EPLAW PATENT BLOG

NL – Sandoz v. Astrazeneca / Supreme Court / Opinion AG – Update

Posted: June 18th, 2019

Sandoz B.V. v. Astrazeneca AB, Supreme Court of the Netherlands, Opinion Advocate General Van Peursem, 10 May 2019, Case No. ECLI:NL:PHR:2019:608

Astrazeneca holds patent EP 1250138 relating to a fulvestrant formulation. A product with fulvestrant is marketed by Astrazeneca under the name FASLODEX and is used for the treatment of oestrogen hormone dependent breast cancers. In earlier preliminary injunction proceedings (which can be read here) the Judge in first instance and the Court of Appeal ruled that Astrazeneca’s patent is deemed to be valid. In cassation proceedings only the question regarding inventive step remains.

The Advocate General is of the opinion that Sandoz’ appeal in cassation should be dismissed.

Full summary:

The A-G opinion was rendered in the Supreme Court appeal against a decision of the Court of Appeal in summary proceedings in relation to the Dutch part of EP 1 250 138 (EP ‘138) held by AstraZeneca.

EP ‘138 claims the use of fulvestrant in the preparation of a pharmaceutical formulation for the treatment of breast cancer by intramuscular injection. Claim 1 of EP ‘138 claims the use of fulvestrant in the preparation of a pharmaceutical formulation for the treatment of benign or malignant disease of the breast or reproductive tract by intra-muscular administration, wherein the formulation comprises fulvestrant in a ricinoleate vehicle, a pharmaceutically acceptable non-aqueous ester solvent, and a pharmaceutically acceptable alcohol, and wherein the formulation is adapted for attaining a therapeutically significant blood plasma fulvestrant concentration for at least two weeks.

Sandoz acquired a market authorization for its fulvestrant formulation in 2015 and included its generic product in the July 2016 G-Standaard which was published on 21 June 2016. AstraZeneca filed summary proceedings before the preliminary relief judge of the court of The Hague amongst others requesting an injunction.

On 27 July 2016 the preliminary relief judge of the District Court of the Hague ruled that the Court in main proceedings would likely find the patent valid and infringed and granted a PI. In relation to inventive step the PI judge found that, starting from an article of Howell (which discloses a formulation of fulvestrant in castoroil) as the closest prior art and combining with a publication of McLeskey (which does not concern therapeutic use of fulvestrant for the treatment of breast cancer, but discloses the effects of estrogen on the growth of tumor cells in mice when its ovaries have been removed) the skilled person would not implement the subject matter of claim 1 of EP ‘138 without inventive step.

The decision of the PI judge was confirmed by the Court of Appeal of The Hague.
Like the preliminary relief judge in first instance, the Court of Appeal considered the objective technical problem to be solved by the skilled person: “to provide a formulation of fulvestrant suitable for treatment of a benign or malignant disease of the breast or the reproductive organs.” The Court of Appeal considered that “suitable for treatment of” (in short) breast cancer, requires that the positive results disclosed in Howell need to be maintained, meaning that the fulvestrant formulation needs to be tolerable and needs to achieve a sufficient plasma concentration during at least two weeks after intramuscular injection. Sandoz first ground for its appeal in cassation concerns this finding of the Court of Appeal.

The Court of Appeal further ruled that McLeskey does not contain any pointers which would instigate the skilled person to further investigate with a reasonable expectation of success that the formulation disclosed would be suitable for use in the treatment of breast cancer. According to the Court of Appeal such reasonable expectation of success is not already present when the skilled person would be able to determine the effects of a specific formulation by means of routine tests or when the skilled person is in a “try and see” situation. According to the Court of Appeal such try and see approach would conflict with the strict could-would approach. Sandoz second ground for its appeal in cassation concerns this finding of the Court of Appeal.

The opinion of advocate General Van Peursem is that Sandoz’s appeal is to be dismissed. The A-G first sets out the problem-solution approach as provided for in the Guidelines of the EPO. The problem-solution approach consists of: (i) determining the closest prior art; (ii) establishing the “objective technical problem” and (iii) considering whether or not the claimed invention, starting from the closest prior art and in view of solving the objective technical problem, would have been obvious to the skilled person. Sandoz’s Supreme Court appeal regards the application by the Court of Appeal of the second and third stage of the approach.

Regarding the formulation of the objective technical problem, the A-G points out that both AstraZeneca and Sandoz relied upon the objective technical problem as formulated by the preliminary relief judge in first instance that was confirmed by the Court of Appeal. According to the A-G Sandoz’s objection is therefore not directed to the formulation of the objective technical problem as such but rather on the finding of the Court of Appeal (in par. 4.20 of the judgement) that “suitable for” in the objective technical problem should be explained as meaning that the fulvestrant formulation needs to achieve a sufficient plasma concentration during at least two weeks after intramuscular injection and needs to be tolerable.

Sandoz argues that the Court of Appeal wrongly incorporated the positive results of Howell in the formulation of the objective technical problem. According to Sandoz this was not allowed as EP ‘138 does not disclose anything new in this regard compared to what is already disclosed in Howell.

According to the A-G, the interpretation of “suitable for” that was adopted by the Court of Appeal resulted from an evaluation of the facts, which cannot be assessed in cassation proceedings. In addition, the A-G finds that a patent does not have to disclose anything new with respect to a feature in order for it to be included in the objective technical problem. Without including the positive results of the formulation disclosed in Howell in the objective technical problem, the invention of EP ‘138 would only consist of providing an alternative formulation of fulvestrant.

This however would not reflect the actual invention of EP ‘138. The A-G mentions that one could also rule differently on this point in which regard he refers to an opposition decision in relation to a divisional of EP ‘138 in which the opposition division ruled that the objective technical problem should actually be to find an alternative fulvestrant formulation.

The A-G further points to the fact that in the meantime decisions have already been rendered between the parties in main proceedings in two instances (the Court of Appeal decision in main proceedings d.d. 27 November 2018 including headnote is published here).

In the main proceedings the positive results disclosed in Howell were also included in the formulation of the objective technical problem both by the first instance court and the Court of Appeal. In this regard the A-G finds that Sandoz does no longer have an interest in receiving a decision in cassation on the aspect of the formulation of the objective technical problem. If the Supreme Court would overturn the Court of Appeal’s PI decision and refer this case back to the Court of Appeal, the Court would have to hand down a judgment that is in line with the decision in the proceedings on the merits and would on this basis still have to include the positive results of the fulvestrant formulation disclosed in Howell in the objective technical problem.

With regard to the third stage of the problem-solution approach (the question whether the claimed invention would have been obvious to the skilled person) the A-G considers that Sandoz’ objection against the decision it is in fact an objection against the Court of Appeal’s evaluation of the facts of the case, which cannot be assessed by the Supreme Court.

Nevertheless, the A-G reviews the material contents of Sandoz’s objection. The A-G agrees with Sandoz that the Court of Appeal made a mistake if it indeed found that a try and see approach would contradict the EPO’s could/would approach. The A-G in this regard mentions that if a skilled person starting from the prior art and confronted with the objective technical problem would not arbitrarily carry out routine tests with the hope to come to a solution, but he would do this based on his common general knowledge. A “try and see” situation is therefore not contrary to the EPO’s could/would approach. The fact that the Court may have erred in this regard does not matter, as the Court of Appeal (implicitly) ruled that the skilled person in this case would not take a try and see approach and also that there was no reasonable chance of success.

The A-G concludes that the Court of Appeal has given a comprehensive reasoning for its decision that the skilled person would not come to the invention of EP ‘138 without inventive step. This reasoning would still stand, even if the Court of Appeal did not exactly and in detail follow the case law of the Boards of the EPO in relation to inventive step (with regard to “try and see”). In this regard the A-G mentions that the problem-solution approach is only a tool to help assess whether inventive step is present.

A copy of the Opinion (in Dutch) can be read here.

Summary: Mattie de Koning and Daisy Termeulen, Simmons & Simmons

Leave a Reply