T 0524/14 16-06-2016

Main request

1. Admission of appellant 2's objection under Article 123(2) EPC

1.1 During the oral proceedings, appellant 2 raised an objection under Article 123(2) EPC against the feature of claim 1 of the main request that the carbohydrate source comprised maltodextrin. According to appellant 2, the application as filed only disclosed that the carbohydrate source was, rather than comprised, maltodextrin. This objection should be admitted into the proceedings since it had already been raised in the statement of grounds of appeal.

1.2 During the entire written appeal proceedings, appellant 2 did not object to the main request under Article 123(2) EPC. More specifically, when discussing in its statement of grounds of appeal the fifth auxiliary request before the opposition division, which corresponds to the present main request, appellant 2 did not raise any objection under Article 123(2) EPC. After the respondent had promoted this auxiliary request to its main request, appellant 2 still did not object to it under Article 123(2) EPC in its reply. On the contrary, all objections under Article 123(2) EPC related to different requests and different features. Even after receiving the board's preliminary opinion, which made no objection under Article 123(2) EPC against the main request, the appellant kept silent and did not raise any objection.

1.3 Appellant 2 argued during the oral proceedings that the board and the respondent should have been aware that the objection raised in its statement of grounds of appeal against claim 6 of auxiliary request 3 before the opposition division applied by way of analogy to claim 1 of the main request. However, claim 6 of auxiliary request 3 is an independent process claim, whereas claim 1 of the main request is an independent product claim. The assertion in the statement of grounds of appeal that the attacked feature was not disclosed in relation to the claimed process does not necessarily imply that such a disclosure is missing also in relation to the claimed product. In fact, the attacked feature that the carbohydrate source comprises maltodextrin is disclosed in relation to the claimed product in claim 2 of the application as filed.

1.4 From appellant 2's conduct during the written appeal proceedings, the respondent and the board had to assume that it had no objections under Article 123(2) EPC against the main request. Therefore, they were taken completely by surprise when appellant 2 raised this objection during the oral proceedings, and were not prepared to deal with it. The board therefore decided not to admit this objection into the proceedings (Article 13(1) and (3) RPBA).

2. Interpretation of claim 1

2.1 Claim 1 is directed to a balanced powder composition which is defined inter alia as being substantially devoid of flavour. The meaning of this feature was a matter of discussion between the parties. In this respect, when addressing the results of the flavour testing described in D23, the respondent made the following remark:

"To the contrary, the results in D23 are diverging in a form that no predominant flavor could be found concerning the composition that could prevail the finding that the composition is indeed devoid of flavor. As a result, the test of D23 rather underlines that the composition is substantially devoid of flavour as clearly required by claim 1 of the claims [sic] of the main request" (emphasis added by the board).

Hence, the respondent acknowledged that a composition with no predominant flavour is covered by the definition in claim 1 of being substantially devoid of flavour.

In view of this, the feature "substantially devoid of flavour" in claim 1 must be interpreted broadly as meaning that no predominant or strong flavour is present.

3. Inventive step

3.1 The patent is directed to a consumer product which can be added to a meal, has a balanced caloric and protein content, is not expensive and does not alter the taste of the basic food (paragraph [0009]).

3.2 In the same way, D28 deals with the problem of providing a powdered composition neutral in taste that can be added to food such as soups (page 3, lines 102 to 118).

D28 (example 4 with back-reference to example 3) discloses a whole-milk powder prepared by concentrating 10000 litres of skimmed milk to a solids content of about 40 wt%, mixing it with 320 kg of butter oil (page 2, lines 88 to 97), and drum drying the resulting mixture. The resulting whole-milk powder contains

26 wt% milk fat, 71 wt% fat-free milk solids and 3 wt% water. Instead of butter oil, deodorised butter oil can be used, from which the natural fat flavour has been removed (page 3, lines 106 to 116). The resulting powder is neutral in taste (page 3, lines 117 to 118).

In line with appellant 1's argument, this modified example 4 with the deodorised butter oil can be considered to represent the closest prior art.

3.2.1 Since, after deodorisation, the milk powder of modified example 4 is neutral in taste, it meets the requirement of claim 1 of being substantially devoid of flavour.

The respondent argued in this respect during the written proceedings that the composition of claim 1 had to be per se substantially devoid of flavour and, contrary to the modified example 4 of D28, did not have to be deodorised or refined subsequently to achieve this. Furthermore, claim 1 excluded the case where flavour was simply masked by deodorisation. This argument is however not persuasive, since claim 1 is directed to a composition and simply requires that it be substantially devoid of flavour, irrespective of how this is achieved.

3.2.2 Since the composition of modified example 4 is prepared from skimmed milk as starting material and since milk contains an emulsifier, it can be assumed that this emulsifier is also present in the composition of modified example 4.

The composition in modified example 4 can be furthermore regarded as suitable to be added to food, as required by claim 1.

Lastly, the amount of 26 wt% milk fat in modified example 4 translates into an energetic amount of 45%, which is within the claimed range (40 to 50%).

3.2.3 However, modified example 4 does not disclose the energetic amounts of carbohydrate and protein.

Appellant 1 argued in this respect that the energetic amounts of carbohydrate and protein in the skimmed milk (fat-free part) of the whole-milk powder of modified example 4 of D28 were identical to those that could be calculated for the skimmed milk disclosed in D25. More specifically, according to the second table from the bottom of the last page of D25, the weight percentages of carbohydrates and proteins in dried skimmed milk were 52.71 and 33.95 wt%, which meant that of the

71 wt% non-fat solids in the whole-milk powder of D28,

43 wt% were carbohydrates and 28 wt% proteins. These weight percentages corresponded to energetic amounts of 33% carbohydrates and 21% proteins, which was within the ranges defined in claim 1 (25 to 35% carbohydrate and 20 to 30% protein).

This line of argument rests on the assumption that the compositions of the skimmed milks of D25 and D28 are identical. Even though the board acknowledges that the amounts of carbohydrates and proteins in two different skimmed milks will certainly be similar, they cannot be assumed to be identical. More specifically, as set out in D3 (first sentence on page 1), the composition of skimmed milk depends on the animal species and specific animal from which the milk is obtained:

"Natürliche Milch enthält im allgemeinen Wasser, Proteine, Fette, Zucker und Salze. Die relativen Mengen dieser Bestandteile differieren von Tierart zu Tierart und in jeder Tierart von Tier zu Tier."

That the skimmed milks of D25 and D28 are not identical is further confirmed by the fact that the fat content of the skimmed milk in D25 is ten times higher (0.92%; first table in the lower part of the fifth page) than that reported in D28 (0.09%; table 1 on page 2).

Therefore, even though the energetic amounts of carbohydrates and proteins of the skimmed milk of D28 are certainly similar to those derivable for the skimmed milk in D25, they cannot be assumed to be identical. Accordingly, even though they are certainly close to the claimed ranges, appellant 1's assertion that they fall within them is not correct.

3.2.4 Furthermore, the carbohydrate and fat in modified example 4 of D28 do not comprise maltodextrin and canola oil as required by claim 1.

3.2.5 The board therefore agrees with the respondent that the subject-matter of claim 1 differs from the composition of modified example 4 of D28 in terms of the energetic amounts of carbohydrates and proteins and in that the oil source comprises canola oil and the carbohydrate source comprises maltodextrin.

3.3 It needs to be examined what problem is solved by these distinguishing features.

During the oral proceedings, the respondent argued that the step of deodorising or refining the butter oil in D28 was very expensive. Accordingly, the problem to be solved in view of this document was the provision of a powder composition which could be prepared in a cost-effective manner.

However, claim 1 merely requires the fat or oil source to comprise canola oil. This claim thus covers compositions containing 99.9% of expensive deodorised or refined butter oil as the fat source. Therefore, the partial problem of providing compositions prepared in a cost-effective manner is not solved over the entire scope of claim 1.

The respondent furthermore argued that the claimed carbohydrate and protein amounts solved the problem of rendering the composition substantially devoid of flavour. However, this argument is not persuasive either. As set out above, the whole-milk powder of modified example 4 of D28 is substantially devoid of flavour, so that the problem of providing a substantially flavourless composition is already solved.

Lastly, the respondent argued that the claimed carbohydrate and protein amounts ensured that the composition was balanced. However, since the whole-milk powder of modified example 4 of D28 contains carbohydrates, proteins and fat, and since the energetic amounts of these components are similar (carbohydrates and proteins) or identical (fat) to those of the claimed powder, the whole-milk powder of modified example 4 of D28 is likewise balanced. The term "balanced" is to be interpreted broadly, since whether a composition is balanced depends on the type of consumers, which are not defined in claim 1. Therefore, the problem of providing a balanced powder composition is also already solved in D28.

Consequently, in line with appellant 1's argument, the objective technical problem must be formulated less ambitiously as the provision of a further substantially flavourless balanced powder composition.

3.4 It needs to be determined whether in view of this problem the claimed solution, i.e. the selection of the claimed carbohydrate and protein amounts and the addition of maltodextrin and canola oil, is obvious.

3.4.1 As set out above, the carbohydrate and protein amounts in the whole-milk powder of modified example 4 of D28 are close to those required in claim 1. In the absence of any effect demonstrably linked to these amounts, their selection is a minor, arbitrary variation of those of modified example 4 of D28. As argued by appellant 1 during the oral proceedings, it is part of the skilled person's routine abilities to adapt and thus vary the nutrient amounts according to the consumers' needs. This variation therefore does not contribute to inventive step.

3.4.2 In the same way, the choice of maltodextrin and canola oil represents an arbitrary selection out of nutrients commonly used in food before the priority date of the patent. That these two components were common constituents of food is confirmed by (i) D15, which discloses maltodextrin and canola oil as preferred carbohydrate and lipid sources of an enteral nutritional composition (column 3, lines 45 to 53 and column 4, lines 20 to 22), (ii) D16, which discloses maltodextrin and canola oil as examples of carbohydrates and oils present in a nutritional supplement (column 3, line 66 and column 4, lines 45 and 63) and (iii) D18, which discloses the two components as examples of a complete nutritional product.

It is furthermore part of the skilled person's common general knowledge that maltodextrin and canola oil have only a mild flavour and thus are, in the terms of claim 1, substantially devoid of flavour (see point 2 above). In fact, the FDA defines maltodextrin as a non-sweet polysaccharide (page 13, lines 16 to 17 of D18).

Therefore the skilled person looking for a further substantially flavourless balanced powder composition would have added maltodextrin and canola oil to the whole-milk powder of modified example 4 of D28. So the selection of these two components in the claimed composition does not contribute to inventive step either.

Even if, for the sake of argument, the skilled person would not have considered the taste of maltodextrin and canola oil to be mild, this would not have deterred him from adding these two components to the whole-milk powder of modified example 4 of D28. It is part of common general knowledge that even tasty components do not add flavour if added in small amounts. The skilled person looking for a further substantially flavourless balanced powder composition would therefore simply have added the maltodextrin and canola oil in such amounts. Due to its "comprising" language, claim 1 covers the result of such additions, namely compositions with only small amounts of maltodextrin and canola oil. Hence, even assuming that the skilled person (wrongly) did not consider the taste of maltodextrin and canola oil to be mild, their addition to the whole-milk powder of modified example 4 of D28 would still not contribute to inventive step.

3.4.3 The respondent argued that D28 (page 3, lines 99 to 102) taught away from adding canola oil (a vegetable oil) to the whole-milk powder of modified example 4 of D28, since this document stated that it was desirable to use milk-fat powder instead of vegetable fat.

The board does not agree. What D28 discloses is the following:

"Traditionally, certain products use fat powder based on vegetable fat. It would, however, be desirable if milk fat powder could be used also for these products."

Hence, rather than teaching away from using vegetable fat, D28 actually discloses these fats as an alternative to the milk fat suggested in this document. Therefore, D28 rather provides the skilled person trying to find an alternative composition with an incentive to look for vegetable fats.

3.4.4 The respondent furthermore argued that D15, D16 and D18 disclosed maltodextrin and canola oil only in combination with further components that had a strong taste. For instance, maltodextrin was disclosed in D18 only in combination with sucrose, which was very sweet. Therefore, the skilled person taking the teaching of these documents into account would not have added maltodextrin and canola oil alone to the whole-milk powder of modified example 4 of D28, but would have done so together with additional components with a strong taste. He would thus not have arrived at the composition of claim 1, which had to be substantially devoid of flavour.

The board does not find this argument convincing either. D15, D16 and D18 only confirm that maltodextrin and canola oil were commonly used in food before the priority document of the patent. Hence, it is not on the basis of any of D15, D16 or D18 that the skilled person would use maltodextrin and canola oil in the whole-milk powder of modified example 4 of D28, but on the basis of his common general knowledge as confirmed by these three documents.

3.5 Therefore, the subject-matter of claim 1 is not inventive in view of D28 in combination with common general knowledge.

4. In view of the above, the board did not need to decide on the appellants' insufficiency objections and further inventive-step attacks.

First, fourth and ninth auxiliary requests

5. Admissibility

Both appellants requested that the auxiliary requests not be admitted into the proceedings. During the oral proceedings the board did admit them, mainly because it was not convinced by the appellants' assertion that the respondent could have filed them - as they now stand - already during the opposition proceedings. In view of the fact that the auxiliary requests lack inventive step (see point 6 below), there is however no need to give detailed reasons for that decision.

6. Inventive step

6.1 Claim 1 of all three auxiliary requests is identical, and differs from claim 1 of the main request in that the energetic amounts of fat and carbohydrate have been restricted to 44 to 46% and 28 to 32%, respectively.

6.2 As set out above, the amount of fat in the whole-milk powder of modified example 4 of D28 is 45%, which is within the restricted range of 44 to 46%. This restriction thus does not lead to any additional distinguishing feature.

6.3 The carbohydrate amount was already a distinguishing feature in claim 1 of the main request. Hence, this restriction too does not lead to any additional distinguishing feature. Furthermore, the respondent has not provided any argument why, let alone evidence that this restriction gave rise to any unexpected technical effect. Hence, as for the main request, the claimed carbohydrate amount is an arbitrary selection and can be obtained by the skilled person by a routine variation of the carbohydrate amount of the whole-milk powder of modified example 4 of D28.

6.4 Therefore, as with claim 1 of the main request, the subject-matter of claim 1 of all auxiliary requests is not inventive in view of D28 in combination with common general knowledge.