England and Wales High Court (Patents Court) Decisions

IN THE HIGH COURT OF JUSTICE

		Rolls Building Fetter Lane, London, EC4A 1NL
		12 February 2016

MR JUSTICE ARNOLD :

Contents
Topic	Paragraphs
Introduction	1-6
The procedural context of the Dextrose Remission Issue in more detail	7-12
The witnesses	12-26
Factual witnesses	12-13
Oncologists	14-15
Endocrinologist	16-17
Pharmacists	18-23
Foreign law experts	24-26
The law with respect to the Dextrose Remission Issue	27-34
Indirect infringement	27-32
Negative declaratory relief	33-34
Factual background to the Dextrose Remission Issue	35-92
A note on terminology	35
Alimta and the Alimta SmPC	36-39
Actavis' application for marketing authorisations	40
The Actavis Products and UK SmPC	41-45
Proposed variation to the UK SmPC for the Actavis Product	46-51
Proposed variations to the CP SmPC for the Actavis Product	52
The German SmPC for the Actavis Product	53
Administration of pemetrexed	54
Dexamethasone	55
Preparation of cytotoxic drugs for administration	56-60
The use of saline and dextrose solution as diluents for chemotherapy drugs	61
Stability data available to pharmacists	62
Storage as aseptic preparations	63
Published stability data for Alimta	64
Actavis' stability for the Actavis Product	65
Supply and distribution of the Actavis Product	67-69
Steps taken by Actavis to prevent the Actavis Product from being diluted in saline	70-77
Steps taken by Actavis in relation to the availability of stability data of its product in saline	78-81
Steps taken by Lilly to prevent the Actavis Product from being diluted in saline	82-83
Diabetes	84-91
Assessment of the Dextrose Remission Issue	92-172
The role of the oncologist	94
The role of the endocrinologist	95
The evidence of the pharmacists	96
The importance of the SmPC	97-99
Is it likely that stability data for the Actavis Product in saline will become available?	100-109
Would there be a motivation to use saline rather than dextrose solution as the diluent for the Actavis Product?	110-136
Italian and Spanish law	137
Is it likely that saline would be used to dilute the Actavis Product despite the steps taken by Actavis to prevent this	138-139
Letters to competent authorities and medical centres	140-148
Contractual terms	149-153
Questions from pharmacists about the use of saline	154-161
Lilly's case on infringement from launch	162-171
The stock-piling argument	163-166
The tender argument	167-171
Conclusion	172
The Letters Issue	173-223
Factual background	173-206
Applicable law and burden of proof	207-208
Actavis' case	297
Lilly's case	210-213
Contract	214-216
The law	214
Assessment	215-216
Promissory estoppel	217-223
The law	216
Assessment	218-223
Should a declaration be granted?	224
Conclusion	225
Summary of principal conclusions	226

Introduction

1. Pemetrexed disodium is a chemotherapeutic treatment for lung cancer which has been marketed by the Defendant ("Lilly") or its subsidiaries under the brand name Alimta since 2004. As explained in more detail below, Alimta is marketed as a lyophilised (freeze-dried) powder with instructions to reconstitute it, and then dilute it, in a 0.9% solution of sodium chloride (known as "normal saline"). Pemetrexed and its pharmaceutically acceptable salts were protected by European Patent No. 0 432 677, which expired on 10 December 2010. The protection conferred by that patent was extended by Supplementary Protection Certificates ("the SPCs") which expired on 10 December 2015. Lilly also owns European Patent No. 1 313 508 ("the Patent") for the use of pemetrexed disodium in combination with vitamin B12 or a pharmaceutical derivative thereof and optionally a folic protein binding agent. The Patent will not expire until 15 June 2021.
2. On 27 July 2012 the Second Claimant commenced these proceedings seeking declarations of non-infringement ("DNIs") in respect of the French, German, Italian, Spanish and United Kingdom designations of the Patent with a view to clearing the way for the launch of a generic pemetrexed product on expiry of the SPCs. Subsequently other Claimants were joined, some of which later ceased to be Claimants, but for convenience I will refer to the relevant Claimant(s) from to time as "Actavis". Actavis initially sought DNIs in respect of a product the active ingredient in which was pemetrexed dipotassium. Subsequently Actavis also sought DNIs in respect of pemetrexed diacid and pemetrexed ditromethamine. Actavis applied for marketing authorisations for their products by reference to Alimta. At that stage Actavis proposed that their products should be reconstituted and/or diluted in normal saline, like Alimta. Actavis did not raise any issue with regard to the use of vitamin B12 etc. After unsuccessfully challenging the jurisdiction of this court with respect to the French, German, Italian and Spanish designations of the Patent (see my judgment dated 27 November 2012 [2012] EWHC 3316 (Pat) and the judgment of the Court of Appeal dated 21 May 2013 [2013] EWCA Civ 517, [2013] RPC 37, "Actavis I"), Lilly counterclaimed for threatened infringement of the UK designation. As explained in more detail below, Lilly also brought parallel proceedings in Germany, which led to Actavis discontinuing its claims in respect of the German designation shortly before the trial here.
3. In a judgment delivered on 15 May 2014 ([2014] EWHC 1511 (Pat)), I held that Actavis' proposed dealings in its products would not amount to either direct or indirect infringement of the Patent. In a judgment delivered on 25 June 2015 ([2015] EWCA Civ 555, "Actavis II"), the Court of Appeal upheld my decision with respect to direct infringement, but reversed it with respect to indirect infringement. In brief summary, the Court of Appeal held that the claims of the Patent were restricted to pemetrexed disodium, and therefore dealings in Actavis' products would not amount to direct infringement; but that, if Actavis' products were reconstituted and/or diluted in a sufficient quantity of saline, the resulting solution would fall within the claims because it contained both sodium ions and pemetrexed ions in a ratio of at least 2:1, and thus the supply of Actavis' products would amount to indirect infringement. Both sides have applied to the Supreme Court for permission to appeal against the Court of Appeal's decision, but unless and until the Supreme Court allows an appeal by one side or the other, the Court of Appeal's judgment is binding upon them and upon this Court. This judgment proceeds upon that basis.
4. When the Court of Appeal's judgment was supplied to the parties in draft, Actavis requested the Court of Appeal to remit for trial by the Patents Court an issue as to whether the supply of their products would constitute indirect infringement of the Patent if marketed with instructions to reconstitute and/or dilute the products with 5% dextrose solution instead of saline ("the Dextrose Remission Issue"). The Court of Appeal acceded to that request: see Actavis II at [155]-[156] and [160]-[162]. Although the Court of Appeal left it open to Lilly to apply to the Patents Court to strike out Actavis' claim for DNIs in respect of the supply of their products with instructions for reconstitution and/or dilution with dextrose solution as an abuse of the process, Lilly made no such application.
5. Actavis subsequently launched a pemetrexed diacid product ("the Actavis Product") in the UK on 21 December 2015. Actavis' evidence is that they are on track to launch the Actavis Product in France, Italy and Spain later this year. As explained in more detail below, the Actavis Product is a liquid product which does not require reconstitution, and the Summary of Product Characteristics ("SmPC") specifies that it is only to be diluted with dextrose solution. Lilly does not dispute that, if the Court of Appeal's judgment as to the scope of the claims is correct and if the Actavis Product is diluted with dextrose solution rather than saline, the supply of the Actavis Product does not amount to either direct or indirect infringement of the Patent. Lilly nevertheless resists the grant of DNIs in respect of the supply of the Actavis Product with instructions for dilution with dextrose solution. Lilly contends that it is foreseeable (and hence obvious) that the Actavis Product will be diluted with saline by some customers (or at least, that Actavis have not proved that it is not foreseeable) even though Lilly does not allege that Actavis are taking any steps to encourage the use of saline. By contrast with its stance at the first trial, however, Lilly has not counterclaimed for infringement of the UK designation.
6. In addition to the Dextrose Remission Issue, there is a separate issue between the parties as to the effect of two letters written by Actavis' solicitors to Lilly's solicitors on 17 April 2013 and 16 September 2013 ("the Letters"). Actavis seeks declarations that the Letters do not constitute legally binding undertakings ("the Letters Issue"). Lilly resists the grant of such declarations, but it has not counterclaimed for any relief. Nor has Lilly attempted to rely on the Letters to prevent the launch of the Actavis Product in the UK or (to date) in any other country.

The procedural context of the Dextrose Remission Issue in more detail

7. Counsel for Lilly emphasised in his submissions that the burden was on Actavis to establish its entitlement to the declarations it sought, not on Lilly to establish that dealings in the Actavis Product amounted to infringement. Actavis do not dispute this. Counsel for Lilly also emphasised that the Patent had over five years to run, and submitted that there was considerable uncertainty as to what the position would be in, say, two or three years' time. As counsel for Actavis pointed out, however, that submission has to be viewed against the stance adopted by Lilly.
8. Initially, Lilly resisted a DNI in respect of the supply of Actavis' products with instructions for reconstitution and/or dilution with dextrose solution on the basis that it was foreseeable that some pharmacists would not follow the instructions contained in the SmPC to use dextrose solution, but instead would reconstitute and/or dilute the products in saline. Given the narrow scope of the dispute, on 16 July 2015 I acceded to an application by Actavis for an order for an expedited trial of the Dextrose Remission Issue in a window commencing on 2 November 2015 for the reasons given in my judgment of that date ([2015] EWHC 2124 (Pat)). On 30 July 2015 Floyd LJ refused an application by Lilly for permission to appeal against that decision as being totally without merit.
9. In its evidence in chief served between 6 and 9 October 2015, however, Lilly advanced an unpleaded case to the effect that, although pharmacists would initially follow the instructions contained in the Actavis SmPC to use dextrose solution, it was foreseeable that they would subsequently switch to saline once stability data for Actavis' products in saline became available because of concerns as to the effect of dextrose on patients with diabetes. Lilly subsequently applied for permission to amend its statement of case to plead this case. That application came before me on 22 October 2015. In their skeleton argument for that hearing, Actavis pointed out that, on Lilly's own evidence, stability data would not become available until at least six months after the launch of Actavis' product, and may be not for a couple of years. Actavis argued that it followed that, on Lilly's own evidence, there was no possibility of infringement at the time of launch. Actavis went on to say:

"This all has trial management implications because what Lilly implicitly seeks the Court to do (even though there is no chance of infringement at launch since SPC expiry) is to speculate about what will happen in one, two or more years' time in relation to events of great uncertainty. If so, that is a very bad use of significant Court time and a better course would be declare that Actavis' activities are, presently, lawful, so that it can launch, with liberty to apply to Lilly if (which we say is unlikely) circumstances change. If Lilly does apply in due course, the Court will know, and not have to guess, what (if any) stability data is available, how Actavis' product is actually used, whether diabetics pose a real issue, and so on."

10. Lilly did not accept this suggestion. Lilly did nevertheless agree to the adjournment of the trial until 25 January 2016, and gave an undertaking not to seek preliminary injunctions to prevent the marketing of Actavis' products prior to judgment, as the price for being granted permission to amend its statement of case. (Actavis also sought an interim declaration by way of judgment on admissions, which gave rise to further issues and applications, but it is not necessary to go into that aspect of the matter for present purposes.)
11. As will appear, I agree with Lilly that there is some uncertainty as to what may happen in a few years' time. I agree with Actavis, however, that the reason why this is a problem is because of Lilly's refusal to accept that dealings in the Actavis Product would not initially infringe the Patent and to reserve its right to apply to the Court if and when circumstances changed, as Actavis proposed. I shall return to this point below.

The witnesses

Factual witnesses

12. Actavis called the following witnesses of fact:

i) Jonathan Wilson: Mr Wilson is the Managing Director of Actavis UK Ltd. He was a straightforward witness. Counsel for Lilly pointed out that much of Mr Wilson's statement consisted of hearsay, being matters of which he had been informed by other persons within Actavis, in particular Derek Brown, who is the Director of Actavis' Hospital Business in the UK, and members of Actavis' regulatory team. Counsel for Actavis submitted that it was convenient and proportionate to call one witness rather than several, and pointed out that both Mr Kopernicky and Mr Hannaby (as to whom, see below) had also included hearsay within their statements. It is fair to say that the hearsay in Mr Wilson's evidence was more extensive than that in Mr Kopernicky and Mr Hannaby's evidence. Lilly did not apply to cross-examine Mr Brown or anyone in the Actavis regulatory team, however. Furthermore, I am not persuaded that it would have made a great deal of difference if Actavis had called the relevant witnesses to give first-hand evidence. Nevertheless, I do accept that some caution should be exercised before placing significant weight on the hearsay statements in Mr Wilson's evidence.

ii) Oliver Jüngst: Mr Jüngst is a German lawyer and a partner in Bird & Bird's Düsseldorf office who has conduct of the parallel proceedings between the parties in Germany. He gave evidence of relevance to the Letters Issue. It was sensibly agreed between the parties that neither Mr Jüngst nor Mr von Falck (as to whom, see below) would be cross-examined and that challenges to their evidence would be confined to submissions.

13. Lilly called the following witnesses of fact:

i) Vladimir Kopernicky: Mr Kopernicky is Lilly's Senior Director, Medical Affairs - Europe. Mr Kopernicky was a straightforward witness.

ii) David Hannaby: Mr Hannaby is Lilly's Commercial and Operations Manager. Mr Hannaby was also a straightforward witness.

iii) Dr Andreas von Falck: Dr von Falck is a German lawyer and a partner in Hogan Lovells International LLP's Düsseldorf office. He has conduct of the parallel proceedings in Germany and his evidence also related to the Letters Issue.

iv) Dr Alfred Millà: Dr Millà has been the director of the medical oncology unit at the Hospital Nuestra Señora dei Remi in Barcelona, Spain for the last 12 years and has been working in oncology since 1974. Lilly initially proposed to call Dr Millà as an additional oncology expert. After I had ruled at a case management conference that it was inappropriate for Lilly to call two experts in the discipline of oncology, Lilly called Dr Millà as a fact witness instead. Despite this, Lilly served statements from Dr Millà containing expressions of opinion. Actavis unsurprisingly took objection to this. The objection was resolved partly by Lilly volunteering to delete passages from Dr Millà's second statement and partly by an agreement that the weight to be attached to the remaining parts of his statements objected to would be a matter for the Court. Regrettably, that was not the end of the dispute with respect to Dr Millà's evidence. Without prior warning to counsel for Actavis or requesting permission from the Court, counsel for Lilly led some supplementary oral evidence in chief from Dr Millà with respect to carboplatin. This necessitated the recall of Ms Juanals (as to whom, see below). Counsel for Actavis submitted that Dr Millà's evidence on this point was confused and inconsistent. I agree with this, as I shall explain below.

Oncologists

14. Actavis' expert was Professor Michael Seckl. Prof Seckl is currently Professor of Molecular Oncology at Imperial College London, where he is head of the Lung Cancer Research Group and of the Experimental Cancer Medicine Research Centre, and an honorary Consultant Medical Oncologist at Imperial College NHS Healthcare Trust. In addition, he is director of the Charing Cross Gestational Trophoblastic Disease ("GTD") Centre. He obtained a BSc in Immunology from University College London in 1983, an MBBS in 1986 and a PhD on the development of novel therapies for small cell lung cancer in 1995. He was appointed as a Senior Lecturer and honorary Consultant by Imperial in 1995, Reader in 2000 and Professor in 2002. His principal research interests are in the fields of small cell lung cancer and GTD. He has used pemetrexed since 2007. Prof Seckl was a very knowledgeable, careful and balanced expert.
15. Lilly's expert was Professor Nicholas Thatcher. Prof Thatcher is currently Professor of Oncology at the Christie Hospital NHS Trust. He obtained an MB BChir from Cambridge University and St Bartholomew's Hospital in 1970 and a PhD from Manchester University in 1979. He was successively Senior Registrar from 1978 to 1980, Senior Lecturer from 1980 to 1989 and Reader from 1989 to 1996 at the Christie before taking up his present position in 1996. He retired from routine clinical work in 2010, but is still involved in clinical trials. He prescribed pemetrexed in his clinical work and has continued to publish papers on it. He has published no less than 364 papers altogether. Although Prof Thatcher was a very distinguished and knowledgeable expert in oncology, I agree with counsel for Actavis that some of his evidence strayed outside the realms of his own expertise and into matters of endocrinology, as to which I prefer the evidence of Dr Powrie (as to whom, see below). I also agree with counsel for Actavis that Prof Thatcher was somewhat dogged in his views despite evidence to the contrary.

Endocrinologist

16. Actavis called Dr James Powrie. Dr Powrie is a full time Consultant Physician and Honorary Senior Lecturer specialising in diabetes and endocrinology at Guy's & St Thomas' NHS Foundation Trust, a position which he has held since January 1995. He obtained an MB ChB from Aberdeen University Medical School in 1982, including an intercalated B Med Biol in Immunobiology in 1979. He undertook his clinical training at Aberdeen Hospitals and University Medical School from 1982 to 1988. From 1988 to 1991 he was a Clinical Research Fellow and Honorary Registrar and from 1991 to 1995 a Lecturer and Honorary Senior Registrar at Guy's & St Thomas'. In 1992 he was awarded an MD by Aberdeen University for a thesis on the study of novel agents for the treatment of type 2 diabetes. From 1998 to 2014 he was the Lead Clinician for the diabetes clinical service at Guy's & St Thomas'. He has published over 50 papers on various aspects of diabetes and endocrinology, and he leads a clinical group which is currently carrying out clinical trials on novel treatments for type 1 diabetes. Counsel for Lilly submitted that Dr Powrie's evidence had failed to take into account the realities of treating extremely ill lung cancer patients. I do not accept that Dr Powrie's evidence is to be criticised on that score, but I do accept that it is a relevant matter for me to take into account in my overall assessment.
17. Lilly chose not to call an endocrinology expert, despite the fact that its case hinges on concerns about giving the Actavis Product to patients with diabetes. Counsel for Actavis submitted that it was to be inferred that Lilly had failed to find an endocrinologist who would support its position. I accept that submission.

Pharmacists

18. In relation to the UK:

i) Actavis' expert was Richard Bateman. Mr Bateman is currently a Senior Quality Assurance Specialist Pharmacist for East and South East England Specialist Pharmacy Services and is based at Guys and St. Thomas' Hospital NHS Foundation Trust. He obtained a BPharm from the Welsh School of Pharmacy in 1989, a Clinical Pharmacy Diploma from the University of London in 1992 and an MSc in Pharmaceutical Technology and Quality Assurance from the University of Leeds in 2000. He has been involved in aseptic preparation and dispensing of medicines within the NHS since 1993. He was responsible for the day to day operation of the aseptic unit at Ealing Hospital from October 1993 to June 1997 and from June 1997 to June 2000 he was the Quality Assurance Manager at Great Ormond Street Hospital NHS Trust with responsibility for ensuring that the four aseptic units there complied with regulatory requirements. From June 2000 to June 2003 he was the Regional Quality Assurance Pharmacist in London Specialist Pharmacy Services. He has held his current position since June 2003. Amongst his responsibilities in this post, he has been responsible for auditing unlicensed aseptic units in 80 NHS Trusts and attending as required at Medicines and Healthcare Products Regulatory Agency ("MHRA") audits of licensed units at 21 Trusts. He has also completed two terms as Chairman of the NHS Pharmaceutical Quality Assurance Committee and one term as secretary, as well as being a member of various bodies. Mr Bateman was very well qualified to give evidence with respect to the issues in this case, and he was a very careful and balanced witness. Counsel for Lilly criticised Mr Bateman for having omitted some matters from his reports, but I do not consider that this criticism is justified.

ii) Lilly's expert was Rajinder Nijjar. Ms Nijjar has been the Lead Cancer Pharmacist at Barts Healthcare NHS Trust and Area Team Cancer Pharmacist at NHS England, London Region since April 2013. She obtained a BPharm from Kings College London in 1994 and a Clinical Pharmacy Diploma from the University of Brighton in 1999. She has worked as an oncologist pharmacist since October 1996, holding positions at Guy's and St Thomas' NHS Trust, the Royal London Hospital, and Barts and the London NHS Trust from then until December 2008. From January 2009 to March 2013 she was Lead Cancer Pharmacist at Barts Healthcare and North East London Cancer Network Lead Pharmacist. Counsel for Actavis accepted that Ms Nijjar's oral evidence was fair and balanced.

19. In relation to France:

i) Actavis' expert was Professor Francois Chast. Prof Chast is currently Head of Clinical Pharmacy at Hôpitaux Universitaires Paris-Centre, which includes supervision of all medicines and intravenous formulations prepared at the pharmacy central unit. Prof Chast obtained a degree in pharmacy from the Faculty of Pharmaceutical and Biological Sciences (Paris XI) in 1972, a Masters in Law of Pharmaceutical Institutions from the University of Paris in 1973 and a Doctorate from University Réné Descartes in 1991. From 1975 to 1979 he was Head of Pharmacy-Toxicology at Hôpital Bicêtre and from 1979 to 2011 he was Head of Pharmacy-Pharmacology-Toxicology at Hôpital Hôtel Dieu, which involved amongst other things compounding of anticancer drugs. He was previously an Associate Professor at University Réné Descartes. Counsel for Lilly suggested that his oral evidence was a long way from his written evidence, but I do not accept this.

ii) Lilly's expert was Dr Jean Vigneron. Dr Vigneron is a hospital pharmacist at the Centre Hospitalier Universitaire de Nancy where he has worked since 1985. He has worked in the centralised unit for the preparation of cytotoxic drugs since 1987. He qualified as a doctor of pharmacy in 1982. He has been the President of Infostab, a French non-profit association whose aim is to promote the appropriate use of medicines in hospitals via preparation, administration, stability and compatibility, since 2006. He is a board member of the French Society of Oncology Pharmacy (SFPO) and a French delegate of the European Society of Oncology Pharmacy (ESOP). Dr Vigneron was the creator of the Stabilis website, which provides a database of stability information for drugs. Among other publications, he has published a number of studies on the stability of drugs. He was thus very well qualified to address that aspect of the case. Indeed, as counsel for Actavis pointed out, he was more knowledgeable about such matters than the average pharmacist. Counsel for Actavis accepted that Dr Vigneron's oral evidence was fair and balanced.

20. In relation to Italy:

i) Actavis' expert was Dr Massimo Medaglia. Dr Medaglia is the Director of the Pharmaceutical Unit at the Hospital L Sacco in Milan, a post which he has held since 2005. Since 2007 he has also been the Director of the Department of Pharmacy of the Hospital L. Sacco. He obtained a Masters in Pharmacy from the Faculty of Pharmacy of the Statale University of Milan in 1989 and a Postgraduate Diploma in Hospital Pharmacy from the same institution in 1994. He worked in the centralised unit for the preparation of sterile oncological preparations in the pharmacy of the Hospital of Vimercate from 1995 to 1996 and participated in the development of such a centralised unit at the Hospital L. Sacco. Counsel for Lilly suggested that his oral evidence was hard to follow, but acknowledged that this was probably due to language issues.

ii) Lilly's expert was Dr Giuseppina Fassari. Dr Fassari is Director of the Pharmacy Azienda Ospedaliera di Rilievo Nazionale di Alta Specializzazione ("ARNAS") Garibaldi in Catania, Sicily and is responsible for the Antitumor Pharmacy Unit. Dr Fassari qualified as a doctor of pharmacy in 1999. She has worked for ARNAS since 2004. Counsel for Actavis accepted that Dr Fassari's oral evidence was fair and balanced.

21. In relation to Spain:

i) Actavis' expert was Dr Antonio Salmerón. Dr Salmerón obtained a degree in Pharmacy from the University of Granada in 2002, a masters in Pharmaceutical Care from the University of Valencia in 2005, a PhD in Microbiology from the University of Granada in 2007 and a masters in Manufacturing Advanced Therapy Medicinal Products from the University of Granada in 2011. Dr Salmerón has been a Specialist Pharmacist in Hospital Pharmacy since June 2007, and has worked as a Specialist Pharmacist in a number of hospitals since then. His roles have included (amongst other things) the preparation of chemotherapy drugs and management of quality control programmes. Counsel for Lilly made no criticism of his evidence.

ii) Lilly's expert was Maria Juanals. Ms Juanals is a deputy head pharmacist at the Hospital Nuestra Señora del Remei in Barcelona and she provides relief cover at the Hospital de Barcelona. She works in the cytotoxic drug preparation unit. She obtained a degree in Pharmacy from the University of Barcelona in 2003. Counsel for Actavis accepted that her oral evidence was careful and fair, and contrasted it with that of Dr Millà.

22. A general point about the evidence of Prof Chast, Dr Medaglia and Dr Salmerón is that each of them was given a copy of Mr Bateman's first report and asked to follow the same structure and use the same terminology where applicable when preparing their first reports for ease of comparison. Counsel for Lilly submitted that this was inappropriate. I disagree, but in any event this does not affect the substance or weight of their evidence.
23. A separate point is that Dr Medaglia and Dr Salmerón expressed opinions on questions of Italian and Spanish law in their first reports. Lilly justifiably objected that Actavis had not pleaded these matters and that Dr Medaglia and Dr Salmerón were not properly qualified to address them. This led to an application by Actavis to amend their statement of case to plead the foreign laws relied upon, and to an agreed order for the exclusion of the relevant passages from Dr Medalgia's and Dr Salmerón's reports and for the admission of expert evidence of Italian and Spanish law (as to which, see below). Despite this, counsel for Lilly proceeded to cross-examine Dr Medaglia and Dr Salmerón on this topic.

Foreign law experts

24. As explained above, Actavis raised issues of Italian and Spanish law which have been addressed by experts in Italian and Spanish law. It was sensibly agreed between the parties that the foreign law experts would not be cross-examined.
25. In relation to Italian law:

i) Actavis' expert was Carlo Piria. Mr Piria obtained a law degree from the Catholic University of Milan in 1968 and was admitted to the Bar in 1971. Since 2000 he has been a lecturer in the Post-Graduate School of Regulatory Disciplines at Pavia University. Since 2001 he has also been a partner in the Life Sciences department of Avvocati Associati Fransozi, Dal Negro, Setti in Milan. Previously he worked as an in-house lawyer, including as general counsel of the Novartis Group in Italy. He has published a number of articles on pharmaceutical law.

ii) Lilly's expert was Sonia Selletti. She obtained a law degree from the University of Pavia in 1991 and qualified as lawyer in 1993. She then spent two years working in-house at a pharmaceutical company. Since 1995 she has been a partner in Studio Legale Astolfi e Associati and chief editor of Rassenga di dirrito Farmaceutico e della salute (Review of Pharmaceutical and Health Law). She has published a number of articles on pharmaceutical law.

26. In relation to Spanish law:

i) Actavis' expert was Professor José Garcia-Trevijano Garnica. He has been Professor of Administrative Law at the Universidad Complutense de Madrid, from which he previously obtained a PhD, since 1982. He has also been a lawyer of Spanish State Council since 1980, where he is currently Chief Counsel of the Health Section. He has published a book on the revocation of administrative acts and a number of articles.

ii) Lilly's expert was Hector Jausàs Farré. He obtained a law degree from the University of Barcelona in 1994. He is a partner in Jausàs law firm and head of its Life Sciences department. He has published a number of articles on pharmaceutical law.

The law with respect to the Dextrose Remission Issue

Indirect infringement

27. It is common ground that, so far as is relevant to the Dextrose Remission Issue, there is no relevant difference between UK law and the laws of France, Italy and Spain with respect to indirect infringement. Accordingly it is only necessary for me to set out the UK law.
28. Section 60 of the Patents Act 1977 provides, so far as is relevant, as follows:

"(2) Subject to the following provisions of this section, a person (other than the proprietor of the patent) also infringes a patent for an invention if, while the patent is in force and without the consent of the proprietor, he supplies or offers to supply in the United Kingdom a person other than a licensee or other person entitled to work the invention with any of the means, relating to an essential element of the invention, for putting the invention into effect when he knows, or it is obvious to a reasonable person in the circumstances, that those means are suitable for putting, and are intended to put, the invention into effect in the United Kingdom.

(3) Subsection (2) above shall not apply to the supply or offer of a staple commercial product unless the supply or the offer is made for the purpose of inducing the person supplied or, as the case may be, the person to whom the offer is made to do an act which constitutes an infringement of the patent by virtue of subsection (1) above."

29. Section 130(7) declares that a number of sections in the 1977 Act, including section 60, "are so framed as to have, as nearly as practicable, the same effect in the United Kingdom as the corresponding provisions of the European Patent Convention, the Community Patent Convention and the Patent Co-operation Treaty have in the territories to which those Conventions apply." Section 130(6) provides that references to the Community Patent Convention ("CPC") are to "that convention as amended or supplemented".
30. Article 26 of the CPC, as revised in 1989, provides as follows:

"Prohibition of indirect use of the invention

1. A Community patent shall also confer on its proprietor the right to prevent all third parties not having his consent from supplying or offering to supply within the territories of the Contracting States a person, other than a party entitled to exploit the patented invention, with means relating to an essential element of that invention, for putting it into effect therein, when the third party knows, or it is obvious in the circumstances, that these means are suitable and intended for putting that invention into effect.

2. Paragraph 1 shall not apply when the means are staple commercial products, except when the third party induces the person supplied to commit acts prohibited by Article 25.

3. Persons performing the acts referred to in Article 27(a) to (c) shall not be considered to be parties entitled to exploit the invention within the meaning of paragraph 1."

31. The background to Article 26 CPC, and hence section 60(2) of the 1977 Act, was explained by Jacob and Etherton LJJ, with whom Sir David Keene agreed, in Grimme Landmaschinefabrik GmbH v Scott [2010] EWCA Civ 1110, [2011] FSR 7 at [82]-[98]. They went on at [105]–[131] to consider the requirements of knowledge and intention in section 60(2). They found helpful guidance in relation to these questions in a number of decisions of the Bundesgerichtshof (Federal Court of Justice) on the corresponding German provision, which also derives from Article 26 CPC. In KCI Licensing Inc v Smith & Nephew plc [2010] EWCA Civ 1260, [2011] FSR 8 at [53] Jacob LJ delivering the judgment of the Court of Appeal summarised the key parts of Grimme with regard to the requirements of knowledge and intention as follows:

"i) The required intention is to put the invention into effect. The question is what the supplier knows or ought to know about the intention of the person who is in a position to put the invention into effect – the person at the end of the supply chain, [108].

ii) It is enough if the supplier knows (or it is obvious to a reasonable person in the circumstances) that some ultimate users will intend to use or adapt the 'means' so as to infringe, [107(i)] and [114].

iii) There is no requirement that the intention of the individual ultimate user must be known to the defendant at the moment of the alleged infringement, [124].

iv) Whilst it is the intention of the ultimate user which matters, a future intention of a future ultimate user is enough if that is what one would expect in all the circumstances, [125].

v) The knowledge and intention requirements are satisfied if, at the time of supply or offer to supply, the supplier knows, or it obvious to a reasonable person in the circumstances, that ultimate users will intend to put the invention into effect. This has to be proved on the usual standard of the balance of probabilities. It is not enough merely that the means are suitable for putting the invention into effect (for that is a separate requirement), but it is likely to be the case where the supplier proposes or recommends or even indicates the possibility of such use in his promotional material, [131]."

32. It is clear from these decisions that it is sufficient that a proportion of users will intend to use the means so as to infringe. Even if the majority of users will not intend to use the means to infringe, that is only relevant to remedies, and in particular financial remedies (see Grimme at [134]-[137]). On the other hand, one should disregard "speculative, maverick or unlikely use" of the means (see Grimme at [116], [124], [127] and [129]-[130] and KCI at [47]).

Negative declaratory relief

33. In Actavis II at [145] the Court of Appeal upheld my conclusion that the law which was applicable to the question of whether or not negative declaratory relief should be granted, or more precisely the conditions which need to be satisfied before a DNI will be granted, was the lex fori, that is to say, English law. Applying English law, this Court has a broad discretionary power under its inherent jurisdiction to grant a negative declaration if it is in the interests of justice to do so: see Messier-Dowty Ltd v Sabena SA [2000] 1 WLR 2040 at [41]-[42] (Lord Woolf MR). The old restrictive approach under which a negative declaration would not be granted unless there was a claim of right (Re Clay [1919] 1 Ch 66) has been abandoned. The modern law is that a negative declaration will be granted if it is right in all the circumstances to do so, and in particular if it will serve a "useful purpose": Messier-Dowty at [41]-[42]. It will do so if the claimant has a "real commercial interest" in the negative declaratory relief sought or a "real commercial reason" for it to be granted: Nokia Corp v InterDigital Technology Corp [2006] EWCA Civ 1618, [2007] FSR 23 at [19]-[20] (Jacob LJ).
34. Counsel for Lilly submitted that a negative declaration can only properly be made where the underlying issue has been "sufficiently clearly defined to render it properly justiciable" as Pumfrey J put it at first instance in Nokia v InterDigital [2006] EWHC 801 (Pat) at [20(iii)]. I accept this. Counsel for Lilly also sought to rely upon Pumfrey J's decision in Niche Generics Ltd v Lundbeck A/S [2003] EWHC 2590 (Pat), [2004] FSR 20 as establishing that it would be wrong in principle to grant a declaration based on assumed, hypothetical or contingent facts. As counsel for Actavis submitted, however, Pumfrey J was concerned in that case with a different question, namely the effect of a declaration of non-infringement under section 71 of the 1977 Act in the event that the beneficiary of the declaration did something different. I do not accept that the court cannot make a declaration based on assumed, hypothetical or contingent facts, but I do accept that, if the facts on which an application for a declaration is based are assumed, hypothetical or contingent, then that may well be relevant to the exercise of the court's discretion.

Factual background to the Dextrose Remission Issue

A note on terminology

35. Dextrose is also known as glucose. I shall generally refer to it as dextrose except when quoting from sources that refer to it as glucose or when discussing diabetes. When I refer to saline, I am referring to normal saline unless otherwise specified. Likewise, when I refer to dextrose solution, I am referring to a 5% solution unless otherwise specified. In the context of this case, saline and dextrose solution are both properly described as "diluents", which is how I shall refer to them, but witnesses on both sides and some of the literature references sometimes referred to them as "solvents", although strictly speaking the solvent in both cases is water.

Alimta and the Alimta SmPC

36. According to the SmPC for Alimta, Alimta is indicated for the treatment of non-small cell lung cancer ("NSCLC") and malignant pleural mesothelioma. In some circumstances it is indicated in combination with cisplatin and in other circumstances as monotherapy.
37. As stated above, Alimta is supplied in lyophilised powder form. It is supplied in vials containing 100 mg or 500 mg of pemetrexed disodium. Section 6.6 of the SmPC for Alimta states as follows in relation to the 100mg vials:

"1. Use aseptic technique during the reconstitution and further dilution of pemetrexed for intravenous infusion administration.

…

3. Reconstitute 100mg vials with 4.2 ml of sodium chloride 9 mg/ml (0.9%) solution for injection, without preservative, resulting in a solution containing 25 mg/ml pemetrexed. Gently swirl each vial until the powder is completely dissolved. The resulting solution is clear and ranges in colour from colourless to yellow or green-yellow without adversely affecting product quality. The pH of the reconstituted solution is between 6.6 and 7.8. Further dilution is required.

4. The appropriate volume of reconstituted pemetrexed solution must be further diluted to 100 ml with sodium chloride 9 mg/ml (0.9%) solution for injection, without preservative, and administered as an intravenous infusion over 10 minutes.

…

6. Parenteral medicinal products must be inspected visually for particulate matter and discolouration prior to administration. If particulate matter is observed, do not administer.

7. Pemetrexed solutions are for single use only. Any unused medicinal product or waste material must be disposed of in accordance with local requirements."

Accordingly, following the SmPC, Alimta is prepared for administration in a pharmacy in two steps: first, reconstitution in saline; and secondly further dilution in saline to 100 ml.

38. Section 6.3 of the SmPC for Alimta states as follows:

"Shelf life

Unopened vial 100mg

3 years.

Unopened vial 500mg

3 years.

Reconstituted and infusion solutions

When prepared as directed, reconstituted and infusion solutions of ALIMTA contain no antimicrobial preservatives. Chemical and physical in-use stability of reconstituted and infusion solutions of pemetrexed were demonstrated for 24 hours at refrigerated temperature. From a microbiological point of view, the product should be used immediately. If not used immediately, in-use storage times and conditions prior to use are the responsibility of the user and would not be longer than 24 hours at 2°C to 8°C."

39. Section 4.4. of the SmPC for Alimta contains the following warning in relation to 500 mg vials (there is no corresponding warning for 100 mg vials):

"500mg vial: This medicinal product contains approximately 54mg of sodium per vial. To be taken into consideration by patients on a controlled sodium diet."

Actavis' applications for marketing authorisations

40. Actavis applied for marketing authorisations for the Actavis Product via both the centralised procedure ("CP") and (through Caduceus Pharma Ltd) the decentralised procedure. By the time of the trial, Actavis had obtained a UK marketing authorisation, was close to receiving a CP marketing authorisation covering France, Italy and Spain and expected to receive a German marketing authorisation shortly thereafter.

The Actavis Product and UK SmPC

41. As indicated above, the Actavis Product is supplied as a liquid concentrate. This contains pemetrexed diacid dissolved in water for injection. The Actavis Product is called "pemetrexed ditrometamol" in the UK SmPC, rather than pemetrexed diacid. This was a change imposed by the regulator, but it does not alter the fact that the active ingredient is the diacid. The Actavis Product is authorised for the same indications as Alimta. It is supplied in vials containing 100 mg, 500 mg and 1000 mg pemetrexed diacid.
42. Section 6.6 of the UK SmPC for the Actavis Product states as follows:

"1. Use aseptic technique during the reconstitution and further dilution of pemetrexed for intravenous infusion administration.

…

3. Pemetrexed must only be diluted with 5% glucose solution, without preservative. The appropriate volume of pemetrexed must be diluted to 100 ml with 5% glucose solution administered as an intravenous infusion over 10 minutes. The diluted medicinal product contains 5 g glucose per dose. This should be taken into account in patients with diabetes mellitus.

…

5. Parenteral medicinal products must be inspected visually for particulate matter and discolouration prior to administration. If particulate matter is observed, do not administer.

6. Pemetrexed solutions are for single use only. Any unused medicinal product or waste material must be disposed of in accordance with local requirements."

Accordingly, following the UK SmPC, the Actavis Product is prepared for administration in a pharmacy in one step, namely dilution in dextrose solution to 100 ml.

43. There are no instructions in the UK SmPC as to how the product could be diluted in saline.
44. Section 6.3 of the UK SmPC for the Actavis Product states as follows:

"Shelf life

Unopened vial

18 months.

Shelf life after dilution of concentrate

Stability of infusion solutions of pemetrexed was demonstrated for 24 hours at room temperature and 14 days at refrigerated temperature (2- 8°C). From a microbiological point of view, the product should be used immediately. If not used immediately, in-use storage times and conditions prior to use are the responsibility of the user and would normally not be longer than 24 hours at 2°C to 8°C, unless dilution has taken place in controlled and validated aseptic conditions."

45. Section 6.3 of the original draft UK SmPC provided by Actavis to Lilly stated that the Actavis Product could be stored for 42 days at refrigerated temperature. At a late stage, Actavis were asked by the regulator to change the wording to "stability infusion solution of pemetrexed was demonstrated for 24 hours at room temperature and 14 days at refrigerated temperature". This change is not significant for the reasons explained below.

Proposed variation to the UK SmPC for the Actavis Product

46. As a result of points taken in Lilly's evidence, Actavis applied on 22 January 2016 to vary the UK SmPC by removing from section 6.6 the words "The diluted medicinal product contains 5 g glucose per dose. This should be taken into account in patients with diabetes mellitus." A similar change is to be made to section 4.4.
47. Mr Wilson's evidence was that he had been informed by Actavis' regulatory department that the reason this wording was included in the UK SmPC was that the regulatory team thought that it was desirable "following a misunderstanding between the regulatory and legal teams" and that no regulator had requested that the wording be included in any SmPC for the Actavis Product. Given (i) the hearsay nature of this evidence, (ii) the absence of disclosure by Actavis on the point and (iii) an argument raised by Lilly after the conclusion of the trial that the evidence was contradicted by a document published on the European Medicines Agency's website on 29 January 2016, I place no weight upon it. More importantly, however, Mr Wilson exhibited the relevant European Commission guidelines and explained why Actavis considered that they would be successful in making this change. I should explain before proceeding further that Mr Wilson did not profess to be a regulatory expert, and I therefore place no weight upon his opinion as opposed to the objective considerations which he identified.
48. The relevant European Commission guidelines are entitled "Excipients in the label and package leaflet of medicinal products for human use" dated July 2003 (CPMP/463/00 Final) issued pursuant to Article 65 of Directive 2001/83/EC. The Explanatory Notes explain that the Annex sets out the name of the excipient, route of administration, threshold for the provision of information and information to be provided where the amount of the excipient is equal to or above the threshold value. The relevant entry in the Annex is as follows:



49. Mr Wilson explained that Actavis considered that they would be permitted to remove the wording in question for two reasons. The first is that dextrose is not listed as an excipient in section 6.1 of the UK SmPC. The second is that, in any event, the 5g threshold is not reached because, as Prof Seckl explained, if the direction to dilute the Actavis Product "to 100 ml with 5% glucose solution" is followed, the resulting preparation will contain somewhat less than 100 ml of dextrose solution, and hence less than 5 g of dextrose. In the case of an average-sized woman, it would contain 3.4 g, and in the case of an average-sized man, it would contain less than 3.4 g.
50. Counsel for Lilly submitted that Actavis would not be permitted to remove the wording in question, but he had no answer to the first point explained by Mr Wilson. In relation to the second point, counsel relied on evidence that, in practice, some pharmacists do not follow the corresponding instructions in the Alimta SmPC, but instead add the reconstituted solution to a full 100 ml bag of saline. If the Actavis Product were to be added to a full 100 ml bag of dextrose solution, then the resulting preparation would contain 5 g of dextrose. But the fact remains that, if the direction in the SmPC is followed, the 5 g threshold will not be reached.
51. Accordingly, I conclude that it is probable that Actavis will be permitted to make this change to the UK SmPC. This conclusion is supported by the fact that the German regulator has already permitted the same change to be made to the German SmPC prior to approval (as to which, see below).

Proposed variations to the CP SmPC for the Actavis Product

52. Mr Wilson gave evidence that Actavis intend to apply for two variations to the CP SmPC once it is granted. The first is to make the same change to the glucose wording as I have discussed in relation to the UK SmPC. The second concerns the stability period specified in the CP SmPC. By contrast with the UK SmPC, the CP SmPC states that the Actavis Product is stable "for 24 hours at refrigerated temperature". Mr Wilson said that he had been informed by Actavis' regulatory team that an issue about the stability data had been raised very late by the regulator, and that Actavis decided to accept a limitation to 24 hours in order to ensure that the SmPC was approved as quickly as possible. I place little weight on this evidence. More importantly, Mr Wilson stated that Actavis intended to apply for a variation to bring the CP SmPC into line with the UK SmPC in this respect i.e. to specify a 14 day stability period. Given that both the UK and German regulators have accepted a stability period of 14 days or more, it is probable that that this variation will be approved as well. Accordingly, in the remainder of this judgment I shall proceed on the basis that there will be no material difference between the UK SmPC and the CP SmPC.

The German SmPC for the Actavis Product

53. The German SmPC for the Actavis Product is not directly relevant to the Dextrose Remission Issue, since Actavis do not seek a DNI in respect of the German designation of the Patent. It is nevertheless evidentially relevant in so far as it casts light on the UK and CP SmPCs. On 5 February 2016 the German regulator approved the German SmPC. Section 6.3 states that the Actavis Product is stable for three days at room temperature and 28 days at refrigerated temperature. As I understand it, Actavis are content with a stability period of 28 days and do not intend to seek a variation to the German SmPC in this respect. Section 6.6 does not contain any statement about the quantity of glucose in the dose or any warning about patients with diabetes, as a result of a successful application by Actavis to remove this wording filed on 5 January 2016.

Administration of pemetrexed

54. Pemetrexed is administered as an intravenous infusion over 10 minutes on the first day of a 21 day cycle.

Dexamethasone

55. Pemetrexed is usually administered in conjunction with dexamethasone, which is a steroid that is given to reduce the side effects of chemotherapy. Dexamethasone is given in high doses with chemotherapy, including with pemetrexed, and it is well known to cause a significant increase in the patient's blood sugar levels. As such, oncologists prescribing pemetrexed (or indeed other chemotherapy drugs) together with dexamethasone would be aware that all patients' blood sugar levels should be measured before treatment. For patients with diabetes, the administration of dexamethasone may require more frequent monitoring of their blood sugar than their normal routine, and they may need to adjust their eating habits or medication.

Preparation of cytotoxic drugs for administration

56. Cytotoxic drugs for intravenous administration such as pemetrexed are prepared for administration within a hospital pharmacy, and not on the wards. They are usually prepared in specialist units within pharmacies called aseptic units. Aseptic preparations are exempt from licensing requirements on the basis that they are prepared by, or under the supervision of, a pharmacist who takes full responsibility for the quality of the product.
57. When drugs are prepared in an aseptic unit, this is done by reference to a formal protocol and a worksheet (also referred to as a "standard operating procedure" or SOP) which sets out a detailed step-by-step procedure describing how to make up the drug, how to store it and its stability. It is common ground that, for any new drug, a new worksheet will be prepared and that, in the case of a new generic drug, if the SmPC is different from the reference drug, then the generic will be treated as a new drug, and so a new worksheet will be prepared. Since the Actavis SmPC is significantly different from the Alimta SmPC (it is for pemetrexed diacid, not disodium; the product is in liquid, not powdered form; it only requires a one-step dilution rather than two steps; and it specifies a different diluent), there is no dispute that a new worksheet will be prepared by any hospital that stocks the Actavis Product.
58. It is common ground that, when the worksheet for a drug is prepared, the pharmacist will always take the SmPC relating to the preparation of the product as their starting point. Therefore there is no dispute that, at least initially, all pharmacists will prepare their worksheet for the Actavis Product using dextrose solution in accordance with the SmPC. Nor is there any dispute that pharmacists will not use an alternative diluent to that specified in the SmPC unless and until reliable stability data is available for the drug with that alternative diluent.
59. There is no dispute that the worksheet must be adhered to, save in exceptional circumstances. Furthermore, the worksheets and other related documentation prepared by pharmacies are regularly audited to ensure quality control. The strict procedures and quality assurance services that are in place have the effect of keeping to a minimum the numbers of errors made in terms of following the worksheets for aseptic preparations. Mr Bateman analysed the frequency of errors relating to the use of the "wrong" diluent (i.e. not the precise diluent shown in the worksheet) in the preparation of aseptic drugs in the period 2010-2015 using data the National Aseptic Error Reporting Scheme ("NAERS"), for which he is responsible. Out of 1,839,831 doses prepared in NHS pharmacy aseptic units recorded in the database during that period, 962 errors (0.05% of all doses) were reported in the "incorrect diluent/infusion fluid" category. All except seven of these were detected during routine checks at the assembly or preparation stage. Of those seven, three were detected during labelling, four during the final check and none after release for administration to a patient. Mr Bateman accepted in cross-examination that the nature of the NAERS scheme was such that it was likely to under-report errors, but the data he provided are the best data available as to the rate of diluent error. In any event, Lilly did not contend that it was obvious that the Actavis Product would be diluted in saline in error.
60. It is common ground that there are some circumstances in which a pharmacist may decide to deviate in some way from the SmPC, either for a specific patient or more generally by amending the worksheet itself so that all future preparations of the drug are made in the new way. In the case of an individual change for a specific patient, this would only arise if there was a particular patient-specific issue leading to a request from the physician to the pharmacist. In the case of a change to the worksheet, this would not be done without good cause. This is for two reasons. First, the data in the SmPC have been scrutinised by the regulators, and so can be taken to be reliable. Secondly, because the liability for the product will shift from the manufacturer to the pharmacist if the product is prepared in a way that deviates from the SmPC. (Actavis contend that in Italy and Spain there are also legal reasons, which I will consider below.)

The use of saline and dextrose solution as diluents for chemotherapy drugs

61. It is common ground that saline and dextrose solution are both common diluents, and that of the two saline is the more common. Between 20 and 30% of chemotherapy drugs are diluted with dextrose solution, with about double that proportion being diluted with saline.

Stability data available to pharmacists

62. It is common ground that:

i) pharmacists will initially take the stability data for a product from the SmPC;

ii) for any particular product, a pharmacist may contact the manufacturer to find out if there is further stability data available, in particular if the SmPC specifies a short stability period and the pharmacist wants to know if there are data showing that the product has longer stability than that specified in the SmPC;

iii) in the absence of data from the manufacturer, a pharmacist may look for published data using resources such as the Stabilis website; and

iv) if there are no published data, a pharmacist like Dr Vigneron may consider doing a stability study himself or herself.

Storage of aseptic preparations

63. Once made up, unlicensed aseptic units in the UK are only allowed to store aseptic preparations for a maximum of seven days. Even licensed units (whether NHS or commercial) would not be concerned about needing stability of over 14 days. This is for two reasons. First, a drug like pemetrexed is normally prepared for administration for each patient or group of patients, it is not held as stock once it has been prepared for administration. (By way of exception to this, such drugs may be supplied in a "dose banded" form which is pre-diluted to deliver specific doses, but it was not suggested that this was common with pemetrexed.) It follows that it is only necessary to store the solution for a few days, and only then when occasion demands, such as if a patient is ill and so the treatment must be postponed or if it is desired to administer the drug over a weekend when the aseptic unit is closed. Secondly, even if the drug is physically and chemically stable in the diluent, storage risks bacteriological contamination. This is particularly true of storage at room temperature, which is why storage at refrigerated temperature is frequently indicated. It appears that the situation is similar in France, Italy and Spain.

Published stability data for Alimta

64. As set out above, the SmPC for Alimta specifies a stability of 24 hours at refrigerated temperature. Data have, however, been published in two papers (Y. Zhang and L.A. Trissel, "Physical and chemical stability of pemetrexed solutions in plastic syringes", Ann. Pharmacother., 39(12), 2026-8 (2005) and L. Rondelot et al, "Stability of pemetrexed 25 mg/mL stored in a PVC container after storage for one month at 2-8^oC", Eur. J. Hosp. Pharm. Sci., 13(1), 14-16 (2007)) showing that Alimta is stable for up to 28 days or 31 days (depending on the container) at refrigerated temperature and protected from light. This is subject to a qualification which I shall discuss below.

Actavis' stability data for the Actavis Product

65. Mr Wilson gave evidence that Actavis have data showing that the Actavis Product is stable in dextrose solution for up to 42 days at 2-8^oC and that Actavis will inform any hospital or pharmacy which asks about the stability of the Actavis Product about this longer stability period and offer to provide the data. Counsel for Lilly pointed out that these data had not been disclosed to Lilly in these proceedings, although he did not suggest that they did not exist. Given that the data have not been disclosed, I accept that no assumptions can be made about the reliability of the data beyond what is apparent from the positions taken by the regulators discussed above.
66. Actavis do not advance any case that the Actavis Product is significantly more stable in dextrose solution than in saline. Actavis have not admitted that they have any stability data for the Actavis Product in saline, although it is probable that they do.

Supply and distribution of the Actavis Product

67. Actavis UK Ltd has launched the Actavis Product in the UK and is supplying the product through a third party distributor, Pharmaxo Pharmacy Services Ltd ("Pharmaxo"). Actavis UK has entered into a contract with Pharmaxo under which Actavis UK supplies the Actavis Product to Pharmaxo which supplies it to customers. Nevertheless the principal sales and marketing effort is conducted directly between Actavis' sales force and the customers.
68. [REDACTED].
69. Actavis are supplying the Actavis Product through two different sales routes in the UK. One is through sales made to individual hospitals (or groups of hospitals). The other is through a tender process whereby the NHS opens the supply of pemetrexed to all those interested in bidding for the supply contract. [REDACTED]. It appears that similar sales routes will be followed in France, Spain and Italy, although the details will vary from country to country.

Steps taken by Actavis to prevent the Actavis Product from being diluted in saline

70. Actavis have taken, or are going to take, a number of steps to prevent the Actavis Product from being diluted in saline.
71. First, as explained above, the SmPC for the Actavis Product states that the product must be diluted with dextrose solution, and there is no instruction for how to dilute the product in saline. (The same applies to the package leaflet.) Furthermore, the SmPC only contains stability data for dilution in dextrose solution.
72. Secondly, Actavis have taken the additional precaution of sending letters to the relevant competent authorities and medical centres in the UK explaining that the Actavis Product should not be diluted with saline, and Actavis will do the same in all the relevant jurisdictions upon launch in those countries. The letters to competent authorities are in the following terms:

"IMPORTANT INFORMATION IN RELATION TO [NAME (PEMETREXED)]

Dear [ ]

[NAME] —Pemetrexed 25 mg/ml concentrate for solution for infusion indicated for the treatment of malignant pleural mesothelioma and non-small cell lung cancer

Following the recent grant of marketing authorisation, Actavis is launching [NAME], a generic version of pemetrexed, in the [UK/France/Spain/Italy].

The purpose of this communication is to request the issuance of central guidance and/or notification for appropriate dissemination in relation to the directions and information for preparation and storage in the Summary of Product Characteristics (SmPC) and Package Leaflet (PL). Actavis makes this request since there are some differences in comparison to Alimta® of which all relevant personnel should be aware. In particular, Actavis requests that all relevant institutions and personnel should be informed that [NAME] must only be diluted in 5% glucose solution.

Directions for preparation

[NAME] is supplied in vials as a liquid concentrate for dilution. No reconstitution step is required.

The relevant sections of the SmPC and PL direct as follows:

'[Pemetrexed] must only be diluted with 5% glucose solution, without preservative. The appropriate volume of pemetrexed concentrate must be diluted to 100 ml with 5% glucose solution and administered as an intravenous infusion over 10 minutes. The diluted medicinal product contains 5 g glucose per dose. This should be taken into account in patients with diabetes mellitus.'

Shelf life/Storage

The relevant sections of the SmPC and PL provide extended stability information for the prepared infusion solution prepared with 5% glucose solution and no other diluent:

'Chemical and physical in-use stability of infusion solution of pemetrexed was demonstrated for 24 hours at room temperature and 14 days at refrigerated temperature (2-8°C). From a microbiological point of view, the product should be used immediately. If not used immediately, in-use storage times and conditions prior to use are the responsibility of the user and would not be longer than 24 hours at 2°C to 8°C, unless dilution has taken place in controlled and validated aseptic conditions.'

Patent Rights of Eli Lilly

There has been a finding by the English Court of Appeal that use of saline for dilution of [NAME] will constitute infringement of Eli Lilly's patent EP 1313 508 in [the UK, France, Italy and Spain]. Accordingly, use of saline as a diluent within a hospital or medical centre, whether in a pharmacy or otherwise, in [the UK/France/Italy/Spain] will be unlawful. On the basis of the Court of Appeal judgment, it would also be an infringement for Actavis to sell [NAME] to a hospital or medical centre insofar as it knew or it was obvious that the hospital or medical centre would use saline for dilution of the product sold. In the circumstances if hospitals or medical centres were to use saline supply could be interrupted. Please would you ensure that this is drawn to the attention of all relevant institutions and personnel and direct that for this reason as well they do not specify or use saline but only 5% glucose.

If you have any questions in relation to Actavis' pemetrexed product [NAME] and/or there are any issues with this request, please contact our Medical Information Department on [number relevant to UK/France/Italy/Spain] or email: [address relevant to UK/France/Italy/Spain]."

73. Thirdly, Actavis have implemented a policy that, when representatives from Actavis (or Actavis' affiliates) contact hospitals promoting the Actavis Product, they explain the requirement that the Product be diluted in dextrose solution and make sure that the hospital is aware that it cannot be diluted in saline. This includes a stipulation that, if anyone contacts Actavis asking about use in saline, they will be informed that the Product must not be diluted in saline, and will be provided with a copy of the letter referred to in the preceding paragraph.
74. Fourthly, Actavis have set up the UK supply operation so that supply to customers can be turned on and off at will. Actavis will ensure that supplies in France, Spain and Italy are similarly controllable. If Actavis become aware of the use, or intended use, of saline with the Actavis Product by any customer, Actavis will threaten to withhold the Product from that customer unless the customer confirms that they will use dextrose solution, and if that threat is not sufficient then Actavis will withhold the supply of the Product. It is likely that Actavis will know about any proposed use of saline before it happens, since it was common ground between the pharmacist experts that anyone intending use outside the SmPC would be likely first to contact the supplier of the drug.
75. Fifthly, Actavis are checking that worksheets have been updated in an appropriate manner as regards the Actavis Product prior to supply being commenced, and following up on a regular basis to check that there have not been any further amendments as regards the diluent.
76. Sixthly, Actavis are pursuing the possibility of contractually binding their customers to terms that prevent them from diluting the Actavis Product in saline (including to prevent them from conducting stability tests in saline). This process is complicated and has particular difficulties (a) in relation to sales done by tender and (b) if Actavis need to or wish to supply the Product through their usual wholesalers. In the UK, however, Actavis is attempting contractually to bind the customers through a letter which is sent by Pharmaxo to all prospective customers. There is a dispute about the enforceability of these terms, but in any event Actavis' position is that the letter (whether or not it is strictly enforceable) is an important additional deterrent to pharmacists diluting in saline.
77. Lilly has not suggested any further steps that Actavis should take despite being requested to do so.

Steps taken by Actavis in relation to the availability of stability data of its product in saline

78. As explained above, Lilly's case hinges on the availability of stability data for the Actavis Product in saline. Since Lilly formulated its case in this way, Actavis have taken a number of steps to avoid publication of stability data of the Product in saline.
79. First, Actavis have undertaken to the Court until judgment not to disclose any stability data they may have for the Actavis Product in saline save to appropriate regulators and/or as required by the Court on terms as to confidence. As I understand, Actavis is willing to extend this undertaking after judgment. Lilly suggests that this will put Actavis in breach of the Association of the British Pharmaceutical Industry ("ABPI") code of practice, but Lilly has no pleaded case on this issue. Nor has Lilly explained why this should be so (indeed, counsel did not even refer me to the relevant provisions of the code), although I was informed that Lilly had written to the ABPI on the matter. Lilly's stance is remarkable given that Actavis' purpose in withholding such data is in order to respect Lilly's patent rights and to avoid acting unlawfully.
80. Secondly, the letter that is being sent out to all prospective customers referred to above states that it is a term of the contract of sale that they are not to carry out stability studies with the Actavis Product in saline. As explained above, whilst there is a dispute as to whether this constitutes a contractual term that is enforceable, Actavis contend that the letter itself will deter customers from conducting such stability studies.
81. Thirdly, Actavis have put in place procedures such that, if any hospitals or pharmacies enquire about the stability of its Product in saline, Actavis will remind the relevant enquirer that they are not permitted to perform stability testing in saline and will provide another copy of the letter referred to in the preceding paragraph. An example of this is discussed in paragraph 156 below.

Steps taken by Lilly to prevent the Actavis Product from being diluted in saline

82. There is no evidence that Lilly has taken, or is planning to take, any steps to prevent the Actavis Product from being diluted in saline. There is a striking contrast in this regard between Lilly's inaction and the extensive steps taken by Pfizer in the recent case concerning pregabalin: see Generics (UK Ltd v Warner-Lambert Co LLC [2015] EWHC 2548 (Pat) at [459]-[509]. Counsel for Actavis submitted that it was to be inferred that Lilly did not want to prevent infringement of the Patent. I am not sure that that inference is justified, but I do consider that it is legitimate to infer that Lilly is content for Actavis to have to assume the entire burden of policing Lilly's patent. I shall return to this point below.
83. Actavis have expressed concern that Lilly may instigate the publication of stability data for the Actavis Product in saline. Lilly has, however, stated that it has no intention of carrying out, or encouraging, studies of the stability of the Actavis Product in saline. Accordingly, I do not need to consider what the position would be if Lilly were to do so.

Diabetes

84. It is common ground that around 25% of lung cancer patients in the UK also have diabetes. It is likely that the proportion is similar in France, Italy and Spain.
85. Diabetes is a condition in which the body's ability to remove glucose from the blood and store it (or use it) within cells is impaired. This results in the blood glucose levels (also referred to as the blood sugar levels) of a person being too high. If high levels of blood glucose are sustained for a long period of time (i.e. for many years), serious organ damage may result. Extremely high blood glucose levels sustained for shorter periods of time (i.e. days) can cause dangerous metabolic disruption.
86. Carbohydrates can be divided into simple and complex carbohydrates. Simple carbohydrates are monosaccharides (single sugar units, such as glucose) and disaccharides (double sugar units, such as sucrose). Complex carbohydrates contain three or more sugar units. A common example of a complex carbohydrate is starch, which is a polymer of glucose. When digested, complex carbohydrates are broken down into simple carbohydrates such as glucose. Accordingly, the ingestion of both simple and complex carbohydrates will result in a change in blood glucose levels.
87. Glucose normally enters the blood stream following the digestion of carbohydrate. It is important that the level of glucose in the blood is controlled. This requires the removal of glucose from the blood as its level rises following a meal. Glucose is taken up by cells where it is either stored as glycogen or used as a source of energy. The cellular uptake of glucose is mediated mainly by the hormone insulin.
88. A person's blood glucose levels will fluctuate over the course of a day. In the fasted state, a healthy person will have a blood glucose level of about 4-6 mmol/l. After a meal, however, blood glucose levels are usually significantly higher than in the fasting state.
89. Diabetes results from either a complete lack of insulin (type 1 diabetes) or insufficient insulin and/or insulin resistance (type 2 diabetes). Type 2 diabetes is much more common, accounting for 90% of adult patients with diabetes. In diabetes, blood glucose levels usually remain higher than is normal, even when the condition is well treated and this leads to a risk of damage to various body organs.
90. Diabetes UK recommends that for people with diabetes who are overweight, carbohydrate intake (which includes any sugar) should generally be limited to 230 g for women and 300 g for men. Within this, Diabetes UK recommends that sugar intake is limited to around 25 g per day. For people with diabetes who are not overweight, slightly more carbohydrates may be consumed, but this should still be part of a healthy balanced diet.
91. Besides eating a balanced diet and taking exercise, the management of diabetes typically consists of monitoring blood glucose levels and administration of medications as required to keep levels as close to the normal range as possible. Type 1 diabetes requires patients to administer multiple daily insulin injections, whereas type 2 diabetes is often treated with tablets, such as metformin. Patients are generally taught and encouraged to take responsibility for managing their condition and making adjustments to their treatment regimes to try to obtain good control of blood glucose levels. The vast majority of diabetic patients are able to take some such responsibility and live a relatively normal life.

Assessment of the Dextrose Remission Issue

92. The Dextrose Remission Issue can be formulated in this way: do Actavis know, or would it be obvious to a reasonable person in the circumstances, that the Actavis Product will be diluted in saline by some of Actavis' customers? As I have explained, speculative, maverick or unlikely uses of the Actavis Product are to be disregarded. Mr Wilson gave unchallenged evidence that Actavis do not know that the Actavis Product will be diluted in saline. Accordingly, the issue is whether it is obvious that this is likely to occur at least in some cases. Given that I am concerned with what will happen in the future, the question is whether this is foreseeable. As I have explained, the burden lies on Actavis to establish that it is not foreseeable. Furthermore, the question must be considered bearing in mind that the Patent will not expire until June 2021.
93. As I have explained, it is common ground that, to begin with, pharmacists will follow the SmPC for the Actavis Product and will dilute it with dextrose solution. The likelihood of pharmacists departing from the SmPC depends on the answers to three main questions. First, is it likely that stability data for the Actavis Product in saline will become available? Secondly, if stability data for the Actavis Product in saline became available, would there be a motivation to use saline rather than dextrose solution as the diluent? Thirdly, even if stability data became available and there was a motivation to use saline rather than dextrose solution, is it likely that saline would actually be used as the diluent despite the steps which Actavis are taking to prevent this? I consider each of these questions in turn, but before doing so it is convenient to address a number of preliminary matters.

The role of the oncologist

94. Pemetrexed is prescribed by an oncologist. Furthermore, the oncologist will decide upon the other aspects of the patient's treatment, such as administration of dexamethasone and cisplatin. Prof Seckl and Prof Thatcher were agreed, however, that pharmacists are ultimately responsible for decisions about the preparation of cytotoxic drugs in aseptic units, although a clinician would be involved in finalising a draft protocol.

The role of the endocrinologist

95. It is common ground that endocrinologists are not involved in writing or approving protocols for the preparation of chemotherapy drugs. There was a dispute between the oncologists as to whether or not an oncologist would be concerned about administering pemetrexed diluted in dextrose solution to patients with diabetes which I will consider below. Prof Seckl and Prof Thatcher were agreed, however, that expert advice would be obtained from an endocrinologist or diabetologist if there were any particular concerns about treating a diabetic patient. Furthermore, Prof Thatcher himself pointed out that, if an endocrinologist was asked about a diabetic patient's chemotherapy treatment regimen, the endocrinologist would not simply defer to the oncologist.

The evidence of the pharmacists

96. At the instigation of Lilly, directions were given for separate expert evidence from pharmacists in the UK, France, Italy and Spain. By the end of the evidence, however, it was clear that there was very little difference between the experts from the different countries on the matters that are relevant to the Dextrose Remission Issue. (In saying this, I put on one side Actavis' case with regard to Italian and Spanish law, which I shall consider separately below.) Accordingly, I shall deal with all four countries together.

The importance of the SmPC

97. The starting point is that, as indicated above, all the pharmacists agreed that the SmPC was the primary source of information for preparing a worksheet and that the SmPC would only be departed from where there were good reasons to do so.
98. Furthermore, it was clear from the pharmacists' evidence that it is rare to depart from the SmPC except where the SmPC specifies a short period of stability in a diluent, but data are available establishing that the drug in question is stable for a longer period in that diluent. In those circumstances, it is common to depart from the SmPC to that extent. Thus Lilly relied upon a paper by J. Vigneron et al, "SFPO and ESOP recommendations for the practical stability of anticancer drugs: an update", Eur. J. Onco. Pharm., 8(2), 3-13 (2014), Table 1 of which gives recommendations only a minority of which are to follow the SmPC; but most of the other recommendations are of a longer stability period in the diluent (or one of the diluents) specified in the SmPC. (Even then, as I shall explain below, the evidence shows that the SFPO/ESOP recommendations are not necessarily followed by pharmacists in practice.)
99. Other, less common, circumstances where the SmPC may be departed from are where the drug is being administered for a different indication and/or by a different route of administration to the indication or route of administration specified in the SmPC.

Is it likely that stability data for the Actavis Product in saline will became available?

100. As I have explained, Actavis have so far undertaken that they will not make any stability data which they may have for the Actavis Product in saline available, and I understand that Actavis are willing to continue that undertaking. It follows that a pharmacist who was interested in ascertaining the stability of the Actavis Product in saline would have to obtain data from elsewhere. There would be only two possibilities: to obtain data from a published source or to carry out a stability study to generate data.
101. There was a consensus between the pharmacy experts who had experience of stability studies that, in general, hospitals and other third parties would not carry out stability studies without having a good reason to do so. Thus Dr Vigneron said that the realisation of stability studies depends on the availability of the product, the availability of the laboratory and a student to do the work, and the financial resources available; and that therefore a choice has to be made based on the cost and usefulness of the information which you hope will come of it. In the case of an expensive drug like pemetrexed, the cost of the product is a significant factor. If it is hoped that the study will generate extended stability data which avoids wastage of the drug in clinical use, then the cost may be justified. Otherwise, the cost will be more difficult to justify.
102. In the case of the Actavis Product, the evidence was clear that there would be no motivation from a stability perspective to carry out further stability studies. In relation to the UK, the SmPC for the Actavis Product specifies that it is stable in dextrose solution for 14 days. The UK experts agreed that 14 days was more than enough. Indeed, Mr Bateman was not aware of any hospital storing Alimta for more than 24 hours once made up, and Ms Nijjar confirmed that Barts Healthcare did not store it for longer than 24 hours. In relation to France, Italy and Spain, for the reasons given above I consider it is probable that the CP SmPC will be varied to specify 14 days. Even if it is not, a pharmacist doing a search for stability data for the Actavis Product would be likely to find the UK SmPC and could then rely on that. Lilly suggested that certain minor differences between the CP SmPC and the UK SmPC meant that the pharmacist would not be sure that they related to the same product; but if the pharmacist had any doubt about this, he or she would be likely to ask Actavis, who would confirm that they did. Furthermore, it is also clear from the evidence of the French, Italian and Spanish experts relating to Alimta that there is no need for any longer period of stability: on the evidence, the longest period for which Alimta is stored in those countries is 48 hours.
103. Even if someone was motivated to carry out a study of the stability of the Actavis Product in dextrose solution in order to establish that it had a longer period of stability than specified in the SmPC, it does not follow that that motivation would extend to studying the stability of the Actavis Product in saline. The only motivation suggested by Lilly (other than the motivation relating to diabetic patients considered below) was that researchers would want to study the stability in saline for comparison with the stability in dextrose solution. I accept that there are a few examples in the literature where the stability of a drug has been studied in both saline and dextrose solution, but this was not established to be a routine course. Moreover, the cost, effort and time required would be increased, at least to some extent.
104. This may explain why there are very few examples of reliable stability studies (which the evidence indicates means studies carried out since 2000, due to higher standards being required nowadays) of chemotherapy drugs in a diluent not specified in the SmPC being carried out purely for stability purposes. (There are also a few examples of studies in alternative diluents for off-label indications and/or routes of administration, but that is a different matter.)
105. A further factor in the cost-benefit analysis in the present case is that the researcher would be likely to contact Actavis to ask for their data and/or to obtain the Actavis Product. They would then learn that Actavis would not let them have the Actavis Product for saline stability studies, that Actavis' terms of supply forbid its use for this purpose and that the Actavis Product could not be supplied for clinical use in saline because of the Patent. This would discourage the researcher from wasting precious resources on attempting such a study. In those circumstances, a researcher who was interested in establishing whether the Actavis Product was stable for a longer period than that specified in the SmPC would be likely to stick to dextrose solution.
106. Even if stability data were produced by someone other than Actavis, the experts agreed that they would want to be satisfied of its reliability before relying upon it. Thus Dr Vigneron agreed that the quality of stability studies varies, as can be seen from his Stabilis website, which grades stability studies from A to D. Furthermore, even if a study establishes that a drug is physically and chemically stable in a diluent, there may be issues as to the formation of microparticles which inhibit reliance upon that longer stability period.
107. Alimta is an example of this. As noted, above, stability data has been published in two papers showing that it is stable for at least 28 days at refrigerated temperature. Accordingly, the SFPO/ESOP recommendation is to follow the published data, rather than the SmPC. The SFOP/ESOP publication notes, however, that microparticles may be formed, and therefore says that an in-line filter has to be used. As discussed above, however, in practice there is no evidence of anyone storing Alimta for longer than 48 hours. It appears that concern over microparticles is one reason for this, although no doubt other reasons are concern over bacterial growth and the sheer lack of any real need to store Alimta for longer than 48 hours.
108. For the reasons given above, I conclude that it is unlikely that data showing that the Actavis Product is stable in saline will be published in the foreseeable future, and still less likely that such data would be relied upon by pharmacists from a stability perspective. I cannot rule out the possibility that this may occur before June 2021, however.
109. As to the likely timing of the publication of such data, if it does occur, the witness who was best qualified to express a view on this was Dr Vigneron. His estimate was that it was likely to be two to three years from the launch of the Actavis Product before such data was published. It follows that it is unlikely that such data will be published any earlier than this, even if it does happen.

Would there be a motivation to use saline rather than dextrose solution as the diluent for the Actavis Product?

110. Although it was faintly suggested in Lilly's evidence in chief that three motivations for using saline to dilute the Actavis Product were that (a) saline was the "standard" or "default" diluent, (b) dextrose solution posed a greater risk of phlebitis (inflammation of the veins) and (c) dextrose solution posed a greater risk of bacterial growth, all of these suggestions fizzled out at trial when it became clear that dextrose solution was not only a commonly used diluent, albeit not as common as saline, but also was the only diluent recommended for use with half a dozen chemotherapeutic drugs.
111. Furthermore, the evidence establishes that the mere availability of stability data for the other diluent, or for a different concentration of the same diluent, does not lead to hospitals changing the diluent they use for treatment. A number of examples of this were canvassed in evidence:

i) There are stability data available for Alimta in dextrose solution, yet there is not a single example of Alimta being diluted in dextrose solution by a hospital.

ii) Dr Vigneron gave evidence that there was some published stability data for thiotepa in dextrose solution (which is not in the SmPC), but he explained that he had not actually used thiotepa in dextrose solution as his hospital stuck to the diluent specified in the SmPC (namely saline).

iii) Dr Vigneron also gave evidence that there was a stability study for melphalan at a higher concentration of saline than that specified in the SmPC, but that this had not led to the higher concentration being used in practice.

iv) Similarly, Dr Vigneron gave evidence that bendamustine had been tested in some different concentrations of saline compared with the SmPC, but nevertheless made it clear that hospital pharmacy practice was to stick with the normal saline specified in the SmPC.

112. The only remotely plausible motivation for the use of saline identified by Lilly is a possible concern on the part of oncologists and/or pharmacists about the effect of administering dextrose to diabetic patients. (Although Prof Thatcher also referred to the position of pre-diabetic patients, it is clear that there will not be a concern about these patients if there is no concern about diabetic patients.)
113. Before addressing this in detail, it is convenient to deal with three preliminary points. The first is that there was some debate in the evidence and submissions as to the extent of the motivation which was required in order for a pharmacist to depart from the SmPC. Some of Actavis' experts said in their reports that a "compelling" reason was required, whereas some of Lilly's experts appeared to suggest in their reports that any reason would do. In their oral evidence, however, there was much more common ground between the experts. Ms Nijjar was representative when she said that "if you do something outside the SmPC, you have to have a very good reason for doing it".
114. Secondly, most of the evidence proceeded on the assumption that the quantity of dextrose which would be delivered to the patient with each three-weekly administration of the Actavis Product was 5 g (being the amount of dextrose contained in a 100 ml bag of 5% dextrose solution). As I have explained above, however, this is inaccurate. In fact, assuming that the direction in the SmPC is followed, the quantity administered to an average-sized patient will be 3.4 g or less.
115. Thirdly, when cross-examining Prof Seckl, counsel for Lilly asked questions about the procedure for priming and flushing the dripline from the bag of diluted pemetrexed to the patient which were designed to establish that, in addition to the quantity of dextrose solution which was used to dilute the Actavis Product, an additional quantity of dextrose solution would be delivered to the patient during priming and flushing. The cross-examination was rather confused, and did not clearly establish how much additional dextrose solution could be expected to be delivered in this way. This was not a point which Lilly had pleaded; nor was it mentioned in either of Prof Thatcher's two reports or in any of the eight reports by Lilly's pharmacists; nor was it mentioned in Lilly's 389 paragraph-long written opening submissions; nor was the point put to Dr Powrie (who gave evidence before Prof Seckl). Assuming that this was not a deliberate attempt to ambush Actavis, it can only be explained as a very late afterthought indeed on the part of someone on the Lilly team. Even then, however, Lilly made no application to amend its statement of case. Nor did it lead any evidence in chief from Prof Thatcher, or any of its other witnesses, on the point. In those circumstances, I conclude that this point is not open to Lilly. Even if it was open to Lilly, it is plainly not a point of any substance. Although priming and flushing of the line may perhaps lead to a couple more grams of dextrose being administered to the patient, this is not something that would concern the oncologist or the pharmacist (because if it did, Lilly would have pleaded it and led evidence about it). Nor would it lead to the total amount of dextrose being administered exceeding 10 g in any event.
116. The only evidence about the actual effect of the amount of dextrose in the Actavis Product when diluted on diabetics from any expert in diabetes is the evidence of Dr Powrie, who assumed that the amount of dextrose that would be administered was 5 g. Dr Powrie was firmly of the view that an endocrinologist would not be concerned about administering pemetrexed with 5 g of dextrose, even if administered following dexamethasone, because 5 g of glucose is such a small amount of additional sugar. As he explained, it is the equivalent of 100 ml of skimmed milk or a single jelly baby. This is much less than in a normal meal. He also explained that a 5% dextrose solution can be administered continuously intravenously to diabetic patients, sometimes for prolonged periods (e.g. as part of a rehydration regime), and that many medicines for intravenous infusion can be given in 5% dextrose. He gave the examples of carboplatin and oxaliplatin (also chemotherapy drugs), which are routinely administered in 5% dextrose solutions, but do not cause problems for diabetic patients. I shall return to this point below.
117. Furthermore, Dr Powrie's evidence was that, whilst administering steroids such as dexamethasone would be significant for a diabetic patient, all patients being treated with dexamethasone would need to have their blood sugar levels assessed before and during administration of dexamethasone (which would pick up any potentially pre-diabetic patients). Diabetic patients would be informed about the likely effects of dexamethasone, and would be advised to monitor their glucose levels more frequently and to adjust their medication (if they were on medication) to minimise the increase. In general, patients would be able to go through chemotherapy and steroid treatment without any significant destabilisation of blood sugar levels as long as they followed the recommended adjustments. Dr Powrie explained that the addition of 5 g of glucose to the chemotherapy regime would not have any impact and would not cause any changes to the routines in place for diabetics being given dexamethasone.
118. Dr Powrie's evidence was completely unshaken in cross-examination. It is only necessary to note four points. First, he was asked about type 1 diabetics. It was common ground between Dr Powrie and Prof Thatcher, however, that type 1 diabetes is not really the issue and that the vast majority of diabetics who receive chemotherapy have type 2 diabetes. In any event, a guide for type 1 diabetics which was put to Dr Powrie in cross-examination states that only snacks containing more than 10 g of carbohydrate need to be taken into account when matching to insulin.
119. Secondly, Dr Powrie agreed that 5 g of glucose administered intravenously over 10 minutes would initially lead to a higher blood sugar level than the same quantity ingested orally, but pointed out that the total load of glucose going into the body was the same. Even if one notionally increased the quantity of glucose by 25-30%, as was suggested to Dr Powrie, the quantity of glucose would still be well under 10 g.
120. Thirdly, Dr Powrie was asked to assume that in fact the amount of glucose administered with the Actavis Product was 10 g, but even on that assumption he was clear that it would still not have any significant effect on patients with diabetes.
121. Fourthly, Dr Powrie accepted Prof Thatcher's point that lung cancer patients receiving chemotherapy are often very unwell and have problems eating. He was clear, however, that this did not detract from the fact that, for a diabetic patient, 5 g, or even 10 g, of glucose was not a significant quantity.
122. As counsel for Actavis submitted, Dr Powrie's evidence receives strong support from the evidence of the pharmacists. None of the pharmacists except for Dr Fassari (whose evidence I shall consider below) had ever been asked by a clinician to change a diluent for a product from dextrose solution to something else (either because of a diabetes concern or for any other reason). Furthermore, none of the pharmacists themselves had been concerned about giving dextrose to diabetic patients. For example, Dr Vigneron explained that, even where he has had a choice of diluent (as with carboplatin), he has never consulted an oncologist or endocrinologist about a problem with dextrose solution.
123. This evidence is particularly powerful because it has been standard practice for the last 20 years or so to treat cancer patients, including diabetics, with a number of drugs which are routinely diluted with dextrose solution. Pertinent examples of this are carboplatin, oxaliplatin and docetaxel.
124. Carboplatin has been used, and continues to be used, extremely widely in hospitals throughout the countries that are of interest in this case. The carboplatin SmPC includes both dextrose solution and saline as diluents. Carboplatin is usually administered diluted with a 500 ml bag of 5% dextrose solution, which works out as 25 g of dextrose, over a period of between 15 minutes and one hour. (It can also be administered with a 250 ml bag, which delivers 12.5 g.) There is no dispute that dextrose solution is used rather than saline due to its superior stability profile. Carboplatin is routinely administered with dexamethasone, which significantly raises glucose levels and requires extra monitoring to be done for all patients, as discussed above.
125. The overwhelming majority of clinicians and pharmacists have never had any problems with administering carboplatin in dextrose solution, including with dexamethasone. Strikingly, there is not a single example in the evidence of any patient with diabetes receiving carboplatin in dextrose solution with dexamethasone and then having any problems with their blood sugar control over and above the usual issues that arise when taking dexamethasone.
126. The only evidence of anyone suggesting the use of saline with carboplatin is as follows:

i) Prof Seckl, having wondered if his own practice in relation to carboplatin was unusual, spoke to 15 of his colleagues to ask them what their practice was. With one exception, they told him that they all prescribed carboplatin diluted in dextrose solution with dexamethasone, and took care to monitor patients' blood sugar levels. None of these colleagues had any concern about administering this quantity of dextrose to diabetic patients. One colleague from a large treatment centre, who I will not name because he has not had the chance to defend himself, but will simply refer to as X, expressed a different view. X told Prof Seckl, much to Prof Seckl's surprise, that he did not monitor patients' blood sugar when administering carboplatin with dexamethasone, a practice Prof Seckl described as "extraordinary". X then said that, if he knew that a patient was a diabetic, he would ask for the carboplatin to be diluted in saline. When Prof Seckl said to X that using saline would not make any real difference, because the dexamethasone was much more significant, X agreed and said that the change of diluent was "just me being silly". Prof Seckl described X as a "maverick" in this respect (a choice of expression which Prof Seckl confirmed was his own, and not one which had been suggested to him by Actavis' legal team).

ii) Dr Fassari mentioned that she had had one case where an oncologist had asked her to change the diluent for carboplatin for a diabetic patient to saline. Dr Fassari was unable to give any information about the clinical status of the patient or any of the relevant surrounding circumstances (including whether the patient had type 1 or type 2 diabetes, whether they were on high dose steroids or whether there were any other circumstances that had led to that request). Furthermore, Dr Fassari explained that the reason she was able to approve the change was because there was no compatibility problem with carboplatin and saline, a fact that she knew because saline was a diluent included in the SmPC. Still further, Dr Fassari's own opinion was that 5 g of dextrose could be managed by the patient, and indeed she thought that the greater amount (12.5 g) involved in the particular episode she described could have been managed by normal methods (monitoring etc.).

iii) Dr Millà mentioned for the first time in his oral evidence in chief that he was aware of carboplatin occasionally being administered with saline. Perhaps as a result of the point being raised at the last minute in this way, Dr Millà's evidence was confused and inconsistent. Thus he said that it happened a couple of times a year, but he also said that the last occasion he could recall was about six years ago. Ms Juanals, who worked at the same hospital, gave evidence that she had carefully checked the hospital's records and found only one instance of carboplatin being diluted in saline, which had taken place before she started working there in 2007. That being the case, she could not shed any light on the circumstances. I have no hesitation in preferring Ms Juanals' evidence on this point.

iv) In addition to the evidence considered above, Lilly relied upon a paper by A.L. Myers et al, "Stability study of carboplatin infusion solutions in 0.9% sodium chloride in polyvinyl chloride bags", J. Oncol. Pharm. Practice, 22(1), 31-36 (2016). None of Lilly's experts gave any evidence about this paper and its relevance, if any, to the use of pemetrexed in dextrose solution. It was introduced into the case by Lilly for the purposes of the cross-examination of Dr Powrie, but in the event it was not put by counsel for Lilly to Dr Powrie but instead to Prof Seckl (who had therefore not pre-read the paper and was not properly in a position to comment on it). Nor was there any evidence of what internet searches Lilly's team had carried out to find the paper. This paper was relied upon by Lilly for its passing reference to a statement by the authors that "At our institution (The University of Texas M.D. Anderson Cancer Centre; UTMDACC) we prepare carboplatin admixtures in NS [i.e. normal saline] for special patient populations at risk (e.g. diabetics, obese) who are at a heightened risk of hyperglycaemia^28-30 [emphasis added]". The underlying references supporting this statement are not in evidence, but from the Myers paper appear to be confined to a number of uses of carboplatin which are indeed "special", but about which there is no evidence as to the circumstances surrounding the administration of carboplatin and the quantities of dextrose that would need to be used in those circumstances. Thus reference 28 is a paper about dextrose control in the intensive care unit; reference 29 is about intraoperative hyperthermic intraperitoneal chemoperfusion with oxaliplatin (i.e. not carboplatin and an off-label use of oxaliplatin); and reference 30 is a case study about using dextrose solution as a diluent for hospitalised patients.

127. As counsel for Actavis submitted, the evidence about these examples is mostly second or third hand, very patchy, and at most merely shows some extremely rare uses of saline which do nothing to undermine the overwhelming evidence showing no concerns at all about the use of dextrose solution in the context of carboplatin.
128. Turning to oxaliplatin, this is another important example of a drug routinely used to treat cancer patients which is diluted in dextrose solution. Indeed, it cannot be diluted in saline. Oncologists would know about oxaliplatin and its dilution in dextrose solution. As with carboplatin, oxaliplatin is given with far higher doses of dextrose than the Actavis Product – usually around 25 g. Yet none of the witnesses who used oxaliplatin had any concerns with using dextrose solution.
129. There is no evidence of oxaliplatin ever being diluted in saline for administration for an indication and route of administration covered by the SmPC. The best Lilly could do was a paper by A.M. Mehta et al, "Stability of oxaliplatin in chloride-containing carrier solutions used in hyperthermic intraperitoneal chemotherapy", Int. J. Pharm., 479, 23-27 (2015). Again no evidence was given by any of Lilly's witnesses about this paper, it was introduced into the case by Lilly as cross-examination material (but this time for Mr Bateman) and there was no evidence of what internet searches Lilly's team had carried out to find the paper. The authors state that "The use of large intraperitoneal volumes of 5% dextrose is associated with serious hyperglycaemias and electrolyte disturbances…. [emphasis added]". It is not clear what is meant by "large volumes". In any event, Mr Bateman explained that this was for a niche indication that was completely off-label.
130. As for docetaxel, the SmPC provides that it can be diluted in saline or dextrose solution. Mr Bateman explained that he was aware of a number of hospitals in London and the South East which used dextrose solution, and that there are hospitals where the docetaxel is dose-banded in dextrose solution (i.e. is pre-diluted with dextrose solution, and so cannot be switched to saline). Likewise Ms Nijjar said that she knew of some hospitals that use dextrose solution, but did not consider that they were doing anything dangerous or wrong.
131. The highpoint of Lilly's case was the evidence of Prof Thatcher. In essence, he made two main points. The first was that he would be concerned about giving any unnecessary dextrose to a diabetic patient, even if it was as little as 5 g once every three weeks, since it was an unnecessary risk which could be avoided by diluting the Actavis Product in saline. The second was that he considered that the wording in section 4.4 of the Actavis UK SmPC was an "unusual warning" which would be a "flag to the pharmacist" who would raise the issue with the oncologist when setting the protocol for the Actavis Product.
132. So far as both these points are concerned, I consider that Prof Thatcher's evidence is entirely unpersuasive when tested against the evidence, not only of Prof Seckl and Dr Powrie, but also of the eight pharmacists, and in particular the evidence concerning the administration of much larger quantities of dextrose together with carboplatin and oxaliplatin. (I would add that, as discussed above, the second point will no longer apply once the relevant wording is removed from the SmPC, as it already has been in Germany.)
133. Furthermore, Prof Thatcher eventually accepted that, if an oncologist or pharmacist were to have any concern about administering dextrose to diabetic patients, they would consult an endocrinologist. It is clear from Dr Powrie's evidence that the endocrinologist would reassure them that there was no need for concern. This is supported by the following evidence given by Prof Seckl:

"Q. If there was an option, professor, of taking the pemetrexed either in saline or dextrose, [Prof Thatcher] takes the view that it is not right to expose [diabetic] patients to any additional risk, even if small. You would not need to. Why would you do it?

A. So I can understand his point of view, but I do not agree with it, with respect. Diabetic patients have to eat. They have to consume glucose. It is not that glucose is forbidden. This is a tiny dose of glucose. When I first read his report, I thought, 'Gosh, maybe I have got it wrong', which is precisely why I went and asked a number of my diabetes colleagues, who manage diabetes, and they laughed at me. They said, 'It is nonsense. The issue here is the dexamethasone. You are monitoring the sugar. This is such a tiny dose of sugar. It is not an issue for these patients.'"

134. I would add that another question which divided Prof Seckl and Dr Powrie on the one hand and Prof Thatcher on the other hand was whether patients should be warned about the 5 g of dextrose they would be receiving. Neither Prof Seckl nor Dr Powrie considered that this was necessary, because it was such a small amount. Prof Thatcher was firmly of the view that the principle of informed consent meant that patients should be informed. In my judgment Prof Thatcher's evidence is difficult to reconcile with the absence of any evidence that patients treated with carboplatin and oxaliplatin are warned about the extra 25 g of dextrose they receive. (Nor is it clear that patients who are given 500 mg doses of Alimta are warned about the sodium they are receiving.) But in any event, even if the oncologist took the view that it was good practice to inform the patient about the 5 g of dextrose, it does not follow that this would provide a reason to switch the diluent from dextrose solution to saline. On the contrary, informing the patient would (if anything) assist the patient to manage their blood sugar levels.
135. Finally, as counsel for Actavis submitted, it is striking that in the real world there are no examples of hospitals or individual pharmacists or clinicians switching diluents in the manner that Lilly suggests is foreseeable. First, there are no examples from any country of a hospital, pharmacist or clinician using a drug with a diluent that is not in the SmPC for an indication and a route of administration that is in the SmPC. Secondly, the only example in the entire case of any hospital, pharmacist or clinician using a drug for an indication specified in the SmPC with a diluent that was not specified in the SmPC was for an antibiotic used with a diluent called extraneal. But this was for a niche route of administration which was not indicated in the SmPC for that antibiotic, namely intraperitoneal dialysis to improve ultrafiltration and fluid management. This meant that it was impossible to use a diluent which was specified in the SmPC. This is very remote from the circumstances of the present case. Thirdly, there are also a couple of examples of different diluents apparently being used for off-label indications, but again these shed no light on the likelihood of a different diluent being used for the indications specified in the SmPC.
136. Accordingly, I conclude on the evidence before me that, even if stability data for the Actavis Product in saline were to become available, there would be no motivation to use saline rather than dextrose solution to dilute the Actavis Product. Nevertheless, I accept that I do not have a crystal ball, and I cannot exclude the possibility, remote as it presently seems, that in a few years some pharmacists will decide to switch diluents, perhaps for a reason which has not been canvassed in evidence.

Italian and Spanish law

137. Actavis contend that, in Italy and Spain, a further reason why pharmacists would be slow to switch from the diluent specified in the SmPC to an alternative diluent is that this would be illegal in those countries, at least unless special approval was obtained. Lilly disputes this. As indicated above, each side obtained expert evidence supporting its position. It appears from this evidence that the legislative provisions in each country are open to interpretation and there does not appear to be any case law on the point. It is not necessary to reach a conclusion about which out of each pair of experts is right. This is for a number of reasons. First, because I have already concluded that there would be no motive to switch anyway. Secondly, it seems clear from the evidence of the Italian and Spanish pharmacy experts that, even if it is technically illegal, that would not deter them from switching if there was a sufficiently good clinical or pharmacy reason to do so. Thirdly, and in any event, it is probably sufficient for present purposes to conclude, as I do, that Actavis' interpretation of the legislation is, as matters stand, an arguable one.

Is it likely that saline would be used to dilute the Actavis Product despite the steps taken by Actavis to prevent this?

138. Even if stability data for the Actavis Product in saline became available, and even if there was a motive to switch from dextrose solution to saline, Actavis contend that the steps which Actavis are taking which I have outlined above would be effective to prevent this from actually happening. Lilly disputes this. A number of points were raised by Lilly in this regard.
139. Before turning to consider these, it should be noted that, if oncologists and/or pharmacists did have concerns about diluting the Actavis Product in dextrose solution, then the most straightforward solution would be to administer Alimta diluted in saline. Lilly's only answer to this is to suggest that the purchasing decision may be outside the control of the oncologist and/or pharmacist. But if the oncologist and/or pharmacist was genuinely concerned about administering the Actavis Product in dextrose solution, they would be likely to inform their management, so that this concern could be taken into account when making purchasing decisions.
140. Letters to competent authorities and medical centres. Lilly questioned whether these letters would be effective in so far as they requested the issuing of central guidance and the dissemination of the information. As far as the issuing of central guidance is concerned, it is fair to say that there is no evidence that NHS England has issued central guidance or is likely to do so. Central guidance is less important in this case than in the pregabalin case referred to above, however, since pemetrexed is not prescribed by GPs, but by oncologists in hospitals. Furthermore, as explained above, pemetrexed is prepared in aseptic units according to standard procedures. Thus what matters is whether the information gets to hospitals, and in particular the relevant pharmacists. In this regard, Mr Bateman gave evidence about an email he (and many others) had received from the Specialist Procurement Pharmacist for the Thames Valley & Wessex Pharmacy Procurement Service dated 6 January 2016 explaining both the advantages of the Actavis Product compared to Alimta and that it could not be diluted in saline and attaching a copy of the letter. Furthermore, Mr Bateman also gave evidence about protocols for pemetrexed which had been updated in the light of this guidance. Accordingly, it appears that, in the UK, the guidance is being disseminated to those who need to know it.
141. In addition to questioning whether the information contained in the letters would be disseminated, Lilly contended that it was wrong. Indeed, counsel for Lilly went so far as to submit that the letters contained an "intentionally misleading threat to pharmacists". This is a serious allegation and, since the letters were obviously drafted by Actavis' legal team, it is an allegation against them as much as against Actavis. There was never any basis for this allegation, and it should not have been made.
142. In the alternative, counsel for Lilly submitted that, even if unintentionally, the letters were legally inaccurate, and that any competent legal advisor would so advise any authority or medical centre which sought advice. Counsel for Actavis disputed this. The subject of this dispute is the statement that "use of saline as a diluent within a hospital or medical centre, whether in a pharmacy or otherwise, … would be unlawful". Counsel for Lilly submitted that this was inaccurate because pharmacists would have a defence under Article 27(c) CPC, which in the UK is implemented by section 60(5)(c) of the 1977 Act.
143. Art 27(c) CPC provides:

"The rights conferred by a Community Patent shall not extend to:

…

(c) the extemporaneous preparation for individual cases in a pharmacy of medicine in accordance with a medical prescription nor acts concerning the medicine so prepared".

144. Similarly, section 60(5)(c) provides:

"An act which, apart from this subsection, would constitute an infringement of a patent for an invention shall not do so if—

…

(c) it consists of the extemporaneous preparation in a pharmacy of a medicine for an individual in accordance with a prescription given by a registered medical or dental practitioner or consists of dealing with a medicine so prepared".

145. Counsel for Actavis submitted that these were narrow provisions: they were restricted to "extemporaneous preparation" "for individual cases" (or "for an individual"), and thus did not extend to the case where a pharmacy systematically prepared all doses of a product in a particular way following a worksheet and a protocol and without individual assessment by the pharmacist. I was not referred by either side to any commentary, let alone any case law (whether of the UK or any other state which has implemented the CPC), which assists in determining whether this is correct or not. I think it would be undesirable for me to attempt to decide this question without the benefit of such assistance. It is sufficient for me to say that in my view it is arguable that counsel for Actavis is correct about this, and it is also arguable that these provisions should be interpreted more widely. As at present advised, I do not regard the answer to the question as obvious.
146. Counsel for Actavis also submitted that, even if the pharmacist had a defence under Article 27(c), it did not follow that the hospital would have a defence. He argued that the hospital would be equally as guilty of indirect infringement by supplying the Actavis Product to the pharmacy if it was foreseeable that the Product would be diluted in saline as Actavis would be in supplying it to the hospital. Counsel for Lilly submitted that this would drive a coach and horses through Article 27(c), but again I was referred to no commentary or case law which assists on this question. Again, I think it undesirable to attempt to decide the question without such assistance. Again, it seems to me that both contentions are arguable and that the answer is not obvious.
147. Finally in relation to this point, I would note that Lilly is complaining about a threat of infringement of Lilly's patent alleged to be being made by Actavis to Actavis' customers. Normally, of course, it is competitors who complain about threats of patent infringement made by patentees to the competitors' customers. Counsel for Lilly accepted that he was unaware of any precedent for this. Thus Lilly is not merely seeking to place the entire burden of enforcing the Patent on Actavis, but also Lilly is complaining that Actavis are going too far in attempting to safeguard Lilly's rights. Again, this is a remarkable attitude for a patentee to adopt.
148. As counsel for Actavis submitted, even if the sentence objected to by Lilly is legally inaccurate, it by no means follows that the letters will have no effect. This is for two reasons. First, even if hospitals have a defence to infringement if the Actavis Product was diluted in saline, hospitals would surely not want to be involved in infringement by Actavis. Apart from anything else, it is arguable that hospitals would commit the tort of procuring breach of statutory duty, by intentional dealings with the Actavis Product which caused infringement by Actavis. Secondly, and even more importantly, hospitals would plainly not want to cause, by their own conduct, an infringement by Actavis which had the consequence of their supply of the Actavis Product being cut off, because that would prevent hospitals from obtaining the cheaper Actavis Product. Even if Actavis could not or did not immediately cut off supply, in the longer term supply would be cut off because Lilly would rely upon such dealings to seek an injunction against Actavis.
149. Contractual terms. It is common ground that imposing contractual terms of the kind which Actavis are attempting to impose is not easy. It does not follow that it is impossible, as Lilly appears to contend, still less that it is not worth trying.
150. On this question, Mr Wilson gave evidence about discussions between Actavis and NHS Wales. Some of Mr Wilson's evidence was based on what he had been told by Mr Brown, but importantly Mr Wilson disclosed an email from the responsible individual at NHS Wales dated 19 January 2016. The name of the individual had not been provided by Actavis to Lilly as it should have been, but I directed that this be provided. As a result, Lilly's solicitors were able to speak to the individual in question. In addition, I ordered Actavis to disclose the email from Mr Brown to which the individual was responding. Counsel for Lilly submitted that Mr Wilson's account had been contradicted, or at least qualified, by what the individual had told his instructing solicitor and/or by the email from Mr Brown. I do not accept this. In my judgment the email dated 19 January 2016, when read in conjunction with the "statement 1" to which it refers, makes it clear that NHS Wales is prepared to enter into a contract with Actavis for the supply of the Actavis Product on the terms specified by Actavis, which require (i) the product to be diluted only in dextrose solution, (ii) NHS Wales to issue central guidance in line with the letters discussed above and (iii) worksheets to be updated accordingly.
151. As counsel for Actavis submitted, even if the contractual terms which Actavis are attempting to impose are not enforceable, the very act of trying to impose them will assist to alert Actavis' customers to the issue and to bring home to them the consequences of diluting the Actavis Product in saline.
152. There is also a related dispute between the parties as to whether Actavis could cease to supply the Actavis Product where Actavis have won a tender. Lilly's position appears to be that (i) Actavis would be contractually obliged to supply the Actavis Product, and therefore (ii) Actavis would continue to supply it even if doing so would constitute an infringement of the Patent. Actavis' position is that, if it would be unlawful for them to supply a hospital with their product, then they would not continue with such supply irrespective of whether or not that supply was pursuant to a tender. If or to the extent that any penalties were imposed on Actavis pursuant to the tender contract as a result of them ceasing supply, then Actavis would suffer those penalties.
153. As I drew to the parties' attention during closing submissions, it was held by the Court of Appeal in Microbeads AG v Vinhurst Road Markings Ltd [1976] RPC 19 that the sale of a product which infringed a third party's patent amounted to a breach of the warranty of quiet possession implied into the contract of sale by section 12(2) of the Sale of Goods Act 1893 (now section 12(2)(b) of the Sale of Goods Act 1979). In those circumstances it seems to me that it cannot be correct that Actavis would be obliged to continue to supply the Actavis Product even if it would amount to an infringement of the Patent, as Lilly suggests. On the contrary, Actavis would be in breach of contract if it continued to supply. Once again, it is remarkable that a patentee can have advanced such a suggestion.
154. Questions from pharmacists about the use of saline. Mr Wilson gave evidence, based on information from Mr Brown which Mr Brown had obtained from Actavis' Key Account Manager ("KAM") Shaun Kneeland, about questions which had been asked by Chris Watson, who is the Lead Aseptic Pharmacist at St Bartholomew's Hospital (and hence a colleague of Ms Nijjar's), about diluting the Actavis Product in saline. Despite resisting the admission of this evidence, and criticising Actavis' failure to call Mr Brown, counsel for Lilly positively relied upon this evidence in his closing submissions as showing what was likely to happen. (Furthermore, Lilly adduced no evidence from Mr Watson to contradict the account given by Mr Wilson.)
155. In summary, the evidence is that, at a meeting on 14 December 2016, Mr Watson raised issues as to (i) whether it was necessary to follow the SmPC, (ii) doing stability tests in saline and (iii) concerns about diabetic patients. Mr Kneeland replied (i) that use of saline would be a patent infringement, (ii) what was the point of carrying out a stability test (to which Mr Watson acknowledged that 14 days' stability was sufficient, consistently with Ms Nijjar's evidence) and (iii) that the amount of dextrose was very small. Despite this, Mr Watson subsequently contacted Mr Kneeland on 14 January 2016 asking for "equivalence data" for the Actavis Product.
156. It is not clear what Mr Watson meant by this request. Be that as it may, Mr Wilson exhibited a letter from Actavis PTC to the Chief Pharmacist at Barts Health NHS Trust (i.e. Mr Watson's superior) dated 19 January 2016. This letter merits being quoted in full:

"We refer to our letter of 18^th December, 2015 and the letter dated 30^th December, 2015 you received from Pharmaxo. In those letters you were informed that Actavis' pemetrexed product must only be diluted in 5% glucose solution, as set out in the summary of Product Characteristics and Package Leaflet and that any purchase order received would be on contractual terms that prevented the dilution of Actavis' pemetrexed product in saline for any purpose.

Whilst Actavis would be delighted to supply St Bart's Hospital NHS Trust with Actavis' pemetrexed product, we are concerned about recent enquiries which have been made to Actavis on behalf of St Bart's hospital by Mr Watson, who we understand is the Lead Aseptic Pharmacist. Mr Watson's enquiries relate to the dilution of Actavis' pemetrexed product in saline.

We would like to take this opportunity to reiterate that due to patent reasons Actavis' pemetrexed product MUST ONLY be diluted in 5% glucose and should NEVER be diluted in saline, whether for administration to patients, for research (including the generation of stability data) or for any other purpose. We also reiterate our request that this is drawn to the attention of all relevant personnel (including Mr Watson) and that any standard operating procedure specifies dilution in 5% glucose solution only.

We should be grateful if you would confirm in writing that the above steps will be taken in order to ensure that there is no disruption to any supply of Actavis pemetrexed product to St Bart's hospital NHS Trust"

157. Counsel for Actavis submitted, and I agree, that this letter demonstrates what will happen if questions are raised by Actavis' customers or potential customers about the use of saline: they will be told in no uncertain terms not to do it for any purpose (including generating stability data) and that their supply of the Actavis Product will be cut off if they do.
158. Mr Wilson's evidence, based on what he had been told by Mr Brown, was that Mr Watson was the only person to have raised such queries in over two hundred institutions contacted. In this regard, Mr Wilson gave some evidence based on information obtained from Actavis' KAMs. They had been asked to record information about how many people they had spoken to had either questioned the use of dextrose or asked about the use of saline, among other things. This information had been collated into a spreadsheet which Mr Wilson exhibited. The spreadsheet shows that only 3% of those spoken to had mentioned saline at all.
159. Counsel for Lilly vigorously criticised this evidence for a number of reasons. First, he submitted that it was a survey and that the safeguards required for surveys had not been complied with. I do not agree that it was a survey. Secondly, he again criticised Actavis for not calling Mr Brown. But Mr Brown was simply the person who collated the information. It would plainly have been impractical, or at least disproportionate, for Actavis to have adduced evidence from all of the KAMs. Thirdly, he complained that Actavis had not disclosed the documents underlying the spreadsheet. This is a fair point, but having regard to the evidence given by Mr Wilson as to the manner in which the information was obtained, it seems reasonably clear that the additional documents would not have been very informative. Nevertheless I accept that, in the absence of such disclosure, it would be wrong to place much weight on this evidence.
160. A separate point taken by counsel for Lilly was that, if the information in the spreadsheet was taken at face value, it showed that 6, 1 and 3 people had questioned the use of dextrose, and 5, 2 and 3 people had asked about use of saline in the first three days, whereas after that there had been very few such questions. Counsel for Lilly submitted that "something must have happened". It is not clear to me what he was suggesting might have happened. In any event, I agree with counsel for Actavis that the pattern of the answers is unsurprising, since one would expect more queries at the beginning of the campaign, when awareness of the Actavis Product, and in particular the fact that it must be diluted with dextrose solution, was low, than later on.
161. As I have said, I do not place much weight on this evidence. Taken together with all the other evidence referred to above, however, the overall picture that is presented is that Actavis' safeguards appear to be working. Accordingly, I conclude that, even if stability data for the Actavis Product in saline were to become available, and even if there was a motive for pharmacists to switch from dextrose solution to saline as the diluent, it is probable that the steps taken by Actavis will be effective to prevent this being done. I accept, however, that this cannot be guaranteed.

Lilly's case on infringement from launch

162. As discussed above, it is common ground that, to begin with, pharmacists will follow the SmPC for the Actavis Product and will dilute it with dextrose solution. It is also common ground that it will take some time for stability data for the Actavis Product to become available, even assuming that this does happen. I have concluded that this is likely to take two to three years, based on the evidence of Lilly's own witness Dr Vigneron. Despite this, Lilly strenuously resists the apparently obvious conclusion that Actavis are not presently infringing the UK designation of the Patent and will not infringe the French, Italian and Spanish designations when they launch later this year. Lilly contends that Actavis will infringe the Patent from the moment of launch in each country, although as noted above Lilly has not actually brought a counterclaim in respect of the UK launch. Lilly advances two arguments in support of this contention.
163. The stock-piling argument. Lilly points out that, prior to dilution, the Actavis Product has a shelf life of 18 months. There is no dispute that, in order to exploit their "first mover" advantage as the first generic supplier to launch a generic pemetrexed product, Actavis will want to sell as much of their product as they can in the first months after launch. (As counsel for Lilly pointed out, however, Mr Wilson did not disclose how much Actavis had forecasted that they would sell.) Lilly contends that it follows that it is foreseeable that some of the product sold by Actavis at launch will still be held in stock somewhere in the supply chain after stability data have become available and that such stock will then be diluted in saline.
164. Counsel for Actavis rightly submitted that this argument was entirely speculative and depended upon a concatenation of events, and thus fell a long way short of establishing that it was foreseeable that any stock sold at launch would be diluted in saline. It is sufficient to give two answers to it. First, given my finding that stability data will not become available until two or three years after launch, the shelf life of product sold at launch will have long since expired.
165. Secondly, and in any event, the evidence establishes that pemetrexed is not kept in stock for a long period of time, but on the contrary is rapidly consumed and re-ordered. One reason for this is that it is an expensive drug, and therefore hospitals have a strong economic incentive not to keep funds tied up in stocks of the drug. (The same goes for distributors.) Another reason is that, since pemetrexed is only administered when prescribed by oncologists and is prepared for administration for each patient or group of patients, there is no need to keep it in stock for very long.
166. Mr Wilson gave unchallenged evidence that, in the UK, Pharmaxo would not supply hospitals with more than two to three weeks' worth of the Actavis Product at a time, and generally less than that. It was put to Mr Wilson that Actavis' distributors in France, Italy and Spain could not be controlled to the same extent as Pharmaxo, but he explained that Actavis PTC would be party to the contract with each distributor and so would be able to exercise control. In any event, the evidence of the pharmacists showed that, for the reasons given above, hospitals would not stock the product for long. Thus Dr Vigneron said that he tended to reorder expensive products like pemetrexed on a weekly basis.
167. The tender argument. Lilly's second argument is that, where Actavis win a competitive tender to supply pemetrexed, their bid constitutes an offer to supply the Actavis Product for the whole period of the tender. Lilly contends that this includes an offer to supply the Actavis Product at a time after which it is foreseeable that stability data will have been published and pharmacists will have switched to diluting the Actavis Product with saline.
168. This argument is even more speculative and far-fetched than the stock-piling argument. It fails for a number of reasons. It is sufficient to give three. First, there is no evidence that contract periods exceed two years. On the contrary, such limited evidence as there is suggests that tenders tend to be annual, or at most bi-annual. Secondly, Mr Wilson gave unchallenged evidence that Actavis would not supply against a tender if they knew or foresaw that the Actavis Product would be diluted in saline. Thirdly, for the reasons discussed above, Actavis would not be contractually obliged to supply the Actavis Product if that amounted to an infringement of the Patent, and on the contrary would be in breach of contract if they did so.

Should DNIs be granted?

169. For the reasons given above, I conclude that, at present, it is not foreseeable that the Actavis Product will be diluted in saline. Accordingly, for the foreseeable future, the supply by Actavis of the Actavis Product will not amount to an indirect infringement of the Patent. I have no doubt whatsoever that Actavis have a real commercial interest in obtaining DNIs and that granting DNIs would serve a useful purpose, since this would make it clear that the supply of the Actavis Product is lawful in circumstances where Lilly has resisted that conclusion tooth and nail.
170. The only tenable argument advanced by Lilly against this is that, as I have acknowledged a number of times, I cannot be sure what will happen in the future. It is theoretically possible, if presently unlikely, that in, say, three years, circumstances will have changed. But at the point the Patent will still have over two years to run.
171. I accept the submission made by counsel for Actavis that the correct way to deal with this uncertainty is not by refusing to grant DNIs now, but by giving the parties liberty to apply in the event of a material change in circumstances. In that way, if it turns out at some point before 21 June 2021 that the Actavis Product is being diluted in saline, then Lilly will be a position to obtain appropriate relief notwithstanding the grant of the DNIs at this stage. For the avoidance of doubt, I agree with counsel for Actavis that both sides should have permission to apply, and not just Lilly, since that will enable Actavis to be pro-active, rather than merely re-active, in the event of a change of circumstances.

Conclusion

172. For the reasons given above I will grant DNIs in respect of each of the four designations of the Patent in issue.

The Letters Issue

Factual background

173. The Letters Issue arises out of the jurisdictional battle between the parties at an earlier stage of the proceedings, and needs to be seen in that context. I set out some of the relevant factual background in my judgments of 27 November 2012 [2012] EWHC 3316 (Pat) at [6]-[21] and 15 May 2014 [2014] EWHC 1511 (Pat) at [5] and [238]-[260]. I will take those passages as read, but I will nevertheless recapitulate some of the procedural steps set out in more detail there.
174. On 12 July 2012 Actavis' solicitors wrote to Lilly seeking an acknowledgement that dealings by Actavis in pemetrexed dipotassium would not infringe the UK, French, German, Italian and Spanish designations of the Patent. Lilly did not give that acknowledgement.
175. On 27 July 2012 Actavis issued the Claim Form in the First Action seeking DNIs in relation to dealings in pemetrexed dipotassium in respect of the UK, French, German, Italian and Spanish designations of the Patent. This was served on Lilly on 1 August 2012. Lilly subsequently disputed the jurisdiction of this Court in respect of the French, German, Italian and Spanish designations.
176. On 30 July 2012 Lilly issued proceedings against Actavis PTC and Actavis Deutschland GmbH & Co KG ("Actavis Germany") for threatened infringement of the German designation of the Patent in the Düsseldorf Landgericht (Düsseldorf Regional Court). Those proceedings were served on 9 August 2012. Lilly's argument in essence was that the salt form of the pemetrexed was irrelevant. Thus although the complaint only related to pemetrexed dipotassium, it was equally applicable to pemetrexed diacid and pemetrexed ditromethamine. Actavis subsequently challenged the jurisdiction of the German court on the ground of lis alibi pendens, since this Court was first seized of the dispute.
177. From that point onwards, it was important to Actavis not to threaten to commit acts which (if Lilly was right as to the scope of the Patent or as to what would constitute indirect infringement) would infringe the Patent except contingently. This was for two closely-related reasons.
178. First, as explained above, Actavis' objective was to clear the path for the launch of a generic pemetrexed product in the UK, France, Germany, Italy and Spain after expiry of the SPCs by obtaining DNIs in a single set of proceedings. Accordingly, Actavis wanted to avoid the complication and extra expense of separate proceedings by Lilly with respect to the French, German, Italian and Spanish designations in those jurisdictions. In order to bring proceedings in those jurisdictions, Lilly would need to establish a sufficient threat on the part of Actavis to commit allegedly infringing acts.
179. Secondly, Actavis wanted to avoid the complication and extra expense of preliminary injunction proceedings. Under the law of England and Wales, it is well established that, when considering whether to grant an interim injunction to restrain actual or threatened infringement of a pharmaceutical patent by a generic supplier, one of the factors this Court will take into account is whether the generic supplier has cleared the way in advance of launch by revoking the patent or obtaining a DNI. In accordance with that jurisprudence, Actavis wished to avoid having to defend applications for preliminary injunctions by Lilly by obtaining DNIs in good time before launch.
180. By judgments dated 3 December 2012 and 28 January 2013 the Landgericht Düsseldorf (Düsseldorf Regional Court) rejected Actavis' jurisdictional challenge to the German proceedings brought by Lilly. On 4 March 2013 the Oberlandesgericht Düsseldorf (Düsseldorf Higher Regional Court) dismissed Actavis' appeal. As a result, there were concurrent proceedings for the next 13 months between Actavis and Lilly with respect to the issue of whether Actavis' proposed dealings in pemetrexed dipotassium would infringe the German designation of the Patent pending before the courts of both this country and Germany.
181. As explained in the second of my previous judgments, Actavis sent a series of "taking position" letters and initiated a series of further actions in this country. This was partly in order that additional Claimants obtain relief, partly in order to obtain relief in respect of pemetrexed ditromethamine and pemetrexed diacid as well as pemextexed dipotassium, and partly in order to try to circumvent objections raised by Lilly that Actavis had not satisfied various procedural pre-conditions for the grant of DNIs in respect of the French, German, Italian and Spanish designations under the laws of those countries (if, as Lilly contended but Actavis disputed, those laws were the applicable laws for that purpose).
182. In particular, on 17 April 2013 Actavis issued and served the Claim Form in the Third Action seeking DNIs in relation to the use of pemetrexed ditromethamine and pemetrexed diacid. This Claim Form was served under the cover of a letter of the same date, which is the first of the Letters. This Letter included the following passage:

"We confirm that, while the Patent remains in force in a relevant jurisdiction (e.g. unless and until revoked by the EPO), the launch of any such product will not take place until after both expiry of the relevant SPC and a final unappealable judgment (from the English Court, assuming the English court proceeds to hear the case over the relevant designation and does not dismiss or stay it on jurisdictional grounds; otherwise from whichever other Court has jurisdiction over the relevant designation) that provides Actavis with the declarations that its product will not infringe the relevant designations. For the avoidance of doubt, this means that all companies within the Actavis group of companies (not just those listed as Claimants) hereby undertake to be bound by this confirmation. We confirm that we have authority from all companies in the Actavis group to give that confirmation and undertaking.

Further, we also confirm that only those companies listed as Claimants will be involved in the launch or sale of a pemetrexed product in any of the relevant jurisdictions. The relevant Claimants for each jurisdiction are all of the Icelandic companies for all jurisdictions and the companies incorporated under the laws of the relevant individual jurisdiction. For example, the only companies that might (but not necessarily will) be involved in a launch or sale in Germany are Actavis Group ehf, Actavis Group PTC ehf, Actavis Deutschland GmbH & Co. KG, Medis ehf and Medis Pharma GmbH. For the avoidance of doubt, this means that all companies within the Actavis group of companies (not just those listed as Claimants) hereby undertake to be bound by this confirmation. We confirm that we have authority from all companies in the Actavis group to give that confirmation and undertaking.

If this leaves you in any doubt as to the risk of an imminent launch in any jurisdiction we should be grateful if you would inform us immediately so that any doubt in your mind can be removed and any unnecessary litigation can be avoided."

183. On 15 May 2013 Actavis PTC and Actavis Germany filed their initial substantive response to the Düsseldorf proceedings. It is common ground that one of the issues in the Düsseldorf proceedings was whether there existed a "real danger (or threat) of first infringement" by the defendants in those proceedings. According to the jurisprudence of the Federal Court of Justice, this will exist only if there are serious and tangible indications that the defendant will behave illegally in the near future. Actavis PTC and Actavis Germany contended that there was no real danger of first infringement by them. Although they did not refer in terms to the 17 April Letter, they did state (in paragraph 19 of the English translation, emphasis added):

"Actavis companies have made clear that they do not intend to launch the relevant products on the market in Germany before the expiry of the SPC in December 2015 and before they have obtained a judgment from a competent court holding that their relevant product does not infringe."

They went on (in paragraph 23) to say that "an unconditional decision to commit an infringement … has not yet been made".

184. On 21 May 2013 the Court of Appeal in Actavis I dismissed Lilly's appeal contesting jurisdiction in respect of the First Action and on 27 June 2013 the parties agreed that Lilly's jurisdictional challenge in respect of the Third Action should be dismissed.
185. In the meantime, on 14 June 2013 Lilly amended its Düsseldorf proceedings to bring a claim against Kálmán Petró, the managing director of Actavis Management GmbH (the managing partner in Actavis Germany), in respect of Actavis' proposed dealings in pemetrexed diacid and pemetrexed ditromethamine as well as pemetrexed dipotassium. As Dr von Falck frankly acknowledged in his witness statement, this was in order to get round the fact that, by virtue of the Third Action, this Court was first seized of the dispute between Actavis and Lilly with respect to pemetrexed diacid and pemetrexed ditromethamine. Furthermore, as Mr Jüngst pointed out, if Lilly had considered that Actavis were legally bound by the statements in the 17 April Letter, there would have been no reason for Lilly to enlarge its claim in this way.
186. In its amended claim, Lilly argued that for various reasons there was a real danger of first infringement by each of the defendants in respect of each of Actavis' proposed products. It referred to the 17 April Letter, but only to argue that this did not show that there was no such danger.
187. On 16 September 2013 Actavis sent a further taking position letter to Lilly, preparatory to the Fifth Action, which is the second of the Letters. This Letter included the following passage:

"As you are also aware, Actavis intends, on expiry in December 2015 of the SPCs that cover compound pemetrexed, to launch products of the following types (the 'products') in the territories of France, Italy, Germany, Spain and the United Kingdom (the 'Territories'). Actavis has already taken real and effective preparations for the manufacture of the Products in the Territories as specified here in after.

However Actavis has confirmed in on-going proceedings before the English Court (HC12E02962 and HP13A01487) that it does not intend to launch in a relevant Territory until after Actavis has obtained a judgment that provides Actavis with a declaration or similar decision that its Products will not infringe the respective designation in the relevant Territory."

188. In their subsequent "taking position" letters, Actavis reiterated and relied upon what they had said in the 16 September Letter. Like the 16 September Letter, these letters (and other statements by Actavis of their position in correspondence with Lilly, such as Actavis' fifth letter dated 8 April 2014) are expressed in terms of intentions and plans.
189. In written submissions to the Düsseldorf Regional Court dated 8 November 2013 Actavis made reference to the 17 April Letter. In paragraph 19 of the English translation, it was stated that "The Actavis Group has agreed in entirety not to launch any pemetrexed products prior to a final decision by the English Courts". As Mr Jüngst explained, however, this statement was made in the context of Actavis' argument that procedural economy militated in favour of a stay of the claims against Mr Petró.
190. More pertinently, in the context of its argument on "real danger", Actavis stated in paragraph 28 (emphasis added):

"On the contrary Defendant 2 – like all companies of the Actavis Group has repeatedly expressed its intention to respect the patent in suit. It would only launch a pemetrexed product which has first been found non-infringing in a court decision, If, contrary to expectation, the Actavis Group does not succeed in achieving such a 'Freedom to operate', the Defendants will not launch any of the pemetrexed products in suit. Plaintiff does not doubt the seriousness of this declaration by the Actavis Group. It therefore has not reason to fear an infringement of the patent in suit."

191. Actavis went on at paragraph 39 (emphasis added):

"A procedural declaration can be considered an assertion of patent claims only if, upon evaluation of the overall circumstances of the individual case, the declaration indicates the willingness to act in such a way immediately or not in the near future. However, this is precisely not so in the present case, because in the cover letter of 17 April 2013 …. Defendant 2 made it explicitly clear that it does not in fact intend unconditionally to act according to its legal view, but to submit to the English court's assessment. The declaration of intending to launch a pemetrexed is now subject to the clear condition of a court decision that is favourable for defendants."

192. As Mr Jüngst explained, Actavis' argument was that (among other factors relied on) the 17 April Letter showed that there was no real danger of first infringement.
193. In Lilly's written submissions in reply to the Düsseldorf Regional Court on 17 December 2013, Lilly stated among things (at paragraph 8 on pages 40-41 of the English translation):

"A danger of first infringement substantiated by arrogation can only be eliminated through a sufficient undertaking to cease and desist. However, it is obvious that the Defendants do not want to submit such an undertaking. The statement that they want to comply with a decision of the English court … cannot even slightly be compared with a sufficient undertaking to cease and desist. "

194. As Mr Jüngst pointed out, Lilly was thereby arguing that the Letters were insufficient to prevent there being a real danger precisely because they did not amount to binding undertakings. Furthermore, as Mr Jüngst explained, Lilly's submissions distinguished between the "declaration" or "confirmation" in the 17 April Letter and an undertaking which Actavis had given to this Court on 27 November 2012 because the Letter did "not constitute an obligation under penalty of law" and therefore did "not offer Plaintiff any legal protection which would come even half-way close to an enforceable order to cease and desist" (page 15 paragraph 5).
195. As for the 16 September Letter, Lilly argued (at page 37) that by this letter "Defendants once again expressed an arrogation" i.e. threatened to commit an infringement. Thus Lilly did not argue that this Letter amounted to a binding undertaking, but the opposite.
196. On 3 April 2014 the Düsseldorf Regional Court gave judgment. It declined to stay the proceedings against Mr Petró as Actavis had sought. It held that there was a real danger of first infringement by Actavis PTC in relation to pemetrexed dipotassium, because the letter dated 12 July 2012 was an unconditional threat of infringement which had not been neutralised by anything which had been said or done subsequently, including the Letters (pages 54 to 60 of the English translation). But it held that there was no threat by the other defendants, including by Mr Petró in relation to pemetrexed diacid and pemetrexed ditromethamine, because they were not party to the letter dated 12 July 2012. In particular, the Court held that the 16 September Letter did not evidence such a danger, in essence because it was a conditional statement of intention (pages 63 to 66). As for the 17 April Letter, the Court considered that this showed that the Third Action was not sufficient evidence of a real danger because Actavis had stated that they intended to act in accordance with the English court's decision (page 67). The Court therefore considered the merits of Lilly's claim in respect of pemetrexed dipotassium, but not the other products, holding that the scope of the claims of the Patent extended to pemetrexed dipotassium.
197. In the light of this judgment, Actavis decided to discontinue their English claims so far as they related to the German designation.
198. Actavis PTC appealed against the decision of the Düsseldorf Regional Court with respect to the merits of Lilly's claim in relation to pemetrexed dipotassium. Lilly did not appeal the decision with respect to pemetrexed diacid and pemetrexed ditromethamine. According to Dr von Falck, Lilly accepted that an appeal was useless because the Düsseldorf Higher Regional Court would come to the same conclusion with respect to the issue of real danger of first infringement.
199. On 5 March 2015 the Düsseldorf Higher Regional Court gave judgment on Actavis' appeal, holding that Actavis' proposed pemetrexed dipotassium product was not within the scope of the claims of the Patent (consistently with the decision of the English Court of Appeal in Actavis II) and that dealings in that product would not amount to indirect infringement even if it was diluted in saline (inconsistently with the decision of the Court of Appeal in Actavis II). In its judgment the Higher Regional Court characterised the 17 April Letter (at page 5 of the English translation) as a letter in which Actavis' solicitors "explained the intention to launch two embodiments pemetrexed ditromethamine and pemetrexed diacid only after expiration of the [SPC] for the substance pemetrexed, and only if there were in addition a legally binding judgment that assured to Actavis that its product did not infringe the patent in suit" (emphasis added).
200. Actavis' position is that, although the Higher Regional Court's reasoning is formally restricted to pemetrexed dipotassium, it is equally applicable to pemetrexed diacid and pemetrexed ditromethamine. It is difficult to see how it could be otherwise.
201. Lilly did not suggest that Actavis were bound by the statements contained in the Letters in any jurisdiction until after receipt of the draft judgment of the Court of Appeal in Actavis II on 21 May 2015, when it raised the Letters Issue in its submissions with respect to the order. Actavis disputed that they were bound by those statements in their submissions, and the Court of Appeal did not decide the issue.
202. On 27 July 2015 Actavis' solicitors wrote to Lilly's solicitors stating:

"In our letter of 10 July 2015 and in its Statement of Case on the Remission, Actavis has set out the steps which it will take to secure that Actavis AIs are reconstituted/diluted only with dextrose.

Actavis considers that, if those steps are taken (whether or not the additional step of supply of dextrose is required), it is clear that supply of the Actavis AIs will not infringe. Accordingly, in the United Kingdom, it is Actavis' intention to begin sales of its diacid pemetrexed product shortly after SPC expiry, and it is preparing accordingly. Obviously, if the trial of the remission occurs before SPC expiry, and in the unlikely event that it is determined against Actavis in any respects, the implications of any such decision will be considered and we will inform you of any changes to Actavis' stance accordingly.

In relation to France, Italy and Spain, Actavis will not seek to launch any Actavis AI prior to SPC expiry. However, although Actavis would be ready to launch shortly after SPC expiry in those territories from a commercial and regulatory perspective, in the present legal context Actavis has not yet finally determined any date on which it will seek to launch in those countries. We will update you with Actavis' position on this in due course and will provide you with at least 3 months' written notice of any intention to launch in those territories. Again, if the trial of the remission occurs before SPC expiry, and in the unlikely event it is determined against Actavis in any respects, we will inform you of any revisions to Actavis' stance accordingly.

You will be aware that it is Actavis' position that the 17 April letter had already been replaced by the 16 September letter. In any event, it follows from the above, and we hereby state, that the letters of 17 April 2013 (if still in existence) and 16 September 2013 are of no further effect, to the extent they may have been applicable in the present, much-changed circumstances (which is not accepted, and you are aware that our clients dispute the letters were ever legally binding and/or enforceable by your client; and further contend, in any event, that the letters were satisfied to the extent necessary by the effect of the Court of Appeal's judgment of 25 June 2015). Those earlier letters (to the extent still in existence) should be treated as having been replaced, in all respects, by this letter."

203. Despite this, Lilly did not apply for a preliminary injunction in the UK, France, Germany, Italy or Spain.
204. On 18 September 2015 Actavis notified Lilly that they intended to launch shortly after SPC expiry in France, Italy, Germany and Spain (although, in the event, launch has slipped back in those countries). Again, Lilly did not apply for a preliminary injunction in any of those countries. This is despite the fact that, according to Mr Jüngst, the statements made by Actavis in the letters dated 27 July and 18 September 2015 were "undoubtedly sufficient to constitute a threat of first infringement by Actavis" in Germany.
205. As noted above, on 22 October 2015 Lilly gave an undertaking not to seek a preliminary injunction in the UK, France, Italy and Spain as part of the price of being permitted to make a late amendment to its statement of case in respect of the Dextrose Remission Issue.
206. Lilly has recently been given permission to appeal to the Federal Court of Justice against the decision of the Düsseldorf Higher Regional Court on the merits of its claim with respect to pemetrexed dipotassium.

Applicable law and burden of proof

207. Both sides have approached the Letters Issue on the basis that the applicable law is English law. This is despite the fact that, as will appear, Lilly's case of reliance relates largely, if not exclusively, to Germany.
208. As with the Dextrose Remission Issue, the burden of proof lies upon Actavis as the party seeking negative declaratory relief. In this context the burden of proof is rather less significant, however, since Lilly has pleaded an affirmative case, the evidence is all in writing and there is very little dispute as to the facts.

Actavis' case

209. Actavis' case is that each of the Letters was a serious statement of Actavis' intentions, which was intended to be taken as such by Lilly, but that they did not give rise to contracts, nor were they otherwise legally binding. Actavis further contend that the 17 April Letter was replaced by the 16 September Letter, and that the Letters were revocable and have been revoked, in particular by the letter dated 27 July 2015. Finally, Actavis say that the terms of the Letters have in any event been satisfied by the judgments of the English Court of Appeal in Actavis II and of the Düsseldorf Higher Regional Court (at least if the reference to an unappealable decision in the 17 April Letter is taken to have been revoked by the 16 September Letter).

Lilly's case

210. Lilly's pleaded case in relation to the Letters is as follows:

"18. … The letters of 17 April 2013 and 16 September 2013 are contractual, legally binding and irrevocable. The letter of 16 September 2013 did not replace the letter of 17 April 2013. It merely referred back to it to confirm the existence and terms thereof and did not vary the terms as alleged or at all. At trial the defendant will rely upon the terms of the said letters for their full meaning and effect.

19. Actavis is not entitled to revoke the letters. Lilly has relied on and continues to rely on the terms of the letter of 17 April 2013, alternatively 16 September 2013, by not seeking interim relief in any of the relevant jurisdictions. Further the undertakings have been relied upon by Actavis, Lilly and Dusseldorf Landgericht in the proceedings in Germany as follows: Actavis in seeking to assert a lack of threat to infringe the German designation of the Patent in Germany in relation to two of Actavis' proposed products; the Dusseldorf Landgericht in holding there was no such threat (decision of 3 April 2014); and Lilly is not appealing that finding, to its detriment in that Lilly will now only be able to obtain a fully reasoned judgment in Germany in relation to infringement by the said two products by commencing fresh proceedings which will cause Lilly to incur to incur additional costs and suffer delay in obtaining judgment on such matters. Actavis is, accordingly, estopped from resiling from the undertakings."

211. Actavis interpreted this as a plea of contract, alternatively promissory estoppel. In Lilly's opening written submissions at trial, however, the principal case advanced by Lilly was one of estoppel by representation. Actavis immediately objected that this case was not open to Lilly since it had not been pleaded, and pre-emptively submitted that Lilly should not be given permission to amend its statement of case at such a late stage. Counsel for Lilly submitted in his closing submissions that Lilly's statement of case embraced a plea of estoppel by representation. He made no application to amend the statement of case if I did not accept that submission.
212. In my judgment Lilly's statement of case does not include a plea of estoppel by representation, as distinct from a plea of promissory estoppel. Lilly's statement of case does not plead (i) any representation of fact alleged to be conveyed by the Letters or (ii) Lilly's justifiable belief in the truth of that representation of fact.
213. Since there is no application by Lilly to amend, it is not necessary for me to say any more about this aspect of the matter. I would nevertheless add that I do not accept that the relevant passages of the Letters contain any representation of fact anyway (save as to immaterial matters, such as Bird & Bird's authority). Lilly's argument appears to be that they contain representations as to the legal effect of the Letters. But the first Letter does not contain any representation as the legal effect of the "confirmation" it conveys, or as to the legal effect of the statement that all companies in the Actavis group "undertake to be bound" by that confirmation. In particular, it does not represent that those statements have contractual force or are otherwise legally binding. Still less does the second Letter contain any such representation. Nor did Lilly adduce any evidence that it believed such a representation to be true, or even that that is how it understood the statements made.

Contract

214. The law. It is trite law that, in order for a contract to be formed, there must be (i) an offer, (ii) acceptance of that offer, (iii) an intention to create legal relations and (iv) consideration moving from the promisee. These matters must be objectively assessed.
215. Assessment. Lilly's statement of case fails to plead a case as to how Lilly contends that a contract was formed. This deficiency was not remedied in either Lilly's written opening submissions or written closing submissions, both of which contained nothing more than a bald assertion that "the Letters are contractual". When I asked him about this in his oral closing submissions, counsel for Lilly submitted that the Letters constituted offers, that they were accepted by Lilly by conduct and that Lilly gave consideration by relying upon the Letters.
216. In my judgment neither of the Letters contained any offer. True it is that the first Letter contained an undertaking, but the undertaking was a unilateral one, that is to say, a promise or statement of intention. Even if either of the Letters contained an offer, counsel for Lilly identified no conduct by Lilly which communicated to Actavis any acceptance of that offer. Rather, the conduct he relied upon was internal to Lilly. Even if there was offer and acceptance, there was no consideration moving from Lilly. Lilly did not promise Actavis to refrain from applying for preliminary injunctions (or to refrain from doing, or to do, anything else). Nor was there any other consideration. For each of these reasons, neither of the Letters gave rise to any contract. I would add that, if Lilly had wanted a contractual undertaking from Actavis, it would have been a simple matter to request one. Lilly did not do so.

Promissory estoppel

217. Although counsel for Lilly advanced no argument of promissory estoppel in his written opening submissions, he made it clear in his closing submissions that Lilly did maintain such a case.
218. The law. It is sufficient for the purposes of the present case to quote the formulation of the requirements of promissory estoppel set out in the 32^nd edition of Snell's Equity, which was cited with approval by Morgan J in Crossco (No.4) Unltd v Jolan Ltd [2011] EWHC 803 (Ch), [2011] NPC 38 at [332] (affd [2011] EWCA Civ 1619, [2012] 2 All ER 75), and is repeated in the 33^rd edition at 12-018 (footnotes omitted):

"Where, by his words or conduct one party to a transaction, (A) freely makes to the other (B) a clear and unequivocal promise or assurance that he or she will not enforce his or her strict legal rights, and that promise or assurance is intended to affect the legal relations between them (whether contractual or otherwise) or was reasonably understood by B to have that effect, and, before it is withdrawn, B acts upon it, altering his or her position so that it would be inequitable to permit the first party to withdraw the promise, the party making the promise or assurance will not be permitted to act inconsistently with it. B must also show that the promise was intended to be binding in the sense that (judged on an objective basis) it was intended to affect the legal relationship between the parties and A either knew or could have reasonably foreseen that B would act on it. Yet B's conduct need not derive its origin solely from A's encouragement or representation. The principal issue is whether A's representation had a sufficiently material influence on B's conduct to make it inequitable for A to depart from it."

219. Assessment. In my judgment Lilly's case of promissory estoppel falls at the first hurdle, since neither of the Letters contained a promise or assurance that Actavis would not enforce Actavis' strict legal rights. So far as is relevant to the present case, Actavis did not have, or even claim to have, any legal rights against Lilly. Rather, Actavis were in the process of seeking declarations from this Court that Lilly had no legal rights (by virtue of the Patent) against Actavis. In the passages relied upon by Lilly, Actavis were not making any promise or assurance about their legal rights, rather they were making a promise or assurance about their future conduct.
220. This is not to say that the relevant passages in the Letters were empty statements devoid of legal consequences. They were intended to have, and did have, legal effect by virtue of containing statements as to Actavis' intentions. If Actavis changed its intentions, and thereby broke their promise, Lilly would have had a remedy because (assuming that Lilly was otherwise entitled to do so) Lilly could have obtained a preliminary injunction in the relevant territory (or territories).
221. Nor has Lilly established that it relied upon any such promise or assurance. Lilly adduced no evidence from any witness from Lilly on this point. Instead it relied upon the evidence of Dr von Falck. Lilly's pleaded case is that it relied on the Letters by not seeking interim relief in any of the relevant jurisdictions. Since Dr von Falck said nothing about the position in the UK, France, Italy or Spain, he did not provide evidence of reliance in relation to those countries. In any event, Lilly plainly did not rely upon the Letters, since Lilly made no application for a preliminary injunction after receipt of the letter dated 27 July 2015.
222. As for Germany, although Dr von Falck used the phraseology of reliance in places in his witness statement, upon analysis his evidence does not establish reliance by Lilly even with respect to Germany, particularly when it is read together with Mr Jüngst's evidence. In my view it is clear from the evidence that Lilly did not rely upon the promise or assurance which Actavis gave in the Letters, but on the contrary relied upon the Letters as supporting Lilly's arguments with respect to real danger of infringement (or at least, as not contradicting those arguments).
223. Lilly also claims that it also relied upon the Letters in deciding not to appeal against the decision of the Düsseldorf Regional Court. In my view it is clear from the evidence that Lilly did no such thing. Rather, it accepted that the Düsseldorf Regional Court had reached the right decision. I would add that Lilly's suggestion that it has been prejudiced because it has not obtained a reasoned decision in respect of pemetrexed diacid and pemetrexed ditromethamine does not hold water either. If Lilly is successful in persuading the Federal Court of Justice that dealings in pemetrexed dipotassium amount to either direct or indirect infringement, it is plain that the decision will be equally applicable to pemetrexed diacid and pemetrexed ditromethamine, as Actavis have expressly accepted. Indeed, Actavis have offered Lilly a contractual agreement to this effect, but Lilly has not accepted that offer. Counsel for Actavis submitted that it was to be inferred that Lilly preferred the potential uncertainty due to the absence of a decision with respect to pemetrexed diacid and pemetrexed ditromethamine. I accept that submission.

Should a declaration be granted?

224. In my judgment Actavis have a real commercial interest in establishing that the Letters are not legally binding. Although Lilly has not attempted to enforce the statements made in the Letters, it has not merely steadfastly refused to accept that they are not legally binding, but also positively contended that they are. Making declarations will therefore serve a useful purpose.

Conclusion

225. I will make declarations along the lines proposed by Actavis. I will hear counsel as to the precise wording of the declarations.

Summary of principal conclusions

226. For the reasons given above, I conclude that:

i) At present, it is not foreseeable that the Actavis Product will be diluted in saline. Accordingly, for the foreseeable future, the supply by Actavis of the Actavis Product will not infringe the Patent. I will therefore grant DNIs in respect of the UK, French, Italian and Spanish designations of the Patent. I will give both sides permission to apply to this Court in the event of a material change of circumstances.

ii) The Letters did not give rise to any contract between Actavis and Lilly, nor did they found any promissory estoppel binding upon Actavis. I will therefore grant Actavis appropriate declaratory relief.