Christopher Floyd QC
Introduction |
1 |
The Patent |
14 |
The Claims |
27 |
The issues |
30 |
Obviousness |
31 |
Law |
31 |
The expert witnesses |
37 |
Construction and the Inventive concept |
41 |
The skilled addressee |
44 |
The common general knowledge |
50 |
Obviousness over common general knowledge |
55 |
Secondary evidence |
74 |
Conclusion on obviousness over common general knowledge |
84 |
The prior art |
86 |
Conclusion on obviousness over the prior art |
92 |
Sufficiency |
93 |
Law |
93 |
The pleaded case |
96 |
What is required for sufficiency in the present case? |
99 |
The evidence of insufficiency |
101 |
Does the Patent teach a' principle of general application? |
106 |
Conclusion |
110 |
Introduction
- This is a case about medicines for cattle and other animals. Schering-Plough, to whom I shall refer as "S-P", apply to revoke European Patent (UK) No. 0 980 248 ("the Patent"). The Patent belongs to Norbrook Laboratories Limited ("Norbrook"). It is entitled "Improvements in or relating to long-acting antimicrobials". It has an application date (without benefit of Convention priority) of 6th May 1997.
- The undisputed technical background necessary for an understanding of the issues in the case is as follows. The invention is concerned with a veterinary medicine which is a combination of two separate active ingredients: an antimicrobial and an anti-inflammatory. The combination can be used to treat a range of infectious diseases in animals, including bovine respiratory disease, ("BRD").
- Bovine respiratory disease ("BRD") is a common disease, particularly in young calves reared in close quarters under intensive farming methods. It may be viral or bacterial in origin. When a calf gets BRD the main symptoms are that its temperature goes up (the medical term is "pyrexia") and it shows respiratory distress ("dyspnoea"). It may also show loss of appetite, a raised respiratory rate and increased lethargy.
- The term "antimicrobial" covers at least two classes of drugs: those which act by killing microbes (bactericidal) and those which act by preventing microbe replication (bacteriostatic). It covers the natural and synthetic penicillins as well as numerous classes of the more modem synthetic agents. Antimicrobials will not kill viruses: so, if an infection is viral in origin, antimicrobial therapy is only effective against secondary bacterial invaders.
- It was well known at the priority date to make "long-acting" versions of antimicrobials. A long acting effect can be achieved by the inherent properties of the drug, by the manner in which it is formulated (in the simplest case simply by administering a higher dose) or by a combination of these. By making long-acting veterinary antimicrobials, the number and frequency of doses given to the animal can be reduced, sometimes to a single dose. The advantages of such a dosage regime over more frequent dosing were obvious and were also well known. First, the animal needs to be caught and restrained each time a dose is administered. This results in stress for the animal and work for the vet or farmer: a long acting dose reduces the number of times this needs to be done. Secondly, with a long-acting formulation, the level of the drug in the animal's blood is retained: repeat dosing introduces the possibility that the blood levels may drop below a therapeutic level.
- Antimicrobials do not directly reduce the inflammation of the tissues which is caused by and accompanies infection. Instead, if acting on their own, they will simply remove the cause of the infection, and the inflammation will be left to recover on its own with the aid of the animal's own defence system. To treat inflammation, there is a well known class of drugs called anti-inflammatory agents. These drugs tackle the inflammation associated with infection, rather than removing the pathogen which has caused the infection ..
- The inflammatory response seen in the animal suffering from BRD is in fact the result of the animal's own defence system tackling the infection. Thus, although clinicians' views might have differed in 1997, suppressing the inflammatory response in all cases might not be productive. However, anti-inflammatories can make the animal feel more comfortable while it is ill, and cause an earlier return of appetite. Moreover the inflammation can itself be life threatening, or cause long term tissue damage to the animal, and in severe cases of infection may need to be reduced while the antibiotic tackles the underlying cause. For this reason it was in fact common in 1997 for vets to administer both an antimicrobial and an anti-inflammatory concurrently, at least in more severe cases of infection. Of course the anti-inflammatory would never be used as the sole agent to treat BRD.
- Several antimicrobial drugs for use for treatment of BRD were known. Of these the most commonly prescribed was oxytetracycline. This was available in both long and short acting formulations, with dosage for up to 5-6 days. Another well known drug was florfenicol, dosed at day zero and again at two days. This too was available in short and long acting forms. And there were others still.
- The anti-inflammatory drugs included steroids such as cortisone ("corticosteroids") and the non-steroidal anti-inflammatories ("NSAIDs"). The familiar human headache pills, paracetamol and ibuprofen, belong to this latter class. NSAIDs were known to have three effects: anti-pyretic (reducing fever, temperature); analgesic (pain-killing) and anti-inflammatory. Flunixin meglumine ("flunixin") had been licensed as an anti-inflammatory for respiratory disease in animals since before 1984 and was the best known veterinary NSAID in 1997. It was indicated for daily administration for up to 5 days, and was frequently administered concurrently with antimicrobials such as oxytetracycline.
- Medicines such as oxytetracycline and flunixin would normally be administered to a calf by injection ("parenterally"). Injections can be given ·into the vein ("intravenously") or into the muscle ("intramuscularly"). If a dose is given intravenously the drug will reach its maximum plasma concentration very rapidly. In intramuscular dosing it will take longer, and the release of the drug into the blood can be controlled by the so-called "depot effect", making it longer-acting. A depot of the drug is formed in the tissue which releases over time. A downside of this is that the animal can experience irritation at the injection site, and its meat must be withheld from sale for a longer period than with intravenous administration. Flunixin was licensed for intravenous, but not intramuscular dosing.
- Combination products are single products designed to administer two therapeutic agents in the same dose. Combination products are to be distinguished from concurrent therapy in which the two drugs are administered separately but at the same time. There were arguments for and against combination products. In the veterinary context such a product obviously saved the need to administer a double injection to the animal, reducing the stress it experiences. In the context of treating an entire herd, the saving in work for the vet could be significant. On the other hand there is . a disadvantage. Combination products lock the two individual components together, reducing flexibility.
- In 1997, in veterinary medicine there were relatively few combination products on the market, although some combinations of antimicrobials existed which were intended to combine the spectra of pathogenic activity of two narrower spectrum agents.
- Nevertheless, in 1997, S-P had been marketing a combination antibiotic and anti-inflammatory injectable product called Finabiotic since 1986. It was a combination of flunixin and oxytetracycline. The oxytetracycline was short acting. The product was formulated overall to be administered once daily.
The Patent
- The opening paragraphs of the Patent, which both sides accepted represented common general knowledge, are a compact summary of some of the technical background which I have attempted to set out above. In short, long acting antimicrobials were a well known way of reducing the handling stress associated with antimicrobial administration and of maintaining blood levels. On the other hand, they have disadvantages, namely (a) they may cause skin irritation at the site of injection and (b) persistence of drug at the site of injection for a long time: see paragraph [0003]. At paragraph [0004] the authors say:
"A further difficulty with long acting or single treatment products can be that ,they may not give as high blood levels immediately following administration as do the repeat treatment products. Whilst this may not affect the overall level of efficacy of the product it may result in a slower rate of recovery which can in some cases lead to the increased rate of long term damage to affected tissues and organs. An alternative explanation for the cause of tissue/organ damage is that it results as a consequence of the animals own inflammatory response to infection. As well as causing damage to the tissue this inflammatory process may also reduce the diffusion of antimicrobial to the site of infection/inflammation. One means of preventing this happening is to administer an anti-inflammatory drug. Such a drug on its own will reduce the inflammatory response but will not reduce the incidence of bacterial infection and so it is also necessary to administer an antimicrobial. One such product containing both an anti-inflammatory drug and an anti-microbial is commercially available, namely Finabiotic, Schering-Plough Animal Health, however, use of this product requires daily treatments in order to be effective."
- The specification then says in paragraph [0005] that it is an object of the invention
"to obviate or mitigate the aforesaid disadvantages by providing a long-acting or single treatment formulation which has both antimicrobial and anti-inflammatory effect"
One of the areas of dispute in this case is precisely what this statement is promising, and which of the various disadvantages earlier referred to the patentee is, by implication, suggesting he had overcome.
- At paragraph [0006] the specification goes on to say that, in accordance with the invention, there is
"provided a veterinary product comprising an anti-microbial and an anti-inflammatory agent in intimate admixture ... wherein the anti-microbial is selected from the group consisting of long-acting antimicrobials or depot antimicrobial formulations."
- At paragraph [0009] of the specification it says:
"Advantageously this invention provides a veterinary product which on administration produces reduced irritation at the site of administration and reduced levels of antimicrobial found at the sites of administration over a period of time"
There is, again, a dispute about whether this passage is saying that these are advantages which necessarily follow from the use of the claimed invention, or merely that they are advantages which may attach to some, but not necessarily all of the claimed embodiments.
- The commendably short specification then proceeds by way of three specific examples. Example 1 is a trial of a combination product in accordance with the invention (long acting oxytetracycline and flunixin) against long acting oxytetracycline alone. Sixteen cattle were inoculated with a bacterium known to cause BRD. Those animals which, 48 hours after inoculation, showed the symptoms of BRD (pyrexia and signs of respiratory difficulty) were allocated into two groups~ One group was given a single administration of the combination product, whilst the other group was given antimicrobial alone at the same dosage rate. The animals in both groups were monitored for response to the treatment (a) by body temperature and (b) by scoring on a weighted symptom scoring system. The data was summarised by taking mean values for each group at each time point.
- Table 2, which is reproduced below, sets out the results for temperature for the two groups. It shows that in the group which received both antimicrobial and anti-inflammatory (i.e. the invention) the animal's temperature fell rapidly to around 102°C and remained at that level. By contrast, in the group which received the antimicrobial alone, the temperature remained above 103°C at each time point. In paragraph [0013] it is claimed that the results in a majority of the time points are statistically significant.
Table 2:
Timepoint |
Test Article (antimicrobials + anti-inflammatory) |
Control Article (antimicrobials alone) |
Pre-inoculation |
102.2 |
102.1 |
Pre-treatment |
104.2 |
104.1 |
Post treatment |
|
|
3 hours |
101.9 |
103.8 |
hours |
101.8 |
103.2 |
hours |
102.1 |
103.7 |
hours |
102.3 |
103.9 |
hours |
102.4 |
103.1 |
hours |
102.5 |
103.5 |
hours |
102.1 |
103.1 |
- Table 3 shows the results of the clinical scores. A statistically significant improvement for the combination product over antimicrobial alone is claimed using a "t-test". In paragraph [0016] the patentee claims for the combination product "greater short term and long term therapeutic efficacy than the long acting antimicrobial alone".
Table 3:
Timepoint |
Test Article (antimicrobials + anti-inflammatory) |
Control Article (antimicrobial alone) |
Pre-inoculation |
0.6 |
0.86 |
Pre-treatment |
23.3 |
22.7 |
Post treatment |
|
|
3 hours |
14.4 |
22.7 |
hours |
13.3 |
19.4 |
hours |
15.4 |
19.4 |
hours |
16.7 |
21.6 |
hours |
7.1 |
13.9 |
hours |
4.9 |
14.9 |
hours |
4.3 |
10.9 |
- Professor Lees, who was Norbrook's expert, thought that Tables 2 and 3 showed the anti-inflammatory was exerting its effect beyond its expected duration of 24 hours, and that this was surprising. He could not think of any explanation for this effect. Professor Elliott, on the other hand, was sceptical of the ability of the data to show statistical significance. The statistical t-test was in his view inappropriate for this type of analysis. Moreover, Professor Lees accepted in cross examination that once an anti-inflammatory has brought the temperature down you would not expect the temperature to go up again: there was not normally as he put it, a "biphasic" response. That being so it is unlikely that the results would teach the reader any more than that the anti-inflammatory has, on average, worked within the first 24 hours. No comparison is made with concurrent administration of the two drugs.
- Example 2 consists of a further trial to determine the amounts of residual antimicrobial in tissue. The comparison again is between the combination product of the invention and antimicrobial alone at the same dosage rate. The results showed much reduced (of the order of one hundred fold) levels of oxytetracycline at the site of administration at both 21 days and 28 days after administration. Thus the reader of the Patent would understand that the addition of flunixin to oxytetracycline was helping the oxytetracycline to diffuse away from the site of administration. Whether this was an advantage over concurrent administration is not shown.
- Example 3 attempts a determination of the degree of irritation following injection. In this case a single injection of the patented combination was compared with concurrent administration (i.e. two injections: one of antimicrobial and one of anti-inflammatory). Tissue irritation was measured by the presence of the enzyme AST. AST is released into plasma when body tissue is damaged. So a difference in AST could be used as a measure of tissue damage following injection. The results, which are just mean figures with no statistical analysis, show less damage with the single injection. There was a dispute as to how much this showed when the control animal was not given a second injection: in other words whether the difference could simply be due to the absence of a second injection site in the control group. It seems beyond dispute, however, that the skilled reader would understand that the advantage of less tissue irritation was being claimed for the combination tested.
- In paragraph [0021], in a passage which appears to relate to the invention as a whole rather than the specific combination of the examples, it is said that:
"The combination offers advantages by way of a more rapid and complete treatment of infection than delivery of either of the two individual components separately"
That the combination achieves a more rapid and complete treatment than anti-inflammatory alone is of course hardly surprising as anti-inflammatories alone would not combat the infection at all. The assertion that the addition of anti-inflammatory treatment is speeding up recovery as compared with antimicrobial alone is presumably based on Example 1, which, as I have said, suggests that the anti-inflammatory is doing its job of relieving symptoms and bringing down temperature, on average in the first 24 hours.
- In paragraph [0022] there is the following prediction:
"It should be noted that there are various possible combinations of a long acting or single treatment antimicrobial with anti-inflammatory agent for use in veterinary medicine. In this invention the antimicrobial drug could usefully be [a long list of antimicrobials] alone or in combination. Other antimicrobials may also be usefully employed where they can exist in long acting or single treatment formulations. The anti-inflammatory drug could usefully be flunixin [another long list] or other drug possessing anti-inflammatory properties"
- I should note in passing that the specification is more or less devoid of any detail as to how the products are to be formulated. Moreover the only agents tested in the examples were oxytetracycline and·flunixin.
The Claims
- After exchange of expert evidence, and not long before trial, Norbrook announced an intention to apply to amend some (but not all) of the claims of the Patent. In view of the proximity of the trial it was not practicable for the amendments to be advertised in the normal way so as to allow third parties affected by the amendment to oppose. In order to avoid an adjournment, which would have been both costly and disruptive, the parties initially reached an agreement between themselves that the allowability in principle of the amendments should be decided at the trial, with questions relating to the exercise of the court's discretion to be decided at a subsequent hearing if necessary. It seemed to me, however, that whilst having attractions for the parties, this course was potentially unfair to third parties who might wish to argue aspects of allow ability of the amendments not argued by S-P. So 1 ordered that all issues relating to the amendments should be dealt with separately. This trial therefore proceeded on the basis of proposed amended claims. If the amended claims are held valid, Norbrook will face a further hurdle in showing that the amendments are allowable as a matter of law and discretion. If all the claims are held invalid, or only claims unaffected by the amendments are held valid, then those further questions will not arise.
- Norbrook do not accept that the unamended claims were invalid, but made no attempt to defend them. Henceforth I will refer to the proposed amended claims as "claims". Claim 1 reads as follows:
A method for producing an improved veterinary product comprising bringing a selected long-acting antimicrobial formulation into intimate admixture with a predetermined amount of an anti-inflammatory agent and preparing the admixture for parenteral administration, wherein said anti-inflammatory agent is selected from indo lines such as indomethacin, oxicams such as piroxicam, acetic acid such as diclofenac, fenamates such as tolfenamic acid, propionic acids such as ketoprofen, or sulphonanilides such as nimesulide.
Claims 5, 6 and 7 are as follows:
5. A veterinary product comprising a long-acting antimicrobial, flunixin and a physiologically acceptable vehicle or carrier.
6. A veterinary product according to claim 5 comprising oxytetracycline, flunixin and a physiologically acceptable vehicle or carrier.
7. A veterinary product according to claim 6 comprising 300 mg/ml oxytetracycline, 20 mg/ml flunixin and a physiologically acceptable vehicle or carrier.
- So claim 1 is any long;-acting antimicrobial together with an anti-inflammatory selected from the cited classes, which include flunixin. Claim 5 is for any long-acting antimicrobial with flunixin. Claim 6 is long-acting oxytetracycline and flunixin and claim 7 is a particular strength of long acting oxytetracycline with a standard dose of flunixin.
The issues
- S-P contend that all the claims are invalid for obviousness over common general knowledge, a paper by Deleforge and two papers by Selman; and that claims 1 to 5 are invalid for insufficiency. Norbrook dispute all of this and say further that if claim 1 is invalid, it does not follow that claims 5, 6 and 7 are also.
Obviousness
Law
- Both parties addressed me on the basis of the structured approach to the issue of obviousness approved by the Court of Appeal in Windsurfing v Tabur Marine [1985] RPC 59 and subsequent authorities. That approach requires me, amongst other things, to identify the inventive concept of the Patent and ask, in the end, whether the difference or step between the prior art and that inventive concept is obvious. There is some helpful guidance in the authorities about how to set about the task of identifying the inventive concept.
- In Unilever v Chefaro [1994] RPC 567 at page 580 Jacob J (as he was then) said, having quoted section 125(1) of the Act:
"So when the Act says, in section 3:
"An invention shall be taken to involve an inventive step if it is not obvious to a person skilled in the art, having regard to any matter which forms part of the state of the art....."
it means that which is in the claim under consideration. It is the "inventive concept" of the claim in question which must be considered, not some generalised concept to be derived from the specification as a whole. Different claims can, and generally will, have different inventive concepts. The first stage of identification of the concept is likely to be a question of construction: what does the claim mean? It might be thought that there is no second stage-the concept is what the claim covers and that is that. But that is too wooden and not what courts applying Windsurfing stage one, have done. It is too wooden because if one merely construes the claim one does not distinguish between portions which matter and portions which, although limitations on the ambit of the claim, do not. One is trying to identify the essence of the claim in this exercise."
- Laddie J in Brugger v Medicaid (No.2) [1996] R.P.C. 635 said that:
"It is not legitimate to define the inventive step as something narrower than the scope of the relevant claims. In particular it is not legitimate to identify a narrow sub-group of embodiments falling within the claim and which have certain technical advantages and then to define the inventive step in terms which apply to that sub-group but not the rest of the claim."
It is only where technical advantages are possessed by everything falling within the claim, or at least all sensible embodiments of what is claimed, that it is legitimate to bear in mind the technical advantages of the invention. This is an important consideration in the present case. A wide inventive concept which includes but is not limited to embodiments with technical merit may be more vulnerable to an attack on the ground of lack of inventive step than a narrow one which is limited to or traps the advantageous features.
- In answering the statutory question it may help to show that the benefits which the invention necessarily brings with it are unexpected. Again, by "necessarily" I mean that everything falling within the claim will bring those benefits. That is because, where a seemingly simple step brings with it a valuable and unexpected benefit, it may only be obvious with hindsight to show that the step was one which would have been taken by the skilled person. But as Hallen v Brabantia [1991] RPC 195 shows, if the invention as claimed is obvious for one reason, the fact that it brings with it another benefit does not stop it being obvious. A fortiori if it is only certain embodiments falling within the claim that do so. That is because the statutory purpose is to allow workers in the art to take obvious steps. If in doing so they hit on something useful, that is a bonus they are entitled to take without fear of being sued for infringement of a patent.
- An invention may simply consist in an idea which, once it has been conceived, is one which will obviously work. For those cases a party attacking the patent only needs to show that the idea was an obvious one. But there are other cases where the invention involves something more than the bare idea, because it is not immediately apparent that the idea could be made to work. In these cases the attacking party needs to show something more: that it was obvious to have the idea and to try it to see whether it would work. The argument that it was "obvious to try" was always qualified by the further requirement that there must have been a reasonable prospect of success. A reasonable prospect of success was said to be one which was sufficient to warrant trial. And so the argument came full circle. Recently in St Gobain v Fusion Provida [2005] EWCA (Civ) 177 (unreported) Jacob LJ said that at [35]
"The "obvious to try" test really only works where it is more-or-less self-evident that what is being tested ought to work"
I propose to apply that test here.
- The primary evidence on obviousness is the opinions of the experts: their reasons for saying the invention would be obvious or not rather than their mere assertions that this is so. Secondary evidence of non-obviousness, such as what people did and did not do, or said or did not say at the time can be of assistance but must be kept in its place: Molnlycke v Procter & Gamble [1994] RPC 49. Secondary evidence is relied on here to show what was done by S-P and Norbrook, both together and in isolation.
The expert witnesses
- I have had the benefit o(hearing from two most distinguished experts in the field of veterinary pharmacology. For S-P I heard from Professor Jonathan Elliott who is Professor of Veterinary Clinical Pharmacology at the Royal Veterinary College, University of London. Although his professional experience related primarily to small animals and horses, his expertise was not limited to those animals and his teaching and training had extended to farm animals generally. For Norbrook I heard from Professor Peter Lees who is the Emeritus Professor of Veterinary Pharmacology, also at the Royal Veterinary College.
- Mr Richard Arnold QC, who appeared on behalf of Norbrook with Mr .Thomas Mitcheson, accepted that Professor Elliott was a distinguished scientist, and that as such he was doing his best to assist the court with his evidence. He submitted, however, that he was not representative of the skilled reader of the patent because of his particular research interests and because he had not carried out research into antimicrobials or anti-inflammatories to treat BRD. Mr Arnold went on to refer to the literature search carried out by Professor Elliott as being inappropriate, and the fact that in some respects Professor Elliott had diverged from evidence he had given in his report.
- I have no hesitation in rejecting all of this submission. Professor Elliott was a well informed and scrupulously fair witness, well qualified to assist on the matters. in dispute. It must be recalled that experts are not called as living embodiments of the unimaginative and uninventive skilled person: so it is not a contest to see whose expert most closely represents the skilled person. If the literature search was inappropriate, that may have consequences for S-P's case, but it does not reflect on Professor Elliott as a witness. His readiness to concede points of amplification or clarification from his report is not a ground for criticism either.
- I also heard from two formulation experts, Professor Buckton for S-P and Dr Mullen for Norbrook. I found them both to be helpful witnesses as well.
Construction and the Inventive concept
Claim 1
- There was a dispute about what was meant by "long-acting antimicrobial" in the context of the Patent. In the event it did not affect the way the obviousness case was argued. Norbrook said it meant that the drug was therapeutically active for 48 hours or longer. Norbrook rely on paragraph [0002] of the Patent which says that an advantage of "long-acting" is that animals do not have to be caught and restrained on a daily basis. S-P said there was no such hard and fast definition, and a drug may be long acting provided it is longer acting than shorter acting versions. I suspect that Norbrook's definition may be too rigid, but as it is not necessary for me to decide the point I do not do so."
- The real dispute about construction concerns the inventive concept. Mr Arnold argued that the inventive concept of claim 1 was:
"the combination of a long-acting antimicrobial together with an anti-inflammatory agent to give clinical benefits over the use of those agents individually or concurrently"
Mr Simon Thorley QC, who appeared for S-P with Mr Richard Meade, submitted that the inventive concept is just what is claimed. In their opening skeleton S-P said:
"The basic inventive concept is using a long-acting antimicrobial and an anti-inflammatory in a combination drug, for administration by injection.
In my judgment Mr Thorley's formulation is closer to the inventive concept of claim 1 of the Patent. I would add only two points, the second of more substance than the first. Firstly the last four words are not strictly necessary: the claim is wide enough to cover combination products for administration by other routes. Secondly it is implicit. in the inventive concept of claim 1 that the long-acting effect of the antimicrobial is retained in the combination product. So the invention is not just the idea of the combination product, but a combination product which retains the long-acting effect of the long acting antimicrobial. I explain my reasons in more detail below.
- 1 have reviewed the disclosure of the specification above. The skilled person comes to claim 1 having absorbed that disclosure and armed with his common general knowledge. In the introductory passages he has been told of the advantages of combination products {reduced handling stress), the disadvantage of long acting antimicrobials (irritation and site persistence), the disadvantage of the existing combination product (daily dosing and handling). It is said that it is an object to obviate or mitigate the disadvantages. He is given a particular example of a combination product (oxytetracycline and flunixin) which appears to do just that. He is then told that there are other combinations which "could usefully be employed".
- In my judgment the skilled person would not understand the disclosure of the patent to be telling him that every combination within the scope of claim 1 would produce benefits such as lower irritancy and less site persistence. This would be contrary to his experience that different combinations of drugs would be expected to behave differently. It would require a far firmer teaching than anything to be found in the patent specification for that message to be delivered. On the other hand he would clearly expect the long-acting effect of the antimicrobial to be demonstrated in the combination product: that is the obvious and overriding purpose of the combination.
- Norbrook's formulation of the inventive concept is not what the claim says: claim 1 merely speaks of an improved veterinary product without identifying the nature of the improvement or the comparison with which the improvement is made. There is no doubt that this is deliberate. The veterinary product will be improved if the antimicrobial is long-acting, as no combination of long-acting antimicrobials with anti-inflammatories was known. It will be improved because of the handling advantage. No sensible patentee would wish to limit his claim to cases where additional clinical benefits could be shown: in order to prove infringement he would have to do controlled tests in animals.
- Moreover Norbrook's formulation uses the term "clinical benefits" which Mr Arnold uses in his obviousness argument to denote something different from the mere handling/welfare advantages of a single injection: but the claim is not limited to products which provide such clinical benefits. Someone who sold a combination of the two types of agent for injection (or other parenteral use) would infringe and it would be no defence to show that he was not achieving a clinical benefit compared with individual or concurrent use.
- Finally the Norbrook formulation of the concept requires clinical benefits over individual or concurrent use. It is presumably intended that the "or" is conjunctive, so that the claim requires that there are advantages over both individual and concurrent therapy, otherwise the reference to concurrent use just makes the inventive concept appear more ambitious than it really is. However, the patent does not promise clinical benefits generally over concurrent administration. Example 3, has a concurrent control and shows (subject to questions about significance) that that combination is less irritant than the two injections given concurrently. But the skilled reader would not be taught by this that all the claimed combinations would display this effect; particularly as he has been told in paragraph [0002] only that long-acting formulations "may cause irritation", not that they will. Further, the second sentence of paragraph [0021] only makes a general claim for advantages over administration of the individual components, not over concurrent administration. In my judgment the skilled reader would not take out of the specification any promise that each combination will achieve clinical benefits over concurrent administration. There is accordingly no basis for implying any reference to benefits over concurrent use into the claim.
The skilled addressee
- There was a dispute over the proper description of the addressee in the present case. For S-P it was contended that the skilled addressee was a team comprising a veterinary pharmacologist or a veterinarian with pharmacological experience and a pharmaceutical chemist specialising in formulation. Norbrook contend that the addressee is only the pharmacologist/veterinarian and that the formulator does not form part of the skilled team. However Norbrook accept that in practice the work of formulating products in accordance with the invention would be handed to a specialist formulator.
- In my judgment the specification of the atent is addressed to a team consisting of the veterinary pharmacologist or veterinarian with pharmacological experience and a specialist formulator. The Patent is for a formulated product: claim I requires ingredients to be mixed intimately; and claims 5,6 and 7 require a vehicle or carrier. By remaining silent on all issues relevant to formulation, the patentee is assuming that the skilled addressee does not need to be told how to set about the work of formulation. The Patent is clearly addressing itself to a team which involves someone who will have no difficulty making up combination products without being given specific instructions on how to do so.
The common general knowledge
- I have recited much of the necessary common general knowledge above. There were, however, some areas of dispute as to what further common general knowledge the skilled team would have.
- One question was the extent to which it would be supposed that anti-inflammatories such as flunixin would cause irritation if injected intramuscularly. Professor Elliott was asked about this in cross-examination:
Q. It is right, is it not, that many NSAIDs such as flunixin were known to cause local irritation and even tissue damage if administered intramuscularly?
A. Yes, that was known at the time.
Q. That was one of the reasons why flunixin was only licensed for intravenous administration?
A. I am not sure how irritant flunixin is by the intramuscular route. There may be many reasons why a company would licence a drug by a particular route to begin with. The Finabiotic obviously does have flunixin going by the intramuscular route. So it does seem to me that it is feasible to give flunixin by the intramuscular route.
In re-examination Professor Elliott was shown an extract from a 1991 publication called Bovine Medicine. This is a substantial volume running to almost 1000 pages and including 56 chapters. The Professor was specifically directed to the following sentence on page 851 in the chapter called "Inflammation and anti-inflammatory drugs":
"Some NSAIDs are very irritant to tissues when injected perivascularly but flunixin, which is used as the meglumine salt, can be given intramuscularly when absorption is very rapid and tissue irritation is not great."
Having seen that passage the Professor then said that it would seem that flunixin was not irritant when injected by the intramuscular route. Professor Elliott's evidence as a whole, in my judgment, serves to show that it was common general knowledge that NSAIDs could be irritant if injected IM, but that it was not common general knowledge that flunixin was an exception. If the latter were not so I have no doubt that Professor Elliott would have mentioned it in his report, would not have given the answer he did in cross examination and would not have had to be re-examined. I am also fortified in this view by the fact that flunixin was not licensed for intramuscular use: I doubt that the skilled team would have expected to find much information in the textbooks on which they could rely about the IM use of flunixin. Professor Lees accepted that Bovine Medicine was a reasonably well respected textbook (which is not surprising as he co-wrote the relevant chapter) but that does not establish that every utterance in it would have been found by the skilled person, or if found be regarded as a basis for action. Moreover the passage merely says that the irritation is not great, not that it does not occur at all.
- Professor Elliott also considered that the use of anti-inflammatories was well known to combat irritation at the site of injection caused by the antimicrobial. He supported his view by reference to an article called Adawa which he found by doing a search. In my judgment Adawa was not part of the common general knowledge. Moreover I do not think it was established that the skilled person knew that anti-inflammatories would reduce irritation caused by anti-microbials. If asked to think about it, the skilled person might come to the conclusion that they might help, but he would not be administering the anti-inflammatory for that purpose. And he would not know whether the extra irritation caused by the injection of the anti-inflammatory would make things worse.
- There was also some evidence about what the skilled team would think about the necessary or desirable dosing regime for an anti-inflammatory such as flunixin when administered concurrently with an antimicrobial such as oxytetracycline. It will be recalled that flunixin was licensed for daily use for up to 5 days. Professor Elliott accepted that the standard regime was to administer it for at least three days. Clearly sometimes only a single dose would be administered. Dr Varma of S-P gave evidence that in his experience veterinarians usually only administered a single dose. It would all depend on the clinician's view at the time.
- Overall I think that the evidence established that the team would know that a single dose of flunixin would in many cases not be adequate, but that it would be adequate in some. The animal would have to be looked at on the day after the first injection to see whether the flunixin had done its job. A further injection would then be given as necessary. The evidence did not support the assertion that you would always need anti-inflammatory cover for more than 24 hours.
Obviousness over common general knowledge
- The argument starts from the known product Finabiotic which was a combination product including short-acting oxytetracycline and flunixin which required daily administration. If the skilled team were looking to improve the product it would be a natural thing to think of using a long-acting version of oxytetracycline to achieve the handling advantages and reduced stress for the animal. Long-acting oxytetracycline was a known product. So it was obvious, so S-P contend, to combine it with flunixin. In Windsurfing terms, the difference between the common general knowledge and the inventive concept was the incorporation of long-acting oxytetracycline into Finabiotic in place of the short-acting version: and that, say S-P, was obvious.
- Norbrook do not expressly concede that it would be obvious to have the bare idea of a combination of flunixin and long acting oxytetracycline, but Mr Arnold, on their behalf, focused his submissions on this part of the case on what the skilled team would have thought if the idea had occurred to it. Mr Arnold contends that the team would reject it as not being worthwhile. Any new pharmaceutical product requires expensive research and development. There would have been inadequate incentive to pursue the idea, and so the team would never have arrived at a veterinary product within the scope of claim 1.
- The first and most fundamental consideration for the skilled team would be whether, if they combined the long-acting oxytetracycline with flunixin, the long-acting effect of the antimicrobial would be retained. I think that it would be far from self evident that this would be the case. Professor Lees, in paragraph 50 of his report said this:
"It was not known in 1997 whether or not when a long-acting antimicrobial drug and a NSAID are co-administered at the same injection site in a single product, the absorption rate and hence the action of one or both the drugs would be altered (i.e. does one drug have an adverse effect on the other in terms of magnitude~ and time-course of action?). ill fact the studies conducted by Norbrook and Schering Plough in the 1980s, that are described in the following section, demonstrate that co-administration of flunixin with oxytetracycline does in fact alter the magnitude and time-course of at least the flunixin component of such a combination product (compared to when flunixin is administered on it's own)."
- Professor Lees' evidence suggests that the reader of the Patent would not know whether the interaction between the drugs would be such as to affect adversely their absorption and consequent action. Professor Lees was asked to expand on this in his cross examination:
MR. THORLEY: .......... In the third
19 bullet point, the second one down on page 16, you say it was
20 not known in 1997 whether or not with co-administration the
21 absorption rate and therefore the action of one or both drugs
22 would be altered. Would you just amplify upon that a little
23 for us please?
24 A. If you give two drugs in combination, as Prof. Buckton has
25 discussed at some length, then each drug, each component, may
2 have an impact upon the absorption of the other, These things
3 can be difficult to predict, but you have to do
4 the experiments in order to determine whether that is so. What we
5 do have is a series of studies that were undertaken by
6 Norbrook but in collaboration with Schering-Plough in the
7 1980s. From those studies -- I think there is a total of four
8 studies the data that clearly came out of them was that at
9 that point in time there were significant differences between
10 the plasma concentration time profiles of the flunixin and/or
11 the oxytetracycline when the compound were given singularly,
12 compared to when they were given in a test pilot formulated
13 product. Therefore, I put the point that at that point in
14 time, there having been no success in co-formulating the two
15 active ingredients to give the same plasma concentration
16 profiles as when administered as separate actives, it would
17 have been a discouragement to me to proceed further with the
18 development of a long-acting product of this kind.
19 Q. You could not predict before you tested what interaction, if
20 any, there was going to be?
21 A. You always have to make a formulation which you hope will give
22 you the same plasma concentration time profiles. You cannot
23 absolutely predict it. If you do not get the same plasma
24 concentration times profiles, if you are dealing with two well
25 established actives, like oxytetracycline and flunixin, then
2 you have to go back to the drawing board and say to the
3 formulator, "We have not succeeded. Please try again. We
4 want a formulation that gives us the bioequivalence for
5 separate
6 Get back to your desks and come up with an answer?
7 A. Yes, that is right.
And then later he asked this:
15 Q. I am now posing the hypothesis that you are dealing with
16 a less well known antimicrobial and you therefore do not have
17 the assistance of data to help you work out what are the
18 necessary plasma levels and so on. The problem is that much
19 more severe.
20 A. Yes. If you are dealing with what we might call a novel
21 compound, a novel molecule, then the studies that would have
22 to be undertaken would be considerably greater.
23 Q. There will be some fundamental research involved.
24 A. Yes.
- It seems to me therefore that it would not have been self-evident that the combination ought to retain the long-acting action of oxytetracycline. The skilled team would be concerned that the co-administration of flunixin at the same site would affect the absorption rate to the extent that the long acting effect would be lost.
- MY Arnold made a number of further points which I can summarise in the following way
1.The use of long-acting antimicrobials was associated with an increased risk of tissue damage and a longer meat withholding period compared to short-acting antimicrobials.
2.Some NSAIDs were known to cause localised irritation, which would be thought to compound problem 1 rather than reduce it if the NSAID is injected at the same site.
3. Flunixin was only licensed for intravenous administration whereas antimicrobials such as oxytetracycline were usually delivered by intramuscular injection.
4. It would have been illogical to have combined a standard anti-inflammatory drug which required daily dosing with a long-acting antimicrobial that lasted much longer - their durations and therefore actions were out of phase.
5. The convenience benefit of a combination product is lost if the need for daily treatment with the NSAID remains.
6. Combining the two drugs in a single product would lead to a loss' of flexibility for the clinician.
7. Even if the addressee contemplated the possibility of combining a long acting antimicrobial with an anti-inflammatory drug, his first step would be to test for bioequivalence with the component drugs. The evidence from tests carried out by SP and Norbrook before the filing date shows that such tests would have indicated that the combination was not worth pursuing - see further below.
8. If it was obvious to combine a long-acting antimicrobial with an anti-inflammatory drug to achieve clinical benefits, it would surely have been done before 1997. All the information and ingredients required were available to the skilled person by 1984. On SP's case there is no explanation as to why it was not done before.
- The first six points are those that would be discussed by the skilled team before advancing the project beyond the conceptual stage. These need to be assessed primarily by reference to the evidence of the experts, together with a further point not stressed by Mr Arnold in his submissions. The seventh is a point which would only be encountered if the team had put aside these initial considerations and gone ahead with trials for bioequivalence, and is based on secondary evidence. The final point is the familiar forensic cry, and also involves an appeal to secondary evidence. I consider each of the points made on the primary evidence in turn before turning to the others.
- The use of long-acting antimicrobials was associated with an increased risk of tissue damage and a longer ,meat withholding period compared to short-acting antimicrobials. This point loses its force, in my judgment, when one recalls that long acting oxytetracycline was on the market. The skilled team would not consider this as a freestanding reason for not proceeding with the project. However the irritancy of the antimicrobial would be a baseline when they began to consider the impact' of adding the anti-inflammatory at the same site.
- Some NSAIDs were known to cause localised irritation, which would be thought to compound the problem rather than reduce it if the NSAID is injected at the same site. It is true that some NSAIDs were known to cause irritation if injected intramuscularly. I have reviewed the common general knowledge about this above. The skilled team would not have known that flunixin did not add to the irritation. Even if I am wrong in holding that Bovine Medicine is not common general knowledge, the team would only know that the degree of added irritation from flunixin was "not great".
- I have to bear in mind here as well that Finabiotic was on the market. So the skilled team would know that the combined level of irritation of short acting antimicrobial and flunixin was not too great to prevent use in a commercial product. But that would not tell the team that they could get away with it in a long acting formulation. Professor Elliott was cross examined about that: in the end his view was that he thought that the addition of the NSAID would not make the position worse than it did with the existing product and there was a chance it might make it better. Professor Lees thought there was a danger that adding one irritant substance to another would lead to a summation of their irritant effects. He was asked this in cross examination about choosing an anti-inflammatory to go with a long acting antimicrobial:
Q. Certainly they would have been conscious of the fact that in choosing their anti-inflammatory, they should use an anti-inflammatory which itself did not cause any further inflammation.
A. That would be the ideal drug to choose, yes.
Later the passage from Bovine Medicine was put to Professor Lees. But it was not put to him that it showed that flunixin was a drug which would not cause any further irritation, merely that the take-home message from the passage was that the irritation was not great, which is of course what it says.
- I think Mr Arnold is right that the skilled person considering a combination of the long-acting antimicrobial oxytetracycline and the NSAID flunixin would be concerned that the cocktail might prove more irritating when administered by injection into the muscle. In order to know how much more irritating he would have to conduct the kind of trial which was done by the inventors in Example 3 of the patent, but he would not have been able to predict in advance the result of such a trial.
- Flunixin was only licensed for intravenous administration whereas antimicrobials such as oxytetracycline were usually delivered by intramuscular injection. This point does not seem to me to amount to a free-standing reason why the team would not proceed. They would know that in Finabiotic the flunixin goes in intramuscularly; and the dose of flunixin in the proposed product would be the same. But they would nevertheless have in mind that in the proposed product, both drugs would have to be administered at the same. site, so that the irritant effects at that site would be additive.
- It would have been illogical to have combined a standard anti-inflammatory drug which required daily dosing with a long-acting antimicrobial that lasted much longer - their durations and therefore actions were out of phase. The idea of giving the long-acting form of antimicrobial is to avoid the need to catch the animal and inject it the following day. For those animals requiring a single dose of NSAID the combination product is just what they want: for those requiring two of three days of NSAID, they will require to be restrained and injected on the subsequent days in any event. The problem is that one does not know on day 0 what regime of NSAID will be needed. So, says Mr Arnold, the combination product is not a. rational one to make.
- Mr Thorley says that the combination product would be thought of as a good starter injection for all animals requiring NSAID therapy. If it turns out that further NSAID therapy is necessary that can be administered by standard daily injections. The proposed product would be thought a useful addition to the vet's armoury by allowing him to give one injection rather than two on the first day, and possibly avoid subsequent injections altogether.
- Mr Arnold says that the "one injection not two" would be seen as a mere convenience benefit, insufficient to secure regulatory approval. But in my view that is to say that the invention was not commercially obvious: it cannot stop the idea being seen as beneficial from a practical and technical perspective.
- Both sides were supported by their experts in the points that they made. Professor Lees thought in particular that the team would want to be sure that the product provided anti-inflammatory cover for more than one day before it would go forward. Professor Elliott thought that the combination would be useful even if it provided 24 hours cover.
- I think Mr Thorley is right to the extent that the mismatched product would not be rejected on these grounds. Extending the period of anti-inflammatory cover would be thought a benefit, but not essential. But looking at it from the other point of view, the team would have mixed feelings about the usefulness of a mismatched product.
- The convenience benefit of a combination product is lost if the need for daily treatment with the NSAID remains. It is true that some of the convenience is lost, but the need for daily treatment is eliminated in those animals who only require a single dose of NSAID; and the animal receives One injection not two on the first visit.
- Combining the two drugs in a single product would lead to a loss of flexibility for the clinician. This assumes that the clinician only carries the combination product: but I am satisfied that the team would not think this to be so.
Secondary evidence
- Both sides rely on aspects of what was done by S-P and Norbrook to support their cases. I should start by summarising what occurred. The evidence is given by Dr Varma of S-P, who exhibits the relevant documents. Norbrook chose not to cross-examine him.
- In November 1985 Schering in France proposed a combination product of oxytetracycline and flunixin. The proposed dosing was 20mglkg of oxytetracycline and 4 mg/kg of flunixin, i.e. a double dose of both. The memo suggesting the product states simply:
"To develop this product we must know:
• what is the pharmacokinetic behaviour of oxytetracycline and flunixin when their posologies are doubled ....
• What are the risks of toxicity and local safety with these new posologies?"
It is clear that this proposal led to some research either in France or. the United States, because in September 1987 a Project Authorisation Request of Schering US records that research has been done on a combination product at 20mglkg of oxytetracycline and 2 mg/kg of flunixin, giving 48 hours of antimicrobial and 24 hours of anti-inflammatory cover. The document states that there were "distinct therapeutic benefits over the use of oxytetracycline alone", but does not identify what these are.
- In my judgment. this evidence is consistent with the view I arrived at as to the thinking of the skilled team. The initial reaction was to try and produce a "matched" product, but this is later modified to the less ambitious task of an unmatched product. The 1985 memorandum throws some light on the concern about tissue irritation.
- In the meantime S-P had entered into an agreement with Norbrook to collaborate on the production of a combination of Norbrook's long-acting formulation of oxytetracycline and flunixin. The agreement was signed on May 18th 1987. The work conducted under that agreement, and a successor agreement of January 1989 is summarised by Professor Lees in his report. He concludes that the work showed that it was not possible to establish bio-equivalence for flunixin for products containing flunixin alone and formulations containing flunixin and oxytetracycline. He says that he would have been concerned in particular about the peak concentration of flunixin in plasma which was consistently lower. He would therefore not have been able to assume that the combination product would have been as good as for the same doses of the drugs administered by separate injections.
- The lack of bio-equivalence is something which the skilled team would only encounter if they had found the objective of a combination product something sufficiently promising to warrant trial. Once they had set off on the trials, I doubt that the lack of bioequivalence amounts to a technical reason for not completing them: it would merely mean that it would have to ascertain whether the peak plasma levels were still adequate for therapeutic efficacy. Of course, bio-equivalence would encourage the skilled team that the regulatory pathway would be fairly clear. But as Richardson Vicks illustrates, regulatory and technical considerations need to be kept separate.
- I think these trials serve as no more than an illustration of the fact that it could not have been assumed by the skilled team, before they commenced their work, that the long-acting effect of the combination would be preserved. I have already taken this factor into account in my assessment of the primary evidence.
- Finally S-P continued with work on a long-acting combination of florfenicol and flunixin. This work continued intermittently through the 1990's. In 2003 they filed a patent application, claiming priority from 2002 for this combination. The patent application states that it would be expected that flunixin injected intramuscularly would have reduced bioavailability and hence less clinical efficacy than when administered intravenously, and that it is surprising that flunixin when administered intramuscularly with florfenicol is as efficacious. This serves as another illustration of why the skilled team would not be entitled to expect the combination to work exactly as the sum of its parts: but it is no more than that.
- Mr Thorley invites me to look at the totality of the secondary evidence and to see it as supporting S-P's case of obviousness. He says that Mr Arnold's "lions in the path" fear of added irritation, mismatching activity, lack of bioequivalence - did not in fact stop the idea being proposed and tried by S-P and Norbrook, and on several occasions. Thus he says that the French Schering came up with it; they proposed it to the Americans; and Norbrook and Schering proposed collaboration with each other, all without .objection that it is not worthwhile. Moreover he secured admissions from Professor Lees that at least some of the steps taken were reasonable things to do.
- In my judgment the evidence relied on by Mr Thorley is wholly inadequate to support the use which he seeks to make of it. The original French proposal was coupled with a concern about toxicity and local safety which was not explained. Thereafter the actions of S-P were conditioned by tests which had been done. The evidence does not show that those involved were not and had never been concerned about whether the combination could be produced without unacceptable irritation at the site of injection, or without deleteriously affecting absorption behaviour.
- I do not think I gain much assistance from asking the forensic question ("Why was it not done before?") in the present case. S-P were concerned about a Pfizer patent, and in any case the decision on which products to take beyond the concept stage was more of an economic than a technical one. In any case it would require a much more detailed, and accordingly disproportionate investigation, to determine whether the various delays were really occasioned by the fact that the inventive concept was not obvious.
Conclusion on obviousness over common general knowledge
- Although not expressly conceded by Norbrook, I conclude that the idea of combining a long acting antimicrobial such as oxytetracycline with an anti-inflammatory such flunixin would be one which would occur to the skilled team without invention. The prior availability of Finabiotic and long-acting oxytetracycline make this conclusion more or less inescapable. I have nevertheless come to the conclusion that the inventive concept of claim 1 involves an inventive step as compared with common general knowledge. The skilled team would not have known that the product would work in the sense that the long-acting effect of oxytetracline would be preserved. Further, they would have been concerned that the combination product would have been unacceptably irritant. Although the skilled team would appreciate that the product would have had some limited utility if it worked, their state of mind would not have been that it was self evident that the combination would work.
- I believe a similar result would have been arrived at by applying the European Patent Office's problem and solution approach. If Finabiotic is taken as the closest prior art, then the "problem" for claim 1 would be whether one could produce a combination product with long-acting oxytetracycline and flunixin in which the long-acting effect of oxytetracycline will be retained. The answer would be that the common general knowledge did not teach you whether this would be so.
The prior art
Deleforge
- Deleforge was published in 1994. In it the authors studied the effect of administering a long-acting formulation of oxytetracycline alone with long-acting oxytetracycline co-administered with tolfenamic acid, an NSAID, the latter either administered as a single dose or as two doses two days apart.
- There was much debate between the experts as to how much was scientifically established, as opposed to claimed, by Deleforge. Much of the data is stated to be statistically not significant. Importantly, however, the paper shows that a second dose of the NSAID was important.
- In my judgment it would not be obvious to make a combination product within claim I on the strength of the teaching in Deleforge. The skilled team's concerns about absorption and irritation when injection occurs at the same site are not displaced by Deleforge.
Selman 1
- Selman 1 was published as part of the proceedings of the Fourteenth World Congress on Diseases of Cattle held in Dublin in August 1986. It discloses benefits said to be achieved by treating calves with a combination product containing both short-acting oxytetracycline and the anti-inflammatory flunixin, compared to that of treatment with antibiotic only. The product involved is Finabiotic, although it is referred to by another trademark.
- I have held that Finabiotic was common general knowledge. Selman 1, it seems to me, adds nothing to the case based on common general knowledge. S-P rely on it in addition to show that the industry treated Finabiotic seriously. I am in no doubt that it did treat the product seriously- according to Dr Varma it was a strong product in Europe. But it does not make claim 1 invalid for obviousness any more than does the common general knowledge.
Selman 2
- Selman 2 was published as part of an International Symposium on Nonsteroidal Anti-inflammatory Agents in Orlando, Florida in January 1986. It presents more or less the same data as Selman 1, but with this additional statement:
"Furthermore the intramuscular administration of the combination product (Finoxine) appeared to be painless and did not result in any significant local (i.e. injection site) reaction: in fact, it appeared far better tolerated than oxytetracycline hydrochloride by itself."
This is a qualitative observation, unsupported by any data. If the skilled team with its common general knowledge had. read Selman 2, they would have considered this statement in relation to their concern about injection site irritation. But they would not be satisfied that if they adopted long-acting oxytetracycline and flunixin they would observe a corresponding benefit. Moreover the paper would do nothing to allay the concern over absorption rate. It therefore does not change my view that claim I was not obvious.
Conclusion on obviousness over the prior art
- I conclude that none of the cited prior art documents leads to a conclusion different from that which I reached on the basis of the common general knowledge. The obviousness attack fails ..
Sufficiency
Law
- In Biogen v Medeva [1997] RPC 1 at 48, Lord Hoffmann said:
... the specification must enable the invention to be performed to the full extent of the monopoly claimed. If the invention discloses a principle capable of general application, the claims may be in correspondingly general terms. The patentee need not show that he has proved its application in every individual instance. On the . other hand, if the claims include a number of discrete methods or products, the patentee must enable the invention to be performed in respect of each of them.
- In Kirin Amgen v Hoechst Marion Roussel [2005] RPC 9 Lord Hoffmann explained what was meant by a principle of general application:
112. This gave rise to a good deal of argument about what amounted to a "principle of general application". In my opinion there is nothing difficult or mysterious about it. It simply means an element of the claim which is stated in general terms. Such a claim is sufficiently enabled if one can reasonably expect the invention to work with anything which falls within the general term. For example, in Genentech I/Polypeptide expression (T 292/85) [19891 OJ EPO 275, the patentee claimed in general terms a plasmid suitable for transforming a bacterial host which included an expression control sequence to enable the expression of exogenous DNA as a recoverable polypeptide. The patentee had obviously not tried the invention on every plasmid, every bacterial host or every sequence of exogenous DNA. But the Technical Board of Appeal found that the invention was fully enabled because it could reasonably be expected to work with any of them.
113. This is an example of an invention of striking breadth and originality. But the notion of a "principle of general application" applies to any element of the claim, however humble, which is stated in general terms. A reference to a requirement of "connecting means" is enabled if the invention can reasonably be expected to work with any means of connection. The patentee does not have to have experimented with all of them.
- The burden of proving insufficiency is on the party alleging it. Mr Thorley argues that once the patentee responds to an allegation of insufficiency by amending his claims, the burden shifts to the patentee to disprove insufficiency. I do not see why this should be. A wide claim may be insufficient, but a narrower claim already present in the patent may turn out not to be so. Or the patent may be amended in response to official objections that the claims are too wide. The burden of showing insufficiency remains with the party alleging it in that case. I see no reason why it should not in the present case also. However, I do not think that the conclusion in this case turns on where the burden of proof lies.
The pleaded case
- S-P's case is that the wide claims of the Patent cannot be justified on the basis of the disclosure of the Patent. Their pleaded case was originally as follows:
The teaching of the Patent does not enable a person skilled in the art to work the alleged invention across its whole scope without undue effort and/or invention. In particular a person skilled in the art would not be able, without undue effort and/or invention, to practise the alleged invention of the Patent with long-acting anti-microbial formulations other than those specifically taught in the Patent. Further the teaching of the Patent does not give rise to a reasonable expectation that the alleged invention could be practised without undue effort or invention other than with the long-acting anti-microbial formulations specifically taught.
- It will be seen that the plea as it stands accepts the sufficiency of combinations of oxytetracycline and flunixin combinations, as these are specifically taught. It is an attack on the width of the claim beyond that specific combination.
- When S-P's evidence came to be served it included the expert evidence of a skilled formulator, Professor Buckton. S-P agreed, in the light of protests from Norbrook, to provide a statement of their case of insufficiency on formulation matters. That case alleged a number of difficulties of formulation: .
the difficulty of finding a long-acting anti-microbial where one was not previously available;
the fact that there were only a few known, long-acting anti-microbials; the difficulty of dealing with release rates;
the lack of direction in the patent as to whether a depot effect was to be exploited or whether the specific characteristics of the formulation should be used;
the difficulties arising from combining a long-acting anti-microbial with an anti-inflammatory;
the fact that· there was no way of making predictions about how different combinations would work.
What is required for sufficiency in the present case?
- A habit of patent lawyers (which it is easy to fall into) is to talk about directions being sufficient to enable the invention to be "worked" without specifying what working the invention really means. It is essential in a case such as this to identify what is meant. Moreover it is important to be even-handed with the issue of obviousness.
- In the present case the invention of claims 1 and 5 is a veterinary product which is a combination of two different types of active ingredient in which the long acting effect of the antimicrobial is retained in the combination product. It Follows that for those claims of the Patent to be sufficient the skilled person must be capable of making an combination of those two ingredients which has that effect.
The evidence of insufficiency
- In paragraphs 50 to 55 of his first expert report Professor Elliott noted potential difficulties with some of the claimed combinations of drugs covered by the unamended claims. That evidence was given before the amendments were proposed, and the combinations he considers are not within the claims.
- Thus S-P's case is essentially one about the difficulties or impossibilities of formulating some combinations within the claims so as to achieve a product at all or one having the necessary retention of long-acting effect. Professor Buckton explained that these difficulties would be non trivial even if starting from an existing long-acting formulation. He said that "the presence of the second drug could alter the release properties of the first (by altering solubility I diffusion / exchange of counterions and for a number of less obvious reasons), just as the existing formulation mayor may not alter the release properties of the second component." He also said that "it would not be a trivial thing to add a second drug (one of a great many anti-inflammatories) to an existing long acting formulation. To do so would require a formulation to be tested to ensure release rates of the two drugs are controlled in the desired way, that they liberate concentration I time profiles that achieve the required therapeutic effect whilst minimizing toxicity, and also that the product is chemically and physically stable over its entire shelf life." He maintained his view in cross examination, explaining that there were problems enough with one active ingredient, but where you have two and one is to be long-acting there was "very very much more complexity". On many occasions you would not be able to make a product because there will be too many problems.
- Dr Mullen who was Norbrook's expert was asked in cross examination to consider a number of formulation problems, dealing with physical incompatibilities such as differing solubilities. His evidence was that this work might throw up problems, sometimes easy to solve and sometimes more difficult or impossible. He tried to epitomise the differences between his evidence and that of Professor Buckton's as follows:
20 Q. Sometimes you can solve them easily and sometimes you can
21 solve them with difficulty.
22 A. Yes, it is a very inexact science. With the best skill in the
23 world and understanding of physical chemical interactions,
24 sometimes they are unpredictable. I think perhaps where
25 myself and Prof. Buckton differ is in terms of emphasis.
2 I believe perhaps Prof. Buckton was unduly pessimistic. I was
3 trying to give my Lord a kind of more balanced idea. If it
4 does not sound too flippant, perhaps one way to look at it is
5 that Prof. Buckton's attitude was that the glass was half
6 empty. My attitude was that the glass was half full.
- The problems are illustrated to some extent by some evidence which Norbrook introduced in the cross-examination of Professor Elliott. It consisted of a declaration by a Dr Robert Simmons of S-P in which he explained amongst other things the problems encountered in the course of determining whether the florfenicol/flunixin combination could be made to work. A complex of the two products was formed. Although the formation of this product was minimised by a stabiliser, and it was clearly possible to test the combination without the use of the stabiliser, it is nevertheless the case that the Patent teaches nothing about how to avoid these drug-drug interactions. The fact that they apparently do not occur for flunixin and oxytetracycline provides no assurance that they will not occur for other combinations. It is simply not possible to extrapolate.
- In my judgment it is inconceivable that all the combinations falling within claims 1 and 5 could be made to work in the sense I have indicated without undue research or experiment. The invention requires more than the ability to get the two products into the same injection: it requires the formulator to be able to do so and retain a long acting effect. This is too great a challenge to enable a claim of the width of claims I or 5. As Mr Thorley submitted, a half empty glass is, in the end, half empty, whether viewed by the optimist or the pessimist.
Does the Patent teach a principle of general application?
- Of course the fact that there will be some, or indeed many cases where the invention cannot be made to work without undue effort, or not be made to work at all will not matter if the patent can be said to disclose a general principle. Does the patent disclose a principle of general application, in the sense that it was a reasonable prediction that all combinations claimed would have been combinations in which the long-acting effect of tile antimicrobial would be retained?
- I think Professor Lees' evidence, which I have quoted at paragraph 58 above, is fatal to this argument. It is precisely because absorption in combination products is unpredictable that the claims need to be confined to those combinations which will work without undue research. The claims under consideration are far too wide for that purpose.
- Professor Lees did say that he thought that the Patent did disclose a "class effect". But it was plain that this view was based in part on knowledge he acquired after the relevant date. And his evidence as to why this class effect existed was not persuasive, principally because he accepted that he could not explain why the effect occurred. Professor Elliott was not prepared to accept the existence of a class effect disclosed by the Patent.
- I would add only that if I am wrong about the inventive concept, and it is Norbrook's formulation or something like it which represents the target for insufficiency, then it is clear that the task facing the skilled team is even more challenging and the case of insufficiency is stronger still.
Conclusion
- In the result, claims 1 to 5 are invalid for insufficiency, but claims 6 and 7 are valid. I will hear counsel on the form of order and the consequences of my findings for Norbrook's amendment proceedings.