T 1738/17 26-10-2021

1. The subject-matter of the contested patent

1.1 The patent relates to a barbed suture for closing or binding together wounds in human or animal tissue, an example of which is shown in Figures 3A-3B, reproduced below. Compared to a conventional suture, a barbed suture (30) includes one or more spaced barbs (35, 37, 39) which project from the surface of the elongated body (32) of the suture. The barbs are arranged to allow passage of the barbed suture through tissue in one direction but to resist movement of the barbed suture in the opposite direction. Hence, barbed sutures do not have to be knotted, like conventional sutures (paragraphs [0002], [0004]).

FORMULA/TABLE/GRAPHIC

The barbs can conveniently be made by cutting, for example by means of a cutting machine including a plurality of blades configured for escarpment of barbs onto a suture filament, usually one set of axially spaced barbs at a time (paragraphs [0030]-[0032]).

1.2 More particularly, the patent aims to provide an improved barbed suture which has not only an increased tensile strength, but also an enhanced ability in holding and maintaining wound edges together (paragraph [0011]).

In accordance with the patent, these improved properties result from the combination of a specific spatial arrangement of the barbs on the suture body along with an optimised geometry of the individual barbs:

(a) as shown in Figures 3A-3B above, the barbs are disposed on the suture body in a 120-degree staggered disposition including three sets of barbs (35, 37, 39), each comprising a plurality of aligned barbs axially spaced along the longitudinal axis of the suture. The sets are staggered with respect to each other (paragraph [0021], Figure 3A) as well as radially spaced about 120 degrees from each other (paragraph [0049]; Figure 3B). Such a disposition maximises the retention force of the suture (paragraph [0069]);

(b) the barbs are cut into the filament so as to maximise the barb density along the suture filament, which increases the ability of the barbed suture to anchor tissues, without compromising the filament integrity (paragraph [0124]). The optimal barb density is defined by the fact that a first parameter, barb cut distance P, is in a particular range (claim 1). P is illustrated in Figure 7B, albeit for a different type of suture (namely twist cut sutures; see point 5.1 below):FORMULA/TABLE/GRAPHIC

(c) the barbs have an individual geometry defined by three further geometric parameters: barb cut depth D, barb cut length L and barb cut angle theta, as shown in Figure 7A:

FORMULA/TABLE/GRAPHIC

D, L and theta are not independent but linked by the following equation (paragraph [0111]):

sin(180°-theta)=sin(theta)=D/L

The patent discloses optimal ranges for the parameters D, L and theta (granted claims 1, 2; paragraphs [0120]-[0122]).

2. Admittance of the appellant's submissions filed on 25 October 2021

2.1 In its statement of grounds of appeal, the appellant did not raise any objection or reasoning relying on the diameter or the shape of the cross-section of the sutures, or on Figure 3 of D1. It was only on 25 October 2021 (the day before the oral proceedings), i.e. after notification of the summons to oral proceedings, that the appellant made such submissions. These submissions, and a fortiori the new, annotated, higher-quality Figure 3 accompanying them, therefore constitute amendments to the appellant's case.

2.2 According to Article 13(2) RPBA 2020, any amendment to a party's appeal case made after notification of a summons to oral proceedings shall, in principle, not be taken into account unless there are exceptional circumstances, which have been justified with cogent reasons by the party concerned.

2.3 Contrary to the appellant's view, there are no such exceptional circumstances in the present case that would justify the filing of these submissions only at such a late stage in the proceedings.

2.3.1 Decision T 1064/15, on which the appellant based its reasoning on insufficiency of disclosure, was published in 2018, i.e. it was available to the appellant more than two years before the summons in the present case was issued in February 2021. It is immaterial whether the appellant only came across this decision late, because such circumstances cannot outweigh the need for procedural economy and the principle of fairness to the other party.

2.3.2 The appellant also explained that the higher-quality Figure 3 was available for downloading together with D1. This figure could thus already have been filed with the appellant's statement of grounds of appeal to support the inventive step objection raised therein in view of D1.

In this respect, the board is not convinced by the appellant's argument that filing the new figure had been triggered by new matter raised in the board's preliminary opinion. Indeed, the board's statement in point 5.3 that "D1 itself does not contain any pointer towards any numerical ranges for the barb geometry parameters" did not introduce any new matter into the proceedings but merely concurred with the respondent's view, presented in the latter's reply to the statement of grounds of appeal (point 5.5.1: "D1 is completely silent with regard to specific barb geometries (...) D1 provides no guidance whatsoever to lead him in that direction"). Moreover, deriving barb parameters from the figures of D1 was already an argument presented by the appellant in the first-instance proceedings and addressed in the appealed decision (point 4.6.1). Thus, the new Figure 3 and the reasoning based on it could and should have been filed earlier.

2.3.3 A change of representative by the appellant, even shortly before the oral proceedings, cannot justify such a late filing either.

2.3.4 Furthermore, contrary to the appellant's argument, it is irrelevant whether the respondent was already aware of decision T 1064/15 or of the suture depicted in Figure 3, or even that Figure 3 appeared on the same page as other figures of D1 already under discussion. The respondent cannot be expected to speculate on potential evidence or lines of argument that the appellant has not presented.

2.4 For these reasons, pursuant to Article 13(2) RPBA 2020 the board decided not to take the appellant's submissions filed on 25 October 2021 into account.

3. Main request - added subject-matter

3.1 The parties dispute whether the particular combination of numerical ranges for barb cut depth D, barb cut length L and barb cut distance P defined in claim 1 as granted can be directly and unambiguously derived from the root application as originally filed.

3.2 The claimed ranges for barb cut depth D, barb cut length L and barb cut distance P are disclosed in the root application together with a number of other geometric features of the barbs, also presented as advantageous, such as a barb cut angle theta in a specific range, a corrugated underside, an arcuate base, and varying barb sizes (page 3, line 30 - page 4, line 3; page 34, lines 2-4, and the discussion of parameters L, D, theta and P on pages 34-35).

While the description of the root application does indeed mention that these features may be present in any combination (page 4, line 3: "or combinations thereof"; page 34, line 3: "and/or"), there is no suggestion in the root application of the particular selection defined in claim 1 as granted. Although conceptually encompassed by the original disclosure of the root application, this selection does not emerge clearly and unambiguously from the original disclosure as a whole and therefore constitutes new technical information extending beyond the content of the root application as filed, contrary to the respondent's view.

3.3 The respondent has also referred to original claim 21, which explicitly discloses the combination of the parameter ranges for D, L and P, but with an additional parameter range for barb cut angle theta.

The argument that removal of the latter (thereby leading to the combination defined in claim 1 as granted) does not introduce added subject-matter, in particular as the "three-point test" would show, does not convince the board.

Indeed, by defining particular ranges for D, L, P and theta, original claim 21 defined a particular set of barb geometries, with barbs of specific thickness, length and cut orientation that were disposed along the suture filament in a particular density. The parameters D, L and theta cannot be varied independently of each other, but are geometrically linked by the formula indicated in point 1.2(c) above. Hence, by removing the limitation on theta, a different, broader set of barb geometries is defined in claim 1 as granted: however, this set does not correspond to the set defined by original claim 21, nor is it derivable from the description, as discussed in point 3.2 above. This results in presenting the person skilled in the art with new technical information. There is therefore no basis for omitting the angle theta from the combination originally disclosed in claim 21.

In this respect, it is emphasised that the "three-point test" is only meant to provide an aid to determining the allowability of an amendment, and cannot take the place of the assessment made above (see for instance T 1852/13).

3.4 It follows that, contrary to the respondent's argument and to the opposition division's finding (point 2 of the appealed decision), the subject-matter of claim 1 of the main request does not meet the requirements of Article 76(1) EPC.

4. Auxiliary request 1 - added subject-matter

4.1 Compared to claim 1 as granted, claim 1 of auxiliary request 1 further includes the limitation that the barb cut angle theta ranges from about 140° to about 175°, i.e. it contains the same limitation on theta as originally disclosed in claim 21 of the root application.

It follows that claim 1 of auxiliary request 1 includes the same four parameter ranges for D, L, P and theta as defined in original claim 21 (see point 3.3 above).

The board is therefore satisfied that the subject-matter of claim 1 of auxiliary request 1 meets the requirements of Article 76(1) EPC. The appellant had no objection in this respect.

4.2 The description of the application on the basis of which the contested patent was granted contains the description of the root application verbatim, plus the claims of the latter redrafted as aspects of the invention. It follows that the requirements of Article 123(2) EPC are also met. The appellant did not contest this point either.

5. Auxiliary request 1 - sufficiency of disclosure

5.1 It is true, as put forward by the appellant, that the patent specification only includes an explicit definition of the barb cut distance P for twist cut sutures (paragraph [0114]), which are different from the staggered sutures claimed.

However, the person skilled in the art would understand from paragraph [0114] that, for a twist cut suture, the barb cut distance P is measured as the longitudinal distance between two adjacent barbs after the barbs have been aligned longitudinally by twisting the suture back into the state in which it was cut: in other words, the barb cut distance P corresponds to the longitudinal distance between the respective terminus of the cuts of the two adjacent barbs when the aligned set of barbs was cut (see Figure 7B).

Since the staggered suture is cut in an untwisted state as described in paragraphs [0064]-[0066], no twisting is necessary to return the suture to the state in which it was cut. Therefore, contrary to the appellant's view, the person skilled in the art would infer by analogy that the barb cut distance P for a staggered suture is the longitudinal distance between two adjacent barbs of the same set of aligned barbs, i.e. the distance called P1 by the appellant.

The appellant submitted that a person skilled in the art applying the measurement methodology described in paragraph [0114] to staggered sutures would instead apply a twist to align the two nearest barbs 35, 37 of different sets, and would thus measure the distance P2 instead of P1. The board disagrees. Due to the staggered disposition, applying such a twist would in fact align only the two nearest barbs under investigation with the longitudinal axis of the suture, and not the remaining barbs. Hence, in contrast to the situation described in paragraph [0114] for twist cut sutures, where the whole set of barbs is aligned longitudinally, no "longitudinal cut distance P between two adjacent barbs" could be measured unambiguously; in particular, it would not be possible to calculate the average for several pairs of adjacent barbs, as described in paragraph [0114]. The person skilled in the art would therefore recognise that applying a twist to staggered sutures to measure barb cut distance P does not make any technical sense.

The interpretation of P as P1 that has been adopted above is also consistent with the fact that, as the appellant itself noted, the only distance mentioned with respect to staggered sutures is "the longitudinal distance between two of barbs 35" referred to in paragraphs [0054]-[0055]. The fact that these paragraphs obviously contain an error is irrelevant, since this error merely concerns the offset between the sets of barbs, i.e. the distance P2 and not P1.

Hence, contrary to the appellant's view, the person skilled in the art would derive from the patent specification a clear and unambiguous definition of the barb cut distance P for the staggered sutures as well.

5.2 The appellant further argued that the invention as claimed could not be carried out for some combinations of the parameters D, L, P and theta that did not work, although they were within the ranges defined in claim 1.

In particular, the appellant referred to the requirement that the barbs of the staggered suture should not overlap, which resulted in the criterion P>L (paragraph [0142]). Given the range claimed for L, according to the appellant this prevented the person skilled in the art from producing the staggered sutures claimed for all values of P between about 0.1 and about 0.2 (in units of the elongated body diameter).

Irrespective of whether the appellant's assertion that the barbs of a staggered suture as claimed should not overlap is correct, the board notes that the allegedly problematic values of P remain close to the lower limit claimed and that the extent of the corresponding range (about 0.1 to about 0.2) remains very small compared to the overall extent of the range claimed (about 0.1 to about 6). In fact, when taking the inclination of the barbs into account, it can be inferred that the lowest permitted value Pmin of P at which the barbs are still prevented from overlapping is obtained when L=0.2 and theta=140° (see point 1.2(c) above):

Pmin = L|cos theta| = 0.2x|cos 140°| APPROX 0.15

which is even closer to the lower limit of the claimed range for P of about 0.1, if not even identified with it in practice due to the uncertainty arising from the term "about".

Therefore, if the range claimed for P encompasses values incompatible with staggered sutures, as alleged by the opponent, these incompatibilities are in any case restricted to only the very limits of the ranges defined in claim 1, hence to the edges of the claim. Specific combinations of the geometric parameters that do not work would be immediately ruled out by the person skilled in the art as being outside the scope of practical application of the claimed subject-matter, so their existence does not justify an objection of insufficiency of disclosure.

5.3 For these reasons, the board concludes that, contrary to the appellant's argument, the invention as defined in auxiliary request 1 is disclosed in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.

6. Auxiliary request 1 - inventive step

6.1 The appellant raised an inventive step objection in relation to claim 1 of auxiliary request 1, taking the barbed suture disclosed in Figure 4 of D1, reproduced below, as the starting point. It is common ground that D1 discloses the same geometric parameters D, L, P and theta to characterise the barb configuration of the suture as those in the contested patent (see Figure 5 reproduced below, and the paragraph above the figure in D1). However, D1 does not disclose any particular numerical ranges for these parameters.

FORMULA/TABLE/GRAPHIC

The point disputed by the parties is whether the particular combination of ranges for the parameters D, L, P and theta defined in claim 1 of auxiliary request 1 is inventive over D1.

6.2 The appellant first argued that there was no convincing evidence demonstrating that these particular ranges actually led to an improvement in staggered sutures and thus solved the technical problem mentioned in paragraph [0011] of the patent, namely optimising the performance of a barbed suture.

It is true that the experimental data presented in the description, especially in the various tables, concerns twist cut sutures and not the claimed staggered sutures. However, the description discusses in general terms in paragraphs [0120]-[0125] how the various geometric parameters involved are functionally interrelated and directly or indirectly influence both the tensile strength and the retention force of the barbed suture. In view of these explanations, and despite the absence of comparative data, the board sees no reason to question the explicit statement in paragraph [0119] that similar considerations also apply to staggered sutures. The board considers it credible that the particular ranges defined in claim 1 have a technical effect and solve the technical problem stated in the patent.

Contrary to the appellant's view, under these circumstances the burden is on the opponent to prove the opposite. In the present case, the appellant (as the opponent) has not submitted any proof or evidence convincingly casting doubt on the assertion that the problem has been solved by the patent. In particular, the fact that the same considerations may additionally apply to further types of barbed suture (as mentioned in paragraph [0119]), or the fact that twist cut sutures might in fact have better wound holding capabilities than staggered sutures (as disclosed in paragraph [0105]), does not mean that the determined ranges do not lead to improved performance in staggered sutures.

The fact that some combinations of the parameters claimed may be incompatible (point 5.2 above) does not contradict the conclusion above, either, since the person skilled in the art would regard these incompatible combinations as not covered by the claim.

6.3 The barbed suture of claim 1 is distinguished from that shown in Figure 4 of D1 by no fewer than four interrelated parameters, each in a specific range. Contrary to the appellant's argument, a mere routine optimisation starting from D1 would not lead the person skilled in the art to the set of barb configurations claimed.

6.3.1 While D1 does indeed suggest optimising the barb geometry in order to improve the performance of the suture (page 5, section "Future Trends", lines 12-16), D1 itself does not contain any pointer towards numerical ranges for the barb geometry parameters, and does not provide any guidance or direction which would suggest the barb configuration specified in claim 1 to the person skilled in the art.

In this respect, Figure 5 of D1 is a schematic drawing merely illustrating relevant parameters to characterise the barb configuration. Contrary to the appellant's argument, the person skilled in the art would not be able to infer any angle or dimensional ratio from this purely schematic drawing.

6.3.2 The appellant further argued that the person skilled in the art would find suggestions of the combination of ranges claimed in D2 or D3, which both disclose barbed sutures. The board disagrees on this point as well.

Indeed, while a barb cut depth D falling within the claimed range is disclosed in D2 (paragraph [0017]) and D3 (page 14, lines 28-35), neither of these documents directly and unambiguously anticipates the ranges for L, P and theta that are also claimed, let alone a combination of these ranges. A barb cut angle theta is not even mentioned in either of the documents. The "interval" 7 referred to in D2 (paragraph [0017]) does not correspond to the barb cut distance P used in the contested patent (see point 5.1 above), but rather to the length of barbless filament between two adjacent barbs, as illustrated in Figure 1. However, this is not directly comparable to the barb cut distance P. Similarly, the definition of the "barb spacings" mentioned in D3 (page 14, lines 31-32) does not unambiguously correspond to the barb cut distance P. Furthermore, the figures of D2 and D3 are also merely schematic drawings (and moreover, they are of low quality or even hand sketched) from which the person skilled in the art would not be able to infer an angle or dimensional ratio.

6.3.3 The appellant's further argument, that the ranges specified in claim 1 are so broad so that they prima facie encompass substantially all the barb configurations that would be seriously contemplated by the person skilled in the art, does not convince the board either. The appellant itself submitted, in the written procedure, that D3 (page 14, lines 23-35) disclosed a barb cut spacing P of 10, i.e. one much larger than the claimed maximum of 6.

6.3.4 The board thus concludes that, contrary to the appellant's argument, the person skilled in the art starting from D1 would not have arrived at the claimed invention in an obvious manner. The subject-matter of claim 1 of auxiliary request 1 therefore involves an inventive step (Article 56 EPC).

7. Auxiliary request 1 - further objections

The appellant had no further objections to the claims of auxiliary request 1, or to the description being adapted by incorporating the amendments to paragraph [0122] that were filed by the respondent during the oral proceedings before the board on 26 October 2021. The board had no objections either.