A quick recap: Actavis had challenged validity of AZ’s Swiss part of EP 2 266 573 back in 2015, and the FPC had indeed held in 2017 that EP 573 was invalid for lack of inventive step; see this Blog here. However, the Supreme Court overruled that decision and remitted the case back to the FPC for re-assessment of obviousness; see this Blog here. In the second round, the FPC held that the subject-matter of EP 573 did involve an iventive step; see the decision O2018_009 of 27 May 2019. In brief, the FPC held that neither
Howell (referred to as D15) in further view of McLeskey (referred to as D13); nor
rendered the claimed subject-matter obvious. That decision was again appealed to the Supreme Court, and I have been eagerly awaiting a good read. But nothing like that. The Supreme Court only published a dismissal order. The parties have apparently reached an amicable settlement; court fees are shared.
AZ’s request for interim injunctive relief w/o hearing the defendant beforehand was dismissed with decision of 20 February 2019. Hearing the defendant didn’t change the outcome; the request was also dismissed with decision of 9 April 2019. Both requests failed for lack of urgency, given the fact that Sandoz’s generic has been on the market since 2.5 years.
At the face of it, this appears to be pretty straight forward, in particular in view of the most recent decisions S2018_006 (¶13) and S2019_001 (¶6) emphasizing the 14 months time bar in no uncertain terms.
But the devil is in the detail. Maybe.
This is not the first time that the FPC had to deal with EP’573. The patent had been revoked for lack of inventive step over Howell in view of McLeskey; see this Blog here. But AZ appealed and essentially argued that the FPC erroneously assumed a (concrete) pharmaceutical formulation in Howell, thus incorrectly defined the distinguishing features and the (objective) technical problem, and then wrongly concluded for lack of inventive step.
And, indeed, the Supreme Court agreed and remitted the case for re-assessment of inventive step; see this Blog here. From the reasons of the present decisions, we can now catch a glimpse of what is currently going on in the remitted proceedings (formerly O2015_011, now O2018_009): The judge-rapporteur held on 23 November 2018 that the patent was valid. But it is not only that the wind has changed at the FPC. The Gerechtshof Den Haag also held that the patent was valid, in second instance proceedings. Likewise, a Board of Appeal at the EPO overruled the first instance revocation of an opposition division and finally ruled on 24 January 2019 that the patent was valid.
On the other hand, this is also not the first time that AZ sought injunctive relief for Sandoz’s generic. AZ had asserted a different patent against the same generic already in summary proceedings S2016_007, i.e. EP'138. It becomes clear from the present decisions that these earlier proceedings had been terminated because AZ had withdrawn the requests.
Now, here is a timeline events which is colored to reflect my very personal view on the prima facie validity of EP’573 over time:
One may conclude from the above timeline that there had been some red(-ish) flags concerning the validity of EP’573 for quite a long time.
The decisions note in passing that main infringement proceedings with case no. O2017_014 concerning Sandoz’s generic are pending in parallel, but based on a different patent (EP'195); from the case no. it is clear that this suit has been brought in 2017, but no further information is available to date.
Further, the decision indicates that in some proceedings with case no. O2018_010 the very same EP’573 is at stake as in the present matter, but the defendant is not revealed.
AZ essentially argued that it had been prevented from bringing the request for interim injunctive relief earlier because of the previously negative assessment of validity EP’573 by the EPO, the FPC (O2015_011) and the district court of The Hague (NL), and that the wind has changed only very recently; see timeline above.
The single judge did not buy into that and emphasized that AZ had undoubtedly been aware of Sandoz’s generic since 26 July 2016; AZ could have lodged main proceedings at any time since then. The decision further holds (¶22; inofficial translation):
[T]he decision of the Board of Appeal of the EPO of 24 January 2019 […] may substantiate the claim to which the plaintiff is entitled with regard to the validity of the patent in suit, but in no way justifies urgency. […]
Ultimately, however, all parallel proceedings have no influence on the purely procedural question of urgency in the present summary proceedings. […]
The tactical awaiting of a foreign parallel decision on the patent in suit before an action is lodged does not belong to the subjective but rather to the subjective circumstances.
It surely is a tough exercise to fit the gist of a decision in a single tweet; but the FPC’s tweet on the decision is straight to the point:
A decision of a Board of Appeal of the EPO does not give rise to urgency where the contested product has been on the market for 30 months.
Urteil v. 9. April 2019 i.S. AstraZeneca AB g. Sandoz Pharma AG re vorsorgl. Massnahme (#Fulvestrant): ein Entscheide einer Beschwerdekammer des EPA begründet keine Dringlichkeit, wenn das angegriffene Produkt schon seit 30 Monaten auf dem Markt ist, https://t.co/iinRKiYDWk
The approach taken in the present decisions in my understanding focuses much more on the total time than on the apparent change of prima facie validity of the patent in suit over time. While the 14 months time bar is comparably generous (German courts typically deny urgency after 1-2 months, in my experience), it seems to be a pretty rigid time bar nowadays.
In an earlier case at the FPC, urgency had still been acknowledged for a request that had been filed five months after a BoA of the EPO had maintained the patent in suit which had been revoked in first instance by an opposition division (S2013_004, decision of 12 May 2014, ¶4.8).
It will be interesting to see how the FPC’s approach in the assessment of urgency will develop.
In (very) brief, AstraZeneca’s EP(CH) 2 266 573 B1 had been revoked for lack of inventive step over Howell in view of McLeskey. The parties heavily dissented on whether or not Howell contained an enabling disclosure or not. The FPC held that it did, and that the objective technical problem to be solved in accordance with the problem-and-solution approach was only to provide an alternative formulation for sustained release.
AZ appealed and essentially argued that the FPC erroneously assumed a (concrete) pharmaceutical formulation in Howell, thus incorrectly defined the distinguishing features and the (objective) technical problem, and then wrongly concluded for lack of inventive step.
The Supreme Court indeed agreed and remitted the case for re-assessment of inventive step on the basis of an ‘objectively correct’ definition of the problem to be solved; see ¶2.3.5. Further, the Supreme Court notes in ¶2.3.3 that a correct definition is:
Provision of a castor oil-based formulation (i.e. a formulation consisting of castor oil and known suitable solvents and adjuvants in a certain composition and amount) for the administration of up to 250 mg fulvestrant for the treatment of breast cancer, which is well tolerated, has a uniform release profile and achieves the therapeutically decisive concentration of fulvestrant in the blood plasma over a longer period of time.
It is rare (to say the least) that the Supreme Court re-defines the objective technical problem to be solved in the assessment of inventive step. It did so here because the FPC got the meaning of an enabling disclosure wrong:
[…] weil [die Vorinstanz] von einem unzutreffenden Begriff der Ausführbarkeit der Lehre ausgegangen ist […]
Really? The FPC got such a fundamental point of law just wrong!?
This demands for a closer look!
The Supreme Court holds that the FPC failed to recognize the concept of feasibility or sufficient disclosure with the conclusion that ‘the technical teaching of [Howell] could basically be reworked.’ For a technical teaching consists not only of the problem but also of the solution.
Die Vorinstanz hat den Begriff der Ausführbarkeit oder der hinreichenden Offenbarung verkannt mit dem Schluss, dass ‘die Lehre [in Howell] grundsätzlich nacharbeitbar war’. Denn eine technische Lehre besteht nicht nur aus dem Problem, sondern auch aus der Lösung.
However, Howell only suggests that there was a compatible and pharmaceutically effective depot formulation available — but the actual composition of this formulation is not disclosed.
Further, the Supreme Court holds that the FPC did not expand on which concrete formulation the skilled person(s) would have found on the basis of the information in Howell without undue burden and without involvement of an inventive step. The general knowledge that steroids such as fulvestrant can be dissolved in castor oil with certain excipients and solvents in such a way that compatible injections can be produced is not sufficient for the feasibility of a technical teaching, in the Supreme Courts’ view.
In sum, the Supreme Court thus holds that the FPC applied an incorrect legal concept of feasibility with the assumption that no concrete technical formulation was required in order to affirm that there well is a technical teaching in Howell; ¶2.2.4.
Die Beschwerdeführerin rügt im Ergebnis zu Recht, dass die Vorinstanz von einem unzutreffenden Rechtsbegriff der Ausführbarkeit ausgegangen ist mit der Annahme, es bedürfe keiner konkreten technischen Formulierung, um die Offenbarung der technischen Lehre — d.h. hier der in [Howell] beschriebenen Depot-Formulierung — bejahen zu können. […] Die Vorinstanz hat die Offenbarung einer technischen Lehre durch [Howell] zu Unrecht bejaht.
Further down the road, the Supreme Court concluded that the FPC failed in the identification of the differentiating features. It just could not do it correctly because there is no specific disclosure in Howell.
But still, Howell undoubtedly is pre-published. Now, what to do with it?
The FPC had left it undecided whether or not a non-enabling disclosure is ‘prior art’ under Art. 54(2) EPC. But the Supreme Court took over — and answered even more than that. The Supreme Court holds that Howell belongs to the state of the art (contrary to what the EPO typically does, i.e. to just ignore it in toto; see Guidelines, G-IV, 2). But, in the Supreme Court’s view, such a piece of prior art cannot be assessed with the problem-and-solution approach because the objective technical problem is then just to find a working solution to what is insufficiently disclosed therein, and there is an inherent motivation to search for that solution; ¶2.3.2.
Der Beschwerdegegnerin ist zwar zuzustimmen, dass die Information von [Howell], in der das Problem formuliert und Ansätze für die Lösung mitgeteilt werden, zum Stand der Technik gehört. Wird jedoch ein Dokument als ‘nächstliegender Stand der Technik’ beigezogen, das keine technische Lösung offenbart, […] wird der ‘Aufgabe-Lösungs-Ansatz’ verlassen. Denn die objektive technische Aufgabe ergibt sich unmittelbar aus [Howell], wenn darin dem fachkundigen Adressaten mitgeteilt wird, dass es eine Lösung für das Problem gibt […]. […] Eines Vergleichs von Merkmalen zur Ermittlung der objektiven technischen Aufgabe bedarf es in diesem Fall nicht. […] Aufgrund der Information in Howell, dass es eine Formulierung [gibt], wird das Fachteam jedenfalls dazu motiviert, geeignete Depotformulierungen zu suchen.
Frankly, it is now me who is not (yet) enabled to fully grasp the implications of this decision. Time will tell, as always.
As could be expected after the main hearing, assessement of inventive step in view of McLeskey and Howell was key. The panel of judges essentially followed the expert opinion of the judge-rapporteur and denied patentability for lack of an inventive step in both cases. Besides the (always) very case-specific assessment of inventive step, I feel that some other aspects of these decisions are of more general interest.
Partial priorities / ‘self-collision’
Plaintiff argued that the divisional application EP 1 669 073 A2 is novelty-destroying for its parent, EP’138. With reference to the decision G 1/15 of the Enlarged Board of Appeal (EPO), the FPC did not buy the self-collision argument.
In principle, the claims at stake could well be split into parts that did / did not enjoy priority (‘generic or’-claims) – even though that might well be a challenging task. However, since there was no intervening prior art on file, there was no need to go through that academic exercise. Partial priority was acknowledged, ruling out any self-collision issue.
Es scheint aber insbesondere im Hinblick auf die Entscheidung G 1/15 angebracht, im vorliegenden Fall Teilprioritäten grundsätzlich anzuerkennen und zwar selbst dann, wenn keine explizite Aufteilung der Patentansprüche in die Prioritäten gültig beanspruchende bzw. nicht gültig beanspruchende Teilbereiche vorliegt.
The ‘self-collision genie’ is pushed back in the bottle.
Prior public use / clinical studies
It was beyond dispute that Faslodex® had been in phase-III clinical study in 1998, well before the earliest priority date. The decisions refer to the study as 9338IL/0020, but this appears to be a typo. I trust it should read 9238IL/0020; see EMA ‘Assessment Report for Faslodex’ of 2010, at the bottom of p 29.
Plaintiff argued that clinical studies would generally make the pharmaceuticals publicly available, in the absence of any proof to the contrary. It was referred to T 0007/07 in this respect. Further, it was argued that the patients must have been informed about the exact formulation; and even if this had not been the case, the formulation could have been identified / analysed with ease.
The FPC did not agree. In the case at hand, two ‘Clinical Trial Agreements’ from this study contained confidentiality clauses, and the ‘Written Informed Consent Form’ apparently did not reveal details as to the exact formulation. Finally, the patients got a monthly injection into the gluteal muscle. This is unlikely being done by the patient herself, and thus it is unlikely that the patient ever had access to the pharmaceutical for further investigation of the formulation.
Under the given circumstances, the clinical study was thus held to not constitute a public prior use. Novelty was acknowledged.
Insufficient disclosure to be considered as prior art?
Howell discloses a dosage of 250 mg fulvestrant and mentions an injection volume of 5 ml. Plaintiff argued the this results in a solution with a concentration of 50 mg/ml. Defendant disagreed: The solubility of fulvestrant would be too low to obtain a solution in castor oil. Rather, a dispersion would be obtained. Further additives would be necessary to obtain a solution with such a high concentration of fulvestrant. But there is no indication of any such additives in Howell. Defendant thus argued that Howell must not be considered as prior art, for lack of an enabling disclosure.
Subject-matter can only be regarded as having been made available to the public, and therefore as comprised in the state of the art pursuant to Art. 54(1), if the information given to the skilled person is sufficient to enable him, at the relevant date (see G‑VI, 3), to practise the technical teaching which is the subject of the disclosure, taking into account also the general knowledge at that time in the field to be expected of him (see T 26/85, T 206/83 and T 491/99).
Where a prior art document discloses subject-matter which is relevant to the novelty and/or inventive step of the claimed invention, the disclosure of that document must be such that the skilled person can reproduce that subject-matter using common general knowledge (see G‑VII, 3.1).
This appears to be a sensible approach at first sight. But is it really in line with the definition of ‘the state of the art’ codified in Art. 54(2) EPC?
The state of the art shall be held to comprise everything made available to the public by means of a written or oral description, by use, or in any other way, before the date of filing of the European patent application.
Frankly, I am not conviced. What is published, is just out there as-is. What the skilled person actually can do based on a specific publication at a given point in time, that’s a different story to me. Anyway, the FPC did not need to decide on this issue; but I catch the notion between the lines that it is not carved in stone that the FPC would follow the EPO’s approach.
Lässt man die Praxis des EPA unberücksichtigt, ist die D15 vollumfänglich dem Stand der Technik zuzurechnen. […] Ob der erwähnten Praxis des EPA zu folgen ist, kann deshalb hier offen bleiben.
In the case at hand, the FPC held that Howell did contain an enabling disclosure for the skilled person, and it was considered as prior art.
Belated limitation not admitted into the proceedings
Defendant limited one of his auxiliary requests in O2015_012 with an additional feature (marked-up below) as follows:
Use of fulvestrant in the preparation of a pharmaceutical formulation for the treatment of breast cancer by intra-muscular administration to human patients, […] wherein […] the formulation is adapted for attaining a therapeutically significant blood plasma fulvestrant concentration of at least 8.5 ng ml-1for at least two weeks.
Defendant argued that this further limitation was a partial acknowledgment of the complaint which is possible at any stage of the proceedings; Art. 241 CPC. The FPC did not agree. The further limitation only concerned an auxiliary request, i.e. it only came into play when the court had already decided on the main requests of the defendant: Denial of the complaint, without any amendments to the claims. At that stage, there is no room anymore for a partical acknowledgment of the complaint. The FPC held that the defendant in fact present the court with new facts to be considered as an auxiliary measure. This is possible in general, but only within the time limits for submission of novae; Art. 229 CPC. Consequently, the newly drafted auxiliary request was not admitted into the proceedings.
UPDATE October 17, 2017:
The patentee has lodged an appeal to the Supreme Court in both cases.
UPDATE May 30, 2018:
AZ’s appeal re O2015_012 has apparently been dismissed, and the appeal re O2015_011 partially approved. No more information available, for the time being.
UPDATE June 5, 2018:
The Supreme Court has published decisions 4A_541/2017 (appeal against O2015_011) and 4A_543/2017 (appeal against O2015_012) on its website earlier today. Further, 4A_541/2017 is scheduled for publication in the official collection.