Note that Hepp Wenger Ryffel is involved in this case on behalf of the defendant.
This case is about an alleged infringement of Gilead’s SPC C00915894; the basic patent is EP 0 915 894 B1 (see EPO Register and Swissreg). Gilead’s product is Truvada®, a pharmaceutical for the treatment of HIV-1 infection. The present main proceedings are following-up on summary proceedings S2017_006 when the FPC granted interim injunctive relief; see this Blog here.
Validity of the SPC is not an issue anymore, it has already been affirmed by the Supreme Court; see this Blog here.
The dispute is all about whether or not the scope of an SPC that specifically refers to a unique salt form does extend to other salts. The SPC concerns tenofovir disoproxil fumarat + emtricitabin (emphasis added) .
The parties agreed to disagree on this issue in the hearing. Gilead insisted that the ‘typical product definition’ has to be applied, in a sense that the fumarate is only to be understood as a ‘reference’ for the assessment of whether or not yet another salt form has the same pharmacological effect. If it had, then it should be considered ‘the same product’.
On the other hand, Mepha argued that the term ‘product’ is defined in Art. 140a(2) PatA, and it is to be construed within the context of the PatA; the Therapeutic Products Act (TPA) was not relevant in this respect. Thus, infringement has to be assessed as it is routinely done, i.e. deciding on whether there is literal infringment or infringement under the DoE. Mepha argued that the SPC is neither literally infringed, nor under the DoE.
The expert opinion of the judge-rapporteur held that the SPC was infringed, apparently essentially along the line of Gilead’s arguments. Still, in case the court followed the expert opinion, Mepha requested the court to provide a full assessment of infringement in the judgment, also including an assessment of literal infringement and infringement under the DoE. The Supreme Court would then be in a position to review the whole matter at once and in a timely manner, without the need to remit the case back to the FPC to carry out yet another assessment.
As mentioned above, Truvada® is a pharmaceutical for the treatment of HIV-1 infection. But Mepha recently obtained the market authorization of its Emtricitabin-Tenofovir-Mepha® in Switzerland also for PrEP – which is short for Pre-Exposure Prophylaxis; see Swissmedic Journal 02/19. More information on PrEP can be found on Gilead’s website here. However, Gilead currrently has no MA for PrEP in Switzerland; Mepha’s MA is the only one in Switzerland for this indication.
PrEP likely is the reason why this case attracted quite some public attention. There has even been an organised Facebook event in St. Gallen; and flyer have been distributed, see here and here.
On a sidenote, when a member of the audience wore a shirt with one of these slogans on it, the presiding judge held that this was an attempt to influence the judges, and thus ordered him to either hide the slogan / change clothes; or to leave the court room.
The inventions are all about the capacitive testing of yarns or fabric, wherein the dielectric property of a capacitor arrangement is determined.
It became clear from the pleadings that the defendant’s main request in both cases was that the patent be maintained in amended form only. In my understanding, the expert-opinion of the judge-rapporteur had held that the main request in O2016_010 / EP’250 was allowable, but not in O2016_011 / EP’018.
Novelty has not been an issue anymore, in both cases. Only obviousness, undue extension of subject-matter (interim generalisation) and clarity (of the additional feature) have been discussed. D2 (US 4,843,879; referred to in EP’250, ¶) was apparently agreed to be the closest prior art, and was combined in the assessment of obviousness with either D4 or D11 — both of which have unfortunately not been identified.
For me as a chemist, the highlight of the day was the plaintiff’s emotional release:
That’s not chemistry here; it’s physics!
The formulation of the ‘objective technical problem’ was quite a big deal in the hearing. This is what EP’250 says in ¶ (emphasis added):
It is therefore the aim of the present invention to provide a method for the symmetry adjustment of a device for the capacitive examination of a moving elongated textile test sample which does not have the above disadvantages. The device should be simple, fast, cost-effective and, in particular, automatically adjustable.
Now, can automation be considered as the ‘objective technical problem’? In the plaintiff’s view, the ‘in particular’ is an emphasis of the aim of automation. The defendant disagreed: Automation was an add-on, and the other advantages could well be realised without automation. On the funny side, the following example has been used:
She likes flowers, in particular roses.
Now, imagine that the rose-loving woman visits a florist. Which flowers will she recognise first, amidst all the other flowers? The roses, of course!
Compensation of the ‘necessary expenses’ for the assisting patent attorney pursuant to Art. 3 lit. a and Art. 9(2) CostR-PatC was yet another point of discussion. The plaintiff requested reimbursement of about kCHF 45 for each case and argued, that the defendant triggered the additional costs because the amended claims had only been filed with the rejoinder which then had effectively restarted the whole discussion. I did not get the value in dispute, but the presiding judge indicated that kCHF 45 is about twice as much as the amount that would be awarded for legal respresentation, i.e. ‘according to the tariff’ that is based on the value in dispute.
Now, why is that important?
Since O2012_043, the FPC’s practice is that, as a rule, the ‘necessary expenses’ for the patent attorney may well be within the same order of magnitude of the costs for legal representation. But what about nullity cases? A patent attorney could do that on his/her own, without any legal representative; Art. 29(1) PatCA. Compensation would then be awarded according to the tariff for legal representatives; Art. 9(2) CostR-PatC.
The defendant questioned whether costs incurred for the assisting patent attorney are ‘necessary expenses’ at all, on top of the costs for legal representation according to the tariff, if he/she could have done the whole case on his own.
Use of the technical infrastructure in the court room
It is quite hard for the public to follow the arguments when the representatives just argue with acronyms and references. It’s sometimes frustrating, but I got used to it. But still, one of the basic ideas of a public hearing is to give the public the opportunity to follow the proceedings, isn’t it? But how would you even get a glimpse of what is really at stake when the parties only refer to ‘EP’111’ or ‘D1’ or ‘feature 1da’. That’s not information — it’s just noise.
I was very happy that the recent hearing in case S2018_006 was different (see this Blog here): Whenever someone referred to a specific document, the President put the same on the screen for it to be easily recognisable by the whole audience and the judges; projector and screen are fixedly installed in the ceiling anyway (see yellow arrow in the image below). Much appreciated!
But that’s not only great for the audience; I strongly feel that it is also beneficial for representatives when arguing their case. When referring to a certain paragraph in a specific document in your pleadings, or a technical drawing, you always want to make sure that all the judges are on the same page, right? But how would you know that, with the judges sitting on a podium that even has an elevated front side that makes sure that nobody reveals a hand up there, behind the monitors. Yes, there are fixedly installed monitors (green arrows in the above image).
Now, did they all pick up the document that you are upon to discuss? You never know. So, why not actually use the monitors that each judge has in front of him/her, and the projector/screen for the audience, to make sure that everybody sees what you want to be seen? I could not think of any more powerful way to focus the discussion and thoughts on a specific argument.
Unfortunately, it didn’t work out that way in this hearing even though the defendant suggested it. I hope this was for technical reasons only, and does not indicate a general change of practice again.
Both parties had lodged an appeal — but the Supreme Court dismissed both appeals. While the FPC’s judgement was quite a booklet of 50+ pages, it took the Supreme Court only 7 pages to deal with both appeals. Most of the judgment indeed is pretty straight forward. Still, the judgment has quite some interesting aspects, in my perception:
Assignment of a patent does not imply assignment of the universal right to the invention per se
Guenat had requested that Swiss Finest‘s Swiss patent be annulled based on Art. 26(1) lit. d PatA, i.e. for lack of entitlement. However, such request can only be brought by the entitled person; Art. 28 PatA. Guenat alleged that Frédéric Garinaud transferred ‘full and complete ownership’ of the European patent application to Guenat. But that did not help. Frédéric Garinaud specifically assigned the European patent application to Guenat, but not the universal rights in the ‘invention’ itself. Guenat’s request for declaration of nullity based on Art. 26(1) lit. d PatA was thus held inadmissible for lack of standing.
On appeal, Guenat criticises the FPC’s ‘artificial dichotomy’ between the invention and the related patent application, and claims complete control over the invention in dispute. The Supreme Court didn’t buy into that. The decision holds in ¶3.2 that Guenat’s argument is based on an assumption about the parties’ intention in respect of the transfer agreement of April 2013. This is not a question of law, but of fact. But the FPC had not found that the real and common intention the parties had been to assign all intellectual property rights to the invention and the right to be granted all related patents. On the contrary, according to the FPC, the agreement in question ‘clearly refers only to the European patent application and the resulting European patent, and not to the invention as such’ — which finding is binding on the Supreme Court since it is not ‘arbitrary.’
Thus, the Supreme Court apparently had no doubt that the assignment of a specific patent does not necessarily imply the assignment of the universal rights in the invention itself.
Assessment of inventive step is a question of law, but …
Guenat argued on appeal that the FPC had violated the principle of party disposition because it took facts into account that had not been alleged by the defendant.
Now, what had happened?
The FPC had held that a specific feature (C3-1) was not disclosed in the closest prior art document. Guenat had a different understanding of the specific feature, and had argued that based on a ‘correct’ understanding of the feature it had well been disclosed in the closest prior art.
On appeal, Guenat argued that Swiss Finest had not even alleged that feature C3-1 was not disclosed in that document. Unfortunately, it is not readily clear from the decision to which version of feature C3-1 that allegation pertained: The feature as construed by the Guenat, or the feature as construed by the FPC?
The Supreme Court held that it had not been disputed that all the factual elements necessary for the analysis of obviousness had been brought on file by the parties. Further, the Supreme Court held that Guenat had not asserted that the FPC had to supplement the facts on its own, and that such conduct was also not apparent from the file. In the Supreme Court’s view, the FPC was thus in a position to assess obviousness, which is a question of law. It is irrelevant in this respect that the presence or absence of specific features in the prior art and/or the patent have been alleged by the defendant or the plaintiff. On the basis of the facts gathered by the parties in order to consider a question of law, the court did not violate the principle of party disposition.
Without any in-depth knowledge of what had or had not been argued by the parties, I just cannot say whether I would agree with the assessment on the merits of the case. But the Supreme Court’s general considerations on questions of law vs. questions of fact trigger some further thoughts.
Clearly, assessment of obviousness is a question of law — but the underlying factual elements still need to be asserted by the parties. But what are the ‘factual elements’? They must be something more specific than just ‘the patent in suit’, ‘D1’ or the like. Else, a court could just combine whatever is on file in its assessment of obviousness, fully detached from the pleadings. This cannot be it. Now, let’s assume that both parties come up with a somewhat strange claim construction, and base all their obviousness attacks / defenses on a wrongly construed claim. In my understanding, claim construction clearly is a question of law, too (see e.g. X ZR 255/01 – ‘Bodenseitige Vereinzelungseinrichtung’ of the German Federal Supreme Court in ¶5, with further reference). But what if no party ever pointed to the relevant underlying ‘factual elements’ (i.e. specific sections of the patent in suit) that support the ‘correct’ claim construction? May the court then even come up with a different / the ‘correct’ claim construction at all?
UPDATE 21 March 2019:
Note that the FPC had held in the recent decision S2018_006 that claim construction is a question of law. In that case, both parties had not construed the feature in question at all, and the FPC came up with a construction on its own:
Wie der Begriff korrekt auszulegen ist, tragen die Beklagten hingegen ebenso wenig wie die Klägerin vor. Da die Auslegung des Patentanspruchs eine Rechtsfrage ist,15 geht die Annahme der Beklagten fehl, das Gericht dürfe den Anspruch mangels entsprechenden Vortrags der Klägerin nicht so auslegen, dass er auch etwas anderes als einen Schneidprozess erfasst.
But, again, what if both parties came up with a claim construction mutually agreed upon — but which, in the courts view, is just wrong? Or, what if both parties come up with different constructions which are both wrong in the court’s view? Can the court then correct such (a) construction(s) if it would have to rely on ‘factual elements’ that have never been pointed at by the parties, just because it is a question of law? And if that was the case, then where is the limit in the assessment of obviousness — which is a question of law, too?
The theory of separating questions of law from questions of fact is simple. But life is complicated …
On a sidenote: Wouldn’t it be good to establish common ground on what the claim actually is all about before diving any deeper? This is what a Markman hearing in the U.S. is all about. Even though I am not overly enthusiastic about quite some aspects of patent litigation in the U.S., this one certainly is useful.
These proceedings are following-up on the case O2014_002 (discussed on this Blog here in more detail) where the FPC held that certain non-return valves for urinals produced by von Allmen AG were infringing EP 1 579 133 B1.
From what has been argued at the hearing, it appears that the defendant had obtained valves from a Swiss supplier (presumably Enswico AG; meanwhile liquidated) which in turn had obtained the valves from von Allmen AG. Thus, infringement was not an issue anymore.
Based on the information obtained in proceedings O2014_002, the plaintiff alleged that Aquis Systems had obtained 17’500 infringing valves. On the contrary, Acquis Systems argued that it had only obtained 4’700 valves in total, that those valves had been produced while a license agreement had still been in place and that it had believed in good faith that a license fee of CHF 1,85 per piece had been paid to Mr Dahm. Note that the patent had initially been granted to Mr Dahm, but has been assigned to the plaintiff Daspaco AG in 2013; see Swissreg.
Further, Aquis Systems argued that Daspaco had no standig to bring this suit, since it had not been the proprietor of the patent at the relevant time.
The presiding judge pushed the parties to have settlement discussions, given the small amounts of money involved. And, indeed, both parties agreed.
If in a patent infringement proceeding the plaintiff files before the instructional hearing a reply limited to the defense of the validity and no pleadings take place at the instructional hearing, the plaintiff may submit new allegations of fact, and therefore also amended patent claims, in the completed reply filed after the instructional hearing without having to meet the requirements of art. 229 para. 1 CPC.
Please find some background information about the subject-matter at stake on this Blog here, and a report about the main hearing of 29 October 2018 on this Blog here.
The only independent claim 1 of CH’755 as granted reads as follows:
Durchflussmessfühler (10) mit einem ein zylindrisches Gehäuse (10) definierenden Durchgang mit einer ersten Durchgangsöffnung (13) und einer zweiten Durchgangsöffnung (29); einem im Durchgang des Gehäuses angeordneten Strömungswiderstand (23), welcher das Gehäuse in einen ersten und einen zweiten Gehäuseteil (11 und 27) unterteilt; einer ersten Anschlussstelle (17) mit einer Verbindung zum Innern des ersten Gehäuseteils (11); und einer zweiten Anschlussstelle (19) mit einer Verbindung zum Innern des zweiten Gehäuseteils; dadurch gekennzeichnet, dass die ersten und zweiten Anschlussstellen (17, 19) in einem Abstand voneinander auf dem gleichen Gehäuseteil angeordnet sind.
Inofficially translated to English language:
Flow sensor (10) with a passage defining a cylindrical housing (10) with a first passage opening (13) and a second passage opening (29); a flow resistor (23) arranged in the passage of the housing which divides the housing into a first and a second housing part (11 and 27); a first terminal (17) having a connection to the interior of the first housing part (11); and a second terminal (19) having a connection to the interior of the second housing part; characterised in that the first and second terminals (17, 19) are spaced apart on the same housing part.
Fig. 1 of CH’755 surely helps to understand what the invention is about:
The defendant denied an infringement, and argued for nullity as a plea in defense.
The verbal limitation of the patent
This is what the hn is all about. Imagine a plaintiff / patentee in infringement proceedings who is confronted with a plea for nullity in defense: When is his last chance to formally assert his patent only to a more limited and presumably valid extent?
First, it is important to understand how the FPC conducts the proceedings. If the defendant raises a plea for nullity in his answer to plaintiff’s statement of claim in infringement proceedings, the FPC invites the plaintiff to file a reply which is strictly limited to only the nullity aspects of the case. What follows next is an instructional hearing. In preparation of that hearing, the limitation of plaintiff’s reply to only the nullity aspects is to assure that the court has a single exchange of opinions at hand for both infringement and nullity. If the case moves on after the instructional hearing, it is again up to the plaintiff to complete his reply. Thereafter, the defendant provides the rejoinder re both infringement and nullity. This is the end of the regular exchange of briefs in which the parties can present facts and arguments without the need to comply with Art. 229(1) CPC (novae). However, the plaintiff may of course comment on new allegations and facts presented by the defendant in his rejoinder — within a deadline set by the FPC (typically two weeks, extendable by one further week; see this Blog here), or under the unconditional right to be heard (within 10 days).
The below is an illustrative summary of this chain of events:
Statement of Claim
w nullity plea in defense
re alleged nullity only
Instructional hearing w/o pleadings
re infringement and nullity
re infringement and nullity
on new factual allegations in the rejoinder, if any
Judge rapporteur’s expert opinion
Note the subtitle to the second reply in the above: The decision holds that the plaintiff may present arguments and facts in his second reply after the instructional hearing not only relating to infringement aspects, but rather also to the nullity aspects of the case.
In the defendant’s view, the plaintiff thus had three chances to deal with validity / nullity of the patent, i.e. in his initial statement of claim and both his replies. The FPC disagreed: There is no reason for a plaintiff to deal with nullity in the beginning. How would he even do so in the absence of any knowledge of the defendant’s potential arguments? Thus, the FPC held that both parties had exchanged their views on both aspects of the case for the first time only after plaintiff’s first reply (limited to nullity). With the second reply and the rejoinder, both without limitations re novae, the double exchange of briefs is concluded.
Undoubtedly, an amended claim is to be considered a new fact (decision 4A_543/2017 of the Supreme Court, ¶2.3); as such, it would be an inadmissible novum after the normal exchange of briefs.
The plaintiff ‘verbally limited’ the claim only in his second reply. The defendant argued that this second reply was limited to only infringement aspects; it should only complement the first reply that had already dealt with nullity. Defendant argued that this was established practice at the FPC. Again, the FPC disagreed: There was not an established practice of the FPC in that sense.
Interestingly, the former President Dieter Brändle had apparently indicated in the instructional hearing that the patent could not be asserted in a more limited form any more after the instructional hearing. Defendant argued that it is against good faith if the limitation was now accepted nevertheless. However, the decision holds that it is up to the whole panel of judges to finally rule on admissibility of new facts; Mr. Brändle only gave his personal, preliminary opinion on this issue at that time. When the defendant had been invited to provide his rejoinder, this came along with an explicit note to deal with plaintiff’s reply as a whole, and that the whole panel of judges would finally decide on admissibility of the limitation of the patent. Thus, the defendant had no procedural disadvantage.
On a sidenote, I am wondering what the last point of time would be for a patentee / plaintiff to limit the claim when faced with a full-fledged counterclaim for nullity, i.e. when there are two parallel tracks of claim and counterclaim:
Statement of Claim
Answer re c’claim
Instructional hearing w/o pleadings
on new factual allegations in the rejoinder re c’claim, if any
Judge rapporteur’s expert opinion
Would the plaintiff’s reply after the instructional hearing then be limited to the infringement part, and the final chance to limit the claim be the rejoinder re c’claim? This remains to be seen in a case to come.
General remarks on claim construction
The decision summarises certain key aspects of claim construction in ¶25 (and, later on, in ¶47). I do readily agree with most of it, but stumbled upon the following:
When case-law refers to ‘broadest interpretation’ of claim features,15 the feature must still be capable of fulfilling its purpose in the context of the invention. This means that the claim must not be interpreted under its wording, but also not in such a way that embodiments are covered which do not achieve the inventive effect.
15 FPC, decision O2013_008 of 25 August 2015, ¶4.2 — ‘elektrostatische Pulversprühpistole’ [see this Blog here]
The decision cites ample case-law, but there is no fn in the last sentence. Now, where does that come from? The sentence somehow reminds me of hn 3 of the ‘Spannschraube’ decision of the German Federal Supreme Court, X ZR 85/96 of 2 March 1999. The German Supreme Court had held:
The scope of protection of a European patent cannot be extended to embodiments which wholly (or to an extent which is practically insignificant) waive the success sought by the patent.
However, the Swiss Supreme Court had to assess the same subject-matter shortly thereafter, and explicitly disagreed; 4C.348/1999 of 12 April 2001: What had been considered to be no infringement in Germany, was well an infringement in Switzerland!
The Swiss Supreme Court did not see any reason to deviate from the literal and technically plausible meaning of the feature — even though an explicit object of the patent that had been recited in the specification was not achieved by such embodiments, i.e. the use of very short clamping screws (see EP 0 319 521 B1, col. 1, l. 54-57):
Der Erfindung liegt die Aufgabe zugrunde, eine Rohrschelle der eingangs genannten Art zu schaffen, die sich auch mit einer sehr kurzen Spannschraube gut handhaben und leicht schliessen lässt.
It remains to be seen whether the present decision might indicate a gradual departure from the old Spannschrauben approach of the Swiss Supreme Court.
The claims at stake can be summarized as follows, structured into features:
Flow sensor (10) with
a passage defining a cylindrical housing (10) having a first through opening (13) and a second through opening (29);
a flow resistor (23) arranged in the passage of the housing, which divides the housing into a first and a second housing part (11 and 27);
a first connection point (17) having a connection to the interior of the first housing part (11);
a second connection point (19) having a connection to the interior of the second housing part;
the first and second connection points (17, 19) being arranged at a distance from one another on the same housing part;
claim 4 as granted:
wherein flanges are provided on the first housing part (11) and on the second housing part (27) for connecting the housing parts with flange pressure surfaces (15, 16, 31, 32) are formed;
first variant of claim 10 as granted:
wherein the connections run partly in the flange pressure surfaces,
second variant of claim 10 as granted:
[wherein the connections] exit through the flange pressure surface as openings on the sealing surface side (17, 19, 21, 33, 35, 37) .
Features a) – f) correspond to claim 1 as granted (marked in bold in the above); the whole set of features, i.e. a) – f) and j) – k), corresponds to the claim as limited with plaintiff’s second reply. Don’t get confused by the missing features g) and h): The parties referred to these features in their feature analysis; but they are irrelevant for the present decision since they only relate to dependent claims 2 and 3 which are not at stake.
The decision holds that claim 1 as granted lacks novelty over document E10, i.e. (JP S61-205023); see Fig. 1:
Next, the decision assesses and acknowledges novelty of the more limited claim over E1 (US 6,585,662 B1) which lacks features j) and k). Defendant had also argued that a prior use of the device of E1 had occured (prior use ‘Jones’). However, the decision holds that this prior use does not need to be assessed any further: Defendant had argued that the prior use disclosed a device according to the teaching of E1. Thus, the decision holds that any such prior use would necessarily also have lacked features j) and k).
Moving on to the assessment of obviousness, the decision emphasizes that it is up to the party that alleges nullity of a patent to make the relevant allegations. The court only examines whether the grounds put forward lead to the invalidity of the patent. Even though obviousness is a question of law, the relevant factual assertions must be made by the party that asserts nullity. In the following, therefore, the decision only deals with the assessment of obviousness starting from document E2 that has been referred to as ‘closest prior art’ by the defendant, i.e. US 4,083,245.
Features f), j) and k) are missing in E2.
With E2 as closest prior art, the decision holds that documents E7 (DE 32 25 114 C1, erroneously referred to as DE 32 25 115 C1 in the decision), E8 (US 1,768,563 A) and E9 (JP H03-21735 U) fail to render the limited claim obvious — mainly because these devices are made from metal / intended for use in high pressure applications, contrary to flow sensors in the medical field which are typically made of plastic, using die casting methods.
In sum, the more limited claim that the plaintiff had submitted with his second reply was thus held to be valid.
Defendant most fiercely contested that the adult version of the flow sensors has flange pressure surfaces and, even if there were flange pressure surfaces, that the connections from the connection points to the inside of the first/second half of the housing run partially into or through the flange pressure surfaces. Defendant referred to the computer tomographic images that had been submitted by the plaintiff:
The above figure shows a cross-section through theflanges of the two halves of the housing. The flow resistor which is clamped between the flanges is marked with 23. Identifiers 45 and 47 mark a groove and comb structure. Defendant argued that the gas-tight connection is made exclusively via the groove/comb structure. However, the flange surfaces would not touch each other (red arrows in the above figure); only the planes of the comb and groove that protrude perpendicularly from the flange surface would be in contact (white arrows in the above figure). Defendant argued that this is intentional because it allows flow resistors of different thicknesses to be used without the housing halves having to be designed differently, contrary to the teaching of the patent in suit. Since, by definition, flange pressure surfaces would have to lie against each other under pressure, flange pressure surfaces are missing in the adult version, in the defendant’s view. And even if the area in which the flow resistor is trapped by the flanges (marked with a yellow arrow in the figure above) were still called the ‘flange pressure surface’, no connections would occur there in any case. Defendant referred to the below figure in this respect (with the white squares / annotations added by the court):
The above figure again shows a cross-section of an adult version of the flow sensor. The thin vertical line is the flow resistor; identifier 33 refers to one of the connections to a connection point which seems to end on the right because it runs diagonally in the Z plane. The decision holds that ‘flange pressure surface’ and ‘flange surface’ are used as synonyms in the patent. A flange pressure surface in the sense of the claim is the surface of a flange facing away from the body part to which the flange is attached and serving to connect the body part to another body part, whereby the connection need not necessarily be made over the entire surface.
Thus, the flange pressure surface of the attacked embodiment runs over the entire width of the flange, i.e. in the above figure from point a) to point b). Consequently, the decision holds that the connections from the connection points to the inside of the first/second housing parts also run through the flange pressure surfaces, as can be seen in the figure above.
The decision holds that the defendant may have improved the teaching of the patent in suit, and the improvement might even be patentable. This does not, however, lead out of the scope of protection of the earlier patent (cf. Art. 36 PatA). In sum, the decision holds that the version for adults is literally covered by the limited claim of the patent. Likewise, the version for children is also held to be literally covered by said claim.
Wording of the injunctive relief
The defendant objected to the wording of the request for injunctive relief, for that it is too indefinite and merely recites the wording of claim features that are under dispute between the parties (such as e.g. the term ‘flange pressure surface’). The decision holds that the features are defined in the reasons of the decision beyond any doubt, so that a mere factual check by the competent authorities is well sufficient. Thus, it is neither necessary nor helpful for each and every feature to be defined in the prayer for relief itself.
Again, this reminds me a bit of the German approach where the mere claim wording is typically used in prayers for injunctive relief, and the exact scope is then to be deduced from the reasons of the decision. I do have some symphathy for the FPC’s approach that keeps the verdict clean and simple, with all the potential ambiguities being resolved by way of definitions given in the reasons.
In sum, the defendant was ordered to abstain from further use of the infringing embodiments, and to provide information and accounting concerning the same.
The parties commented on the decision with the following press releases:
The decision is not yet final; an appeal is pending at the Supreme Court.
The patentee who deletes their patent from the patent register after the filing of a nullity action is liable to pay the procedural costs even if they have not been forewarned before the filing of the action.
The hn highlights in no uncertain terms the risk associated with maintaining patents of doubtful validity in Switzerland: Even in case of a nullity action out of the blue, and even if the patent is then withdrawn immediately, the patentee / defendant still has to bear the procedural costs.
Frankly, this does not come as a complete surprise. The tendency has already been clear since O2015_010, decision of 5 January 2016. The present order again confirms that Swiss law (unlike Germany, cf. § 93 DE-ZPO) does not provide for an obligation to issue a prior warning letter. With reference to the practice of the cantonal court of Zug (A3 2010 58, ‘Geburtsgel’), the order holds that clapping all costs on the defendant is justified if the defendant had
[…] given the impression by his pre-litigation conduct that he had not had the patent cancelled on mere warning.
Noteworthy, the present order holds that this conclusion can already be drawn from the mere existence of a formally valid patent. This is because the patent is cancelled if the renewal fees due are not paid on time (Art. 15(1) lit. b PatA). It would be unreasonable to assume that the patentee would pay the annuities on time but cancel the patent on first demand.
And, in fact, the patentee in the present matter did not respond immediately to the (informal) notification of the action by cancelling the patent, but rather offered the plaintiff a licence. This suggests that even in the case of a pre-litigation request, the patent would not have been cancelled immediately.
The order has been published in highly truncated form only. But still, I am reasonably sure about the patent at stake. Searches in Swissreg give only a single perfect fit, i.e. EP(CH) 1 204 164 B1 of Feller AG; see Swissreg. Annuities had been validly paid until 30 November 2018, but the patent had been withdrawn shortly before that date, with letter of 6 November 2018 (published in Swissreg on 8 November 2018).
The invention of EP’164 apparently is all about a screwless connecting terminal for electrical conductors, with more secure releasability of the conductors. Claim 1 reads as follows:
Device (1) with screwless terminals for connecting electrical conductors, with at least one spring comprising an elastic terminal arm (7) and a brace (2) which encloses it, where a contact tongue (8) of the terminal arm (7) cooperates with a retaining part (5) of the brace (2) so as to maintain a conductor introduced between the contact tongue (8) and the retaining part (5), where the brace (2) leaves the contact tongue (8) at least partially accessible on both sides of the retaining part (5), so that it can be acted upon from outside the brace (2) on both sides of the retaining part (5), characterized in that the brace (2) comprises a base (3) and a lateral wall (4) the upper part of which is bent so as to form the retaining part (5) in such a way that this retaining part (5) is only connected through its longer side with the lateral wall (4), and that the latter comprises at least one recess (11) in the vicinity of the extremity of the contact tongue (8).
All this is more readily understandable with Fig. 1 of EP’164 at hand:
Unfortunately, I failed miserably to track the plaintiff of this case. If someone out there does know more, please let me know.
I had reported about the ownership dispute in summary proceedings between between David Bensoussan and Rock dental AG on this Blog here.
Except for the complaint, nothing had surfaced in main proceedings O2017_026, though. See the full complaint below. I now came across an order that had been issued on 11 September 2018: The parties apparently found an amicable solution. The order had not been published by the FPC, but the terms agreed upon by the parties do not contain any confidentiality obligation. It is rare that one can catch a glimpse of how cases are settled at the FPC: The full order is available in the file wrapper at the EPO in relation to the European phase of the patent application concerned. In brief, the parties agreed as follows:
Marc Fehlmann and David Bensoussan are joint inventors.
Rock dental AG shall be the sole owner of the patent applications / patents.
David Bensoussan shall be reimbursed for assignment of his share of rights in the patent applications / patents with a 20% share of the license income.
As a result of the settlement, case O2017_026 was written off as settled.
Salmon Pharma as the defendant in SPC infringement proceedings neither disputed validity of the basic patent EP 0 716 606 B1 of Genzyme Corporation, nor that the subject-matter of the SPC C00716606/01 is actually covered by the basic patent or that the attacked embodiment (sevelamer carbonate) is covered by the SPC. Rather, the defendant (only) alleged that the SPC was invalid because the IPI wrongfully allowed re-establishment of rights (Art. 47 PatA) with respect to the time limit for filing the SPC application under Art. 140f PatA.
The question as to whether the lists of grounds for nullity in Art. 26 and Art. 140k PatA are exhaustive had indeed not yet been explicitly decided by the Supreme Court before. However, the present judgment refers to an indication in 4A_52/2008, ¶2.2, according to which the grounds for invalidity for the certificate are listed in Art. 140k PatA. Furthermore, in application of the Patent Act of 1888, the Federal Supreme Court had already held that the grounds for invalidity of patents were exhaustively listed therein (28 II 309, ¶5). I just love to read such old decisions. Amazingly straight to the point; and the type face is so stylish:
Unsurprisingly, the Supreme Court takes the case-law of the Court of Justice of the European Union (ECJ) on Art. 15 of Regulation (EC) No 469/2009 (or Regulation (EEC) No 1768/92) into account, in view of the legislator’s explicit intent to bring the Swiss rules into line with European law.
However, the Supreme Court could not infer anything from the ECJ’s case-law in support of Salmon’s position. Rather, the Supreme Court confirms the FPC’s reasoning: The ECJ only affirmed the nullity of certificates due to the non-observance of standards which are not explicitly mentioned in Art. 15 of Regulation No 1768/92 by referring to Art. 15 as interpreted in view of Art. 3.
For example, in C-127/00 — Hässle ./. Ratiopharm, the ECJ held that failure to comply with the transitional regime of Art. 19 of Regulation No 1768/92 may result in the certificate being invalid. The requirement under Art. 19 is to be understood as an implicit substantive requirement additional to the conditions laid down in Art. 3 of the Regulation. A certificate which was granted even though an initial marketing authorisation had already been granted before the date laid down in Article 19 of the Regulation is therefore null and void; ¶¶84-92).
Likewise, in C-195/09 — Synthon ./. Merz Pharma, the ECJ held that Art. 3 of Regulation No 1768/92 necessarily referred to a product falling within the scope of that regulation within the meaning of Art. 2 thereof. A certificate granted for a product falling outside the material scope of the Regulation should be considered null and void; ¶56.
It cannot be inferred from these judgments that the list of grounds for nullity in accordance with Art. 15 of the Regulation is not to be understood as exhaustive. On the contrary, the ECJ stated that it could not be inferred from either the wording or the history of Art. 15 of the Regulation that the list of grounds for nullity of the certificate referred to therein was not exhaustive (Synthon, ¶55; Hässle, ¶¶90-91).
The Supreme Court thus held that the lists of grounds for invalidity of both Art. 26 and Art. 140k PatA cannot be supplemented. Consequently, in the context of a patent nullity action or a nullity action relating to an SPC, only the grounds provided for in these provisions may be invoked. The Supreme Court notes in passing that legal certainty demands for this, too.
Angesichts dessen, dass es sich bei der Nichtigkeit eines Patents bzw. Zertifikats um eine schwerwiegende Folge handelt, entspricht dies dem Gebot der Rechtssicherheit.
Next, the Supreme Court assessed whether non-compliance with the deadline for filing the request pursuant to Art. 140f PatA or wrongful reinstatement pursuant to Art. 47 PatA, could be subsumed under a ground for invalidity pursuant to Art. 140k PatA. However, this is neither apparent nor does it result from the relevant case-law of the ECJ.
Finally, the Supreme Court notes that SPCs and patents are granted by order of the IGE (144 III 285, ¶3.2). Nullity under Art. 26 and Art. 140k PatA is not to be mixed up with the administrative contestability or nullity of the dispositions of the IPI on which the grant of a patent or SPC is based. In other words, the fact that the catalogues of grounds for invalidity under the PatA are exhaustive does not mean that these dispositions cannot be contestable or void, which must be assessed according to general principles of administrative law.
Dass die Kataloge der patentrechtlichen Nichtigkeitsgründe abschliessend sind, bedeutet […] nicht, dass diese Verfügungen nicht anfechtbar bzw. nichtig sein können, was nach allgemeinen verwaltungsrechtlichen Grundsätzen zu beurteilen ist.
Interestingly, the Supreme Court explicitly left the question open whether the decision of grant of the SPC (or the earlier decision of 4 April 2005 concerning reinstatement) could have been challenged by the appellant, as the FPC had assumed: Any relevant time limit had expired since long, anyway. On the other hand, the nullity of a decision must be observed by all authorities applying the law at all times (137 I 273, ¶3.1; 133 II 366, ¶¶3.1-3.2; 132 II 342, ¶2.1; 129 I 361, ¶2; each with further references to case-law).
Incorrect administrative acts are usually not null and void, but only contestable; and they become legally effective when not challenged. Invalidity, i.e. absolute ineffectiveness, of an order is only accepted if it is afflicted with a profound and substantial defect, if this serious defect is obvious or at least easily recognisable and if the legal certainty is not seriously jeopardised by the acceptance of invalidity. Deficiencies in content only rarely result in the nullity of an order, in exceptional cases; this requires an extraordinarily serious deficiency. The main grounds for invalidity are the functional and factual lack of competence of an authority and serious procedural irregularities (such as the fact that the person concerned did not have the opportunity to participate in the proceedings). If an order is not legally binding in this sense, any authority dealing with the matter must comply with it at all times and ex officio (see, inter alia, 138_II_501, ¶3.1; 137 I 273, ¶3.1; 136_II_489, ¶3.3).
Salmon had argued that reinstatement was granted ‘arbitrarily and contrary to the clear and decades-long practice’ of the IPI. Again, the Supreme Court left undecided whether the IPI indeed wrongly allowed reinstatement because the alleged misjudgment could not be a defect that leads to the nullity of the decision. In particular, the Supreme Court held that any such deficiency would be neither obvious nor easily recognisable: By definition, reinstatement is a remedy when something went wrong, i.e. in the event that a time limit prescribed by the law was not observed.
Finally, in simple terms, the Supreme Court notes: The IPI had come to the conclusion that the conditions for reinstatement pursuant to Art. 47 PatA were met. The appellant may not share this view, but the administrative act is still not null and void.
Das IGE ist zum Schluss gekommen, dass die Voraussetzungen für die Wiedereinsetzung gemäss Art. 47 PatG gegeben waren. Die Beschwerdeführerin mag diese Einschätzung nicht teilen; ein nichtiger Verwaltungsakt liegt jedoch nicht vor.
Note that Hepp Wenger Ryffel is involved in this matter on behalf of the plaintiff / appellant.
Please see this Blog here for a discussion of the underlying decision of the FPC. In brief, it is all about an assignment action of Marcel Riendeau against Zehnder Group International AG, concerning two European patent applications, i.e.:
Both EP2 and EP3 had been filed in Zehnder’s name only; the plaintiff had been designated as inventor. The plaintiff had partially suceeded in first instance when the FPC held that the parties are jointly entitled to both patent applications.
On appeal, the plaintiff objected that the facts had been wrongly established in the first instance decision in various aspects, that the burden of proof had been incorrectly allocated and that the FPC had not dealt with the plaintiff’s entitlement to a transfer of the patent applications based on a contractual undertaking.
To cut a long(er) story short: The Supreme Court did not agree and dismissed the appeal. The FPC’s decision is thus final now, and it remains to be seen how the parties get along with their joint entitlement to both EP2 and EP3.
The Supreme Court now dismissed Omega’s appeal, i.e. confirmed the FPC’s decision. EP(CH) 1 837 719 B1 is thus finally declared invalid.
The issues dealt with by the Supreme Court in the decision, in brief:
Age of a prior art document
On appeal, Omega has argued that the FPC took an unrealistic piece of prior art into account in the assessment of inventive step. Indeed, the FPC had considered a document that was published already about a century ago, i.e. US 759,914. In view of the evolution of watchmaking since then, Omega considered it unrealistic that such a document was actually considered as a starting point in the assessment of inventive step. Consequently, Omega alleged a violation of Art. 56 EPC.
The Supreme Court did not buy this argument. It is held in no uncertain terms that everything that had been made publicly available — including old documents — forms part of the prior art. One cannot exclude a document from consideration in the assessment of inventive step just because of its age; see ¶3.1.3.
Écarter un document de ceux que consulterait l’homme de métier en raison de son ancienneté reviendrait à priver les brevets ayant dépasé un certain âge de toute valeur dans le cadre de l’analyse de l’effet invenif. Un telle pratique ne saurait être déduite de l’art. 65 CBE.
Reference to some case-law of the Boards of Appeal of the EPO did not help either: The mere age of a document has not been decisive in T 479/00, T 366/89, T 1000/92 and T 334/92. The Supreme Court notes that the obsolescence or outdatedness of a technology had been the key issue in these decisions, which excluded the respective documents from consideration by the skilled person. Apparently, no such case had been established by the appellant in the present matter.
Incompatibility of technical teachings
The balance wheel of US 759,914 has inertia blocks that must have been placed from the inside, as claimed by the patent in suit; see the screw head on the inner side of the felloe in the figures below.
But what is still missing here is the stud (‘plot’) according according to the patent in suit. Omega argued that the skilled person would not have added studs to the balance of US 759,914 in view of the risk to impair the automatic adjustment of its temperature dependent moment of inertia, which is the sole purpose of this invention. However, the Supreme Court holds that this is not in agreement with the fact findings of the FPC, and thus did not consider this argument any further.
And the Supreme Court indeed acknowledges that it would have been desirable if it had done so.
Ainsi, il eût été souhaitable que la juridiction précédente détaille la première étape de l’approche problème-solution comme elle l’a fait pour les autres étapes de son examen.
But still, in view of the FPC’s detailed discussion of how the skilled person had arrived at the invention starting from US 759,914 without inventive merit, it became sufficiently clear how close both inventions actually are, and why US’914 thus is a reasonably chosen closest prior art.
Late filed auxiliary requests / claim limitations
Omega had filed yet another, even more limited auxiliary request (AR3) with the response to the expert opinion of the judge-rapporteur. The FPC had not admitted this request into the proceedings anymore, with reference to O2015_012.
Omega had challenged this on appeal. But while this appeal has been pending, the Supreme Court ruled on the appeal against O2015_012 — and backed the practice of the FPC in this respect; decision 4A_543/2017 of 08 May 2018, see this Blog here. Accordingly, Omega did not succeed with this argument, either.