1. Danisco/Novozymes (EP 1 804 592)

DK – Denmark

Danish High Court of 25 May 2012 – Danisco A/S (today a part of DuPont) and Novozymes A/S

Keyword: novelty – inventive step

The patent in question (EP 1 804 592) claimed a method for the manufacturing of an animal feed composition in the form of steam-treated pelletized feed. Claim 1 related to such a method comprising the steps of mixing feed components with granules comprising a core and a coating, wherein the core comprises an enzyme, and steam treating and pelleting said composition, with the characterising feature that "the coating comprises a salt which has a constant humidity at 20°C which is above 60%". Novozymes alleged infringement, while Danisco maintained that the disputed patent was invalid for lack of novelty. On 7 July 2011, the EPO's opposition division declared the patent void for lack of novelty in view of a particular prior art document called "Jacobsen".

The High Court found that the salts calcium carbonate and magnesium silicate might be included in the coating of the enzyme-containing granules in prior art document "Jacobsen", and that the parties agreed that the claims in the disputed patent would therefore lack novelty if calcium carbonate and/or magnesium silicate were to be found to have a constant humidity at 20°C that is above 60%. The court found it probable that calcium carbonate had a measurable constant humidity at 20°C that was higher than 60% and lower than 100%.

Accordingly, and since it did not appear from the definition of constant humidity in the disputed patent nor in the patent description with the clarity necessary that the salts that were encompassed by claim 1 of the disputed patent had to have a certain – not further specified – solubility and hyproscopicity, the High Court found that a skilled person would find it obvious that at least calcium carbonate was useful as a coating salt according to the disputed patent. On this basis, the High Court found that the disputed patent in the form in which it was issued lacked novelty in view of "Jacobsen".

Furthermore, the opposition division found that the auxiliary requests put forward by Novozymes lacked inventive step in the light of the prior art. The parties obtained testimonies from their respective experts, but these differed in opinion on inventive step. The High Court thus did not find that there was a sound basis for assuming that the opposition division's decision, according to which the subsidiary claim sets lack inventive step, was incorrect. It held that the subsidiary claim sets lacked inventive step and that Novozymes did not establish that there was a right that could be infringed.

Editor's note: The board of appeal confirmed that the patent was invalid with its decision T 1839/11 of 31 July 2012, just two months after the Danish High Court's decision.

NL – Netherlands

Rechtbank 'S-Gravenhage of 22 June 2011 – Danisco v Novozymes

Keyword: novelty – inventive step

The patent in question (EP 1 804 592) claimed a method for the manufacturing of an animal feed composition in the form of steam-treated pelletized feed. Claim 1 related to such a method comprising the steps of mixing feed components with granules comprising a core and a coating, wherein the core comprises an enzyme, and steam treating and pelleting said composition, with the characterising feature that "the coating comprises a salt which has a constant humidity at 20°C which is above 60%". Danisco applied to have the patent revoked, inter alia due to an alleged lack of novelty and inventive step. Novozymes counter-claimed for alleged infringement.

In the attack on novelty, a particular prior art document called "Jacobsen" was cited. Jacobsen explicitly described a method such as the one claimed by the patent in question, except for the characterising feature. It was argued before the court, however, that this feature was also disclosed because Jacobsen claimed as a coating the salts magnesium silicate or calcium carbonate that had a constant humidity above 60% at 20°C.

The court acknowledged that the calcium carbonate referred to in Jacobsen, which reached almost 100% of constant humidity at 20°C, had a constant humidity above the 60% as claimed by the patent in question. However, the court stated that this did not mean that the characteristic measure of the patent in question was also directly and unambiguously disclosed in Jacobsen. The court held that such reasoning would ignore how the skilled person, using his common general knowledge, would understand the characterising feature of claim 1 of the patent. With regard to novelty, the claim of a patent should be read in the context of the description and the drawings.

The court concluded that Jacobsen did not disclose a salt according to the claim. The skilled person would understand claim 1 of the patent, which required the determination of a salt's "constant humidity", to refer to a soluble salt with some degree of hygroscopicity because "constant humidity" was a term of art that only had relevance for such salts. The salt used in Jacobsen was, however, insoluble and thus could not be used to determine the constant humidity.

Regarding inventive step, the court applied the problem-solution approach. It determined the technical problem to be objectively solved as providing improved enzyme activity/stability in steam pelleting and found the patent to be inventive. The prior art showed the application of a salt coating in one prior art document and the use of a method described in another. However, the court did not see how the one document pointed to another and considered the reasoning which combined the two as based on hindsight. It stated that it could not be assumed that the average skilled person, starting from the closest prior art, would have arrived at the solution of a salt coating as in the patent in dispute. In conclusion, the court held Novozymes' patent valid and infringed by Danisco.

Editor's note: On 5 August 2011 the opposition division of the EPO revoked the patent in its entirety. It was followed by board of appeal decision T 1839/11 of 31 July 2012, which dismissed Novozymes' appeal and made the revocation of the patent final.

2. Quetiapine (EP 0 907 364)

BE – Belgium

Antwerp District Court judgment of 20 December 2013 – Sandoz v AstraZeneca

Keyword: sustained release formulation – decisions from other jurisdictions – inventive step

The court noted that a number of revocation actions had been brought in Europe against various national parts of the European patent in question. Although each national decision was equally important and it could be presumed that the legal issues raised were competently examined in all these countries, it was notable that even those countries considered to be among the leaders in the patent world reached different outcomes (see the Netherlands, UK and German judgments), while the countries which reached the same outcome gave different reasons.

It was impossible to ascertain whether the same arguments were put forward before the various national courts. The defendant's approach in these proceedings was very different from that taken in England. The present court did not have access to all submissions made and documents produced before the foreign courts and, even if it did it would not be for it to comment on those courts' assessment of them. It thus made its own assessment independently of the parallel proceedings.

The objective technical problem was defined to be to provide a quetiapine formulation resulting in a uniform and stable level of plasma concentration (i.e. a uniform and constant release ratio) over a longer period without having to increase the dosage frequency. This differed from the objective technical problem as put forward by the defendant, who argued that this was the development of a new medicine designed to release the active ingredient quetiapine in a particular way for treating psychotic disorders. This was rejected by the court. The court applied the problem-solution-approach on the basis of its above definition of the objective technical problem and concluded that the indications in the prior art were such that the skilled person, when faced with the objective technical problem, would opt to develop a sustained-release formulation using a gelling agent, such as the one in the patent.

Based on the above, the court held that the Belgian part of European patent 0 907 364 had to be revoked for lack of an inventive step.

DE – Germany

Federal Patent Court of 13 November 2012, joined cases 3 Ni 43/10 (EP), 3 Ni 24/11 (EP) and 3 Ni 25/11 (EP) – Quetiapine

Keyword: inventive step – sustained-release formulation

European patent (DE) 0 907 364 was granted for a sustained-release formulation of quetiapine marketed as Seroquel XR. Quetiapine has dopamine-blocking effects, and is used to treat psychosis or hyperactivity. The claimants argued that the patent's claims were not novel, and its teaching not inventive. Revoking the patent for Germany, the Court ruled that its subject-matter – as granted, and in the defendant's fallback versions – did indeed lack inventive step.

The Court found that the skilled person wishing to solve the problem addressed by the patent could draw inter alia on a published study showing that once-daily administration of an immediate-release quetiapine composition was unlikely to be effective, but reducing the dosage frequency was indeed desirable for achieving better compliance amongst schizophrenia patients. However, skilled persons wishing to achieve a once-daily dose of quetiapine would be prompted to consider a release profile other than the known, immediate-release one by a press release announcing plans to develop a "once-daily dosage form". They would associate that term with formulations releasing the active ingredient over an extended period, i.e. dosage forms known in the field as "sustained-release formulations". It was therefore obvious to consider such a formulation in order to improve patient compliance and regulate plasma levels.

Nor could the feature claiming the use of a gelling agent to formulate the quetiapine composition give rise to inventive step. At the time in question, it was known from the art relating to sustained-release drug administration that matrix systems based on gelling agents such as hydroxypropyl methylcellulose were suitable for formulations involving numerous active ingredients whose structure – and therefore also solubility – vary widely. Nowhere did the patent indicate that formulating quetiapine appropriately required special measures extending beyond the average skill in the art. It therefore lacked inventive step.

Editor's note: For appeal decision see below.

DE – Germany

Federal Court of Justice of 13.01.2015 (X ZR 41/13) – Quetiapine

Keyword: inventive step – sustained-release formulation

On 13 January 2015 (X ZR 41/13), the Federal Court of Justice rejected the appeal against the decision of the Federal Patent Court and confirmed that the patent was obvious in view of the state of the art. It did note, however, that the problem as defined by the Federal Patent Court was too narrow. The problem underlying the patent in question was not the formulation of the quetiapine composite with the most constant release profile possible over the longest period of time possible, but rather to make available a formulation of the quetiapine composite with improved effectiveness. In its headnote, the Federal Court of Justice stated: "In defining the technical problem underlying an invention, it is not permissible to assume that it would have been appropriate for the skilled person to deal with a particular problem. Rather, the technical problem should be defined so generally and neutrally that the question of what pointers the prior art contained in that respect will be confined to the question of inventive activity."

GB – United Kingdom

Patents Court of 22 March 2012 – Teva UK Ltd v AstraZeneca AB [2012] EWHC 655 (Pat)

Keyword: inventive step – parallel proceedings

AstraZeneca's patent was for a sustained release formulation of an anti-psychotic drug known as quetiapine. Teva contended that it was invalid on the ground of obviousness over an item of prior art referred to as Gefvert.

The judge applied the familiar structured approach to the assessment of allegations of obviousness set out in Windsurfing International Inc v Tabur Marine (Great Britain) Ltd [1985] RPC 59 and re-stated in Pozzoli v BDMO SA [2007] EWCA Civ 588 (see also second edition "Case Law from the Contracting States to the EPC" 2004 – 2011, OJ SE 3/2011,p 124). It was common ground that the difference between Gefvert and claim 1 was that Gefvert did not disclose a sustained release formulation of any kind.

Considering the further evidence and arguments from each side, the judge found that the skilled team would conclude from Gefvert that a single 450 mg dose of an immediate release formulation daily would not be efficacious. The skilled team would regard one daily administration as desirable. To achieve once daily administration, a sustained release formulation and a higher dose of an immediate release formulation would both be obvious possibilities. Furthermore, the skilled team would not be deterred from developing a sustained release formulation. Nor would its expectation of success be adversely affected. The judge therefore found that the skilled team would expect to be able successfully to formulate a sustained release formulation of quetiapine using HPMC. The solution was an obvious one.

The District Court of The Hague (in Case 397921/HA ZA 11-1977 of 12 March 2012) had given judgment on a parallel claim by Sandoz to revoke the Dutch counterpart of the patent and had rejected Sandoz's argument that the patent was obvious over Gefvert. In summary, the Dutch court considered that quetiapine had not yet been proven to be sufficiently effective without having serious side effects and the average person skilled in the art would have a very limited motivation to choose that particular substance to create a sustained release formulation. Nor would the skilled person have high expectations of successfully developing a sufficiently effective sustained release formulation of quetiapine.

The judge regretted that different European courts considering the validity of the same patent should reach opposite conclusions. The Dutch judgment merited respect and had been carefully considered. Different conclusions were reached for these reasons:

1. The evidence before the respective courts was different. The hearing before the Dutch court lasted only one day; there was no cross-examination of expert witnesses.

2. Although a lot of the same articles and textbooks appeared to have been placed before each court, there were certain differences.

3. The Dutch court proceeded on the basis that the patent was not entitled to priority and so relied upon a number of articles published after the priority date. There was no challenge to priority before the UK court.

4. The arguments before each court were different.

5. In the light of the evidence and arguments before him, the UK judge was unable to agree with the Dutch court's conclusions with regard to either motivation or expectation of success.

6. The Dutch court appeared to have attached no weight to the fact that the problems described in the patent were illusory and that the "lions in the path" relied upon by AstraZeneca were not mentioned in the patent.

Editor's note: The decision of the Patents Court was upheld on appeal (see next summary).

GB – United Kingdom

Court of Appeal of 30 April 2013 – AstraZeneca AB v Hexal [2013] EWCA Civ 454

Keyword: inventive step – role of the Court of Appeal – parallel national decisions

This was an appeal from the judgment of Arnold J in Teva v AstraZeneca [2012] EWHC 655 (Pat) (see above). He revoked the patent, which was for sustained release formulations of an anti-psychotic drug called quetiapine, on the ground that, in the light of the prior art and the common general knowledge, the relevant claim was obvious.

Ongoing litigation in the courts of other countries about the validity of patents equivalent to the patent at issue had been a factor affecting the trial judge's decision to grant permission to appeal. After the hearing of the appeal the parties notified the court of further developments in other jurisdictions.

Mummery LJ stated that, fortunately, the Court of Appeal did not have to decide whether the judgments of other courts in different jurisdictions were right or wrong; all that it had jurisdiction to decide was whether the order made by Arnold J was, on the evidence and arguments before him, wrong. Its role was not to retry the issue of obviousness. Its function was not one of allowing the whole issue to be re-argued from scratch with a view to its making a fresh evaluation of the evidence on what was a multi-factorial issue and in the expectation that a contrary determination of the merits of the issue would be substituted for the decision of the trial judge. The issue of obviousness was quintessentially a matter of fact, degree and overall impression for the trial judge. His function was to evaluate the evidence as a whole and to reach a judgment on it. The Court of Appeal was reluctant to interfere with a finding of obviousness, unless the judge had gone wrong on a point of legal principle or if, for some other reason, the decision was plainly wrong.

Furthermore, the trial judge was rightly respectful in his discussion of the Dutch decision, which had been drawn to his notice as a judgment relating to the same patent, but reaching a different conclusion. However, it was a different case decided by different judges on the basis of different evidence and argument. The judge was neither bound by it nor was he obliged to justify his own judgment in the light of it.

The Court of Appeal thus found that the decision of Arnold J on the obviousness of claim 1 was not wrong in principle nor was it plainly wrong for some other reason. There was ample evidence to underpin his assessment. His conclusion was justified on a proper application of his correct understanding of the relevant principles of patent law to the facts found by him on the evidence. It was not the role of the Court of Appeal on the issue of obviousness to become embroiled in a detailed re-consideration of particular findings of fact for which there was some evidential basis. The appeal was dismissed.

NL – Netherlands

The Hague Regional Court of Appeal of 10 June 2014 – Accord v AstraZeneca

Keywords: inventive step – skilled person – expectation of success

AstraZeneca was the proprietor of European patent 0 907 364 for a sustained-release formulation with the active ingredient quetiapine (Seroquel), used in treating schizophrenia, bipolar disorder and depressive disorders.

The Hague Court of Appeal revoked the patent due to a lack of inventive step, reversing the decision of the court of first instance.

The court stated that there was no inventive step if, in view of the closest prior art, the average skilled person would – and not only could – have solved the problem using the claimed approach. An invention was obvious not only if its results could be clearly predicted but also if there was a reasonable expectation of success, i.e. the skilled person could reasonably predict that a research project would be successfully completed within an acceptable timeframe (see "Case Law of the Boards of Appeal", 7^th ed. 2013, chapter I.D.7). A mere "hope to succeed" was insufficient. The skilled person could access everything in the "state of the art" and had "the normal means and capacity for routine work and experimentation" at his disposal. He had a conservative attitude, took no incalculable risks and carried out all routine experiments (see "Case Law of the Boards of Appeal", 7^th ed. 2013, chapter I.D.8). It might be relevant to consider whether testing would be time-consuming and/or complex or whether only routine tests would be needed (see "Case Law of the Boards of Appeal", 7^th ed. 2013, chapter I.D.8).

The court defined the average skilled person as a team comprising a clinician in the field of psychological disorders (psychiatrist) and a formulation expert.

For the purpose of assessing inventive step, the court established that the document Gefvert (1995) was the closest prior art, because it also illustrated the clinical application of immediate-release quetiapine in practice. It disclosed both the substance quetiapine and its pharmacological properties as atypical anti-psychotics.

The court stated that the problem to be solved by the skilled person was to develop a quetiapine formulation (for oral administration) achieving stable and desirable plasma levels at a reduced dosage frequency. In deciding whether this was solved, it was irrelevant whether the skilled person would have any motivation to develop a quetiapine formulation and, if so, whether there was any need for a reduced dosage frequency. Motivation might be relevant in deciding whether the skilled person would have chosen a particular approach to attempting to solve the problem, but that was something different from asking whether the skilled person would actually be motivated to solve the problem in the first place.

The court considered that the skilled person would in any case be motivated to develop a sustained-release quetiapine formulation since there was ample evidence at the priority date on which to base an expectation that quetiapine was (or would be) effective, whereas there was no reason to believe that the risks were such as to dissuade the skilled person from developing the formulation.

The invention was therefore obvious for the skilled person.

3. Olanzapine (EP 0 454 436)

AT – Austria

Supreme Patent and Trademark Chamber of 29 June 2011 (Op 3/11)

Keyword: selection invention

The patent in suit (EP 0 454 436) was for a composition called olanzapine, used as an anti-psychotic.

In revocation proceedings, the Supreme Patent and Trademark Chamber held that a "selection invention" involved choosing certain matter from within a known larger area, i.e. picking out an example with special qualities not previously known. Such an invention was new if the selection met all three of the following criteria (Op 5/05):

1. It was narrower than the known area.

2. It was far enough away from prior-art embodiments and the known area's end-points.

3. It was not arbitrarily picked out as a mere prior-art embodiment, but was a new invention (deliberate selection, new technical teaching) exhibiting special qualities not previously known.

Pre-published document Annex 3 contained numerous benzodiazepines with a Markush formula which included olanzapine.

On criterion 2, the Chamber ruled that whilst there might not be much real "distance" between two similar compounds, when chemical substances were used in pharmacy their effect was unpredictable if substituents were changed even very slightly. For example, replacing an ethyl group with a methyl group could significantly change a drug's effectiveness or range. So even if substituents were very alike structurally, the pharmaceutical effect would not necessarily be similar. Criterion 2 – Olanzapine being sufficiently distant from the Annex 3 compounds – was therefore fulfilled.

On criterion 3, the Chamber ruled that only the patent in suit described olanzapine's unexpected and surprisingly high effectiveness (potential suitability as a neuroleptic, even in small doses; no extrapyramidal side-effects; no agranulocytosis; fewer instances of elevated liver enzymes); Annex 3 did not show this. So criterion 3 – an unexpected effect – was also fulfilled.

NL – Netherlands

The Hague Court of Appeal of 27 September 2011 – Eli Lilly v Ratiopharm

Keyword: novelty – mistake in prior art – selection invention

Eli Lilly was the proprietor of a patent (EP 0 454 436) concerning olanzapine, the active ingredient of a drug used for the treatment of disorders of the central nervous system.

Ratiopharm sought the revocation of the Dutch part of the patent and argued that olanzapine had already been disclosed in a study. The title of the study was inconsistent with the structural formula depicted therein, which in turn differed from the structural formula of olanzapine. Ratiopharm argued that this difference was an obvious mistake which the skilled person would immediately recognise and correct. Therefore, the study disclosed olanzapine directly and unambiguously.

The Hague Court of Appeal ruled that even if the skilled person recognised the study as being manifestly erroneous, he would consult the cited references to ascertain its "technical reality" and then come to the conclusion that the disclosure of flumezapine (and not olanzapine) was intended. Therefore, the study was not novelty-destroying.

The court further stated that the problem-solution approach should also be followed when assessing whether a selection invention involves an inventive step, and held that the skilled person would not select olanzapine because there were no indications in the closest prior art (not even when read together with other documents) that olanzapine had the desired combination of properties, i.e. a good anti-psychotic effect in the absence of certain side-effects.

In conclusion, the court upheld the validity of the patent.

ES – Spain

See "Supreme Court (Civil Chamber) of 10 May 2011 – (309/2011) Laboratorios Cinfa et al. v Eli Lilly and Company Ltd" under chapter VI.4. "Reservation to the EPC".

4. Occlusion device (EP 0 808 138)

GB – United Kingdom

See "Patents Court of 22 July 2014 – AgaMedical Corporation v Occlutech (UK) Ltd [2014] EWHC 2506 (Pat)" under chapter I.C.1. "State of the art".

NL – Netherlands

Supreme Court (Hoge Raad) of 25 May 2012 – AGA v Occlutech

Keyword: claim interpretation

AGA was the proprietor of European patent 0 808 138, which was granted for an occlusion device used to treat structural heart disease, including structural heart defects. The medical device comprised a metal fabric formed of braided metal strands characterised by two clamps adapted to clamp the strands at opposite ends of the device. Occlutech produced dumbbell-shaped medical devices for occluding cardiac septum defects, with only one clamp at one end of the device.

In the previous instances, AGA's patent was held not to be infringed by Occlutech's, since it did not fall within the scope of protection of AGA's patent. The court regarded clamping at opposite ends of the device to be an essential feature of the claim. Such essential features could not be interpreted too broadly unless the description and the drawings provided clear indications for a different approach. The Court of Appeal ruled that there was no basis in the description or drawings for AGA's inventive-thought-based interpretation, which included devices with only one clamp, but which was incompatible with the wording of the claims.

The Dutch Supreme Court dismissed the appeal and concurred with the Court of Appeal's decision. It observed that how a patent was to be interpreted was generally very closely interwoven with factual assessments. This means that, when an appeal was brought before it, it could only review the accuracy of interpretation to a limited extent. It held that Art. 69 EPC, together with Art. 1 Protocol for the application of Art. 69 EPC, could be used in determining the scope of protection of a patent, as well as the essence of the invention and the inventive thought behind the patent claims. It found no fault with the appellate court's interpretation that the scope of protection did not extend to devices with only one clamp at one end of the device.

Editor's note: For the appealed decision see second edition "Case Law from the Contracting States to the EPC" 2004 – 2011, OJ SE 3/2011, p. 251.

SE – Sweden

District Court of Stockholm of 4 March 2011 – Occlutech GmbH v AGA Medical Corporation, AGA Medical Corporation v Occlutech International AB and Tor Peters (joint cases)

Keyword: claim interpretation – scope of protection

AGA was the owner of European patent 0 808 138 for Sweden. The patent concerned a medical device. AGA alleged that Occlutech had infringed the patent, but the infringement claims were rejected by the District Court. In its decision, the court had to decide on the scope of protection of AGA's patent.

AGA claimed that the invention concerned a medical device that had a broad range of uses and was also intended to be used for the treatment of septum defects in the heart. The purpose of the patent was stated in the description as follows: "a dependable embolisation device that is both simple to situate and that can be positioned correctly in a vessel." AGA asserted that the patent included every collapsible device that could be used for any medical purpose; there was no limitation regarding certain types of treatment or exclusion of treatment for septum defects. Furthermore, the claimed invention also encompassed devices with one clamp, in AGA's view. According to Occlutech, the patent claim was limited to embolisation devices.

The court held that, according to § 39 of the Swedish Patent Law, the extent of patent protection was defined by the patent claims. It was further indicated that guidance for understanding the patent claims could be found in the description. The patent description in the case at hand clearly stated that the device concerned comprised an embolisation device. Guidance could also be obtained from other sources such as from the application procedure. The response by an applicant to a notification from the authority could also possibly offer support for a certain interpretation of the patent. In the case at hand, what occurred at the EPO during the application procedure before the patent was granted was of significance for the interpretation of the feature. The applicant had changed the patent claim during the examination procedure with the following explanation: "Nevertheless, to clarify this distinction over the prior art, we have included in claim 1 the feature of clamps provided at either end of the device to clamp the strands together. We submit that this clearly specifies the shape of the device, requiring that the outer ends do not flare outwardly but are constricted to give the required dumbbell shaped expanded configuration." The District Court found that it was clear that the meaning in this part was that the device had to have two clamps, one on each end of the device.

The District Court found nothing to indicate that the invention concerned a medical device that had a broad range of uses and was also intended to be used for the treatment of septum defects in the heart. The patent description clearly stated that the device concerned comprised an embolisation device as asserted by Occlutech. Other analysis did not contradict this. The District Court found that the medical device in the sense of the patent consisted of an embolisation device and that the patent was not infringed.

5. Calcipotriol (EP 0 679 154)

DE – Germany

Federal Court of Justice of 15 May 2012 (X ZR 98/09) – Calcipotriol monohydrate

Keyword: inventive step – reasonable expectation of success

The invention related to calcipotriol monohydrate, a new crystalline form of calcipotriol with, so the patent stated, technical and stability properties superior to the anhydrous calcipotriol known in the prior art.

The Court held that the discovery and provision of calcipotriol monohydrate lacked an inventive step (Art. 56 EPC).

The defendant had argued that the formation of crystalline forms, and so too the formation of hydrates, had not been foreseeable in 1993 (priority date), and still was not today. The Court, however, took the view that the decision as to whether an inventive step was involved did not necessarily depend on the foreseeability of certain results. The crucial question was whether the prior art would have motivated the skilled person to take measures described there and apply them to a known substance. In answering this question, it might be relevant to consider whether the skilled person could reasonably have expected that such action would successfully solve the technical problem (Federal Court of Justice of 6 March 2012 – X ZR 50/09; and of 10 September 2009 – Escitalopram; Swiss Supreme Court of 27 March 1995 – Manzana II; EPO T 60/89 of 1 August 1990 – Fusion proteins).

Those conditions were met in this case. The citations examined by the Court (relating to substances structurally similar to calcipotriol) motivated the skilled person to take the measures described there – dissolving a solid substance in organic solvent while adding water – and apply them to calcipotriol, the outcome of which would have been calcipotriol monohydrate. Moreover, he could reasonably have expected these measures to succeed, and the effort involved – use of organic solvent and water – was proportional to the likely result.

IT – Italy

Turin Ordinary Court of 11 February 2011 (1415/2011) – Sandoz v Leo Pharma – calcipotriol monohydrate

Keyword: inventive step

In these revocation proceedings the Court applied the problem-solution approach followed by the EPO boards of appeal and noted that the objective technical problem addressed by the patent was calcipotriol's low stability in anhydrous crystalline form if exposed to aqueous media (e.g. for production in cream or gel formulations).

The solution claimed was to use a new crystalline – monohydrate – form of the compound which was more stable than the anhydrous form.

The patent taught that the monohydrate was produced by dissolving crystalline or non-crystalline (amorphous) anhydrous calcipotriol in an organic solvent and adding water and possibly a non-polar solvent.

The Court ruled that the patent was not inventive. The skilled person, facing the technical problem of the low stability of the anhydrous crystalline form of a vitamin D3 analogue such as calcipotriol, would have tried to solve it by looking for another crystalline form of the same substance. He could then have found, from the prior art, a crystalline form more stable than the known (anhydrous) one. Looking for more stable crystalline forms was routine for pharmaceutical companies. The skilled person would have known that the crystalline hydrate form was more stable than the anhydrous compound. In aiming to solve the problem of anhydrous calcipotriol's low stability, he would have looked in the prior art which contained chemical compounds that were very similar to calcipotriol and which contained descriptions of processes for obtaining the crystalline monohydrate form.

NL – Netherlands

The Hague Regional Court of Appeal of 9 April 2013 – Sandoz v Leo Pharma

Keyword: inventive step – problem and solution approach – skilled person – expectation of success

In this infringement case, the validity of Leo Pharma's patent for a "New crystalline form of a Vitamin D Analogue" (a medicinal product for the topical treatment of psoriasis which contains the active ingredient "calcipotriol") was examined.

The decision hinged on inventive step. Although not always appropriate, the problem-solution approach could be applied in this case to assess inventive step. The parties agreed that the closest prior art was WO 834, in which the substance calcipotriol in crystalline form (i.e. anhydrate) was described for the first time. The court defined the objective technical problem as follows: how can the storage stability of calcipotriol in a form suitable for therapeutic use be improved over that of the known calcipotriol anhydrate? This was in line with the objective technical problem formulated in the Italian proceedings (see above). The German court identified a two-part problem, one part being the improvement of storage stability (see above).

The court examined whether, in the light of the crystalline calcipotriol anhydrate disclosed in WO 834, it should have been obvious to the average skilled person at the priority date (15 January 1993) to search for a crystalline form of calcipotriol that was suitable for therapeutic use and more stable than the anhydrate during storage (i.e. showing less degradation) and whether he would have found calcipotriol monohydrate.

The court found that the patent lacked inventive step for the following reasons:

The skilled person in the field of vitamin D derivatives was a team consisting of a formulation expert with experience in topical formulations and an analytical pharmaceutical chemist.

In the early 1990s, this skilled person worked in an environment in which the importance of searching for (pseudo)polymorphs of active ingredients in the development of medicinal products was known. The skilled person knew that polymorphism affected the properties of an active ingredient, including its bioavailability but also its (various forms of) stability. It was irrelevant whether a comprehensive polymorphism study was standard procedure (particularly with regards to vitamin D derivatives) in early 1993. What mattered was that the skilled person was familiar with research into polymorphism and its importance.

The court found that the skilled person would have researched the existence of other crystalline manifestations of calcipotriol possibly having better storage stability.

The fact that, as Leo Pharma contended, the outcome of any such polymorphism testing could not be predicted on the basis of theory would not have deterred the skilled person from carrying out such tests, precisely because testing was simple and relatively cheap. It was already known from WO 834 that calcipotriol could be crystallised and the skilled person was aware that it was possible to form a monohydrate from the three vitamin D analogues similar to calcipotriol in structure which were described in the prior art, and that this could be done in a very simple manner. The likelihood of success was sufficient for the skilled person to start testing. Therefore, the skilled person could have established the formation of calcipotriol monohydrate at the priority date.

SE – Sweden

District Court of Stockholm of 20 May 2011 – Leo v Sandoz

Keyword: inventive step – novelty

Leo was the proprietor of European patent 0 679 154. Sandoz alleged that the patent was invalid for lack of novelty as well as lack of inventive step. As to novelty over the "Acne Patent" (WO 91/12 807 A) and the "Base Patent" (EP 0 227 826 B), Sandoz alleged that by carrying out the teaching of the "acne patent", the skilled person would inevitably arrive at a composition containing calcipotriol monohydrate. The court held that this assumed that the anhydrous form of calcipotriol would be produced using other processes than those available at the time the "acne patent" was filed (1991). Furthermore, it followed from the statements made by the expert witnesses that once a particular crystalline form had been discovered and reproduced, it was practically impossible to avoid contamination by seeds of the new crystalline form. Hence it could not be ruled out that the anhydrous form used in the comparative experiments had not been contaminated by monohydrate crystalline seeds.

As to inventive step, the court considered the anhydrous form of calcipotriol known from the "Base Patent" as the closest prior art. The problem indicated in the application had to do with the need to regulate the size of the crystals by means of a wet ball mill process. According to the description, wet milling had proven to be difficult to carry out with use of the water-free crystal form described in the Base Patent. The difficulties were indicated to be due partly to wettability and partly to the circumstance that a stable foam was produced made it difficult to obtain a suitable small and uniform particle size. The differences between the anhydrous form and the monohydrate that were relevant in formulation of the problem concerned the characteristics in wet milling. They were better with the monohydrate, since the crystals were easier to wet and no stable foam was formed during milling, which meant that the size of the crystals could be controlled. With this background the objective technical problem should be formulated as follows: the provision of a suspension of calcipotriol that is suitable for use in pharmaceutical compounds for the production of creams and gels.

The solution to the problem was to use monohydrate of calcipotriol, which crystal form allowed for regulation of the particle size by means of wet milling. The next question was whether there was anything in known technology at the priority date that would have led an expert to solve the given problem by producing the monohydrate of calcipotriol. This step could also be expressed in such a way that there should have been an incentive in the known technology that would have led an expert to the solution to the problem or, in other words, that it would have been clear to an expert against the background of known technology to try to produce the monohydrate of calcipotriol as a solution to the problem. The court found this not to be the case.

It held that the object of the patent was to solve a specific foaming problem of a crystal suspension of calcipotriol and that there was no pointer for solving this problem by forming a monohydrate in the prior art. The court was not of the opinion that a skilled person would have investigated whether calcipotriol formed a monohydrate anyway, e.g. in order to improve the stability of the anhydrous form or because the formation of a foam occurred during a wet ball milling process wherein an aqueous milling fluid was used. The court also did not consider that the possibility of a spontaneous formation of a hydrate of calcipotriol in an (aqueous) suspension cream, which Sandoz argued must happen for thermodynamic reasons, was sufficient motivation for a skilled person to investigate whether such hydrates might form or not. Sandoz cited several documents relating to three vitamin D analogues for which the only stable crystalline form was monohydrate. The court held that no firm conclusions could be made for the crystalline properties of a particular compound, here the wettability, based on the crystalline properties of similar compounds. Therefore, the skilled person would not find any guidance from these documents. Rather than searching for new crystalline forms of calcipotriol, the skilled person would seek to modify the process conditions of the wet milling.

6. St Gobain/Knauf (EP 0 399 320)

BE – Belgium

Liège Court of Appeal of 19 September 2013 (2011/RG/1503) – Saint-Gobain v Knauf

Keyword: scope of protection – interpretation of claims

Saint-Gobain was the proprietor of European patent 0 399 320. The claim at issue was directed to the use of glass fibres.

The Court cited proceedings in France – notably before the Paris District Court, which had declared Saint-Gobain Isover's submissions inadmissible and dismissed Knauf's revocation action (as later revised by the Paris Court of Appeal's judgment of 16 May 2014, see next summary).

The original application – especially its claim 1 – had been amended several times before being granted in the final version cited against Knauf. The Court ruled that if a company such as Saint-Gobain itself decides to amend its own claims during grant proceedings, it cannot later go back on what those amendments mean.

Citing basic principles, Saint-Gobain argued that claims could never be construed strictly or literally; they always had to be interpreted in the light of the description and drawings. So the skilled person reading the term "diameter" (of fibre populations) and required to say exactly what it meant in claim 1 would give it its true technical meaning of "median diameter". The Court however ruled that Saint-Gobain was defending a reading which boiled down to going back on its earlier amendment. It was trying to reinsert implicitly the term "median", which it had agreed to delete during proceedings before the EPO, whilst maintaining it (as "mittleren" in German) in claim 2.

Deleting "mittleren" from claim 1 had restricted the scope of protection. Saint-Gobain could not be allowed to use interpretation as a way of undoing that, and restoring protection it had consciously forfeited. Saint-Gobain's "interpretation" of claim 1 was designed not to reveal its essential meaning but to revert to the earlier, broader scope of protection conferred – which however was prohibited by Art. 123(3) EPC.

Ruling on infringement, the Court said the theory of equivalents could not be used to restore matter dropped by the patentee itself during grant proceedings.

FR – France

Paris Court of Appeal of 16 May 2014 (12/06678) – Knauf v Saint-Gobain

Keyword: glass fibres – revocation of European patent – insufficiency of disclosure – added subject-matter

The Court noted that Saint-Gobain had sued Knauf in both Belgium and Germany for infringing European patent 0 399 320. The present case concerned claim 1 of the patent's French part.

The patent was for "glass fibres with increased biological compatibility". The Court began by reviewing in detail how knowledge about the carcinogenic properties of asbestos had developed over time. Its persistence in the lungs depended on the composition and size of the fibrous dust.

The "skilled person" was a team comprising a chemist with a good knowledge of mineral wool fibres and a pathologist/toxicologist.

The Court ruled that the skilled person did not know whether it was better to use fibres with a smaller diameter, as stated in the description, or a larger one, as suggested by the properties of the fibres according to claims 1 and 2. There were inconsistencies regarding the diameter of the fibres, and formulations such as "which exhibit no carcinogenic potential" or "having no carcinogenic properties" were functional definitions not sufficiently defined in the patent, which also did not refer to any standards. The tests described were outside its scope. Lastly, its teaching did not on its own disclose the invention sufficiently clearly and completely to enable the skilled person to carry it out.

Knauf argued that Saint-Gobain had extended the subject-matter, in particular by incorporating into claim 1 just a part of claim 3 as originally filed, and by deleting the reference to the fibres' median diameter. On the former issue, Saint-Gobain countered that the amendment had been accepted in EPO opposition proceedings as complying with Art. 123 EPC. But the Court ruled that the features presented as essential had changed significantly. The skilled person had no reason to believe there was any correlation between the fibres' composition and their median diameter, and had no incentive to combine the two aspects.

After considering all the pleadings, the Court set aside the Paris District Court's judgment of 11 December 2011 and revoked European patent 0 399 320 for added subject-matter and insufficiency of disclosure.