Still warm from the press and courtesy of Matthieu Dhenne, come tidings of the fall of another important pharma IP, namely the Atripla SPC (Supplementary Protection Certificate).
Atripla is marketed as a pink tablet with “123” impressed on one side. It contains a combination of three anti-HIV drugs, namely efavirenz, emtricitabine and tenofovir.
The U.S. pharmaceutical giant Merck Sharp & Dohme Corp. (MSD) owns European patent No. EP 0582455, entitled “benzoxazinones as inhibitors of HIV reverse transcriptase“. The patent was filed on August 3, 1993.
Two SCPs were filed and granted in France based on the EP’455 patent, and on two successive marketing authorizations (MAs):
- The first one, FR01C0012, was filed on April 10, 2001 and granted on May 18, 2001. It protected the active efavirenz per se. This SPC expired on November 20, 2013.
- The second one, FR08C0021, was filed on May 27, 2008 and granted on November 20, 2009. It protects the triple therapy combination of efavirenz, emtricitabine and tenofovir (marketed as Atripla) and is set to expire on August 2, 2018.
On September 20, 2016, Mylan initiated legal proceedings against MSD in France, claiming that the FR’021 SPC is invalid. The Paris Tribunal de grande instance (TGI) issued its judgment on November 30, 2017.
The judgment is interesting both regarding the admissibility of the action and the merits.
As far as admissibility is concerned, the nullity defendant claimed that Mylan was time-barred from requesting the nullity of the SPC.
As a first line of response, Mylan argued that the general statute of limitations in our Code civil, which provides a five-year limitation period for “personal or movable actions“, is not applicable to actions for nullity of an IP right. Unsurprisingly, the court disagreed, in keeping with recent case law at the first instance and appeal levels. The TGI made in particular reference to a trademark ruling by the Cour de cassation dated June 8, 2017. For the court, applying this ruling by analogy leads to the conclusion that an action for revocation of an SPC is indeed subject to the limitation period under ordinary law.
That being said, the real interesting point is the determination of the starting point for the five-year limitation period. Although there has been a lot of discussion (including on this blog) concerning the starting period for patent nullity cases, there has been no clear guidance for SPC nullity actions, as far as I am aware of.
MSD’s case was that the starting point for the limitation period was the publication of the SPC application. The court disagreed and set the following principles.
The starting point for the limitation period must be set to the day, determined in concreto, when Mylan knew or should have known, because it intended to market a generic version of the drug which received an MA on December 13 , for the combination of [the] three actives, which is protected by the SPC, which represents an impediment for its business.
So, we all get the idea there – although a couple of words may be missing, which happens from time to time when your sentences are too long, and this is probably why my blog software keeps blaming me for using more than the recommended threshold of 25% of sentences containing more than 20 words.
The general principle of an in concreto assessment is in keeping with the TGI’s previous decisions in patent revocation cases. The court went on:
[…] Only the SPC matters as an impediment, and not the patent.
One should not refer to the date of grant of the patent, since the validity of the patent is not challenged by Mylan, which acknowledges that the efavirenz active compound is the subject-matter of the invention protected by the EP’455 patent and then by the [FR’012] SPC which expired on November 20, 2013.
Only the validity of the [FR’021] SPC […] is challenged […].
Thus the publication of the grant of the patent cannot be set as the starting point for the limitation period, as it would in fact require an unrealistic watch from stakeholders and is unrelated to the development of the project which gives standing to sue.
Mylan’s standing does not derive from the publication of the title, be it the patent or the SPC, but from its concrete intent to market the same drug.
In this case, they have to check that this intent to market the product does not infringe any IP, and if this is the case, to seek its revocation before launching.
Watching patent or SPC registers cannot be required from stakeholders before they intend to develop a competing product.
[…] In the present case, the first MA for Atripla […] was granted on December 13, 2007. In view of article R. 5121-28 of the Code de la santé publique, the generic company can only apply for an MA starting from the eighth year after the grant of the MA for the original drug, and cannot be granted one before ten years.
Therefore, Mylan could not apply for an MA before December 13, 2015, and could not obtain it before December 13, 2017. As a consequence, the date at which Mylan’s standing can be taken into account is December 13, 2015, which is the date starting from which it could apply for an MA. Thus, Mylan is not time-barred as it had until December 13, 2020 to start legal action.
What is somewhat paradoxical is that the TGI calls for an in concreto appraisal but then defines what could possibly be a general rule for SPC cases, namely that the starting point for the limitation period is the date at which third parties may start applying for their own MAs.
We will need to wait for further cases to know for sure whether this is indeed a general rule or not.
Turning to the merits of the case, the discussion and the ultimate reasoning of the court are extremely similar to what can be found in the recent decision on Truvada, also reported on this blog a few weeks ago.
Truvada is another anti-HIV drug based on the combination of tenofovir and emtricitabine. The SPC at stake in today’s decision relates to the combination of the same compounds, plus a third one, efavirenz. And the problems raised by this other combination are analogous.
According to article 3(a) of the SPC regulation (regulation (EC) No. 469/2009 of the European Parliament and of the Council), an SPC “shall be granted if, […] (a) the product is protected by a basic patent in force“.
How to determine whether a product can be considered as being “protected” by a basic patent has been the subject of intense litigation and numerous rulings from the CJEU, which are mentioned in the TGI’s judgment. Again, readers of this blog can refer to the Truvada post, which contains a short summary of the most important CJEU case law prepared by Lionel Vial.
In the present case, none of the claims of EP’455 explicitly recites the combination of the three active compounds of the combination. Instead:
- efavirenz is covered by a generic formula in claims 1 and 5 and is singled out in claims 2 and 12 (as well as in claims 8 and 9 but in combination with other drugs different from tenofovir and emtricitabine);
- tenofovir and emtricitabine are not cited in the patent;
- claims 7 and 16 relate to the combination of a generic formula (covering efavirenz), or of efavirenz specifically, together with a nucleoside analog;
- tenofovir and emtricitabine belong to this category of nucleoside analogs.
According to the court, this is insufficient to consider that the combination of the three active compounds is protected by the EP’455 patent pursuant to article 3(a).
Says the court:
It turns out that the description never explicitly cites either tenofovir or emtricitabine which are not identified in the EP’455 patent, be it individually or collectively in a composition. And in addition the specific combination claimed as an active product “efavirenz + emtricitabine + tenofovir” is not implicitly but necessarily and specifically taught in the description, and no indication makes it possible for the skilled person to select emtricitabine and tenofovir as nucleoside analogs.
In fact, if I understand correctly, emtricitabine and tenofovir were not even identified and known yet as anti-HIV drugs at the filing date of the EP’455 patent.
Furthermore, the court refused to consider the claims relied upon by MSD (reciting nucleoside analogs) as “functional claims” because “they do not describe the structure which should be present nor the function that the second and third products should have in this structure“.
For the sake of completeness, the court stated that even if the claims were considered as functional, the four-step test established by the Dutch patent office would then not be satisfied. Again, this same test was discussed in the previous Truvada post, so I will not describe it again here.
As a consequence, the SPC was found to be invalid under article 3(a).
By way of overkilling, the court added that the SPC was also invalid in view of article 3(c) of the SPC regulation, per which an SPC “shall be granted if, […] (c) the product has not already been the subject of a certificate“.
In this case, another SPC had been granted based on the same EP’455 patent, namely the efavirenz SPC. MDS relied on the Georgetown CJEU decision (C 484/12). According to this decision, article 3(c) does not preclude the grant of one SPC for a combination of active ingredients, and another SPC for a single active ingredient, based on the same patent.
Nevertheless, according to the TGI, Georgetown is only applicable if the mono and combo products are separate inventions.
In one brief paragraph, the court then held that:
the combination of efavirenz with emtricitabine and tenofovir does not represent a separate invention which may give the right to a second SPC. For this second reason, SPC [FR’021] is invalid under article 3(c) of the regulation.
Those readers in favor of pan-European consistency (which probably means most readers of this blog) will be happy to know that the TGI’s decision mirrors a similar ruling in the UK handed down on March 21, 2017, per which the corresponding UK SPC was declared invalid by Mr. Justice Arnold.
CASE REFERENCE: Tribunal de grande instance de Paris, 3ème chambre 1ère section, November 30, 2017, Mylan SAS v. Merck Sharp Dohme Corp., RG No. 16/14466.