England and Wales High Court (Patents Court) Decisions

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
PATENTS COURT

		Royal Courts of Justice, Rolls Building Fetter Lane, London, EC4A 1NL
		21/05/2015

Mr Justice Birss :

Topic	Paragraph
Introduction	1
The issues	8
The witnesses	13
The skilled person	28
Common general knowledge	29
The patents	68
Claim construction	78
Novelty:	87
Novelty – prior sales of Copaxone	90
Novelty – Lemmon	96
Inventive step	104
Inventive step – levels of bromine and use of a bromine scavenger	106
Inventive step – metal ion impurities	128
Insufficiency	133
Added matter	140
The Dutch decision	161
Conclusion	165
Annex 1 - the claims	Annex

Introduction

1. This case concerns patents EP (UK) 2 177 528 and EP (UK) 2 361 924, both entitled "Process for the preparation of mixtures of trifluoroacetyl GA using purified hydrobromic acid". The patents are divisionals. They claim priority from a US filing on 9^th September 2004. The 528 patent was granted on 7^th December 2011 and the 924 patent on 11^th December 2013. The patents belong to the defendant, Teva. The claimant, Synthon contends they are invalid and applies to revoke them in this action.
2. The parent case (EP (UK) 1 799 703) was revoked by the EPO in opposition proceedings pursuant to Teva's withdrawal of it after a negative preliminary opinion. The 528 patent was upheld by the Opposition Division of the EPO in amended form (removing claim 13). That opposition is under appeal. The 924 patent has also been opposed in the EPO but that process has not yet reached the OD. Recently the Dutch designations of both the 528 and 924 patents have been revoked by the Hague District Court. Teva will appeal that decision. Teva contends that the reasons for the decision in the Netherlands are different from those in the pleadings and evidence here.
3. The patents relate to glatiramer acetate. Glatiramer acetate consists of a mixture of synthetic polypeptides made from four amino acids. A low molecular weight fraction of glatiramer acetate is sold as a pharmaceutical product under the trade mark Copaxone by Teva for the treatment for relapsing remitting multiple sclerosis. Copaxone is one of a limited number of drugs that have been found to be effective in the treatment of this debilitating disease of the nervous system.
4. Glatiramer acetate was first formulated in the late 1960s and was later tested in an experimental animal model for multiple sclerosis called autoimmune encephalomyelitis. Research showing its effectiveness which had been undertaken at the Weizmann Institute in Rehovot, Israel was published in 1971 and suggested the product as a potential treatment for multiple sclerosis. A US patent No. 3 849 550 was granted covering it. By 1987 it had been demonstrated to be a potentially effective treatment for exacerbating-remitting multiple sclerosis. The product's efficacy may derive from its composition as a synthetic analogue of a natural component in the myelin sheath. The myelin sheath surrounds and protects nerve fibres and is damaged in multiple sclerosis.
5. Both the 1971 paper and the 550 patent disclose the basic process for making glatiramer acetate. An international patent application (WO 95/31990) for an improved process was published on 30^th November 1995. The applicants were Lemmon Corporation and Yeda Research and Development Co. Ltd and the document has been referred to in this case as Lemmon. Lemmon was directed to producing the low molecular weight fraction of glatiramer acetate. This reduced toxicity while preserving efficacy. The subject matter of Lemmon came before the Patents Court. The Lemmon patent was found valid by Arnold J ([2012] EWHC 1848 (Pat)) and that judgment was upheld on appeal ([2013] EWCA 925 Civ).
6. Today Copaxone is responsible for $4.2 billion worth of annual sales worldwide, which represents 21% of the Teva's group's total revenue.
7. The two patents in this case relate to what are said to be improvements in the process for making glatiramer acetate.

The issues

8. Synthon challenges the validity of the patents on four grounds, as follows:

i) Each of the product claims of the 924 patent lack novelty in the light of pre-priority date sales of Copaxone;

ii) All of the claims of both the 924 and 528 patents lack novelty and/or lack inventive step over Lemmon;

iii) Both patents are insufficient;

iv) Certain claims in the 924 patent are invalid for added matter.

9. A large number of claims are alleged to be independently valid. For the 528 patent they are claims 1, 2, 3, 6 and 10. For the 924 patent they are claims 1, 8, 11-13, 16, 22, 23, 24, 25, 26, and 29. All the claims are set out in Annex 1. I will refer to allegedly independently valid claims as "i.v. claims".
10. There are four points of detail to note about the i.v. claims. First, claims 23 and 24 of 924 are only alleged to be independently valid for novelty. They are not said to be independently inventive. Second, some of the i.v. claims can only be understood having regard to other claims from which they depend but which are not themselves alleged to be independently valid. They are identified in Annex 1 as well. Third, although claim 22 depends ultimately on claims 19 or 20, Teva only relies upon claim 22 to the extent it depends on claim 20. Fourth Synthon's added matter arguments and Teva's defence of added matter covered more claims than the i.v. claims and can only be dealt with that way.
11. I am bound to say Teva's approach to the independent validity of the claims of these patents is unconvincing. I can quite see how there might be bare novelty of multiple claims with the cascading levels of bromine or metal ions but it is hard to credit the thought that there might be independent inventiveness in the large majority of the claims relied on.
12. There is an uncontroversial application to amend the 528 patent to bring it into line with the form of the claims allowed by the OD of the EPO by deleting claim 13. I will allow the amendment.

The witnesses

13. Synthon called Professor Alethea Tabor. Prof Tabor is Professor of Chemical Biology at University College London. She is a member of the Royal Society of Chemistry and, within the RSC, a member of the Protein and Peptides Science Group Committee. Her current research interest is in the field of solid-phase peptide synthesis. Prof Tabor produced a single expert's report.
14. Teva called Professor Ben Davis. Prof Davis is Professor of Chemistry at Oxford University. His election as a Fellow of the Royal Society was announced on 1^st May 2015. His research centres on biomolecular functions with an emphasis on proteins and carbohydrates. Of some 200 papers naming Prof Davis as an author, 110 are in the field of proteins/peptide science. Prof Davis produced an expert's report and a report in reply to Prof Tabor's report.
15. Teva accepted that Prof Tabor had given her oral evidence fairly but mounted a sustained attack on Prof Tabor's written evidence, submitting that it fell a long way short of complying with the obligations of an expert. A particular criticism was that there were important matters on which she agreed with points made by Prof Davis in his reply report which qualified evidence stated in her main report however, when Prof Tabor came to swear to the accuracy of her report, she did not correct those matters although they had already been pointed out. When this was put to Prof Tabor in cross-examination she was clearly embarrassed about it and explained that she had drafted various things in response but had been advised by the legal team acting for Synthon that there was no need for her to serve a reply report and the points were better raised in cross-examination. During closing speeches Mr Silverleaf took responsibility for this and submitted it did not and should not undermine Prof Tabor's credibility as an expert witness and her ability to help the court on the issues to be decided. He also submitted that the criticisms were grossly overstated.
16. This is the first occasion on which Prof Tabor has given evidence in a patent case. It is unfortunate that she was advised not to produce a reply report. Expert witnesses owe a duty to the court (CPR r35.3(1)). This duty overrides any duty they have to the party and its team of legal advisers and CPR r35.3(2) makes that expressly clear. Inevitably experts rely on what they are told by the legal team and this is particularly true in patent cases when the nature of the subject matter means that the experts are often not experts at being expert witnesses. Prof Tabor is at fault for swearing to a report which contains statements which, by the time she swore to it, she knew needed qualification in material ways.
17. The relevant instances are: Prof Tabor's evidence on benzyl cations (I will address the detail below), her unqualified statement that all halogen acids may contain traces of free halogen (an overgeneralisation), her unqualified statement that tyrosine is particularly susceptible to reaction with any electrophile (in fact it only reacts with certain electrophiles and conditions also matter), her statement that certain materials would be "free" of metal ions/ bromine/ brominated material when that is not the case, and her statement that final glatiramer acetate would be "free" of metal ions was overstated. The qualifications were given readily when the questions were put and I am quite certain Prof Tabor in no sense set out to mislead. However Prof Tabor should have exercised her own judgment on the matter. It hardly needs saying that experts bear a personal responsibility for their evidence. If she had then I am sure a reply report would have been written and when Prof Tabor swore to the truth of the two reports, what she was swearing to would have fairly represented her actual opinions.
18. Another criticism levelled by Teva is that Prof Tabor's report includes a quotation from a student primer text book by Prof John Jones without exhibiting the extract. That is unfortunate because the quoted paragraph is highly selective, omitting material just before it which shows why it is of much less relevance. Again Prof Tabor accepted this in cross-examination. The decision not to exhibit the extract seems to have been taken by Synthon's legal team. The passage should not have been selectively quoted in this way at all, but at least if the full extract had been exhibited, the problem would have been mitigated.
19. Prof Tabor's personal position is mitigated very considerably by the advice she was given by the legal team. That sort of advice puts an expert in an impossible position. Despite the fact that clients may not like it, experts need to be advised fully about their personal responsibilities and role. They need to be advised that they must exercise their own judgment in matters of this kind. Whether an expert serves a reply report is first and foremost a matter for the expert not a matter for the lawyers. Prof Tabor should have been advised accordingly but she was not. The real cause of the problem in this case was not Prof Tabor.
20. The Patents Court is fortunate to have the assistance of very well qualified persons who are genuine experts in their field in order to help the court in deciding this kind of patent dispute. In order to have experts selected by the parties who are selected for their technical experience rather than their skills as witnesses, inevitably the kinds of people chosen will have less experience as expert witnesses than they might otherwise. That makes the experts reliant on advice and support from the legal team. This episode is a lesson in what can go wrong.
21. Teva levelled another criticism of Prof Tabor's report in that it failed to mention things it ought to have done. They were: that she had not used HBr/AcOH to deprotect, that she had searched textbooks on peptide synthesis to support her view on common general knowledge but had not found support, and that she had found a paper by Merrifield (1966 see below). The first two points should have been mentioned. The third point seems to have been another result of advice from the legal team.
22. There was a confidentiality regime in place to deal with some disclosure documents from Teva. Oddly Synthon decided not to introduce Prof Tabor into that regime. Some documents related to prior art Copaxone but I did not understand Teva's criticism of Prof Tabor here since those documents played no part in the trial. However other documents related to research done into red coloured Copaxone (discussed below) which Prof Davis relied on expressly in his report. Prof Tabor ought to have been shown the documents. If there was a problem with the confidentiality regime that should have been raised with the court in advance of trial. It was not.
23. I have accepted many of the criticisms of Prof Tabor's written evidence levelled by Teva but it would not be fair to her to leave this issue without paying tribute to her oral evidence. The contrast between the two is very striking. In the witness box Prof Tabor was a model witness: she is clearly an expert in her chosen field, gave candid evidence and answered questions without equivocation. I infer that the root cause of the problem was the manner in which Prof Tabor's expert's report was prepared, the fact that she had not undertaken this exercise before and her over reliance on the advice of the Synthon legal team.
24. Synthon's criticisms of Prof Davis were not as wide ranging or extensive as Teva's of Prof Tabor but they were significant nevertheless. They were pressed on closing primarily because Teva had contended that the court should prefer Prof Davis on all points over Prof Tabor because of the impact of the points raised above.
25. Synthon contended that at times it was difficult to get Prof Davis to answer the question, that at times he chose not to engage with the issues, and that he was on occasion argumentative. These points related to particular occasions in the cross-examination, particularly questions about (i) the colour of HBr/AcOH and the possible effect of free bromine on that colour and (ii) Example 1 of the patent concerning amino acid analysis and chromatography conditions. There is some force in Synthon's submission in relation to Example 1 and I will deal with that in context. Part of the problem in the cross-examination about colour was due to Mr Silverleaf's questions but in my judgment Prof Davis did have the occasional tendency to get carried away, finding fault in the questions rather than seeking to help the court. However overall I reject the suggestion (if made) that these aspects of Prof Davis' testimony undermine his reliability as a witness in this action.
26. Standing back I completely reject Teva's wholesale submission that I should prefer Prof Davis' evidence to that of Prof Tabor. Teva were right and entitled to raise the issues they did concerning Prof Tabor's report but having seen both witnesses in cross-examination, the wholesale submission is over the top and unfair. Before me at trial both experts set out to give their fair and unbiased opinions to the court. I understand the issues in this case much better as a result of the work of both Professors and I am grateful to them for their testimony at the hearing.
27. Teva also served a number of Civil Evidence Act notices to rely on internal documents. I will deal with these in context.

The skilled person

28. The skilled person relevant for all purposes in this case is a synthetic chemist with practical experience of peptide synthesis.

Common general knowledge

29. I will take the law on common general knowledge to be that in the statement made by Arnold J in KCI Licensing Inc v Smith & Nephew plc [2010] EWHC 1487 (Pat) at paragraphs 105 – 115 which was approved by the Court of Appeal [2010] EWCA Civ 1260 at paragraph 6.
30. The parties prepared an agreed summary of the technical background to the case in the form of a technical primer. It was common ground that subject to three qualifications, the whole of the primer represented part of the common general knowledge of the skilled person. The three qualifications were about paragraph 30 (which related to tyrosine), paragraph 31 (which related to glatiramer acetate) and figure 6 (a sample amino acid trace). The points on tyrosine were dealt with by the experts and I will address them below. The point on glatiramer acetate is simply that although it is necessary and convenient to summarise the known synthesis of glatiramer acetate as part of the background, that information was not itself part of the common general knowledge. Neither side suggested otherwise and Prof Tabor was unaware of the synthesis until becoming involved in this case. The point on figure 6 was that the experts agreed that the technique of amino acid analysis would have been common general knowledge. The particular trace was not itself common general knowledge (nor was it said to be). There was a debate about whether it was representative and I will deal with that in context.
31. The main points in the primer were as follows.
32. Proteins are polymers made up of amino acids joined together by peptide bonds. Each amino acid has a carboxylic acid group and amino group. Each one also has a distinct side chain. Amino acids are linked together to form peptides by covalent bonds formed when the carboxylic acid group of one amino acid joins to the amino group of another. The four amino acids of interest in this case are alanine, glutamic acid, lysine and tyrosine. They are:

33. Some side chains are inert in the conditions used to make amino acid polymers and some side chains are able to react during the polymerisation reaction. It is common practice when amino acids have reactive side chains to mask or protect those side chains during the polymerisation reaction. Three of the four relevant amino acids have reactive side chains which may potentially engage in side reactions during polymerisation. They are lysine, glutamic acid and tyrosine.
34. The terms peptide and polypeptide do not bear a precise meaning but are used frequently without difficulty. They are both polymer molecules consisting of amino acids joined together with peptide bonds. A smaller chain is often called a peptide whereas the word polypeptide conveys the idea of a longer chain.

Peptide synthesis

35. In order to facilitate formation of the peptide bonds, it is common practice to activate the amino acid. This involves creating a leaving group at the acyl carbon of the carboxylic acid group. Activation is necessary because otherwise carboxylic acids react with amines at ambient temperature to form salts.
36. One method used to make peptides and polypeptides is solid-phase peptide synthesis (SPPS). This technique is used to make polymers with a defined sequence of amino acids. Normally the structure is built up from the carboxylic acid terminated end (the C-terminus). The C-terminal residue is fixed to a solid support via the carboxylic acid group. The next amino acid in the sequence is introduced. It has the amino group protected to prevent unwanted reactions and the carboxy group activated. The first amide bond is formed. The protecting group on the end of the chain is removed to reveal an amino group ready to react with the carboxy group of a third amino acid.
37. A method used for making random polypeptides is random polymerisation. This leads to polypeptides in which the amino acids have been attached in an unpredictable order, thereby leading to a mixture of molecules with many different sequences of amino acids. One such method for synthesising random polypeptides is via the use of an activated amino acid in N-carboxyanhydride (NCA) form. The NCA forms of alanine, glutamic acid, tyrosine and lysine are:

38. NCAs can be reacted together to form random polypeptides in the following way. The NCAs are treated with a small amount of an amine initiator in an organic solvent. The initiator reacts with the carboxyl moiety of the NCA and forms an amide bond. This also leads to loss of carbon dioxide from the intermediate, which unmasks the amine group of the NCA. The unmasked amine group can then react with another NCA activated amino acid at the carboxyl and unmask the amine group on that molecule and so on.
39. Use of more than one NCA results in a mixture of polypeptides with variable chain length, composition and sequence. If each NCA in the reaction mixture reacts at the same rate then the sequence and composition of the polypeptide will be random. If some NCAs react faster than others then there will be a tendency for these NCAs to incorporate preferentially but the resulting sequence will still be random. Varying the amount of initiator varies the chain length since more chains will be started for a given supply of NCAs.
40. The idea of masking or protecting side chains is mentioned above. This is undertaken by using a protecting group. A useful protecting group is one which can be introduced easily, poses no problems of its own whilst in position, stays firmly in place as long as needed and can be removed easily under conditions which do not have any unwanted effects. A well-known protecting group for glutamic acid is a benzyl group to form a ?-benzyl ester and a well-known protecting group for lysine is a trifluoroacetyl (TFA) group to form the corresponding amide. The glutamic acid and lysine NCAs with those protecting groups are:

41. There are various possibilities for removing the ?-benzyl protecting group from the glutamic acid side chain. One is hydrogenolysis and another is acidolysis:

i) One version of hydrogenolysis involves the protected amino acid polymer interacting with a palladium charcoal matrix solid under an atmosphere of hydrogen gas. The advantage of using this technique is that it is possible to carry out hydrogenolysis under mild conditions compatible with other parts of the polymer chain.

ii) Acidolysis uses reagents such as hydrobromic acid (HBr) in glacial acetic acid (AcOH) or trifluoroacetic acid (TFA). HBr in acetic acid is a convenient reagent to use as many protected peptides are soluble in glacial acetic acid. Following reaction with HBr in acetic acid, the deprotected product may be precipitated out as a hydrobromide salt by the addition of a suitable solvent e.g. diethyl ether.

42. HBr is a strong acid, and HBr in acetic acid is a highly corrosive reagent. Due to its highly corrosive nature, the reagent is supplied and stored in glass or plastic bottles or containers and reactions involving it may be carried out in non-metallic vessels.
43. One means of removing the trifluoroacetyl protecting group from the lysine residue is to use a base such as piperidine.

Characterisation of polypeptides

44. Random polypeptides will be made up of molecules with undefined sequence and length. Although it is very difficult to separate and analyse every different peptide component from the polymer mixture, it is desirable to be able to characterise the amino acid composition, the range and distribution of molecular weights, and levels of impurities in the polymer. Techniques routinely used for this analysis are:

i) Amino acid analysis. This involves breaking down the polypeptide into its constituent amino acids and determining the amounts of different amino acids present, usually by chromatography.

ii) Size Exclusion Chromatography (SEC). This is a chromatographic technique using gel permeation or gel filtration. It involves separating the polypeptide molecules by "size". The retention time on the column is a function of size (strictly hydrodynamic volume).

iii) Nuclear Magnetic Resonance Spectroscopy (NMR). This technique is used to investigate whether chemical functional groups are present in a sample and if so, in what relative amounts. Proton NMR spectra of random polypeptides can provide information about the relative ratio of incorporation of the amino acids (by comparing the ratios of the peaks characteristic for each amino acid) and the degree to which polymerisation has occurred (DP) (by comparing the ratios of the characteristic amino acid peaks to the peaks corresponding to the amine initiator).

iv) Matrix Assisted Laser Desorption/Ionisation Time of Flight Mass Spectrometry (MALDI-TOF). This is a type of mass spectrometry. When applied to a random polypeptide it can provide information on molecular weight distribution.

45. The foregoing matters were common ground. Before I deal with the contentious aspects of common general knowledge, it is convenient to summarise the synthesis of glatiramer acetate in order to put them into context.

Technical background - the synthesis of glatiramer acetate

46. The same basic synthetic scheme is set out in the 1971 paper (Eur J Immunol [1971] 1:242 Teitelbaum et al) and in the US 550 patent mentioned above. As mentioned already glatiramer acetate is a random synthetic polypeptide that is a mixture of the four amino acids: alanine, glutamic acid, lysine and tyrosine in no pre-determined order. The length of each polypeptide chain and its composition varies randomly from polypeptide to polypeptide, although overall the ratio of the amino acids is approximately 6:2:5:1.
47. The first step is to polymerise the NCAs of tyrosine, alanine, ?-benzyl glutamate and N-trifluoroacetyl lysine. The side chains of glutamic acid and lysine are protected by a ?-benzyl and trifluoroacetyl protecting group respectively. Although tyrosine has a free hydroxyl group which can be (and sometimes is) protected, no tyrosine protecting group is used in the synthesis of glatiramer acetate. The reaction takes place in solution in dioxane. Amounts of the four NCAs are mixed together and an initiator added to start the polymerisation process. The amounts of the amino acids incorporated in the polypeptides are determined by the ratio of starting NCAs used and their relative reactivity.
48. The next step is deprotection of the ?-benzyl group used to protect the free carboxylic acid group in glutamic acid. This first deprotection step is central to this case. The protecting group could be removed by hydrogenolysis or acidolysis. Acidolysis was more practical for large scale reactions. The 550 patent described using 33% HBr/AcOH to deprotect the glutamic acid residue. Lemmon teaches the application of the HBr/AcOH for a longer time and at a rather warmer temperature in order to achieve two purposes. First is deprotection to remove the ?-benzyl group and second is cleaving the polypeptide chains into shorter chains, producing a lower average molecular weight in the resulting mixture. Example 4 of Lemmon describes this method.
49. The product produced by the first deprotection step is sometimes referred to as TFA-glatiramer acetate because the lysine residues in the polypeptide mixture are still protected by the TFA groups.
50. In the second deprotection step the TFA protecting group is removed by treatment with piperidine in water. Finally there is a purification step using dialysis. The resulting product is glatiramer acetate.

Elements of common general knowledge in issue

51. Prof Davis expressed the view that peptide synthesis was not a vital area of chemical research at the priority date (2004). His point was that by this time automated solid phase synthesis had become routine and prevalent. This point cannot be taken too far. It was by no means a moribund field, but nevertheless I accept the Professor's characterisation. It is relevant to the thinking of the skilled person.
52. Synthon submitted that it was well known at the priority date that HBr could contain traces of free bromine (Br₂). Whereas HBr and HBr/AcOH were essentially colourless, bromine is a deep red-brown colour and even the presence of a little bromine could impart colour to the gas or solution. Commercially available HBr/AcOH is specified as being colourless to yellow or brown in colour. Synthon submitted that the primary source of this colour is bromine. These points were supported by Prof Tabor. Prof Davis key points in response were two. First was that there is not a quantitative relationship between colour and bromine level. For example any free bromine in HBr/AcOH could equilibrate with the bromide ion Br^- from the HBr to form Br₃- which is itself colourless. Second was that free bromine could be a likely potential cause of such a colour but there were other possible causes and a skilled person would not make assumptions and would not act on assumptions without evidence.
53. On this I preferred Prof Tabor's evidence as reflective of the thinking of the skilled person if they were prompted to consider what might be the reason why a sample of HBr/AcOH before them was slightly coloured. They would infer that any yellow or brown colour was likely to be caused by some free bromine in the solution. I accept that the skilled person would not expect to be able to quantitate the bromine level based on colour and they would not assume there was very much free bromine at all. Bromine is highly coloured and very little may be all that is needed to cause colouration. Finally they would know that the colour might not be due in whole or in part to free bromine but they would infer that this was the most likely explanation.
54. Prof Tabor had looked for support for her view about the presence of free bromine in HBr/AcOH in six textbooks but not found it. That is a different point, relevant to the next issue. What I am dealing with at this stage is what a skilled person would infer if prompted to think about the colour of a yellow/brown HBr/AcOH solution. They would think it likely to be caused by free bromine.
55. A different but related question is about how much free bromine the skilled person would think was in HBr/AcOH which was colourless or more or less colourless, for example because it had been freshly made. Prof Tabor said in her report that the skilled person would assume that colourless HBr/AcOH was free of bromine impurities or only had trace amounts. The significance of the word "free" is that it appears in the patent claims. In cross-examination the Professor qualified her evidence about her usage of the term free (both for bromine and other impurities). I do not accept that the skilled person would make the assumption described in Professor Tabor's report. They would, on the premise they thought about it, assume that the level was low whether the solution was light yellow or colourless. Despite Prof Davis' evidence, I find that they would also assume it was likely to be lower in less coloured material but they would not know, and would know they did not know, what the level actually was.
56. A different question again is how much free bromine is in fact present in a given batch of HBr/AcOH. That is an issue of underlying fact rather than one about the state of knowledge of the skilled person. It was not established. The highest free bromine level mentioned in any of the patent claims is 0.5%. Having heard the evidence in this case, I suspect that the bromine in freshly prepared HBr/AcOH kept away from light and oxygen is at a concentration below that value, but suspicion is not the right approach. There is no reliable evidence to which my attention has been drawn. Synthon has not established that such HBr/AcOH has a level of free bromine below 0.5%. It has certainly not established that the level would be below 0.1% (claim 1 of 924) nor has it established that HBr/AcOH would be "free" of free bromine.
57. The next issue is whether the common general knowledge included knowledge of a problem or potential problem due to free bromine present in HBr/AcOH when used in a deprotection reaction reacting with the phenol group in tyrosine residues. And as part and parcel of that, whether the common general knowledge included the idea of avoiding this problem either by using HBr/AcOH free of bromine by removing the free bromine using a scavenger or obtaining bromine free HBr/AcOH in some other suitable way. Synthon submitted these matters were common general knowledge, supported by Prof Tabor while Teva did not agree, supported by Prof Davis.
58. This was a major dispute in the case but in the end I think the position was quite clear.
59. First the basic chemistry is not in dispute. When free halogen molecules approach electron rich regions such as those in benzene or phenol, the halogen molecules can become polarised and can act as an electrophile. For example bromine in water will readily brominate phenol. The relevant reaction in this case is bromine in an HBr/AcOH solution brominating the phenolic side chain of tyrosine. This chemistry supports Synthon.
60. On the other hand, second, the experience of the experts on this issue was quite different. Prof Davis routinely uses HBr/AcOH deprotection reactions in his laboratory and he explained that when he does this he does not have any concerns over HBr/AcOH. He regards it to be a robust unproblematic reagent. He explained that when he instructs his students to perform this deprotection step using HBr/AcOH he does not advise that the bromination of tyrosine is a possible concern and does not advise the addition of a scavenger. On the other hand Prof Tabor had never performed a deprotection reaction on a polypeptide using HBr/AcOH. She had to check whether there was any HBr in her lab at all as she did not know because she did not use it regularly. For these reasons I prefer the opinion of Prof Davis on this issue to that of Prof Tabor.
61. Third the position on the textbooks favours Teva. No textbook presented in evidence mentioned any concern about brominating tyrosine residues as a result of carrying out deprotection using HBr/AcOH. The six textbooks considered by Prof Tabor did not support her view. Extracts from two textbooks by Professor John Jones were in evidence. One was the student primer. It did discuss using HBr/AcOH deprotection to remove benzyl protecting groups and problems with deprotection and residue specific concerns but did not mention the risk of bromination of tyrosine. A small but not irrelevant point also was that it did mention contaminants, so might have mentioned bromine in HBr/AcOH if that was thought to have been a situation in which contamination mattered.
62. Fourth, specification sheets for HBr/AcOH were in evidence but while they show that HBr/AcOH could be coloured, they do not demonstrate that those skilled in the art were concerned about any free bromine in the solution causing unwanted reactions.
63. Fifth, Synthon relied on a number of papers put to Prof Davis in cross-examination. They do show the use of a scavenger such as phenol (and others) in analogous contexts but none of them is concerned with liquid phase deprotection reactions using HBr/AcOH. A paper by Marglin and Merrifield in 1966 was put. Prof Merrifield won a Nobel Prize for his work on SPPS. The 1966 paper is part of his work in this area but is not the frequently cited paper, which came earlier. In the 1966 paper HBr is used to cleave the peptide from the resin support. The paper mentions first bubbling the HBr gas through a solution of resorcinol in TFA to remove trace bromine gas before then applying the solid resin in another TFA solution. The HBr gas can undergo photolysis under light and bromine is formed. The bromine can react with tyrosine residues and so a scavenger is added. The same work as Merrifield 1966 is cited in a Merrifield text book from 1980. Extracts for a textbook Stewart & Young (1984) was in evidence. It has a foreword by Prof Merrifield and mentions the reaction described in the 1966 paper. I find neither the text book nor the paper were common general knowledge at the priority date.
64. The other point mentioned in Stewart & Young is high temperature thermolysis for amino acid analysis of proteins. To perform amino acid analysis the peptide bonds have to be broken to split the chain back to its constituent amino acids so that they can be analysed. This is performed by heating at 110ºC for 24 hrs with HCl. In these conditions the HCl forms some Cl₂ which can halogenate the tyrosine residues. Phenol is added to prevent chlorination of tyrosine. Synthon also put two other documents (the European Pharmacopoeia and a paper by a company called Bio-separations) which addressed the same point. The same use of phenol in the same context can be seen in the patent itself in the amino acid analysis Example 1 at paragraph [0103]. The expedient of adding a scavenger like phenol to deal with this potential chlorination problem in this kind of amino acid analysis was clearly common general knowledge. However it does not follow that the skilled person would link this, as a matter of common general knowledge, to the use of HBr/AcOH in deprotection during synthesis.
65. Synthon also relied on a Novabiochem catalogue. This related to the use of HBr in TFA and concerned a problem caused by benzyl cations reacting with residues such as tyrosine and the use of scavengers to deal with the problem. The scavenger mentioned which is added to TFA is thioanisole. However the fact this document is concerned with HBr in TFA rather than HBr in AcOH is significant. There is an important difference between these two systems. It was common general knowledge that using HBr in TFA in benzyl deprotection proceeded by an S_N1 mechanism and produced benzyl cations. Benzyl cations do create a risk of side reactions and scavengers were employed to prevent them. However when HBr is employed to carry out the benzyl deprotection in solution in AcOH rather than TFA, the alternative S_N2 mechanism is promoted. That minimises the production of benzyl cations such that there is no need to add a scavenger to deal with benzyl cations.
66. Synthon submitted that the wide range of publications in which this teaching appeared, with papers at every level of practical demonstration and in different contexts, demonstrated the point was common general knowledge. I do not agree. These papers do not represent a single teaching. I accept that, looked at with the knowledge of this case, one can see chemistry in them which is related to the chemistry applicable to the use of a scavenger in HBr/AcOH to prevent bromination of tyrosine but these publications taken alone or together do not support Synthon's point. They are in fact concerned with reactions in a variety of different circumstances and conditions. None of them are the circumstances arising in this case. What has happened is that Synthon has looked back into the literature knowing about the reaction in issue in this case and found disparate references to similar reactions, some in closer circumstances than others. That does not demonstrate common general knowledge.
67. In conclusion, I find that the skilled person did not, as a matter of common general knowledge, have a concern about a risk of a side reaction with a tyrosine residue and free bromine present in HBr/AcOH when used in a liquid phase deprotection reaction or cleavage reaction. A skilled person taught to perform that reaction would not set out to use HBr/AcOH free of bromine nor would they seek to remove free bromine by using a scavenger. They would not think about free bromine in HBr/AcOH at all. The fact that, given a prompt to do so, they would conclude that colour in HBr/AcOH is most likely caused by some free bromine makes no difference. First they would not be prompted to consider it. HBr/AcOH would be regarded as a robust reagent. Second even if they thought there might be some free bromine in the material they would not have any reason to expect it to create an appreciable risk of a side reaction with tyrosine in the relevant conditions.

The patents

68. The specifications of the patents are sufficiently similar for it to be convenient to deal only with the specification of the 924 patent. The patents refer to glatiramer acetate as GA. They describe the synthesis of GA and TFA-GA. The term TFA-GA refers to GA in the form in which the lysine residues are still protected by TFA.
69. The core synthetic steps disclosed in the patents have been described already: polymerisation, benzyl deprotection, polypeptide cleavage and deprotection of TFA lysine residues. Paragraphs [0090] – [0099] of the specification explain the two aspects of the patents as relating to free bromine and metal impurities.
70. Paragraphs [0090] to [0093] state that during the development of the production process for GA it was found that some of the tyrosines were brominated and that it was found that free bromine present in the HBr/AcOH used in the synthesis of TFA-GA caused the bromination of the tyrosine residues. The patent describes how this reaction can be prevented by adding a bromine scavenger, such as phenol or sodium bisulfite, to the HBr/AcOH solution before it is reacted with protected TFA–GA. The specification goes on to explain (at paragraph [0108]) that to reduce the level of the brominated tyrosine residue impurity (bromotyrosine) in the final product to a level of less than 0.2% in the final product, the level of free bromine in the HBr/AcOH must be lowered by the addition of a bromine scavenger.
71. Example 1 describes the measurement of the amount of bromotyrosine present in batches of TFA-GA and GA. The example also describes experiments aimed at demonstrating the effect of the amount of free bromine added to HBr/AcOH on the level of bromotyrosine impurity in the resulting product. Table 1 shows that as the amount of free bromine added to the HBr/AcOH increases, so does the amount of bromotyrosine recovered from resulting batches of product. Even where no bromine is added, bromotyrosine at levels of 0.1% and 0.2% was recovered from the TFA-GA and GA respectively.
72. Examples 2 and 3 relate to the production of HBr/AcOH in a glass lined reactor and using a phenol scavenger.
73. In Example 4 the level of the bromotyrosine in eight different batches of GA was measured. Five batches (A to E) were made using the so called "old method" using commercially sourced HBr/AcOH. The remaining three (X, Y and Z) were made by the "new method". The "new method" is the method of producing HBr/AcOH in examples 2 and 3. The "old method" used HBr bought from external suppliers. GA made by the "old method" resulted in batches with between 0.14 and 0.32% bromotyrosine. GA made by the "new method" did not contain any detectable bromotyrosine. In opening Synthon suggested it might advance a case based on some inference to be drawn from the numbers in the table in Example 4 and the molar ratios of amino acids in glatiramer acetate but by the closing speeches whatever the point had been, it was not pursued.
74. The second aspect of the patents relates to the level of metal ions. Paragraphs [0094] - [0099] explain that although pre-filled syringes of Copaxone were generally clear, a red colour was detected after the solutions were kept at room temperature for 12-24 hrs and the source of this colour was unknown. Exposure to metal led to metal impurities in the HBr. These metal impurities could then chelate with the GA to form coloured complexes.
75. The patents explain that by taking precautions, such as using a glass lined reactor to produce HBr/AcOH, the formation of metal impurities can be avoided. In addition, the patents explain that by avoiding bringing the HBr/AcOH into contact with metal surfaces, for example by using Teflon-lined piping, the formation of trace metal ions can be prevented. Eliminating the formation of metal ions, prevented the formation of red coloured GA.
76. The patents also state (at paragraph [0098]) that solutions of HBr/AcOH with a colour of below 2000 APHA led to GA without a red colour. The American Public Health Association (APHA) colour index is a yellowness index. Paragraph [0115] describes how that measurement is made. The APHA scale is defined such that each APHA unit is based on the dilution of a 500ppm stock solution of a yellow platinum-cobalt compound. The index is such that each APHA unit is generated by the equivalent of 1 mg of platinum per litre of solution, thus an APHA 20 standard solution contains 20 ppm of platinum.
77. Example 5 describes judging the colour of HBr/AcOH solution by visual comparison of the solutions with standard solutions from the APHA colour index. A table containing the APHA values for five batches of HBr/acetic acid that were manufactured using non-metal apparatus is provided (at paragraph [0118]). The APHA values for the five batches are between <300 and 700 APHA. The specification states (at paragraph [0119]) that the colour of the five batches indicated that the batches were free of bromine and free of metal ion impurities.

Claim construction

78. The law is not in dispute. The two key cases are Kirin-Amgen Inc v Hoechst Marion Roussel [2004] UKHL 46 and Virgin Atlantic Airways Ltd v Premium Aircraft Interiors UK Ltd [2009] EWCA Civ 1062.
79. The claims can be grouped into convenient categories in different ways. Most of the claims are to processes. Those processes are either for making glatiramer acetate or the TFA-glatiramer acetate intermediate. There are claims to the glatiramer acetate product itself (of the i.v. claims, those are claims 22-26 of the 924 patent).
80. Claim 1 of 528 is to a process for making TFA-glatiramer acetate in which the glutamic acid deprotection step involves using HBr/AcOH with a defined level of metal ions (less than 1000 ppm) and a defined level of free bromine (less than 0.5%). Claims 2 and 3 of 528 depend on claim 1 and reduce the permitted levels of metal ion and free bromine. The claims include a cascade of alternative levels defined in numerical terms ending with "free of [metal ion impurities/free bromine]". In this context "free" would be understood by the skilled reader to be undetectable and at least less than the lowest numerical limit. This reasoning applies to claims 2 and 3 of 528 and to claim 7 of 924 (not itself independently valid but i.v. claims depend on it).
81. Claim 6 of 528 depends on i.v. claims 1, 2 or 3 and requires a further step of pretreating the HBr/AcOH with a bromine scavenger to remove free bromine.
82. Claim 10 of 528 depends on i.v. claims 1, 2, 3 or 6 and defines the colour of HBr/AcOH by reference to a cascade of values on the APHA scale (starting at less than 2000 down to less than 500).
83. Claim 1 of 924 is to a process for making TFA-glatiramer acetate in which the glutamic acid deprotection step involves using HBr/AcOH with a defined level of free bromine of less than 0.1%. Thus it differs from claim 1 of 528 in two ways. There is no limit on the metal ion concentration in the HBr/AcOH and the free bromine limit is lower.
84. Claims 8, 11, 12, 13 and 16 of 924 depend on claim 1 and import now familiar limits relating to the HBr/AcOH in the process, i.e. cascading limits on metal ion concentration (8 - 12) and colour by APHA (13, 16). In amongst the dependencies of these claims are claims with the same scope as claims 1, 2, 3 and 10 of 528.
85. Claims 22 to 26 of 924 depend ultimately on either claim 19 or 20 (via claim 21) but Teva only relies on claims 22-26 insofar as they depend in claim 20. The broadest i.v. claim relied on in this set is claim 22 which therefore amounts to a claim to a mixture of glatiramer acetate with a desired molecular weight and less than 100ppm metal ion impurities. A point on "desired molecular weight" is best dealt with in the added matter section. The narrower claims 23-26 consist of another cascade of metal ion levels down to "free of metal ion impurities" in claim 26. Again "free" must be at least less than 10 ppm given the claim structure.
86. Finally claim 29 of 924 is directed to making a pharmaceutical composition containing glatiramer acetate which has a percentage of bromotyrosine lower than 0.3%. Any synthetic method will do. The key steps in the claim require testing a batch of glatiramer acetate and measuring the percentage of bromotyrosine by hydrolysing the sample and performing chromatography of the hydrolysate against sample solutions. The sample solutions are of the relevant amino acids and bromotyrosine. Only batches which pass the 0.3% threshold are included.

Novelty:

87. To be valid an invention must be new, which means it must not form part of the state of the art (s1(1)(a) and s2(1) of the 1977 Act, Art 54 EPC). The state of the art includes all matter made available to the public before the priority date (s2(2) of the 1977 Act).
88. The House of Lords in Synthon BV v SmithKline Beecham plc [2005] UKHL 59, [2006] RPC 10, held that for an item of prior art to deprive a claim of novelty, two requirements must be satisfied: disclosure and enablement. First, the prior art must disclose subject matter which, if performed, would necessarily infringe that claim. As it was said in General Tire and Rubber Co v Firestone Tyre and Rubber Co Ltd [1972] RPC 457 at 486, "[t]he prior inventor must be shown to have planted his flag at the precise destination before the patentee" (per Sachs LJ). The second requirement identified in Synthon is that the prior art must disclose that subject matter sufficiently to enable the skilled addressee to perform it.
89. There is no issue about enablement in this case, the question arising over the Lemmon prior art is about the first limb of Synthon. The issue is whether the prior art would fall within the claims. The test is a strict one, as the flag planting metaphor employed by Sachs LJ was intended to indicate. The test is one of necessity and inevitability. If a prior document leaves a choice open for the skilled person and if the result only falls within the patent claim if the skilled person adopts one way forward and not the other, then there is no lack of novelty. In that circumstance evidence that a skilled person "would" do something when faced with that choice is evidence relevant to obviousness, not novelty. The claim may lack inventive step but it has not been anticipated. On the other hand patentees will sometimes argue that a choice exists when in fact there is no genuine choice and in fact the patented way forward really is inevitable. If those are the facts then the claim lacks novelty but that is not because the skilled person had to make a choice, it is because there really was no choice at all. Fanciful supposed choices do not count.

Novelty – prior sales of Copaxone

90. Sales of Copaxone before the priority date are admitted. In order to demonstrate lack of novelty of claims 22-26 Synthon needs to establish that the products had a metal ion impurity level within the claims (from less than 100 ppm in claim 22 to "free" in claim 26).
91. Teva provided disclosure relating to the composition of pre-priority Copaxone. In closing (paragraph 64) Synthon accepted that it was not possible from the documents to assess the metal ion content of any batch of pre-priority Copaxone whether in terms of ppm of metal ion impurities or APHA scale colour.
92. There is evidence that the pre-priority material was made to a specified level of heavy metals, and the numerical limits for those are lower than the numbers in some claims but the claims are not directed only to heavy metals, they relate to all metals. There are many metals which are not heavy metals.
93. Synthon asked Teva to admit that prior art Copaxone was made using HBr/AcOH produced in a non-metallic (glass lined or Teflon lined) reactor. Teva admitted glass lined reactors were used on some occasions. Synthon took this to be an admission that the solution had been prepared on those occasions with entirely non-metallic apparatus and so Synthon inferred that the prior art Copaxone must have been free of metal ion impurities since the only expedient taught in the patent to prevent metal ion impurities is to use non-metallic apparatus. Teva pointed out that its admission did not go that far because it was not asked to make any admission about apparatus generally and because there could be other sources of metals. Teva is right. The patent itself mentions using non-metallic vessels and non-metallic tubing in addition to vessels. Moreover metals could come into the process at other stages, e.g. in the second deprotection step with 1M piperidine in water.
94. Synthon has not established that pre-priority date Copaxone fell within any of the relevant claims of 924. Nor do I accept that there is sufficient evidence from which to infer such a conclusion which might shift an evidential onus onto Teva to disprove it. There is simply no direct evidence as to what the metal ion level of Copaxone was and the other evidence and admissions made by Teva do not allow one to draw any kind of inference about what the level of metal ion impurity was.
95. I find that the product claims are novel over prior sales of Copaxone.

Novelty – Lemmon

96. The relevant teaching is Example 4 of Lemmon. The synthetic scheme disclosed in Example 4 has been summarised above. Claim 1 of 528 has two apparent differences over Lemmon, the HBr/AcOH has a metal ion concentration of less than 1000ppm and a free bromine concentration of less than 0.5%. Claim 1 of 924 has one difference over Lemmon, the level of free bromine in the HBr/AcOH must be below 0.1%.
97. Synthon advanced a number of submissions about what a skilled person would do faced with Lemmon. They would be aware of the possible presence of free bromine in HBr/AcOH and that it had the potential to react with tyrosine and so they would ensure they used HBr/AcOH free of bromine. That would be achieved either by discarding coloured HBr/AcOH in favour of colourless HBr/AcOH because they knew that the colour was due to free bromine or by using a scavenger like phenol. Also separately the skilled person would use materials which were as pure as possible in any event. In my judgment even if these propositions were accepted in their entirety, they would not establish that the skilled person had no choice to make in the matter. The points fall to be considered as issues relating to inventive step. They do not establish lack of novelty. As a matter of disclosure, Lemmon does not teach the avoidance of a reaction between bromine and tyrosine, nor does it teach the use of HBr/AcOH free of free bromine, nor does it teach the use of a scavenger. I find that irrespective of the arguments about common general knowledge, the skilled person given Lemmon would not necessarily seek to avoid a reaction between free bromine and tyrosine, would not necessarily use colourless HBr/AcOH (even assuming that solution is free of free bromine) and would not necessarily use a scavenger. At best Synthon's arguments show only that the skilled person had choices. That is not enough.
98. This finding of novelty applies to all i.v. claims which involve a level of free bromine or bromotyrosine (that is all i.v. claims of 528 as well as claims 1, 8, 11 - 16 and 29 of 924). That leaves product claims 22-26 of 924 which only involve a limit on metal ions, not bromine. Claim 29 involves actually measuring bromotyrosine using a bromotyrosine standard solution and is plainly novel over Lemmon on any view.
99. There is no express teaching in Lemmon to produce the HBr/AcOH in a glass lined vessel or use glass lined apparatus. Synthon submitted that no-one would use metallic vessels to produce HBr/AcOH and so since that is the only source of metal ion impurities identified in the patent and since the patent teaches the use of non-metallic equipment to solve the problem, the claims must lack novelty.
100. The specification sheets for HBr/AcOH indicate it is highly corrosive to metals. In my judgment it is more likely than not that a skilled person would only ever use a non-metallic vessel to produce HBr/AcOH. There was evidence about a non-reactive metal called Hastelloy but it was speculative. When asked about some evidence given in the Netherlands Prof Davis explained he did not know about the industrial production of HBr. The idea that the skilled person would have to make a choice about using a non-metallic vessel is fanciful.
101. However Prof Davis gave evidence, which I have already mentioned above in the context of Copaxone, that there are other potential sources of metal ions in the overall process such as the piperidine and process water used in the last deprotection step and also in the amino acid NCAs and dioxane. It is also at least possible that metal equipment such as a metal cannula would be used to transfer materials in a lab (Lemmon is on a lab scale) and that metal parts like valves may be used if one is considering an industrial scale.
102. There is no teaching in Lemmon to avoid potential sources of metal altogether and I am not satisfied it would be inevitable or necessary to do so. I find that glatiramer acetate produced following Lemmon would not necessarily have metal ion impurities at a level required to satisfy the broadest relevant claim, claim 22.
103. Thus I reject the allegation of lack of novelty over Lemmon. Whether or not this result or the result over Copaxone leads to an insufficiency squeeze is something I will address below.

Inventive step

104. To be valid an invention must involve an inventive step, which means it must not be obvious to a skilled person having regard to the state of the art (s1(1)(b) and s3 of the 1977 Act, Art 56 EPC). The structured approach to the assessment of obviousness was set out by the Court of Appeal in Pozzoli v BDMO [2007] EWCA Civ 588. In Conor v Angiotech [2008] UKHL 49 the House of Lords considered the issue of obviousness. There Lord Hoffmann (with whom the others of their Lordships agreed) approved the following statement of Kitchin J made in Generics v Lundbeck [2007] RPC 32:

"The question of obviousness must be considered on the facts of each case. The court must consider the weight to be attached to any particular factor in the light of all the relevant circumstances. These may include such matters as the motive to find a solution to the problem the patent addresses, the number and extent of the possible avenues of research, the effort involved in pursuing them and the expectation of success."

105. The skilled person and the common general knowledge have been identified above. In terms of differences and inventive concept I will start by focussing on the claim features based on levels of bromine and use of a bromine scavenger. Second I will consider levels of metal ions, and APHA colour of the HBr/AcOH. Teva submitted that there was an interaction between these issues based on paragraph [0098] of the patent and on Prof Tabor's acceptance that bromotyrosine / metal ion complexes could be the cause of the colour. After dealing with these issues separately and if it is necessary to do so I will consider whether the points need to be looked at together.

Inventive step - levels of bromine and use of a bromine scavenger

106. The starting point is Lemmon. The difference between Lemmon and claim 1 of 528 is the level of free bromine in the HBr/AcOH being less than 0.5%. There is also a feature about metal ions in the HBr/AcOH but for now I am considering bromine. The only difference between claim 1 of 924 and Lemmon is less than 0.1% free bromine in the HBr/AcOH.
107. Synthon's case on obviousness developed (Teva would say shifted) over time. The main points advanced by Prof Tabor in her report were that it was obvious to get rid of or avoid free bromine in the HBr/AcOH, for two reasons. One was because it was obvious that there was a risk that free bromine in the HBr/AcOH might react with the tyrosine. The other reason was that it was also obvious that there was a need for a scavenger such as phenol to take care of benzyl cations which could arise in the deprotection step and that such a scavenger would remove free bromine as well. Prof Davis did not agree with either proposition.
108. I reject both points. The first argument founders on the common general knowledge I have already addressed. Absent hindsight, nothing in the skilled person's common general knowledge and nothing in Lemmon itself would alert them to the idea that there was a risk of bromination of tyrosine as a result of using the HBr/AcOH used in example 4. That is so despite the rather higher temperature and longer time proposed in order to carry out the cleavage as well as deprotection. Even if their HBr/AcOH was coloured to some extent the skilled person would not foresee any risk caused by an interaction with tyrosine. The skilled person would not regard themselves as using a contaminated reagent. Prof Davis mentioned that he would be prepared to use a 15 year old Winchester (bottle) of HBr/AcOH. I understood him to be putting this forward as an eloquent indication of his attitude to the robustness of the material rather than a particular statement focussed on 15 years. I accept Prof Davis' opinion that the skilled person would just not be concerned with HBr/AcOH.
109. The benzyl cation point was answered by Prof Davis in his reply report, which Prof Tabor accepted in cross-examination. Benzyl cations might be a risk with HBr in TFA and an S_N1 mechanism but not with HBr/AcOH which went by S_N2. I reject the benzyl cation argument as a reason why the skilled person would use a scavenger when performing Example 4 of Lemmon.
110. In opening Synthon added a third approach. This was that the skilled person would routinely perform an amino acid analysis to characterise their product having followed Lemmon and that analysis would reveal the presence of some bromotyrosine. Bromotyrosine is obviously not what the skilled person wanted to make and they would readily identify HBr/AcOH as the source (since it is the only source of bromine atoms). They would realise free bromine in the HBr/AcOH must be brominating the tyrosine at the deprotection and cleavage stage. It would be obvious to remove the bromine and an obvious expedient would be a scavenger like phenol. This approach was put to Prof Davis in cross-examination. He did not accept it but Synthon submits his reasons are not compelling. Synthon also submits that Prof Tabor supported this line of attack. I will address this below.
111. Finally in closing Synthon added a further way of putting the case based on Agrevo/Triazoles [T939/92] submitting that given the way Teva were arguing the case there was in truth no technical advance here. Lack of plausibility was also mentioned. Teva had not identified what was inventive about the contents of the patent and the patent did not do so either. For example the patent does not express surprise in finding bromotyrosine in the glatiramer acetate. Synthon argued that all this invention amounted to was using a pure reagent in a known reaction (pure, bromine free and metal ion free HBr/AcOH) and this in law could not support a patent (EPO Board of Appeal Novartis/Erythro-compounds T990/96).
112. I do not accept the submission from Synthon that there is a legal principle that using pure reagents in a manufacturing process is never patentable. These questions depend on all the facts and circumstances. T990/96 was about a product claim to a single chemical compound defined by its formula in which the novelty was said to reside in a high level of purity. The "general rule" identified by the Board (paragraph 7) was that a document disclosing a low molecular weight compound and its manufacture makes available to the public in the sense of Art 54 EPC that compound at all grades of purity as desired by the person skilled in the art. That general rule is not relevant to the Teva process claims at all. As for the product claims, the ones in this case are defined by reference to a level of an identified class of impurities rather than to a general purity level. I also note that the Board expressly referred to low molecular weight compounds, by which I understand them to be confining themselves to what are sometimes called small molecules in contra-distinction to larger biological molecules like the polypeptides this case is concerned with. All the same if the point was a point of law, the distinction does not seem a good one. Finally the Board also recognised the possibility of exceptions albeit it did describe them as extraordinary situations. I will not grapple with that.
113. Teva objected to the lack of technical advance/plausibility argument, saying it was not pleaded and that one could not therefore criticise the patentee for not calling evidence directed to the question of whether eliminating bromotyrosine or the red colour from Copaxone was a technical advance. Synthon maintained it was covered by the generality of the obviousness pleading. In my judgment the argument is not open to Synthon in the pleadings. This sort of argument is routinely pleaded expressly in proceedings before the Patents Court. An argument of this kind can be and must be expressly pleaded if it is to be argued. It is no hardship to the party challenging validity to require them to do so.
114. In addition however I am wholly unpersuaded by this manner of arguing Synthon's case on its merits in any event. The patent makes it clear that the inventors regarded the steps described to prevent unwanted bromotyrosine from being present in the final product and to eliminate the red colour as improvements to the manufacturing process. This is a process used to manufacture what on the face of the patent is a valuable therapeutic agent, valuable in the sense of having utility in treating what is unquestionably an awful disease. Improvements in processes for manufacturing such things are precisely the sorts of technical advances with which patents can be concerned. Ensuring that an unwanted product such as bromotyrosine is not formed at all so that it cannot end up in the final pharmaceutical composition is plainly a technically worthwhile thing to do regardless of whether bromotyrosine is known to have any relevant biological activity. So too is preventing the formation of a coloured form. A colour change is an indication that something has changed in the pharmaceutical product. Whether it has adverse impact in the sense that it represents either a material reduction in potency or any increase in toxicity is not the issue. Those skilled in the art will not want it to happen in any case.
115. Finally, a word about the patent specification. I do not accept that much can be made of the fact that the patent does not expressly assert that this or that element in the disclosure was surprising or difficult. Such language appears sometimes in patents but it can be a hostage to fortune when it turns out that prior art, which perhaps was unknown to the inventors, shows that something was not so surprising after all. When that occurs it may or may not lead to the invalidity of a given claim but an approach which made invalidity more likely because one element disclosed in the specification in the section describing the invention was not flagged as unexpected would create unwelcome incentives to those drafting the documents. They will get even longer. The primary task of the specification is to explain the invention so that the claims are supported and can be understood and so that the skilled person can perform the invention to the standard required by the law. That is quite enough.

Amino acid analysis

116. I will now turn to Synthon's third approach. By closing a major aspect of Synthon's case was that the skilled person would find bromotyrosine by conventional amino acid analysis and once they had found it, they would use a scavenger (etc.) in the HBr/AcOH. I accept that if a skilled person had found a significant level of bromotyrosine in the products of Example 4 of Lemmon and wanted to prevent its formation, then the most likely culprit would be bromine which must therefore have been present in the HBr/AcOH. It would be obvious to deal with that by reducing the level of bromine in HBr/AcOH to whatever level had the desired effect. One obvious way to achieve this would be to add a scavenger like phenol. The skilled person could not say for sure that this would work without testing but in the relevant circumstances it would be obvious to try with a solid degree of confidence that it was likely to work. The materials and arguments relied on by Synthon to establish that this approach was common general knowledge did not do so, but they do demonstrate in my judgment that by a process of obvious chemical reasoning a skilled person presented with the task of preventing the formation of bromotyrosine would, without inventive step, think of removing free bromine from HBr/AcOH and using phenol etc. as the solution to that problem.
117. The issue is whether this process of reasoning would start at all.
118. Synthon submitted that the skilled person would perform amino acid analysis and this analysis, using routine conditions, would indicate the presence of something extra, i.e. something other than the four amino acid components of glatiramer acetate. That something extra would be bromotyrosine. The patent itself showed that conventional conditions were all that was needed to isolate bromotyrosine using standard chromatography (such as HPLC) of the amino acid analysis mixture made after hydrolysis of the polypeptide under test. Instead of seeing four peaks representing the four amino acids of glatiramer acetate, when the skilled person made glatiramer acetate using the Lemmon process with HBr/AcOH containing free bromine, a fifth peak would be seen. It would be bromotyrosine. Synthon submitted that Prof Tabor's evidence supported these propositions.
119. Teva submitted that this analysis was laced with hindsight. First, Teva characterised this issue as whether the skilled person would "stumble across" bromotyrosine. The point is that Synthon's case is and has to be that a skilled person not looking for bromotyrosine or aware of it as a species which might be present, would nevertheless encounter evidence of its existence, follow it up and find it. Second they emphasised Prof Davis' view which was that the analytical techniques which would be used by a skilled person were NMR, SEC and/or MALDI-TOF. Those techniques would not highlight the presence of bromotyrosine in a way which would allow the skilled person to stumble across it. So, argued Teva, focussing on amino acid analysis involves hindsight. Prof Tabor's evidence in cross-examination was that it would have been impossible to detect bromotyrosine in glatiramer acetate using those three methods. I therefore find that the three techniques, employed by a skilled person analysing the product of Example 4 but not looking for bromotyrosine, would not draw the skilled person's attention to bromotyrosine present in glatiramer acetate made by following that example. Third Teva submitted that Prof Davis' opinion was that there was no evidence a skilled person carrying out amino acid analysis but not looking for bromotyrosine would stumble across it. The fact that conditions exist whereby bromotyrosine can be isolated by chromatography when the skilled person is looking for it does not demonstrate what a skilled person would do if they were not looking for it. Fourth Teva submitted that most of the evidence of Prof Tabor which Synthon relied on did not support the proposition advanced. Fifth Teva contended that the one aspect of Prof Tabor's evidence which did support the issue was given in cross-examination but the reasons she gave for her view were not convincing and in fact supported the points made by Prof Davis.
120. Turning to Prof Tabor's evidence, Teva is right that the evidence in the Professor's report is not directed to the relevant issue. It is directed to the question of whether a skilled person looking for bromotyrosine would be able to find it by amino acid analysis. I am sure they would. It does not show however that a skilled person performing a conventional amino acid analysis of the product of Lemmon simply to analyse the results with no preconceptions about bromotyrosine would happen to choose any particular conditions which would reveal it.
121. In cross-examination [Day 2/213 et seq] Prof Tabor explained her opinion that the skilled person would perform amino acid analysis because they would want to measure the ratios of amino acids and would then find five peaks and this would alert them that there was a problem regardless of whether or not you had reasons to suspect there was bromotyrosine there or not. When asked how Prof Tabor knew the peaks would resolve by HPLC, she referred to three matters. First the levels of bromotyrosine were not negligible, second a paper which she had found the previous week (it turned out to be by Wu et al (Biochemistry 1999, 38, 3538-3548)) and third, the fact this had been done in the patent itself (Example 1). It was put to the Professor that all the paper and the patent showed was that you can set up the HPLC analysis to find it if you know or suspect its presence. Prof Tabor accepted it would be easier in that case but said it was not impossible otherwise and referred to Fig 6 of the primer as representative of a case in which baseline resolution of 20 amino acids is shown. Her view was that bromotyrosine would run at a different place in that chart. When pressed about the difference between picking it up when not looking for it and setting things up when you know what you are looking for, Prof Tabor maintained that the presence of a shoulder on a peak would alert the skilled person to the presence of an unusual or unnatural amino acid. In other words Prof Tabor was not maintaining that five separate peaks would be produced but was expressing her opinion that bromotyrosine present in the glatiramer acetate would produce at least a shoulder on one of the expected four peaks and that would act as the alert.
122. Prof Davis did not accept that bromotyrosine would emerge unlooked for as suggested by Synthon. The chromatography experiment in the patent was put to him. At the beginning of the cross-examination on this point the Professor's evidence was unconvincing. He appeared to be suggesting that there was something inherently inventive (a "discovery") about the conditions described in Example 1 whereby bromotyrosine is measured. His evidence about how the experiment is performed in detail based on spiking was confused.
123. However after the Wu paper was put to him and a page from Wikipedia about HPLC, Prof Davis' evidence became clearer. The Wu paper was an example of workers measuring bromotyrosine when they knew it was there. It contains two different sets of chromatography conditions to detect bromotyrosine depending on whether the sample in which the material is being measured (the analyte) is of one kind or another. This showed that even looking for the same thing in different analytes might require different conditions. I should record that Wu is plainly not common general knowledge.
124. It was put to Prof Davis that the conditions shown as a typical example of HPLC conditions in Wikipedia are very similar to those in the patent. The Professor did not agree and I accept his evidence. First I am not prepared to place weight on a recent printout from Wikipedia on a contentious issue of this kind in any case. Second, the fact that both mention acetonitrile and water does not establish that the conditions in the patent would be chosen, without hindsight, by the skilled person. The conditions stated in the patent are sparse but they are not the same as the ones in Wikipedia. The example given in the latter is 5% acetonitrile in water with a linear gradient over time up to 95% acetonitrile whereas the former amounts to acetonitrile:water:acetic acid in a ratio 95:4:1 using an isocratic gradient (i.e. an unvarying ratio).
125. The only conditions shown to work in the evidence are conditions which have been chosen knowing that one is looking to isolate bromotyrosine. It is plainly possible to select conditions to reveal bromotyrosine if one is looking for it. The patent demonstrates that. Indeed I infer from the sparse nature of the patent disclosure that it is not very difficult once you know what you are looking for. However no convincing evidence was called by Synthon to establish with any specificity what conditions a skilled person would choose when running a chromatography experiment following amino acid analysis of glatiramer acetate made based on Lemmon but not looking for bromotyrosine. The conditions would be chosen based on the fact that the analyte is hydrolysed glatiramer acetate. Fig 6 of the primer was not intended to analyse glatiramer acetate. I infer that the skilled person has essentially an open ended range of conditions to choose from. I bear in mind that this is a pharmaceutical product and no doubt would be analysed carefully but even assuming that the level of bromotyrosine is not negligible, I am not satisfied that its presence would reliably emerge either at all or even as a shoulder prominent enough on another peak to be of sufficient interest to be worth following up. I find that a skilled person carrying out Lemmon and performing amino acid analysis afterwards would not be alerted to the presence of bromotyrosine.
126. In conclusion I find that the bromine level claims (and claim 29) are not obvious over Lemmon. This is for one main and one subsidiary reason. The main reason is that I am not satisfied amino acid analysis conducted after performing Lemmon would alert the skilled person to the presence of bromotyrosine. The subsidiary but not irrelevant reason is that focussing on amino acid analysis itself introduces an element of hindsight into the analysis given the three other analytical techniques available.

Inventive step - metal ion impurities

127. The difference between the glatiramer acetate claimed by claim 22 and Lemmon is the level of metal ion impurities (< 100 ppm) in the glatiramer acetate. Synthon submitted that these claims were obvious as follows. According to the Physicians Desk Reference Copaxone can be kept for a week at room temperature outside a fridge. As the patent explains the problem emerged because some Copaxone kept at room temperature turned red in colour after 12-24 hours. Synthon submitted that if this happened to material produced following Lemmon the skilled person would consider metal ion impurities to be a possible cause. Their presence can be readily established by standard analytical techniques such as ICP-MS which was a technique suggested by Prof Davis. Once metal ion impurities are detected it would be entirely obvious to use non-metallic apparatus to produce the HBr/AcOH since that solution is highly corrosive to metals and would be an obvious likely source of metal ion impurities. That would in fact eliminate the source of metals identified in the patent and from the patent one can take it that this would solve the problem.
128. Synthon's submission is compelling up to a point but in my judgment it fails on the evidence of both experts. The subtlety in Synthon's formulation of its case is that it avoids facing up to Prof Davis' opinion that the standard practice of a skilled person faced with a problem of colouration would be to identify the cause of the colouration and then, having identified the cause, employ a solution based on that knowledge. Prof Tabor agreed with Prof Davis on this point in cross examination. This is fatal to Synthon's case because it is entirely clear on the evidence that identifying the cause of the red colour is challenging. Prof Davis said that identifying the cause may not necessarily be possible and that the required research would by its nature involve skill, judgment and serendipity. Teva produced two versions of a report of work done to identify the cause of the colour. Prof Davis reviewed them and explained that they reinforced his view. Teva contends that the whole report is confidential (that is for another day). It was subject to a hearsay notice and I read both versions. The report is entitled "Pink/red Glatiramer Acetate". It shows the two individuals (I infer scientists) working to determine the cause of the red colour. They investigated a number of hypotheses at least some of which seem to me to be entirely unrelated to issues in this case. Various stages of the synthesis are studied in some depth. A wide range of analytical techniques are used. The nature of the conclusions can be understood from this passage, which is a fair summary:

"Unfortunately, the appearance of the red colour in final product could not be unequivocally related [nor] to certain component[s] used in the synthesis, neither to experimental conditions of the synthesis."

(My typographical corrections added in square brackets.)

129. I find that faced with a red coloured glatiramer acetate product, the skilled person would identify metal ions as one possible cause but it would be one among a number of others. The others mentioned by Prof Davis were a peptide degradation product, a bromine atom containing species (which could generate Br₂) and an impurity which leached from the syringe. Each of these, if one was the cause, would require different resolution.
130. The skilled person would set out to identify the likely cause of the colour. They might well measure the metal ion level as part of that investigation. However to take Synthon's approach and jump to the conclusion that metal ions should be eliminated is not the activity of the uninventive skilled person. I am not satisfied the skilled person would just try things without having identified the cause of the colour. I am not satisfied that even a substantial research project over a number of months would identify that metal ions are the or a likely cause. If having failed to pin down the cause a skilled person then changed tack and started just trying out different expedients, alighting upon elimination of metal ions may or may not be one of the expedients to test. However either way I reject the idea that that sort of exercise would be indicative of obviousness. It is not.
131. Finally, just as in the novelty section, I am not satisfied that using non-metallic vessels would necessarily eliminate the colour (because there may be metal ions from other sources) nor am I satisfied that it would have been obvious to go as far as to eliminate all metal equipment altogether unless the skilled person was motivated to do so by a finding that metal ions were indeed likely to be the cause of the problem. It follows that the step of using non-metallic vessels would not inevitably solve the problem. If it did not then assuming the skilled person had not identified the cause of the problem when doing what Synthon contend for, they are no further forward. As before whether this has an impact on sufficiency is a matter I will address below.

Inventive step - conclusion

132. I conclude that all the claims of both patents involve an inventive step. Since I have reached that conclusion considering the two aspects separately, there is no need to consider whether the two elements interact.

Insufficiency

133. A patent specification must disclose the invention clearly and completely enough for it to be performed by a person skilled in the art (s72(1)(c) Patents Act 1977, Art 83 EPC). There are various kinds of insufficiency. In Sandvik v Kennametal [2011] EWHC 3311 (Pat) at paragraphs 106 to 124 Arnold J identified three kinds of insufficiency as follows: First, where the skilled person is unable to carry out the claimed invention either at all or without undue burden and without needing inventive skill ("classical insufficiency"). Secondly, where the claim is ambiguous so that the skilled person cannot tell when he is within the claim or outside it. Thirdly, where the breadth of the claim exceeds the technical contribution to the art made by the invention.
134. Proving insufficiency usually requires positive evidence but it is possible in some cases to advance so called squeeze arguments aimed at putting the patentee on the horns of a dilemma. This approach may mitigate the need for evidence focussed directly on insufficiency because the argument takes as a premise findings of fact made in another context such as novelty or inventive step. That is Synthon's approach here.
135. Synthon submitted that since the patent taught that the only source of metal ion contamination was metal apparatus and the only solution was not using metal apparatus, it followed that a finding that the claims are novel or inventive over Lemmon because the glatiramer acetate made using HBr/AcOH in non-metallic apparatus might still contain metal ion impurities from other sources must necessarily expose a lack of sufficient disclosure.
136. The difficulty I have with this submission is as follows. The patent discloses that metal ion impurities were chelated by glatiramer acetate and that the metal ion/glatiramer acetate complexes contributed to the colour (paragraph [0096]). It also teaches that using non-metallic apparatus to produce the HBr/AcOH led to a solution free of metal ions and no red glatiramer acetate ([0097]). It contains claims to glatiramer acetate free of metal ion impurities and with metal ion impurities below specified levels. Take claim 22 with a metal ion impurity level of less than 100 ppm. When a skilled person puts the patent's teaching into practice they will no doubt use non-metallic apparatus. Leaving aside consideration of free bromine (which is not really a legitimate thing to do but I will do for present purposes) this activity by the skilled person is the same as carrying out Lemmon with non-metallic apparatus. Given the claim and the specification, the skilled person will expect to see no red glatiramer acetate and a product with less than 100 ppm metal ion impurities. They will measure the metal ion level in the product and look at its colour. Perhaps they will find exactly what the patent leads them to expect. However I have found as a fact that putting Lemmon into practice using non-metallic apparatus will not inevitably have the result in claim 22. Thus by the same token (and again leaving aside free bromine) it follows that putting the patent's express teaching to use non-metallic apparatus in practice will not inevitably produce metal ion levels claimed in claim 22 or colourless glatiramer acetate. There may be metal ions present because they were introduced as an impurity in the 1M piperidine in water used in the second deprotection step.
137. The question is what would a skilled person do if, having taken this course and followed the express teaching of the patent, they nevertheless found they had coloured glatiramer acetate or metal ions above 100 ppm.
138. I doubt a skilled person would have much difficulty with this. Once you know having read the patent that metal ions are a problem, before you throw up your hands and give up, you would take steps to see if metal ions had been introduced at other stages or try taking steps to remove them. But I do not have to make any findings about that because Synthon called no evidence addressing this issue either in chief from Prof Tabor and or in cross-examination with Prof Davis. Synthon has tried to construct the argument based on equating the position of the skilled person following Lemmon with the position of the skilled person reading the patent but they are not the same. Lemmon makes no mention of a required level of metal ion impurities and no mention of colour for which metal ion impurities are a possible cause. Crucially the patent does contain such a teaching. Claim 22 would only be insufficient if a skilled person could not make glatiramer acetate with less than 100 ppm metal ion impurities without undue burden. To establish that in this case requires evidence. There is none.
139. At one stage there appeared to be a further insufficiency argument advanced relating to the APHA colour test but it was not pursued.

Added matter

140. No amendment to a patent or a patent application is permitted if it adds matter as compared to that disclosed in the original application (s76 of the 1977 Act, Art 123(2) EPC). The basic approach to be followed is that explained by Aldous J as he then was in Bonzel v Intervention [1991] RPC 553. Added matter has been considered by the Court of Appeal in Vector v Glatt [2007] EWCA Civ 805 and AP Racing v Alcon [2014] EWCA Civ 40.
141. Synthon contended that claims 5, 20, 27 and 28 of 924 are invalid for added matter. The reason why these claims add matter is said to be because the 924 application does not disclose:

i) glatiramer acetate having a "desired" molecular weight;

ii) glatiramer acetate containing less than 1000 ppm metal ion impurities; or

iii) mixture of TFA-glatiramer acetate or glatiramer acetate with a colour of less than 1000 APHA.

142. In closing Synthon clarified its case and confirmed that the attack on claim 5 was not pursued.
143. Synthon also contended that the added matter attack applied to claims dependent on the claims the subject of the allegation, i.e. to claims 21-28 which depend on claim 20 and to claim 32 which depends on claims 27 and 28.
144. Claim 20 of 924 is focussed on glatiramer acetate. It claims a mixture with a desired molecular weight and with less than 1000 ppm metal ion impurities. Claims 27 and 28 depend ultimately on either claim 19 or 20. Claims 27 and 28 are to mixtures characterised by a colour measured on the APHA scale (less than 1000, 700 or 500). If the claim depends on claim 20 then the mixture is a mixture of glatiramer acetate. If the claim depends on claim 19 then the mixture is a mixture of TFA-glatiramer acetate. Claim 32 is essentially to the use of TFA-glatiramer acetate made in accordance with claims 19, 21-28 or 30 in the manufacture of glatiramer acetate.

Glatiramer acetate having a "desired molecular weight"

145. Synthon submits that the application as filed discloses various mixtures of what are referred to simply as "polypeptides" or "TFA-polypeptides" which may or may not be glatiramer acetate or TFA-glatiramer acetate. The only polypeptide mixtures stated to be glatiramer acetate are disclosed at p6 ln22-p7ln8 and p7 ln10-31. However the molecular weight of these mixtures is not given, nor is it stated whether the molecular weight (or average molecular weight) is "desired" nor are metal ion impurity levels stated for these mixtures, nor colour. Synthon argues that the phrase "desired molecular weight" does not appear in the application. The phrase "desired average molecular weight" does appear but none of this is enough to support claim 20, which refers to a mixture of glatiramer acetate having a desired molecular weight (notably glatiramer acetate not TFA-glatiramer acetate and not a "polypeptide").
146. In my judgment there is nothing arising from the distinction drawn by Synthon between the references to polypeptides as opposed to glatiramer acetate. Reading the application as a whole, it is plainly directed to making glatiramer acetate which is a mixture of polypeptides (see e.g. the Background section at p1 ln 16 et seq). The skilled reader would understand that the manufacturing process which the document explicitly states it is concerned with improving (at p3 ln 20) is the manufacturing process for glatiramer acetate. Synthon is right that in the "Summary of the invention" section from p4 to 9 many of the paragraphs use the term "polypeptide" whereas others refer to glatiramer acetate but the reader would not attribute any major significance to this. The passages at p6 and 7 of the application mentioned by Synthon above are within that section. Then there is a "Detailed Description of the Invention" section which repeats some of the Summary section and consists of a large number of further paragraphs. The application does not contain any claims but these paragraphs would be understood by the reader in the same way. (To someone versed in patent practice they might be identified as statements of invention / consistory clauses). Finally there are three sections which are to all intents and purposes the same as the corresponding parts of the granted patent: a "Terms" section defining terms from p20, a Discussion section from p23 which deals with free bromine and metal impurities (c.f. granted paragraphs [0090] – [0099]), and then Examples 1 to 5. These sections clearly relate to glatiramer acetate.
147. The references to a mixture of polypeptides in the application would be understood by the reader to cover the possibility that the mixture was not glatiramer acetate (perhaps it could have a very different molar ratio) but the reader would understand the whole disclosure to be applicable to making glatiramer acetate as the paradigm case.
148. The term "average molecular weight" is defined at p20 ln 24 and appears and is used e.g. in the Detailed Description section in clauses related to TFA-polypeptides. Synthon is correct that the phrase "desired molecular weight" does not appear in the application. The phrase "desired average molecular weight" is used both for TFA-polypeptides (e.g. p4 ln 8) and polypeptides in their final form (p5 ln26-p6 ln20). As I have said, the reader would regard all this as applicable to glatiramer acetate or TFA-glatiramer acetate as the paradigm case of what is under discussion.
149. I do not believe there is added matter in the difference between desired molecular weight and desired average molecular weight. No new information is taught to the skilled reader by the former phrase as compared to the latter. I suppose in theory (there was no evidence dealing with this to which my attention was drawn) the phrase without the word "average" might be apt to cover something more (but what?) however the legally significant point is that if no new information is conveyed, as I find to be the case, then there is no added matter even if the coverage is broader (see AP Racing v Alcon).

Metal ion impurity levels

150. The point on metal ion impurity levels is as follows. Synthon argues that the 924 Application discloses TFA polypeptides containing less than 1000 ppm metal ion impurities and processes involving the use of a solution of HBr/AcOH containing less than 1000 ppm. However, it is not disclosed that the glatiramer acetate obtained from a mixture of TFA-glatiramer acetate which itself contained less than 1000 ppm metal ion impurities results in a glatiramer acetate product containing less than 1000 ppm metal ion impurities. Nor is it disclosed that glatiramer acetate made in a process using HBr/AcOH which contained less than 1000 ppm metal ion impurities results in a glatiramer acetate product containing less than that level of metal ions.
151. Teva's response is that the application contains a clear disclosure of the need to avoid metal ion impurities. I agree. The disclosure (application pages 24-25) is the same as paragraphs [0094] - [0099] as granted which I have addressed above (using non-metallic apparatus for HBr/AcOH, chelating and glatiramer acetate / metal ion complexes etc.). Teva points to the disclosure of using HBr/AcOH with less than 1000 ppm metal ions and a cascade of lower levels and to the disclosure of TFA- polypeptides with less than 1000 ppm metal ions and the same cascade of values.
152. Teva then argues that although the application does not specifically list out glatiramer acetate with the same decreasing cascade of metal ion impurity levels as it does for HBr/AcOH and TFA polypeptides, in the light of the general overarching disclosure to avoid metal ion impurities in glatiramer acetate and the disclosure of HBr/AcOH and TFA polypeptides with low metal ion levels, it would be clearly disclosed to the skilled person that they should have the corresponding low levels of metal ions in their final glatiramer acetate product in order to avoid the colouration problem.
153. Teva also submits that this argument was raised in third party observations under EPC Art 115 by Synthon before grant but the EPO examiner allowed the claims all the same. It argues that if anything the EPO are seen as being stricter on added matter than the national courts (possibly sometimes too strict) and so that is an indication there is no problem here.
154. I am concerned by the third party observations point at least to some extent. When the court is considering something like added matter after grant, necessarily the examiner must have allowed the claim but usually one cannot say whether the point was drawn to their attention. The position is different when, as Teva submit, the very point has been formally put to the examiner. I must say I also agree that as a generalisation the approach which has in the past been taken to added matter objections in the EPO has often appeared quite strict, perhaps unduly so on some occasions.
155. However in the end the court is not bound by the examiner's decision. The fact is that the application does not contain any statement that the level of metal ions in the final glatiramer acetate product is necessarily the same as the level of metal ions in TFA-glatiramer acetate intermediate. Nor does it teach that if the metal ion level in TFA-glatiramer acetate is less than a given amount (e.g. 1000 ppm) it follows that the metal ion level in glatiramer acetate will also be less than that same given amount (in the example 1000 ppm).
156. The arguments on novelty, inventive step and insufficiency have a bearing on this. Teva submitted and I have accepted that metal ion impurities could be introduced after TFA-glatiramer acetate has been formed, at the second deprotection step when the TFA-blocked material is converted to glatiramer acetate. There could be metal ions in the piperidine or the water. That fact illustrates why it does not follow to the skilled reader of the application that the metal ion level in glatiramer acetate will always be the same as the level in TFA-glatiramer acetate (or in HBr/AcOH). The application teaches that metal complexes in TFA-glatiramer acetate and glatiramer acetate are the cause of the colour but I do not see any basis from which the skilled reader would take it that the levels at one stage necessarily followed from the levels at another stage.
157. To avoid added matter there must be a clear and unambiguous disclosure. That is English law as much as the law in the EPO. In my judgment a description of glatiramer acetate itself with a specified metal ion concentration presented in the granted patent is new information to the skilled reader as compared to the information in the application. Thus claim 20 adds matter and in so far as claims 21-26 depend on claim 20, they are referring to metal ion concentration in glatiramer acetate (as opposed to TFA-glatiramer acetate) and add matter too.

APHA colour levels of the glatiramer acetate mixture

158. As to colour levels of the glatiramer acetate mixture, Synthon argues that the 924 Application does not disclose glatiramer acetate or TFA glatiramer acetate as having a specific APHA colour value. The only disclosure of APHA values relates to HBr/AcOH. Teva argues that the avoidance of metal ions would avoid colouration and so the claims do not add matter. I accept that the application (like the patent) teaches that the red colour was eliminated. There is also a statement linking colour of the HBr/AcOH and the glatiramer acetate in both documents, as follows: "HBr solution with a color below 2000 APHA was shown to produce glatiramer acetate without red color." (patent paragraph [0098]). This simply does not specify the colour of the glatiramer acetate by the APHA scale either expressly or by implication. If anything the contrast between the two colours is clear. One is measured by a scale and the other is determined by a qualitative expression - red.
159. In my judgment claims 27 and 28 disclose new information as compared to the application and are invalid for added matter insofar as they depend on claim 20.

The impact of conclusions on added matter

160. Subject to any further submissions, claim 20 will need to be deleted but claims 21-26 would remain but have their dependencies amended to reflect the removal of claim 20. They would remain dependent on claim 19 and they would relate only to TFA-glatiramer acetate and to metal ion levels in that material. There is no added matter there. Equally claims 27 and 28 would also need to be amended to reflect the removal of claim 20 and to remain dependent ultimately only on claim 19. And claim 32 would need to have its dependencies amended to reflect that, leaving the claim dependent only on claim 19 and claims 21-28.

The Dutch decision

161. On 22nd April 2015 the Hague District Court decided to revoke the Dutch designations of the 528 and 924 patents. As here, the proceedings in the Netherlands were between Synthon and Teva. In the Netherlands Teva did not defend claims 20-28 or 31 of 924 and so no question of lack of novelty had to be decided (decision paragraph 5.2.1) and presumably no added matter was raised. As before me, Lemmon was taken as the relevant prior art (5.3.1) and the skilled person was the same (5.3.2). Free bromine and metal ion concentration were taken as distinct partial problems for the purposes of inventive step following a problem/solution approach, with no synergistic effect between them (5.3.1). The claim features based on non-metal apparatus, metal ion levels and colour values were held not to contribute to inventive step because the skilled person would use non-metal apparatus anyway and thereby unavoidably solve the metal ion problem without inventive step (paragraphs 5.3.4 to 5.3.9). The claim features based on bromine were not inventive because the problem to be solved was how to modify the process to prevent contamination by bromotyrosine and the solution was obvious. The only possible source of bromine in the process was HBr/AcOH, it was obvious that free bromine in that mixture must be brominating the tyrosine and therefore using bromine free HBr/AcOH or employing a scavenger was obvious (5.3.10-5.3.21). Given that the Dutch court has reached a different conclusion from me on the same legal issues, I will address the decision in more detail.
162. As regards metal ions: the Dutch court held that the skilled person would prepare HBr/AcOH in glass lined apparatus because that material would corrode metal apparatus and the skilled person would have in mind that HBr/AcOH prepared in such a way would be contaminated by such corrosion (paragraph 5.3.4). The evidence of Teva's expert Dr Bille was said to be that not using metal apparatus was industry standard. Avoiding the red colour was a bonus effect of the application of common general knowledge and not inventive. The only solution posed by the patent was to use non-metal apparatus. The "mere insight" that metal ion concentration in HBr/AcOH can be critical cannot render the claims inventive because it would allow the claim to monopolise the application of the state of the art. Teva's argument that further measures might be necessary to lower metal ion concentration was contrary to the patent's teaching.
163. As for free bromine: starting from Lemmon Example 4, Synthon contended the problem to be solved was how to modify the process to prevent contamination by bromotyrosine. The Dutch court rejected Teva's submission that Synthon's problem inappropriately contains pointers to the solution. That was because the skilled person would come across the contamination problem based on an FDA report by a Dr Jessop which seems to have related to toxicity studies of Bromo-Copolymer-1 (i.e. brominated glatiramer acetate) in rats and which seems to have taught that levels of bromotyrosine should be kept as low as possible (5.3.13). An argument from Teva that this document was not available to the public at the priority date in 2004 was raised too late and not accepted. This FDA report is not in evidence before me. The Dutch court held that if the skilled person wanted to prevent contamination by bromotyrosine, using a scavenger like phenol was obvious.
164. Thus I believe the different conclusions I have reached from those of the Dutch court derive from key differences in the evidence before each court. On bromine, the Dutch court had the FDA report of Dr Jessop whereas I did not and I have had to consider a different issue about amino acid analysis. On metal ions, I have evidence accepted by both side's experts that the skilled person would want to establish the likely cause of a problem before they moved to try out solutions. That is not how the matter was approached in the Netherlands.

Conclusion

165. I reject the attack on the validity of all claims save that claims 20, 27 and 28 of the 924 patent are invalid for added matter. Subject to further submissions, claim 20 will need to be deleted and the dependencies of claims 21-28 and 32 adjusted accordingly.

Annex 1 – the claims in issue

The 528 Patent

1. A process for obtaining a mixture of trifluoroacetyl glatiramer acetate, wherein during the process a batch of a mixture of polypeptides, each of which consists of alanine, ?-benzyl glutamate, tyrosine and trifluoroacetyl lysine is deprotected with a solution of hydrobromic acid in acetic acid, the improvement comprising use of a solution of hydrobromic acid in acetic acid, which solution comprises less than 1000ppm of metal ions, and less than 0.5% of free bromine.

2. The process of claim 1, wherein the solution of hydrobromic acid in acetic acid comprises less than 0.1% of free bromine, less than 0.05% of free bromine, less than 0.01% of free bromine, less than 0.001% of free bromine or is free of free bromine.

3. The process of claim 1 or 2, wherein the solution of hydrobromic acid in acetic acid comprises less than 500ppm of metal ion impurities, less than 100ppm of metal ion impurities, less than 10ppm of metal ion impurities, or is free of metal ion impurities.

6. The process of any one of claims 1-5, further comprising a step of pretreating the solution with a bromine scavenger in order to remove free bromine.

10. The process of any one of claims 1-9, wherein the colour of the hydrobromic acid in acetic acid solution is less than 2000 APHA, less than 100 APHA, less than 700 APHA, or less than 500 APHA.

The 924 Patent

1. In a process for obtaining a mixture of trifluoroacetyl glatiramer acetate, wherein the mixture has a desired average molecular weight and wherein during the process a batch of a mixture of polypeptides, each of which consists of alanine, ?-benzyl glutamate, tyrosine and trifluorolysine is deprotected with a solution of hydrobromic acid in acetic acid, the improvement comprising use of a solution of hydrobromic acid in acetic acid, which solution comprises less than 0.1% free bromine.

8. The process of claim 7, wherein the solution of hydrobromic acid in acetic acid comprises less than 100 ppm of metal ion impurities.

11. The process of claim 10, wherein the solution of hydrobromic acid in acetic acid comprises less than 10 ppm of metal ion impurities.

12. The process of claim 11, wherein the solution of hydrobromic acid in acetic acid is free of metal ion impurities.

13. The process of any one of claims 1-12, wherein the colour of the solution of hydrobromic acid in acetic acid is less than 2000 APHA.

16. The process of claim 15, wherein the colour of the solution of hydrobromic acid in acetic acid is less than 500 APHA.

22. The mixture of claim 21, where in the mixture comprises less than 100 ppm metal ion impurities.

23. The mixture of claim 22, where in the mixture comprises less than 30 ppm metal ion impurities.

24. The mixture of claim 23, where in the mixture comprises less than 20 ppm metal ion impurities.

25. The mixture of claim 24, where in the mixture comprises less than 10 ppm metal ion impurities.

26. The mixture of claim 25, where in the mixture is free of metal ion impurities.

29 A process for preparing a pharmaceutical composition containing a mixture of glatiramer acetate, wherein the mixture has a predetermined percentage of brominated tyrosine acceptable for inclusion in a pharmaceutical composition, which comprises:

obtaining a batch of glatiramer acetate

measuring the percentage of brominated tyrosine of the batch by a process comprising:

a) hydrolysing the batch to obtain a hydrolysate

b) eluting the hydrolysate through a chromatographic column

c) measuring the level of bromotyrosine in the hydrolysate;

d) preparing sample solutions of the amino acid components of the batch and bromotyrosine

e) eluting the sample solutions through the column of step b); and

f) calculating the percentage of brominated tyrosine in the batch; and

including in the pharmaceutical composition a batch only if its percentage of brominated tyrosine so measured is less than 0.3%.

Claims of the 924 patent which are not themselves alleged to be independently valid but from which independently valid claims depend.

Claim 7 (on which claim 8 depends):

The process of any one of claims 1-6, wherein the solution of hydrobromic acid in acetic acid comprises less than 0.05% of free bromine, preferably less than 0.01% of free bromine, more preferably less than 0.001% of free bromine, yet more preferably the solution of hydrobromic acid in acetic acid is free of free bromine.

Claims 9 and 10 (on which claim 11 depends):

9. The process of claim 8, wherein the solution of hydrobromic acid in acetic acid comprises less than 30 ppm of metal ion impurities.

10. The process of claim 9, wherein the solution of hydrobromic acid in acetic acid comprises less than 20 ppm of metal ion impurities.

Claims 14 and 15 (on which claim 16 depends):

14. The process of claim 13, wherein the colour of the solution of hydrobromic acid in acetic acid is less than 1000 APHA.

15. The process of claim 14, wherein the colour of the solution of hydrobromic acid in acetic acid is less than 700 APHA.

Claims 20 and 21 (on which claim 22 depends and Teva relies):

20. A mixture of glatiramer acetate wherein the mixture has a desired molecular weight and less than 1000 ppm metal ion impurities.

21. The mixture of any one of claims 19-20, wherein the mixture comprises less than 500 ppm metal ion impurities.

Claims 27, 28 and 32 (which relate to the added matter issue):

27. The mixture of any one of claims 19-26, wherein the colour of the mixture is less than 1000 APHA, preferably less than 700 APHA.

28. The mixture of claim 28, wherein the colour of the mixture is less than 500 APHA.

32. Use of the mixture of trifluoroacetyl glatiramer acetate any one of claims 19 or 21-28 or the trifluoroacetyl glatiramer acetate of claim 30 in the manufacture of glatiramer acetate.