review of patent cases in the english courts in 2014 patent review_final lres.… · case –...

Review of Patent Cases in the English Courts in 2014 by Brian Cordery, Dominic Adair, Naomi Hazenberg and Sylvia Delbeuf

© Bristows LLP February 2015

2

Quotation of the Year

“I have reversed my decision in your favour and award £5,000,000.00 in damages that would settle the claim in full. I apologise that I did not even question your opponents on this issue or the matters concerning the manipulation of design sheets and copyright dates as at the time I didn’t think it was all that relevant.” Per Richard Perry (impersonating HHJ Hacon in a letter to the Court and the parties) in Brundle v Perry [2014] EWHC 475 (IPEC)

The information contained in this document is intended for general guidance only. If you would like further information on any subject covered by this Bulletin, please e-mail Brian Cordery ([email protected]), or the Bristows lawyer with whom you normally deal. Alternatively, telephone on + 44 (0) 20 7400 8000.

Review of Patent Cases in the English Courts in 2014 1


Index

Quotation of the Year Inside front cover

Introduction 2

Claim Construction and Infringement 2 Indirect Infringement 4 Innocent Infringement 4

Validity 5 Patentability 5 Priority 5 Novelty 6 Obviousness 6 Skilled Addressee 7 Insufficiency 8 Added Matter 8

Supplementary Protection Certificates (SPCs) 8

Damages 9 Compensation pursuant to Cross-Undertakings 9 Account of Profits 10

Costs 10 Employee compensation 11

Patent Attorney Negligence 11

Procedural Issues 12 Summary Judgment 12 Partial Invalidity 12 Amendment Applications 12 Jurisdiction 13 Stay and EPO Oppositions 14 Stays and EPO Central Limitation 14 Expedition 15 Disclosure 15 Final Injunctions 16 Dissemination of Judgments 16

Unitary Patent/Unified Patent Court 16 The Past Year 17 2015? 17 Other Issues and Overall Timings 18

Looking Ahead 19

The Authors 20


2

5 [2004] UKHL 466 [2009] EWCA Civ 10627 Improver v Remington [1989] RPC 69; the term “Protocol Questions” was coined in Wheatley v Drillsafe [2001] RPC 7 by Aldous LJ8 Actavis v Eli Lilly [2014] EWHC 1511 (Pat)9 Hospira v Genentech [2014] EWHC 1094 (Pat)10 [2013] EWCA Civ 93

1 [2014] UKSC 552 Actavis v Eli Lilly [2014] EWHC 1511 (Pat)3 AstraZeneca v Krka [2014] EWHC 84 (Pat)4 [2014] EWHC 2404

Introduction

2014 was another busy year in the English Patents Courts with 79 decisions on substantive issues compared with 81 in 2013 and 82 in 2012. As has been the trend in recent years, the majority of cases in the court lists concerned life sciences and telecoms although this year there were also important cases in other technical fields.

Highlights and points of note from the year included:

• The Supreme Court adjudicated another patent case – Servier v Apotex1 – which concerned the principle of ex turpi causa non oritur actio. The UK’s highest court agreed with the decision of the Court of Appeal but for slightly different reasons;

• Floyd LJ, Birss J and HHJ Hacon both spent a first full year in their respective new roles. All are continuing to make a valuable contribution to UK patent law;

• The Patents Court granted Actavis a declaration of non-infringement not only in respect of a European Patent (UK) but also its French, Italian and Spanish counterparts2;

• There was a comparatively rare decision on the amount of compensation to be awarded under the cross-undertaking given on an interim injunction application3;

• The Patents Court implemented the CJEU’s ruling in the Eli Lilly v HGS case4; and

• The Unitary Patent system, which at times over the last decade or more has seemed to be politically too difficult, now appears to be lumbering inexorably towards coming into effect, although the precise timing is still the subject of debate.

As with previous years, this review attempts to summarise the most important decisions on a topic-by-topic basis. The UK Patents Act 1977 is referred to as “the Act” and the European Patent Convention 2000 as “the EPC”.

As ever, the authors have endeavoured to cover every important development that occurred during the course of the year. However, as this is a condensed summary, not every decision is mentioned.

Claims Construction and Infringement

By and large, the English courts are continuing to construe claims in accordance with the guidance given by Lord Hoffmann in Kirin Amgen5 and the elaboration given by Jacob LJ in Virgin Atlantic v Premium Aircraft6.

As reported in previous reviews, the Improver or Protocol questions7, which were commonly used in the nineties and early noughties to determine whether variants of a claimed product or process infringed, have gradually fallen into disuse. It was therefore something of a surprise when Arnold J deployed them in determining the issue of infringement in the pemetrexed case discussed in more detail below8. Time will tell whether this decision will come to be regarded as an outlier, or whether it will mark the start of a return to fashion of the Protocol questions. Another interesting aspect of the pemetrexed case was Arnold J’s use of the file wrapper as an aid to construing the claims. Although there have been one or two previous decisions in which the actions of the patentee have been taken into account when construing a claim, it has been more honoured in the breach than the observance, to say the least. The Judge opined that “anyone who is interested in ascertaining the scope of a patent and who is professionally advised will obtain a copy of the prosecution file and will consider it to see if it sheds light on the matter.” Finally, on the issue of construction, Arnold J held that although Swiss-type claims were a legal fiction designed to circumvent Article 52(4) EPC 1973, this did not warrant interpreting such claims as if they were claims to a method of treatment. Rather such claims should be considered to be directed to the manufacturer of the medicament in question.

The construction of Swiss-type claims was also considered by Birss J in an important passage of his decision relating to Hospira’s preparations to market its version of Genentech’s cancer medicine Herceptin9. Following Regeneron10, Birss J held that


13 See footnote 6, ante14 Jarden v SEB [2014] EWCA Civ 1629 15 See footnote 14, ante16 Terrell on the Law of Patents, 17th Edition, at paragraph 9-107 17 [2014] EWCA Civ 146218 [2014] EWHC 1559 (Pat)

11 Hospira v Genentech [2014] EWHC 3857 (Pat)12 Kennametal v Pramet, [2014] EWHC 565 (Pat)

the word “treatment” in a claim was to be construed as a “functional technical feature” – i.e. a claim to “X for the treatment of disease Y” would be to something which was indeed effective for the disease in question. Additionally, Birss J held that “for” in such claims meant “suitable and intended for”. Dwelling on the implications of the above findings, the Judge noted that a prior art proposal to administer the drug in the manner claimed was not enough to deprive the claim of novelty because the proposal did not disclose that the treatment was efficacious. Finally, Birss J confirmed that clinical data would not generally be required in a patent provided that the specification already satisfied the requirement of “plausibility” – a phrase coined to characterise what it is that a patent specification must provide in order to be sufficient, short of full clinical proof of efficacy. In a later case between the same parties11, Birss J observed obiter that it was not settled as to whether Swiss-type claims should be construed as product or process claims. Birss J also considered the construction of “product-by-process” claims, and identified the apparent inconsistency in the House of Lords decision in Kirin-Amgen that the scope of such claims can be different for novelty as opposed to infringement/sufficiency. However, he did not depart from that analysis as it is binding on him under English rules of precedent.

In Kennametal v Pramet, the principles of claims construction were applied to a claim that required the cutting edge of a cutting tool insert to have a substantially straight region12. The claimant argued that, as no size limitations were specified within the claim, this feature of the claim, could be satisfied by a substantially straight region that was tiny - that which might occur in an arc of large radius. However, no straight region could be seen in the defendant’s design drawings or in the products if examined by the naked eye or under magnification. Sitting as a Deputy Judge, Henry Carr QC held that the principle of reasonable certainty required that the skilled person should be able to determine whether there was a substantially straight cutting edge region without having to resort to the type of elaborate experiments that were conducted in this case. Accordingly, the claimant’s argument was rejected.

Although it is permissible to use the reference numerals

in a patent specification as an aid to understanding the invention, they must not be used when construing the claims. As Jacob LJ put it in Virgin13: “the numbers help you get the map the right way up, they do not help you read it to find out exactly where you are”. One of the questions on appeal in Jarden v SEB14 was whether Arnold J had erred in using numerals in his construction of the claim to find that the main body of a chip fryer included the lid (when considering the location of a “main heater means”) and hence that Jarden’s fryer, which had a lid-mounted heater, infringed. In a judgment given by Vos LJ15, it was held that Arnold J had fallen into error on this point; he had not used the numerals merely to identify the parts of the device but had impermissibly used them to direct the skilled reader as to which parts of the patented device to include within a particular term in the claim. Construing the claims afresh and without use of the numerals, Vos LJ adjudged that including the lid in the meaning of “main body” ignored the clear language of the claims and that a purposive construction could not overcome the clear restriction put into the later claims that the main body was separate from the lid. As a result, the appeal succeeded and the finding of infringement was reversed.

Should the courts be reluctant to construe a patent in such a way that would lead to a finding that it was obvious over common general knowledge? It is said by the authors of Terrell that they should, for such a construction would be absurd16. Floyd LJ considered this question in the Court of Appeal’s decision in Adaptive v BT17. He examined the two authorities cited by Terrell on this point and declined to follow them, noting that a canon of construction is not a rigid rule. It was possible for a patentee to have been isolated from the common general knowledge, or to have genuinely believed that he had made an invention over it. On the facts of the case at hand he was not persuaded that it was right to give it any weight.

In what turned out to be a moot point, as all of the relevant claims had already been held to be invalid, the construction of “suitable for” was considered in the context of an anticipation attack in Rovi v Virgin18. The challenger alleged anticipation by a piece of prior art because the hardware described was capable of carrying out all of the claimed functions if the relevant software


4

25 See footnote 20, ante26 [2014] EWHC 228 (Pat)

19 [2008] EWHC 329 (Pat)20 Koninklijke Philips Electronics N.V. v Nintendo of Europe GmbH [2014 EWHC 1959 (Pat)21 [2013] EWHC 1227 (Pat)22 See e.g. §§ 177 and 182 of his judgment23 Swarovski-Optik v Leica [2014] EWCA Civ 63724 [2014] EWHC 3916 (Pat)

was added. Mann J rejected this argument, stating that it would lead to the striking result that a large number of computer-based patents would become vulnerable to revocation or a large number of computers would infringe patents even if they had not been configured in such a way as to reproduce the patented activities. It would lead to products infringing merely because they could be made to reproduce the patented activities. Mann J applied Floyd J’s reasoning from Qualcomm v Nokia19 that if a physical modification was required, the apparatus in question cannot be “suitable for” use and cannot infringe. This reasoning was subsequently discussed by Birss J20 when he considered whether a general purpose computer could be considered as apparatus suitable for virtual body modelling.

A curious feature of a decision by Vos J in Swarovski-Optik v Leica21, a case about riflescope optics, was the comment in the final paragraph of his judgment that claim 3 (a dependent claim) may be invalid for obviousness, despite the non-obviousness of parent claim 1. The Judge appeared to take the view that claim 3 provided an alternative to claim 1 insofar as it claimed a different arrangement of lenses22. The case was then appealed23 on a number of points but, in dealing with the issue of claims construction, Floyd LJ took the trouble to point out that in fact Vos J recognised that error at a hearing after judgment but did not consider that it had any bearing on his judgment or order. Debate on the interpretation of the patent claims continued on appeal and the question arose as to whether in fact claim 3 must be narrower than claim 1, since it would not make sense for the draughtsman to include limitations within claim 3 that were already present within claim 1. Floyd LJ noted that although dependent claims are usually narrower in scope that is merely a common canon of construction and not an inflexible rule of law. In the present case, it would lead to an absurd result in the interpretation of claim 1 and hence the claims must be interpreted instead to leave claim 3 redundant for having identical scope to claim 1.

Indirect Infringement

The longest decision of the year was that of Arnold J in Idenix v Gilead24 concerning the former’s patent to certain nucleoside analogues which had activity against

viruses of the Flaviviridae family, including HCV. The proceedings were issued on 12 March 2014 just hours after the patent had granted, and a trial occupying some 11 days of court time was held just seven months later in October. Most of the decision concerns the validity of the patent which, as will be described in more detail below, was held to be comprehensively invalid. However, the decision also contains an interesting finding on infringement - namely that Gilead’s sofosbuvir molecule would have infringed both directly (it being held that the masked phosphate group in sofosbuvir was a “phosphate” within the meaning of the claim) and also indirectly because the metabolism of sofosbuvir led to the production of at least one compound falling within the patent and Gilead possessed the requisite knowledge to be liable for indirect infringement.

A similar finding was reached by Birss J in relation to the Wii console in Philips v Nintendo25. Having found that use of Nintendo’s Wii system with a balance board and the relevant game software directly infringed Philips’ patent, Birss J went on to consider the question of indirect infringement under Section 60(2) of the Act. The Judge was clear that a bundle containing board, console and software was certainly “means essential” for putting the invention into effect but Philips had asserted that the Wii console, sold alone without the balance board or the relevant game software, was also “means essential”. Here again, the Judge agreed. The Wii console was a “critical part” of the patented combination even if sold alone. It was noted that in theory this reasoning could lead to the sale of a product being prohibited under this section even if it pre-dated the patent but the Judge did not regard this as a problem; the test demanded a subjective assessment of whether in fact there is an intention to use for the patented purpose in addition to whether the means are suitable for that purpose when assessed objectively.

Innocent infringement

Readers will be aware that under Section 62(1) of the Act, no damages will be awarded against an innocent infringer provided he proves that he was not aware of the patent and had no reasonable grounds for supposing that it existed. In Collingwood v Aurora26, Roth J rejected Aurora’s argument that it was an


30 [2013] EWHC 1876 (Pat)31 [2014] EWCA Civ 133532 See footnote 9, ante33 T903/05

27 Lantana Ltd v Comptroller-General of Patents, Designs & Trade Marks [2014] EWCA Civ 146328 [2009] EWHC 34329 [2013] EWCA Civ 451, [2013] R.P.C. 30

innocent infringer of Collingwood’s patent. Considering the second, objective, limb of the test, it was adjudged that various adverts for Collingwood’s products in trade literature were sufficient to put Aurora on notice that there might be patent protection. The Judge reached this finding despite the fact that neither the patent nor the patent application was referred to in the advertisements because it was common to seek patent protection in the field of LED lighting. Further, the Court found it striking that there was no direct evidence from any witnesses as to the awareness of Collingwood’s products within Aurora’s company. An additional factor, which the Judge stated should have led Aurora reasonably to have discovered the patent, was the fact that it based the design for its products on Collingwood’s products. In so doing, the design team failed to check if it was protected by a patent and Aurora failed to ask its patent attorney (with whom it regularly worked) to conduct a search for relevant patents.

Validity

Patentability

Following Birss J’s judgment last year and the previous rejection of Lantana’s patent application by the UKIPO, the Court of Appeal upheld the decision that the invention was excluded from patentability for being a computer program as such27. With reasoned judgments given by both Arden LJ and Kitchin LJ, the Court held that there was no inconsistency between the acceptance that a claim was novel and inventive and a finding that its contribution fell solely within excluded subject matter. Arden LJ noted that this was a “deliberate legislative policy”. It was alleged that the Judge had used an erroneous approach in considering the claim broken down into its constituent parts rather than assessing the claim as a whole. This was rejected. As required, Birss J’s judgment focused on the overall contribution that had been made. The judgment of Arden LJ considers the signposts set out in AT&T28 and modified in HTC v Apple29 which can be useful to identify a technical contribution and hence save the invention from being excluded. Commenting on the fifth signpost, she noted that circumvention of a problem can amount to an invention but, in the case at hand, the mode of the solution in the claim was the “perfectly

straight forward use of a familiar medium”. Overall, the AT&T signposts did not assist Lantana – there was no identifiable technical effect outside the computer program itself. On the value of EPO decisions as precedents, Arden LJ noted that, due to the adoption of different methodologies in approaching this type of excluded subject matter, EPO decisions provided only limited assistance to a judge in the English courts.

Priority

In upholding Birss J’s decision30, the Court of Appeal in HTC v Gemalto took into account the skilled person’s background knowledge when determining the extent of the priority document’s disclosure31. The priority document disclosed a particular method of compacting an application program, called namespace mapping. The question for the Court was whether this disclosure was sufficient to support a claim that was broad enough to encompass all compaction methods. The Court of Appeal confirmed that what must be considered in the priority document is what was disclosed to the skilled person, rather than what was obvious in light of the common general knowledge. However, the priority document should not be read in a vacuum. The knowledge and background of the skilled person can be considered in determining what is disclosed. Taking this into consideration, the Court of Appeal confirmed Birss J’s decision that the skilled person would not consider, given their knowledge of many compaction methods, that they had strayed from the teaching of the priority document by using a method other than namespace mapping. Indeed, the priority document’s common theme of space and time efficiency was such that it taught a principle of general application that was reflected in the granted claim. This contextualisation of the disclosure of the priority document seems to provide a slight retreat from the strict application of the law on priority that emerged as a theme from last year’s cases.

Patentees concerned about priority may also be pleased with the way in which Birss J handled the issue in the first Hospira v Genentech32 case. Ultimately, and with some reluctance, Birss J disagreed with the EPO decision in Gemvax33 and held that the priority document must contain an enabling disclosure which, in the context of an invention including the achievement


6

37 Eugen Seitz v KHS and Norgren [2014] EWHC 14 (Ch)38 Pozzoli v BDMO SA [2007] EWCA 58839 [2007] EWHC 1040 (Pat) per Kitchin J at §7440 KCI v Smith & Nephew [2010] EWHC 1487 (Pat)41 [2014] EWHC 2873 (Pat)42 British Acoustic Films ([1936] 53 RPC 221 at 250) per Luxmoore J

34 See footnote 24, ante35 [2009] EWHC 1304 (Pat)36 [2014] EWHC 2506 (Pat)

of a therapeutic effect, meant that the patentee must show that the invention was plausible. However, perhaps softening slightly the fairly rigid approach on priority that the English courts have taken of late, Birss J observed that “in a specification filed without claims, the various features found in the claim of the granted patent are unlikely to be written out in a neat paragraph”. In the circumstances, Birss J found that there had been a direct and unambiguous disclosure of the claimed invention in the priority application.

On the question of formalities, among other important issues in Idenix34 was the finding that a Research and Development agreement between A and B that provided that B “shall at all times be the owner” of any resulting IP was enough for the title in a resulting patent application to pass automatically from A to B without the need for formal assignment. Arnold J held in the alternative that the wording amounted to an agreement to assign and that this created an equitable right which was sufficient to render B the successor in title for the purposes of priority. This case is arguably a further example of the English courts adopting a more lenient approach to what constitutes a successor in title for the purpose of assessing priority, following the strict interpretation by Kitchin J in Edwards v Cook35.

Novelty

The novelty of a patent is destroyed if the invention was made available before the priority date to at least one person who was free in law and equity to use it. Just one person will do. In the case of AGA Medical v Occlutech36, the recipients were three doctors, an anaesthetist and a nurse in a children’s hospital in Bratislava. The disclosure was made by prior use of the invention – the “Amplatzer” medical device for closing a hole in the heart – supervised by the inventor himself, Dr Amplatzer, who had travelled to the Slovak Republic to oversee the insertion of a small number of devices by way of a clinical trial. The critical question was whether the clinical trial was confidential. It was accepted that there was no express legal obligation as to confidentiality but the patentee argued that there was an equitable duty of confidence. The EPO Opposition Division had already taken the view that a clinical trial gives rise to a rebuttable presumption as

to confidentiality (and found no evidence to rebut it). However, on the facts before Roth J, the English court decided that there was no duty of confidence: the trial was not perceived as a commercial venture but rather as a “revolutionary” scientific endeavour, the first of its kind with this device, which would lead the doctors to wish to speak widely about it. Accordingly, the patent was held to have been anticipated.

In contrast, a fax between two separate companies engaged in a Research and Development collaboration was held to be confidential and therefore not part of the prior art for the purpose of assessing the novelty of an invention relating to plastic pistons in the valves used in stretch blow moulding machines37.

Obviousness

The English courts continue to address the question of obviousness using the Pozzoli test38. Most decisions also refer to the summary of the principles set out by Kitchin J in Generics v Lundbeck39. The courts can also take into consideration information neither disclosed in a specific item of prior art relied on nor forming part of the common general knowledge if it can be shown that the skilled person, faced with the problem to which the patent is addressed would acquire that information as a matter of routine40. In finding a patent to a dosage regimen for an emergency contraceptive method obvious, Sales J found that the skilled person would ascertain as a “matter of routine” that the prior art citation contained an error when it referred to a dose of “1.5g” of a particular active ingredient rather than “1.5mg”.

Sales J went further in Teva v AstraZeneca41 in finding that material readily identified by a database search of relevant journal articles would be classified as common general knowledge. As readers will know, traditionally, the common general knowledge has been defined as “knowledge which is generally known and accepted without question by the bulk of those who are engaged in the particular art”42. In holding AstraZeneca’s patent for the combination of the long-acting beta2 agonist, formoterol and the corticosteroid, budesonside to be invalid over a prior art patent in combination with the common general knowledge, Sales J held that the


47 Teva v Leo Pharma [2014] EWHC 3096 (Pat)48 Schlumberger v Electromagnetic Geoservices [2010] EWCA Civ 81949 Environmental Defence Systems v Synergy Health and others [2014] EWHC 1306 (IPEC)

43 See footnote 40, ante44 See footnote 11, ante45 See footnote 24, ante46 E.g. AgrEvo/Triazol sulphonamides, T939/92, [1996] EPOR 171

common general knowledge should be “adapted and kept appropriately up to date for the procedures for the dissemination of scientific knowledge in the age of the internet and digital databases of journal articles.” In the authors’ view such material may well have been taken into account by the more experienced patents judges but probably as the sort of background information that Arnold J specifically held was not common general knowledge in KCI v Smith & Nephew43.

As noted, English courts at all levels recognise obviousness as requiring a multifactorial assessment. Although it could sometimes be of use, Birss J found in the second Herceptin44 case that the so-called “could/would” test was an oversimplification:

“...The issue is multifactorial and based closely on the particular circumstances.... the word “would” is not always straightforward. Sometimes asking simply if a skilled person “would” do something risks placing too much weight on what are really minor or irrelevant factors like cost, instead of focussing on the technical issues. Moreover, the well known 9½ inch plate is not something a skilled person would make. It is more accurate to say that it is not patentable because the skilled person could make it without any inventive step.”

On the subject of 9½ inch plates, “lack of technical contribution obviousness” is becoming an increasingly popular line of attack against patents. Idenix45 confirms that UK law follows the approach of the EPO in relation to this ground of invalidity46. In the case at hand, Arnold J noted that it was not plausible that substantially all the compounds covered by the claims of Idenix’s patent would be effective against viruses of the Flaviviridae family including HCV. Thus the claims were held to lack inventive step.

When Teva challenged Leo Pharma’s patents to an ointment containing calcipotriol and corticosteroids with a solvent from a specified class, it ran its obviousness attack in an unconventional manner – starting from the common general knowledge that a calcipotriol plus corticosteroid combination product would be desirable and then adding in a prior art citation which was a US patent teaching the use of the claimed solvents in

pharmaceutical preparations with anti-inflammatory steroids for topical administration. Overall, Birss J accepted Teva’s argument even though he noted that it only worked if the US patent were notionally presented to the skilled person at a specific point in the sequence of steps to the invention47. Birss J observed that his findings were limited to the case before him. He also confirmed that regulatory considerations could be taken into account for the purposes of inventive step but that their significance would vary from case to case and that “given they are commercial rather than directly technical in nature, these factors are unlikely to outweigh the technical considerations in any but the strongest cases”.

Skilled Addressee

Many readers will remember Jacob LJ’s judgment in Schlumberger48 as authority that the identity of the skilled addressee for assessing priority date considerations such as inventive step may differ from that for assessing post-grant considerations such as insufficiency or claim construction. This becomes important for those rare inventions that marry together elements from very different technical fields, where realising the relevance of the different technology forms part of the inventive step. The IPEC case of EDS v Synergy49 was just such a case. Here, the purported invention was the realisation that absorbent pads (known from personal hygiene products) could be used effectively in sandless bags for flood defence systems. HHJ Hacon decided to determine inventive step as a preliminary issue and found the patent invalid on the basis of evidence that use of superabsorbent polymers in sandless bags was well known before the priority date. The case contains some useful practical observations for pleading cases in the IPEC, where detailed particulars must be provided up front. HHJ Hacon indicated that in their pleadings parties should provide a statement of the skilled addressee’s technical field so that the common general knowledge from that technical field can be addressed. Such a statement would be of particular importance in cases such as this where parties disagree over the identity of the skilled addressee and because it would be rare for more than one expert per party to be permitted in the IPEC.


8

53 See footnote 11, ante54 See footnote 52, ante55 C-493/1256 See footnote 4, ante

50 See footnote 24, ante51 See footnote 9, ante52 [2014] EWCA Civ 40

Insufficiency

As well as being found invalid for obviousness, the claims of Idenix’s UK patent to certain nucleoside analogues which had activity against viruses of the Flaviviridae family were held insufficient due to excessive claim breadth. Further the Judge held that the patent did not enable the skilled team to perform the invention across the breadth of the claims without undue burden. In effect, it had set the skilled team a research project and claimed the results50.

Birss J’s decision in Hospira v Genentech51 once again highlighted the squeeze between inventive step and sufficiency that patents to dosage regimens in particular must often overcome. Genentech relied on construing “for” in its patent for the administration of trastuzumab to a breast cancer patient once every three weeks as meaning “suitable and intended for” thereby bringing efficacy into the assessment of inventive step. Birss J nevertheless found the patent invalid for obviousness in light of the FDA label for Herceptin, which described a weekly administration of trastuzumab with the chemotherapeutic agent paclitaxel being administered every three weeks, and the common general knowledge. As a result, there was no need to consider sufficiency. However, Birss J noted obiter that if the patent did involve an inventive step, then the patent did not contain sufficient clinical data to support a claim that the treatment was efficacious.

Added Matter

The decision of the Court of Appeal in AP Racing v Alcon Components52 will be remembered best for the judgment of Floyd LJ on the issue of added matter. The patent in suit related to disc brake callipers having peripheral stiffening bands (“PSBs”) for use in motor racing and the claim under consideration included an integer that “each of the stiffening bands has a profile that is asymmetric about a lateral axis of the body when viewed in plan.” The teaching of the specification was that the PSBs were hockey stick-shaped in nature. Overturning the first instance finding by HHJ Birss QC in the Patents County Court on added matter, Floyd LJ held that there was no prohibition in claiming an invention in more general terms than that which was described

in the specification. The prohibition was directed to the disclosure of new information to the skilled reader. Thus, “the skilled person would understand from the granted patent, just as in the case of the application, that the PSBs disclosed include those which follow a lateral and leading edge (and therefore are asymmetrical about the lateral axis). The skilled person would also understand that the PSBs are exemplified by the hockey stick shapes described in the specific embodiments. He or she would not, therefore, learn any new information about the invention.”

Birss J took this instruction on board in Hospira v Genentech53 where he noted his error in AP Racing54 and discussed candidly the difficulty of applying the law on added matter when the EPO takes a noticeably different, and stricter, approach. At present, it seems that in some respects the English Patents Courts are applying a more stringent test for priority than added matter which seems a little peculiar given the purpose of the rules and that the priority document will inevitably be prepared at an earlier stage of the development of the invention.

Supplementary Protection Certificates (SPCs)

The unenviable task of implementing the ruling of the Court of Justice of the European Union (CJEU) in Eli Lilly v HGS55 fell to Warren J in the spring of 2014. It is a very poor reflection on the clarity, if not the quality, of the CJEU’s decision that both sides were credibly able to claim that the CJEU had opined in their favour. In a decision dated 18 July56, the Court held that Eli Lilly was not entitled to a declaration that an SPC filed by HGS based on Eli Lilly’s Marketing Authorisation for its antibody tabalumab, would be invalid. In reaching his decision, Warren J accepted HGS’ argument that whether a product was “protected by a basic patent in force” for the purposes of Article 3(a) of the SPC Regulation (Regulation (EC) No. 469/2009) depended upon whether it fell within the scope of the claims of the patent as interpreted in accordance with Article 69 of the EPC/Section 125 of the Patents Act. In this instance, HGS’ patent claimed antibodies with a functional definition – namely that they should bind to the protein neutrokine-alpha and it was common ground


60 Astrazeneca v Krka Order of Vos J dated 8 October 201061 AstraZeneca v Krka [2014] EWHC 84 (Pat)

57 [2009] EWHC 191658 [2008] EWHC 190259 See footnote 55, ante, para. 43

that tabalumab did in fact bind to neutrokine-alpha.

Thus, as a matter of principle, an SPC for A should be allowed in respect of a patent containing a Markush formula which covers A. However, an SPC for A plus B will not be allowed based on a patent claiming A on the basis that although this product would infringe the patent for A, it is not “covered” by it. Warren J noted that this principle should be subject to one proviso – namely that where the claims contain some general wording extending the scope of the claim beyond the principal wording, a product would not be considered to be specified unless it fell within the scope of the claim absent the general wording. In other words if a patent is directed to A but contains a claim to “a pharmaceutical composition comprising A”, A plus B would not be “protected” by this patent even though “comprises” is understood to mean “including but not limited to” under EPC drafting conventions. Thus Arnold J’s decision in Astellas57 would still be correct but whether Kitchin J’s decision in Gilead58 would still be deemed to be correct would depend on how the Court viewed the words “and optionally other therapeutic ingredients”.

Warren J noted that the CJEU seemed to have commented on the so called “Third Party Issue” i.e. the fact that HGS’ SPC would be based on a competitor’s Marketing Authorisation59. However the Judge refused to deal with it as the issue had been formally abandoned by the parties, although he did appear to note that the difficulties with the approach seemingly proposed by the CJEU.

Over the summer, the CJEU gave further guidance on the interpretation of an “active substance” under the SPC Regulation on Plant Protection Products (Regulation (EC) No. 1610/96), which corresponds to the term “active ingredient” under the SPC regime for medicinal products. It ruled that a safener can be an “active substance” and so the subject of an SPC. Safeners have an auxiliary function insofar as they reduce the toxic effects of other active substances. The reference had been made by the German Federal Court in Bayer CropScience (C-11/13) in relation to Bayer’s application for an SPC for a safener (isoxadifen), the German court being uncertain as to whether an indirect effect on plant health was sufficient for a substance to

qualify as an active substance under the legislation. On 19 June 2014, the CJEU followed the AG’s opinion and held that the terms “product” and “active substance” should be interpreted as covering a substance intended to be used as a safener “where that substance has a toxic, phytotoxic or plant protection action of its own”. It was not necessary to make a distinction between direct and indirect modes of action.

Damages

Compensation pursuant to Cross-Undertakings

If the English court grants an interim injunction to a patent holder to hold the ring until trial, the patent holder will invariably be required to provide a cross-undertaking to compensate the defendant if it later transpires that the injunction should not have been granted. In 2010, AstraZeneca (AZ) obtained an interim injunction against Krka to prevent it selling Emozul capsules which were alleged to infringe a patent relating to esomeprazole with a particular optical purity60. A cross-undertaking in the usual terms was given by AZ for Krka’s benefit. Following a finding of non-infringement in a parallel case, AZ withdrew its infringement claim in the summer of 2011 and the injunction was lifted. Krka claimed £32M in damages to compensate it for lost profits for the period of the injunction as well as future losses attributable to the loss of its “first mover advantage” (i.e. the advantage it would have had as the first generic in the market). Significantly, this was not a case involving simple generic substitution as Emozul was a capsule whereas AZ’s esomeprazole medicine was a tablet formulation so pharmacies could not simply substitute the two products. Nevertheless Sales J accepted evidence adduced from Krka that Emozul would have made substantial inroads into the UK market and the parties were directed to agree the amount payable substantially according to the proposal put forward by Krka61.

2014 was marked by the closing of another chapter in the litigation between Servier and Apotex in relation to Servier’s patent for alpha-crystalline perindopril, the Supreme Court giving judgment in Servier’s appeal on the ex turpi causa principle. To recap briefly, the case concerns damages claimed by Apotex under a


10

67 Celanese v BP [1999] RPC 203 at §5168 [2014] EWHC 2924 (IPEC)69 [2014] EWHC 3234 (IPEC)

62 [2011] EWHC 730 (Pat)63 [2008] EWHC 2347 (Ch)64 [2012] EWCA Civ 59365 [2014] UKSC 5566 Taylor v Chester, LR4 QB 309 per Mellor J at §16

cross-undertaking given by Servier, Apotex having been injuncted for a period before the patent was revoked. Arnold J had ruled62 that the award of damages, assessed by Norris J at £17.5 million63, and paid by Servier, should be refunded by Apotex because the damages were claimed in respect of lost sales of a product whose manufacture in Canada infringed Servier’s Canadian patent. He applied the legal principle ex turpi causa non oritur actio (no action can arise from an illegal or immoral act) to prevent Apotex from relying on its own wrongdoing in Canada to make the claim in England in respect of the same product as a matter of public policy.

The Court of Appeal reversed Arnold J’s decision64. It held that although the ex turpi causa rule was engaged, it should be applied proportionately to the illegality involved. Apotex had acted in good faith in Canada, believing the Canadian patent to be invalid and hence its actions were “low on the scale of culpability”. Comments were also made to the effect that Servier should not occupy the higher ground, having enjoyed a highly profitable but unjustified monopoly to the detriment of the National Health Service. Apotex had also conceded before the hearing that it should discount the damages payable by Servier by an amount equal to that which the Canadian court would have ordered by way of damages payable by Apotex for the infringement in Canada.

Lord Sumption gave the leading judgment in the Supreme Court65. Having reviewed the facts of the case and the legal authorities on the ex turpi rule, which as early as 1869 was said to be “as thoroughly settled as any proposition of law can be”66, he decided only one issue: whether patent infringement counts as “turpitude” for the purpose of the defence. The answer was no. Turpitude requires breach of a law relating to the public interest, namely criminal and quasi-criminal law. Patent infringement is a tort and tort is concerned with the obligations of private parties. As such, it does not engage the public interest. There was no need for a value judgment on the significance of the illegality and the injustice of barring a claim on account of it. Lords Mance and Toulson, also giving reasoned judgments, agreed, thus bringing to an end this phase of the damages inquiry under Servier’s cross-undertaking.

Damages in Canada, and hence the value of the actual payment to be made by Servier in England, remain to be determined.

Account of Profits

When evaluating an account of profits following a finding of infringement, consideration must be given to whether the profits made were caused by the infringement. The classic illustration of this principle was provided by Laddie J in his “whistle on a battleship” example67: for a patented whistle, profits cannot be claimed on the sale of a battleship to which the whistle is attached. In Abbott v Design & Display68, HHJ Hacon had to decide whether the profits on the sale of MDF panels were attributable to the patent for a snap-fit aluminium insert. He held that they were - not because the panels incorporating the insert were infringing the claims (like the battleship) but because the inventive concept of the patent was partly embodied in the shape of the slot into which the insert snapped.

Two interesting points arose subsequent to the account69. First, it transpired that the main defendant had given inaccurate information about the scale of the infringement such that the £500,000 IPEC cap on compensation recovery would have been breached. HHJ Hacon refused to allow the cap to be exceeded, even though there were multiple defendants between which the profits could be apportioned. The rule is that the cap applies to the proceedings as a whole and only the parties can agree to lift it. If a claimant anticipates recovering more than the cap allows for each defendant, each should be sued separately and the actions heard together. The second point of interest is whether an additional amount under CPR Part 36.14(3)(d) should be subject to the cap, or whether it can take the total award above the cap. The judge was clear that in this situation the cap can be exceeded – the incentive under Part 36 is a powerful tool for settlement which is important in the IPEC.

Costs

Those unfamiliar with patent litigation may be surprised at what Birss J called “the familiar dispute which now arises on a regular basis at the end of nearly every patent


75 [2014] EWHC 2149 (Ch)70 ASSIA v BT [2014] EWHC 80 (Pat) at §4671 See footnote 70, ante72 [2014] EWHC 475 (IPEC) 73 [2009] EWHC 181 (Pat)74 [2014] EWHC 1647 (Pat)

case. Who is the winner?”70. Such is the complexity of a typical dispute that it is rarely easy to answer this question with absolute certainty. In ASSIA v BT71, two patents were at issue and three different systems were assessed for infringement. ASSIA won on the validity of both patents and won on the infringement of one system. However, if the patents were taken separately, the non-infringement findings made BT the winner in relation to one patent. Nevertheless, Birss J crowned ASSIA the overall winner as there were too many costs in common to both patents to divide the case. Instead, BT’s success on the infringement cases was acknowledged with a discount of 40% on ASSIA’s costs.

No review of 2014’s patent cases would be complete without mention of Brundle v Perry72 in the IPEC. It is the costs decision which is most notable, for here HHJ Hacon provides the startling revelation that not only did litigant in person Mr Perry use “intemperate language and expletives in his pleadings and in his skeleton argument”, he also sent a letter to himself (copied to all, including the court) following judgment but before the costs hearing in which he purported to be the judge. An extract from this letter forms our quotation of the year. Taking a tolerant and generous approach, HHJ Hacon considered that no-one would take the letter (or Mr Perry’s behaviour generally) seriously and although “strikingly unusual”, it did not reach the level of an abuse of process necessary to trigger the lifting of the costs cap and scale costs in the IPEC. Instead, Mr Perry was ordered to pay an extra £2,000 in costs to each defendant.

Employee Compensation

The law on inventor compensation under Section 40 of the Act was dealt with comprehensively by Floyd J (as he then was) in Kelly v GE Healthcare73. Readers will recall in that case that the patents protecting GE’s Myoview invention were held to be of outstanding benefit and that inventors Kelly and Chiu were awarded 2% and 1% of the benefit derived from the patents, respectively, as a fair share. The decision in Kelly set a benchmark and it is not surprising that any subsequent decisions on employee compensation should make frequent reference to it. One such case is Shanks v Unilever74 in which Arnold J dismissed

an appeal by Professor Shanks from a decision of the UKIPO refusing him compensation. The hearing officer had held that Unilever’s gross benefit from Shanks’ patents concerning biosensors for measuring blood glucose levels amounted to £24.5 million, which was not “outstanding” within the meaning of Section 40(1) of the Act. Had it been, a “fair share” under Section 41(1) would have been 5%. Both parties challenged the hearing officer’s calculation of benefit and fair share.

The appeal decision contains several notable points but one of the most interesting concerned the question of whether the benefit was “outstanding”. Shanks argued that to find otherwise would be to decide that Unilever was too big to pay: benefit should not purely be relative to size or a large multinational company would very rarely experience a benefit to be outstanding. Arnold J refused to disturb the Hearing Officer’s decision on this point as a multifactorial assessment had been made, as required by statute. However, Arnold J did confess to being troubled by one feature of the Hearing Officer’s assessment of what constituted outstanding benefit: it was a striking feature of the case that Unilever obtained a considerable rate of return for very little commercial risk. Unilever did not exploit the patents through manufacture but by licensing, in negotiations over which it had a strong bargaining position. Nevertheless, Arnold J acknowledged that the Hearing Officer had taken the risk-to-return point into account; the weight given to it was a matter for the Hearing Officer alone. There being no change to the finding of outstanding benefit, the appeal was dismissed.

Patent Attorney negligence

Judgments on the topic of negligence in the patent professions are rare but in July 2014, Birss J gave a thorough and detailed decision on the question of patent attorney negligence in the case of Baillie v Bromhead75. In some 200 paragraphs, the judgment describes the chronology of events charting the decline in the relationship between patent attorney and client, a glow-in-the-dark plastics business. Birss J then made his findings on the allegations of negligence and loss. No negligence of any consequence was found but the attorney was scolded for his conduct, criticism of which by the claimant was held to be well-founded.


12

79 Koninklije v Nintendo [2014] EWHC 3177 (Pat) 80 Gerber v Lectra, an unreported decision of Aldous J on 5 May 1993; Kirin-Amgen’s Patent [2002] EWHC 471, [2002] RPC 43; Zipher v Markem [2007] EWHC 154, [2007] FSR 18.

76 Starsight v Virgin [2014] EWHC 8 (Pat) 77 [2014] EWHC 2196 (Pat)78 Nampak Plastics Europe Limited v Alpla UK Limited [2014] EWCA Civ 1293

Inevitably, the case turns on its facts but practitioners may be interested to note that, for example, the attorney was held to be negligent in overlooking prior art cited in third party observations on the EPO Register (albeit there were no consequences), but failure to request transfer of the file from an earlier adviser, which would have revealed the existence of an earlier priority document, did not amount to negligence.

Procedural Issues

Summary Judgment

It is an established principle in the English courts that summary judgment is rarely appropriate in patents cases. This is because to determine the issues at trial, the court will need to adopt the mantle of the skilled person which in turn requires expert evidence as to, among other things, the common general knowledge. This principle was applied by Arnold J in dismissing an application for summary judgment on the issue of infringement made by Virgin against Rovi76. In making his decision, Arnold J also took into account the delay in bringing the application, the fact that the trial was close and the fact that if the summary judgment decision were appealed, this could lead to procedural complications.

Nevertheless, bucking the trend, Birss J awarded summary judgment in Alpla’s claim for a declaration of non-infringement against Nampak77 and this was subsequently confirmed by the Court of Appeal78. Nampak had previously commenced infringement proceedings based upon its patent relating to plastic milk bottles. Noting the caveats about expert witnesses mentioned above, the Court of Appeal nevertheless confirmed that in simple cases, such as this, the warnings from the authorities did not apply with such force. It was noted by Floyd LJ that it was necessary for the party resisting such an application to identify, perhaps in only quite general terms, the nature of the evidence of the common general knowledge which it proposes to adduce, and to be in a position to explain why that evidence might reasonably be expected to have an impact on the issue of construction. If that party is not able to do so, it is open to the Court to conclude that he is simply hoping that “something may turn up”. As there were no terms of art to contend with,

the issue of construction turned solely on the reading of the specification. Floyd LJ was unable to see how the conclusions drawn from this reading could be shown to be erroneous by subsequent evidence. The Court of Appeal construed the claim, which related to the length of the sides of a milk bottle and the nature of its truncated corners, and agreed with Birss J’s assessment that Alpla’s product did not infringe.

Partial Invalidity

Following his finding that two of Philips’ patents were infringed but partially invalid, Birss J surprisingly allowed one of the invalid claims to remain in force and on the register79. He did so because the proposed deletion of the invalid claim may have altered the construction of the valid and infringed claim in such a way as to render it invalid. The Court was adjudged to have the jurisdiction to make such a decision under Section 63 of the Act and noted that the authorities reinforced this conclusion80. While the Patents Court would in almost all circumstances require the deletion of an invalid claim, the Court had jurisdiction to grant relief on a partially invalid patent without requiring the deletion of the invalid claim. However, it was noted that this would require truly exceptional circumstances. Birss J did not consider that an outcome whereby the patent was revoked in its entirety was fair unless there was no sensible alternative. Here, although the proposed course would leave a patent on the register that contains invalid claims, this problem was mitigated by two factors. The first of these was that the Comptroller would place a copy of the Court’s order on the register. Secondly, Philips gave an undertaking not to assert the invalid claims, grant exclusive licences in respect of the invalid claims or assign the patent, thereby providing sufficient safeguards to ensure that it would not obtain any unwarranted advantage.

Amendment Applications

Applications to amend patents are relatively common in English patent litigation. Amendments can be unconditional, where the patentee informs the challenger (often for the purposes of the UK litigation only) that it will not contend that the claims as granted are valid, or conditional, where the patentee contends that the patent


86 [2012] EWHC 3316 (Pat)81 Compact GTL v Velocys [2014] EWHC 2951 (Pat)82 See footnote 20, ante83 T307/07 and T1423/0784 [1994] RPC 23185 See footnote 8, ante

is valid as granted but seeks to introduce a fall-back position in case the Court finds otherwise. In 2014, Arnold J confirmed that a conditional amendment to narrow claims will only be allowed if the Court finds the claims as granted to be invalid81. If a broader claim in the patent as granted is held valid, a patentee will not be permitted to amend the patent to include one or more subsidiary claims.

Faced with a proposed amendment application under Section 75 of the Act, Birss J had to consider the rules relating to double patenting in Philips v Nintendo82. Nintendo submitted that even if the amendments found were to be formally allowable, they should nevertheless not be permitted by the Court in an exercise of its discretion. Having reviewed a number of UK and EPO authorities83 including Marley’s Roof Tile84, the Judge derived a number of general principles. Amongst others, these included noting that, whilst double patenting was not a ground for revocation under Section 72 of the Act, the EPO does recognise it as a ground for rejecting amendments to a divisional application provided that the proposed amended divisional claim has the same scope as an independent parent claim and the applicant has no legitimate interest in obtaining both claims. One circumstance in which a legitimate interest could exist is if the divisional claim is broarder than, but encompasses, the parent claim. In considering Marley’s Roof Tile, the Judge noted that this judgment was directed to the construction of Section 73(2) of the Act and was not binding in relation to the exercise of discretion under Section 75 of the Act. It was adjudged that, as a matter of UK law, a double patenting objection could be taken as a ground for refusing a post-grant amendment but should be done so only in limited circumstances.

Jurisdiction

On 15 May 2014, Arnold J handed down a decision85

which may well come to be recognised as marking a paradigm shift in the practice of the English Patents Courts. The case involved pemetrexed, a drug developed by Eli Lilly and used in cancer treatment. Actavis sought to clear the way ahead of the expiry of SPCs for the compound pemetrexed in late 2015. Specifically, Actavis applied for a declaration of non-infringement (DNI) in respect of a Lilly patent to

pemetrexed disodium in combination with vitamin B12 and optionally a folic protein binding agent. Actavis wished to market pemetrexed diacid, pemetrexed dipotassium or pemetrexed ditromethamine. Most interestingly, and somewhat unusually, Actavis sought a declaration from the English court not only in respect of the UK patent but also the French, German, Italian and Spanish designations of the same European Patent. Readers will recall that in late 2012, Arnold J had decided that the English court had jurisdiction to hear the actions and this had subsequently been upheld by the Court of Appeal86.

Having heard the evidence at the main trial, Arnold J applied UK, French, Italian and Spanish law to decide on the question of infringement, considering national laws on claim construction, the doctrine of equivalence, and direct and indirect infringement. As to which law was applicable to the question of whether Actavis was entitled to a DNI in relation to a non-UK patent, Lilly argued that (according to the Rome II Regulation) relevant national law should apply. Arnold J held that rules on grant of DNIs were procedural and, therefore, the applicable law was determined not by Rome II but by English private international law, under which the applicable law was the lex fori, i.e. English law in this case. However, Arnold J held that, in any event, Actavis would have been entitled to DNIs under each of the relevant national laws.

In assessing infringement, a key question was whether the term “pemetrexed disodium” in the patent’s claims should be interpreted to mean only the disodium salt or whether other pemetrexed salts would be covered. Actavis referred to the prosecution history of the patent, noting that during prosecution, to overcome objections, Lilly had amended the term from “an antifolate” to “pemetrexed” and subsequently to “pemetrexed disodium”. Arnold J held that, although courts should be cautious before applying prosecution history to aid claim construction, in this case it was relevant. He also considered both Article 69 EPC (and its Protocol) and the UK Improver questions, and concluded that the scope of the claim was to be interpreted as covering only pemetrexed disodium, and that there was no direct infringement under UK law. He also dismissed Lilly’s argument that there was indirect infringement due to the


14

90 [2014] EWHC 2611 (Pat)91 [2014] EWCA Civ 25092 [2013] EWHC 468 (Pat)

87 [2013] EWCA Civ 149688 [2013] UKSC 46, [2014] A.C. 16089 Actavis v Pharmacia [2014] EWHC 2265 (Pat)

presence of sodium ions after reconstitution or dilution in saline of Actavis’ proposed products. Applying the other national laws, Arnold J’s final conclusions were the same.

Actavis’ application had included a request for grant of a DNI in respect of the German designation of the patent (as well as the UK, French, Italian and Spanish designations). Arnold J noted that, although the English court was first seized, Lilly had brought and pursued an infringement claim in Germany and, on 3 April 2014, the Düsseldorf Regional Court held that use of pemetrexed dipotassium would infringe the German designation. Arnold J said he was unable to agree with that conclusion; apart from the evidence and arguments before the German court being different in some respects, he disagreed with the court’s reasoning on some issues, including its application of Article 69 and the Protocol and its disregard of the prosecution history.

Stays and EPO oppositions

As many readers will know, in considering whether to grant a stay of proceedings whilst an EPO opposition is pending, the English Court will look at the guidelines given by the Court of Appeal in IPCom v HTC87 in late 2013 following the observations of the Supreme Court in Virgin v Zodiac88 earlier that year. These IPCom Guidelines, which empower the Court with a very wide discretion in order to further the balance of justice, suggest that a stay should be the default option, but that commercial certainty may favour allowing the UK proceedings to continue, which may be the more usual outcome.

In 2014, an interesting debate on the issue of stays was played out before Arnold J89. Actavis issued proceedings seeking to revoke Pharmacia’s patent for sustained release dosage forms of pramipexole. The proceedings were commenced before the window for EPO opposition proceedings had closed. Pharmacia applied for a stay of the English proceedings pending the outcome of the EPO proceedings. The evidence before Arnold J was that the English proceedings would be decided at first instance within twelve months and on appeal within two years. On the other hand, the

EPO proceedings would be likely to take over five years to resolve without expedition and probably at least three years with expedition. For its part, Pharmacia offered undertakings: (a) to seek expedition of the EPO proceedings; (b) not to seek an injunction against Actavis or its customers until the determination of the EPO proceedings; and (c) only to seek damages of 1% of Actavis’ net sales until the end of the EPO proceedings. Despite the undertaking not to seek an injunction during the period of the opposition and the evidence that the damages payable to Pharmacia if the patent were held valid and infringed would be modest compared to the costs of the English proceedings, the Judge refused a stay. Although Arnold J considered the issues were “finely balanced”, he held that the key reason for his decision was that the EPO opposition proceedings had only just begun and that commercial certainty for the parties as regards the UK would be achieved much more quickly if the stay was refused.

In a postscript to the above judgment, Arnold J noted that after he had released his judgment in draft, the patentee, Pharmacia, had offered two additional undertakings to Actavis, the potential infringer, in return for a stay: (i) not to seek an injunction in the UK against Actavis and its customers for the life of the patent and; (ii) only to seek damages of 1% of Actavis’ net sales in the UK during the life of the patent. In a further judgment90, Arnold J held that the additional undertakings tipped the balance in favour of a stay as they “substantially eliminate the commercial uncertainty to which Actavis will be exposed in the United Kingdom as a result of a stay.” Arnold J also expressed the hope that the EPO would accede to the joint request of the parties to accelerate the opposition and any appeal.

Stays and EPO central limitations

2014 was the first year in which the Courts had to grapple with the issue of staying English proceedings pending the conclusion of a central limitation application at the EPO. This question first arose before the Court of Appeal in Samsung v Apple91. Following judgment from Floyd J a year earlier that its patents were invalid92, Samsung appealed and the order to revoke the patents was stayed pending the outcome of the appeal, provided that it was pursued with due diligence. Several months


99 [2014] EWHC 3115 (Pat)100 [2008] EWCA Civ 622101 Teva v Boehringer Ingelheim [2014] EWHC 3186 (Pat)102 Consafe Engineering (UK) Ltd v Emtunga UK Ltd [1999] RPC 154 and Alfred Taylor v Ishida (Europe) Ltd [2000] FSR 224

93 See footnote 88, ante94 [2000] RPC 631 95 Case No. C07.085HR96 [2013] UKSC 1697 Kennametal v Pharmet [2014] EWHC 1438 (Pat) 98 [2014] EWHC 1793 (Pat)

later, Samsung filed central amendment applications at the EPO relating to the two patents in suit with a request that the applications be accelerated, and sought an adjournment of the appeal pending their determination.

The Court allowed Samsung’s application to adjourn the appeal. Kitchin LJ, giving the leading judgment, held that the filing and pursuit of the applications by Samsung during the pendency of appeal proceedings could not be characterised as an abuse of process and was not an activity with which the Court could properly interfere. If the proceedings continued there was real possibility of the amendment process completing and the appeal being heard on the basis of claims deemed to never have existed, the amendments having retrospective effect. In coming to this decision, the Court took support from the Supreme Court’s decision in Virgin93 and that of the Court of Appeal in Palmaz94. Further, as this was the first time this issue has arisen in the UK, the Court considered the approach taken by the Dutch Supreme Court in Boston Scientific v Medinol95. There, following completion of central amendment proceedings, the Dutch Supreme Court allowed the appeal against revocation and remitted the case back to the Court of Appeal for rehearing on the basis of the amended claims. Kitchin LJ adopted Lord Neuberger’s comments in Schutz v Werit96 in relation to the German Supreme Court, noting that the Dutch Supreme Court was another court whose decisions must be accorded considerable respect.

Kennametal attempted to rely upon the decision in Samsung in its application that an order to revoke its patent be stayed pending the outcome of its central amendment application97. In contrast to Samsung, Kennametal had chosen not to appeal the decision that the patent was invalid. Henry Carr QC (sitting as a deputy judge) noted that in the absence of an appeal, his first instance decision was final so no concurrent UK proceedings existed. He stated that this was a completely different case from Samsung and refused Kennametal’s application.

The question of whether to adjourn a first instance trial in light of a central amendment application was put to the Court in Rovi v Virgin98. In this case, Rovi had dropped its infringement case so that only the issues of validity

remained live and, approximately six weeks before the scheduled trial, it applied for central amendment. Mann J granted the application for adjournment, considering that the pertinent factors from Samsung operated in this case just as strongly. A major factor in his reasoning was that certainty could not be achieved until the EPO had ruled on the amendment application. Further, it was of particular importance that infringement was no longer in issue. Virgin’s application to require Rovi to make the amendment application under Section 75 of the Act failed, as it could not demonstrate that there was any jurisdiction to order that a patentee must use the English route for an amendment. In allowing the trial to be adjourned, the Judge was clear in stating that a patentee cannot assume that the self-generated uncertainty arising from a late central amendment application will always and easily end in the same result.

Expedition

It is a feature of pharmaceutical patent litigation that generics plan launches carefully and significantly in advance. Birss J noted that this is an important factor when deciding on a request by a generic for expedition, suggesting that it will be difficult to succeed on such requests. However, in Generics v Warner Lambert99, making reference to the relevant principles for deciding requests for expedition set out by Lord Neuberger in Gore v Geox100, he granted a request for expedition by the generic on the basis that (i) the matter was genuinely urgent because the value of the market was high and the patent a real obstacle and (ii) there was space in the court list during the period in which the expedited trial was sought. However, in another case101 where there was likely to be significant disruption to the list, he refused the request.

Disclosure

As many readers will know, in English patent litigation, a party alleged to infringe a patent can avoid giving disclosure on the issue of infringement if it elects to serve a product or process description instead. The rules of Court provide that such description must contain “full particulars of the product or process alleged to infringe.” The regime came into force in the late 1990s and, in two early decisions, Pumfrey J102 made it clear that


16

106 Adaptive Spectrum v BT [2014] EWCA Civ 1513 107 [2009] EWCA Civ 1513108 Directive 2004/48/EC109 [2012] EWCA Civ 1339110 [2014] EWHC 979 (IPEC)111 The authors are grateful to Bristows’ partner Alan Johnson for writing this part of the review.

103 See footnote 98, ante104 Wobben v Siemens [2014] EWHC 3173 (Pat) 105 [1974] AC 133

product/process descriptions must be complete in all relevant areas and provide sufficient particulars to enable all issues of infringement to be resolved. Fifteen years on, Arnold J reminded the parties in Rovi v Virgin103 that Pumfrey J’s observations are as relevant today as they have always been and that the description needed to contain information of at least the specificity of the claims and often more information if the claims are expressed in general terms. The Judge underlined the need for the parties’ advisers to cooperate with one another in relation to product/process descriptions and was critical of the “debacle” that had arisen from the parties’ conduct in this respect.

In an unusual decision concerning a wind turbine patent, the alleged infringer Siemens was ordered by Morgan J to disclose details of its customers to enable Wobben to join them to the action, commence separate infringement proceedings, or to enable them to agree to be bound by any decision against Siemens104. Following the principles from Norwich Pharmacal105, the Court had little difficulty in finding that there was an arguable case of infringement (or threatened infringement, the relevant technology not so far having been universally activated), that Siemens would have facilitated that infringement, that Siemens had the information requested and that Wobben needed this information to bring proceedings. The decision to grant Wobben’s request turned upon the question of whether it was just to do so. In weighing this in the balance, the Judge held that the reputational damage Siemens said it would suffer as a result of Wobben pursuing its customers on the basis of the information provided was less than the damage it would incur if Wobben proceeded as it was entitled to do. Siemens could indicate which customers had activated the technology, such customers being a subset, whereas without that information, Wobben may have no choice but to pursue all customers indiscriminately based upon its own enquiries as to their identity. The Judge noted, obiter, that in other cases it might be pertinent to request that if the Court is minded to grant such an order, the patentee give an undertaking in damages which could later be enforced if the infringement claim failed and the order requiring the disclosure of information caused damage to the defendant.

Final Injunctions

Following a finding that Adaptive’s patents were valid and infringed, the Court of Appeal granted a stay of the final injunction to allow BT to make alterations to its systems to render them non-infringing106. The Court was narrowly persuaded to grant the stay by factors including Adaptive’s statement that, as its business lies in licensing, its main focus was a financial remedy. However, it noted that this did not mean that Adaptive was not entitled to seek an injunction. The stay was ordered on terms that BT pay 10% of its estimated weekly revenue from the infringing products for each week of the stay but, following Virgin v Premium107, it was stressed that this would amount to an interim payment on account of damages rather than reflecting an accurate level of royalty payment. It rejected BT’s request that Adaptive should provide a cross-undertaking in damages in relation to the final injunction to cater for the possibility that its patents could later be revoked or materially amended by the EPO. A subsequent EPO decision would render the injunction ineffective at that point but not ab initio, so there would be no reason for Adaptive to pay for the harm suffered during the period that the injunction was rightly granted.

Dissemination of Judgments

Although Article 15 of the IP Enforcement Directive108 makes no mention of publication of judgment in cases of non-infringement, Jacob LJ used Recital 27 of the Directive and particularly the need to “contribute to the awareness of the public at large” to order publication of a decision of non-infringement in Samsung v Apple109 on the basis that there existed a need to dispel commercial uncertainty in the marketplace. It is notable that HHJ Hacon followed the same approach in Brundle v Perry110, thus bolstering the foundation for orders of this nature in the future.

Unitary Patent/Unified Patent Court111

Looking back at last year’s piece on the UPC, your author must confess to having been guilty of unjustified optimism as to the pace at which events would take place in 2014. Indeed, in large part, this year’s summary


of what to expect in 2015 could be summarised as: “see last year’s report”. But that is explicable when one considers that 12 months ago, official predictions were that the system would be in place in “early 2015”, whereas now the official position (and one which is widely regarded as optimistic) is “not before end 2015”, i.e. early 2016. So why is this? What has happened in 2014, and what lies ahead for 2015?

The Past Year

The major news in 2014 was perhaps that the Advocate General of the CJEU advised that the so-called second Spanish challenge (to the legality of the Unitary Patent and Language Regulations), should be rejected. The Court follows AG’s opinions in around 80% of cases, and given that it seems unlikely that the Court would strike down a system 40 years in the making, we can hopefully (almost) put to one side the possibility that the system could now be thrown out by the Court.

At a more mundane but nonetheless vital level, the Rules of Procedure moved on again to a 17th draft which was debated at a public hearing in Trier (Germany) in November. The meeting was both well-attended and well-run, and users can be in no doubt that their concerns have been listened to, if not always acted upon. There are still wrinkles in the draft rules, but for the most part, the major issues are settled.

There has also been progress on ratifications of the UPC Agreement by Contracting Member States, but not as much as one might have hoped. Of the “big three” only France has ratified. The UK needs to complete various tasks before it can ratify, and the target date of April / May 2015 has now slipped. More significantly, Germany is still barely out of the starting blocks on its progress toward ratification.

But ratification will probably not be the rate limiting step for the start date. Rather it will be practical issues.

Contrary to expectations (and certainly hopes) we did not see in 2014 any announcements about the selection of judges. What we saw instead was a realisation that some countries had no candidate judges with real patent litigation experience, and that basic training would be

required.

Neither was there anything about court fees, opt-out fees, unitary patent fees and the “distribution key” (the way the unitary fees would be shared between participating states). On the plus side, the debate on these issues was started and progress has been made, in particular, on court fees.

We also got to know more about individual Member States’ plans for local and regional courts in the UPC. Sweden and the Baltic states reached agreement to set up a regional division. Rumours have circulated that regional divisions will be hosted by two more groups of states: Romania, Bulgaria, Greece and Cyprus; and the Czech and Slovak Republics. Slovenia and Hungary may join them also. However, nothing has been announced, and recent comments from Dr Margot Fröhlinger suggest that negotiations to form these regional divisions may have broken down. There will be more local divisions than at first seemed likely. Austria, Ireland, Denmark and Finland will all have their own local divisions in addition to the UK, France, Germany, The Netherlands, Belgium, and Italy. The bigger news was on languages in these divisions. Subject to Rule 14.2(c) being agreed to permit more flexibility during proceedings, it seems that the German and French local divisions will offer English as a working language. This makes it likely that every local and regional division will offer English as an option to Claimants.

Most significantly of all, progress was made toward setting up the court computer system, including a demonstration filing system. This is now on track.

2015?

The theoretical aim of the Preparatory Committee remains to have completed its work by 31 December 2015 so that the new system can go live from April 2016. This is ambitious, the timetable for the computer system especially so. But we can expect (in roughly this order):

• News as to the final form of the Rules on rights of representation and training for patent attorneys;

• Proposals for court and opt out fees and a public

18

consultation;

• An 18th and final set of draft rules of procedure (possibly in April);

• Proposals for unitary patent fees and (perhaps at the same time) a distribution key (possibly in June – though that may be optimistic);

• More news about judges – with candidate judge training beginning in the spring; and

• More ratifications – with the UK ratifying by the end of the year, albeit holding back its instrument of ratification and depositing it only when everything else is ready.

Of course there could also be spanners in the works. The CJEU could yet declare the Unitary Patent and Language Regulations unlawful. If so, that would kill the project for at least a few years. Deadlock could be reached on the distribution key – this was, after all, one of the reasons an earlier incarnation of the dossier failed in 2003/04. The UK might hold an EU in-out referendum in 2017 if the Conservatives are elected in the May General Election. Were that to happen and the referendum scheduled for 2017 as expected, it would seem exceptionally unlikely that the system would be brought in before then since an “out” vote would scupper the system for all parties given the UK’s central role. Finally, and perhaps most intriguingly, Germany may decide not to ratify. That may seem unlikely, but it is understood that Germany will need a cost-benefit assessment of some type before it can ratify. Since the system seems likely to be expensive for Germany, it is at least possible that there could be a hiccough here.

Other Issues and Overall Timings

Finally, what are the big issues remaining, and when can we realistically expect to see the system coming into operation?

As to issues, attention continues to focus on the transitional provisions and especially the opt out procedure. Users are pushing for clarification as to the practical implications of the transitional regime and

whether the opt out will really be effective. One relatively new sub-issue is who must file the opt out? We have long known that it must be all “proprietor(s)”, but who is that exactly? The current relevant Rule of Procedure says it is the person(s) appearing on the EPO register, but that may well not reflect the true ownership, given the number of intra-group assignments, not to mention arm’s length assignments which are not recorded after expiry of the opposition period and after which the EPO register becomes redundant. This rule must be changed.

There are also remaining concerns about SPCs for unitary patents. There has been silence on this topic for over a year. Although there is a working assumption that national SPCs will be available based on unitary patents, surely all pharmaceutical companies will be concerned at this lacuna in announcements, and keep unitary patents off their agenda until there is some clarity on this. After all, the SPC period is, for many drugs, the most important period of protection.

And last of all, what of timings? What does the crystal ball look like? In truth, it is distinctly misty. The current target of April 2016 for start up is surely impossible. October 2016 is a more credible target which has been discussed, but it is just that: a target. Hence 2017 is more realistic, and your author’s current best guess is second half 2017. To be predicting anything earlier would be a triumph of hope over experience, and since that slot coincides with the UK holding the EU Presidency, one might think the UK would do its utmost to prevent the project slipping into 2018. There remains, however, (still) a lot to do, and it is better done properly than in haste. Happily, the Preparatory Committee seems to have this firmly on board.


Review of Patent Cases in the English Courts in 2014

112 Warner-Lambert v Actavis [2015] EWHC 72 (Pat)

19

Looking Ahead

2015 is likely to witness the following important developments:

• It is rumoured that a new Patents Court Judge will be appointed in the autumn to relieve some of the pressure that is currently on the Court lists;

• The Court of Appeal will review several important decisions mentioned in this review including the Actavis v Lilly pemetrexed case and the damages award in AZ v Krka. (In relation to the Actavis v Lilly case, it will be interesting to see if the Court of Appeal considers that the Improver questions are still in step with modern patent law);

• Issues concerning second medical use patents will arise in the English Courts. Already in 2015, Arnold J has issued an important decision concerning a skinny-label generic version of Pfizer’s Lyrica medicine112; and

• Although 2014 was comparatively quiet for SPCs, these rights are so valuable to originators in the pharmaceutical industry and the law is so uncertain that it is improbable that there will not be further disputes in 2015.

20

The Authors

01 Brian Cordery Partner

Brian specialises in contentious intellectual property matters. He has been involved in numerous patent disputes, mostly concerning pharmaceuticals but also in other fields. His first Patent Review of the Year was published in 2001 and writing the review has occupied the long winter evenings for him ever since.

02 Dominic Adair Partner

Dominic is a partner in the Intellectual Property Department at Bristows. His practice mainly involves patent litigation and particularly the co-ordination of international patent litigation. Prior to joining Bristows, Dominic obtained a PhD in Zoology.

03 Naomi Hazenberg Associate

Naomi specialises in intellectual property litigation. Although she has experience with a broad range of IP rights, her practice is largely focused on patent disputes. She has a particular interest in technical and telecoms matters having obtained a masters degree in Physics prior to joining Bristows.

04 Sylvia Delbeuf Paralegal

Sylvia works as a paralegal in the Intellectual Property Department at Bristows. She is part of a team which is responsible for the co-ordination of the defence of the intellectual property of a major pharmaceutical product, in which she assists with the management of litigation in a number of jurisdictions including several other European countries as well as Asia and South America.

01

02

03

04


100 Victoria EmbankmentLondon EC4Y 0DHT +44(0)20 7400 8000F +44(0)20 7400 8050

bristows.combristowsupc.com

This paper contains material sourced from responsibly managed and sustainable commercial forests, certified in accordance with the FSC (Forest Stewardship Council). It is produced using recyclable 100% ECF (elemental chlorine free) pulp; an environmentally friendly bleaching process.

review of patent cases in the english courts in 2014 patent review_final lres.… · case –...

Documents