BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

England and Wales High Court (Patents Court) Decisions


You are here: BAILII >> Databases >> England and Wales High Court (Patents Court) Decisions >> Actavis UK Ltd v Novartis AG [2009] EWHC 41 (Ch) (16 January 2009)
URL: http://www.bailii.org/ew/cases/EWHC/Patents/2009/41.html
Cite as: [2009] EWHC 41 (Ch)

[New search] [Context] [View without highlighting] [Printable RTF version] [Help]


Neutral Citation Number: [2009] EWHC 41 (Ch)
Case No: HC07CO 1753

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
PATENTS COURT

Royal Courts of Justice
Strand, London, WC2A 2LL
16/01/2009

B e f o r e :

MR JUSTICE WARREN
____________________

Between:
ACTAVIS UK LTD
Claimant
- and -

NOVARTIS AG
Defendant

____________________

Mr Roger Wyand QC & Mr Piers Acland (instructed by Bird & Bird LLP) for the Claimant
Mr Richard Arnold QC & Mr Mark Chacksfield (instructed by Bristows) for the Defendant

Hearing dates: 16th, 17th, 18th, 21st ,22nd, 23rd, 24th, and 25th July 2008

____________________

HTML VERSION OF JUDGMENT
____________________

Crown Copyright ©

    Mr Justice Warren :

    Introduction

  1. In these proceedings, the Claimant ("Actavis") seeks to revoke European Patent (UK) No. 0948320 ("the Patent") on the grounds of obviousness and insufficiency. The Defendant ("Novartis") does not defend claims 1 and 10 as granted and has applied to amend. In this judgment, references to the claims are to the proposed amended claims unless otherwise stated. Mr Wyand QC and Mr Acland appear for Actavis; Mr Arnold QC (now Arnold J, but whom I will refer to throughout as Mr Arnold) and Mr Chacksfield appear for Novartis.
  2. There is also a pleaded claim by Actavis that claim 10 as proposed to be amended is invalid because it is a method of treatment. Actavis does not pursue this pleaded claim in the light of the decision of the Court of Appeal in Actavis v Merck. Novartis nonetheless asks me to rule on the pleaded claim on the hypothesis that the appeal from the Court of Appeal to the House of Lords is successful. I will come to that request in due course.
  3. Novartis has counterclaimed for infringement. It is accepted by Actavis that its product is within the scope of at least one of the claims as proposed to be amended. Accordingly, the issue of infringement stands or falls with the issue of validity.
  4. The amendment to claim 1 is objected to on grounds of added matter and lack of clarity.
  5. Background

  6. The Patent is concerned with a drug called fluvastatin, a member of a class of drugs known as HMG-CoA reductase inhibitors, commonly called statins, which have the effect of reducing blood cholesterol. They act primarily in the liver to competitively and reversibly inhibit HMG-CoA reductase, an enzyme that catalyses a rate-limiting step in cholesterol biosynthesis. They thereby reduce the blood plasma concentration of low density lipoprotein (LDL)-cholesterol, the deposition of which causes atherosclerosis and hence coronary heart disease. Such drugs are widely prescribed for the prevention and treatment of hypercholesterolemia ie high cholesterol. The first member of this class of drugs to be marketed (in 1987) was the naturally-occurring substance lovastatin. Later two semi-synthetic analogues, simvastatin and pravastatin, were marketed (in 1989 and 1991). And later fluvastatin was marketed (in 1994).
  7. Fluvastatin was the first fully synthetic analogue; it was patented in the early 1980s. That patent has now expired. A supplementary protection certificate in respect of the patent covering fluvastatin expired on 22 August 2008. The compound received marketing approval for use in humans at the end of 1993 and was first marketed in 1994. At the priority date of this Patent (which, as is common ground, is 8 October 1996), the drug was available in the form of capsules containing 20 mg or 40 mg of fluvastatin sodium (a water-soluble salt of the active ingredient). As from 22 August 2008, there has been nothing to prevent Actavis from launching a conventional, immediate release, formulation of fluvastatin, that is to say immediate release capsules.
  8. Evening administration was recommended because cholesterol synthesis peaks at night. Dosage was adjusted at 4 week intervals because the maximum reductions in LDL concentration were generally only seen after 2 - 4 weeks. Fluvastatin sodium did not show any enhanced effect when the same overall daily dose was given twice daily rather than once daily. Fluvastatin sodium was well tolerated. The most frequently reported side effects were minor and transient gastrointestinal problems.
  9. Since the launch of fluvastatin, other fully synthetic analogues have been marketed. Fluvastatin sodium is marketed by Novartis under the trade mark LESCOL. It is a successful drug, but not the most successful statin, which is currently atorvastatin. All statins currently available, including fluvastatin, are sold in the form of conventional (immediate release) formulations. Novartis also markets fluvastatin sodium in a sustained release formulation under the trade mark LESCOL XL, which has proved successful. No other statin has ever been marketed in a sustained release formulation.
  10. Precisely what is meant by a sustained release formulation is something which I will need to deal with in more detail later; but in essence, as the name suggests, it is a formulation that is designed to release active ingredient in a sustained manner over a period of time. It is this formulation of fluvastatin which the Patent seeks to monopolise.
  11. The Patent

  12. Paragraph [0001] of the Patent describes the invention as relating to sustained release pharmaceutical compositions comprising a water soluble salt of fluvastatin as an active ingredient. The claimed sustained release compositions are matrix formulations or diffusion-controlled membrane coated formulations and combinations thereof.
  13. Paragraphs [0002]-[0008] of the Patent describe sustained release compositions.
  14. Paragraph [0002] of the Patent is in the following terms:
  15. "In recent years there has been a large increase in the development and use of sustained-release tablets which are designed to release the drug slowly after ingestion. With these types of dosage forms, the clinical utility of drugs can be improved by means of improved therapeutic effects, reduced incidence of adverse effects and simplified dosing regimens."

    The existence and clinical utility of sustained release formulations is thus acknowledged in a general way although, no doubt, the improvement (if any) in the clinical utility of any particular drug will depend critically on the characteristics and uses of that drug.

  16. Paragraph [0003] (which must be read with paragraph [0002]) explains the way in which the Patent is using the concept "sustained release":
  17. "A sustained-release tablet releases the drug during several hours, typically more than 3 hours and less than 30. Other commonly used terms such as "controlled release", "extended release", "prolonged release", etc., all comply with the definition of a product that releases the drug typically over more than 3 hours."

  18. There are (and have long been) a number of different types of sustained release formulation. The claims of the Patent mention two of them – "matrix formulations" and "diffusion-controlled membrane coated formulations." These were perfectly conventional at the priority date of the Patent. One can see this reflected in paragraph [0004] at sub-paragraphs (i), (ii), and (iii).
  19. Paragraphs [0007] and [0010] refer to prior disclosure of sustained release or time-controlled release formulations of HMG-CoA inhibitors (also known as statins):
  20. "[EP-A-0465096] discloses sustained release tablet formulations of HMG-CoA inhibitors" [That is the correct patent reference: paragraphs [0006] and [0007] mix up this patent and a US patent]
    "The use of some HMG-CoA reductase inhibitors for the preparation of a medicament adapted for time-controlled release administration is disclosed in EP-B-0 375 156"
  21. The Patent returns to water solubility in paragraph [0008], where it explains that "the drug release from sustained release formulations is related to the drug solubility" and therefore "a special challenge is met when trying to formulate water soluble substances for sustained release formulations…..Another possibility is to use a less water soluble salt. However, such a change requires more extensive development work and may also lead to bioavailability problems due to incomplete dissolution".
  22. In paragraph [0009], reference is made to Hypercholesterolemia, HMG-CoA reductase and the four approved statins, all of which were clearly common general knowledge.
  23. The formula for fluvastatin is set out in paragraph [0011]. Paragraph [0012] then states:
  24. "Fluvastatin is a water soluble drug. For example, the solubility of the sodium salt of fluvastatin in water extends to more than 50 g/l. Biopharmaceutical requirements of a sustained release product of this water soluble drug would then at first sight impose formulation problems, as discussed above. Thus, with a diffusion controlled release device for this soluble substance, e.g. an insoluble matrix of a polymer, fast release rates can be expected due to the high solubility of fluvastatin creating high concentration gradients as the driving force for diffusion out of the matrix."
  25. The invention is summarised in paragraphs [0017]-[0018], which state:
  26. [0017] It has surprisingly been found that sustained release compositions, comprising fluvastatin as a water soluble salt, exhibit particularly favourable release characteristics such as unexpectedly long duration and slow rate of drug release. In the present context, the term "water soluble" should be understood as a solubility of more than 30 mg/ml in water at +37oC.
    [0018] Consequently, the present invention provides a pharmaceutical composition for sustained release comprising a water soluble salt, preferably the sodium salt, of fluvastatin as an active ingredient. The sustained release fluvastatin compositions for which these favourable properties are obtained are selected from the group comprising matrix formulations, diffusion-controlled membrane coated formulations; and combinations thereof.

  27. Different excipients used to make the formulation of the invention are described in paragraphs [0019] to [0024], all of which were common general knowledge.
  28. Therapeutic effect is mentioned in paragraphs [27] to [29]. The formulations are "useful for lowering the blood cholesterol level in animals, in particular mammals e.g. humans. They are therefore useful as hypercholesterolemic and anti-atherosclerotic agents"; and "Consequently, the invention provides in another aspect the use of fluvastatin for the manufacture of a pharmaceutical composition for sustained release, for the treatment of hypercholesterolemia". A method of treatment is also disclosed.
  29. Paragraph [31] describes the dosing: generally, dosage will be "in the range 1 to 1000 mg of fluvastatin per day preferably 2 to 200 mg/day".
  30. Paragraphs [0032]-[0049] describe five examples of the invention. Paragraph [0034] states that in examples 1 to 3, drug release was determined by use of an USP II apparatus.
  31. In the first four examples, the release profile of fluvastatin sodium is compared with methylparaben and/or diclofenac in terms of release rates from three matrix formulations and diffusion across a polymer membrane. Methylparaben and diclofenac are said to have lower solubilities in water (~ 2 mg/ml and ~ 5 mg/ml respectively compared with > 50mg/ml for fluvastatin). The fifth example comprises the manufacture of sustained release formulations of fluvastatin sodium. Paragraph [0033] states:
  32. "Methylparaben and diclofenac sodium could be expected to exhibit a somewhat slower release rate and longer duration of release, due to the lower water solubility. However, unexpectedly, the release of fluvastatin was consistently slower than methylparaben and diclofenac sodium for all the tested types of sustained release formulations."

  33. The other experiments are also described as showing unexpectedly favourable extended release properties or an unexpectedly slow release rate. Mr Arnold submits that the results in Examples 1-4 (the data are found in Figures 1-4) are, indeed, on their face surprising, and need explanation. In relation to Examples 1 – 3, he says that the skilled person would know that the transport rate or flux of a drug is usually proportional to its solubility, and inversely proportional to the third root of its molecular weight. As such, his expectation would be that the measured flux of fluvastatin sodium would be approximately 10 times that of diclofenac sodium, and 20 times that of methylparaben. Dr Rue accepted that some of these results would have been seen as a little surprising by the skilled formulator in 1996.
  34. As to Example 4, the skilled person, according to Mr Arnold relying on Professor Collett, would expect the flux of a drug to increase in proportion to its concentration, as is seen for diclofenac sodium. Yet this is not what is found, indicating that fluvastatin sodium exhibits surprising properties in the sustained release formulations of the invention.
  35. The claims in issue

  36. Other than claims 1 and 10, Novartis relies on claims 2 and 3 as being independently valid. Accordingly it necessary to consider only claims 1, 2, 3 and 10.
  37. Claim 1 may be broken down as follows:
  38. [A] A sustained release pharmaceutical composition comprising

    [B] a water soluble salt of fluvastatin as active ingredient and

    [C] being selected from the group consisting of matrix formulations, diffusion-controlled membrane coated formulations and combinations thereof,

    [D] wherein the sustained formulation releases the active ingredient over more than 3 hours.

  39. The pharmaceutical composition of claim 1 does not need to confer any therapeutic effect. The only functional limitation is that the fluvastatin is released over more than three hours. It is common ground that the test is carried out in accordance with the USP Paddle method (at pH 6.8 and 37°C). It does not matter how much or how little (provided that it is not de minimis) of the active ingredient is released after 3 hours.
  40. Claim 2 is as follows:
  41. A pharmaceutical composition according to claim 1 wherein the said water soluble salt of fluvastatin is the sodium salt.

  42. Claim 3 is as follows:
  43. A pharmaceutical composition according to claim 1 or 2 which is an eroding matrix formulation.

  44. Claim 10 may be broken down as follows:
  45. [A] The use of a water soluble salt of fluvastatin for the manufacture of,

    [B] a sustained release pharmaceutical composition for the treatment of hypercholesterolemia,

    [C] said sustained release pharmaceutical composition being selected from the group consisting of matrix formulations, diffusion-controlled membrane formulations and combinations thereof,

    [D] wherein the sustained release formulation releases the active ingredient over more than 3 hours.

  46. Mr Wyand makes a number of submissions about what is not in the Patent. Since they will be relevant later on, I include what he says here:
  47. a. The Patent does not refer to the therapeutic benefits of a sustained release composition containing fluvastatin.
    b. If there was a prejudice against formulating high solubility drugs into sustained release compositions (which is not accepted), the Patent does nothing to overcome that prejudice.

    c. There is no suggestion that pH-dependent solubility is a problem Likewise there is no teaching as to what happens as the tablet goes down the GI tract and reaches pH 8.

    d. The Patent does not suggest any other reason for doubting that a sustained release formulation could be made. If there was any such prejudice (whether concerned with first-pass metabolism, half-life, stability, partition co-efficient or absence of concentration effect) (which is not accepted), the Patent does nothing to overcome that prejudice.

    The Experts

  48. Reports were adduced from five experts. Actavis' experts were Dr Peter Rue, a formulation chemist and Professor Paul Durrington, a clinician specialising in the treatment of hypercholesterolemia. Novartis' experts were Professor Barrie Bycroft, a medical chemist, Professor John Collett, a formulation chemist and Professor Yves Horsmans, a clinical pharmacologist. Professor Horsmans was a comparative late-comer to the fray but, to cut a long story short, there is in the event no objection to his introduction notwithstanding directions for trial which envisaged only two experts from each side.
  49. Professor Bycroft was not cross-examined. According to Mr Wyand, the evidence which he gave is irrelevant to the issues in the action. If, however, he is wrong about that (as Mr Arnold says he is) the professor's evidence is unchallenged.
  50. So far as concerns the four witnesses who were cross-examined, I have to say that they were all excellent witnesses. Not only were they all eminently qualified to give the evidence that they gave, but they gave their evidence in an impressive and authoritative manner. Mr Wyand remarks correctly that, in the case of Professor Collett, his written evidence included clinical topics which were plainly outside of his expertise but he readily accepted in cross-examination that on these matters he would turn to the clinicians on the team. Professor Collett did not seek to give me the impression that he knew more than he actually did.
  51. There is a huge amount of common ground between the experts in each discipline. This is not a case where I have to choose between experts on each side who hold widely differing opinions on fundamental issues. Their opinions do differ in some respects which are of material significance. I will need to consider their evidence on an issue by issue basis; it is impossible to say that one expert is so obviously to be preferred to another that I should necessarily accept the evidence of the one and reject the evidence of the other on all issues.
  52. I found Professor Durrington to be a particularly impressive witness. He was particularly helpful in his attempts always to answer questions as at the priority date in October 1996. He did his best to consider the issues through the eyes of the average clinician working in the area of hypercholesterolemia but his expertise is so great that I detect that he may have found that a difficult exercise to perform. Whether his starting point in terms of the notional task which he, as a member of a hypothetical team, would have been undertaking was the correct one is an issue which I will need to consider later. Mr Arnold submits that Professor Durrington started from the wrong place and has therefore come up with the wrong answers.
  53. Professor Horsmans was also an impressive witness. Although English is not his first language, I do not think that this caused him any disadvantage; indeed, he had a most engaging and fluent style. He was perhaps slightly more discursive than Professor Durrington but he was clearly doing his best to assist the Court.
  54. Dr Rue was also a first-class witness. He has clearly approached his responsibilities with the utmost care and objectivity.
  55. Finally, Professor Collett was also an impressive individual. That is so, not only in the sense of leaving a real impression of his personality behind, but also in displaying the breadth of his knowledge and experience. However, he had more difficulty than the other witnesses in sticking to the questions asked. He was also clearly a man of strong views. It may simply be the expression of those views where they favour Novartis' case that left me with the feeling that if any of the experts was at all partisan it was Professor Collett. Mr Wyand levels some specific criticism at him, set out in his closing submissions. I do not propose to set out the details, but I have formed my own view about these criticisms which I take into account in my assessment of his evidence overall.
  56. Prior Art

  57. The pleaded prior art comprises (1) the March 1996 edition of the British National Formulary (which it is common ground is part of the common general knowledge); (2) US Patent 4,739,073 ("Kathawala"); and (3) PCT Application WO 095/06470 ("Scolnick"). The pleaded prior art does not include the patent application or the patents referred to in the Patent and there is no suggestion that these were common general knowledge. The prior pleaded art does not include papers, such as McClelland et al and Cheng et al which were deployed in expert evidence and cross-examination.
  58. The British National Formulary

  59. The British National Formulary ("BNF") is a bi-annual publication referred to by medical practitioners in the United Kingdom when prescribing drugs to patients. The March 1996 edition contains an entry for fluvastatin. It is a short entry which I set out in full.
  60. "Indications: primary hypercholesterolaemia (hyperlipidaemia type IIa) in patients with cholesterol concentrations of 6.5 mmol/litre or greater who have not responded adequately to dietary control
    Cautions; contra-indications; Side-effects: see under Simvastatin; Interactions: Appendix 1 (fluvastatin)
    Dose: initially 20mg daily in the evening; usual range 20-40 mg daily in the evening, adjusted at intervals of 4 weeks; up to 40 mg twice daily may be required."

    Lescol (Sandoz) is shown as the available product, sold in packs of 20 mg or 40 mg capsules. The cautions and contra-indications under simvastatin read as follows:

    "Cautions: monitor liver function before and during treatment; history of liver disease (avoid if active);….advise patients to report muscle pain…
    Contra-indications: active liver disease; pregnancy (toxicity in animal studies) and breast-feeding; porphyria.."

    Kathawala

  61. Kathawala discloses a class of compounds for use as inhibitors of cholesterol biosynthesis and lowering blood cholesterol levels and therefore in the treatment of hyperlipoproteinemia and atherosclerosis (ie hyper-cholesterolaemia). The specification includes the following:
  62. "the compounds of Formula 1 are competitive inhibitors of [HMG-CoA] reductase, the rate limiting enzyme in cholesterol biosynthesis and, therefore, they are inhibitors of cholesterol biosynthesis. Consequently, they are useful for lowering the blood cholesterol level in animals e.g., mammals, especially larger primates, and, therefore, as hyperlipoproteinemic and anti-atherosclerotic agents….."

  63. Kathawala includes three tests which demonstrate the biological activity of various compounds of Formula I. The first two tests (A and B) are in vitro and the third (C) is in vivo. The first (A) is designed to measure the ability of the tested compounds to inhibit HMG-CoA reductase. The second and third tests (B and C) are designed to measure ability of the tested compounds to inhibit cholesterol biosynthesis. The methods used are described and the results are tabulated. One of the test compounds shown to perform well in the three assays is Example 8. This is fluvastatin sodium.
  64. Kathawala describes the way in which the various compounds of the invention can be formulated into pharmaceutical compositions. That description includes the following:
  65. "The daily dosage is usually divided into two or four equal portions or administered in sustained release form. A typical oral dosage of the compound of Example 8 is indicated to be 1 mg. three times a day. Usually, a small dosage is administered initially, and the dosage is gradually increased until the optimal dosage for the host under treatment is determined….."

    Scolnick

  66. Scolnick relates to an invention for the prevention and treatment of Alzheimer's disease. It relates to the administration to humans of an HMG-CoA reductase inhibitor, expressly referring to Lovastatin, Simvastatin, Pravastatin and Fluvastatin "to lower Apolipoprotein E isoform 4 (ApoE isoform 4) levels in the central nervous system to treat, arrest the development of and prevent the onset of Alzheimer's disease". The named compounds are referred to as "known cholesterol lowering agents".
  67. Mr Wyand summarised the theory as explained in this way. I gratefully adopt his explanation. The presence of a protein called apoE isoform 4 in the brain had been linked to an increased risk of developing Alzheimer's disease. Scolnick postulates that the administration of a statin will cause an up-regulation of the LDL receptor, thereby lowering the concentration of apoE-containing lipoproteins in the bloodstream. This in turn would cause an increase in the flow of apoE isoform 4 from the central nervous system to the bloodstream.
  68. Thus, although the objective of the treatment is to lower apoE isoform 4 in the brain, this is achieved by lowering LDL in the bloodstream.
  69. Scolnick discloses that the statin (in this case, structural formula (I) ) may be administered in a variety of ways (orally, topically, parenterally, by inhalation spray or rectally). It is stated that it is usually desirable to use the oral route. In relation to the oral administration ("in the form of a capsule, a tablet or the like") it is stated as follows:
  70. "The orally administered medicament may be administered in the form of a time-controlled release vehicle, including diffusion controlled systems, osmotic devices, dissolution controlled matrices and erodible/degradable matrices. Doses may be varied, depending on the age, severity, body weight and other conditions of human patient, but daily dosage for adults is within a range of from about 1 mg to 1000 mg (preferably 5 mg to 100 mg) which may be given in a single dose or in two to four divided doses…"
  71. Claim 10 is a claim to a method "wherein 5 to 100 mg of the HMG-CoA reductase inhibitor is administered by a controlled release dosage form".
  72. The issues

  73. The following issues fall to be decided:
  74. Issue 1: Obviousness over the common general knowledge including the British National Formulary

    Issue 2: Obviousness over Kathawala

    Issue 3: Obviousness over Scolnick

    Issue 4: Insufficiency: it is said that the Patent (as proposed to be amended) does not disclose the alleged invention clearly enough and completely enough for it to be performed by a person skilled in the art because the claims all require the making of a "sustained release" composition; the specification does not teach and the skilled person would not know the criteria according to which it can be assessed whether or not a pharmaceutical composition is "sustained release" within the meaning of the claims.

    Issue 5: Amendment of the Patent/Added Matter

  75. As already mentioned, Actavis does not rely upon the method of treatment objection to claim 10 but wishes to preserve its right to raise the matter in a higher court if necessary. The objection does not raise any issue of fact. Novartis invites me, nonetheless, to give a ruling on this objection on the basis that my decision in Actavis v Merck was correct and the Court of Appeal wrong. That would be a curious course for me to adopt given that there are no factual findings which I need to make to enable the method of treatment objection to claim 10 to be dealt with should the House of Lords reverse the Court of Appeal. My disinclination to accept Novartis' invitation is reinforced by this consideration: unless and until the House of Lords actually does overrule the Court of Appeal, it cannot be known precisely why the Court of Appeal is wrong if it is wrong. It is not therefore clear on what hypothetical basis I should proceed. I think it best to say nothing more about this objection.
  76. Cholesterol

  77. Cholesterol is an essential component of all animal cell membranes. Most of the cholesterol in the body is synthesised in the liver although some is derived from dietary sources and some is synthesised in peripheral cells. A high level of cholesterol (hypercholesterolaemia) is associated with an increased risk of coronary heart disease and stroke.
  78. Cholesterol is a lipid and therefore poorly soluble in aqueous media such as the blood. In order to transport cholesterol around the body, it is packaged with other types of lipid (triglycerides and phospholipids) and proteins in the form of globular particles called lipoproteins. These come in varying sizes and compositions and are classified according to their density into four groups: chylomicrons, very low density lipoprotein (VLDL), low density lipoprotein (LDL) and high density lipoprotein (HDL).
  79. Each type of lipoprotein has a different role in the transport of lipids. For example LDL is the main carrier of cholesterol from the liver to the peripheral tissues and accounts for around 60-70% of the cholesterol in the blood. Most hypercholesterolaemia is due to an excess of LDL.
  80. Cells in the body express on their surface a protein called the LDL receptor. As its name suggests, this receptor binds to LDL in tissue fluid, initiating a process by which cholesterol and the other components of the LDL enter the cell. The majority of LDL receptors in the body are usually expressed in the liver.
  81. As stated, most of the cholesterol in the body is synthesised in the liver. This involves a number of different steps, one of which (the crucial rate-limiting step) is catalysed by an enzyme called 3-hydroxy-3-methyl-glutaryl coenzyme A reductase (abbreviated to HMG-CoA reductase). The activity of HMG-CoA reductase is itself suppressed by the presence of cholesterol and metabolites of cholesterol in the cell.
  82. Maintaining a steady state concentration of cholesterol within the liver (and other cells) is achieved by an interplay between LDL receptor expression and HMG-CoA reductase activity. Inhibition of HMG-CoA reductase leads to an increase in the number of LDL receptors on the surface of liver cells. The consequence is a reduction in the levels of LDL cholesterol in the bloodstream.
  83. Statins

  84. The search for a pharmacological inhibitor of HMG-CoA reductase began in the late 1960s. The first "statin" was discovered in the 1970s – a natural fungal product called Mevastatin. This led to the discovery of another fungal product called Lovastatin which was approved for human use in 1987. Subsequent statins have all been semi-synthetic or fully synthetic compounds. Simvastatin was approved for human use in 1989, Pravastatin was approved in 1991 and Fluvastatin in 1993.
  85. The general perception of statins in 1996

  86. Professor Durrington's evidence, in his second report and in his oral evidence, which I accept was as follows:
  87. a. In 1996 there was still considerable scepticism about the benefits of using cholesterol-lowering drugs (including statins) to lower LDL and treat coronary heart disease. He exhibited two prominent articles which had appeared in 1991 and 1992, one by Professor Michael Oliver in the Lancet and the other by George Davey Smith in the BMJ. Both authors were opponents of the use of LDL lowering drugs to treat coronary heart disease. In both cases, the authors were particularly concerned that any reduction in cardiac mortality resulting from statins might be offset by an increase in non cardiac mortality. It should be noted that, by the priority date, Professor Oliver at least (I am not clear whether Davey Smith had also been converted by then) had been persuaded of the real benefit of statins.

    b. The two papers had a significant effect upon the prescribing practices in Britain. The message conveyed to many physicians by Oliver was that statins should not be used to treat patients with high (as opposed to very high) cholesterol levels. Likewise the George Davey Smith paper influenced many doctors not to prescribe cholesterol lowering drugs and caused health authorities to think about whether they should allow doctors to prescribe such treatment. The result was that Britain ended up being the country with the lowest use of cholesterol lowering drugs in the world at one stage, despite having one the highest risks of CHD. Although that summary related to Britain, it needs to be remembered that the skilled addressee is not necessarily to be found in Britain: obviousness needs to be considered on a world-wide basis.

    c. It was following the two papers that the British Hyperlipidaemia Association published an article (see Postgrad Med J 1993; 69: 359-369) on the topic, offering some guidelines about the management of hyperlipidaemia. Its purpose was to counter the negativity of Oliver and Davey Smith and to present a stronger case in the other direction.

  88. Between 1994 and 1996, the results of three large scale clinical trials were published (referred to generally as 4S, WOSCoPS and CARE) involving the use of simvastatin and pravastatin.
  89. The three clinical trials

    A. 4S – Scandinavian simvastatin study group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: Scandinavian simvastatin survival study Lancet 19 November 1994; 344: 1383-1389

  90. This was a landmark study, one of the first properly conducted randomised placebo controlled trials – a "major piece of science" according to Professor Durrington. The patient group was a high risk population with established coronary heart disease – angina or previous myocardial infarction. The conclusions were unequivocally positive at least in relation to secondary prevention ie where patients have already developed coronary heart disease. Professor Oliver himself was converted, stating in a short paper (see BMJ 20 May 1995; 210: 1280-1281) that there was no longer any doubt about the benefit and safety of statins in treating hypercholesterolemia in patients who have had a myocardial infarction; at least the balance of risk had tilted clearly in favour of treatment with statins. The message was that statins should be prescribed for use in high risk patients although the jury was still out in relation to primary prevention.
  91. B. WOSCoPS – Shepherd J et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolaemia The New England Journal of Medicine, 16 November 1995; 333(20): 1301-1307

  92. This study was written for the West of Scotland Coronary Prevention Study Group (hence its reference as WOSCoPS). The results were very positive but did not remove all concerns about the use of statins. In particular, although this was a primary prevention trial, the patient group was high risk – men with relatively high cholesterol and resident in Scotland where the risk of coronary heart disease was amongst the highest in the world. Furthermore Professor Durrington considered that there were concerns about giving statins to people with mild elevations in cholesterol since certain epidemiological observations had suggested that low cholesterol levels might cause bowel cancer; I see no reason to doubt what he says.
  93. Despite WOSCoPS, there remained a lack of confidence on the part of physicians and patients about the use of statins according to Professor Durrington; Professor Horsmans was willing to agree with that, at least so far as the UK was concerned, although things were different in the US and other parts of Europe. The principal manifestation of that caution was that clinicians were more ready to prescribe statins in secondary prevention than in primary prevention as Professor Durrington said and as Professor Horsmans recognised too. The move had been from thinking that cholesterol treatment did not work and was possibly harmful, to thinking that it did work but it might be harmful, as Professor Durrington put it in cross-examination and as Professor Horsmans accepted. Professor Durrington added, I note, that the thinking moved on further to the view that it did work but was too expensive.
  94. Side-effects of statins

  95. For statins, the main side effects of clinical concern are hepatotoxicty (liver damage) and muscle inflammation (sometimes referred to as myalgia, myopathy or myositis).
  96. Hepatotoxicity

  97. Elevated levels of liver transaminase enzymes are a clinical indicator of hepatotoxicity as Professor Durrington agreed. However, he readily acknowledged that elevated levels did not necessarily mean that a patient was suffering from hepatotoxicity and these elevated levels are frequently asymptomatic.
  98. Nonetheless, according to Professor Durrington, it was not known in 1996 that these increased levels might not be an early indicator of a serious (unknown) adverse consequence. He also says that it was (and still is) therefore not uncommon to monitor liver transaminase levels in patients on statins although he doubts that such an exercise is worth doing. That is because he does not believe, as a specialist, that raised transaminases are statin induced at all; the problem, however, is that many doctors still feel there is a significant risk of hepatic damage from statins. The advice being given in 1996 was to monitor liver function every two months during the first year of treatment. As appears from a 1996 paper co-authored by Professor Horsmans:
  99. "Another effect of the HMG-CoA reductase inhibitors concerns the increase in hepatic enzyme activity which may end in acute hepatitis. It is advisable to monitor liver functions every 2 months during the first year of treatment."
  100. According to Mr Wyand's submissions, in 1996 hepatotoxicity was believed to be dose-related. He relies on Professor Durrington who explained that for most clinicians (the general and the specialist) the assumption would be that hepatotoxicity would be dose related and be more likely to be encountered with a higher dose than a lower dose. Mr Arnold pushed him to accept that people did not know what relationship there was, if any, between dose levels and clinically significant hepatotoxicity. The following answer summarises his view:
  101. "I think the clash is not so much between the specialist and the generalist, I think the clash here is between the practicing physician who may be a specialist and the academic. I think we talked about this yesterday when we talked about orthodoxy. So I, as an academic, can entertain a view which flies in the face, perhaps, of what people would consider to be common sense clinical experience and clinical knowledge. And I would, as I have said to you, I think that in terms of treating a patient, as a general principle, doctors would expect to encounter more side effects from medication as they increase the dose. Now, that may be, as you have pointed out, about once a day and twice a day, it is a similar sort of argument. I can accept that you could produce a really good paper which suggests that once a day medication is no better than twice a day. I can accept that that would be possible, but I have already told you that on its own would not change medical opinion and medical opinion would need a lot more to change it away from the belief that once a day medication was better than twice a day. I think we have the similar sort of situation here, that most doctors would assume that if it says in the BNF that elevations of transaminases can occur, the hepatotoxicity can occur with a particular drug, they would anticipate that they would be more likely to encounter that with a higher dose of that drug than with a lower dose. That is a principle which most doctors would adhere to, in my view regardless of whether they are specialists or not. It is only academics who might say, "That is probably wrong. We are going to go out and do a study to prove that it is wrong"."

  102. Mr Wyand refers to a short passage in his cross-examination of Professor Horsmans on this topic in the context of a paper by Conrad B Blum (Comparison of Properties of Four Inhibitors of 3-Hydroxy-3-Methylglutaryl-Coensyme A Reductase: AMJ Cardiol 1994; 73: 3D-11D). When asked whether there was a dose relation, he agreed that there was although I suppose that his agreement could be taken as related only to the proposition that the authors perceived a dose-relation. Mr Arnold says that it is flatly wrong to attribute to Professor Horsmans agreement with the proposition that hepatotoxicity was believed to be dose-related. It is clear that Professor Horsmans himself did not believe there to be such a relationship. But that is not the point I am now addressing which is whether Professor Horsmans considered that the common perception, contrary to his own belief, was that there was a dose relationship.
  103. I should note that Blum is one of the papers which Mr Arnold objects to on the basis that it was not pleaded or even referred to in the expert reports or skeleton arguments. It is, however, a paper which Professor Horsmans accepted he would have expected would be found by a person interested in the treatment of hypercholesterolemia and was himself aware of it. I refer to it, however, not to establish from the paper what was secondary common general knowledge, but to put in context the question asked by Mr Wyand and Professor Horsmans' answer to his question. In any case, the objection is based on the observations of Pumfrey J in Nutrinova (see paragraph 129 below) which deals with the pleading of prior art which is relied on as secondary common general knowledge as that term is explained by Mr Arnold. Blum is not relied on as secondary general knowledge in this way; Mr Wyand has not submitted that anyone looking at Blum would accept it as correct without question or anything of that nature.
  104. Mr Arnold relies on a passage in Professor Horsmans' report and on two passages in his cross-examination. Since this is a very important aspect of the case on obviousness – the problems with hepatotoxicity might provide a motive of developing a sustained release formulation of fluvastatin - I must deal with these parts of his evidence.
  105. In his report, at paragraph 19, he says this:
  106. "For almost all drugs, it is impossible to demonstrate a relationship between dose level and clinically significant hepatotoxicity. This was and is a known problem in clinical pharmacology. Contrary to Professor Durrington's assumptions, it was not possible in 1996 (or indeed now) to understand why or how statins are involved in a drug-induced liver injury process."
  107. To be fair to Professor Durrington, I do not think that is what he did assume. What he said was that it "was understood to be a reflection on how adept the liver is at metabolising [statins] to non-toxic metabolites". He was simply giving an explanation of why it was thought (as, according to him it generally was) that hepatotoxicity was seen as dose-related; it was not a view which he himself held. Moreover, the fact that it was not possible to understand why or how statins are involved in a drug-induced liver injury process does not entail that it was not a commonly held perception that they were so involved. Mr Arnold did not succeed in shaking Professor Durrington at all.
  108. The two relevant passages in Professor Horsmans' cross-examination are not, I am afraid, the clearest questions and answers which I have heard. The first is the passage which Mr Wyand actually relies on mentioned above and which suggests that Professor Horsmans did indeed recognise that it was believed in 1996, that hepatotoxicity was dose-related.
  109. The other passage concerned a possible relation between hepatotoxicity and the concentration effect in the liver. The exchange went like this:
  110. "Q. Hepatotoxicity may be related to the concentration effect in the liver?
    A. If you believe that [quaere whether this should read "what"] most experts thought in 1996, that is correct. Then you are not speaking about hepatotoxicity. We are really speaking about an [unclear] phenomenon, except for Paracetamol which was really well-known in 1996. Hepatotoxicity is not linked with the dose that you give. For most hepatotoxicity drugs, in other words, nearly all except Paracetamol, it is not linked to the dose or the concentration. It is always an immuno allergic phenomenon, and by definition it depends on not the dose but the fact that you have a different pathway mechanism immunology proper to the patients who could explain, partly at least, what is ongoing in the human being. As you know, regarding hepatotoxicity, we have no animal models, except for Paracetamol, but Paracetamol is very particular."
  111. It is clear from his answer that Professor Horsmans was saying that hepatotoxicity is not dose related in nearly all cases but its cause was unclear - "an added phenomenon of the liver" as he put it in another answer. He was not addressing, in this answer, whether there was a general perception that hepatotoxicity was dose-related.
  112. I accept Professor Durrington's evidence that the general perception among general and specialist practitioners was that hepatotoxicity was dose related. I think it likely that the skilled team would have had concerns about hepatotoxicity. But even if that is wrong, the persons actually prescribing any drug which the skilled team was able to formulate and which was subsequently marketed would typically have held that belief in 1996. In terms of motivation, it seems to me not to make a great deal of difference whether the team itself was concerned about hepatotoxicity being dose related or whether it was dealing with a potential market which had that concern.
  113. Systemic side effects

  114. One of the perceived side-effects of statins is muscle inflammation or myositis. Professor Durrington explained (and I accept) that the cause of myositis was (and indeed still is) not well understood but it was known that the statin or active metabolite had to enter the systemic circulation for it to occur. Mr Wyand submits that it was the commonly held view in 1996 that it was desirable to get more of the statin into the liver and less into the systemic circulation. Professor Horsmans accepted that it was "one of the beliefs" by which I understood him to mean that there was at least a significant body of opinion to that effect. However, although Mr Wyand may be right in the limited context of systemic side-effects – almost self-evidently he is right – I do not accept his submission more generally. I return to this aspect at paragraphs 223ff below.
  115. The concerns about myositis are reflected in the advice given to prescribing doctors. The advice in 1996 was that patients receiving statins should be told to report symptoms of muscle pain, tenderness or weakness. Treatment was to be stopped if there were any signs or symptoms of muscle disorders. For example see the following entries for fluvastatin at the priority date:
  116. a. MIMS March 1996: S[pecial] P[recautions]: ……Advise patients to report any unexplained myalgias, muscle tenderness or weakness. .
    b. BNF March 1996 Cautions:….. advise patients to report muscle pain.
    c. ABPI Data Sheet Compendium 1995-96 Precautions: … Patients should be advised to report promptly unexplained muscle pain, tenderness, or weakness, particularly if accompanied by malaise or fever. Lescol therapy should be discontinued if markedly elevated CPK levels occur or myopathy is diagnosed or suspected.

  117. I am satisfied from the evidence of Professors Durrington and Horsmans that in 1996 it was generally thought that myositis was dose-dependent. Professor Durrington expressed the view that that was the general perception. Professor Horsmans' evidence recognised that there was a concern about this: Mr Arnold accepts that it was a commonly held, but not a generally accepted, view that myositis was dose-dependent. What Mr Arnold says, however, is that the quotes from Professor Horsmans' evidence are deeply selective. However, I think it is worth identifying a couple of the extracts:
  118. "Q. Myopathy, in the right-hand column, "Myopathy, although rare, is the most dramatic effect adverse effect of the HMG-CoA reductase inhibitors.....". it talks about what it is. "Very rarely, this proceeds to rhabdomyolysis and renal failure. With lovastatin and probably also with the other HMG-CoA reductase inhibitors, the risk of myopathy is dose related". Again, that was the common belief at that time generally amongst clinicians?
    A. Yes."

    "Q. No, but you know that in 1996 at least systemic side effects were believed to be related to the concentration of the statin in the systemic circulation; yes?
    A. But once again it is fascinating because you are speaking about the population, I am speaking about a patient. On a population base you are fully right….."

  119. I also take account of the final answer in the passage below, a passage dealing principally with Professor Horsmans' response to a question in re-examination about his view of the significance of myopathy being observed even at low doses:
  120. "I think the average skilled person will take his time before prescribing statin, surely for primary prevention, and even some physicians were reluctant to prescribe any statin at any dosage regarding that because it was immediate. Thus, the patients can see in the following days, in the days following the prescriptions, that he has acute myalgia and muscle ache and so on. That is a real discomfort, so that is the reason why I think a lot of people, still in 1996, are in some doubt about the fact that they have to put everybody on statins, even some patients in the context of secondary prevention.
    Q. Does that tell us anything about whether it is a dose-dependent effect or a patient-dependent effect?
    A. I think when you look at myopathy there is both. In other words, when you look at all the things there is clearly a patient-dependent effect. You are not able, and even now nobody is able to predict if a patient will have myalgia, (unclear), myopathy and so on. You can say there is some risk factor, but a patient without any risk factor can develop these kind of things, even at low dose. But it is true that if you take a population based study on population based patients who will receive statins, the more you increase the dose the more you incur this risk."

  121. So Professor Horsmans acknowledged that the average physician would be cautious about prescribing statins for primary prevention and that some were reluctant to prescribe statins at any dose because of the risk of myopathy. Accordingly, it is fair to say that many people in 1996 doubted the merits of widespread use of statins, even in the context of secondary prevention.
  122. Pharmaceutical formulations

  123. A drug must be formulated into a suitable dosage form (eg tablet, capsule, suspension) before being administered as a medicine. The object is always the same – to produce a dosage form that will deliver the drug to the patient in a therapeutically optimum manner and perform consistently.
  124. The choice of dosage form depends on the characteristics of the drug. Orally administered formulations are the most preferred in terms of patient acceptability. The main choice is between tablets and capsules. The former are the preferred route as they are generally cheaper and easier to manufacture than capsules.
  125. Immediate release

  126. In many, indeed probably most, cases, the tablet or capsule is designed to disintegrate rapidly in the stomach allowing the active ingredient to dissolve in the gastric fluid. Such formulations are referred to as "immediate release". The dissolved drug passes out from the stomach and is absorbed across the intestinal wall into the hepatic portal vein. The rate at which the drug enters the blood stream is limited by the rate at which it crosses the intestinal wall rather than the rate of dissolution. The drug then passes via the hepatic portal vein to the liver. A proportion of most drugs is metabolised in the liver. The remainder then passes into the systemic circulatory system which distributes the drug around the body. In its passage around the body, it can be further metabolised and/or excreted. These various processes are complex and vary from drug to drug. ADME (absorption, distribution, metabolism, excretion) studies form an important part of the development programme for any new drug.
  127. Professor Collett and Dr Rue were agreed that immediate release is the default position in the case of a new drug, not least because, as Dr Rue says, one wants to get to the market as soon as possible after which it is possible to extend the product lines with alternative formulations.
  128. Sustained release

  129. For many drugs, immediate release formulations work perfectly well. The concentration of active ingredient in the blood peaks and troughs between doses but not so as to exceed the 'therapeutic range' ie never so great as to be toxic and never so low as to be ineffective. However for other drugs, immediate release formulations may not be ideal because the blood concentration exceeds the therapeutic range – too great, too low or both.
  130. For many drugs, the disadvantages of immediate release (toxicity and/or ineffective clinical response, high frequency of dosing) have resulted in the development of formulations in which the rate of transfer across the intestinal wall is no longer the crucial step. Instead the speed at which the drug enters the blood stream is determined by its rate of dissolution in the gastrointestinal tract. The result is a reduction in the frequency of dosing (hence improved patient compliance) and maintenance of the drug concentration within the therapeutic range. There are several different names for these formulations including "modified release", "controlled release", "sustained release", "extended release" and "delayed release". These terms are not well defined and some are capable of being used to describe different systems.
  131. "Sustained release" compositions are designed to release the active ingredient over an extended period of time. They are designed with the object of making the formulation the rate limiting factor.
  132. In its opening skeleton Novartis contended that sustained release formulations are only considered when there is a therapeutic need. However, Mr Wyand says that the development of sustained release formulations may sometimes be undertaken for purely commercial reasons; he cites as an example the sustained release formulation of ranitidine at Glaxo about which Dr Rue gave evidence and which I accept. The conventional (immediate release) formulations of ranitidine were perfectly satisfactory for treating ulcers on a twice a day regime. But a new class of drugs (proton pump inhibitors) were coming onto the market in once-daily formulations. The development of a sustained release ranitidine formulation at Glaxo was driven by the need to maintain market-share by matching the dosage regime of the new drugs.
  133. Dr Rue and Professor Collett agreed that, both today and in 1996, such formulations might be developed for commercial reasons. By commercial need, Dr Rue explained that this could include a line extension or a change in dosing regimen such as from twice daily to once daily. Professor Collett agreed with Dr Rue about line extensions although he considered that a reduction in the frequency of dosing from twice a day to once a day would count as a therapeutic advantage rather than a commercial advantage.
  134. Although commercial considerations may make it highly desirable to manufacture a line extension, it is not necessarily obvious that it will be technically possible to do so. It might be rational to embark on a project with the hope of producing a new product and, if it is a line extension, there may be factors which lead the project team to think that they might succeed. But that is not to say that they have a reasonable expectation of success let alone an expectation of success. Accordingly, the fact that a course of action might be one which the skilled team would almost certainly embark upon because of commercial pressures from management, does not mean that the team will actually embark upon the project with an expectation (rather than a hope) of success. No doubt if the project team were to tell management that there was only an outside chance of success, management would review the project. I have used the word "success". What counts as success or not will depend on the facts of the case and what the objective in sight is.
  135. Commercial drivers of this sort do not make the invention which might be made as a result of the project obvious if they are not technically obvious; using the word "obvious" (as Mr Wyand does do from time to time) to describe what the team would do because of commercial drivers does not mean that the result, when achieved, is obvious in the sense that word is used in the legislation. Commercial considerations may provide a motivation for pursuing a particular course; I accept, therefore, that an argument against obviousness on the basis that there is no therapeutic need, can be met by showing that there is a commercial need or that it is commercially "obvious" to take a step. But I reject any suggestion that obviousness in the context of inventive step necessarily follows from the correctness of that argument. It makes no difference to that conclusion that, as is established by the decision of the Court of Appeal in Hallen Co v Brabantia (UK) Ltd [1991] RPC 195, if a claimed invention is technically obvious, it is immaterial that the step would not have been taken for commercial reasons ie it was not commercially obvious.
  136. Making sustained release compositions

  137. There are various well-known strategies for making sustained release compositions. For example, the drug can be embedded in an insoluble matrix, through which it gradually diffuses over time. Alternatively the matrix can incorporate a slowly soluble or eroding polymer. Another approach is to encapsulate particles or cores within an insoluble polymer layer through which the drug diffuses. There are numerous variations on these three basic types of system – the composition of the matrix and/or polymer can be tailored to meet the desired release characteristics of the drug in question. Combinations of the three types of system can also be used. Finally a more elaborate strategy is to use osmotic pressure as the driving force to generate a constant rate of release of drug through a small hole in the surrounding semi-permeable membrane – the OROS system.
  138. Not all drugs are necessarily suitable to be formulated as a sustained release composition. At least for a drug whose therapeutic action is related to its concentration in the systemic circulation, certain characteristics militate against a sustained release formulation; it is common ground that the following factors would all be taken into account in assessing suitability for a sustained release formulation. In general they would be regarded as disincentives but, according to Dr Rue, not necessarily such as to put the formulator off such a course. There is a dispute about whether the same applies in the case of a drug whose target organ is the liver, an aspect to which I will come later.
  139. Solubility

  140. Mr Arnold says that low and high water solubility were regarded as a problem. Dr Rue accepted that very high or very low solubility – extremes in aqueous solubility – may be problematic but not otherwise. See further at paragraph 220 below.
  141. pH-dependent solubility

  142. Mr Arnold says that, in addition, pH-dependent solubility is a problem. Dr Rue recognised that pH-dependent solubility could be a problem but, as with solubility generally, only in extreme cases. It is worth setting out his carefully qualified answer which I accept as correct:
  143. "A….It depends on the absolute magnitude of the solubility. If it is still reasonably soluble over the range, then it does not matter. If it varies from very soluble to very highly insoluble, then it is going to cause a problem.
    Q. It is particularly a problem if it is going to be influenced by the variation in pH in the GI tract?
    A. It depends on the lowest solubility that that achieves. Let me take an example. If the drug is more soluble in the acid then as it passes down the intestine it will become less soluble. Nevertheless, if it is still reasonably soluble at pH 7-8, then the variability is not a problem. If, on the other hand, it varies such that when it reaches pH 7, it is 0.01 mg/ml or less, then, yes, it probably would be a problem."

    pKa

  144. The pKa of a drug is an important physical-chemical consideration as it indicates the fraction of the drug which will be ionized and/or un-ionized at various pH values. Since it is generally accepted that most drugs are preferentially absorbed in the uncharged form, acidic drugs with low pKa values will have a higher fraction of uncharged species at the low pH of the stomach and first region of the small intestine. Thus if the drug is weakly ionising, the pKa value is relevant because, as Professor Collett put it "if it lies in the physiological pH range, the drug may well display different dissolution rates throughout the gastrointestinal tract". This was also regarded as undesirable. Dr Rue accepted that variable ionization may cause drug stability problems.
  145. Partition co-efficient

  146. Between the time that a drug is administered and the time it is eliminated from the body, it must diffuse through a variety of biological membranes which act primarily as lipid-like barriers. A major criterion in evaluation of the ability of a drug to penetrate these lipid membranes is its partition coefficient (a measure of differential solubility of the drug). There is an optimum partition coefficient for a drug at which it most effectively permeates membranes and thus shows greatest activity. Drugs with a partition coefficient which is higher or lower than the optimum can be poorer candidates for formulation into sustained release formulations. However, Dr Rue considers that this is a problem only in extreme cases and the context of the answers he gave make it clear that he considered this to be the generally held view in 1996. I accept his evidence on that.
  147. Stability

  148. Most oral sustained-release systems are, of their nature, designed to release their contents over much of the length of the GI tract. Drugs which are unstable in the environment of the intestine present potential problems as Dr Rue accepted. But, as he pointed out, this is the case in any oral dosage form.
  149. Absorption

  150. The rates, extent and uniformity of absorption of a drug are important factors when considering its formulation into a sustained release system. Drugs that are slowly absorbed or absorbed with a variable absorption rate are, according to Novartis' case, poor candidates of sustained release formulations. Again, Dr Rue accepted that extremes of absorption and variability of absorption are problematic. There is a problem where the drug is not effectively absorbed throughout the small intestine. For example, if there is an active transport mechanism high in the small intestine such as for ascorbic acid and riboflavin.
  151. First-pass metabolism

  152. According to Mr Arnold, sustained released formulations were regarded as unsuitable where the rate of first pass metabolism was high or subject to variability within the gastrointestinal transit. Dr Rue was asked some questions on this by Mr Arnold who referred to Welling & Dobrinska: Dosing Considerations and Bioavailability: Assessment of Controlled Drug Delivery Systems. He read this passage:
  153. "Hepatic metabolism is a saturable process. Saturable elimination has been reported for only a small number of drugs following intravenous administration, including alcohol, phenytoin, carbamazepine, valproate, theophylline and probenecid. These observations have been based on drug/concentration dependent elimination rates. The small number of drugs that have exhibited this phenomenon indicates that most drugs do not achieve drug levels that saturate hepatic metabolising enzymes at therapeutic dose levels. After oral dosing, on the other hand, the drug reaches the liver via the portal vein at far greater concentrations than normally observed in the systemic circulation. In fact, the levels may be high enough to exceed the capacity of the hepatic metabolising enzymes. Thus, the higher the oral dose the greater the possibility of saturating hepatic drug metabolising enzymes. Conversely, the smaller the dose, or the slower the dose is released from the formulation, the smaller the possibility of saturating first-pass metabolism. The potential for reduced drug availability due to first-pass metabolism is therefore greater with controlled or sustained release formulations than with conventional dosages."

  154. Mr Arnold put this question: "So what they are saying is that if you have extensive first-pass metabolism, that is a problem for a sustained release formulation?" To which Dr Rue answered: "In terms of the bioavailability and the systemic circulation, potentially, yes" confirming that he meant precisely that and not elsewhere other than the systemic circulation. His evidence, which I accept, is that an extensive first-pass effect is a problem for drugs whose site of action is reached via systemic circulation but goes no further than that. See further at paragraphs 223ff below.
  155. Half-life

  156. Dr Rue agreed that long and short half lives (referring to the rate at which the drug is metabolised) are problematic. But he did qualify this by saying that the half-life may or may not be a problem depending on the dose. If there is a very short half-life and a very high dose, then he thought it almost certain that there would be a problem. But if there is a very low dose then he considered that the half-life is unlikely to be a problem because a huge amount of the drug would not be needed in the tablet. He agreed that, generally speaking, a half life of less than 2 hours or greater that 12 hours is probably not suitable for a sustained release formulation; this was, I consider, part of the general common knowledge.
  157. Dose size

  158. Drugs that needed to be administered in a large dose were regarded as unsuitable. Dr Rue agreed with this taking greater than 1g as large for this purpose. Professor Collett explained that the active ingredient was not the problem but the form (eg the salt) and the excipient might be. In relation to fluvastatin, he accepted that this would not really be a problem; he engagingly said that he would not go home worrying about it.
  159. Molecular size and diffusivity

  160. Drugs with relatively high molecular weight (greater than 500 according to Professor Collett) were regarded as less suitable for sustained release formulation. Although this is a factor to be taken into account by the formulator, there is no evidence to suggest that this has ever been perceived as a problem in the case of fluvastatin
  161. Distribution

  162. Drugs with a high distribution volume were regarded as poor candidates. Again, although this is a factor to be taken into account by the formulator, there is no evidence to suggest that this would have been perceived as a problem in the case of fluvastatin.
  163. Cumulative administration

  164. Professor Collett pointed out that the fact that fluvastatin had a cumulative therapeutic effect over 2-4 weeks meant that it would not have considered suitable for a sustained release formulation. He was not challenged on that. Dr Rue largely accepted this. He agreed that it was a disincentive, not a strong indication either way but not encouraging.
  165. Patient Compliance

  166. Patient compliance has become something of an issue in the present case because it is said by Actavis that there are patient compliance advantages – as well as it is said other therapeutic advantages – in having a single daily dosage. Fluvastatin in a sustained release formulation can be administered once a day whereas the larger dose in immediate release tablet form of 80 mg per day needs to be administered twice daily.
  167. There is agreement that, other things being equal, for an oral dosage form, once daily dosing is generally regarded as optimal; Professor Collett went along with that suggestion.
  168. However, matters went further because there is a dispute about whether patient compliance was in fact better with a single daily dose rather than a twice daily dose. It was generally accepted, and was common ground in the present case, that twice daily dosing achieved better compliance than four times daily or even more; but there is no evidence to establish that single daily dosing achieved better compliance than twice daily dosing. Whatever research might one day reveal on that question, Professor Durrington's clear evidence was that there was in 1996 and remains to this day a widespread belief amongst clinicians, regulatory bodies and drug companies that patient compliance is better with single daily dosing than with twice daily dosing notwithstanding the absence of clinical evidence of any statistical difference to that effect. Professor Collett wisely declined to step into this debate since it was not his field – it was a matter, he said for the clinicians and left it to Professor Horsmans to give his view, as to which see paragraph 275 below.
  169. Professor Durrington's view is at least consistent with an American study in 1990 on which he was cross-examined. The study was an imperfect trial for a whole variety of reasons as he explained and which I do not need to go into. Although the message of the study appears to be that there was no significant difference between once and twice daily, I am sure that Professor Durrington was right to take the view that the paper would not enable him to influence the medical profession, the licensing authorities or drug companies to modify their thinking and to accept that twice daily dosing was as effective, in patient compliance terms, as single daily dosing. As the authors concluded:
  170. "A large number of medications are currently available that are effective when administered once or twice daily. Therefore, probably the simplest and single most important action that health care providers can take to improve compliance is to select medications that permit the lowest daily dose frequency possible"
  171. My conclusion is that it was generally accepted among the medical profession – and clinicians in particular – that once daily dosing achieved better patient compliance than twice daily dosing, albeit that experts such as Professor Durrington might personally have considered this to be a view unsupported by any evidence.
  172. The Skilled Addressee

  173. The skilled addressee can, of course be a team. In the present case, the team includes, as Mr Wyand submits, at least a formulation scientist, a pharmacokineticist and a clinician with an interest in the treatment of hypercholesterolemia. Using slightly different terms, Mr Arnold includes in the team (i) a pharmaceutical formulation chemist (or "formulator" for short), (ii) a pharmacologist or clinical pharmacologist and/or a pharmacokineticist or biopharmaceutist and (iii) a clinician. I think (i) is the same as Mr Wyand's formulation scientist. As to (ii), a pharmacologist is a scientist who is concerned with the mode of action and therapeutic effect of drugs ie the effect of the drug on the body; a clinical pharmacologist is both a pharmacologist and a clinician, and therefore will be concerned with both the mode of action and the administration of drugs; a pharmacokineticist is a scientist who is concerned with what happens to drugs in the body ie the effect of the body on the drug, particularly in terms of absorption, distribution, metabolism and excretion and a biopharmaceutist is a pharmacokineticist who is also concerned with the effect of using different dosage forms. As to (iii), a clinician is a doctor who is directly involved in treating patients rather than undertaking theoretical or laboratory studies, and who will therefore be involved in drug administration and patient care, in particular in clinical trials. (ii) and (iii) match Mr Wyand's second and third team members. So we have members of the team who are represented by Dr Rue and Professor Collett and by Professor Durrington and Professor Horsmans.
  174. Of these, Mr Arnold says that the Patent is primarily addressed to the formulator. I am not sure what that submission is meant to indicate other than that the formulator might be the leader of the team. The Patent is addressed to the skilled addressee who is an amalgam of the individuals making up the team. I do not think any assistance is to be derived from the proposition that the Patent is addressed principally to A rather than B.
  175. Professor Collett's description of the qualifications and experience of the formulator is to be found in his first report – essentially he would have a first degree in chemistry, pharmacy or pharmaceutical science and several years' postgraduate experience in a laboratory, including relevant experience in formulation. He would be assisted by a team of less experienced personnel and have access to analytical laboratories. Dr Rue agreed with Professor Collett's description.
  176. Dr Rue agreed with Professor Durrington (and Professor Collett) that the team would include a clinical pharmacologist. Professor Durrington agreed with Dr Rue (and Professor Collett) that the team would include a pharmacokineticist.
  177. Novartis contends that the skilled team would also include a medicinal or pharmaceutical chemist such as Professor Bycroft. Dr Rue did not agree that the team would include a medicinal/pharmaceutical chemist, but agreed that it would include what he termed a development chemist having a lot of the same expertise at least if it was not a line extension project. Professor Collett was clear that someone like Professor Bycroft would be at the initial meeting, and thereafter a pharmaceutical process chemist would be involved: that was a term Professor Collett used to describe the chemists who "get hold of [the compound]" and carry out all sorts of work on it. Professor Durrington accepted that he was not in a position to say that Professor Collett's and Professor Bycroft's opinion that the team would include a medicinal/pharmaceutical chemist was wrong. Accordingly, Mr Arnold submits that the correct conclusion on the evidence is that the team would include a medicinal/pharmaceutical chemist.
  178. One needs to be a bit careful about what Professor Collett and Professor Bycroft actually say. I get the clear impression from Professor Collett that, although the medicinal/pharmaceutical chemist attends the early meetings, he would soon hand over to other chemists (the physical chemists and the property chemists) with the formulator being very much in charge.
  179. Professor Bycroft agreed with Professor Collett that the formulation chemist commonly works in a team including pharmaceutical chemists, biopharmaceuticals experts, pharmacologists and chemists, agreement which he expresses having been a member of such a team many times in his career. He does not actually identify in his report the nature of the research carried out by the teams with which he was involved. But it seems that they did include some line-extension projects since he says that the team would meet "for example, when considering the potential of novel pharmaceutical compositions of a known compound (such as the development of alternative formulations of fluvastatin)". However, that is entirely consistent with Professor Collett's description of the presence of this team member at the early meetings.
  180. I do not think that Dr Rue would disagree that development chemists would be involved with the formulation of a new drug; where he parts company with the case Mr Arnold presents is the extent of any involvement in line-extension. But it seems to me that both Professor Collett and Professor Bycroft are saying that in their experience, a team concerned with line extension would include a medicinal chemist. I accept that evidence so far as concerns the early stages of such a project. But I agree with Mr Wyand that the evidence for Novartis does not go further than that, and I accept involvement of a medicinal chemist would be comparatively short-lived.
  181. So far as the clinician is concerned, Professor Durrington agreed that the clinician would not need to be someone at the forefront of translating the use of statins into clinical practice, nor an ordinary hospital physician or GP, but someone with experience in testing new drugs and new formulations of old drugs and in particular running clinical trials. He would not be restricted in his role to performing clinical trials. I agree with Professor Durrington when he said that the role of a clinician is to provide information as to the way statins work and their effect on metabolism. As he explained, it would be foolhardy not to consult someone with experience in the field of lipoproteins, lipoprotein metabolism and statins. He readily accepted that there was a need for a team of different disciplines. I do not consider, as has been suggested, that the clinician's only role is to carry out post-formulation trials.
  182. Common general knowledge

  183. It may be thought to be supererogation to say anything in this judgment about the law on a topic so central and well-known in the field of patents as common general knowledge. But because there has been some focus on what has been termed by Mr Arnold as secondary common general knowledge, I need to say something about it.
  184. Common general knowledge means the information which at the date of the patent in question is known and accepted without question by those who are engaged in the art or science to which the alleged invention relates. It is important to differentiate between matter which was in the public domain at the priority date of the patent and matter which can properly be regarded as common general knowledge. Evidence that a particular fact is known or even well-known to a witness does not mean that it is common general knowledge. Likewise, a piece of information disclosed in a scientific paper does not become common general knowledge merely because it is widely read. On the other hand, it is not necessary to show that the information is known in the sense that the skilled person has memorised it. Material which is known to exist and to which the skilled person would refer as a matter of course if he cannot remember it is part of the common general knowledge. See generally Terrell on the Law of Patents (16th ed) at paragraphs 6-36 to 6-38.
  185. This is easy to state but often difficult to apply; and it is important to bear in mind Aldous LJ's salutary warning in Beloit Technologies Inc v Valmet Paper Machinery Inc [1997] RPC 489 at 494 ll. 21-31:
  186. "It has never been easy to differentiate between common general knowledge and that which is known by some. It has become particularly difficult with the modern ability to circulate and retrieve information. Employees of some companies, with the use of libraries and patent departments, will become aware of information soon after it is published in a whole variety of documents; whereas others, without such advantages, may never do so until that information is accepted generally and put into practice. The notional skilled addressee is the ordinary man who may not have the advantages that some employees of large companies may have. The information in a patent specification is addressed to such a man and must contain sufficient details for him to understand and apply the invention. It will only lack an inventive step if it is obvious to such a man…..
    It follows that evidence that a fact is known or even well-known to a witness does not establish that that fact forms part of the common general knowledge. Neither does it follow that it will form part of the common general knowledge if it is recorded in a document…."
  187. As Laddie J explained in Raychem Corp's Patents [1998] RPC 31 at 40 ll. 1-40, what is common general knowledge does not depend on the memory of the skilled person:
  188. "This emphasises the difference between pleaded prior art and common general knowledge. The court is trying to determine in a common sense way how the average skilled but non-inventive technician would have reacted to the pleaded prior art if it had been put before him in his work place or laboratory. The common general knowledge is the technical background of the notional man in the art against which the prior art must be considered. This is not limited to material he has memorised and has at the front of his mind. It includes all that material in the field he is working in which he knows exists, which he would refer to as a matter of course if he cannot remember it and which he understand is generally regarded as sufficiently reliable to use as a foundation for further work or to help understand the pleaded prior art. This does not mean that everything on the shelf which is capable of being referred to without difficulty is common general knowledge nor does it mean that every word in a common text book is either. In the case of standard textbooks, it is likely that all or most of the main text will be common general knowledge. In many cases common general knowledge will include or be reflected in readily available trade literature which a man in the art would be expected to have at his elbow and regard as basic reliable information."

  189. Mr Arnold developed, in the light of the foregoing, uncontroversial, description of the law, the concept of secondary common general knowledge. He refers to two cases which consider the position regarding information which the skilled person would obtain as a matter of routine before tackling the problem at hand. In Nutrinova Nutrition Specialities & Food Ingredients GmbH v Scanchem UK Ltd [2001] FSR 42 at [81] Pumfrey J said:
  190. "There is a category of information which would be accepted generally once discovered and which will certainly be discovered by the skilled unimaginative chemist doing his job, not as a matter of performing a diligent search, following all the leads and cross-references, but as a matter of identifying the publications which are necessary to use as a starting point for tackling the new problem. I have no doubt that publications of this sort must be pleaded, if only to avoid surprise, but if it is established that their disclosure would form the basis of the skilled man's approach to the new problem it seems to me that they will form part of the knowledge with which he will approach other disclosures. To this extent, the information which they disclose must be treated in the same way as information forming part of the common general knowledge."

  191. That is no doubt correct; but, as the judge recognised in a passage appearing just before that quoted, the skilled person will read the document which, as a skilled person, he will discover on his literature search with the eyes of the skilled person and therefore as an unimaginative person skilled in the relevant art.
  192. Lewison J has also considered this aspect in Ivax Pharmaceuticals UK Ltd v Akzo Nobel NV [2006] EWHC 1089 (Pat), [2007] RPC 3. After referring to Nutrinova, he said this:
  193. "[55] I respectfully agree. If that is what would happen in the real world, as a matter of the routine of a skilled but unimaginative person, there is no reason why it should not happen in the world of patents. Although this information may not be part of the common general knowledge, as that expression is usually understood, it is nevertheless to be treated in the same way as common general knowledge. In so far as this process consists of linking one piece of existing knowledge with another, it is an uninventive linkage.
    [I would add that "it" in this context must mean the information which the skilled person would take home from what he read, not necessarily what a real expert would understand from reading the same literature.]

    [58] … It is part of routine formulation to carry out tests on the drug that the formulator is asked to formulate, in order to discover its properties. In the course of such routine work the formulator would discover all he needed to know about tibolone in order to put into effect the techniques of formulation that formed part of his stock in trade. …
    What was the state of knowledge of the skilled person?
    [61] I have phrased the question in this way because, for reasons I have explained, I consider that it is right to imbue the skilled formulator with knowledge of the properties of the drug to which he is to apply his ordinary and unimaginative formulation skills."

  194. In the light of those decisions, Mr Arnold makes the following submission in his closing submissions (the additions of [A] to [E] are mine for ease of reference):
  195. "Novartis accepts that the law was correctly stated and applied by Pumfrey J in Nutrinova and by Lewison J in Ivax. [A] It follows that the knowledge of a formulator asked to formulate a drug includes the knowledge about the properties of the drug (what Lewison J described at [11] as "basic information about the physicochemical properties of the drug" [and, Mr Arnold would add, its pharmacodynamic and pharmacokinetic properties]) which the formulator would as a matter of routine either be given or ascertain from the literature and accept as reliable or determine by testing during the course of pre-formulation. [B] It does not follow that all papers which could be found by a formulator conducting a diligent search forms part of his state of knowledge. [C] Still less does it follow that all papers which could be found by other members of the team conducting a diligent search form part of their common general knowledge. [D] Even if a document could be found, the information contained in it does not form part of the common general knowledge unless it is proved that the information was both generally known and generally accepted by the relevant class of persons. [E] If it is not proved that information contained in a paper is part of the common general knowledge of the relevant skilled person, it cannot be relied upon as part of an obviousness attack unless the document in question has been pleaded as prior art."

  196. That submission raises some difficulties. I readily accept [A] provided that it is not read as implying that certain other knowledge is thereby excluded. I also readily accept [B].
  197. I also agree with [C] to this extent. Just as it is not the case that all papers which could be discovered by a diligent formulator form part of his state of knowledge, so too it is not the case that all papers which could be discovered by a diligent member of the team, for instance the clinician, forms part of the state of knowledge of the formulator; the case is a fortiori. But I would not go much further. In particular, if the state of knowledge of the skilled clinician includes the information contained in a paper which he would have been bound to discover as a matter of routine, then the information is part of his state of knowledge and is part of the state of knowledge of the team (just as the common general knowledge of the team includes the common general knowledge of each member of the team). Since the skilled person is a hypothetical construct (a number of persons making up the skilled team) I see no justification for splitting the (total) knowledge of the construct between its skilled members (remembering that each skilled member bears no relation to a real person: he is the "nerd" as described by Jacob LJ to the distaste of Pill LJ).
  198. I also agree with [D]. But that leaves open the status of the information which is disclosed by a literature search but does not already form part of the common general knowledge. Its disclosure does not turn the information into actual common general knowledge when it is discovered, even if the team carrying out the particular project in hand would have been bound to discover it. It is, however, according to Pumfrey J to be treated in the same way as information forming part of the common general knowledge provided that it "would be generally accepted once discovered" to use his phrase in Nutinova. I confess to having some difficulty with that formulation. How, I wonder, can any person giving expert evidence have any real idea about whether what can be taken away from the paper having read it would have been "accepted generally" or not. The expert can identify what he himself takes away from the paper but that may be different from that which the unimaginative skilled person would have taken away. Be that as it may, one can see the idea which Pumfrey J was articulating and if one were to substitute "would regard as highly probable to be reliable" for "would be generally accepted once discovered", I would not dissent.
  199. As to [E], Pumfrey J certainly took the view that publications of this sort must be pleaded, if only to avoid surprise. That may well be correct in relation to a document which is said to be secondary common general knowledge in the sense in which Mr Arnold uses that term although I rather doubt that Pumfrey J was seeking to lay down an inflexible rule rather than a desirable practice. I do not think anything really turns on this since, in the end, Mr Wyand did not seek to rely on any of the disputed papers for what they contained rather than as a cross-examination tool.
  200. For his part, Mr Wyand says that, in the present case the formulation scientist would be provided with or would find out for himself information regarding the physiochemical properties of fluvastatin sodium – its solubility, pKa, partition coefficient etc. I think that that short statement is consistent with the result that would be reached applying Mr Arnold's approach to what the skilled person would discover.
  201. Obviousness – the Law

  202. There is some dispute between the parties about the effect in the present case of the decision of the House of Lords in Conor Medsystems Incorporated v Angiotec Pharmaceuticals Incorporated [2008] UKHL 49, [2008] RPC 28 ("Conor"). Before coming to that, I should set out some general propositions which I do not think are contentious.
  203. Obviousness is a question of fact. A qualitative, not a quantitative, test is to be applied. The primary evidence is that of properly qualified experts who will assist the court in determining whether the relevant step would have been obvious to a skilled man having regard to the state of the art. All other evidence, such as, for example what persons in the art did or did not do, is secondary to that primary evidence. Secondary evidence could be relevant to corroborate the primary evidence, but it needs to be kept firmly in its place. See Mölnlycke v Procter & Gamble Ltd [1994] RPC 49 at pages 112-114.
  204. The primary function of the expert witnesses is to educate the court in the technology – they come as teachers, as makers of the mantle for the court to don. What matters is how good they are at explaining things. It is permissible for the expert to opine on the 'ultimate question' ie whether or not the invention is obvious. However such conclusions in themselves are of little value. What matters are the reasons for the opinion. See Technip France SA's Patent [2004] RPC 46 at paragraphs 12-16.
  205. In assessing an allegation of obviousness, it is necessary for the court to consider the matter from the perspective of the relevant skilled person. This includes not merely the positive and negative aspects of his common general knowledge, but also his attitudes and perceptions, his preferences and prejudices: see Dyson Appliances Ltd v Hoover Ltd [2001] EWCA Civ 1440, [2002] RPC 22 at [55]-[57] (Aldous LJ) and [87]-[90] (Sedley LJ) and Panduit v Band-It Co Ltd [2002] EWCA Civ 465, [2003] FSR 8 at [27]-[31] (Aldous LJ) and [50] (Mance LJ).
  206. Mr Arnold submits that it is also necessary for the court to consider what problem or need the skilled person is trying to solve or meet: see Hallen Co v Brabantia (UK) Ltd [1991] RPC 195 at 212 l. 22 – 213 l. 6 and the authorities cited therein. As Slade LJ pointed out at the beginning of this passage, "one cannot assume that the skilled man simply makes technical trials for the sake of so doing". This, I agree, is certainly one of the factors the court will take into account. But it does not follow that because the need is already met or a problem already solved that improvements or different solutions cannot be obvious.
  207. A related point is that it is relevant to consider whether there was any motive to take the step in question from the prior art when deciding whether it is an obvious one: see Pharmacia Corp v Merck & Co Inc [2001] EWCA Civ 1610, [2002] RPC 41 at [124] (Aldous LJ).
  208. I should deal with one particular point arising from that paragraph of Aldous LJ's judgment. In its opening submissions, Actavis submitted that it did not matter whether there was any incentive or reason to formulate fluvastatin as a sustained release composition because this was an inherently obvious step to take, like making a 5¼" plate. That submission derived from a passage in the judgment of Laddie J in Hoechst Celanese Corp v BP Chemicals Ltd [1997] FSR 547 at 573 (referring to workshop developments such as a 5¼" plate). Mr Arnold submits that this was a passage which Aldous LJ stated in Pharmacia Corp v Merck & Co Inc [2001] EWCA Civ 1610, [2002] RPC 41 at [124] "should not be followed generally". I do not read Aldous LJ in that way. It is true that he did not agree with Laddie J in introducing a test not in the Act (namely whether a step from the prior art was a workshop modification). But what he was saying should not be generally followed, I consider, was not that gloss on the statutory test; rather, it was Laddie J's remark at the beginning of the passage cited by Aldous LJ namely "Before a step from the prior art can be held to be obvious there must be some reasons why the man skilled in the art would wish to take it". As Aldous LJ said, a "step from the prior art, albeit made without reason, can still be obvious", later holding that whether there is reason for taking the step is an important consideration, although not an essential requirement for a conclusion of obviousness.
  209. There has been some debate in the cases and before me about whether it is correct to ask whether the skilled person could take the step(s) in question or whether he should or would take the step(s). I will not give an answer to the debate at this moment. But I do observe, that, in the past, courts have sometimes asked themselves one or other of those questions. In Hallen v Brabantia at 211 l. 48 – 212 l. 20 the Court of Appeal agreed with Aldous J that the proper question depends on the facts of each case, though always bearing in mind that the onus falls on the party asserting invalidity to show that the invention was obvious. Mr Arnold says that "the settled jurisprudence of the European Patent Office is that an invention is only obvious if the skilled person would take the step, and not merely if he could: Case Law of the Boards of Appeal (5th ed) at 131-132". He has not actually referred me to the cases but only the statement in that citation. Even that citation does not go quite as far as he suggests. The passage actually reads as follows:
  210. "It is the boards' established case law that the question is not whether the skilled person could have carried out the invention, but whether he would have done so in the hope of solving the underlying technical problem or the expectation of some improvement or advantage – the so-called "could-would" approach (T 2/83, OJ 1984, 265; 90/84, T 7/86 OJ 1988, 381; 2000/94, T 885/97). So the point is not whether the skilled person could have arrived at the invention by modifying the prior art, but rather whether, in expectation of the advantages actually achieved (ie in the light of the technical problem addressed), he would have done so because of promptings in the prior art (T 219/87, T 455/94, T 414/98)."

  211. It is in this context always important, in assessing obviousness, as it is with novelty, to bear carefully in mind the statutory words. It is easy to find in the cases words more or less apposite to the facts of the case (eg would/could, motive, expectation of success, workshop variants, whether there is a reason for taking the step from the prior art) to describe how the court has made its decisions, using concepts which cannot be of universal application. Time and time again, the Courts have emphasised that the correct question is that laid down in the statute, namely whether the invention was obvious to the person skilled in the art: see in particular Asahi v Macopharma [2002] EWCA Civ 466 at paragraph 27 ("Evidence as to what could or could not or would or would not be done can be relevant, but the correct question is that laid down in the statute") and Conor. In that case, Lord Hoffmann cited with approval the observations of Kitchin J in Generics v Lundbeck [2007] RPC 32 at 72 in considering how a number of different factors should be taken into account:
  212. "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."

  213. The skilled person is deemed to look at and read publicly available documents, however obscure they may be. He never misses the obvious, nor stumbles upon the inventive. He is deemed to be interested in the prior art, even if a real worker in the field would not be. He reads documents assiduously. He is never bored (and is thus, I add, quite unlike a lawyer or a judge). This reflects the policy of the law, that anything which is obvious over what is available to the public cannot be the subject of valid patent protection: see Pfizer Ltd's Patent [2001] FSR 201 at paragraph 62 per Laddie J. It does not matter whether there are a number of obvious routes. Providing that the route in question is obvious, the claim is invalid: see Brugger v Medic-Aid Ltd [1996] RPC 635 at 661.
  214. It is always important to guard against hindsight. With hindsight, the most meritorious inventions can appear obvious; but the Court's task is to attempt to put itself into the shoes of the skilled person who does not have the advantage of hindsight. Further, arguments that inventions are obvious in the light of common general knowledge need to be scrutinised with particular care. As noted by Aldous LJ in Coflexip SA v Stolt Comex Seaway MS Ltd [2000] EWCA Civ 242, [2000] IP & T 1332 at [45], the advantage of alleging obviousness in the light of common general knowledge is that the argument can be advanced "unencumbered by detail". Such arguments enable the elements of the invention to be picked and mixed from what was known, without proper attention being given to their context. Accordingly, there is a real danger of hindsight creeping in, as was held to be the case in Coflexip.
  215. It is frequently appropriate, but not compulsory, to apply the structured approach to obviousness set out in Windsurfing International Inc. v Tabur Marine (Great Britain) Ltd [1985] RPC 59 as reformulated Pozzoli SPA v BDMO SPA [2007] EWCA Civ 588 [2007] FSR 872 ("Pozzoli") as follows:
  216. 1) (a) Identify the notional "person skilled in the art"

    (b) Identify the relevant common general knowledge of that person;

    2) Identify the inventive concept of the claim in question or if that cannot be done, construe it;

    3) Identify what, if any, differences exist between the matter cited as forming part of the "state of the art" and the inventive concept of the claim or the claim as construed;

    4) Viewed without any knowledge of the alleged invention as claimed, do those differences constitute steps which would have been obvious to the person skilled in the art or do they require any degree of invention?

    Conor

  217. I need now to consider the decision in Conor. Lord Neuberger agreed with the decisions of Lord Hoffmann and Lord Walker, adding that "the decision represents a significant development in United Kingdom patent law" and noting that the House was differing from "the views of highly experienced Judges in that field" [Mummery LJ and Jacob LJ (with Tuckey LJ) in the Court of Appeal and Pumfrey J at first instance]. It is difficult not to agree with that, although it is difficult at the same time to read Lord Hoffmann as seeing himself as doing anything other than applying well-established law to clear facts.
  218. The only claim in issue (claim 12) can be written out, without dependencies on earlier claims, as follows:
  219. "A stent
    for expanding the lumen of a body passageway,
    comprising a generally tubular structure coated with
    a composition comprising an anti-angiogenic factor and a polymeric carrier, the factor being anti-angiogenic by the CAM assay,
    and wherein said anti-angiogenic factor is taxol, or an analogue or derivative thereof,
    for treating narrowing of a body passageway,
    for treating or preventing recurrent stenosis"
  220. Stripped of its verbiage, the invention specified in claim 12 is a stent coated with taxol for preventing or treating restenosis. The dispute was whether a vascular stent in accordance with the claim was obvious.
  221. The claimant contended that the invention of claim 12 was invalid for obviousness in the light of three items of prior art. For present purposes, Mr Wyand finds it necessary to elaborate on only one of these, a patent application referred to as Wolff. Jacob LJ quotes from Wolff over one-and-a-half pages, and then says:
  222. "In summary, therefore, Wolff discloses, amongst other things:
    (a) The idea of a drug eluting stent, the purpose of this being to achieve local delivery of the drug;
    (b) The fact that very little drug would be needed because of local delivery;
    (c) That such a stent might be useful to deal with restenosis;
    (d) That the kind of drug which might be used is an "anti-replicate";

    (e) That "anti-replicate drugs include among others…[a list of 7 drugs but not including taxol]".

  223. Pumfrey J held that claim 12 was invalid. His reasoning (found in paragraphs 61 to 64 of his Judgment) is set out in [26] of Lord Hoffmann's Opinion. Pumfrey J. identified the fundamental question as follows:
  224. "61. Is it sufficient for Conor to show that taxol is an obvious candidate for testing on a drug-eluting stent in addition to the material specifically identified in Wolff, or is it necessary to show that taxol is an obvious, or the obvious, material to use in a drug-eluting stent for administration to human beings? Put another way, is the patent vulnerable only if it can be shown that the skilled person would have an expectation of success sufficient to induce him to incorporate taxol in a drug-eluting stent, or is it sufficient that without any expectation of success he would test or screen taxol?" [my emphasis]

  225. Pumfrey J answered the question in the sense indicated by the words in bold above. He did so on the basis that the patent itself did not demonstrate that a taxol-coated stent would work to prevent restenosis.
  226. "62. In my judgment, this question is to be answered by assessing the contribution to the art disclosed by the specification. For the reasons that I have given above, I am satisfied that the disclosure of the specification is that taxol may be incorporated in a stent. It does not suggest that such a stent would be safe or that such a stent would work to prevent restenosis. I think it is fair to say that the sum of the disclosure of the specification is that taxol should be incorporated in a drug-eluting coating on a stent with a view to seeing whether it works to prevent restenosis and whether it is safe. If it is obvious to the skilled person that taxol should be incorporated in a drug-eluting coating on a stent with a view to seeing whether it prevents restenosis, and is safe, then the claim is invalid, the specification having made no contribution to the art …
    The claim is to a physical device, that is, to a stent upon which is a drug-eluting coating loaded with taxol and optionally with other active ingredients as well. If, as I consider is the case here, the specification provides directions to make such a stent, but provides no data or other material suggesting that such a stent is in fact suitable for the treatment of restenosis, then success in preventing restenosis is not, in my view, a relevant consideration when assessing the obviousness of constructing such a stent. I accept immediately that there must be some motive for making such a stent: but a sufficient motive is the testing of such a stent to see if it has potential in the treatment of restenosis. In the present case, therefore, I reject Mr Waugh's contention that the definite object in view is the treatment or prevention of restenosis. The object in view is the testing of a taxol-loaded stent to see if it is of any use in the treatment or prevention of restenosis: that is all the specification provides."

  227. Pumfrey J concluded that in the light of Wolff, the skilled man would consider taxol to be worth testing as an anti-proliferative drug for preventing or treating restenosis. Accordingly claim 12 was invalid for obviousness.
  228. The Court of Appeal upheld the judgment at first instance.
  229. In the House of Lords, Lord Hoffmann identified the principle at stake in this way: "It is about how you identify the concept embodied in the invention which may constitute the 'inventive step' for the purposes article 56 of the EPC and section 1(1)(b) of the Patents Act 1977".
  230. To identify the concept, it was critical, as Lord Hoffmann explained, to ask the right question. He considered that both Pumfrey J and the Court of Appeal had asked themselves the wrong question; the correct question was whether it was obvious to use a taxol-coated stent to prevent restenosis not whether taxol was an obvious candidate for testing on a drug-eluting stent. In particular see the following:
  231. "17. … The invention means prima facie that specified in the claim: see section 125(1) of the 1977 Act. In the present case, the invention specified in claim 12 was a stent coated with taxol. There was no dispute that this was a new product. The question should therefore simply have been whether it involved an inventive step.
    ………………..
    19. In my opinion, however, the invention is the product specified in a claim and the patentee is entitled to have the question of obviousness determined by reference to his claim and not to some vague paraphrase based upon the extent of his disclosure in the description. There is no requirement in the EPC or the statute that the specification must demonstrate by experiment that the invention will work or explain why it will work. As the Dutch court said (at paragraph 4.17):
    "…it is not required in the view of the court that experimental data concerning such use of taxol stents in humans and the actual prevention of restenosis be included in the patent to further substantiate [the claim].""

    ……………

    In the event, therefore, neither the judge nor the Court of Appeal answered what I consider to have been the correct question, namely, whether it was obvious to use a taxol-coated stent to prevent restenosis.
    ………………
    If Pumfrey J had asked himself this question, he would have found the patent to be non-obvious. …….."

  232. It is not immediately clear to me why Pumfrey J would have answered the correct question in the way indicated by Lord Hoffmann. Perhaps it is because Lord Hoffmann saw the position differently from the Judge and the Court of Appeal in terms of what the patentee had added to the sum of knowledge. The Judge and the Court of Appeal saw what the patentee had done as no more than select one possible drug, taxol, and name it as the selected drug without more. Lord Hoffmann, however, clearly attached importance to the CAM assay: see paragraph 21 of his speech. Thus, after considering Pumfrey J's analysis and his identification of the relevant object in view as the testing of a taxol-loaded stent to see if it is of any use in the treatment or prevention of restenosis, Lord Hoffmann says this:
  233. "27 For the reasons I have given, I am afraid that I cannot agree with this analysis. In my opinion it is absolutely clear that the teaching of the specification, so far as it supported claim 12, was that a taxol-coated stent would prevent or treat restenosis. I agree with the opinion of the Dutch court (at paragraph 4.17):
    "[T]he patentee sufficiently clearly indicates in the patent that it is advantageous to use taxol (inter alia but also specifically for restenosis) and states as reason for this that taxol … scores well in the CAM assay to demonstrate its anti-angiogenic effect, bearing in mind that the patentee saw the solution for restenosis in the use of an anti-angiogenic factor."…"

  234. And later in his speech we find this:
  235. "37 The Court of Appeal upheld the judgment of Pumfrey J on the ground that the patent contained no "disclosure" saying that taxol was specially suitable for preventing restenosis. Again, I agree that the description, though offering a theory (its anti-angiogenic properties) as to why taxol would prevent restenosis, did not offer any evidence that this would turn out to be true. If it had not turned out to be true, the patent would have been insufficient. But there is in my opinion no reason as a matter of principle why, if a specification passes the threshold test of disclosing enough to make the invention plausible, the question of obviousness should be subject to a different test according to the amount of evidence which the patentee presents to justify a conclusion that his patent will work."

  236. However, it seems to me nothing turns on that for present purposes since what is of importance in the present case is the approach of the House of Lords to obviousness, not its resolution of the question of obviousness on the (very different) facts of the case before it. I have thought it necessary to mention it, nonetheless, because it is important to understand why, having identified the correct question, Lord Hoffmann then attributes to Pumfrey J a particular answer to that question. I therefore agree with Mr Wyand when he submits that, insofar as the decision of the House of Lords is authority for a principle of wider application, it is that the question of obviousness is to be determined by reference to the claim and not to some vague paraphrase based upon the extent of the disclosure in the description (see Lord Hoffmann at paragraphs [3], [17] and [19]).
  237. Lord Hoffmann went on to explain the "obvious to try" test in the following terms:
  238. "42. In the Court of Appeal, Jacob LJ dealt comprehensively with the question of when an invention could be considered obvious on the ground that it was obvious to try. He correctly summarised the authorities, starting with the judgment of Diplock LJ in Johns-Manville Corporation's Patent [1967] RPC 479, by saying that the notion of something being obvious to try was useful only in a case in which there was a fair expectation of success. How much of an expectation would be needed depended upon the particular facts of the case. As Kitchin J said in Generics (UK) Ltd v H Lundbeck A/S [2007] RPC 32, para 72:
    "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."

  239. Mr Wyand accordingly submits that in terms of the 'obvious to try" test, the House of Lords did not endorse the proposition in Saint-Gobain PAM SA v Fusion Provida Ltd [2005] EWCA Civ 177, [2005] IP & T 880 that this "really only works where it is more-or-less self-evident that what is being tested ought to work." Far from it, he contends that Lord Hoffmann, approving the Court of Appeal on this, said that the notion of something being obvious to try was useful in a case in which there was a "fair expectation of success". How much of an expectation of success would be needed depended on the particular facts of the case, including such matters as were identified by Kitchin J in Generics v Lundbeck. I agree with that analysis. The approach in Saint-Gobain was no doubt correct on the facts of the case; but each case must be decided on its facts by reference to the statutory criteria according to the approach stated by Kitchin J and endorsed by Lord Hoffmann. It will not necessarily be the case that the Saint-Gobain test is appropriate in every case.
  240. Pozzoli

  241. Mr Wyand relies on Pozzoli. I therefore need to consider what can be derived from it and what, if any, relevance it had in the present case. It was decided by the Court of Appeal after Conor in the Court of Appeal but before Conor in the House of Lords. Mr Wyand submits that it is unaffected by the decision of the House of Lords; it addresses a different issue from what was before the House of Lords in Conor and is entirely consistent with the speeches in that case.
  242. In Pozzoli the Court of Appeal was dealing with a case where an allegation of obviousness over the prior art was being denied on the basis that there was a prejudice against the alleged invention of the patent in suit. Having cited United Carbide v BP Chemicals [1998] RPC 1, Jacob LJ, with whom Mummery and Keene LJJ agreed, analysed the issue in paragraphs [27]-[29].
  243. "27 Patentability is justified because the prior idea which was thought not to work must, as a piece of prior art, be taken as it would be understood by the person skilled in the art. He will read it with the prejudice of such a person. So that which forms part of the state of the art really consists of two things in combination, the idea and the prejudice that it would not work or be impractical. A patentee who contributes something new by showing that, contrary to the mistaken prejudice, the idea will work or is practical has shown something new. He has shown that an apparent 'lion in the path' is merely a paper tiger. Then his contribution is novel and non-obvious and he deserves his patent.
    28 Where, however, the patentee merely patents an old idea thought not to work or to be practical and does not explain how or why, contrary to the prejudice, that it does work or is practical, things are different. Then his patent contributes nothing to human knowledge. The lion remains at least apparent (it may even be real) and the patent cannot be justified.
    29 This analysis does not require a different way of looking at the inventive concept depending on whether or not the patentee has shown the prejudice is unjustified as the judge thought at [67]. It is simply that in the former case the patentee has disclosed something novel and non-obvious, and in the latter not. The inventive concept, as I have said, is the essence of what is in the claim and not dependent on any question about a prejudice being overcome."

  244. Mr Wyand says that this is entirely consistent with Lord Hoffmann's reference to the patentee being entitled to have the question of obviousness determined by reference to the claim. It is necessary to look at the claim in each case and to determine whether what is there claimed is or is not obvious in the light of the common general knowledge and the state of the art, the state of the art including such prejudices as exist in the light of the way that the skilled person reads the available material.
  245. I do not think the effect of Conor (or rather lack of effect as Mr Wyand submits) on Pozzoli is as straightforward as he suggests. There is, in Jacob LJ's judgment, a strong flavour of the reasoning which can be found in his judgment in Conor in the Court of Appeal, with his reference to the contribution to human knowledge made by the inventor, reasoning which did not find favour with the House of Lords. What Conor tells us is that we must look to the claim of the patent when assessing obviousness.
  246. But restricting the enquiry strictly to the claim results in a surprising result. Take the case of a patent such as that in Pozzoli where there is a perceived prejudice – a lion in the path. If the invention works and if the patent tells the reader enough to work the invention, it should not, on this strict approach, be necessary to give any explanation at all about why or how the prejudice has been overcome. That, however, would be an unacceptable result. Can it really be the case that a patent should be granted where the claim is to an invention which is an old idea thought not to work without the applicant giving any explanation at all of why he now considers that the generally-accepted reason for thinking that it would not work (the prejudice) is wrong?
  247. The answer to that question is, I think, "No". That is because, as Lord Hoffmann recognises, there is a threshold test of plausibility. Where the invention involves overcoming a prejudice against an old idea, the specification must disclose enough at least to make the invention plausible. There must be some reason to justify the conclusion that something which was thought not to work would, after all, work perfectly well and if no reason is given, then the patent cannot be justified. However, this is not because the claim is obvious but because the description does not support the claim.
  248. Further, it needs to be noted that Jacob LJ is considering a single prejudice. It is that prejudice which prevents the invention from being obvious. He does not address the position where there is some other factor which might be sufficient to meet an obviousness attack. It might then be asked whether the patentee can rely on any such factor in meeting an obviousness attack if the teaching of the patent has not addressed that factor and provided at least a plausible explanation of why or how the problem with which that factor is concerned has been overcome.
  249. This may seem an odd question to ask. The reason why I do so is that Mr Wyand seems to be saying (a) that the only problem addressed in the Patent is the absolute water-solubility of fluvastatin (b) that, in fact, that was not a problem at all and would not have been seen by the skilled person as a problem and (c) therefore the invention was obvious. He relies on Pozzoli in reaching that conclusion.
  250. The answer to my own question must, I think, be that the patentee does not need to address such a factor in the patent before he can rely on it as a defence to an obviousness attack. There is no warrant for such a requirement in the legislation. The question is of obviousness is one of fact which cannot depend, in my judgment, on what the patentee has included in his patent.
  251. It is a different matter whether the patent should have been granted in the first place or is invalid for some reason other than obviousness. Consider this example:
  252. The patentee alleges that there is a prejudice and purports to explain why it is wrong. There are other problems, not identified in the patent, which make it non-obvious. The invention turns out to work but there is nothing in the patent to show how the other problems have been overcome or to show why the invention might work. The prejudice is shown on the evidence to be no prejudice at all.

    In such a case, the patent teaches nothing new: it does not teach how or why the alleged prejudice has been overcome since it was no prejudice at all and it does not teach how or why the other problems which can be relied on to defeat an obviousness attack have been overcome. If the patent cannot be justified (to use Jacob LJ's words) that is not because it is obvious but because it is not supported by the description. This issue does not arise on the facts of the present case, not least because the Patent was in fact granted and lack of support is not a ground for revocation.

  253. Pozzoli is concerned with prejudice. Whether the present case is really concerned with prejudice in the sense in which patent lawyers use the word, is doubtful. The water-solubility problem might be seen not as a prejudice but simply as a serious problem which it might or might not be possible to overcome. But in that case, the position in relation to reliance on factors not identified in the Patent is precisely the same as if the problem were perceived as so serious as to put off altogether the skilled person from pursuing the alternative formulation. It does not seem to me, therefore, that Pozzoli really helps Mr Wyand. It does not support an argument that, because Novartis has only identified the water-solubility of fluvastatin as a problem, it cannot rely on other problems as a defence to an obviousness attack.
  254. Of course, Mr Wyand says that the whole water-solubility issue is a red-herring (not even a paper tiger): there has never, he says, been an established prejudice against formulating highly soluble drugs into sustained release compositions. That is a matter which I shall come later.
  255. That discussion addresses several points which Mr Arnold makes about Actavis' argument. There are a couple of other aspects which it is convenient to deal with here.
  256. The first is his observation that Actavis' case appears to include this; namely that, other than solubility, the concerns which the formulator would have regarding the suitability of fluvastatin for incorporation in a sustained release formulation are not addressed in the Patent. It will be apparent that I agree with Mr Arnold when he says that, although it is true that they are not addressed in the sense of being explicitly discussed, this is legally irrelevant. However, he says that, in any case, the concerns are addressed by the Examples in the Patent, which demonstrate that, despite the apparent unsuitability of fluvastatin for such a formulation, it is possible successfully to make a sustained release formulation of it.
  257. What the examples actually show is that it is possible to make a formulation which has sustained release characteristics in vitro. They do not show that the formulation is clearly going to work in vivo. I do not see, therefore, how the concerns which Mr Arnold has identified are addressed in the Examples. As will be seen later, I conclude that it is obvious that a formulator asked to produce a formulation of fluvastatin that had the release characteristics described in the Patent in vitro, would have no difficulty in doing so in the light of his common general knowledge and what he would learn about the drug in the course of his routine pre-formulation tests; and would have been able to say, before doing so, that he was confident of success.
  258. The other point is that, according to Mr Arnold, Actavis seeks to distinguish Conor in several ways; one suggested distinction is that in Conor, the prior art suggested a range of things including taxol and it was not suggested that there was a prejudice against taxol, whereas in the present case it was obvious to make a sustained release formulation of fluvastatin and there was a counter-argument that the skilled person would not think that it would work. This is said by Novartis to be a factually incorrect characterisation of the present case because Novartis' case is that the skilled team would not think that there was any incentive to make a sustained release formulation of fluvastatin as well as that it would not have been regarded as a suitable candidate for such a formulation.
  259. That highlights the two distinct ways in which the obviousness attack is resisted.
  260. The first resistance lies in motivation; it is the proposition that the person skilled in the art would not have considered developing a sustained release formulation at all. The argument is that there was no motive for developing such a formulation. Immediate release tablets were effective. Sustained release would not be seen as likely to provide any therapeutic advantage. Even if a commercial advantage is capable in theory of providing a motive, there was no evidence that, at the priority date, any such commercial advantage would have been seen to exist.
  261. The second resistance is the alleged difficulty in developing a sustained release formulation. Even if the skilled formulator would have considered all possible formulations, he would have rejected a sustained release formulation as worthy of consideration because of the difficulties which he would have appreciated would face him. It is not just that there existed a perfectly satisfactory formulation (immediate release tablets) providing a much easier route to delivery; it was that there were problems including, but going beyond the high water solubility of fluvastatin, which would positively have put the formulator off. These prejudices are not restricted to the one identified in the Patent – the problem of high water solubility – but went further in the way investigated in the expert evidence. Further, a skilled team which had decided to look at the possibility of a formulation other than immediate release would have gone down one of the other routes available which would be seen as more likely to produce a result.
  262. I should also mention that these two lines of resistance are, in a real sense, incompatible. If a person skilled in the art would have had no motive to develop a new formulation, he would not even begin to consider the difficulties which he might face were he to attempt to do so. It is not that he would be put off developing the new formulation; it is rather that it would not occur to him to attempt to develop it in the first place.
  263. Construction

  264. There is a dispute about what the claims in the (unamended) Patent mean in referring to "sustained release". Mr Wyand submits that it simply means that the release of the active ingredient is maintained over an extended period of time, with no minimum or maximum period being specified. The proposed amendments to claims 1 and 10 bring certainty by adding "wherein the sustained release formulation releases the active ingredient over more than 3 hours". The point of construction is important since the claims relate to "sustained release" compositions. To determine what it does mean, it is necessary to carry out a rather detailed analysis of certain paragraphs of the specification.
  265. Quite apart from the fact that the Patent is dealing with a pharmaceutical composition, it is clear from paragraph [0002] (see paragraph 12 above) that "sustained release tablets" are ones which are "designed to release the drug slowly after ingestion". That paragraph is dealing with the development of these types of dosage forms for drugs generally and is not focussing on fluvastatin or indeed statins as a whole.
  266. It is also clear that a "sustained release tablet" is seen by the writer of the Patent as one which "releases the drug during several hours, typically more than 3 hours and less than 30". Since there is a reference to several hours, it cannot mean one hour. But could it mean, for instance 2 hours or 3 hours? As a matter of ordinary language, it would not include only 2 hours; rather, "several" means more than 2 but not many. One might question whether 3 would ordinarily be included within "several" but there is no need to pause to answer that question since it is clear that 3 is enough. This can be seen from the use in the same sentence of "typically over more than 3 hours" (and a similar reference in the second sentence in the context of other commonly used terms) which makes it clear in my judgment that "several hours" must include 3 hours. In my judgment, the words "sustained release" are being used to describe a formulation which releases its active ingredient over a number of hours more than 2 (and thus at least 3, since there is no justification for considering parts of hours so that 2½ hours could fall to be treated as "several" hours because it is more than 2 hours). Whether this is the usual meaning of "sustained release" I do not need to decide. As Dr Rue said, that may not be the ordinary concept of "sustained release" but, as he agreed with Mr Arnold, that is what the reader of the Patent would take away as the concept being used in it.
  267. But that is not the end of the point. Another question is the context in which that period of 3 hours is being used. The Patent is concerned with a drug; and although it is clear (eg from paragraph [27]) that the patentee is concerned with animals, it would be idle to suppose that his focus was not really on a drug for use in humans.
  268. The ordinary meaning of "sustained release formulation" is in the context of the delivery of the drug to a subject, typically a human. Both Professor Collett and Dr Rue, in their reports, reflect this reality, each referring to the therapeutic advantages or to disadvantages of sustained release in the context of its actual use in the human. Professor Collett (at paragraph 30 of his first report) includes a graph measuring blood drug concentration against time, plotting a curve for each of sustained release dosage form and a conventional dosage form. Dr Rue caries out the same exercise.
  269. In addition, Dr Rue explains, from paragraph 3.22 onwards of his report, the concept of the release rate of the drug. One starts (see paragraph 3.22) with an in vitro dissolution test. He explains that an acceptable specification is more complex in the case of sustained release than in the case of immediate release. He clearly used "acceptable specification" in the context of these in vitro dissolution tests. One can see that an "unacceptable specification" in vitro would not lead to satisfactory in vivo results. Indeed, in paragraph 3.27 he goes on to say that, following preliminary in vitro dissolution tests, human volunteer studies will be used to define the precise dissolution specification for any given sustained release formulation.
  270. Paragraph [0002] of the Patent refers to "the development and use of sustained-release tablets which are designed to release the drug slowly after ingestion". It seems to me to be clear that this is to be taken as including a reference to tablets which are taken by humans – indeed, that may well be the principal use of sustained release formulations. Accordingly, reading the (unamended) Patent in the context of what has just gone before, and without taking account of what appears later in the Patent, the reference to "typically more than 3 hours and less than 30 hours" in paragraph [0003] could be taken as a reference to the release period in vivo. That conclusion is one which seems to me to be consistent with the parts of the expert reports which I have just mentioned. I will call this "the in vivo construction".
  271. Another view is, however, possible; there is nothing logically impossible in the reference to 3 hours being a reference to a period of release in vitro. Paragraph [3] would then be read this way
  272. "A sustained-release tablet [that is to say a composition which is designed to release the drug slowly after ingestion (ie an in vivo)] releases [that is to say in vitro] the drug during several hours, typically more than 3 hours and less than 30 hours."

  273. Accordingly, the patentee would, on that approach, be using the term "sustained release" to describe a tablet with two characteristics: first, it releases slowly after ingestion; secondly, an in vitro dissolution test would show the drug releasing over the stated period.
  274. Mr Arnold effectively adopts this approach and says that the Patent is using the words "sustained release formulation" as one where the active ingredient is released over more than 3 hours; the Patent, he says, defines sustained release by reference to the paddle test adding that that "is how you know whether you are getting the 3-hour release rate or not". Perhaps he goes further than the way I have described it in the immediately preceding paragraph, since, as Mr Arnold describes it, in vivo performance does not appear to come into the picture at all, so that a formulation which released in accordance with the paddle test would be a sustained release formulation within the meaning of the (unamended) Patent even if it acted, in vivo, in the same way as an immediate release formulation. I will call this "the in vitro construction".
  275. Mr Arnold finds support for his approach to construction from the examples set out in paragraphs [0032] to [0043]. The examples were given to "exemplify the unexpectedly favourable properties of fluvastatin" in various formulations. The results show fluvastatin releasing under the paddle test over a period greater than 3 hours and to be less soluble than the comparator drugs. The section of the Patent is headed "EXAMPLES OF THE INVENTION"; and paragraph [0034] refers to the "tested types of sustained release formulation". Accordingly, it is said that the formulations tested are seen by the patentee as falling within the invention by virtue of their in vitro release characteristics.
  276. Dr Rue agreed that the formulator reading the first sentence of paragraph [0003] in the Patent would have no difficulty in understanding it. He also agreed that it was clear from that sentence that, regardless of what the reader might otherwise understand by "sustained release", the patentee regarded "sustained release" as a formulation which released for more than three hours. Unfortunately, neither the questions put to Dr Rue nor his answers to them reveal whether he considered the 3 hour period referred to in paragraph [0003] was a release period in vitro or in vivo. However, Dr Rue did accept that Figure 3 showed that both the diclofenac and the fluvastatin SR released the active ingredient over a period of more than 3 hours, and that it was quite clear that the patentee regarded both the diclofenac and the fluvastatin SR as being sustained release formulations. Mr Arnold submits that this evidence confirms that the Patent is perfectly clear as to the meaning of the expression "sustained release pharmaceutical composition" in claims 1 and 10; it is a composition which has the relevant release period in vitro.
  277. I agree with his conclusion and consider that the correct construction is the in vitro construction. It is not, I think, as clear as it might be, but reading the Patent as a whole, I consider that the patentee is using the words "sustained release" as exhibiting a characteristic of release over a period of more than 3 hours and that the Examples show this to be the period shown in the dissolution tests carried out. If that is wrong, then it seems to me that the in vivo construction must be correct. I reject Mr Wyand submissions that the in vitro construction gives no meaning to the inclusion of the word "typically" and that "sustained release" refers to any formulation which release the drug over a period of time which may or may not be longer than 3 hours or shorter than 30 hours.
  278. Dr Rue accepted that the formulator reading paragraph [0034] and the examples which follow would have no difficulty in carrying out a dissolution test of that kind using the apparatus described. Mr Arnold submits that the Patent therefore provides a clear test for determining whether a particular composition is a sustained release formulation within the claims. I agree with that if, as I consider to be the case, the in vitro construction is correct.
  279. Whichever of the two possible constructions is correct, the proposed amendment adds nothing, although it does not, I consider, assist in resolving which of the two constructions I have considered is correct: one is still left asking under what conditions the release period is to be measured.
  280. 80 mg/d dosing

  281. In March 1996, the maximum daily dose of fluvastatin was increased from 40 mg to 80 mg. The change was announced in the Monthly Index of Medical Specialities ("MIMS") in the following terms:
  282. Lescol's max. dose increased
    The maximum daily dose of the
    lipid lowering agent, Lescol
    (fluvastalin) has been increased
    from 40 mg to 80 mg.
    Clinical studies have shown
    that at a dose of 80 mg daily,
    usually taken as 40 mg twice
    daily, fluvastalin reduces LDL
    -cholcstetol by as much as 37%.
    Some patients with coronary
    heart disease may benefit from
    higher doses of fluvastatin and
    the product is generally well
    tolerated at this dose level.
    A calendar pack containg 56 x
    10 mg capsules in blister strips
    marked with the morning and
    evening doses has been produced
    for patients requiring this dose.

  283. MIMS, like BNP, is referred to by medical practitioners in the United Kingdom when prescribing drugs to patients.
  284. There is one discrete issue on dosing which, in the light of that change, I should deal with at this point. Novartis say that as of October 1996 there was no satisfactory evidence as to whether there was any difference in efficacy between administering fluvastatin once daily and twice daily. It was understood to be very well tolerated, even when given as 80 mg per day and there was no evidence that it made any difference whether the 80 mg dose was divided or not. Those propositions are based on four publications referred to in Mr Arnold's closing submissions; I consider them in the next four paragraphs.
  285. MIMS for March 1996: We are concerned with the "New This Month" page described by Professor Durrington – accurately I consider – as "just a news item that had been prepared by a medical journalist as a result of information given by the company". This news item announces that the maximum dose of Lescol has been increased from 40 mg to 80 mg, although it does refer to that larger dose as "usually taken as 40 mg twice daily" and refers to a calendar pack marked with morning and evening doses. I would have thought the message coming even from this news item was that the dosage would be twice daily; but Mr Arnold was insistent that there was nothing to suggest concern about the larger dose even if taken in one go. Professor Durrington however made the comparison with what is said in the prescribing information on page 20 of the same publication:
  286. "Q. Yes. To repeat my question, there is no hint here of any concern whatsoever that doubling the dose of fluvastatin would cause any increase in side effects at all. On the contrary, it is stated to be generally well tolerated.
    A. There is not any hint there but if you look at page 20 in the same edition of MIMS, if you look in the bottom of the left-hand column under Lescol, the very bottom of the column it says: "Initially 20 mg once daily at night, if necessarily increased at monthly intervals to max 40 mg twice daily." There is nothing there about taking 80 mg in one go.
    Q. Indeed, but then it depends on how much significance we read into that and I think ----
    A. You would have to read because that is what it is telling doctors to do. The other is, as I said, just a news item that has been prepared by a medical journalist as a result of news information given by the company."

  287. Ilingworth & Tobert: A Review of Clinical Trials Comparing HMG-CoA Reductase Inhibitors, Clin Ther, 16(3), 366-85. As its name suggests, this was a review of different statins. It refers to doses of fluvastatin of 40 and 80 mg/d. There is no indication as to how the 80 mg was administered – whether as a single or a divided dose.
  288. Zavarol et al: Efficacy of Fluvastatin. This too refers to various doses of fluvastatin but again the authors do not state how the dose was administered. In periods I and II, the dosage was 40 mg twice daily.
  289. Eliav et al: High-Dose Fluvastatin and Bezafibrate Combination Treatment for Heterozygous Familial Hypercholesterolemia. This study assessed the long term use of fluvastatin alone or in combination. In periods I and II of three periods, the 80 mg/d dose was expressly stated to have been in two doses of 40 mg, once at breakfast and the other at bed time. In period III, there was forced titration to 80 mg but it is not expressly stated in the body of the paper (in contrast with the abstract) how the 80 mg dose was administered.
  290. It is true to say that there is no evidence that it made any difference whether the 80 mg dose was divided or not; but there is no evidence that 80 mg once a day had even been tried let alone that it was well-tolerated. Professor Horsmans considered that 80 mg in one go would lead to much more fluvastatin getting past the liver and into the systemic circulation than with 40 mg twice daily which would suggest that a single dose at 80 mg would be preferable, other things being equal. The prescribing information for fluvastatin in the BNF (which Novartis accept to be common general knowledge) includes an express instruction to use single doses for 20 mg and 40 mg and divided doses for 80 mg. The common perception would therefore have been that 80 mg once a day was not recommended.
  291. Common general knowledge – the facts

  292. Mr Wyand submits that the following were common general knowledge at the priority date. In the light of the factors I have already addressed, I agree with his submissions save to the extent indicated.
  293. For the clinician:
  294. a. Statins were gaining acceptance but were still regarded with some suspicion. They were used for secondary treatment, ie following coronary heart disease, but were still not widely adopted for primary treatment.
    b. The main site of action of the statins was known to be in the liver.

    c. There was some concern over side effects, mainly hepatotoxicity and myopathy, which were generally believed to be dose related. Myopathy was believed to be related to the presence of the statins in the systemic circulation. However, on the whole, Fluvastatin sodium was well tolerated. The most frequently reported side effects were minor and transient gastrointestinal problems.

    d. It was known that the statins were subject to a high first pass metabolism ("first pass effect"). Mr Wyand suggests that a high first pass effect was generally regarded as being a good thing because it limited the amount of statin in the systemic circulation. This is not common ground: see further at paragraph 262 below.

    e. The first pass effect was known to be saturable for fluvastatin so that increasing the immediate release dose resulted in an increased percentage getting into the systemic circulation.

    f. Fluvastatin was the least effective of the four statins available, requiring a larger dose to achieve the same therapeutic effect as the other three.

    g. Peak cholesterol synthesis occurred at night although the precise timing of it was not known. Evening administration was recommended for this reason.

    h. The dosage of statins was started low and titrated up every 4 weeks or so until the level of LDL lowering desired was achieved. Dosage was adjusted at 4 weeks intervals because the maximum reductions in LDL concentration were generally only seen after 2 - 4 weeks. Fluvastatin sodium did not show any enhanced effect when the same overall daily dose was given twice daily rather than once daily.

    i. In March 1996 the maximum dose of fluvastatin was increased from 40 mg taken in the evening to 80 mg per day, but taken as 40 mg twice a day. Fluvastatin was available in blister packs of 40 mg tablets marked AM and PM.

    j. The other statins could be taken up to their maximum dose at once a day in the evening.

    k. For patient compliance, once a day is preferred: see paragraph 110ff above.

    l. Patients being treated with statins were often taking other medications. There were concerns over drug/drug interactions. Taking fluvastatin once a day would give the largest window of opportunity to take a second medication without risking a drug/drug interaction. See further at paragraph 263 below.

  295. For the formulator:
  296. a. Fluvastatin was soluble but not highly soluble according to the Pharmacopoeia definition.
    b. The different sustained release systems referred to in the Patent were common general knowledge in 1996.

    c. Even highly soluble drugs could be formulated as sustained release but at the extreme they might be limited in the technology that could be used.

    d. Fluvastatin sodium immediate release was rapidly and effectively absorbed in the GI tract.

  297. There are other matters which Mr Arnold identifies as common general knowledge some of which I have already dealt with and others of which I should mention (perhaps with risk of repetition). There was a large measure of agreement between Professor Collett and Dr Rue. The following arise for consideration.
  298. Common general knowledge - the formulator

  299. Drug properties that are relevant to the preparation of formulations generally. These include (i) aqueous solubility and dissociation constant; (ii) partition coefficient; (iii) stability; (iv) absorption; (v) distribution; (vi) metabolism; and (vii) elimination and biological half-life. These are all matters which I have addressed to some extent in considering sustained release formulations generally; but see also paragraph 220ff below. One of the objectives of the pre-formulation study would be to gather data under each of these headings. Dr Rue's evidence was that "within the research based pharmaceutical companies such information is very rarely available in the literature. The majority of this information is generated by the team." I accept that evidence. Like secondary common general knowledge gained as the result of a literature search, this sort of information forms part of the material on the basis of which the question of obviousness is to be addressed.
  300. Immediate release formulations. The majority of formulations on the market in 1996 were immediate release formulations. The vast majority of drugs are originally sold in an immediate release formulation, which is the default position at least for a new compound, something which may well be driven by commercial concerns to get the drug on the market. Again, I accept that, but would add that, ex hypothesi, it is not the default position when consideration is being given to a line extension.
  301. Patient compliance. I have already dealt with this aspect, concluding, contrary to Mr Arnold's position, that the accepted position, whether right or wrong, in 1996 (and possibly even today, although it is not necessary to decide) was that better patient compliance would be achieved with one dose daily rather than two.
  302. The rationale behind sustained release formulations. Mr Arnold says that sustained release formulations are developed in order to improve drug therapy over that achieved by non-sustained release formulations. Commercial considerations apart, I accept that Professor Collett and Dr Rue agreed that what the skilled person would be looking for in a sustained release formulation was a "significant clinical improvement" over an immediate release formulation. But this is a statement of what real people in the real world would do and is directed at the question of motivation. No doubt the skilled person, just as much as the real formulator, should be taken as appreciating that actual development may require a motive. But that does not necessarily answer the questions of obviousness. Motivation is certainly an element but it is not necessarily a decisive element (consider again the 5¼" plate). Thus, when Mr Arnold says that sustained release formulations are developed in order to achieve improved therapy, he diverts attention away from the real question – was this obvious within the meaning of the statutory words – to a different question which is whether a real formulator would actually proceed along a particular line of development. Suppose, for instance, that the evidence clearly established that any formulator, if asked by his commercial bosses whether a sustained release formulation could be made and would work, would say, "Yes, it's easy to make and I am very confident that it will work." To which the clinician adds "But it won't work any better than an immediate release formulation and it will be more expensive. Why on earth do you want to do it?". Then there is no therapeutic improvement; and his questioner might respond "Oh I see, I'll take it no further". Does that necessarily make a sustained release formulation non-obvious? I think not.
  303. It is said by Mr Arnold that the skilled person would appreciate, as part of his general knowledge, that one important potential advantage of a sustained release formulation is where it produces a blood plasma concentration of the drug which stays within the effective therapeutic range, and thus neither falls below the minimum effective concentration nor exceeds the maximum safe concentration, over the intended period of action of the drug. Mr Arnold points out that Dr Rue accepted that if the formulator does not know what the effective therapeutic range is, then he will not know whether smoothing out the peaks and troughs in the plasma concentration profile will have any effect in terms of efficacy. Still less so if the desired duration of action is uncertain. That makes sense in a context where the therapeutic action is related to its concentration in the systemic circulation; but this point is not, in my view, of significance where the therapeutic action or side-effect is not dependent on, or at least is recognised not to be dependent on, plasma concentrations.
  304. The skilled person would, I agree, also appreciate that one disadvantage of sustained release formulations is that they are more expensive to develop and manufacture. He would also recognise other potential disadvantages of sustained release formulations: (i) an increased risk of "dose-dumping" ie sudden release of a large dose if the delivery system fails; (ii) longer duration of action possibly leading to increased duration of side effects or unpredicted drug response; and (iii) loss of flexibility in dosage adjustment, eg where non-uniform blood concentration is an advantage. There is nothing to suggest that (iii) has any relevance in relation to fluvastatin; and whilst (ii) is a possibility, it is also a possibility that there will be lower blood plasma levels using sustained release with a decrease in side-effects.
  305. Developing a sustained release formulation is more difficult than an immediate release formulation. Chang and Robinson in Lieberman Lachman and Schwartz (2nd ed, 1990) says that it is "normally a very difficult task" to develop a sustained release formulation. Dr Rue did not agree with this, but he did agree that it was more difficult than producing an immediate release formulation; and he agreed that the opinion of Robinson was one that was to be respected in 1996 although he added that he did not consider that many formulators would have considered the general proposition to be correct. I accept, of course, that it was commonly, perhaps universally, accepted that development of a sustained release formulation was more difficult than producing an immediate release. I do not consider that Chang and Robinson establishes that, in 1996, it was "normally very difficult" to do so, but even if it was, it would need to be asked whether fluvastatin was, in that sense, a normal case, about which Chang and Robinson say nothing.
  306. Drug properties that are relevant to the preparation of sustained release formulations. I have already considered these in the discussion above of sustained release formulations. There are some additional points which arise in the light of Mr Arnold's submissions.
  307. a. Aqueous solubility and dissociation constant. Low water solubility was regarded as a problem, but so was high water solubility. In addition pH-dependent solubility was regarded as a problem. With the substitution of "very" in front of "low" and "high", I agree. But it is only extremes of absolute solubility which are properly to be regarded as a problem. There was nothing in the material put to Dr Rue or in Professor Collett's own evidence which would lead to the same conclusion in cases other than very high or very low solubility. Further, I accept Dr Rue's evidence that high solubility merely limits the type of technology and that this would accord with the formulator's understanding in 1996. Further, I accept his evidence in relation to pH-dependent solubility recorded in paragraph 98 above.

    Dr Rue's evidence was backed up with his own experience with ranitidine. I accept, of course, that the commercial driver there was very different from that which might be said to exist in the present case, but Dr Rue was very confident of being able to overcome such problems by selecting the appropriate technology. I found that evidence very persuasive and I accept it so far as concerns the absolute solubility of fluvastatin. In this respect Dr Rue's evidence was also entirely consistent with Langer & Wise (see at paragraph 328 below).

    b. Partition coefficient. Extremes of partition coefficient were a problem. So too was pH-dependent partition coefficient. Dr Rue considered these were problems only in extreme cases, evidence which I accept. There is nothing to suggest that this would have been seen as a problem in 1996 in other than extreme cases.
    c. Absorption. Drugs that are slowly absorbed or absorbed with a variable rate were poor candidates. See further at paragraph 102 above. I consider that it was common general knowledge

    d. Distribution. Drugs with a high distribution volume were regarded as poor candidates. Although this is a factor to be taken into account by the formulator, there is no evidence to suggest that this has ever been perceived as a problem in the case of fluvastatin.

    e. Metabolism. Sustained released formulations were regarded as unsuitable where the rate of first pass metabolism was high or subject to variability within the gastrointestinal transit. I have already considered Dr Rue's evidence about this, which I accept, at paragraph 104 above.

    f. Elimination and biological half-life. Drugs with short and long half-lives were both regarded as unsuitable. In general it was thought that drugs with a half-life of less than two hours were unsuitable for sustained release.

    g. Cumulative effect. Drugs that need to be administered cumulatively for a period of time in order to be effective were regarded as less suitable. Again, I accept what Dr Rue said about this: see paragraph 109 above. There is nothing to suggest that fluvastatin was even perceived by anyone as falling into this category.

  308. There are two specific areas – water solubility and the liver as the target organ – about which more needs to be said. Although strictly, I suppose, falling under the later heading of "Obviousness – the facts," I find it convenient to deal with them as part of the present discussion of common general knowledge.
  309. First, Mr Arnold has more to say about water solubility specifically. He refers to Dr Rue's reliance, in this context, on Table 1 in his reports as showing that it was not difficult to formulate water soluble drugs as sustained release formulations. Dr Rue was addressing the assertion in the Patent that "Biopharmaceutical requirements of a sustained release product of the water soluble drug would then at first sight impose formulation problems as discussed above….except perhaps only for a first initial short time during hydration and swelling of the matrix". Dr Rue states, in his report, that that is "is simply not correct. For a drug such as fluvastatin sodium having a solubility of greater that 50 g/ml, there would have been no difficulty in developing a sustained release formulation". It is in that context that he set out in the Table, to illustrate this point, some of the sustained release dosage forms of the more soluble drugs on the market in October 1996.
  310. Matters are not quite that straightforward, however. In cross-examination Dr Rue agreed with a number of propositions put to him:
  311. a. The Table gives no information as to how easy it was to develop the formulations, what steps the formulators took or what types of formulations were adopted;

    b. With one or two exceptions, the drugs all needed to be kept above a certain minimum therapeutic concentration throughout the day. That may be true; what it would show is that there might be a special motivation for producing a sustained release formulation but does not indicate one way or the other whether it would be difficult to do so.

    c. The drugs listed in the Table were salts of weak bases. Professor Collett's evidence was that these were easier to formulate in sustained release dosage forms than salts of weak acids. Dr Rue did not agree with that. Mr Arnold records that Professor Collett's evidence was unchallenged; that is true in the sense that Dr Rue had not dealt with this point and Mr Wyand did not cross-examine on it. However, Mr Arnold asked Dr Rue questions about it and I am not simply going to ignore his evidence. Professor Collett had not thought Dr Rue's report on this point to have been sufficiently important to respond to in his own first report. He only did so in a second report.

    d. Dr Rue disagreed with Professor Collett (except in relation to cases of very insoluble compounds) because in making a controlled release dosage form, one is attempting to put the control of release into the dosage form and take it away from the drug; therefore the solubility of the compound is not, in his opinion, particularly important and so whether it is a weak base or a weak acid is not particularly significant. As to Professor Collett's own reasoning, it was that it is generally easier to slow down (or sustain) the release of a drug when its solubility is decreasing or had reached equilibrium solubility. Conversely it is generally more difficult to slow down the release of a drug when its solubility is increasing. Dr Rue did not disagree with that. Rather, his objection was that Professor Collett's approach assumes that the dissolution rate is being controlled by the solubility of the drug. That would be so, Dr Rue agreed, with some technologies used to produce sustained release dosage forms but not with others. He expressed himself to be very confident of being able to avoid the problem. If I am entitled to take account of Dr Rue's objection as well-founded without the matter having been put to Professor Collett, it strikes me as a convincing explanation; but I am aware, of course, that an explanation can often seem convincing until it is shown to be wrong.

    e. Even if Professor Collett's evidence concerning acid/basic salts were to stand unqualified (on the basis that it was not challenged by Mr Wyand in cross-examination), his point was a general one, not focused on fluvastatin in particular. As to the general point, there remains the evidence which Dr Rue did give in his report that the characteristic of "water soluble" would not have been a concern. He explains that the whole purpose of a sustained release formulation is to slow down the rate at which the drug is released from the dosage form so that the release of the drug, rather than the absorption, is rate limiting. The solubility of the drug is only one of many factors which can affect the release rate from a sustained release dosage form. As to fluvastatin in particular, his view, expressed more than once in cross-examination, was that its solubility would simply limit the choice of technology. His answers addressed the changing solubility of the drug as it passes through the GI tract. Professor Collett did not disagree. When asked whether high solubility was going to make sustained release formulation impossible he said "It should not make it impossible. It is going to make life more difficult", recognising that high solubility would simply limit the technology available. This answer was given in the context of a question about pKa which would be, he accepted, a problem only if the variation in solubility (as a result of pH dependency) ranges from the very high to the very low.

    f. The next proposition is that some of the drugs listed in Table 1 were covered by patents. More specifically, as Professor Collett notes, the patents were granted in respect of novel, non-conventional, delivery systems and thus, he says, implies that the preparation of such formulations was not considered standard or routine. This was not challenged. There is obvious force in the point and it does show that limited assistance can be obtained from the comparisons in Dr Rue's Table.

  312. Secondly, as to drugs targeted at the liver, it has been suggested by Actavis that the skilled formulator would not think that the factors considered above were disincentives in the case of a drug which was targeted at the liver. Mr Arnold submits that there is simply no foundation for this suggestion. Dr Rue did not say this either in his report or (save in one instance which I come to in a moment) in cross-examination. There is, he says, no support for it in any of the standard textbooks. The general proposition was not even put to Professor Collett.
  313. That may be so. But equally there is nothing in the text books which have been drawn to my attention or in Professor Collett's evidence to show that some, at least, of these disincentives were seen as such other than in the context of a discussion relating to drug plasma levels. The skilled person would surely have some understanding of why the disincentive is seen as such. If the formulator did not already know, the clinician would tell him. As Mr Wyand puts it, the skilled formulator is not going to say, "I do not care because it says in this textbook and all my experience is with drugs that do not have a primary target in the liver, I am not going to do it because it has a high first-pass metabolism." It is, he says, nonsense to think that the formulator will say "I have this check-off list here that tells me high first-pass metabolism not suitable for sustained release." - end of story. I agree: it will not be because the textbooks say something about disincentives in the context of drug plasma levels. Rather, it will be because, notwithstanding that the target organ is the liver, there are perceived and identified disadvantages which have nothing to do directly with plasma levels.
  314. In his first report, Professor Collett explained his opinion that although the target site for fluvastatin was the liver, the skilled team would still be concerned that the high levels of metabolism would reduce the inhibitory activity of fluvastatin saying that the contrary would only be the case if in some manner fluvastatin was able to bind to HMG-CoA reductase before encountering metabolic enzymes for which he was not aware there was any evidence. Slowing the release of the drug from the formulation could lead to reduced therapeutic effect: there will be less fluvastatin available to avoid the metabolic enzymes and compete with HMG-CoA for the HMG-CoA reductase active site ie there will be a reduced local bioavailability within the liver. He concluded that "Preparing sustained release formulations would therefore go against the established practice at the time".
  315. Professor Collett reiterated that view in answer to the reports of Dr Rue and the first report of Professor Durrington. He adds this:
  316. "First-pass metabolism is potentially a problem with sustained release formulations of drugs whose therapeutic effect is related to its plasma concentration in the systemic circulation because slowing down the rate of release of the drug or the amount available for absorption can result in a decrease in therapeutic effect. This decrease occurs because a greater proportion of the administered dose is metabolised and does not reach the systemic circulation intact. The concern therefore relates to the amount of drug that is available to carry out its therapeutic effect. In my opinion, this concern does not disappear just because the metabolic enzymes and the target enzyme (HMG-CoA reductase) are both present in the liver. A high degree of first-pass metabolism will still mean that a reduced amount of fluvastatin is available to interact with its target enzyme, as there will be a reduction in the local concentration of unmetabolised or active fluvastatin within the liver. It would have been expected that this reduction in concentration would lead to a reduction in therapeutic effect. Furthermore, slowing down the rate of release of fluvastatin (as would occur in a sustained release formulation) and slowing down its rate of delivery to the liver could result in a further reduction in its therapeutic effect, as a greater proportion of the administered dose could be metabolised before it is able to interact with its target enzyme. In relation to this latter point, I noted in my First Report that in 1996 it was thought that the enzymes that metabolise fluvastatin were saturable. It was also known that fluvastatin exerts its therapeutic effect by binding to HMG-CoA reductase in place of its natural substrate, HMG-CoA, i.e. by competitive inhibition. Increasing the rate of delivery of fluvastatin to the liver would therefore have been expected to, increase its therapeutic effect as it will ensure that an increased amount of fluvastatin avoids metabolism and is free to interact with its target enzyme in place of HMG-CoA."

  317. Those points were not, in my judgment, adequately met by Mr Wyand. He acknowledges that Dr Rue accepted that the skilled team would be concerned about the possibility of the high first-pass metabolism reducing the inhibitory activity of fluvastatin but this, he says, was on the assumption that the molecule was being metabolised in preference to interacting with the HMG-CoA reductase. (As to selective binding, see in more detail at paragraphs 241ff below). He takes that from a passage of Dr Rue's cross-examination where he was being asked about the relevant passages from Professor Collett's first report:
  318. "Q. Prof. Collett deals with the specifics in paragraph 78 and he says there: "Although the target site of action for fluvastatin ... is the liver, the skilled team would still be concerned that the high levels of metabolism would reduce the inhibitory effect of fluvastatin" and it says "(the contrary would only be the case if in some manner fluvastatin was able to bind to HMG-CoA reductase before encountering the metabolic enzymes - for which I am not aware there was any evidence)." He is right about that, is he not?
    A. Yes, indeed, yes. But, as far as I am aware, there was no evidence.
    Q. And, therefore, it is also right to say that the skilled team would be concerned that the high levels of metabolism would reduce the inhibitory activity of fluvastatin?
    A. On the assumption that it is getting to the metabolic enzymes first, which is an interesting conjecture. I have no idea whether it would or not.
    Q. And there was no information available in 1996 on that point, was there?
    A. No, there was not. I would suggest that that would make it a very interesting trial to carry out."

  319. This is a somewhat unsatisfactory exchange. It is not clear whether Dr Rue's first answer is directed at the proposition that the skilled team would be concerned, or whether it is simply agreeing with the proposition that "the contrary would be the case…." only if fluvastatin selectively bound to HMG-CoA reductase. In his second answer, however, he appears to be agreeing with Mr Arnold about the concerns of the skilled team, but only on the assumption that the drug is getting to the metabolic enzyme first. He does not expressly say that there would be a concern otherwise. Of course, if it was the case, and had it been known to be the case (which it was not) that the drug selected in favour of HMG-CoA reductase, everyone would agree that there would be no concern on the part of the skilled team. Equally, if it was known that the drug selected in favour of the metabolic enzyme, there would be a perception of a real problem. This aspect is discussed further below when considering what Mr Arnold has to say about the common general knowledge of the clinician.
  320. The way Professor Collett put it in his report was that the skilled team would think that "drip feeding" the drug into the liver would simply reduce the therapeutic effect relative to the amount of drug administered because more drug would be metabolised before it could be of any therapeutic benefit". In cross-examination, he answered the point this way:
  321. "Q. Therefore, first-pass metabolism is not a high first-pass metabolism, is not going to give you cause for concern; or is that for Professor Horsmans?
    A. I am sorry to say it is Professor Horsmans' concern. It is a different sort of condition or site of action that I have ever come across before. The number of drugs that are active in the liver, as opposed to being metabolised in the liver, it is a novelty for me to see one that is going to be acting on the liver, so I am on uncharted waters. I am not [I think this "not" has crept in by error] going to pass this one on to Yves [Professor Horsmans]. What I am saying is that as a formulator, they say to me, 'We want this, that and the other' but I have never heard of one where they say they want it for the liver.
    Q. Fair enough. So you would not know whether a measurement of bioavailability in the systemic circulation system would have any relevance to what is actually going on in the liver?
    A. The liver, as far as I would be concerned, there would be a black box. What I can say is that if I see something appearing in the plasma, then it tells me that there has been some absorption from the dosage form that I have produced. So whoopee, it has got there, but whether it is going to go and do the business, I cannot tell you."

  322. Now, the liver may have been a black box to Professor Collett, but it was not a black box to Professor Durrington or Professor Horsmans and thus not to the skilled person. The problem from Actavis' point of view is that even the clinicians could not say that fluvastatin selected in favour of HMG-CoA reductase rather than in favour of the metabolic enzyme.
  323. The question then, it seems to me, comes down to this: Is the skilled team to be taken to have a concern only if the evidence available to it established at least a likelihood that fluvastatin would bind selectively to the metabolic enzyme? Or is it to be taken to have a concern for one (or both) of two reasons: first, because there is generally a concern about high first-pass metabolism and, in the absence of evidence, it could not be assumed that there would not be a problem where the drug is targeted at the liver; and secondly, because an understanding of the competitive binding to HMG-CoA reductase and the metabolic enzyme would lead to a concern unless it was shown positively that fluvastatin bound selectively to HMG-Co reductase? In my judgment, it cannot be said on the basis of the evidence which I have received the skilled person would not at least be seriously concerned. The fact that fluvastatin is targeted on the liver is not enough to eliminate the high first-pass metabolism of the drug as a disincentive to developing a sustained release formulation (although whether it is enough to defeat a claim of obviousness is another matter).
  324. As to the short half-life of fluvastatin, this again, is said by Actavis not to be a disincentive to the formulation of a sustained release dosage form. So far as Dr Rue is concerned, Mr Arnold relies on precisely the same passage of his evidence as Mr Wyand relies on to support the view that the formulator, knowing that fluvastatin acts in the liver, would not have had a concern about the short half life. The actual interchange went like this:
  325. "Q. ……..We saw in the books we looked at this morning repeated statements to the effect that candidates with half-lives shorter than 2 hours were not thought to be suitable for sustained release formulations. So again, on that basis, this would not have been thought to be a suitable candidate?
    A. That is too simplistic, I am afraid. For compounds whose therapeutic action is directly correlated with plasma concentration, that is true, providing the dose is fairly high. Now, I am not an expert pharmacologist or clinician, so I am not unhappy to say, you know, I think I would have said to the clinicians and the pharmacologists, 'Hang on, we are not actually concerned with the plasma concentration half-life of this drug because it is acting in the liver. I do not know what is likely to happen. Do you?' I have no idea what answer I would have got.
    Q. Would you agree with this, that the formulator who was following the advice in the standard textbooks would have thought that this was an unsuitable candidate?
    A. Not knowing what he would know about fluvastatin. In general terms, yes, if you are talking about generalities, that would be true. But you have to consider what you know about the specific drug compound."

  326. That supports Mr Wyand in the sense that the impact of a short half-life may be different when one focuses on the liver rather than plasma concentrations. The skilled team would know that the reason why a short half-life is regarded as a disincentive to a sustained release formulation is because of the reduced therapeutic effect which would be obtained where that effect depends on systemic levels of the drug. But where the target organ is the liver, there would be no reason to think that systemic levels were of any relevance, or at least of the significant relevance which they would normally have, to the therapeutic efficacy.
  327. But that is not an end of this aspect. In his second report, Professor Collett expressed the view that the concerns which the skilled formulator would have had over the short half-life of fluvastatin sodium would not have disappeared just because fluvastatin sodium exerts its therapeutic effect in the liver. A short half-life, according to him, is potentially a problem for sustained release formulations because the formulation may require unacceptably large doses to maintain a therapeutic concentration. Given the uncertainty about metabolism of fluvastatin in the liver and the lack of information on the concentration of fluvastatin in the liver required for therapeutic effect or the time period over which that concentration must be maintained for therapeutic effect, the skilled formulator would have remained concerned about the short half-life of fluvastatin sodium.
  328. I accept what Professor Collett says to the extent that the skilled team would have some concerns about the short half-life. But I do not think that the skilled team would have seen its concerns as a significant disincentive. It would not be put off by this factor to the extent of dismissing fluvastatin as a candidate of sustained release formulation.
  329. What I take away from the totality of the evidence is firstly that it was common general knowledge in October 1996 that the main site of action of the statins was known to be in the liver. It was well-known that the therapeutic effect of statins peaked during the night but it was not known precisely when. The precise way in which statins worked was still not well-understood, however. The skilled person would know that, as a general rule, a short half-life made the drug unsuitable for sustained release, but he would also know that the reason for that was concerned with plasma levels. Where the therapeutic effect of the drug was known to be in the liver, the formulator, at least, would not know what was likely to happen but he would have no reason to believe that the half-life either was a relevant factor. However, if he found something appearing in the plasma, he would, were he a pupil of Professor Collett, shout Whoopee, but his shout, it seems to me, is nothing to do with half-life but with the satisfaction that "something has got through". In contrast, the high first-pass metabolism of fluvastatin would have been perceived generally as a disincentive.
  330. What, then, of Mr Arnold's so-called secondary general knowledge, a matter I have considered in general terms already?
  331. The formulator's first priority in searching the literature would be to obtain data on the relevant properties of fluvastatin. I agree with Mr Arnold that the best evidence of what would be found by the formulator conducting a literature search as part of his pre-formulation study is contained in Professor Collett's first report, since he did actually undertake his own literature search for the purposes of preparing his report. The papers (exhibited as JC5 to his first report) were selected as being representative and reliable. Dr Rue accepted that these papers would be found.
  332. Dr Rue agreed with Prof Collett's account of the key information about fluvastatin which the formulator would have known, ascertained from his literature search or have been provided with by other members of the skilled team as part of the pre-formulation study. Some of it is summarised in the passage above under the heading Background.
  333. In addition, Dr Rue agreed that Professor Collett's report accurately stated the properties of fluvastatin sodium to be, and that the skilled team would have ascertained them to be, as follows:
  334. a. Its water solubility is 80 mg/ml at 25oC, and ranges from 1 to 100 mg/ml between pH 5.5 and pH 7.5 at the same temperature. The solubility would be greater (perhaps double) at 37oC. Its pKa is about 5.5.
    b. Its partition coefficient is 1.5 and is pH-dependent.
    c. It is sensitive to heat, light, moisture and especially low pH.
    d. Its molecular weight is 411.
    e. It was dosed at 20 – 80 mg per day as set out above.
    f. It is rapidly and almost completely absorbed from the gastrointestinal tract when administered in immediate release form.
    g. Its distribution volume is small.
    h. It is subject to extensive first pass metabolism in the liver, the first pass metabolism is saturable and its metabolites were known to be therapeutically inactive.
    i. It was reported to have a short terminal half-life of between 0.5 and 2 hours.

    Common general knowledge - the clinician and the clinical pharmacologist

  335. I need to say something more about the action of statins in the liver itself, about which there was a great deal of uncertainty. The evidence establishes the following: Once in the liver a statin molecule might either be irreversibly metabolised by one of a number of enzymes (particularly but not only CYP2C9), or bind reversibly to HMG-CoA reductase (and thus exert its therapeutic effect). It was not known, however, what the relative binding affinities of the statins (and in particular fluvastatin) were for the HMG-CoA reductase or the various metabolic enzymes in vivo or what the relative kinetics and thermodynamics were. Similarly it was not known what concentration of statin in the liver was achieved, nor how it varied over time. The skilled person would, however, think that the concentration in the liver was one of the factors determining the desired therapeutic effect as would be the duration of that concentration within the hepatocyte. It was not known how one could estimate concentration in the liver from the serum concentration, but that was the least bad guide to liver concentration.
  336. What was not known either was whether systemic exposure to statins was a good thing or a bad thing. The uncertainties are well illustrated by the 1995 edition of Professor Durrington's own book Hyperlipidaemia: Diagnosis and Management:
  337. "The hydrophilic statins inhibit cholesterol biosynthesis more specifically in the liver, which they enter as the result of an active transport mechanism. This has been hailed both as an advantage and a disadvantage."

  338. As Professor Durrington explained, for those who were concerned about inhibiting cholesterol biosynthesis outside the liver or muscle side effects it was a disadvantage if the statin went outside the liver, but for those who believed that statins had a pleiotropic effect on the arterial wall or reduced fracture rates this was an advantage. Thus there was uncertainty as to the role of statins in extrahepatic tissues. It was a subject of intense debate and interest. Mr Arnold himself relies on the fact that some people thought that release of statins into the systemic circulation was beneficial but, as I understand the evidence, reduction in blood cholesterol levels was not one of the benefits. A person concerned only about blood cholesterol levels would have his primary focus on the liver and would not be interested in seeing release of statins into the systemic circulation.
  339. Common general knowledge - the pharmaceutical chemist

  340. As discussed, there was some dispute in the evidence as to whether the fifth member of the skilled team should properly be called a pharmaceutical or development chemist or pharmaceutical process chemist, but it matters little as their common general knowledge would be the same. What matters more is the relevance of the chemist's skill to the question of obviousness. Let me set out what Mr Arnold (accurately in my view) takes away from Professor Bycroft's evidence (which evidence stands unchallenged).
  341. Such a chemist would assist the formulator by proposing and producing different salts, pro-drugs or analogues of the active in question with a view to altering its physico-chemical characteristics (and in particular solubility, stability and partition coefficient) to assist the formulation chemist in achieving the desired dose profile. With regard to fluvastatin specifically there were a number of alternative salts available which would be likely to lower its solubility and rate of dissolution. The salts could then be formulated into a suitable dosage form. Preparing such salts would have been a relatively straightforward task.
  342. An alternative route to altering the drug's physico-chemical properties would have been the development of a prodrug. In relation to fluvastatin, the drug could have been converted into its lactone form, reducing its solubility and therefore rate of release. The original drug would then be released by the action of metabolic enzymes, regenerating the original hydroxyacid form of fluvastatin within the body. The development of prodrugs to change the characteristics of a given compound was a common technique in 1996. Dr Rue was happy to acknowledge prodrugs as one possible route but considered that there were many reasons why a formulation team would not go down it.
  343. Yet another possibility might be synthesising analogues of fluvastatin, to identify a compound with more suitable physico-chemical properties. But Mr Arnold acknowledges that this route would be more time consuming and expensive and would only be considered as the last of the three options.
  344. This may all be correct; indeed it is to be taken as correct in the absence of any challenge. Mr Wyand asks, in effect, So what? It is a good question. The ultimate issue for me is one of obviousness and the fact that there may be other, perhaps even preferable, courses open to the skilled team, does not render the particular course of developing a sustained release formulation of fluvastatin ipso facto non-obvious. The statutory question is not what the skilled person would have done but whether a particular invention is obvious having regard to the state of the art. I do not really find Professor Bycroft's evidence of much assistance.
  345. Obviousness – the starting point

  346. In considering the legal principles involved, I have mentioned the related aspects of purpose and motive, that is to say what problem or need the skilled person is addressing, and whether there was a reason for the skilled person to take any steps at all. This is of some importance because identification of the task which the skilled addressee is faced with can have a huge impact on – and can sometimes even answer – the question of obviousness. One strand of Novartis' case is that the team would not have considered a sustained release formulation (except perhaps to dismiss it out of hand) because there was no motive to do so, because of the difficulties which would be faced and because of other different routes available. However, if the assumption is made that the skilled team is interested in developing an alternative to the currently available immediate release formulation, and in particular a sustained release formulation, then that removes one hurdle in the way of a finding of obviousness. Identifying the interest of the team in this way thus defines the task with which the team is concerned in a way which assumes that the team certainly will be looking to develop a sustained release formulation; so that obviousness would then fall to be assessed by what was or was not obvious to the team which had decided to embark upon the development of a sustained release formulation. It eliminates the argument that the sustained release formulation was not obvious because no-one would have had a motive to proceed along that route in the first place.
  347. Mr Arnold raises this point as part of a wider preliminary criticism of Actavtis' case, namely that it is based entirely on hindsight. In referring to the evidence of both Professor Durrington and Dr Rue, he says this:
  348. a. Both of them were shown the Patent at an early stage after they were instructed. Neither was asked to express an opinion on what the skilled team would have done before being shown the Patent. That is true, as a matter of fact. I take it into account when assessing their evidence and reaching my conclusions. But I do not think that the point is so damaging that I can really place little reliance on what they say: Professor Durrington in particular was careful to ground his answers as of the priority date. In any case, I see this as giving rise to little difficulty in relation to the clinician, Professor Durrington, rather than the formulator, Dr Rue.
    b. Furthermore, Mr Arnold complains that both were well aware of Lescol XL (Novartis' sustained release formulation of fluvastatin). There is not much that could be done about that, I think. It is hardly likely that any expert worth his (fluvastatin) salt could have been found who had not heard of Lescol XL.

    c. Even worse, both assumed that the skilled team was "involved in the development of a sustained release formulation of fluvastatin". Thus both Professor Durrington and Dr Rue said in their reports that they believed the skilled team would comprise of a team involved in the development of a sustained release formulation of fluvastatin sodium (using identical words no doubt coming from Actavis' lawyers). Both confirmed that their reports were predicated on that assumption. Dr Rue also says in his report that he had been asked to consider how in October 1996 the formulation scientist "would have approached the question of developing a sustained release formulation for fluvastatin sodium". In cross-examination he said "I think that having seen the patent, I made the assumption that somebody in the organisation had said, 'We would like a sustained release tablet of this drug, what can you do?'". While both witnesses gave evidence that they had made that assumption themselves rather than being asked to make it, it is manifest according to Mr Arnold that they were led by their instructions to make it.

  349. As to c., above, Mr Arnold is correct in the way he describes the approach of Actavis' experts (at least with the omission of the pejorative "Even worse"). I am not sure that I agree with his conclusion that it is manifest that they had been led by their instructions to make the assumption which they did. Nor does their approach to the preparation of their reports make the reports of doubtful utility. If I am satisfied that the team would in fact have thought a sustained release formulation to be worthy of consideration and would not have dismissed it out of hand, then the reports of both Professor Durrington and Dr Rue address the issues which would then face the team. Further, their function is to educate me so it is really for me to assess the extent to which what I learn is relevant to the question which I need to answer. Moreover, their evidence, as a result of cross-examination, went beyond what is found in their reports. Mr Arnold is of course correct to insist that the question to be decided is whether it would be obvious to develop a sustained release formulation. I bear that in mind – constantly – in addressing all the submissions made to me about obviousness.
  350. It is not suggested that the members of the team would be any different, or their skill-sets different, depending on whether they were simply interested in the formulation of fluvastatin to treat hypercholesterolemia or were tasked with developing a sustained release formulation of fluvastatin. The question of obviousness is therefore directed to the same team in either case.
  351. It is not, in any case, immediately apparent that Professor Durrington and Dr Rue have in fact adopted an incorrect approach. It must be remembered that the Patent deals with a known product (fluvastatin) which is to be exploited using a well known technique (sustained release): sustained release was a well-known and often exploited technology and there can be no doubt that the skilled team would have been entirely familiar with the various different methods of producing a sustained release formulation. Even if the skilled team is simply that – a notional skilled person with the skills relevant to formulation of fluvastatin – and does not come to an assumed task of developing a sustained release formulation, it does not follow that such a formulation will not form part of its thinking. It is to my mind inconceivable that the possibility of such a formulation will not pass through the mind of the skilled team since formulation is its notional stock-in-trade. It may be that an absence of purpose or motive will result in out-of-hand dismissal of such a route (and if it does, the question of whether that is sufficient to defeat an obviousness attack will arise) but once it is established that the team considers such a route as worthy of further investigation, the question of obviousness will turn, I consider, on precisely the same issues as those which would arise if the team were tasked, in the first place, to develop a sustained release formulation (as assumed by Professor Durrington and Dr Rue). In particular, if the skilled team is confronted with what Mr Arnold submits would be a difficulty in the way of a sustained release formulation, or if there is an alleged prejudice against a sustained release formulation of fluvastatin, (in either case whether because of water solubility or otherwise), the question whether that difficulty or prejudice is such as to defeat an obviousness claim cannot, in my judgment, depend on whether or not the skilled team is one which is involved in the development of a sustained release formulation of fluvastatin. In saying all of that, I am not pre-judging anything. There may be powerful reasons for saying that sustained release formulations of all types were non-obvious; and it is possible that particular claims in the Patent are non-obvious even if sustained release generally was perceived as a way forward.
  352. Obviousness over the pleaded prior art

  353. I propose to adopt the structured approach as re-stated in Pozzoli. I have already considered the notional "person skilled in the art" and the relevant common general knowledge of that person. I take the prior art in turn.
  354. BNP/common general knowledge

  355. I have set out at paragraph 43 above the relevant parts of the BNF. The reference to "twice daily" in the dosing regime means one in the evening and one in the morning. I should add the following under this heading:
  356. a. Professor Durrington's evidence, which I accept, was that the prescribing information is very likely to reflect a clinical trial. Often, the clinical trial protocol is taken (eg 40 mg twice daily) and, so long as the prescribing advice to the doctor closely reflects that, regulatory approval will be given. Professor Horsmans agreed with that, qualifying his answer by pointing out that it is not known whether the clinical trial was designed with a divided dose in order to avoid systemic side effects.

    b. In his first report, Professor Durrington expressed the view that a clinician with an interest in the treatment of hypercholesterolemia would have interpreted the twice daily regimen for 80 mg of fluvastatin as an indication that there were concerns about hepatotoxicty and systemic side effects with a single 80 mg immediate release dose. I note that that is an unqualified expression of view: it does not depend in any way on the clinician being a member of the skilled team being tasked with the development of a sustained release formulation. Mr Arnold's criticism of the starting point adopted by Actavis is not relevant in this context. There is force in Mr Wyand's submissions (a) that Professor Durrington's interpretation is entirely rational having regard to the common general knowledge in 1996 and (b) that it is also entirely consistent with Professor Horsmans' evidence, in particular that there was a desire to reduce or eliminate the incidence of systemic side effects associated with statins as far as possible and that of an 80mg dose of fluvastatin in one go, the result will be much more fluvastatin getting past the liver and into the systemic circulation. I was wholly unconvinced by Professor Horsmans' rejection of the idea that the dosing directions had nothing to do with concerns about side-effects and everything to do with how the clinical trial was in fact conducted (as to which there is no evidence). I therefore accept Professor Durrington's view as set out in the first sentence of this paragraph.

  357. Mr Wyand identifies the inventive concept for the purposes of the structured approach as the making of a sustained release composition using a water-soluble salt of fluvastatin as the active ingredient, to which I would add that the concept of sustained release in this context, on the construction of the Patent which I have reached, envisages release over a period of more than 3 hours in vitro. So far as claim 10 is concerned, he would presumably identify an additional inventive concept namely the use of the composition for the treatment of hypercholesterolemia. And so far as claim 3 is concerned, there is also the addition of the eroding matrix as the specified method of formulation.
  358. Novartis does not dispute that the BNF was common general knowledge which does not, accordingly, require separate consideration. On that basis, Mr Wyand submits that the only difference between the inventive concept and the BNF/common general knowledge is that there is no disclosure of the making of a sustained release composition: that is the only difference.
  359. Mr Arnold says the description of the inventive concept is inaccurate; it leaves out other elements of the claim in a way which expressly subsumes only integers [A] and [B]. What he leaves out are integers [C] and [D]. It is, I think, correct that the substance of [C] is omitted since the group listed is not exhaustive of the possible sustained release formulations: the inventive concept is thus narrower than Mr Wyand submits. I am not sure that anything at all turns on that in relation to claim 1 since it is not suggested that claim 1 is obvious because of a particular formulation which is not a member of the group. As to integer [D], on the construction which I have concluded is correct, the concept of "sustained release" as used in the Patent already has built into it release over a period of 3 hours or more in vitro. It might be more accurate, I suppose, to refer to the inventive concept in a way which includes the period of release over 3 hours or more. But again, it is not suggested by Mr Wyand that the Patent is obvious (in contrast with insufficient or unclear) in relation to "sustained release" as ordinarily understood but not "sustained release" as used in the Patent.
  360. I consider the inventive concept is thus a sustained release composition using a water-soluble salt of fluvastatin as the active ingredient releasing over 3 hours or more in vitro being selected from the group of matrix and or diffusion-controlled membrane coated formulations. The active ingredient was well-known, the formulation was well-known and the immediate release formulation of the active ingredient was well-known. On that basis, the differences are that the BNF/common general knowledge do not disclose the making of a sustained release composition (ie one releasing in vitro over 3 hours or more) using the specified delivery methods.
  361. As to the fourth step (ie viewed without any knowledge of the alleged invention as claimed, do those differences constitute steps which would have been obvious to the person skilled in the art or do they require any degree of invention?), Actavis has two strands to its case on purpose and motive. Mr Wyand submits that in October 1996 there were "a host of therapeutic and/or commercial reasons for developing a sustained release formulation of fluvastatin sodium". Although therapeutic improvement had formed part of Actavis' case before the hearing began, the commercial reasons were only developed as the case proceeded (Mr Arnold would say developed on the hoof to rescue a collapsing case, something I will need to decide).
  362. Actavis now relies on these reasons to establish a motive for proceeding along the path to a sustained release formulation of fluvastatin:
  363. Reduction in systemic side-effects

  364. I have already discussed these at length. So far as material is concerned, what I derive from the evidence is that the team in October 1996 would have thought that a sustained release formulation would quite possibly reduce systemic side-effects but could not really predict that with any confidence. As to hepatotoxicity, the team would not know one way or the other what the results of a sustained release formulation would be. Considerations of systemic side-effects would not deter the team from embarking on a project; nor would hepatotoxicity, but the team would be more cautious about that. Indeed, so far as hepatotoxicity is concerned (not a systemic side-effect) Professor Durrington accepted that a reduction in side-effects could not be postulated. He agreed that the team in October 1996 would have no evidence to lead them to think that sustained release would decrease liver toxicity. Professor Horsmans was of the view that there would be an increased risk because of increased exposure of the liver to fluvastatin.
  365. Although Mr Wyand puts this element of his case perfectly generally, it is apparent from the totality of the evidence that the point really stems from the increase in the maximum dose from 40 mg to 80 mg coupled with the change to twice-daily dosing. I suspect that if such a change in dosing had not been announced, it is unlikely that Mr Wyand would be saying that there was any real motivation to produce a sustained release product. It is to be noted, however, that Professor Durrington did not see the exercise as restricted to 80 mg, envisaging the possibility that the maximum dose could increase even beyond that in the future.
  366. Reduction in drug-drug interactions

  367. A once daily dose (as opposed to twice daily) will lessen the potential for drug-drug interactions – a wider window of opportunity to take another drug. This is because the sustained release composition of fluvastatin would be taken in the evening at bed-time, thus providing wider window than a dosage in the morning and in the evening. A sustained release formulation delivering over say 3½ hours or even 6 hours taken at bedtime would have less scope for interaction with other drugs taken in the morning and in the evening; at least there would be less scope than if the sustained release formulation were taken in two immediate release doses, one in the morning and one at bedtime. Professor Horsmans accepted that result in his cross-examination even though paragraph 33 of his report might be read in a different way. He added that fluvastatin was not, in fact, responsible for any such drug-drug interactions but acknowledged that this fact had not become generally accepted until after the priority date. I agree that there would have been a perceived advantage in the context of drug-drug interaction.
  368. Potential improved therapeutic effect

  369. Mr Wyand says that a sustained release formulation would be an obvious way to increase the duration of inhibition in the liver. Professor Horsmans agreed with that proposition. However, he is not, by that acceptance, to be taken as accepting also that the efficacy would be increased: it might be of equal or less efficacy. It was common knowledge in 1996 that the period of peak synthesis of cholesterol was in the middle of the night although it was not known precisely when. Professor Horsmans agreed with Mr Wyand that it was commonly believed by physicians in 1996 that the peak cholesterol activity in the night was a significant factor and explained why statins were to be taken in the evening. Whilst agreeing with that, Professor Horsmans did not consider that the data justified the commonly held view, the data showing no impressive difference between the time of day at which the drug was taken and its efficacy.
  370. This increase in the period of inhibition achieved the perceived desired effect of having the statin operate over the period of peak synthesis of cholesterol in the middle of the night. Absent specific knowledge as to window of peak synthesis, Professor Collett considered it desirable for the formulation to release for more than three hours.
  371. According to Mr Wyand, this was, in fact, the driving force behind Novartis' development of the LESCOL sustained release formulation. The following comes from a confidential internal Novartis Formulation Development Report dated 25 August 1998.
  372. "The objective behind the development of a modified release (MR) formulation is to achieve about 40% reduction in the LDL-C levels with once a day dosing."

  373. Mr Arnold refers to this document too. I find the following extract from it of relevance too:
  374. "The proposed strengths for this development are 80 mg and higher. The hypothesis behind the MR [Modified Release] development is the efficient utilization of the drug by the target organ (liver). The difficulty however, lay in identifying the release rate of the drug from the MR form which would result in optimum efficacy (% reduction in LDL-C). The issue is complicated due to the high first pass metabolism and the lack of Pharmacokinetic/Pharmacodynamic (PK/PD) correlation."

  375. Obviousness is not, of course, to be judged by the amount of work needed to reach the invention. But if one sees the actual expert team identifying difficulties, it might cast doubt on whether the skilled addressee (in contrast with the real-life team) would have perceived those difficulties as paper tigers, especially in the light of later parts of the document. For instance, it is also recorded that "due to these difficult issues" it had been decided to develop 6 different dosage forms which were then tested. Then it is said "A lot of preformulation trials had to be conducted to produce dosage forms with the desired release rates."; and later "In addition, the solubility of fluvastatin in water is very much pH dependent…..This will make the development of a pH independent MR form very challenging, if not impossible.".
  376. In any case, I do not read the evidence to which Mr Wyand refers as establishing that it could have been predicted that a sustained release formulation would give improved therapeutic effect. The reality is that nobody could have said what would happen. Professor Durrington himself recognised that the skilled person would be faced with two plausible and mutually exclusive theories about what would happen and with contrary results in terms of efficacy. He did think that the skilled person would have found literature which suggested that sustained release compositions of lovastatin and simvastatin might be more efficacious in reducing LDL but the only conclusion he drew from that was that it "might have tipped the commercial mind towards perhaps investing in this project".
  377. Professor Horsmans was firm in his evidence that no prediction could be made. He says, and I accept, that it could not be predicted whether a sustained release formulation of fluvastatin would give a better efficacy, the same efficacy or worse efficacy than an immediate release formulation. Mr Arnold comments that there would be no point developing an 80 mg sustained release formulation unless it was expected to give better efficacy than 40 mg twice daily or 80 mg once daily immediate release, or to have the same efficacy and other advantages. If one includes commercial advantages in that, what Mr Arnold says seems to me to be absolutely right. But that is subject to the observation that, as explained elsewhere in this judgment, there is nothing to suggest that 80 mg immediate release (or 2 x 40 mg taken at one time) had been tried or that it was well-tolerated; the prescribing directions suggest otherwise.
  378. Mr Wyand himself acknowledges the unpredictability. Thus, Dr Rue said that the formulator would not have been able reliably to predict whether a sustained release formulation of fluvastatin would have improved clinical utility (improved therapeutic effects, reduced incidence of adverse effects or simplified dosing regimen). The team could not be certain of success in being able to develop a sustained release formulation, taking it into clinical trials and showing therapeutic benefits. Nor was there any guarantee that a formulation which worked satisfactorily in vitro would perform satisfactorily in vivo.
  379. Nevertheless Dr Rue was very confident that the formulator could develop a tablet giving whatever in vitro dissolution rate was required. Professor Collett was very confident that he would be able to make a formulation which would work in vitro but would not have been able to guarantee success in vivo (whether in humans or dogs) – "we could give it a shot but I cannot guarantee it for you". But if asked to make a sustained release formulation, the formulator would, he accepted, see what he could do; he would not say to his commercial boss "Oh, no, I do not think it is going to work so I am not even going to try it".
  380. Patient compliance

  381. I have discussed patient compliance at some length and concluded that the common perception in October 1996 was that a reduction from 2 doses daily to a single dose would result in improved patient compliance. Whether this factor by itself would justify, in the context of fluvastatin, an expensive project to bring a sustained release formulation to the market must be doubtful. On the other hand, the skilled person, alive to the improved patient compliance and ignoring commercial considerations, might ask himself whether a single dose formulation could be made and having embarked on consideration of that issue, might consider that it was clear that such a formulation could be made; in which case it might be said to be obvious.
  382. Competition with other statins

  383. I have already considered patient compliance. There was a widespread belief in 1996 (as now) amongst clinicians, regulatory bodies and drug companies that patient compliance was better with once daily dosing than with twice daily dosing (notwithstanding the absence of clinical evidence of any statistical difference to that effect). Whether a reduction in dosing frequency from twice a day to once a day is regarded as therapeutic benefit, as Professor Collett views it, or is a commercial benefit as Dr Rue views it, the resulting formulation, according to Mr Wyand, is likely to compete better with the other once-daily statins. It is true that there is no evidence about the commercial benefits of a sustained release formulation. However, I consider that I am entitled to take account of this factor which both sides see, with different spectacles, as a benefit. It can, I consider, be brought into account as a potential motive for proceeding along the route of producing a formulation that would deliver the drug over a period of time. If Dr Rue's approach is correct, that is because it is a commercial advantage.
  384. Similarly, Mr Wyand notes that the dosing of fluvastatin at 80 mg instead of 40 mg is accompanied by a change in dosing regimen from once daily to twice daily. This, according to him, could be seen as disadvantageous and might lead the physician to change to prescribe one of the more potent statins instead. He relies on Professor Horsmans' answers to his questions, but I do not think that the Professor was as clear on this as Mr Wyand might like. What he actually accepted was that a change to 80mg taken in two separate 40mg doses might result in a move to another statin and he accepted that one interpretation of such behaviour is that there is a disadvantage in moving to twice daily. But he also said that another interpretation, another reason for the move to another statin, is that fluvastatin is not the best statin, a drug he himself never prescribed. He always started (and implicitly finished) with a different statin. He did not go so far as to say that that made it non-obvious to formulate a sustained release dosage form of fluvastatin (on the footing that it would be pointless to formulate the drug at all), but he did say that the suggested commercial advantage in having one dose rather than two was tiny. So I am left being not quite clear whether the move away from fluvastatin at a daily dose of 80mg would be because the dosing requirement would then be for two doses of 40mg or because a patient who needed 80mg would be better off with another statin even if he could take a single 80mg dose of fluvastatin.
  385. Summary

  386. Mr Wyand submits on the basis of all of the above that the motivation for the skilled team to develop a sustained release formulation is overwhelming. There were, he says, perceived therapeutic and commercial benefits which would have provided the motivation. Dr Rue's evidence was that the formulator would be very confident that he could deliver something that would work to sustain release over more than 3 hours in vivo. Although Professor Collett was not sure that he would succeed in vivo, on his evidence the formulator would at least have a good reason to try – he would not put his foot down and say "this can't be done". This suggests that the skilled person is not going to be put off even considering sustained release as a way forward. The rider of the obviousness horse is not therefore going to fall at the first hurdle.
  387. There are some points which I need to make about (and following on from) that submission before turning to what Mr Arnold has to say.
  388. The first point is that the change to 80 mg dosing is of great importance. It formed the central plank in Actavis' case as presented in the skeleton argument and has remained the starting point for the skilled team in closing submissions. Without it, it would be much harder for Actavis to argue that there was a motivation for developing a sustained release formulation. Indeed, one of the suggested benefits relied on is an improvement in patient compliance as compared with twice daily dosing, an improvement which I have held (contrary to Mr Arnold's submissions) to be an established common belief.
  389. The second point relates to side-effects. On the evidence, it was impossible to predict that there would be any therapeutic improvement – it was only a possibility, but so was a lessening of efficacy. The same goes for the effect on hepatotoxicity. In relation to systemic side-effects, there were reasons for thinking that there might be an improvement, but this could not be predicted with any confidence.
  390. The third point is that, in referring to a "sustained release formulation", Mr Wyand must be taken to mean a formulation which can be taken by a human and which will have some therapeutic effect.
  391. The fourth point is that the therapeutic effect must at least match the effect of the current dosage of 40 mg twice daily. I can understand that if the new formulation has certain benefits such as reduction in side-effects and/or better patient compliance, there might be a motive to proceed even if there was no improvement in efficacy. Similarly, I can understand that if the new formulation has improved efficacy then again, whether or not it reduces side-effects, there might be a motive to proceed. But the present case does not fit into either of those categories. Instead it is a case where the efficacy of a sustained release formulation could not be predicted; it might be the same, it might be better but it might even be worse. In order to provide a motive for developing a sustained release composition against that uncertainty, the other factors – in particular reduced side-effects and better patient compliance – would have to be seen by the team as real benefits and benefits which were likely to be achieved. In that context, the suggested reduction in side-effects would have to be compared not with a single 80 mg does of fluvastatin, but with the twice daily regimen.
  392. The fifth point follows on from that. It is that, in terms of motivation, the object of the development programme is to produce a sustained release formulation of fluvastatin which can be taken by humans and which will result in better efficacy than a twice daily dose. The object of the programme is not to produce a formulation which will display certain release characteristics in the paddle test. The success of the project is not to be judged by reaching the invention claimed in claim 1 of the Patent or indeed in any of the other claims; the success of the project is judged by the production of a drug which has real benefits as compared with the available formulations, whether benefits of increased efficacy, reduced side-effects, better patient compliance or otherwise. If it could be shown at the outset that there was no reasonable prospect of success of achieving the end-result, there would be no separate motivation to produce a formulation which had the release characteristics shown in the paddle test: there would simply be no point in making such a formulation when it would not lead to the only real objective.
  393. But even if there is no motive for embarking on the project because the end result is thought to be one incapable of achievement, it does not necessarily follow that the claims of the Patent would be non-obvious. It is logically possible that those claims (or some of them) are obvious technical developments from the prior art even though nobody would ever think of taking the relevant steps because, in practice, they would only be taken as part of a larger project designed to produce something better than that which already exists.
  394. Indeed, one way in which Mr Wyand puts Actavis' case is to say that the claims of the Patent fall within this category and that there is no need for any motive at all. That submission, if it is correct, means that there is no need to consider at all what might happen in vivo and whether the skilled team might be put off embarking on the project in the first place. I will return to this aspect of his case later.
  395. Whatever the motivation, it is a fact that nobody actually proposed, prior to the priority date, that a sustained release formulation of fluvastatin – or any other statin – would be a good idea. It was not a suggestion made by either Professor Durrington or Professor Horsmans both of whom were active in the field and publishing around that time although, given their view of fluvastatin as compared with other statins, that might be seen as unsurprising.
  396. Assuming in favour of Mr Wyand that there was a motivation from the clinician's point of view to develop a sustained release formulation of an 80 mg dose of fluvastatin, the question then is whether the formulator would consider it to be a possible candidate. Mr Arnold submits that the answer must be that he would not, for four reasons.
  397. His first reason relates to concentration effect. Dr Rue said, and Professor Collett agreed, that the usual circumstance in which a sustained release formulation would be regarded as appropriate is where repeated administration of an immediate release formulation resulted in a plasma concentration profile that exceeded the effective therapeutic range. That evidence, of course, is focusing on cases where the therapeutic range of the drug is, or is thought to be, reflected in the plasma levels. It is common ground that no information was available as to the effective therapeutic range of fluvastatin. It was also understood that blood plasma concentration did not give any guide as to the concentration in the liver, still less the concentration over time profile. Mr Arnold has set out in his closing written submission some passages from Dr Rue's cross-examination on this aspect (for reference, at Day 2 pages 138 – 140 of the transcript) from which he extracts four points:
  398. a. It confirms that fluvastatin does not fit the usual case for developing a sustained release formulation. That is true; but it does not necessarily follow from the fact that the usual case is different that fluvastatin is a case where the formulator would be put off because of that difference although he would, no doubt, look carefully to see whether the reasons in the usual case did or did not apply to fluvastatin.
    b. Dr Rue did not know whether the potential advantage in terms of patient compliance of going from twice daily to once daily dosage would "be sufficient for the company to invest the time and cost in developing such a product". I am sure Dr Rue is right in expressing that caution. What he says, however, is a clear reflection of the view that this change in dosage would be seen as having some advantage.

    c. The best Dr Rue could do was to speculate that releasing the drug over 12 hours might provide a benefit and that would be "worth having a look".

    d. He accepted, according to Mr Arnold, that if the clinical pharmacologist was unable to say that there was a likelihood of increased efficacy it would not be worth the expense and even more so if the clinical pharmacologist said that it was just as likely and possibly even more likely to be less efficacious. I am not sure that he did accept that, although he did say it was a reasonable view – if there is no likelihood of it, why would one do it? He also said that it would be the company which would make the decision, effectively viewing it as a commercial decision. I must note, however, that Dr Rue did not give his answers in the context of a question which reflects the first sentence of this paragraph: he was not asked the question on the footing that the clinical pharmacologist was unable to say that there was a likelihood etc but on the basis that the clinical pharmacologist has said that the available information suggests that a sustained release formulation would not be likely to give any improvement. Those are very different matters. A person who is told that X is unlikely to work may well not proceed; a person who is told that it is not known whether or not X will work might have a very different approach.

  399. As to the absence of knowledge about the therapeutic range of fluvastatin, the dosing regimen in the British National Formulary would, in my view, nevertheless give the formulator, at the very least, pause for thought and would provide some incentive to try a sustained release 80mg formulation. As Dr Rue said, it would be "worth having a look". Further, if there was a possibility of improved efficacy (which would include reduced side effects), this would be a driver for developing a sustained release product.
  400. Even accepting, as I do, that it could not be predicted what the outcome would be, I do not accept Mr Arnold's argument on concentration effect. The formulator would not, I consider, reject fluvastatin as a candidate for sustained release on this basis. He would, however, be cautious and certainly not see any benefit here.
  401. Mr Arnold's second reason relates to high first pass metabolism and saturable effect. I have discussed this already at paragraphs 103 and 223 above. I concluded that the fact that fluvastatin is targeted on the liver is not enough to eliminate the high first-pass metabolism of the drug as a disincentive to developing a sustained release formulation. But equally, it is not a sufficient disincentive to rule out fluvastatin altogether as a candidate for sustained release formulation.
  402. Mr Arnold's third reason relates to short half life. Again I have discussed this already at paragraphs 231 above concluding that there would be concerns but not so serious as would rule fluvastatin out as a candidate.
  403. Mr Arnold's fourth reason relates to cumulative effect. Again I have discussed this already at paragraph 109 above concluding that this is a disincentive. It is, however, only a factor and certainly not enough of itself to rule out fluvastatin as a candidate
  404. In my judgment, Actavis establishes that there was a real motivation for the skilled team to develop a sustained release formulation of fluvastatin. I do not go so far as to agree with Mr Wyand that the motivation was overwhelming.
  405. The next question to be addressed, a very important question indeed, is whether the formulator would have expected to be able successfully to develop a sustained release formulation of a water-soluble salt of fluvastatin.
  406. In summary, Mr Arnold submits that on the evidence, the answer to this question must be that (i) the formulator would not have considered that it was impossible to formulate fluvastatin in a sustained release formulation, (ii) the formulator would have considered that there were a number of difficulties to be overcome and (iii) the formulator would not have been confident of success.
  407. In this already over-long judgment I have considered all the suggested problem areas and reached some conclusions. I can deal with the difficulties which Mr Arnold submits would have faced the skilled team and see whether they would have been easily overcome.
  408. Water solubility: I have already dealt with this exhaustively and conclude that absolute water solubility was only a problem at the extremes. High or low (non-extreme) solubility would limit the available technology but whether one takes Dr Rue's position (confident of success) or Professor Collett's position ("should not make it impossible", an answer which in context did not mean he was not sure whether it could be done, but only that it could be difficult) the result is the same, which is that absolute water solubility was not a problem which would prevent the claims of the Patent being obvious.
  409. The pH dependent solubility of fluvastatin: this is another aspect I have already dealt with at some length. It has not been suggested that this is the case with fluvastatin. The skilled team would have known that fluvastatin did not fall within the extreme case or alternatively would have had no reason to think that it did.
  410. In that context, the pH-dependent solubility of fluvastatin was never explicitly put to Dr Rue. He was only cross-examined on pH-dependent solubility as a matter of generality, and never in the context of fluvastatin. Accordingly, Mr Wyand submits that this is not a matter on which Novartis should be permitted to rely. I do not agree. The point was not a new one which came to light only in Professor Collett's cross-examination – albeit that he did emphasise the point a number of times – it is to be found at paragraph 75 of his first report and in paragraph 19 of his second report. In any case, as Dr Rue said, pKa, absorption and solubility are all interdependent factors which must be taken together. The thorough cross-examination of Dr Rue certainly dealt with the pKa value and addressed the variable ionisation state as the drug moves through the GI tract.
  411. However, even if Novartis is allowed to rely on the point, it does not follow that it is a good one. The evidence shows that the pH-dependent solubility would only cause a problem if it varied from very soluble to very insoluble over the relevant range. That was Dr Rue's evidence. I think that Professor Collett agreed; that, at least, is how I understand the evidence which, for reference, is to be found in the transcript for Day 4 of the hearing from pages 472 to 479. But even if I have misunderstood what he was saying, I do not think that there is enough to establish that the formulator would have expected that he would not be able to achieve success.
  412. Stability: again, this has been discussed to some extent. Professor Collett considered that the instability of fluvastatin would have been a concern. Dr Rue disagreed with this on the basis that this did not appear to be a problem with the immediate release formulation. It was put to Dr Rue that this was because the immediate release form of fluvastatin in 1996 had an enteric coating. On the assumption of the truth of that assertion he accepted that that is why the stability of the immediate release tablet in the stomach would have been seen as a problem. But there is, on the evidence, no truth in that assumption and therefore Dr Rue's agreement was obtained on a false premise. To be fair to Mr Arnold, he does not rely on this (non-)point in his closing submissions.
  413. Instead, he relies on the proposition that the immediate release formulation contains a stabiliser. The stabiliser employed in the immediate release formulation (a combination of sodium bicarbonate and calcium carbonate) was found by Novartis not to be suitable for its sustained release formulation. The stabiliser in fact used by Novartis for the sustained release formulation (potassium bicarbonate) also acts as a micro-environmental pH buffer, but this was not common general knowledge in 1996 and Professor Collett did not know whether Novartis was aware of this. I do not understand how this buffering effect is to be taken into account; Professor Collett said the chemical was put there as a stabiliser not as a buffer.
  414. pKa: again, this has been discussed to some extent. However, I must add to what I have already said that Professor Collett considered that the changing ionisation state of fluvastatin would have been regarded as problematic. Dr Rue's view was that this was "not necessarily" so, explaining that what Professor Collett was referring to in his report was a system in equilibrium. Mr Wyand put the following to Professor Collett:
  415. "Q. But you heard Dr Rue say that the partition coefficient as such is not that relevant because what you have is you have the drug going through the gut wall so you are constantly having to pull, as it were, through the gut wall?
    A. Yes, that is what I was saying….It is a dynamic situation."

  416. Accordingly, Mr Wyand says that having a pKa of 5.5 is not necessarily a problem, because the intestinal system is not in equilibrium – the rate of absorption will affect the rate at which ionized fluvastatin becomes un-ionized. Nonetheless, I accept Professor Collett's view that the changing ionisation state of fluvastatin would have been regarded as problematic, not so problematic as to put the skilled team off, but one of the factors that would make achievement of success unpredictable.
  417. Partition co-efficient: again, this has been discussed to some extent with mention of Dr Rue's evidence that it is a problem in extreme cases. However, Professor Collett considered that the changing partition coefficient of fluvastatin would, once again, have been something regarded as problematic. Dr Rue, when questioned on the relevant paragraph of Professor Collett's report accepted that whether the drug was absorbed further down the gut was an unknown, although he added that it was well known that fluvastatin was well-absorbed from an immediate release tablet. Professor Collett said nothing which could be taken as qualifying the view expressed in his reports. Mr Wyand says that the partition coefficient was not, in any case, established on the evidence. I think that it was established, and I am willing, on the evidence, to accept the figure of 1.5, a figure accepted by Dr Rue. In any event, Professor Collett explained that what was more important than the precise figure was the pH-dependency of the partition coefficient. As with pKa, I conclude that this the partition coefficient would be seen as possibly problematic, not so problematic as to put the skilled team off, but one of the factors that would make achievement of success unpredictable.
  418. Absorption: Professor Collett considered that the variable absorption of fluvastatin would have been regarded as of concern. Dr Rue agreed that, if one was considering the table of characteristics set out in Chang and Robinson (for reference Lieberman, Lachman and Schwartz: Pharmaceutical Dosage Forms 1990 Chapter 4) and one was on the wrong side of the line so far as water solubility, pKa and absorption was concerned, then one's conclusion was probably going to be that this was an unsuitable candidate. But I am not sure where that gets Mr Arnold: the wrong side of the line would mean (i) extremes of aqueous solubility or pH-dependent solubility, particularly in the physiological pH range (ii) drugs that are very lipid or very water soluble ie extreme in partition coefficient and (iii) slow absorption or variable absorption. But only (iii) was the case for fluvastatin. In any case, according to Dr Rue, it was only extremes of absorption and variability of absorption which were problematic, evidence which I accept. But taking the related factors of pKa, partition co-efficient and absorption together, the conclusion must be that the skilled team would have perceived problems which would make the prospect of success uncertain.
  419. Perhaps more important than all of this detail are the overall views of Dr Rue and Professor Collett about the prospect of success in formulating a sustained release dosage form of fluvastatin. After Mr Arnold had taken Dr Rue through what he, Mr Arnold, considered to be disincentives (discussed above) the following exchange took place:
  420. "Q. It follows that if the manager of the team were to put the formulator on the spot and say, 'Mr. Formulator, if I give you a budget to develop a sustained release formulation fluvastatin, are you going to be successful?' the formulator would be bound to reply, would he not, 'I cannot guarantee that'?
    A. The formulator -- can we be clear about what the formulator is? Are you asking me whether the pharmacist who is developing the tablet, the dosage form, would be able to say, 'I will be able to give you a tablet that will dissolve at the appropriate rate', or are you asking me whether the team itself would be successful in developing a product, taking it into the clinic to show that there would be benefits?
    Q. The team could not assure their manager that they would be successful, could they?
    A. No, the team could not.
    Q. And what they would have to say is, 'Well, it is uncertain'?
    A. Yes."

  421. Mr Arnold says that that shows Dr Rue's opinion to be that success was uncertain. Defining success in the sense of "taking it into the clinic to show that there would be benefits" he is of course correct. But what Dr Rue did not accept – he was not asked – was whether the formulator would expect to be able, if asked to do so, to produce a tablet that will dissolve at the appropriate rate in vivo. It is quite clear from his evidence that he considered that this could be virtually guaranteed in vitro and that there would be a high expectation of success in vivo.
  422. Professor Collett, for his part, was not prepared to admit to a strong expectation of success in producing a formulation with the required dissolution rate in vitro. He would not even "reckon he could do it" in vivo that is to say to produce a formulation of fluvastatin which would release over a period longer than 3 hours in vivo regardless of any therapeutic effect; nor would he even agree to the proposition that "there might be a problem but I would be pretty confident that I would be able to do it". This was the one aspect of Professor Collett's evidence where I think he really was being over-cautious. Or, to put it another way, the sensible and realistic reservations which he actually held (ie that he could not guarantee anything; there could always be problems which had not been foreseen) were expressed in this part of his evidence in language which expressed rather greater reservations than were justified.
  423. I take quite shortly three further points which Mr Arnold relies on:
  424. a. Dr Rue agreed with Professor Collett that, if the skilled team were deciding how to formulate fluvastatin, it was overwhelmingly likely that they would decide to formulate it as an immediate release formulation. I think that he only agreed that in relation to a new drug where it is clearly correct that that would be the first approach. But it does not begin to answer whether other approaches would or would not be obvious, particularly where there is already a well-known immediate release formulation on the market.
    b. Professor Collett was unshaken in his opinion that, if the team were pressed to change the release profile of fluvastatin, the simpler course would be to change the salt. In that way release could be slowed down without changing from an immediate release formulation. Dr Rue agreed that a change of salt was something which was sometimes done, although he regarded it as uncommon. He also agreed that using a prodrug was an approach that was repeatedly mentioned in the standard textbooks, but his view was that it was rarely used. Professor Bycroft's evidence that the medicinal/pharmaceutical chemist would consider changing the salt or a prodrug or a structural analogue was unchallenged. I accept that some people would do what Professor Collett says; equally I accept that there are others who, as Dr Rue says, would not. In either case, I do not consider that the other routes available show that the formulation of fluvastatin in a sustained release dosage form is not obvious, or also obvious. In any case, if one of the motivations for a change in the release profiles is to enable an 80 mg dose to be given in one go, there is nothing to suggest that either a different salt or a prodrug would achieve that result any more or less easily that using the existing salt.

  425. Although I have referred to commercial obviousness a number of times in this judgment, I do need to say more about it. Mr Arnold has included a number of detailed submissions on this topic towards the end of his written closing submissions. I am not going to deal with them in detail but limit myself to the following:
  426. a. Mr Arnold describes Actavis' new case on obviousness as an attempt to argue that, even if it was not technically obvious to develop a sustained release formulation of a water soluble salt of fluvastatin, it was a commercially obvious step to take. He says that that is legally erroneous: the question is whether the invention is technically obvious to the relevant skilled person. I agree with that.
    b. But, to be fair to Mr Wyand, I do not think that his case on commercial obviousness is quite as Mr Arnold categorises it. He does not, at least as I understand his case, say that there was a compelling commercial driver to develop a sustained release formulation and that therefore such a formulation was obvious. Clearly, such an argument could not hope to succeed, at least unless there was a real expectation of success, an expectation sufficient to pass the "obvious to try" test and to defeat all the apparent lions on the path. Rather, he relies on the commercial driver to establish motivation and thus to satisfy the requirement which Mr Arnold says exists to show motivation. Of course, Mr Wyand says that motivation is not, in fact, a necessary element, but that is a different matter which I will address later.

    c. Mr Arnold says that, in any case, the new case is evidentially unsupported. He points out correctly that Actavis did not call any expert with expertise in marketing pharmaceuticals to say that it would be commercially obvious to develop a sustained release formulation; and says that, if it had, Novartis would have considered calling its own expert. He also points out that Actavis did not even adduce any evidence in chief to this effect from either Dr Rue or Professor Durrington or advance any such case in its skeleton argument. The point appears to have started with the reference to commercial need made by Dr Rue during the course of cross-examination. I agree with Mr Arnold when he says that the new case is a late afterthought.

    d. Nonetheless, I do not think that I should ignore certain aspects of what might be seen as the commercial case. Mr Arnold himself relies on lack of a commercial driver as part of his own case. He says that no-one is going to put time, money and effort into developing a new formulation if it produces no improved therapeutic benefit. But this, surely, is because there would be no (commercial) point in doing so since formulators work in the real world and do not produce new products for fun. I think that Mr Wyand is entitled to rely on commercial incentives the other way. In particular, the reduction in dosing from twice a day (2 x 40 mg immediate release) to once a day (1 x 80 mg sustained release) would have been perceived as an advantage by the skilled team in terms of patient compliance (although not by Professor Horsmans or even Professor Durrington) and whether one sees that as a therapeutic benefit or a commercial benefit it is one which I consider that I can take account of in assessing motive.

  427. In the light of the totality of the evidence, the essentials of which I have discussed above, I reach the following conclusions:
  428. a. The skilled team is to be seen as engaged upon the task of formulating fluvastatin for use in humans for the treatment of hypercholesterolemia. There is no suggestion that the team would be formulating the drug for the treatment of different animals or for the treatment of any other condition.
    b. Part of the common general knowledge of the team includes the fact that an immediate release formulation was already available. That formulation could be taken at a dose of up to 40 mg once a day and up to 80 mg per day in two doses.

    c. Whether or not the team was expressly directed to consider formulating fluvastatin in a sustained release dosage form, it would at least consider such a formulation as a possible way forward and not dismiss it out of hand.

    d. The skilled team would seek to identify the benefits (or incentives) and problems (or disincentives) of a sustained release formulation.

    e. The team would regard improved patient compliance as a benefit which would certainly be obtained if a satisfactory formulation of an 80 mg dose could be successfully achieved.

    f. The team would be concerned about clinical efficacy. It would be uncertain about whether improvement could be achieved. It could not be ruled out. But it certainly could not be said with any confidence that any improvement would be achieved, nor that there was a strong expectation that it would be achieved.

    g. The team would also be concerned with side-effects. Again, it would be unable to predict with anything approaching certainty that any reduction in the risk of side-effects would be achieved, but there would be some hope of reduced risks of hepatotoxicity (on a population basis as explained above) and a reasonable expectation of improvements in myopathy. However, neither of these was seen as serious problems with fluvastatin at the priority date.

    h. The team would have, as a result of common general knowledge and secondary common general knowledge, an expectation of being able to develop an 80 mg sustained release formulation which released over at least 3 hours in vivo and would be confident that it would have some clinical efficacy. But it would have no confidence that it would achieve better, or even the same, efficacy as the existing maximum dosage of 2 x 40 mg daily immediate release. It would expect, but could not be certain, that this would not produce a greater risk of side-effects than the existing maximum dosage.

    i. It would have been rational, from a technical perspective, for the team to have produced a sustained release formulation for 80 mg once daily dosage with a view to improved patient compliance, anticipated reduction in side-effects and possible improved efficacy. But in the light of the uncertainties, such a course might well be seen as commercially unjustifiable; but that would be a matter for the decision of commercial people not of the skilled team.

    j. The team, if it were asked to do so, would be confident of producing a sustained release formulation of fluvastatin at various doses which had the release characteristics in vitro described in the Patent. It would be confident that such a formulation could be taken by humans and would have some therapeutic effect albeit perhaps of lesser efficacy than the equivalent dosage taken in immediate release form as a single dose (up to 40 mg daily) or two doses (up to 80 mg daily). There would, however, be no reason for producing such a formulation for its own sake if it were thought impossible or highly unlikely that such a formulation would have, or would lead to a product which would have, improved efficacy and/or reduced side-effects.

  429. It follows from those conclusions that even if the skilled team might have had a motive for developing a sustained release formulation of fluvastatin, it had nothing approaching certainty or even a strong expectation that it would achieve success in the sense of producing such a formulation which releases in vivo over a period of 3 hours or more and which had improved clinical efficacy (other than improved patient compliance) or the same efficacy with a reduced risk of side-effects. Its prospects of success in the project which it is to be assumed is being undertaken can, at best, be described as uncertain.
  430. It also follows, from paragraph j. above, that the skilled team would have no separate motivation simply to produce a formulation which released over 3 hours or more in the paddle test. There would be no reason to do so unless it were thought that the resulting formulation dosed at 80 mg daily might also produce the benefits for which the team is searching and in relation to which, for reasons already given, success was uncertain.
  431. Unless the claims in the Patent can be said to be obvious over the common general knowledge quite apart from the motivation for, and prospects of success of, the project with which the skilled team is notionally engaged (ie as described in paragraph 312.a above) the inevitable result of these conclusions must, in my judgment, be that the claims in the Patent are all non-obvious. Applying the approach set out by Kitchin J in Generics v Lundbeck, I consider the prospect of success of the project is the most important single factor; that prospect was, I have held, uncertain. Further, although there was a motive for developing a sustained release formulation, the only benefit which would have been perceived as certain was improved patient compliance; a reduction in the risk of side-effects also provided a motive, but the risks were not, even in 1996, regarded as great. Moreover, although there can be more than one obvious route, the fact that there were alternative lines of development is a factor which is to be taken into account is assessing obviousness.
  432. I do not consider that Actavis can successfully argue that claims in the Patent are obvious because (i) the skilled team would have embarked on the development project and (ii) if it had done so, it would inevitably have made the invention. There are two reasons why it cannot argue in that way.
  433. The first is that this is, in my judgment, a legally inadmissible argument. Such an argument could, in my judgment, run only if it was obvious (in the statutory sense) that the team would meet with success in its project. It is not enough that the team would (whether possibly, probably or even virtually certainly) have embarked on the project; rather, it must be obvious for it to do so in the same sense that something may be "obvious to try". Whether or not Mr Wyand is correct in saying that a fair expectation of success is enough, I do not consider that, on the facts of this case, the skilled team could have had even that measure of confidence. The second reason is that it is not established on the facts that the skilled team would inevitably have made the invention.
  434. Can the claims in the Patent then be said to be obvious quite apart from the notional project? This takes us back to the 5¼" plate on which Mr Wyand relies. The reason that the plate (assuming that it is novel in the first place) is obvious over the prior art is because it involves no technical advance (whereas a 10 foot plate might well do so in the light of the different structural strength required). The person skilled in the art would see the differences between the new plate and the old plate as involving only steps which were obvious. It does not require a motive for making the new plate before it can be said that the allegedly inventive steps are obvious.
  435. I do not think that, in a case of that sort, it is a requirement that the person skilled in the art "would" have taken the steps in question. I do not consider that the cases in this jurisdiction or in the Board of Appeals prevent that conclusion. In particular, the "could-would" approach of the Board – at least insofar as is revealed in the summary with which Mr Arnold has provided me – is directed at a case where there is some identified underlying technical problem or where there is some identified improvement or advantage. It may then be appropriate as a general rule to ask whether the skilled person, addressing that problem or attempting to obtain that benefit, would, rather than could, carry out the invention. But where there is no problem and no improvement or advantage which is sought to be addressed by the patent, it is essential to go back to the words of the statute and simply ask whether the claimed invention is obvious to the person skilled in the art having regard to any matter which forms part of the state of the art. In the case of the 5¼" plate, there is no problem and no advantage which a proposed patent would address and the "could-would" question is not, in my view, a relevant question to ask.
  436. In the present case, the situation is different because the Patent is on its face seeking to deal with a problem, namely the high water solubility of fluvastatin. The Patent claims that, surprisingly, fluvastatin sodium exhibits particularly favourable release characteristics. The evidence, however, shows that the high absolute solubility of a drug is a problem only at very high and very low levels. The evidence also shows that the high (but not very high) solubility of fluvastatin would not have been seen as a serious problem at least in vitro; it would only have been seen as limiting the available technology for producing a sustained release formulation releasing in vitro over a period of 3 hours or more. This depends, of course, on secondary common general knowledge: the actual solubility of fluvastatin would not have been part of the common general knowledge of the team but would have been learned only as a result of the pre-formulation investigations which it is common ground would have been carried out. But that does not make any difference: the skilled team embarking on a formulation project for fluvastatin would, as a matter of course, obtain information about its solubility which is to be treated, for the purposes of assessing obviousness, in the same way as actual common general knowledge.
  437. Neither claim 1 nor claim 3 indicates the use of a technology which would not have been seen as an available technology. Claims 1, 2, 3 and 10 are not restricted to any particular dosage level nor in the way in which the different formulations mentioned might be used. Accordingly the claims cover formulations which the skilled formulator would have been able (in the light of the common general knowledge and secondary common general knowledge) to formulate using some appropriate technology within each claim had he been asked to do so; and since the dosage is not specified, the skilled team (including clinician) would be confident that some level of dose could be given in sustained release form which it would also be confident (i) would have some efficacy (perhaps not as good as the existing immediate release formulations) and (ii) would not produce unacceptable side-effects.
  438. In those circumstances, it seems to me that the only argument against the obviousness of claim 1 is that, in fact, nobody – whether clinician, formulator or commercial boss – would in fact have thought of actually producing a formulation within claim 1 as an end in itself; there would simply have been no motivation to do so. To put that another way: nobody would have asked the skilled formulator to produce such a formulation and he would not have thought of doing so himself without a request.
  439. In my judgment, this absence of motivation does not prevent claim 1 from being technically obvious. The skilled team would, for reason already given, not reject a sustained release formulation out of hand. Even though its focus will be on the end result of an 80 mg once daily formulation having improved efficacy and less serious side effects, coupled with better patient compliance, the concept of sustained release will be present. It does not need a commercial boss to ask the question "Can you make a formulation [within claim 1]" for the skilled formulator to appreciate its technical obviousness any more than it needs a commercial boss to ask a similar question of the plate designer for that designer to appreciate the technical obviousness of the odd-sized plate. I conclude that claim 1 is obvious over the common general knowledge.
  440. I confess to finding this a very finely balanced issue. This is particularly so given that the necessary state of knowledge of the skilled team depends on secondary common general knowledge and that by that stage of the analysis the skilled team's focus, as I have said, will be on the end product. Further, it is implicit that I reject any suggestion that the conclusion on obviousness is based on hindsight. It is not always easy to put the benefit of hindsight to one side; and in the present case I am conscious of the influence which it might have. Different minds might take different views but my conclusion is as I have stated.
  441. Accordingly, claim 1 is invalid for obviousness over common general knowledge.
  442. As to claims 2 and 3, Mr Arnold accepts that claim 2 falls with claim 1. However, he says that Claim 3 stands in a different position.
  443. Claim 1, it will be remembered, is a sustained release composition selected from the group consisting of matrix formulations, diffusion-controlled membrane coated formulations and combinations thereof. Claim 3 is a composition according to claim 1 which is an eroding matrix formulation.
  444. Dr Rue's evidence on the question of solubility was that the formulator would not think that the solubility of fluvastatin sodium was so high as to make it an unsuitable candidate for a sustained release formulation at all, but rather would think that it was unsuitable for some types of sustained release formulations and that this was a problem which could be solved by using a diffusion-controlled membrane formulation. Consistently with this, a membrane was suggested to Professor Collett in cross-examination as the first choice with an osmotically-controlled system being the second. Even if claim 1 is obvious, this evidence, it is submitted, shows that claim 3 is not obvious.
  445. Whilst I accept what Mr Arnold says about Dr Rue's evidence, what Dr Rue said more than once was that the technology would be limited, without stating precisely how; and his evidence that the problem could be solved using a diffusion-controlled membrane was in answer to a direct question by Mr Arnold by reference to that formulation. It cannot be taken as an acceptance that there would be a problem with an eroding matrix formulation (within claim 3). Further, in one of the passages to which Mr Arnold refers, Dr Rue, in quoting from one of the books to which he was taken (Langer & Wise: Medical Applications of Controlled Release: Vol II: Applications and Evaluation) accepted that with matrix formulations a high ratio of insoluble matrix to drug may be required, but he did not accept that it was difficult let alone impossible. The passage which he quoted, and on which Mr Arnold seems to rely, does not support the proposition that an eroding matrix formulation would be difficult let alone impossible, only that it would be more difficult than a diffusion-controlled membrane.
  446. Moreover, the alternatives put to Professor Collett were not ranked in order of preference but simply listed in a particular order; an eroding matrix was mentioned as a third possibility in the same passage of his cross-examination as that referred to by Mr Arnold.
  447. In my judgment, claim 3 falls for the same reasons as claims 1 and 2. I do not consider that it is possible to draw the distinction which Mr Arnold seeks to draw.
  448. So far as claim 10 is concerned, the use of fluvastatin to treat hypercholesterolemia was, of course, very well known and certainly part of the common general knowledge of the skilled team. It would have been obvious, in my judgment, to the skilled team that some dosages within claim 1 would have had some therapeutic effect in the treatment of that condition although it would not be known how any particular dose would have compared with any particular dose or doses of an immediate release formulation. Such comparisons form no part of claim 10. In my judgment, claim 10 is also invalid as being obvious over the common general knowledge given the invalidity of claim 1.
  449. These conclusions make it unnecessary to consider obviousness of Kathawala and Scolnick. I should nonetheless consider them.
  450. Kathawala

  451. I have set out the relevant passages at paragraphs 44 and 46 above. Mr Wyand submits that it can be seen that Kathawala expressly teaches the idea of making sustained release compositions; and the specification refers to a typical oral dose of fluvastatin sodium being taken three times a day.
  452. He then identifies the difference between Kathawala and the inventive concept in this way: Kathawala does not expressly disclose the making of a sustained release composition using fluvastatin sodium.
  453. This is said not to involve any degree of invention because (a) Kathawala teaches that the compounds of the invention can be formulated in one of two ways (immediate release or sustained release) and (b) there can be no invention in making a sustained release composition with fluvastatin sodium, one of the best performing compounds in the three biological assays.
  454. Mr Arnold says, in effect, that this is nonsense. He says that it is pure hindsight to read obviousness into the reference to sustained release in the passage cited at paragraph 46 above, and that it would not begin to be obvious to the skilled team on reading Kathawala without knowledge of the Patent to make a sustained release formulation of fluvastatin at all, let alone one falling within the claims of the Patent for the following reasons:
  455. a. Kathawala is a patent which is primarily addressed to a pharmaceutical chemist (or even a synthetic organic chemist). Its focus is on the novel compounds disclosed and methods of synthesising them. It does not purport to disclose any new ways of formulating such compounds. It contains nothing that would be of interest or use to a skilled team formulating fluvastatin, and in particular the formulator, since (subject to the points made below) it contains nothing about formulating the compounds of Formula I except banal generalities that the formulator would be well aware of from his common general knowledge.
    b. It contains a laundry list of possible dosage forms, dosage sizes and routes of administration. It gives the reader no incentive to adopt any one of these possibilities as opposed to the others.

    c. The reference to administration in sustained release form is simply a throw away suggestion in the course of reciting a whole series of other possibilities.

    d. There is nothing to suggest to the skilled team that a sustained release formulation of Example 8 or any other compound of Formula I is either desirable or feasible or how to go about formulating it. There is obvious force in all of that although, of course, it might be that all formulations in the laundry list are obvious

  456. Mr Arnold reminds me that Dr Rue agreed that Kathawala would tell the formulator nothing about formulating the compounds of Formula I that he was not already aware of from his common general knowledge. It follows, he says, that Kathawala is hopeless as a starting point for an obviousness attack.
  457. I agree with that conclusion. In my judgment the claims of the Patent are not obvious over Kathawala if I am wrong in my conclusion concerning obviousness over the common general knowledge.
  458. Scolnick

  459. I have referred to the relevant passages at paragraphs 47 to 50 above. Scolnick teaches the use of the Lovastatin, Simvastatin, Pravastatin and Fluvastatin in the prevention and treatment of Alzheimer's disease.
  460. Mr Wyand submits that Scolnick expressly teaches the idea of making sustained release compositions; although the clinical data referred to in Examples 1 and 2 involve the use of Lovastatin and Simvastatin, there is a clear teaching that other statins can be used, including Lescol (fluvastatin sodium). I do not think that that is correct reading of Scolnick. The passage referring to time-controlled release vehicles is expressly referring to specific compounds (in fact lovastatin and simvastatin); there is nothing stated expressly in Scolnick about a suitable formulation for fluvastatin. Following that passage, in a new paragraph, reference is made to the Examples: Example 2 describes a protocol using simvastatin, but stating that other HMG-CoA reductase inhibitors could be substituted. There is nothing to suggest that the Example teaches the use of a sustained release formulation. I do not consider that there is any clear teaching of the idea of making a sustained release formulation of fluvastatin. Accordingly, the claims of the Patent are not obvious over Scolnick.
  461. However, that is not all. Mr Wyand points out that, although Scolnick is addressed to the treatment of Alzheimer's disease, it is dealing with the administration of the statins which are used for the treatment of hypercholesterolemia and were well-known for that treatment. He also says that a relationship between the diseases was shown by Professor Durrington's answers in cross-examination: that is true, but it cannot be said that such a relationship was part of the common general knowledge of the skilled formulation team.
  462. Mr Wyand identifies the difference between Scolnick and the inventive concept in this way: Scolnick does not expressly disclose the making of a sustained release composition using fluvastatin sodium.
  463. This is said not to involve any degree of invention. Actavis' case is that Scolnick teaches that the fluvastatin sodium is one of a small number of compounds which can be formulated in a number of ways. Oral dosage forms are preferred and these can be sustained release. There can be no invention in carrying out such teaching.
  464. As with Kathawala, Mr Arnold says, in effect, that this too is nonsense. Again, he says that it is pure hindsight and that it would not begin to be obvious to the skilled team on reading Scolnick without knowledge of the Patent to make a sustained release formulation of fluvastatin at all, let alone one falling within the claims of the Patent for the following reasons:
  465. a. Scolnick is a patent application which is addressed to a clinician who is treating patients with Alzheimer's disease. It does not purport to disclose any new compounds. On the contrary, it expressly acknowledges at p. 3 ll. 32-35 that lovastatin, simvastatin, pravastatin and fluvastatin are known cholesterol lowering agents and at p. 4 ll. 28-33 that these and related compounds are disclosed in the prior art. Rather, the disclosure is of a second medical indication for a broad class of these known compounds represented by Formula (I). Nor does Scolnick disclose any new formulations of such compounds. It contains nothing that would be of interest or use to a skilled team formulating fluvastatin, and in particular the formulator, since it contains nothing about formulating the compounds of Formula (I) except, again, banal generalities that the formulator would be well aware of from his common general knowledge.
    b. The passage referring to the different possible formulations is another laundry list of possible dosage forms, dosage sizes and routes of administration. Scolnick gives the reader no incentive to adopt any one of these possibilities as opposed to the others.

    c. The reference to administration in time-controlled release form is again a throw away suggestion in the course of reciting a whole series of other possibilities. There is nothing to suggest to the skilled team that a sustained release formulation of fluvastatin or any other compound of Formula (I) is either desirable or feasible or how to go about formulating it other than by listing a number of standard techniques.

    d. The statement at p. 11 ll. 13-14 that the dose may be given in a single dose or in two to four divided doses would again point towards an immediate release formulation.

  466. Mr Arnold makes the same point that Scolnick adds nothing to the common general knowledge. Again I agree. This is another reason why the claims of the Patent are not obvious over Scolnick if I am wrong in my conclusion concerning obviousness over the common general knowledge. Moreover, there is nothing in Scolnick which would solve the problems addressed above in relation to obviousness over the common general knowledge; Scolnick provides no additional motivation to develop a sustained release formulation of fluvastatin.
  467. Insufficiency

  468. Mr Wyand now puts Actavis' case on insufficiency this way. Professor Collett identified the distinction between the prior art disclosure (which he described as much like a shopping list at a supermarket) and the Patent as proposed to be amended with the instruction that the release was to be sustained over a period of more than 3 hours. Such an instruction is not, according to Mr Wyand, contained in the unamended patent and Professor Collett was not convinced that the diclofenac example in Figure 3 described as "sustained release" did release over a period of more than 3 hours.
  469. I do not understand why this would amount to insufficiency even if it were correct. As Mr Arnold says, it is manifest that, in its common general knowledge meaning, the expression "sustained release formulation" is a relative term: it refers to a formulation which releases the active ingredient over an extended period, that is to say, over a longer period than an immediate release formulation.
  470. But in fact I have found, as a matter of construction, that the Patent does tell us that the sustained release must be over a period of more than 3 hours. Accordingly, I do not consider there is anything in this argument. I should add that a rather different argument in Actavis' skeleton argument was not pursued in closing. The artificiality of this objection is confirmed by the fact that Actavis had no difficulty in determining, and admitting, that its product falls within claims 1 and 10 and specifically that it is a sustained release composition which releases its active ingredient over more than 3 hours.
  471. Amendment

  472. In the light of my conclusion concerning the meaning of "sustained release" in the Patent, the proposed amendment simply clarifies, but does not change, the meaning of the Patent in any away. The amendment, on that basis, clearly does not add matter. I allow the amendment accordingly.
  473. Conclusion

  474. Actavis' attack on claims 1, 2, 3 and 10 of the Patent of obviousness succeed. Its attack based on insufficiency fails. There is no infringement. Novartis' claim to amend the Patent succeeds.


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/ew/cases/EWHC/Patents/2009/41.html