T 0034/14 10-03-2016

Reasons for the Decision

Main request

1. Inventive step

1.1 The invention concerns a method of promoting growth in VLBW infants with a birth weight below 1500 g (claim 1). It addresses the problem of ensuring such growth while avoiding strain on the infants' immature metabolism through protein overload (paragraph [0008]).

1.2 Both appellants attacked inventive step starting from D9 as the closest prior art.

1.2.1 D9 is directed to a method of ensuring growth in infants (page 1, lines 3 to 6 and claim 19) while avoiding metabolic overload (page 1, line 30 to page 2, line 9). The infant formula may be used for both preterm and fullterm infants (page 6, lines 3 to 4).

D9 is thus concerned with the same field and problem as the opposed patent. It can therefore be considered to represent the closest prior art.

1.2.2 The respondent contested this and argued that VLBW infants with a birth weight lower than 1500 g had very specific problems. These however were not addressed at all in D9, which only referred to preterm infants in general. Therefore, rather than starting from D9, the skilled person would choose a document such as D5, which was concerned with the specific patient group of VLBW infants with a birth weight lower than 1500 g.

The respondent's assertion that VLBW infants had very specific problems rested on paragraph [0007] of the opposed patent which mentions "particular problems linked to both the infants' low weight and degree of immaturity". However, this paragraph refers explicitly to infants having a birth weight between 500 g to 600 g only. The "particular problems" in question therefore do not necessarily occur in infants with a birth weight just below 1500 g as still covered by claim 1.

Furthermore, this paragraph does not specify the particular type of problems that occur. The only specific problem referred to by the respondent during the oral proceedings was the avoidance of protein overload and its side-effects. However, this is exactly the problem D9 solves (point 1.2.1 above).

It has thus not been shown by the respondent that VLBW infants as defined in claim 1 have any specific problems not addressed in D9. There is therefore no reason why the skilled person should not start from this document as the closest prior art.

1.3 Example 2 of D9 discloses a formula suitable for preterm infants containing 3.5 g hydrolysed whey protein per 100 kcal (see in particular the first table on page 12).

The whey protein in the example of D9 is hydrolysed by a double hydrolysis, applying first "the enzyme Novozyme" for 4 hours at 55°C and then pancreatin for 8 hours at 55°C.

1.3.1 The protein amount of 3.5 g per 100 kcal in the example of D9 is within the range defined in claim 1. The whey protein must be hypoallergenic as required by claim 1, because it is hydrolysed. This has never been disputed by the respondent.

1.3.2 However, D9 does not disclose the degree of hydrolysis of the hydrolysed whey protein used in example 2.

Appellant 2 argued in this respect that the double hydrolysis applied in example 2 of D9 inherently led to the claimed degree of hydrolysis. As shown in D1 (pages 8 and 9), a double hydrolysis as applied in D9 resulted in a degree of hydrolysis of about 17.3, i.e. within the claimed range. This argument is however not convincing. The double hydrolyses in D9 and D1 are entirely different. In example 2 of D9 the second step of the hydrolysis is carried out for 8 hours at 55°C, while in D1 the conditions are different, namely 1 hour at 55°C. Apart from that, example 2 of D9 uses "the enzyme Novozyme" in the first step, while D1 uses Trypsin, which is not necessarily the same. Hence, contrary to appellant 2's assertion, it cannot be concluded that the degree of hydrolysis in example 2 of D9 is the same as that obtained in the double hydrolysis described in D1.

Appellant 2 further argued that a single hydrolysis in D1 led to a degree of hydrolysis of 14 and a triple hydrolysis to a degree of hydrolysis of 35 (pages 8 to 10 of D1). Therefore, the degree of hydrolysis resulting from the double hydrolysis as carried out in D9 had to lie between 14 and 35. There was thus an "overwhelming probability" that the degree of hydrolysis in example 2 of D9 was within the claimed range of 8 to 20. The board does not agree with this argument either. First of all, most of "the range of 14 to 35" - assuming this range actually exists - is outside the claimed range of 8 to 20, so there is in fact no "overwhelming probability" that the degree of hydrolysis in example 2 of D9 is as claimed. Secondly, whether a claimed feature is disclosed in the prior art depends not on likelihood but on whether the prior art disclosure contains the same technical information as the claimed subject-matter (T 231/01, point 5.6).

The degree of hydrolysis is thus a first distinguishing feature.

1.3.3 As set out above, the composition of example 2 of D9 is suitable for preterm infants. Preterm infants include but are not identical to VLBW infants with a birth weight lower than 1500 g. The specific type of infants (patient group) is therefore a second distinguishing feature.

1.4 The patent does not provide evidence of any unexpected technical effect linked to the claimed degree of hydrolysis. In particular, no degree of hydrolysis is given in the patent for the experimental diet in example 2, the only diet for which any effect on growth has been demonstrated in the patent. This rather implies that the degree of hydrolysis is not critical. In fact, this is corroborated by the study in D5 which found that there were no differences in growth rate depending on the degree of hydrolysis (abstract and right-hand column of page 306).

Furthermore, there are no data in the patent or on file showing that the specific patient group, i.e. preterm infants with a birth weight of less than 1500 g, is linked to any unexpected technical effect.

1.5 There is thus no effect due to the distinguishing features which could be used to formulate the objective technical problem. Therefore the board agrees with the appellants that the problem solved in view of D9 is merely the provision of an alternative composition for promoting growth in an alternative patient group.

1.6 The respondent contested this and argued that VLBW infants with a birth weight below 1500 g had on the one hand a need for fast growth, and on the other, as stated in paragraph [0007] of the patent, a very fragile metabolic state which involved particular problems and furthermore could be easily strained by protein overload. Accordingly, the opposed patent solved the problem of ensuring particularly fast growth whilst also addressing the particular problems of these specific VLBW infants without any negative side-effects linked to protein overload. These problems were not addressed in D9, which was not even "enabling" as regards any therapeutic use for VLBW infants with less than 1500 g birth weight.

The board does not find this argument convincing.

There is no support and in particular no evidence in the opposed patent that the growth rate achieved with the composition of the patent is higher than that achieved with D9. In fact, the composition of example 2 of D9 is nearly identical in terms of its amino acid profile to that exemplified in paragraph [0017] of the patent. There is thus no reason to assume that there is any difference in the growth rates achieved in D9 and the patent.

Furthermore, as set out in point 1.2.2 above, the "particular problems" mentioned in paragraph [0007] of the patent do not necessarily occur in infants with a birth weight just below 1500 g and can thus not be considered to be solved over the entire scope of claim 1.

Lastly, growth is achieved in D9 in the same way as in the patent, namely without overloading the infant's metabolism (page 2, lines 4 to 9 of D9).

1.7 Hence, even in the light of the respondent's arguments, the problem solved in view of D9 remains the provision of an alternative composition for promoting growth in an alternative patient group.

1.8 As a solution to this problem, the opposed patent (claim 1) proposes the selection of a specific degree of hydrolysis of 8 to 20 to be administered to VLBW infants with a birth weight of less than 1500 g.

1.9 As set out above, no unexpected technical effect is obtained by selecting the degree of hydrolysis claimed. This is thus an arbitrary selection within the routine abilities of the skilled person. Furthermore, hydrolysed whey protein with the claimed degree of hydrolysis for feeding premature infants is already known from D1 (page 1, lines 6 to 8 in conjunction with hydrolysates 1 and 2 disclosed on page 8, line 19 to page 9, line 16).

As also set out above, the selection of the specific preterm infants with less than 1500 g birth weight is not linked to any unexpected technical effect either. Hence this selection too is arbitrary and within the routine abilities of the skilled person. In fact, as evidenced by D3 and D5 (D3: background and results section of the abstract; D5: title in connection with the birth weights given in table 4), the term "preterm infant" was regularly associated with VLBWs of around or below 1500 g before the priority date of the patent.

The skilled person looking for an alternative composition for promoting growth in an alternative patient group would thus - using his routine abilities or considering D1, D3 and D5 - have arrived at the degree of hydrolysis and the patient group claimed.

Regarding the respondent's argument that there was a technical prejudice in the art against feeding a protein amount as high as that disclosed in D9 to VLBW infants, the board does not agree. Firstly, the assertion is contradicted by the patent itself. Paragraph [0031] of the patent states that the American Academy of Pediatrics recommends a protein intake of up to 3.3 g per 100 kcal for infants with a birth weight of 800 g or more, i.e. a value within the claimed range. Hence, the protein amount defined in claim 1 has even been generally recommended in the US. Secondly D9 already indicates that preterm infants need a higher protein amount than fullterm infants. D9 (page 8, lines 20 to 25) discloses that for fullterm formulae the protein amount is preferably less than 2 g per 100 kcal, while for preterm formulae it can be as high as 4 g per 100 kcal.

1.10 The subject-matter of claim 1 is thus not inventive in view of D9 alone or in combination with D1, D3 and D5. The main request is hence not allowable.

2. The appellants also contested novelty of the subject-matter of claim 1 in view of D1, D2 and D9, and raised objections under Article 100(b) and (c) EPC. These issues were discussed at the oral proceedings and the board came to conclusions in favour of the respondent. Since, however, the main request was not allowable for lack of inventive step, there is no need to elaborate on these issues.

Fourth auxiliary request

3. Inventive step

The claims of the fourth auxiliary request differ from those of the main request in that in claim 1 the range for the protein content has been restricted to 3.4 to 3.7 g/100 kcal.

As set out above, the protein amount in example 2 of D9 is 3.5 g per 100 kcal. This is within the restricted range. This restriction thus does not lead to any further distinction over D9. Therefore, for the same reasons as given above with regard to the main request, the subject-matter of claim 1 of the fourth auxiliary request lacks inventive step. The fourth auxiliary request is thus not allowable.

Fifth auxiliary request

4. Amendments - Article 84 EPC

4.1 The claims of the fifth auxiliary request differ from those of the fourth auxiliary request in that at the end of claim 1 the wording "wherein promoting growth in an infant means assisting the VLBW infant to achieve a rate of growth comparable to that exhibited by a foetus of the same gestation [st]age in utero" has been added.

4.2 The amended wording requires that the infant be assisted to achieve a certain growth rate. The meaning of the term "assisting" is not clear. It is in particular unclear whether what is to be achieved by the use according to claim 1 is still the promotion of growth or rather assistance in the promotion of growth, in the sense of creating the conditions for growth to occur at a later stage.

The respondent argued that it was clear that the claim required the promotion of growth rather than merely the creation of the conditions for it. The board acknowledges that this would indeed be clear if the amended wording in claim 1 read "wherein promoting growth in an infant means achieving a rate of growth comparable to that exhibited by a foetus of the same gestational [st]age in utero". But the amended wording reads "wherein promoting growth in an infant means assisting the VLBW infant to achieve" such growth (emphasis added by the board). Either this wording does indeed mean that growth needs to be promoted - in which case however the expression "assisting the VLBW infant" has no meaning and is thus superfluous, rendering the claim not concise - or the expression "assisting the VLBW infant" somehow narrows down the growth promotion but in a manner which is not clear.

4.3 The amended wording furthermore requires the growth rate to be "comparable" to that of a fetus at the same gestational (st)age in utero, but it is not clear within which quantitative limits the growth rate has to be "comparable" to that in utero.

The respondent argued that the growth rate of a fetus in utero would be considered by the skilled person to be a standard curve with a certain margin - the so-called 90th percentile - around the mid-value. A growth comparable to that in utero thus implied that this growth had to lie within the 90th percentile. However, a margin defined by a 90th percentile is specified neither in claim 1 nor in the patent specification, let alone the term "comparable" being defined such that the growth curve has to be within the 90th percentile. In fact, this term could equally imply that the growth curve has to be reasonably close to the 90th percentile rather than actually within it.

4.4 Therefore, the amendment of claim 1 renders it unclear. Since the amended wording was not present in any of the granted claims, claim 1 can be objected to under Article 84 EPC, and so the fifth auxiliary request is not allowable.

Apportionment of costs

5. The opposition division decided to apportion costs against appellant 2.

5.1 This decision was based on the following sequence of events: in the opposition proceedings, appellant 2 filed a new document D17 two days before the oral proceedings scheduled for 18 April 2013. During these oral proceedings, the respondent requested that the oral proceedings be adjourned in order to have sufficient time to prepare a declaration and file experimental evidence if D17 was admitted into the proceedings. The opposition division decided to admit D17 into the proceedings and consequently adjourned the oral proceedings to 10 October 2013, as requested by the respondent.

5.2 In its decision, the opposition division essentially reasoned as follows: there had been no abuse of procedure by appellant 2, but its late filing of D17 was irresponsible and gave rise to unnecessary costs for the respondent. However, as one of the present inventors had co-authored D17, the respondent had been involved in writing it, and must therefore have known that it existed, and what was in it. In view of this, it was equitable for appellant 2 to bear only the respondent's travel and accommodation costs for attending the second oral proceedings held on 10 October 2013.

5.3 Appellant 2 requested that this decision be set aside, arguing that the respondent had been familiar with D17 well before the first oral proceedings, and its request for adjournment of the first oral proceedings had therefore not been justified.

5.4 The board does not agree. The respondent did not ask for an adjournment of the oral proceedings in order to familiarize itself with D17 but to have sufficient time to prepare a declaration or file experimental evidence in respect to the appellant's arguments relating to it. This was a justified request, since even if the respondent was familiar with D17 it did not need to file any submissions about it until it was brought into the proceedings by appellant 2. It was therefore perfectly reasonable for the opposition division to allow the respondent's request for adjournment, and in consequence to award costs against appellant 2.

Incidentally, an opposition division's decision on apportionment of costs is a discretionary one that should be overruled only if the board comes to the conclusion either that the first instance department exercised its discretion in accordance with the wrong principles or in an unreasonable way, and has thus exceeded the proper limits of its discretion (G 7/93; point 2.6).

5.5 Appellant 2 cited decisions T 330/88 and T 336/86, which however are not relevant to the present case:

5.5.1 In the case underlying T 330/88, a new document had been filed by the opponent during oral proceedings before the opposition division. After the oral proceedings, the proceedings were continued in writing. In subsequent appeal proceedings, the board held that the appellant (proprietor), who had requested an apportionment of costs, had had sufficient time to consider the late-filed document during the oral proceedings before the opposition division, which had lasted for two consecutive days. The board therefore rejected the appellant's (proprietor's) request for apportionment of costs. This is different from the present case, where there was not enough time during the oral proceedings to react properly to late-filed document D17.

5.5.2 In T 336/86, after the opponent had filed a new patent publication during the oral proceedings before the opposition division the proceedings were continued in writing. In the subsequent written proceedings, the proprietor requested the revocation of its patent and an apportionment of costs, since the oral proceedings could have been avoided if the opponent had submitted the new patent publication earlier. The board rejected the request for apportionment of costs basically on the grounds that the proprietor was also the holder of the late-filed patent. The board held that the failure of both parties to exercise all due care over this document had resulted in oral proceedings which proved to be superfluous and that in these circumstances it was not justified to award costs to either of the parties.

That case too is different from the present one because the new document was so highly relevant that the patent could no longer be defended, and the respondent (proprietor) actually requested its revocation. Hence, while it may be true that if the proprietor had exercised due care, it would have taken this document into account even before it was filed by the opponent, the same is not necessarily true in the present case.

5.6 The board therefore in the present case found that the opposition division's decision on costs was justified.