in detail to assess whether the criteria outlined in G2/21 and T116/18 have been applied by the Boards of Appeal, and if so, what impact they have had on ongoing proceedings.
1/ Assessment of Inventive Step
In the inventive step analysis, post-filing experimental data may be submitted to support two key stages of the EPO’s well-known problem–solution approach: (i) the definition of the technical problem to be solved, and (ii) its resolution by the claimed invention. Practitioners are well aware that it is possible, during examination or opposition proceedings, to rely on a technical effect different from the one originally proposed in the application—for example, when new prior art is identified and cited. In such cases, the technical problem to be solved is often framed as the improvement of this effect over an alternative solution disclosed in the prior art. Fortunately, it is not required that the application include experimental data proving this improvement, since the most relevant prior art may not have been known at the time of filing, and therefore necessary comparisons could not have been made. Moreover, the effect in question may not have been studied—let alone demonstrated—by the inventors at the time of filing, even though it may be intrinsically achieved by the invention (which itself has not changed).
In this context, it is important to assess whether post-filing experimental data can be taken into account in the evaluation of inventive step, as this may allow for a more favorable formulation of the objective technical problem and/or its solution for the patent proprietor. As we will see, the key question is often whether the technical effect demonstrated by the data is (or is not) described or suggested in the application.
1.1. Disclosed Technical Effect but Improved
The originally filed patent application generally contains no data showing that the invention provides a better effect than the solution proposed in the closest prior art. Nevertheless, during examination or opposition, the proprietor often argues that the technical problem solved by the invention is not merely to achieve the effect mentioned in the application (which would make the invention a mere "alternative" to existing means), but rather to improve that effect compared to existing means—based on appropriate experimental data submitted during the proceedings. The Boards of Appeal have had many opportunities to rule on the admissibility of such a "change" in the technical problem.
T0258/21
This decision of 23 July 2023 concerns the use of clevidipine to treat hypertension in patients who have suffered a stroke. The original application proposed using this drug to treat any type of vascular accident (whether hemorrhagic or ischemic). During the proceedings, the proprietor limited the patient group to those who had suffered an ischemic stroke, although the application did not include experimental data showing clevidipine’s effect on this specific patient group. During opposition, the patentee submitted additional data demonstrating not only that clevidipine was effective in these patients, but also that it provided better treatment outcomes and fewer side effects than other known antihypertensive agents, particularly for this patient group. According to the patentee, the technical problem to be solved should therefore consider the improvement in patient outcomes compared to previously described drugs. In this case, Board 3.3.07 refused to revise the technical problem or accept the comparative data, on the grounds that the improved effect in ischemic stroke patients was not mentioned in the original application, which instead proposed using clevidipine to treat all types of vascular accidents (hemorrhagic or ischemic). The improved effect in ischemic stroke patients was therefore not “derivable” from the original application (page 10). The Board also concluded that the additional data, even if accepted, would not have demonstrated a genuine improvement, as the results provided by the proprietor were not convincing (page 10). Whether or not the data were taken into account thus had little impact on the outcome of the case—possibly explaining why criteria 1 and 2 of G2/21 were not thoroughly analyzed in the decision. The invention’s inventive step was ultimately not recognized, since earlier documents already suggested using clevidipine to treat patients with either ischemic or hemorrhagic stroke, and the patentee could not rely on a surprising treatment effect in these patients—except by raising an objection of insufficiency of disclosure (page 14).
T2716/19
This decision, published on 10 January 2024, concerns a method for manufacturing 9-[2-(phosphonomethoxy)propyl]adenine (PMPA) involving the reaction of specific magnesium oxides. The application as filed identified these magnesium oxides as preferred but did not provide comparative data to demonstrate how their use improved the method. The opponent argued that the application did not disclose any technical effect associated with choosing the claimed magnesium oxides. During opposition, the patentee submitted experimental data showing that using these oxides increased the yield of the PMPA manufacturing reaction and thus argued that the technical problem to be solved was to identify an improved manufacturing process, particularly with respect to yield. Board 3.3.02 noted that the two claimed oxides were indeed identified as preferred in the original application, and that one example described how to implement the invention using these two oxides, including the yield associated with this method. The skilled person would therefore have “immediately recognized that improving the yield of the target product, here PMPA, was a fundamental objective of the method” (pages 11–12). The technical effect of increased PMPA yield is thus, according to the Board, “encompassed” and “embodied” in the originally described invention, in line with the requirements of G2/21. As a result, the change in the technical problem and the additional data demonstrating the yield improvement were accepted, and the inventive step of the claimed method was ultimately recognized.
T1551/22
This decision, dated 29 January 2024, concerns a cosmetic composition containing solid particles and a surfactant. The patentee submitted post-filing data showing that this composition foams faster and more abundantly than known compositions in the prior art. The patentee requested that the technical problem not be limited to providing an “alternative” composition but rather an “improved” one compared to the prior art, based on this new data. Board 3.3.07 accepted this reformulation of the technical problem, as the application as filed did state that the objective of the invention was to rapidly produce a rich foam, and that the composition did indeed achieve this (examples 2 and 3). This improved effect is thus, according to the Board, “encompassed” and “embodied” in the originally described invention, in accordance with the requirements of G2/21. The post-filing data were therefore admitted, and the inventive step was acknowledged.
It therefore appears possible to have additional data accepted and to revise the technical problem retrospectively to include the improvement of a technical effect—provided that the original application demonstrated the effect in question (supported by results). In decision T1989/19 (to be discussed in more detail below), Board 3.3.02 explicitly endorsed this practice by explaining that a technical effect, within the framework of an invention, is always intended to be “improved.” Thus, when a technical effect is demonstrated in the application as filed (or even “derivable” from its teaching, as suggested in T1989/19), the technical problem may be reformulated to aim at the “improvement” of that effect compared to other known means, even if this was not explicitly planned in the application.
However, when the improved effect relates to a “synergy” not originally described in the application, the position of the Boards of Appeal varies:
1.2. Synergy Not Initially Identified
When an invention involves the use of two compounds, the existence of a synergy between them is often a strong argument in favor of inventive step. This synergy may not have been identified at the time of filing, and it is only later that the applicant demonstrates it and asks for it to be considered in defining the technical problem. In the decisions published since G2/21, Boards 3.3.07, 3.3.02, and 3.3.06 have approached this issue differently.
T873/21
Decision T0873/21, published on 20 June 2023, was the first to apply the two criteria proposed by G2/21. In this case, Board 3.3.07 interpreted criteria 1 and 2 quite broadly: the existence of a synergy between two compounds known to treat the claimed diseases was accepted as the objective technical problem, even though the application merely alleged (without demonstrating) that the combination of the two compounds was “preferred” and would “improve” insulin sensitivity (without mentioning “synergy”). During proceedings, the patentee submitted data demonstrating that the two compounds had a synergistic effect on insulin sensitivity. According to the patentee, this synergy should be considered in defining the technical problem, as it was “plausible” based on the individual effects of each compound. Criterion 1 was deemed satisfied because the synergy demonstrated post-filing was considered to be “encompassed” within the alleged “improvement” effect described in the application as filed, and criterion 2 was satisfied because the claimed combination was explicitly indicated as preferred (cf. points 3.3.2 and 3.3.3). Thus, no experimental proof or technical reasoning was required to support the plausibility of the technical effect (synergistic or otherwise); simple assertions in the description were sufficient in this case to establish that a synergistic effect was “derivable” from the application, allowing the post-filing data to be admitted. Surprisingly, the absence of any experimental evidence in the original application showing the combination’s effect on insulin sensitivity was not addressed.
T116/18
In this decision discussed earlier, the question was whether post-filing data describing a synergy between two compounds of the invention against a new moth species mentioned in the application could be accepted, and whether the technical problem could thus be reformulated as producing a synergistic effect against that species. In the application as filed, synergies had been demonstrated against other insects, and the new moth species was mentioned. Board 3.3.02 ultimately considered that this new synergistic effect was encompassed by the teaching of the application as filed—criterion 1 of G2/21 was thus fulfilled. The opponent could not prove that there was legitimate reason to doubt that this synergistic effect existed for the new moth species, and the application contained many examples showing the effect of different compositions on various insects, so criterion 2 was also fulfilled (point 17). Therefore, the additional data were accepted, and the technical problem was reformulated, even though the application originally contained no data specific to the new moth species.
T681/21
This noteworthy decision, published in October 2023 (after T116/18), involved a case where the new technical effect demonstrated by the patentee was the existence of a synergy between two components (a silicone and a CPP polymer) for fabric softening. The application as filed contained no experimental evidence of any synergistic effect between the two combined components. The CPP was merely described as a “preferred” polymer. The technical effect presented in the application was an “improvement” in fabric softness. Board 3.3.06 found that criterion 1 of G2/21 was not fulfilled because the synergistic effect was not “described” in the application, and the patentee had provided no evidence that the silicone and CPP could interact synergistically (point 1.2.5). A less ambitious technical effect (a simple increase in softening) in auxiliary request no. 4 was, however, considered to meet criteria 1 and 2 of G2/21, as the “softening” effect had been explicitly mentioned in the application (point 5.2.1). In this case, the reformulated technical problem did not lead to recognition of inventive step, as each component of the combination was already known to have softening properties.
While the first two decisions (T873/21 and T116/18) suggested that a synergy demonstrated post-filing could be enough to broaden the technical problem and make up for the lack of data in the original application, the third decision, which was much stricter, suggests that it is important for the synergy to have been mentioned in the application from the outset. T873/21 shows that data demonstrating a synergistic effect can be accepted to support inventive step, even if the original application contains no data proving the effect—provided that the effect was at least initially described.
The consideration of experimental data demonstrating an effect not disclosed in the original application remains more problematic:
1.3. Technical Effect Not Mentioned in the Application
Applicants sometimes submit data at a later stage showing that the invention produces a technical effect that was not mentioned in the original application and request that this new technical effect be taken into account in defining the objective technical problem. The key issue for the Boards of Appeal is then to assess whether this new technical effect is “derivable” from the teaching of the originally filed application, in accordance with criterion 1 of G2/21. If so, the new data are accepted and the technical problem may be reformulated.
We have identified several decisions illustrating how the Boards of Appeal may approach such situations:
1.3.1. Technical Effect Derivable from the Teaching of the Original Application
T1045/21
In this case, the difference between the invention and the closest prior art document was the administration of the claimed drug (edaravone) to a subgroup of patients with amyotrophic lateral sclerosis (ALS) who had a forced vital capacity (FVC) greater than 80%. The application contained phase II and III clinical trial results, which did not show any advantage of administering the drug to this specific patient group, and the invention was thus deemed non-inventive by the opposition division. On appeal, the patentee submitted statistical analyses based on data already disclosed in the application (“post-hoc” analysis), allegedly demonstrating a beneficial effect in patients with FVC above 80%. Board 3.3.07 accepted to consider these analyses, though without explaining why. The fact that the analyzed values were already present in the application likely contributed. Ultimately, however, the Board concluded that this retrospective analysis involved potential biases and did not sufficiently demonstrate that the improved effect was solely due to the claimed FVC threshold. The technical problem could therefore not be reformulated, despite the additional data being considered.
T1989/19
This decision by Board 3.3.02, published on 30 June 2023 (prior to T116/18), concerned a crystalline micronized form of tiotropium bromide with specific particle size characteristics. The patentee submitted post-filing data showing improved stability of the micronized form and requested that the technical problem be defined as improving the stability of the micronized particles. The application did not mention the stability of the invention’s micronized form. However, it did describe the composition as “inhalable,” and both the application and prior art taught that particle size was crucial for inhalation. The Board concluded that the skilled person would have understood, from the application and general knowledge, that the particle size must remain stable over time for effective inhalation. Since the opponent did not raise legitimate doubts, the technical effect of “stability”—though not disclosed—was considered “derivable” from the teaching of the application and “embodied” in the invention, since addressing that problem “did not alter the nature of the claimed invention,” thereby satisfying criteria 1 and 2 of G2/21 (point 3.3.14). The Board also accepted that the problem could be reformulated as “improving stability,” given that any technical effect in an invention is inherently meant to be “improved.” The additional data were thus accepted, and inventive step acknowledged.
T1445/21
The invention concerned a liquid cleansing composition containing fragrance microcapsules and a minimal amount of water (an “anhydrous” composition). The difference with the closest prior art was the low water content, which, according to the patentee, enhanced the olfactory performance of the product. This was supported by an experimental report submitted during opposition. The patentee asked to shift the technical problem, initially related to stability, toward olfactory performance, and to consider the experimental report in the inventive step reasoning. Board 3.3.07 accepted the revised technical problem, reasoning that “it is obvious that the olfactory performance of the composition is directly linked to the stability of the fragrance-containing microcapsules” (page 14), noting that the patent itself monitored composition stability by measuring perfume content and thus scent performance. Therefore, olfactory performance was considered “encompassed” by the patent’s technical teaching, fulfilling criterion 1 of G2/21 (criterion 2 was not discussed). However, the data submitted were not sufficient to show that the improved olfactory performance was due to the low water content, the only distinguishing feature over the prior art. As a result, the technical problem was ultimately reduced to providing an alternative composition, and the claimed solution was considered obvious.
1.3.2. Technical Effect Not Derivable from the Teaching of the Original Application
We also identified several decisions illustrating how the Boards of Appeal reason when the technical effect cannot be deduced or “derived” from the original application:
T887/21
The claims related to a nutritional composition used to prevent secondary bacterial infections following an initial viral infection (e.g., influenza). The application contained no experimental results but theorized how the invention could act more effectively after influenza by modulating cell susceptibility to bacterial infections. Post-filing data submitted by the patentee showed a direct antibacterial effect. Board 3.3.04 held that this new effect (direct antibacterial activity) could not be taken into account, as it was not “encompassed” by the technical teaching of the application (which described a post-infection modulation mechanism). Since the post-filing data demonstrated a mechanism not disclosed in the original application, criterion 1 of G2/21 was not met (point 2.15.4). The Board concluded that the effect demonstrated in post-filing data must not only align with the claimed outcome (here, bacterial infection prevention) but must also adhere to the mechanistic/technical narrative presented in the original application—even if only theoretical or qualified as non-binding (e.g., “without wishing to be bound by theory”). The comparative data were also not compelling, and even if admitted, the conclusion would have remained the same: the problem (providing improved antibacterial effect) was not solved (point 2.15.5).
T0364/20
To support inventive step of a paraffin-based composition, the patentee argued during opposition that the claimed composition produced lower volatile organic compound (VOC) emissions—an effect not mentioned in the application. Board 3.3.02 held that the technical problem could not be based on this effect, as it was not described in the application (point 11.5.3), and did not examine whether the effect might be inferred through general knowledge. The objective technical problem was thus reduced to providing an “alternative composition,” and the invention was deemed non-inventive as a “reasonable alternative” easily derived by the skilled person without prior art suggestion.
T1994/22
In this decision of 15 April 2024, Board 3.3.02 (also behind T116/18 and G2/21) rejected the patentee’s attempt to reframe the technical problem around a specific effect not disclosed in the original application—improved photostability of the claimed crystalline form II. The original application merely stated that all crystalline forms were “high quality” and “easily processable.” The patentee argued that photostability contributed to crystal quality, fulfilling criterion 1 of G2/21, and that there was no reason to doubt that form II was more photostable than other disclosed forms (thus meeting criterion 2, per T116/18). The Board disagreed, stating that even under a broad reading (cf. T116/18), criteria 1 and 2 could not be fulfilled. Specifically, vague phrases like “high quality” or “easy processing” did not include a specific technical effect like “photostability,” and no advantageous effect was attributed to form II in the application. Therefore, the alleged improved photostability could not be considered to be “necessarily embodied” by the claimed invention. This new technical effect was deemed too specific in light of the application’s general assertions, and the additional data were not examined.
We can conclude from these decisions that criterion 1 of G2/21 is the key hurdle for accepting additional data demonstrating a new technical effect not originally disclosed. If the new effect is not “encompassed” by—or at least “derivable” from—the teaching of the original application, the new results will not be examined, and the technical problem will not be reformulated. Moreover, if the Board can reasonably “doubt” whether the new effect is due to the distinguishing feature of the invention over the prior art, inventive step is undermined.
2/ Assessment of Sufficiency of Disclosure
Post-filing data can also be crucial in demonstrating the sufficiency of disclosure of an invention—particularly in confirming that the technical effect alleged in the application (or even claimed, in the case of second medical use claims) is indeed achieved by the invention as claimed. Examiners must first assess the admissibility of such data.
The key issue remains what the skilled person understands from the content of the application as filed, in light of their general knowledge at the filing date. According to the Enlarged Board of Appeal, this point must be assessed even more strictly when the claim concerns a second medical use, as the therapeutic effect is then inherently part of the claim. As a result, if the application does not demonstrate the therapeutic effect and contains no other technical information making that effect sufficiently credible, the application is insufficiently disclosed and no post-filing data can cure that defect (point 71 of G2/21).
This position has been confirmed in all decisions published since March 2023:
T2735/19
In this decision predating T116/18, Board 3.3.07 accepted additional data showing that the invention’s drug (TAS-102) could be used against all types of cancer (including blood cancers), even though the data in the application only concerned patients with breast cancer (point 5.2). These data were admitted because (i) the application provided evidence of the drug’s effectiveness against a specific type of cancer, and (ii) the skilled person was familiar with the mechanism of action of one of the combination’s components on DNA replication (making the anti-cancer effect credible for any type of cancer). This decision confirms that post-filing data can be accepted to demonstrate sufficiency of disclosure in second medical use claims when the skilled person’s general knowledge, combined with the teaching of the application, makes the effect shown in the data “credible.”
T728/21
The application underlying this decision explicitly stated that the claimed combination (ivacaftor and lumacaftor) was an embodiment “according to the invention,” but included no results showing that it achieved the claimed technical effect (in this case, treating cystic fibrosis in specific patients). However, the description provided detailed information about each compound’s therapeutic effects. According to Board 3.3.07, sufficiency of disclosure in a second medical use claim may be based on statements in the application’s description, even in the absence of experimental results, provided those statements are specific and verifiable. Such statements may, for example, rely on information that makes the claimed therapeutic effect credible and was accessible to the skilled person at the filing date—even if not part of the general knowledge (points 3.2 and 3.3). The Board also noted that there was no serious doubt regarding the achievement of the claimed therapeutic effect, contrary to the opponent’s assertions (point 3.4). The invention was therefore considered sufficiently disclosed.
T0025/20
In this decision dated 11 January 2024, Board 3.3.02 (i.e., the same that decided T116/18) found that the examples in the originally filed application, which were intended to demonstrate the effectiveness of cyclobenzaprine in improving sleep in patients with post-traumatic stress disorder (PTSD), were “prophetic” (i.e., they described clinical studies yet to be conducted). The Board nevertheless examined the skilled person’s general knowledge at the filing date and noted that cyclobenzaprine was known to alleviate sleep disorders in general, and that sleep disorders were a major issue in PTSD patients. However, the opponent demonstrated that although certain cyclobenzaprines were known to improve sleep generally, they were not known to do so in PTSD patients at the filing date. In the absence of any other information in the application concerning the mechanism of action of these compounds in this patient group, the Board concluded that there were substantial doubts about the effectiveness of the cyclobenzaprines in treating sleep disorders in PTSD patients, making the claimed effect merely speculative. The additional data submitted by the proprietor to address the deficiencies of the application were therefore not accepted (point 6.4).
T0853/22
This decision was published on 7 February 2024. The patent application at issue claimed the use of glatiramer acetate (GA) to treat patients with multiple sclerosis, with a specific dosage regimen. The original application did not include experimental data but described a clinical trial to be initiated, specifying all intended measurements. Board 3.3.04 found that announcing a clinical trial—even with all the necessary details to reproduce the invention—was not sufficient to suggest that a therapeutic effect exists. The opponents provided numerous documents showing that the desired technical effect (namely, reducing brain atrophy in multiple sclerosis patients) was very difficult to achieve within a reasonable timeframe, along with an expert statement noting that brain atrophy in such patients rarely correlates with other parameters tested in the application. The Board ultimately considered the alleged effect not credible and refused to take the post-filing data into account. It conducted a detailed analysis of the skilled person’s knowledge at the filing date and concluded that there were too many serious doubts for the technical effect to be considered credible at that time (points 18, 23, 24, 28 and 29).
T0209/22
This decision by Board 3.3.04 was published on 21 March 2024. The case concerned a claim for the daily administration of a combination of two molecules to treat asthma or chronic obstructive pulmonary disease (COPD). The two active ingredients (umeclidinium and vilanterol) had already been proposed individually in the prior art for treating these conditions, and their 24-hour efficacy as monotherapies was demonstrated. The data in the original application showed that the combination was well tolerated by healthy volunteers, and a significant difference in bronchodilation was observed in those subjects. The Board concluded that the claimed combination had a credible effect at the claimed dose:
“Based on the information provided in the application as filed, there is thus a strong presumption that dual therapy with umeclidinium/vilanterol would be effective in the treatment of asthma or COPD, and that a dosage regimen of once-daily administration would be feasible. Both aspects would have been regarded as credible at the effective date.” (point 5.4.5).
The additional phase III data, obtained in COPD and asthma patients, were thus accepted, and sufficiency of disclosure was confirmed.
T0197/22
This decision by Board 3.3.07, published on 22 March 2024, confirms that if the application lacks sufficient experimental data, post-filing data submitted to overcome an insufficiency objection will not be accepted if there is doubt about the effectiveness of the claimed invention. The invention concerned the use of liposomes containing mRNA to treat genetic diseases via gene therapy. Although the application contained in vivo data (showing that mRNA reached and was transfected into the target cells in mice), these were not deemed sufficient to demonstrate that transgene expression levels in the target cells were adequate for therapeutic effect. Opponents provided documents showing that high transfection efficiency is required for gene therapy to work, and that many known methods at the filing date did not achieve this (points 2.3.2 and 2.3.3). Due to these technical prejudices, the examples in the application (which only demonstrated proof of concept with luciferase) were not considered sufficient to make the claimed use credible, and the additional data were therefore not accepted (point 2.4).
The analysis of decisions published since March 2023 confirms that post-filing data can be accepted to overcome an objection under Article 83 EPC when the teaching of the application and the general knowledge of the skilled person at the filing date make the alleged effect “credible” (and not just “plausible”). Conversely, such data are often rejected when opponents can highlight technical prejudices or gaps in the patent’s teaching that cast doubt on whether the effect can in fact be achieved by the invention.
We may therefore conclude from the above case law that, when it comes to admitting additional data that go beyond a simple confirmation of what is already in the application, demonstrating the “plausibility” or “credibility” of the claimed effect remains essential. This approach—based on a detailed technical analysis of the application’s content and the skilled person’s knowledge—aligns with criterion 1 of G2/21 and is most often followed by technical boards.
Patent proprietors are therefore advised to ensure that their applications include as much technical information as possible about the effect(s) induced by the invention. For opponents, it is always useful to demonstrate the existence of technical prejudices at the filing date that would lead the skilled person to doubt the credibility of the claimed technical effect—thereby blocking the admissibility of post-filing data.
By Gabrielle FAURE-ANDRE
