[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> GW Pharma Ltd & Anor v Otsuka Pharmaceutical Co. Ltd [2022] EWCA Civ 1462 (08 November 2022) URL: http://www.bailii.org/ew/cases/EWCA/Civ/2022/1462.html Cite as: [2022] EWCA Civ 1462

[New search] [Printable PDF version] [Help]

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
INTELLECTUAL PROPERTY LIST (ChD)
PATENTS COURT
Mr Ian Karet sitting as a Deputy Judge of the High Court
HP-2021-000039

		Royal Courts of Justice Strand, London, WC2A 2LL
		08/11/2022

This judgment was handed down remotely at 10.00am on 8^th November 2022 by circulation to the parties or their representatives by e-mail and by release to the National Archives. .............................

Lord Justice Birss:

1. This is an appeal from the 11 May 2022 order of Mr Ian Karet sitting in the Patents Court as a Deputy Judge of the High Court. The order dismissed the application of the defendants made under CPR Part 11 contesting the court's jurisdiction and seeking a stay of proceedings. In giving permission to appeal Arnold LJ noted that grounds 1-3 raised important issues as to the jurisdiction of the court to determine the validity of foreign patents, relating to the Moçambique principle, derived from British South Africa Co v Companhia de Moçambique [1893] AC 602 and the foreign act of state doctrine.

Background

2. The claimant ("Otsuka") is a Japanese pharmaceutical company. The defendants ("GW Pharma") are UK based R&D companies whose work is focussed on possible pharmaceutical applications of compounds derived from cannabis. There is no need to distinguish between the two defendants. In 2007 the parties entered into a Research Collaboration and Licence Agreement. Essentially the parties agreed to collaborate in the research and development of different cannabinoids as potential drug candidates for the treatment of nervous system disorders and cancer. Both parties had the option of obtaining ownership of a candidate and developing it further afterwards. The party who chose to do so would then pay the other one royalties on net sales of any product covered by patents arising from the collaboration.
3. The agreement is governed by New York law. It does not contain a jurisdiction clause. There is an arbitration clause (clause 15.2) under which the arbitration would proceed in London if Otsuka initiated the dispute, or in New York if GW Pharma initiated the dispute. There is a carve out from the arbitration clause, excluding issues of "patent scope, validity or infringement".
4. The collaboration ran from 2007 until 2013. Much of it took place in the UK, including at Reading University. At the end of the research period Otsuka elected not to pursue clinical development of any candidate product.
5. In June 2018 GW Pharma or their affiliates started obtaining marketing authorisations for a product known as Epidiolex or Epidyolex (depending on the country). Epidiolex is indicated for the treatment of seizures associated with various conditions or epileptic syndromes. The active ingredient in Epidiolex is the cannabis derivative cannabidiol. First marketing was in the USA, where the bulk of the sales have been so far. Marketing authorisations have now been granted in other places, including the UK, Australia, Switzerland and Israel. Worldwide sales so far have exceeded $1.4 billion. The product is manufactured in the UK.
6. Otsuka contends that Epidiolex is a product of the research collaboration and is subject to the agreement, so that substantial royalties are due from GW Pharma. In terms of the agreement, Otsuka's case therefore has two aspects. First Epidiolex is a "GW Pharma Product" as that term is defined in the agreement for a product of the collaboration which GW Pharma choose to develop and sell. Second Epidiolex is "Covered" by a "Valid Claim" of the relevant patents, both of which are again defined terms.
7. A "Valid Claim" means essentially a claim which is in force and has not been held to be revoked or invalid by a competent court or government agency, and "Covered" means essentially that an act of commercialisation such as manufacture or use of the product would infringe that claim but for a licence. It is not necessary to decide precisely what these terms mean in this appeal, but it is worth noting that these sorts of definitions are fairly common in international patent licences. They are similar to the terms of the licence in the judgment of Henry Carr J in Chugai Pharmaceutical Co. v UCB Pharma SA [2017] EWHC 1216 (Pat) which, at paragraph 39, the judge noted were in turn similar to the licence in Celltech v Medimmune [2004] EWCA Civ 1331. In Celltech Jacob LJ had held that the clause meant that royalties will be payable if the product does fall within the scope of the relevant claims regardless of its validity unless and until the patent is finally declared invalid; and as Jacob LJ also explained (at paragraph 13):

"To those unfamiliar with the world of patents this might sound irrational – why should royalties be payable (and apparently be unrecoverable) for a licence under a patent claim ultimately held invalid? There is a rational answer to this. Everyone knows that patent law can, at least at its edges, be uncertain – and that different results can arise in parallel litigation in different countries. By taking such a licence MedImmune have achieved a number of clear desirable commercial objectives: (1) they know they are free to develop and market products irrespective of the question of validity; (2) they are free so to do across the globe; and (3) their competitors may have to face litigation from Celltech in a variety of jurisdictions whereas they, MedImmune, stand in the shelter of the patents, valid or invalid. It is not irrational that they should pay for such benefits."

8. GW Pharma do not agree that royalties are due. They contend, amongst other things, that Epidiolex is not subject to the agreement at all because it is not a GW Pharma Product since it was developed independently, nor is it Covered by a Valid Claim of any of the relevant patents. As to the latter, GW Pharma contend that the relevant patent claims do not cover Epidiolex as a matter of claim construction and that the claims would be invalid if they were construed to cover Epidiolex. The second of these arguments is known as a squeeze. Finally GW Pharma also argue that certain claims of the relevant US patent (US 9,066,920 ("US 920")) are invalid in any event.
9. Before going further it is worth putting some of these arguments in context. If the interpretation of "Valid Claim" which is explained above applies in the present case then it has no impact on GW Pharma's arguments that the claims ought not to cover Epidiolex as a matter of construction nor on the squeeze arguments. However it would mean that there is no point in raising the final argument, which is a simple contention that the patent is invalid in any event. The point Jacob LJ was making in Celltech is that such an argument would not provide a defence to a contract claim and, if that really was the argument a licensee wished to rely on, they would need to go to the relevant national court and apply to revoke the patent. Once the national courts had finally ruled the patent was invalid then there would no longer be a Valid Claim, but not before that.
10. However in this case GW Pharma seek to rely on a decision of the US Supreme Court called Lear (Lear v Adkins 395 US 653 (1969)). As I understand it GW Pharma seek to deploy Lear at least in part as an antidote to a prospective argument from Otsuka based on Celltech that the definition of Valid Claim means that invalidity is not a defence to the contractual claim for royalties, especially given that the licensee can always go to the relevant local court and revoke the patent. Lear concerns the question whether under US law a licensee in a contract claim brought against them for royalties can defend the claim on the basis that the US patent is invalid. Thus if GW Pharma are right about Lear and its application in this case (given that the agreement is governed by New York law), it may mean that they would wish to bring a direct challenge to validity as part of their contract defence to the claim. I emphasise that it is not necessary to decide any of these points on Lear or the true construction of the licence, but for reasons I will explain below, it is worth illuminating the issues.
11. There were negotiations that commenced in May 2019 but they did not resolve the dispute. In June 2021, Otsuka commenced an arbitration in London claiming royalties. In August GW Pharma challenged the jurisdiction of the arbitration to determine issues within the carve out relating to patents. Accordingly Otsuka commenced proceedings in the Patents Court, with the claim form issued on 29 October 2021. The proceedings claim declarations that royalties are due under the agreement because the product is a GW Pharma Product and is covered by the relevant patents. The relevant patents are in the USA, UK, Germany, Switzerland, Australia and Israel. They consist of two families, including US 920, EP (UK) 2 448 637 and EP (UK) 2 661 263 (as well as the German and Swiss designation of the same EPs).
12. Proceedings were served within the jurisdiction, since the GW Pharma companies are UK companies (based in Cambridge). The acknowledgement of service indicating that GW Pharma would contest jurisdiction was served on 12 November 2021.
13. On 7 January 2022 GW Pharma commenced proceedings in the New York State Court. In the US Complaint GW Pharma contend that no royalties are due on the basis of the contentions I have summarised above. So far in the USA, Otsuka has filed a motion to dismiss on jurisdiction grounds, which awaits determination.
14. Also in January the UK arbitration was formally constituted with the agreed removal of the patent issues from it. The hearing to determine the remaining issues is set for October 2023.
15. In the UK, the jurisdiction application was listed to be heard in March 2022. Before the hearing GW Pharma provided a letter outlining the basis on which it would defend these proceedings if the jurisdiction challenge failed. The terms of the letter mirror the US Complaint. The jurisdiction application was heard by the judge in March 2022, with judgment handed down on 3 May ([2022] EWHC 1012 (Pat)). After that GW Pharma filed their Defence in these proceedings in May, adopting points from the earlier letter, and in June 2022 Otsuka filed a Reply. It is worth pointing out that as presently constituted all the relief sought in these proceedings is contractual in nature: orders to pay royalties and declarations couched in terms of the agreement.

The judgment

16. The judgment dismissing GW Pharma's application decided three issues: jurisdiction under the Moçambique principle, foreign act of state and a distinct application for a stay on forum non conveniens grounds.
17. On jurisdiction GW Pharma's case was that in their defence they would challenge the validity of foreign patents, particularly US 920, and so the court had no jurisdiction, applying the Moçambique principle as it applies today following the Supreme Court in Lucasfilm Ltd v Ainsworth [2012] 1 AC 208 and as applied and explained in Chugai by Henry Carr J. Paragraphs 62-79 of the judgment address this, holding that if the challenge to a foreign patent is not "direct" (as that term was used in Chugai) then the Moçambique rule is not engaged. The judge then applied those principles, holding (paragraphs 74-78) that that was the position in these proceedings for three reasons. First, GW Pharma's principal defence is that the product was independently conceived. If that (or an argument which may arise that GW Pharma are estopped from challenging validity) succeeds then validity of the patents is irrelevant. Second, if in due course issues of validity of foreign patents do arise which do engage the Moçambique principle then that problem can be avoided by case management, e.g. by determining the issues relating to the UK patents and staying the remainder of the dispute pending decisions in foreign courts. Third, a dispute of this kind should be capable of resolution somewhere, whereas the logic of GW Pharma's argument would mean that no other court, such as the US court, was able to do so. It is unlikely that the parties can have intended that when entering into the agreement.
18. On foreign act of state, the judge at paragraph 80 rejected GW Pharma's case that jurisdiction should also be declined because the grant of a patent is an act of state, holding that such a grant is not an act of state at all, citing Chugai.
19. On forum non conveniens, the case was argued on the conventional basis of the principles in Spiliada Maritime Corp v Cansulex Ltd [1987] AC 460. GW Pharma argued on various grounds that New York was an available forum where the case may be more suitably tried for the interests of all parties and the ends of justice, and that it was clearly and distinctly the more appropriate forum. The judge addressed this in paragraphs 81-92, holding that GW Pharma had not demonstrated that New York was a more suitable forum. Three reasons were identified in favour of the Patents Court (at paragraph 92). The first was that the questions at the centre of the dispute about the nature and extent of the collaboration appeared likely to be resolved by witnesses located in the UK. The second was that greater progress had already been made in the UK proceedings. The third was that there was a lack of certainty that there would be jurisdiction to hear the whole dispute in New York at all.

The grounds of appeal and respondent's notice

20. GW Pharma's grounds 1 and 2 address the Moçambique principle and its application. GW Pharma contend that the judge erred in applying an overly restrictive test for the purposes of the Moçambique principle and further erred in his application of that test to the facts.
21. Ground 3 addresses the foreign act of state doctrine, and the common law public policy exception. The submission is that the judge erred in law in holding that the act of state doctrine (or common law public policy) did not require the court to decline jurisdiction.
22. Ground 4 relates to forum non conveniens, contending that the judge erred in declining a stay on those grounds.
23. Otsuka's case is that the judge was right for the reasons he gave but Otsuka also advances two additional points in support of the judge's overall conclusion. The first point is that as well as the exception to the Moçambique rule based on whether a validity challenge is direct or not which the judge applied, there is a second exception - for claims which relate to a contract. This case would also fall within that exception. The second point is a submission that GW Pharma's case would necessarily involve a country-by-country approach, contrary to the approach adopted by the English courts in related contexts (citing the Supreme Court in Unwired Planet v Huawei [2020] UKSC 37). The relevant principles ought not to be applied so as to prevent Otsuka from bringing its contractual royalty claim against GW Pharma in a single set of proceedings in GW Pharma's home jurisdiction.
24. The convenient approach will be to take Otsuka's respondent's notice points into account at the same time as dealing with grounds 1 and 2. The country-by-country argument also arises to some extent on ground 4 (forum non conveniens).

Grounds 1 and 2 – the Moçambique principle

25. Moçambique itself was a dispute in England between a Portuguese company and a company controlled by Cecil Rhodes concerning mining rights in what is now Mozambique. The House of Lords held that the courts of England and Wales had no jurisdiction to hear a claim to determine title to land situated abroad or for the tort of trespass in relation to that land. The modern authority on the Moçambique principle, as it may be applied to intellectual property matters, is Lucasfilm v Ainsworth. In Lucasfilm the claimant brought claims against a defendant in the UK for infringement of copyright concerning the design of the famous helmet worn by stormtroopers in the Star Wars films. One of the claims was for infringement of US copyright by acts committed by the defendant in the USA. The Court of Appeal had held that a claim for infringement of US copyright was not justiciable in the English court. This was based on what were by then regarded as fairly well established common law principles which took their origin from cases applying Moçambique in intellectual property disputes, including Potter v Broken Hill [1905] VR 612; (1906) CLR 479 and more recent domestic cases like Tyburn Productions v Conan Doyle [1991] Ch 75 and Pearce v Ove Arup [2000] Ch 403. The Supreme Court examined Moçambique and subsequent developments in the law that to a significant extent undermined aspects of that case and its application beyond cases about title to foreign land, and concluded (at paragraph 105) that a claim for infringement of foreign copyright was one over which the English court has jurisdiction, provided that there is a basis for in personam jurisdiction over the defendant.
26. Lucasfilm examines the position at common law because the foreign copyright in question was US copyright and so the then applicable Brussels Regulation (Council Regulation (EC) No 44/2001 of 22 December 2000) did not apply to it. Of course since Lucasfilm, the UK has left the EU and so the Brussels Regulation does not apply in any event. Nevertheless, the fact the Regulation does not apply is a different thing from the question whether aspects of the thinking behind the Brussels Regulation may illuminate questions which do arise.
27. The Supreme Court in Lucasfilm also took care to point out at paragraph 101 that the issue on the appeal was a narrow one and that the parties did not take issue with the application of the Moçambique rule to cases where what is in issue is the validity of a foreign patent, as opposed to its infringement.
28. At the end of paragraph 105, the Supreme Court summarised the modern Moçambique principle in these terms:

"All that is left of the Moçambique rule (except to the extent that it is modified by the Brussels I Regulation) is that there is no jurisdiction in proceedings for infringement of rights in foreign land where the proceedings are "principally concerned with a question of the title, or the right to possession, of that property". So also article 22(1) of the Brussels I Regulation does not apply to actions for damages for infringement of rights in land."

29. Then at paragraph 106 the court explained that it was possible to see how the rationale for the rule, based as it is on comity and the avoidance of conflict with foreign jurisdictions, can be applied to patents "at any rate where questions of validity are involved". In the remainder of that paragraph and subsequently, a number of reasons were given why this conclusion would follow. The main point was made at paragraph 108, that the modern trend was in favour of the enforcement of foreign intellectual property rights, particularly where there is no issue as to validity.
30. After Lucasfilm, in Chugai, Henry Carr J considered the application of the Moçambique principle in its modern form in the context of a dispute about a patent licence which included a US patent. Chugai sought a declaration that it was not obliged to pay royalties to the licensor UCB. Aspects of the case which Chugai sought to advance to show it was not liable to pay under the licence raised questions of the validity of the US patent. At one stage UCB had contended that this meant the entire claim was non-justiciable but by the time the matter came to the judge, the jurisdiction challenge was focussed only on the aspects of Chugai's case which raised validity arguments. Henry Carr J held that the Moçambique principle did not affect the court's jurisdiction in the case. The judge's reasoning contains a number of steps.
31. First, in order to decide if the dispute is concerned with or principally concerned with a challenge to validity of a patent, it is necessary to examine the substance of the dispute (paragraph 29). Before the judge this arose first in the context of the application of Arts 24 and 27 of the Brussels (I) Regulation but given that the Supreme Court in Lucasfilm adopted the same language ("principally concerned with") then in my judgment the same logic applies.
32. Second, by reference to the CJEU in GAT v Luk (Case C-4/03) and two decisions of Arnold J (as he then was), Anan v Molycorp [2016] EWHC 1722 (Pat) and Actavis v Eli Lilly [2014] EWHC 1511 (Pat), Henry Carr J decided that in a case in which the court has no jurisdiction to hear a validity challenge, there is nevertheless a distinction to be drawn between a challenge to validity, which would be impermissible; and a court taking into account the consequences, for validity, of an argument about claim construction in an infringement case (paragraphs 26-32). Therefore on the facts of Chugai the judge held:

33. In its claim for a declaration in these proceedings, Chugai does not claim that the 771 Patent is invalid. Obviously, it does not seek a declaration of invalidity, but this is not merely a question of form, it is one of substance. Chugai contends that its construction is correct and does not contend, on this construction, that the 771 Patent is invalid. As a subsidiary argument, it points to the consequences for validity if UCB's construction is correct. If Chugai succeeds on its subsidiary argument, this would lead to a finding that its tocilizumab product falls outside the scope of the claims of the 771 Patent, and therefore no royalties are payable under the Licence. It is not asking the English court to conclude that the 771 Patent is invalid, as it contends that UCB's construction should be rejected.

34. In the circumstances, I reject UCB's submission that Chugai is challenging the 771 Patent's validity by formulating its claim as a contractual one for a declaration concerning royalties, or by characterising it as one concerned with infringement of the 771 Patent. I accept Chugai's submission that it does not claim that the 771 Patent is invalid, but merely requires the court to ask itself, as a guide to construction, what would be the hypothetical consequences for validity of the rival interpretations. Furthermore, the arguments advanced in the Disputed Paragraphs are incidental to the essential nature of Chugai's claim, which is a claim for determination of its royalty obligations under the Licence.

33. In Chugai these conclusions meant that the claim was not barred by the Brussels (I) Regulation, but they were also relevant to the application of Moçambique. After examining Moçambique and Lucasfilm, the judge took the final step and held that Moçambique did not mean the court should decline jurisdiction over the disputed aspects. That was because the dispute was clearly not principally concerned with validity (paragraphs 57-58) and that in any event the contractual exception to Moçambique applied (paragraphs 59-60).
34. The reasons why the dispute was not principally concerned with validity were the same reasons why the Brussels (I) Regulation challenge had been rejected. Chugai did not contend that the patent was invalid and sought no relief as to invalidity. Its arguments about validity were made to support the arguments about claim construction. As the judge put it, there was no "direct" challenge to validity.
35. The judge went on to consider foreign act of state, which I will come back to in context below, and then finally (obiter) considered the status of direct challenges to validity (paragraphs 70-74). There at paragraph 73 the judge identified what he regarded as powerful arguments that direct challenges, where validity is the principal issue, are not justiciable:

"i) There is basis for drawing a distinction between claims for infringement and invalidity of patents. A claim of infringement is an action in personam, which affects the parties to the action. A patent is a monopoly right in rem, which applies to the entire population of the territory in which it is granted.

ii) This distinction is reflected in the allocation of jurisdiction in the Brussels I Regulation. Article 24(4) compulsorily allocates jurisdiction over a dispute concerning the validity of a patent to the courts of the Member State in which (or for which) that patent is registered; the article does not apply to claims for infringement.

iii) The rule in Moçambique no longer applies to claims for damages for trespass. However, it continues to apply to actions for the determination of the title to, or the right to possession of, foreign land. Infringement of patent is analogous to trespass, whereas validity is analogous to a challenge to the title to or right to possession of land.

iv) As well as comity, the Moçambique rule is founded on the principle of territoriality. Lord Neuberger stated in Shergill v Khaira [2014] UKSC 33; [2015] AC 359 at [41] that the rule was "probably best regarded as depending on the territorial limits of the competence of the English courts or of the competence which they will recognise in foreign states".

v) Patents are local monopolies which involve local policies and local public interest. Their effect is territorially limited. Their validity should be matters for the local judges of the country in which the patent right was first created: see Lucasfilm [2009] EWCA Civ 1328; [2010] Ch 503 at [175] per Jacob LJ."

36. Subject to a point on the contract exception below, I agree with all of Henry Carr J's reasoning in Chugai to which I have referred. The sub-paragraphs above give sound reasons why the courts of England and Wales should decline jurisdiction to determine direct challenges to the validity of foreign patents. Sub-paragraph (ii) still has force today despite the fact that the Brussels (I) Regulation is not part of UK law. Its relevance to the point being made then was and remains as an indication of how states have approached the allocation of jurisdiction internationally, drawing a distinction between validity and infringement.
37. In Chugai there is reference to both the idea of whether a validity challenge is a direct one and also to whether proceedings are "principally concerned with" validity. These two expressions are performing different tasks and it is worth keeping them distinct. A claim consisting of nothing other than a claim for infringement, in which the defendant does not claim that the patent is invalid, but merely requires the court to ask itself, as a guide to construction, what would be the hypothetical consequences for validity if there was infringement, does not involve a direct challenge to validity. Such a claim is also not principally concerned with validity. On the other hand a claim consisting of nothing other than a request for revocation on the ground of invalidity or a declaration of invalidity would be a direct challenge to validity, and would be principally concerned with validity. However a claim raising multiple issues might well properly be said not to be principally concerned with validity, even if one of the subsidiary issues was a direct challenge to validity; but in such a case the court's response would depend on the circumstances. The court might not decline jurisdiction over the dispute as a whole but might address individual issues separately. If the direct challenge only arises on a contingent basis then the right response might involve case management. Unlike the judge below, I would not describe this latter situation as one in which what was really a direct validity challenge was rendered not a direct challenge owing to its subsidiary nature in the action as a whole. The nature of the challenge is a direct one, but its status in the proceedings as a whole means that they are not principally concerned with it.

The Moçambique rule and contracts

38. To complete the picture it is necessary to consider the long standing exception to the Moçambique rule concerning contracts and equitable obligations (see Lucasfilm, paragraph 54, referring to Penn v Lord Baltimore (1750) 1 Ves Sen 444, in which the dispute before the English court was about the boundaries between Maryland and Pennsylvania). The exception was also identified by Lloyd-Jones LJ (as he then was) in Hamed v Stevens [2013] EWCA Civ 911 at paragraph 11:

"…the court has no jurisdiction to entertain proceedings for the determination of the title to, or the right to possession of, immovable property situated outside England, except where:

a) the claim is based on a contract or equity between the parties; …"

39. Also in Hamed, at paragraph 14 Lloyd-Jones LJ made the point that before the Moçambique rule can apply at all, the proceedings must raise directly the issue of title to foreign land. He then went on to consider how the exception would apply in the case before the court – which was a claim for restitution of money paid for a property in Egypt which, it was alleged, had not been transferred to the purchaser, both parties to the contract being before the English court. The conclusion was that there was jurisdiction in Hamed because the case was to enforce contractual or equitable rights as between parties amenable to the jurisdiction (paragraphs 21-24).
40. There are two questions in the present case about the contract exception. One is whether it depends on the existence of an exclusive jurisdiction clause in the contract and the other is about the extent of the exception itself. Would it, for example, allow the court to entertain a direct challenge to the validity of a foreign patent which the court would not have had jurisdiction to determine in the absence of the relevant contract (or equitable obligation)?
41. On the first issue the answer is clear. The contract exception to Moçambique does not depend on the existence of an exclusive jurisdiction clause. There has been such a clause in some of the cases in which the exception has been applied. Good examples are Chugai itself (e.g. paragraph 60) and also the earlier judgment of Peter Prescott QC sitting as a Deputy Judge in Griggs v Evans [2005] Ch 153 at p133 which I will set out now in full because I will come back to it below. Peter Prescott said:

"…in one sense it has always been possible to call into question both the validity and the scope of a foreign intellectual property right. For instance, where the defendant has agreed to pay royalties to the claimant on any product covered by a valid claim of a foreign patent and the agreement is governed by English law and confers jurisdiction upon the English courts. A recent instance is Celltech (Adair's) US Patent [2004] FSR 35 : indeed at first instance Jacob J said [2003] FSR 433 , para 8: "... I found myself receiving submissions on US case law just as if I were a US district judge." When this happens our courts are not considered to act in breach of comity. Even though they might appear to be inquiring into the validity or scope of an intellectual property right granted by a foreign sovereign. But in truth, they are not purporting to tell the American public, say, that one of their patents is invalid, or that the scope of its claims is not what it might appear to be. They are merely settling the rights of two private litigants who have chosen to submit their dispute to the adjudication of our courts. Once again, this involves rights in personam, not rights in rem."

42. However the exception has also been applied in cases without such a clause, such as Hamed itself and statements of the principle were made without reference to such a clause in the cases cited in Hamed (Pattni v Ali [2007] 2 AC 85 at paragraph 26 and Deschamp v Miller [1908] 1 Ch 856). I cannot see any justification based on principle for requiring such a clause. In my judgment the principle must be that the exception applies when the parties to the contract are amenable to the court's jurisdiction. That may but need not arise from the existence of an exclusive jurisdiction clause.
43. The second issue is more involved. In a way the question is whether the exception really is an exception to a rule that the court has no jurisdiction to determine a claim principally concerned with title (etc.) to foreign land or whether it is really just a manifestation of the proper application of the test for what it does or does not mean to say that a claim is principally concerned with title (etc.). Or putting it another way, can the court, when considering a contract claim, decide on title to foreign land, and by extension the validity of a foreign patent?
44. The answer I believe can be found in Hamed, particularly at paragraphs 21 -24. At paragraph 21 the point was made that, if the court finds that an enforceable personal obligation exists, it may order specific performance of that obligation to transfer property abroad. In the next paragraphs the court is then concerned to identify exactly what rights and obligations it was that the court was being asked to enforce; and the answer was the personal obligations of the defendant to the claimant and not their rights under the law of Egypt as to the ownership of property. Therefore the court had jurisdiction over the claim. In other words there are circumstances in which the contractual exception means that the court will grant relief which alters the title or right to possession of foreign property.
45. Applied to patents, in my judgment the answer to the question is this. If all the court is being asked to do is enforce a contract – such as the royalty provisions of a patent licence – then if the validity of a foreign patent is relevant to that contractual question under the applicable law – e.g. if the Lear case applies in the manner I understand GW Pharma to contend for – then the court can address that issue even though it involves what is a direct challenge to validity. The reason is because all that the court is being asked to do is decide whether or not to enforce a personal contractual obligation. That is similar to the point made in Hamed at paragraph 24 in which the court held that even if the personal obligations being enforced involved foreign law, that was not a bar to the exercise of jurisdiction over defendants amenable to the jurisdiction. However a critical feature of this would be that it could not depend on the grant of relief such as an order purporting to revoke a foreign patent as a necessary step in deciding whether or not to enforce the contract. The contract exception only gives the court jurisdiction to enforce the contract, nothing more. Even if the context in which this issue arose was a patent licence contract, a court purporting to revoke a foreign patent would be going beyond enforcing the personal obligations of parties amenable to the jurisdiction. Thus in the context of patents, in that limited sense, the contract exception seems to me to be capable of going further than the ambit of the Moçambique rule itself, absent a contract.
46. The distinction I am drawing is I think the same as the distinction between a decision in rem and one in personam. The appellants submit that the in rem/in personam distinction may be analytically relevant but is not determinative. In my judgment it is at least a label useful to characterise a relevant distinction. The contract exception does not allow the court to make a decision about the validity of a foreign patent in rem but it would allow the court to address the validity of a foreign patent in the course of making a decision concerning contractual rights in personam, assuming (such as if the Lear point does work in the way I have described) such a question was relevant to the contract decision.
47. In this respect I am differing from the judge below at paragraph 71 but I am in agreement with an aspect of Griggs v Evans in the passage cited above and I think also with Henry Carr J in Chugai at paragraph 73. I would add that I am not persuaded that debates about the preclusive effect under US law of a decision of an American court relating to validity under the Lear jurisdiction have any bearing on this issue (see Blonder Tongue v University of Illinois Foundation 402 U.S. 313 (1971)). At the risk of repetition, absent a contract, the Moçambique rule in its current form applies and the English court has no jurisdiction to rule on a direct challenge to the validity of a foreign patent.
48. Bearing all this in mind, I would state the Moçambique rule as explained and formulated in Lucasfilm, and as it applies to patents in the following way:
49. First, in a case in which the courts of England and Wales have in personam jurisdiction over a defendant, then the courts have jurisdiction in proceedings for infringement of a foreign patent save where those proceedings are principally concerned with a question of the validity of that patent. The proceedings will not be principally concerned with validity only because the defendant, who does not claim that the patent is invalid, requires the court to ask itself as a guide to construction, what would be the hypothetical consequences for validity if there was infringement. However what the rule does not permit is a direct challenge to the validity of a foreign patent, and (subject to the exception below) the court has no jurisdiction to determine a claim that the foreign patent is invalid.
50. Second, this Moçambique principle is also subject to a contractual exception. If the case is one in which the court is asked to enforce a contract between the parties then in addition to questions of patent scope/infringement, if and only to the extent that questions of the validity of foreign patents need to be addressed in order to decide on the true nature and scope of the parties' contractual obligations to one another, then the court can do so.

Assessment on grounds 1 and 2

51. As to ground 1 of the appeal, GW Pharma submit that the judge misapplied Chugai by finding that a validity challenge is only direct if it is the first one in the sequence of arguments raised by a party. This is said to overlook the fact that secondary or alternative arguments are often the true centre of gravity of a given case. The order in which the arguments are presented has no bearing on their importance. GW Pharma also argue that avoiding the application of the Moçambique principle through case management is not a solution, as it would lead to frequent stays in favour of foreign proceedings to determine the issues of validity under foreign law. GW Pharma submit that whether the Moçambique principle is engaged depends on the substance of the dispute and whether the dispute calls into question the validity of the patent "to a sufficient extent', arguing that this case resembles Anan, since Otsuka's claim is also built on the premise that the patents are valid.
52. On ground 2, which is convenient to address at the same time, GW Pharma submit that the judgment fails to engage in any detailed analysis of the weight and prominence of GW Pharma's invalidity defence as it relates to the case overall. GW Pharma contend that the validity issue will almost inevitably arise, because invalidity (at least regarding the US patent) is a necessary consequence of construing the US patent claims so broadly so as to cover Epidiolex. As a distinct point, GW Pharma argue that the in rem / in personam distinction cannot be determinative of the issues in the case. Finally GW Pharma refer to an estoppel issue mentioned by the judge and submits it is unlikely to feature prominently.
53. Otsuka contends that the judge made no such errors, supporting the judge's analysis of the weight and prominence of the issues. Otsuka argues that in substance in the present case, if and insofar as the issue of validity arises, it only does so incidentally to the claim. If a direct challenge does arise then case management may be the solution, but it may never be required. On the claim being founded on the premise that the claims are valid (referring to Anan) Otsuka submits that if this were sufficient to engage the Moçambique principle, then any assertion of invalidity would exclude the Court's jurisdiction and no royalty claim could ever be brought in circumstances where there is a multi-jurisdictional patent licence.
54. Taking the points in turn, in my judgment the judge was right to identify the role which the questions of validity actually played in the arguments. GW Pharma's point on sequencing would be a good one if the judge had blindly adopted an approach based on which argument was pleaded first, or something of that sort, but he did no such thing. The judge considered the issues as a whole and looked at the matter in substance. He summarised the dispute and the way in which the issues arose at paragraphs 73-78. He decided that patent validity was conditional and that the court may well be able to decide the dispute without having to consider validity at all. Examining the logical sequence and the relative significance of the various issues is the right approach. The question of independent development is an instructive example. Although in strict logical terms, the two questions of independence and patent validity are distinct, it is manifest, looking at this case as a whole, that the independent development issue is, as the judge held, GW Pharma's "principal defence". In other words it is the, or at the very least a, major centre of gravity of the proceedings. Therefore the most that could be said is that challenges to validity might (or might not) arise as the case develops, and that can be addressed in due course. The judge made no error in this respect.
55. Focussing on the validity arguments, some of them amount to squeeze arguments. These do not involve a direct challenge to validity at all because in the end the court is only being asked to decide on patent coverage. The presence of arguments of this sort is not capable of making it a claim principally concerned with validity, no matter how important they are in the proceedings. They pose no jurisdictional difficulty and do not need further consideration.
56. In relation to the argument GW Pharma have raised that the claims are invalid irrespective of construction, that is a direct challenge to the validity of a foreign patent. However, as explained already, it is conditional and may never arise (e.g. if GW Pharma win their case on independence or Otsuka wins on its case that GW Pharma are estopped from challenging validity) and so it does not justify declining jurisdiction over the proceedings as a whole. But if it does arise then case management might be required (but see the contract exception next).
57. The first point on the contract exception is that the defendants are amenable to the court's jurisdiction in any event. Therefore the contract exception can apply in this case even though the agreement has no exclusive jurisdiction clause.
58. Looking forward, it is notable that GW Pharma's defence has been filed and there is no counterclaim for relief such as a declaration of invalidity. As presently constituted the whole of these proceedings are nothing other than a claim for enforcement of the agreement. No in rem relief is sought by either party. That may be sensible because on one view if GW Pharma wish to avail themselves of a validity challenge to one of the patents, given the terms of the contract and the definition of a Valid Claim, there may be no point asking the English court to consider a direct validity challenge, what would be required would be for GW Pharma to bring invalidity proceedings in the relevant foreign court. On the other hand if, under the applicable law, Lear was found to apply as GW Pharma contend for then there I would hold that there is a basis for that kind of challenge to the validity of the US patent as part of the contractual dispute, but these issues can be dealt with as the proceedings progress.
59. Finally, I will refer to Anan. The appellants' reliance on that case is misplaced. In Anan the issues of infringement and validity of a German patent in that case were found to be inseparable and so jurisdiction was declined because validity was the exclusive preserve of the German court. However this aspect of Anan was rightly distinguished by Henry Carr J in Chugai at paragraph 29. Whatever the position was on the facts of Anan, as a general proposition not every infringement dispute is one which is principally concerned with validity.
60. I would therefore dismiss this appeal on grounds 1 and 2 relating to the judge's application of the Moçambique rule but also on the respondent's notice ground that the contract provides a free standing basis for the court's jurisdiction.

Ground 3 – foreign act of state

61. The foreign act of state doctrine has been applied as a distinct justification for the non-justiciability of foreign trade marks and patents, because those intellectual property rights may be said to depend on the grant or registration by the state. So by contrast the doctrine has not been applied to copyright, because that right arises automatically as a matter of law and does not depend on grant.
62. Lucasfilm examined the foreign act of state doctrine from paragraph 81 onwards. The Court referred to several US authorities on the subject (at paragraphs 81-84) before turning to Potter v Broken Hill, which was the only Commonwealth authority where the act of state doctrine prevented an action for the infringement of foreign intellectual property rights (paragraph 85). Also in that paragraph the Court cited with approval two passages from US decisions. The first was the dissent of Circuit Judge Newman in the CAFC in Voda v Cordis Corp, 476 F 3d 887 (Fed Cir 2007). That judge, with unrivalled experience of patent law, had observed that 'not every governmental action is an act of state' (at p 914). Second, the Court noted the decision of the Court of Appeals for the Third Circuit in Mannington Mills, Inc v Congoleum Corpn, 595 F 2d 1287 (3^rd Cir), rejecting the view that 'the mere issuance of patents by a foreign power constitutes … an act of state' (at p 1293-94).
63. The majority in Lucasfilm therefore concluded (at paragraph 86):

"… in England the foreign act of state doctrine has not been applied to any acts other than foreign legislation or governmental acts of officials such as requisition, and it should not today be regarded as an impediment to an action for infringement of foreign intellectual property rights, even if validity of a grant is in issue, simply because the action calls into question the decision of a foreign official."

64. This passage is put so as to cover both IP rights like trade marks and patents which may be said to depend on grant by a government official, and also unregistered IP rights like copyright.
65. Importantly Lord Mance abstained from expressing a definite view on the application of the foreign act of state doctrine to issues of validity of foreign intellectual property rights which "may be said to depend upon state grant" (at paragraph 115). In other words his abstention was limited to that class of IP rights, which includes patents.
66. However, Lord Mance returned to this point in Belhaj v Straw [2017] AC 964, which concerned the alleged complicity of UK officials in torts allegedly committed by the UK and other states overseas, with the act of state doctrine raised as a defence. Lord Neuberger gave the leading judgment, with which Lord Mance concurred in a separate judgment. He cited Lucasfilm as an authority for the proposition that 'acts of officials granting or registering intellectual property rights have been held to be outside any doctrine of foreign act of state.' (at paragraph 73(iv)). The reference to grant and registration is worth noting because that was the very aspect of Lucasfilm from which Lord Mance had abstained before.
67. Lucasfilm was also cited with approval in Maduro Board of the Central Bank of Venezuela v Guaidó Board of the Central Bank of Venezuela [2022] 2 WLR 167, in which the act of state doctrine was raised as part of the question whether courts in this jurisdiction may consider the validity of appointments made to the Central Bank of Venezuela by Mr Guaidó, who had been recognised by the then UK Foreign Secretary as Venezuela's interim president. Lord Lloyd-Jones (for the unanimous Supreme Court) identified the formulation at Lucasfilm paragraph 86 as stating one of the exceptions to the act of state doctrine (at paragraph 136(7)).
68. Finally, in Chugai, Henry Carr J (at paragraph 67) accepted a submission that the Supreme Court's reasoning in Lucasfilm was obiter but held that nevertheless it 'cannot and should not be ignored', concluding (at paragraph 68) that the decision of a patent examiner to grant a patent is not an act of state.
69. Turning to the submissions, those from GW Pharma largely focus on characterising the discussions in both Lucasfilm and Chugai as obiter. GW Pharma say that the correct view was that expressed by Professor Briggs at paragraph 21.09 of Civil Jurisdiction and Judgments (7^th Ed) that 'the grant of a patent right is closer than many to an act of sovereign power' and they draw support from Lord Mance's abstention at paragraph 115 in Lucasfilm. In support of the submission, GW Pharma cite:

i) Blad v Bamfield [36 ER 992]; (1674) 3 Swan. 604, in which the English Court of Chancery refused to judge the validity of letters patent granted by the King of Denmark,

ii) comments to the effect that granting a patent 'was an exercise of national sovereignty' in the Jenard Report on the Brussels Convention OJ 1979 C59, pp 1, 36, cited by the CJEU in GAT v LuK at paragraph 23, and

iii) Article 24(4) of the Recast Brussels Regulation (Regulation (EU) No 1215/2012).

70. In response, Otsuka submits that there is no basis for going behind the statements in Lucasfilm, which have been cited with approval in Belhaj and Maduro above and also in Yukos Capital Sarl v OJSC Rosneft Oil Co (No 2) [2014] QB 458 (at paragraphs 63-64 per Rix LJ). To do so, Otsuka submits, would substantially limit the availability and effectiveness of remedies before the English Court for holders of intellectual property rights.
71. Otsuka also submits that the ratio of Lucasfilm did in fact encompass the applicability or not of act of state to patents because in its reasoning on the points at issue under Moçambique the Supreme Court needed to and did examine Potter v Broken Hill, which is a patents case in which act of state was applied.
72. In my judgment, looking at the state of the authorities today, whether the point in Lucasfilm itself was to be characterised as ratio or highly persuasive obiter, paragraph 86 is now to be regarded as authoritative, having been cited with approval twice by the Supreme Court in Belhaj and Maduro. If it was necessary to do so I would hold that the reasoning on foreign act of state is in fact part of the ratio of Lucasfilm for the reason given by Otsuka. Although at paragraph 101 the Supreme Court recognised the narrowness of the issue to be decided, nevertheless examining the limits of the act of state doctrine as it applied to registered intellectual property rights like patents was a necessary part of the court's chain of reasoning which involved undermining Potter v Broken Hill (see the whole passage at paragraphs 60–88) and thereby overruling the copyright case Tyburn v Conan Doyle [1991] Ch 75, which was based on it.
73. In any case, even absent the authorities I would hold that as a matter of principle the modern grant of a patent for an invention does not fit within the act of state doctrine as it stands today for two reasons. The first reason relates to the exercise of grant itself. The very word "grant" harks back to a past time, before the Statute of Monopolies 1623, when letters patent were granted on the whim of the Stuart monarchs (and similarly I suspect the Danish monarchy in Blad v Bamfield). Today there is no such condescension by the sovereign power in the grant of a patent by the Comptroller of the Patent Office. Once a properly constituted patent application has been examined and found to comply with the requirements of the law, the Comptroller is required by statute to grant the patent. The relevant words are in s18(4) of the Patents Act 1977 which provide essentially that if the applicant's application is all in order then 'the comptroller shall … grant him a patent.' The second reason follows on from this and was given by Henry Carr J in Chugai at paragraph 68. He observed that once the patent had been granted, any party can challenge the validity of the patent and then can do so in a manner and on grounds which are quite different from an attempt to challenge legislation or government acts such as requisition.
74. As far as I know while many details differ, all states have broadly similar processes for granting and challenging validity of their patents and so, before applying foreign act of state to a patent, I would need to be shown evidence that mechanisms of a kind I have described do not apply in that state.
75. As explained already above in addressing grounds 1 and 2, the courts here should decline jurisdiction to determine direct challenges to the validity of foreign patents (subject to the contract exception). However on grounds of authority and principle, I agree with the judge below that the act of state doctrine is not relevant to the analysis of the court's jurisdiction in this case. Neither side before us suggested that looking at the matter through the guise of public policy made any difference to this analysis.

Ground 4 – forum non conveniens

76. On ground 4 GW Pharma submit the judge adopted a too narrow approach to the Spiliada principles, and failed to conduct a holistic evaluation of all of the potentially relevant factors and the interplay between them. They submit that the judge discounted the significance of the fact that the agreement is governed by New York law, arguing that the fact that a dispute is governed by foreign law is a prima facie reason why the courts of that jurisdiction are more appropriate. Since all of the claims and defences will raise issues of New York law, this will require New York law expert evidence. Moreover, GW Pharma argue that the judge's reliance on the fact that the agreement provided for arbitration proceedings to be held in London if commenced by Otsuka was misplaced because it would be subject to New York governing law and US discovery proceedings anyway. GW Pharma also submitted that potential witnesses are just as likely to be in the US and that the importance placed by the judge on the (perceived) progress of the English proceedings was misplaced for two reasons. First there is no reason to suppose that the US proceedings will take longer to be resolved than the ones in London, and second the question of appropriate forum should not depend on the relative speed of the competing fora.
77. In response, Otsuka submits that there is a serious dispute about whether the New York State Court would permit GW Pharma to rely upon the invalidity defence as regards any of the patents in issue (i.e. even the US patent), let alone all of them (i.e. all the non-US patents). The judge was right (at paragraph 92) to rely on this lack of certainty about the New York jurisdiction. Second, Otsuka submits that there is no basis for interfering with the judge's conclusion that GW Pharma failed to show that the New York State Court was clearly or distinctly a more appropriate forum. The judge weighed the factors identified by the parties before reaching his conclusion and so the court should be slow to interfere with the conclusion unless it has identified an error in the reasoning, which it has failed to do.
78. At paragraphs 81-92 the judge conducted an appropriately holistic evaluation of all of the potentially relevant factors and the interplay between them. He came to a conclusion which was plainly open to the court on the material. Taking the specific points, while no doubt not every witness will come from the UK, the independent derivation issue is at the centre of the dispute and relates to what happened at Reading University and the judge was entitled to put weight on the location of the witnesses. Second, as the judge noted, a dispute of this kind, concerning an international patent licence, ought to be able to be resolved somewhere if at all possible, thereby minimising the need for parties to take proceedings on a country by country basis to resolve what is in truth a single dispute. Accordingly the lack of certainty about whether the only other candidate forum (the New York State Court) had jurisdiction to deal with the whole dispute was another important factor, as the judge recognised.
79. I would agree with GW Pharma that the relative timing as between the proceedings in London and New York is not a powerful factor but the judge did not treat it as determinative and made no error mentioning it. The terms of the arbitration clause were not something the judge placed significant weight on and so the point GW Pharma takes about that does not assist the appeal.
80. The judge had well in mind that to try the case in London would require substantial US law evidence. That will be evidence of US patent law, including evidence on the point about Lear. However the Patents Court is able to handle foreign law of this kind (see e.g. Celltech and also the judgment at the trial which I heard following the jurisdiction challenge in Chugai itself (Chugai v UCB [2018] EWHC 2264 (Pat)).

Conclusion

81. I would dismiss this appeal.

Lord Justice Baker:

82. I agree.

Lady Justice Thirlwall:

83. I also agree.