For the sake of the argument

As I have already mentioned here, one reason why I do not often comment on decisions from the Boards of appeal of the EPO on this blog is that any interesting decision that comes out is instantly shot at by all good bloggers around the place before I even get a chance to draw my metaphorical gun.

So, sorry folks, no breaking news from me today, as Laurent Teyssèdre has already been here, as well as others.

But I could not let this one decision T 1914/12 go without a comment or two.

First, because it is highly interesting. Second, because it is in French (so, maybe some non-French speaking colleagues would benefit from further highlights). Third, because I was involved in the case at hand in the first instance proceedings, so this one brings back a number of memories.

The catchword of the decision can be translated as follows:

Boards of appeal do not have any appraisal power with respect to the admissibility of late arguments which are based on facts that are already part of the proceedings (decision diverging from T 1621/09).

A brief presentation of the facts of the case will probably suffice, before summarizing the very thorough legal discussion in the decision.

The patent at stake is entitled “Method of adhesively bonding a strip of flexible fabric to a substrate“. It was opposed after grant and revoked by the opposition division, for lack of inventive step (as well as extension of subject-matter as far as some of the auxiliary requests are concerned).

In the course of the appeal proceedings, the opposition was withdrawn, and the patent was transferred to the previous opponent. Not that this circumstance is really critical.

The key issue is that, for the first time on appeal, the new patent proprietor identified a further distinguishing feature relative to one of the main cited prior art references.

In a written communication, the Board expressed the preliminary opinion that this was a late argument and that its intention was not to admit it into the proceedings. But the Board, for once presided by the legal member, later changed its mind and came to the opposite conclusion. As a result, the first instance decision was set aside, and the case remitted to the first instance for further prosecution.

But before reaching this conclusion, the Board performed a very thorough analysis of the law pertaining to late facts and arguments.

A fancy EPO appeal clock.

The starting point for the discussion is article 13(1) of the Rules of procedure of the Boards of appeal (RPBA), per which:

Any amendment to a party’s case after it has filed its grounds of appeal or reply may be admitted and considered at the Board’s discretion. The discretion shall be exercised in view of inter alia the complexity of the new subject matter submitted, the current state of the proceedings and the need for procedural economy. 

So the question is: does the discretionary power of the Board apply to a situation such as the one at hand, where a new distinguishing feature is identified by the proprietor at a late stage?

As a first step (section 7.1 of the reasons), the Board reviewed in great detail the notions of “facts” and “arguments” and their respective definitions, based on the common understanding of the terms, of the legal understanding of the terms, and on the specific use of the terms in the EPC and case law.

Clearly distinguishing between facts and arguments is not just a matter of hair-splitting. Depending on whether a late submission is viewed as a fact or an argument, the fate of the submission may differ – even in first instance.

As a side note, in one opposition I have handled, a new inventive step attack based on a timely filed document only later identified as a possible closest prior art document was considered by the opposition division as pertaining to a “new fact” and not merely a “new argument” and it was thus not admitted into the proceedings. The Board set aside the first instance decision – but I don’t know yet if this point will be discussed in the reasons or not.

Anyway, the definition offered by the present Board (reasons, 7.1.4) is the following (in my own translation as always):

A “fact” can be understood as a factual (or allegedly factual) element or a circumstance on which a party bases its case, whereas an “argument” designates a proposition that a party bases on one or more facts and which supports the case that it puts forward. 

The Board then gave the example of a lack of novelty attack:

  • The party’s case (“moyen” in French) is that the claimed subject-matter is comprised in the state of the art.
  • One argument is e.g. that the subject-matter is disclosed in paragraph [0017] of D1.
  • Another argument is e.g. that the skilled person understands the paragraph in a certain manner, in view of common general knowledge, and thus views a certain feature as implicitly disclosed.
  • The textual content of the paragraph is a fact.
  • The copy of the document or of its translation is a piece of evidence.

As a next step, the Board addressed the central issue of whether late arguments can be disregarded or not.

Here, the first interesting remark is that the text of the EPC is not the same in the three official languages.

The English version of article 114(1) reads:

In proceedings before it, the European Patent Office shall examine the facts of its own motion; it shall not be restricted in this examination to the facts, evidence and arguments provided by the parties and the relief sought. 

Note the term “arguments” here. But the French version is in fact quite different, as it solely refers to “moyens invoqués” and “demandes présentées“. I would tend to translate the general term “moyens invoqués” by “the party’s case“. The German version does not specifically mention arguments either.

The Board noted that the English version seems to more accurately express the intentions of the drafters of the EPC (is that generally the case? the Frenchman asks), and therefore adhered to this English version in the rest of the reasoning.

Now the key part of the reasoning is that the English version of the second paragraph of article 114, on which the discretionary power of the EPO to disregard late submissions is based, does not refer to arguments:

The European Patent Office may disregard facts or evidence which are not submitted in due time by the parties concerned. 

This is in contrast to the first paragraph mentioned above, which deals with examination of the EPO’s own motion.

Based on the difference between the two paragraphs, the Board concluded that the discretionary power of the EPO is strictly limited to facts and evidence and does not extend to arguments.

Now going back to article 13(1) RBPA, there is a reference here to an “amendment to a party’s case“, which is not very explicit. But the notion of a party’s case is further defined in article 12(2):

The statement of grounds of appeal and the reply shall contain a party’s complete case. They shall set out clearly and concisely the reasons why it is requested that the decision under appeal be reversed, amended or upheld, and should specify expressly all the facts, arguments and evidence relied on. […]

So, on the face of it, it can be surmised that article 13(1) RPBA, via its implicit reference to article 12(2), empowers the Board with the right not to admit a late modification of the arguments of a party on appeal.

But, reviewing the Travaux préparatoires for the current version of the RPBA, the Board found that a proposed article 13(1) containing an explicit reference to arguments was rejected in the drafting process.

Thus, the Board held, the intent seems to have been in fact not to encompass arguments within the purview of article 13(1). See section 7.2.1 of the reasons.

Then, the Board reviewed the relevant case law in detail (section 7.2.2 of the reasons), and found a number of decisions pursuant to which arguments are not subjected to the Boards’ discretionary power.

But then there are T 1069/08 and above all T 1621/09, which decided otherwise. Especially in T 1621/09, an analysis of the travaux préparatoires was performed but the deletion of the word “arguments” in the drafting process of new article 13(1) was not noted.

In the wake of these two decisions, the Board counted 17 later decisions which relied on the principles set out in T 1069/08 and T 1621/09 without questioning them. A couple of recent decisions have however decided the other way around.

As a conclusion, the Board expressed its disagreement with the approach of T 1069/08 and T 1621/09 (section 7.2.3 of the reasons).

The final reason put forward by the Board for such disagreement is that the RPBA should not “lead to a situation which would be incompatible with the spirit and purpose of the Convention” (article 23), which would precisely be the case if article 13(1) RPBA were to be interpreted as encompassing arguments, although article 114(2) EPC, which prevails, does explicitly not refer to arguments, among the late submissions which may be disregarded.

One first question immediately comes to mind: is this question ripe for a referral to the Enlarged board of appeal?

After all, there are a number of diverging decisions, it seems. I guess the answer may depend on whether the Boards which issued T 1069/08 and T 1621/09 will stand their ground or change course.

The second question that I have is: what about the upcoming revised version of the RPBA?

A few months ago, there was a public consultation at the EPO on a draft. I do not know where this project stands right now. But, for what it is worth, there was a new paragraph 2 in article 12 in this proposal, which reads:

In view of the primary object of the appeal proceedings to judicially review the decision under appeal, a party’s appeal case shall be directed to the requests, facts, objections, arguments and evidence on which the decision under appeal was based.

Beside the fact that there is a new concept here (namely “objections“), this provision seems to introduce a limitation to what kind of arguments a party is allowed to file on appeal – namely, those on which the decision under appeal was based.

If one follows the reasoning of T 1914/12, wouldn’t this proposed provision breach article 114 EPC?


CASE REFERENCE: T 1914/12, Board 3.2.05, June 13, 2018, Gaztransport et Technigaz.

Standard or not standard?

Traditionally, non-practicing entity patent infringement suits have been  relatively few and far between in France.

Without the huge financial stakes of U.S. litigation, and without the lure of German bifurcation, the French legal system has, to some extent, resisted the trend so far. There are a few counter-examples, though.

Like the case discussed today, between famous Intellectual Ventures (IV) and SFR, one of the major French telecommunications companies.

Among the many patents owned by IV is EP 1304002, entitled “Arranging Data Ciphering In a Wireless Telecommuncation System” and originally filed as a PCT application by Nokia Mobile Phones Ltd, in Finland.

As you have probably already guessed, IV initiated legal proceedings in front of the Paris Tribunal de grande instance (TGI), alleging infringement of the EP’002 patent. The alleged infringement mainly relates to the WiFi technology.

You will not know who won at the end of the post, because no one did – the lawsuit is not over yet. In fact, the pleadings hearing is presently scheduled to take place on November 23, 2018. So there is still plenty of time left for you to take your morning off at the new courthouse.

The decision discussed today is only an interim order issued by the judge in charge of case management (JME in the local jargon).

Can you believe there are actually patent applications on methods for identifying patent troll activity? Do some of them get acquired by NPEs, just for the irony of it?

Among the various defenses raised by SFR, there is a rather interesting one, which is that the EP’002 patent is a standard essential patent (SEP).

Two standards were mentioned by SFR: 3GPP TS 33.234 (a 3G standard issued by ETSI – short for European Telecommunications Standards Institute), and 802.11.2012 (a WiFi standard issued by IEEE – short for Institute of Electrical and Electronics Engineers). According to SFR, the EP’002 patent was not declared as essential by its then owner Nokia – a member of both ETSI and IEEE – although it should have been. Due to this omission, they argued, the infringement action should be deemed inadmissible.

Now, apparently SFR did not have all the necessary evidence at hand.

They thus wrote both to ETSI and IEEE and requested the “travaux préparatoires” (one of the most well-known French expressions among lawyers, ranking third after “bonjour” and “merci”) for the relevant standards, as well as relevant correspondence between Nokia and both organizations.

France-based ETSI replied that they would provide the requested documentation, but IEEE turned down SFR’s request.

Thus, SFR filed a motion with the JME, requesting a stay and further requesting that IEEE be ordered to hand down the relevant documents. 

IV fought this motion.

First, IV claimed that SFR’s motion was inadmissible.

Indeed, IEEE is based in the U.S., and the French judge has no jurisdiction to directly order a third party to the litigation to hand down evidence, they argued. The only possible procedure would be the letter rogatory, i.e. the French judge should ask U.S. authorities to issue the order.

The judge closely looked at the written response made by IEEE further to SFR’s request. The first part of this response was the following:

Please be advised that it is IEEE’s policy not to provide information that may be used in litigation without a subpoena. This is to protect IEEE’s neutrality in any dispute. If that is necessary, please feel free to let me know and we can arrange the details of service. 

The JME deduced from this response that IEEE was willing to hand down the evidence at stake, and that they just required a formal judicial order for doing so. The judge was happy to oblige, and the formal and cumbersome process of the letter rogatory could thus be avoided.

I note that IEEE’s response does not specify whether the requested subpoena should be specifically issued by a U.S. court, or whether any subpoena would do. It was probably the judge’s understanding that, in the absence of any specific mention, they were open to any form of subpoenaing.

As a second defense against the motion, IV argued that SFR’s request was useless and unjustified.

SFR raised the SEP argument late in the case management procedure, IV said, and only they should have the burden of providing their own evidence. Besides, the second part of IEEE’s written response to SFR reads as follows:

However, a subpoena may not be necessary in this instance as IEEE 802.11 does host a public website that contains a wide variety of working group materials. The IEEE 802.11 website is located http://www.ieee802.org/11/.

But the judge held that SFR was free to raise a new defense at any point within the case schedule (which, in case you are wondering, is nothing like the extremely tight schedule envisioned for the future UPC). Besides, SFR stated that they could not find the relevant documents on the IEEE website, and that in particular the correspondence with Nokia is not supposed to be available on this site. Finally, the supply of evidence by the standard organizations should not take long.

As a result, the judge granted SFR’s request and issued an order for IEEE to send SFR all documents and reports of the working group relating to the relevant part of IEEE 802.11, as well as the relevant correspondence with Nokia, within one month.

On the other hand, the request for stay of proceedings was dismissed – probably because this timeline is supposed to be still compatible with the planned date for the main hearing.

Access to evidence can be challenging in this country especially in the absence of discovery / disclosure. French judges are often reluctant to order a taking of evidence, such as the provision of documents. But in a case such as this one, it does seem rather fair that a defendant should be allowed to explore a particular line of defense by requesting third party-held documents.

We will thus stay tuned, and I am sure the decision on the merits will be most interesting to read – if the lawsuit goes all the way of course.


CASE REFERENCE: Tribunal de grande instance de Paris, 3ème chambre 4ème section, ordonnance du juge de la mise en état, June 7, 2018, Intellectual Ventures LLC v. Société française du radiotéléphone, RG No. 16/16346.

Money money money

It is always about the money, isn’t it?

No, not life. But patent litigation? Maybe so.

Today’s lawsuit pits an optician businessman – and inventor – against the major French-based ophthalmic optics group, Essilor.

This inventor filed a French patent application concerning a nose pad invention in April 2013. On January 1, 2014, he assigned his invention to Essilor International. I should add that he was independent from Essilor, so we are not talking about an employee’s invention here.

The litigation between them later arose because the inventor was dissatisfied with the money he made from the assignment.

We thus need to look at the consideration provided in the assignment for the transfer of rights.

Based on the excerpts mentioned in the judgment, a lump sum of 400,000 euros was provided. In addition, a variable price was set, over a period of time of 19 years, according to a sophisticated formula taking into account the number of countries in which patent applications would be filed and patents would be granted, as well as the number of devices incorporating the invention which would be sold.

As far as I understand, only the lump sum was paid. In 2016, the assignor claimed additional money corresponding to the variable part of the deal. Essilor replied that there were difficulties in working the invention, and that prosecution of the patent applications on the invention was still ongoing.

In January 2017, settlement discussions were initiated, in which Essilor offered 85,700 euros as a complementary payment – to no avail. In May 2018, the inventor filed a complaint with the Paris Tribunal de grande instance (TGI), alleged a breach of contract and claimed damages.

The inventor provided several estimates for these damages. His first estimate amounted to more than 3 million euros for the period of 2015-2017, and then more than 2.7 million euros per year from 2018 to 2034. His second, auxiliary estimate was 4 million euros in total. The third, less preferred estimate was 600,000 euros in total.

Essilor fought back and requested that the action be dismissed. They also provided their own calculation for the minimal amounts to be due, as an auxiliary argument, namely 16,666 euros from year 2 to year 10 and 3,333 euros from year 11 to year 20. This corresponds to the amounts due in consideration of the current state of the patent portfolio, and does not take into account any sale.

After the nose pads… muzzle pads.

Turning now to the judges’ decision, the first interesting point relates to the issue of whether the negotiations between the parties were loyal or not.

The inventor claimed that he was not properly assisted during the negotiations, which were uneven and led him to agree to a clause on the variable part of the consideration which was unfavorable to him.

The court looked at various email exchanges between the inventor and Essilor before the agreement was executed and could not find any evidence of unfair or disloyal negotiations. The court made the following comments:

It is thus the content of an oral exchange in a time frame very close to the execution of the agreement which led to the adoption of the proportion of granted and non-refused patents as the adjusting parameter for the “provisional amount” or minimal amount, in the drafting of the agreement. But it cannot be deduced that this criterion was never previously discussed […]. Contrary to the defendant, [the inventor] provides little information on the exchanges prior to the execution of the agreement, which, based on their contents, do not reflect the alleged weakness of the plaintiff but at best his desire to rapidly achieve the execution of the assignment agreement. 

[…] Even if it is surprising that the provision relating to the variable part of the price, which is so important […], was significantly modified without any written proposal before the final version, this circumstance is not sufficient to qualify the discussions leading to the execution of the assignment as dishonest. The correspondence filed as evidence […] actually proves the very active part taken by the claimant in the different steps of the negotiation, the stakes of which he always clearly understood. 

Next, the inventor blamed Essilor for filing a PCT application based on his first French application.

Indeed, using the PCT route instead of direct national filings results in a delay in the grant proceedings. I would add that this is actually one of the main reasons why applicants file via the PCT in the first place. But in this case, any delay in the grant proceedings leads to less money payable to the inventor – in view of the definition of the variable part of the price.

The court noted that the agreement does not explicitly mention that the patent filings abroad should be done through the PCT route. Nor does it exclude the PCT route. Therefore, the court had to determine the common will of the parties. The court first noted that, according to another clause of the agreement, the assignor explicitly represented that he had not filed any direct or PCT application. The fact that a distinction between PCT and non-PCT was made in this clause (unrelated to future filings), was interpreted by the judges as a pointer to the fact that a future PCT filing should have been explicitly mentioned, in order for it to be allowed.

As a main reason and as a conclusion, the court then held that:

In view of the importance of the patent grant timeline for the general object of the contract, clause 5.2 must be interpreted, despite its lack of precision in this respect, as not allowing Essilor to file an international application. This choice was thus made in violation of the provisions of the [agreement]. 

So that’s a first rebuke for the assignee.

Then came a second one, relating to the absence of exploitation of the invention.

The court first held that, when the assignment price comprises a royalty which depends on the exploitation of the invention by the assignee, this implies a duty for the assignee to exploit the invention; otherwise the clause would be wholly one-sided.

This obligation concerns the means implemented by the assignee, and not the outcome. In other words the assignee does not have to succeed in implementing the invention, but they have to try. And the assignee has the burden of showing that they have used all appropriate means for exploiting the invention.

In this respect, Essilor referred to two messages evidencing some difficulties in the development of the nose pads of the invention.

The court commented as follows:

These two sole exhibits relating to a time span of almost two years cannot possibly show that there is a technical impossibility of exploiting the invention. They merely reveal that each version at stake turned out to raise different development issues, and that in 2016 only the “tube” version was deemed as being able to be made and marketed. 

So, that was held as another fault.

Finally, what is generally the stodgiest part of a judgment (this case being no exception): the assessment of damages.

The court started by holding that the damages should compensate for the loss of opportunity for the inventor to get royalties. Then, they took note of the objections raised by the various patent offices and the amendments made in response which, they implied, lowered the prospects of exploitation of the invention.

The court also took into account:

  • the initial estimate of the annual royalty of 100,000 euros which had been made by Essilor during the negotiation phase;
  • the defects of one of the versions of the device which led to dropping production launch;
  • the fact that the inventor subsequently filed another patent application on a nose pad.

All in all, the court discarded both the claimant’s and the defendant’s estimates, and came up with a global figure of 300,000 euros of damages compensating for the inventor’s loss of opportunity.

There is a very French expression for this approach: “à la louche“, literally “with a ladle“. It means a very rough estimate, close to guesswork. And, à la louche it is in this case, I think.

On the other hand, we cannot really blame the court for this. First, the parties came up with hugely discrepant figures, as is not unusual. Second, loss of opportunity is all about what could have been in an alternate universe, which leaves a lot of room for imagination. Third, the court was not assisted by an expert and had only the partial evidence provided by the parties at hand.

I can’t help but wondering: the parties will continue to be bound by the agreement for many years to come. So, would a settlement not be still the best option for all of those involved?


CASE REFERENCE: Tribunal de grande instance de Paris, 3ème chambre 3ème section, May 25, 2018, Jean-Luc B. v. SA Essilor International, RG No. 17/06753.

It-which-must-not-be-named

Some pharma cases are somewhat delicate to discuss in a blog post.

Case in point, if I provide the commercial name of the drug at stake in today’s litigation, I am afraid that this post may be classified as a spam and may thus never reach my email subscribers.

You see, it is the sort of drug which is prescribed for the treatment of erectile dysfunction, and which keeps coming up in these pestering unsolicited email messages that you may receive on a daily basis.

Just to be clear, today’s drug-which-must-not-be-named is not the famous one that starts with a V (containing sildenafil as an active compound), but the other famous one that starts with a C (containing tadalafil as an active compound).

He-who-must-not-be-named.

Icos Corporation (of the Eli Lilly group) is the owner of a number of European patents in connection with the C. drug.

First, there is EP 0740668, which was the basic patent for a French Supplementary Protection Certificate (SPC No. FR 03C0017), which expired in November 2017. Second, there are EP 1173181 and EP 1200092, designated as “secondary patents” by the Paris Tribunal de grande instance (TGI).

In November 2014, generic drug company Mylan obtained a marketing authorization (MA) related to the C. drug. In January 2016, Mylan initiated nullity proceedings with respect to the EP’181 and EP’092 patents in front of the Paris TGI. The parties later reached a settlement agreement with respect to EP’092, so that only the fate of EP’181 remained to be decided upon. Icos Corporation and the French distributor Lilly France counterclaimed for infringement of EP’181. The first instance judgment was issued in May 2018.

EP’181 or equivalents thereof were or are also litigated in other countries. According to the summary provided by the court, the patents were revoked in Germany, the United Kingdom, Canada and Japan. It may thus come as little surprising that the same outcome was achieved in this country. On the other hand, the ground for nullity that the TGI took into consideration is relatively unexpected, as will be apparent below.

But before getting there, let’s first look at the statute of limitations defense raised by Icos.

Mylan argued that the statute of limitations is not applicable to patent nullity suits. This argument was rejected by the court, in keeping with earlier decisions.

Turning to the determination of the starting point for the limitation period, the court recalled its now established principle of an in concreto determination.

The court thus explained that the grant of the EP’181 patent was not the starting point for the limitation period. The general principle is the following:

The starting point for the limitation period must thus be set at the date, determined in concreto, at which Mylan was or should have been aware of EP’181, due to its intent to market a generic of the drug [C.], which led to the MA obtained on November 21, 2014, since this patent is an impediment to its exploitation.  

In this case, a determining factor to be taken into account was the date at which Icos obtained its own MA:

In this case, the first MA for [C.] was granted in November 2002. By way of application of article R. 5121-28 of the Code de la santé publique, the generic company can only apply for an MA as from the eighth year after the grant of the originator’s MA, and cannot be granted one before the tenth year. Therefore, Mylan could not file an MA application before November 2010.

This reasoning is fully consistent with that applied in another recent case which already involved Mylan.

However, this is not the end of the story here. The court further held:

In this case, an additional fact should be taken into account in the in concreto analysis of standing and the starting point for the limitation period. […] [Namely, Icos corporation] filed a request for limitation of the EP’181 patent on February 14, 2014 with the European patent office, and the limitation of the patent was published on March 25, 2015. 

Thus the patent enforceable against Mylan could only be known on this date, so that the starting point for the limitation period is March 25, 2015. 

In another recent case, the starting point of the limitation period was postponed by a court to the date of the decision of the Board of appeal of the EPO in the opposition appeal regarding the patent at stake. The relevant paragraph of this decision may be worth quoting again here:

[…] It is only on [July, 7, 2014, i.e. the date of the Board of appeal’s decision] that the drafting of the patent which is sought to be revoked was stabilized and that Ethypharm was able to precisely know the content of the claims of said patent as well as all the facts making it possible for them to act, so that the action is not time-barred and is admissible. 

We now have a confirmation that limitation proceedings, just like opposition proceedings, may result in a postponement of the limitation period for nullity actions.

It remains to be seen how general this principle is and in particular whether it extends e.g. to the impact of other lawsuits involving third parties.

Turning now to the merits of the case, claim 1 of EP’181 as limited reads as follows:

A pharmaceutical unit dosage composition comprising 1 to 5 mg of [tadalafil], said unit dosage form suitable for oral administration up to a maximum total dose of 5 mg per day.

Independent claim 10 is a Swiss-type claim containing similar features.

Mylan raised all classical grounds for nullity, but the court focused on insufficiency of disclosure.

After reviewing the description of the patent, the court noted the following facts:

  • There are several molecules belonging to the class of type 5 phosphodiesterase (PDE5) inhibitors.
  • Among them, particular reference may be made to sildenafil, the active compound of V., marketed at the priority date of the patent in doses of 25, 50 and 100 mg.
  • However, sildenafil generates a number of side effects, such as facial red patches, or a lowering of blood pressure.
  • The invention thus relates to a low dosage of the known alternative drug tadalafil, in order to provide an effective treatment of erectile dysfunction without the side effects associated with sildenafil.
  • The patent also contains a number of examples showing the efficacy and the absence of side effects of low dosage forms of tadalafil.

The court was apparently quite puzzled by the patent as a whole:

The problem expressed in the description of the patent is to provide a principle which avoids the issues of red patches and side effects of sildenafil by a particular dosage of tadalafil. 

Indeed, and as rightly noted by Mylan, no side effect associated with tadalafil is mentioned in the patent, so that the dosage suggested for tadalafil curiously addresses a problem associated with another active compound. 

The court then referred to a standard mentioned in the so-called “finasteride” judgment of December 6, 2017 by the Cour de cassation, commented on this blog:

[…] When a claim relates to a [second] therapeutic application of a substance or composition, obtaining this therapeutic effect is a functional technical feature of the claim. Therefore, in order to meet the requirement of sufficiency of disclosure, it is not necessary to clinically demonstrate this technical effect; but the patent application must directly and unambiguously reflect the claimed therapeutic application, so that the skilled person can understand, based on commonly accepted models, that the results reflect this therapeutic application.

The court then came back to the technical problem presented in the patent:

Icos Corporation and Eli Lilly do not dispute that no prior art document describes any side effect related to the use of tadalafil.
And they cannot validly argue that the absence of documentation in this respect does not amount to the absence of a problem, because the onus is on them to show that there was a problem to be solved and that it is solved by the teaching of the patent.
It thus appears that the problem described in the patent relates to sildenafil and not tadalafil, and it cannot be extrapolated that both active compounds have the same side effects, unless one were to admit the resolution of artificial or speculative problems.
In fact, the examples cited in the patent demonstrate that the dosage mentioned in the patent does not address the listed “problems”. 

In summary, the problem to be solved cannot be considered as the reduction in the side effects of tadalafil, because such side effects were not known in the prior art – only side effects of sildenafil were known.

Most of the examples of the patent also do not demonstrate the existence of side effects of tadalafil associated with higher dosages, so that these were held not to “reflect” the alleged therapeutic application (using the wording of the Cour de cassation).

The conclusion reached by the three-judge panel will not doubt cause a stir, as the invention recited in claim 1 was found not to be sufficiently disclosed in the patent.

The finasteride case related to a second therapeutic application invention, for a known molecule. It is well accepted both at the EPO and in French national courts that the new therapeutic application has to be demonstrated in a plausible manner in the patent, otherwise the patent is insufficient.

Yet, in the present case, claim 1 is a classical product claim, with no functional feature. According to EPO case law, there should be no problem of insufficiency of disclosure, because the skilled person is able to manufacture the composition containing the active substance at stake in the claimed dosage range. The question of whether said claimed dosage range provides any technical benefit or not only pertains to the appraisal of inventive step.

Now, as regular readers of this blog are well aware, the French approach to validity is much more fluid than the EPO’s.

If a court is convinced that an invention does not properly solve the alleged technical problem, or that the technical problem is artificial, this can give rise to a number of invalidity objections, including insufficiency of disclosure. My understanding is that the technical problem tends to be viewed by French courts as an integral part of the claimed invention itself.

But there is yet another cause for controversy in the judgment.

I mentioned above that most of the examples of the patent do not demonstrate the existence of side effects of tadalafil associated with higher dosages. That said, there is one example, namely example 7, which does analyze in detail the occurrence of various side effects depending on the dosage of tadalafil. The table of results is in fact even reproduced in the judgment. The court first remarked that some side effects are not present at all at any dosage. So far so good. But, regarding those side effects which are indeed shown to be less frequent in the claimed dosage range than at a higher dosage, the court noted:

Regarding headache, back pain and myalgia […], the reasoning is the same because these effects were never previously observed.

This part of the judgment seems to imply that, at least in the context of drug dosage patents, the existence of the technical problem to be solved must be acknowledged in the prior art, and cannot be demonstrated for the first time in the patent itself.

The invention can thus not be a so-called “problem invention“.

Things should be put into perspective, though, and the present case may not necessarily be generalized. Maybe the court did not believe that example 7 was convincing at all. At the very least, the fact that the dosage originally claimed in the patent, namely from 1 to 20 mg, had to be later restricted to 1 to 5 mg, due to some relevant prior art, certainly contributed to the court’s perception of the patent being invalid.

In fact, the court reviewed all the following claims and concluded that they suffered from the same deficiencies as claim 1, mentioning a lack of inventive step in passing for some of them. Fluidity of the grounds for nullity indeed.

As a final note, this is probably one of the last judgments penned by Ms. Courboulay, who, given her seniority and her involvement in many conferences and events, was often considered as the leading judge in the 3rd (IP) chamber of the Paris TGI.

Ms. Courboulay has now officially retired; but given the large number of important rulings which she authored, there is little doubt that her influence will continue to be felt in the coming years.


CASE REFERENCE: Tribunal de grande instance de Paris, 3ème chambre 1ère section, April 5, 2018, Mylan v. Lilly France & Icos Corporation, RG No.16/05073.