T 0570/18 (Dairy Nutrition with Cereals/GERVAIS DANONE) 21-06-2022

Reasons for the Decision

1. Patent

The patent aims at promoting the consumption of a fermented dairy product contained in a container, and involves mixing the dairy product with cereals. The dairy product's specific viscosity

- prevents the cereals from being immediately and completely covered with the dairy product;

- calls for not much stirring (paragraphs [0012] and [0013]); and

- is stated to be associated with crunchiness of the cereal even after contact with the fermented dairy product (example 6).

2. Main request - inventive step

2.1 In the decision under appeal, the opposition division found that claim 1 involved an inventive step over the prior-art products disclosed in D2.

2.2 The appellant agreed that D2 was the closest prior art. However, it contested the opposition division's finding on the patent's technical effect and the problem solved by the claim.

2.3 D2 and distinguishing feature

2.3.1 D2 is an excerpt from the Mintel database GNPD. It describes three packaged commercial products sold before the patent's priority date. One of the entries in the database is the product with the record ID 1278892 ("Rumblers Oat Crunch"), sold in April 2010. This is the product that is referred to as D2 in the following.

2.3.2 The package of D2 "Rumblers Oat Crunch" comprises a cereal product in a first container and a yogurt (i.e. a fermented dairy product) in a second container. The label instructs the consumer to pour the yogurt onto the cereal product and to stir (figure, page 4).

2.3.3 D2 is a commercial product designed for the same purpose as the product referred to in the patent. It qualifies as the closest prior art.

2.3.4 It was common ground amongst the parties that the claimed use differed from D2 only by the viscosity range in claim 1.

2.4 Technical problem in the decision under appeal

2.4.1 In the decision under appeal, the opposition division held that the technical problem was to provide an improved use involving pouring yogurt over cereal product. It explained that the experimental data summarised in a table on page 11 of the decision demonstrated that the viscosity of claim 1 caused a superior interaction between yogurt and cereal (i.e. improved crunchiness). It added that

"the fact that the crunchiness of a cereal product as a function of the viscosity of the fermented dairy product in contact with said cereal appears to follow a curve which is low at low viscosity, high around the data point of 342.5 mPa.s and then turns to be low again at 1050 mPa.s, especially with respect to the last bit, is in fact surprising." (page 13)

2.4.2 The appellant contested this part of the decision: it argued that the data was irrelevant because it was not representative of the closest prior art.

2.4.3 In view of this, the following has to be done:

- first, the data on file has to be assessed;

- then it has to be established whether the effect of improved crunchiness is demonstrated over the closest prior art;

- only then is it possible to formulate the technical problem.

2.4.4 This is done in the following.

2.5 Experimental data on crunchiness

2.5.1 The table on page 11 of the decision is copy-pasted below:

FORMULA/TABLE/GRAPHIC

2.5.2 The data in the table is based on the patent's example 6 and table II (products 1 to 4) and additional experimental results filed during the examination proceedings (product 5). Products 1, 2 and 5 concern products having a very low viscosity (between 2.5 and 83.5 mPa.s) and product 4 concerns a viscous product (1050 mPa.s). Example 3 is the only product exemplifying the viscosity called for in claim 1.

2.5.3 The number of peaks, measured as described in the patent (paragraphs [0083] and [0084]), indicates the crunchiness of the cereals after contact with the fermented dairy product: the higher the number of peaks, the higher the crunchiness of the cereal. The data in the table shows that the highest value indicating crunchiness is achieved at a viscosity of 342.5 mPa.s. At a higher viscosity (1050 mPa.s), fewer peaks are observed, i.e. a lower crunchiness is achieved (example 4).

2.5.4 In view of this, the respondent argued that increased crunchiness of the cereal did not correlate with increased viscosity. The data measured defined "a curve which reaches a maximum around 342.5 mPa.s" (reply to the statement setting out the grounds of appeal, page 3; also page 6). This was unexpected since it showed that the crunchiness deteriorated both at lower and at higher viscosities.

2.5.5 The opponent contested this. It carried out experiments during the opposition proceedings (letter dated 13 October 2017). In these, fermented dairy products with three different viscosities (1-2 mPa.s; about 300 mPa.S; about 770 mPa.s) were combined with cereals and the number of peaks was established. It arrived at the following conclusion:

"The opponent has attempted to reproduce the result reported in Table II of the patent and has been unable to do so. The opponent has only noted that, as expected, a more viscous product leads to increased crunchiness presumably as a result of the reduced transfer of water."

2.6 Technical effect over the closest prior art

2.6.1 Next, it has to be examined whether the data on file allows the conclusion that the viscosity called for in claim 1 provides an improvement in crunchiness over D2.

2.6.2 As concerns the nature of the compositions used to provide the comparative data, Case Law of the Boards of Appeal of the EPO, 9th edition, 2019, Chapter I.D.10.9 sets out that

"a surprising effect (advantageous effect or feature) demonstrated in a comparative test can be taken as an indication of inventive step. If comparative tests are chosen to demonstrate an inventive step on the basis of an improved effect, 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".

2.6.3 In the patent in suit, the comparative fermented dairy products have a viscosity far from the viscosity range called for in claim 1 and there is only one data point within this range. No data was provided for the commercial product of D2.

2.6.4 A decisive point lies in comparing the claimed subject-matter with the closest prior art D2. There is no disclosure in D2 of the viscosity of the fermented dairy product. During the opposition proceedings, which started in April 2016, it was not possible to establish the viscosity of the fermented dairy product of D2 at the point of time when it was sold.

2.6.5 The respondent argued that the product of D2 had a very low viscosity, similar to that of a drinking yogurt.

In this context it referred to a blogger's report on the product "Rumblers Oat Clusters", published shortly after the patent's filing date (DYC1). The report describes the blogger's tasting experience and assessment of the product and includes three photos (the packaged product, the two separated containers, and the yogurt being poured onto the cereal).

The respondent's reading of DYC1 was that the yogurt was runny and that the cereal became soggy. In view of this, it concluded that the viscosity of the yogurt was very low, similar to that of a drinking yogurt.

However, the individual blogger's perception that the yogurt was "runny" and rendered the cereals "soggy enough" has to be read in the context of the statement that yogurt from "normal pots" is not suitable for mixing with cereals. No conclusion can be drawn from this as regards the viscosity of the yogurt tested in DYC1. In addition, the photo of the yogurt being poured on the cereals in DYC1 shows that the cereals are not immediately and completely covered with the yogurt. Part of the yogurt remains on top of the cereals. This further indicates that the term "runny" must be understood in comparison to a typical yogurt sold in "normal pots".

2.6.6 In 2017, the opponent attempted to obtain further information on the viscosity of the product of D2. In the meantime, products sold under the brand "Rumblers" were no longer available. These products had been rebranded in 2014 as "Nomadic Oat Clusters", as press release D8 shows. For a product belonging to this brand, the opponent established the viscosity of the yogurt at about 300 mPa.s.

2.6.7 The respondent considered that the viscosity measured was irrelevant. In its view, there was no certainty that the product of D2 and those sold under the brand "Nomadic Oat Clusters" were identical. The rebranding could have been associated with a modified composition and in particular a modified viscosity.

2.6.8 However, there is no evidence that the rebranding exercise changed the product of D2 in terms of its textural properties.

2.6.9 Rather, the contrary applies. The press release D8 informs the public that the product formerly known as "Rumblers" was rebranded to "Nomadic Oat Clusters" and this involved "combining the same taste and creamy Irish yogurt, just with a new name and a great new look". It is more likely than not that the rebranding only changed the presentation of the product, while the textural properties or viscosity remained as they were. Moreover, the appellant is correct that it would be improbable and uncommon for changes influencing the consumer's perception of the product not to be mentioned in the press release.

2.6.10 Therefore the board has no reason to believe that the rebranding has led to a change in the viscosity of the yogurt of the "Rumblers" product.

2.6.11 In reaction to the opponent's measurement of the viscosity of the yogurt of the "Nomadic Oat Clusters", the patent proprietor also carried out its own experiments in late 2017. It found that, depending on the flavour, the yogurt's viscosity ranged between 183 mPa.s and 226 mPa.s, which was lower than the value the opponent had measured in a yogurt of undisclosed flavour (about 300 mPa.s).

2.6.12 Considering all this, the board has no evidence supporting the contention that rebranding has led to a (significant) change in viscosity. The measurements carried out by both parties on the product sold under the brand "Nomadic Oat Clusters" lead to the conclusion that the product of D2 has a viscosity which is well above that of drinking yogurt, and in particular that of product 5 (see above, points 2.5.1 and 2.5.2). In short, the viscosity of the product of D2 is closer to the viscosity called for in claim 1 than that of any of the comparative examples on file.

2.6.13 In view of this, the appellant is correct that the data available focuses around two extremes, namely some low-viscosity dairy products and a viscous dairy product. These measurements are not representative of the closest prior art and cannot be taken into consideration to support any specific technical effect.

2.6.14 Contrary to the opposition division's assessment, the subject-matter of claim 1 cannot be regarded as a selection invention in which the viscosity of the fermented dairy product has been deliberately restricted to achieve an improved level of crunchiness.

2.6.15 The respondent referred to the opponent's test results filed by letter dated 13 October 2017 (see above, point 2.5.5). The respondent's argument was that the fermented dairy product having a viscosity of about 770 mPa.s, which was associated with the highest peak value and crunchiness, supported the finding that the viscosity called for in claim 1 provided improved crunchiness.

2.6.16 However, this argument is not convincing.

First, the closest prior art has a viscosity below the lowest viscosity value of the range called for in claim 1. What would be relevant in this context would be data showing what the effects on crunchiness are in the viscosity range between that of the closest prior art and that of the range called for in claim 1. No such data is available.

Second, the viscosity value of 770 mPa.s is above the highest viscosity value of the range called for in claim 1. It is outside the range of claim 1, and not relevant in view of the closest prior art's viscosity.

2.6.17 Finally, it is observed that there is no (experimental) evidence supporting the respondent's contentions that there is a threshold viscosity above which the peak value (i.e. crunchiness) changes considerably or a maximum peak value at a viscosity around 342.5 mPa.s.

2.6.18 Moreover, even if it were conceded that there was a threshold viscosity, there is no evidence that this threshold viscosity is not already achieved in D2.

2.6.19 Therefore there is no effect to be considered in the formulation of the technical problem.

2.7 Consequently, starting from D2, the technical problem is the provision of an alternative use.

2.8 Obviousness

2.8.1 The skilled person would consider modifying the runny property of the fermented dairy product, i.e. its viscosity. The board fails to see anything other than an arbitrary selection in the distinguishing feature.

2.8.2 Furthermore, the appellant correctly argued that to arrive at the viscosity range called for in claim 1 would be a matter of routine. The skilled person would modify the runniness (i.e. viscosity) of the yogurt while ensuring that it adequately coated the cereal and met customers' preferences.

2.8.3 Moreover, fermented dairy products having a viscosity which varies over a wide range from thin to thick are known in the art. Fermented dairy products of the prior art with a viscosity between 17.5 mPa.s and 1050 mPa.s were discussed in these proceedings.

2.8.4 The respondent argued that the skilled person would have had no motivation to modify a commercial product such as the product of D2.

2.8.5 This argument fails to persuade the board.

2.8.6 It may be that the manufacturer of the product of D2 itself has no interest in providing a different product.

2.8.7 However, the exercise of assessing inventive step of claim 1 using the problem and solution approach is a different one. The claim under scrutiny involves a novel (i.e. a different) use, and the question is whether the skilled person starting from the closest prior art would have provided the solution claimed. It is recalled that the notional person skilled in the art referred to in Article 56 EPC is constantly looking at further developing the closest prior art while maintaining its advantageous properties (T 505/88, Reasons 3.3). This applies in the present case too.

2.8.8 To conclude, the subject-matter of claim 1 would have been obvious to the skilled person.

2.9 The ground for opposition under Article 100(a) EPC in conjunction with Article 56 EPC prejudices maintenance of the patent.

3. Auxiliary request

3.1 In the auxiliary request, the viscosity range is more restricted, and spans from 300 mPa.s to 400 mPa.s.

3.2 There is no evidence that the more restricted range is associated with a technical effect and consequently a different technical problem compared with claim 1 of the main request.

3.3 The respondent argued that the appellant had not shown that fermented dairy product with the viscosity called for in claim 1 existed. The argument is understood to be that the viscosity of claim 1 involved a non-obvious alternative.

3.4 However, the skilled person starting from the closest prior art which has a relatively low viscosity would have envisaged working in the viscosity range called for in claim 1, in particular at around 300 mPa.s.

3.5 To conclude, the subject-matter of claim 1 does not involve an inventive step (Article 56 EPC).