England and Wales High Court (Patents Court) Decisions

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
PATENTS COURT

		Royal Courts of Justice Strand, London, WC2A 2LL
		27/11/2003

Mr Justice Laddie:

1. This is an action for infringement of Patent EP (UK) No. 0,402,152 ('the Patent'). The claimant patentee is Merck & Co. Inc. The defendant is Generics (UK) Limited. The Patent relates to a method of making a compound called 4-amino-1-hydroxybutylidene-1,1-bisphosphonic acid monosodium salt trihydrate, that is to say the monosodium salt of alendronic acid. This salt is more commonly called monosodium alendronate, or simply "alendronate". I will refer to it by the latter name. Structural drawings of this compound and its parent acid, alendronic acid, are shown below.

2. Alendronate is a valuable pharmaceutical. It has proved to be of use in inhibiting bone resorption. For this reason it has application in the treatment of diseases such as osteoporosis. It is, apparently, the largest selling osteoporosis medicine in the world.
3. The defendant is a supplier of pharmaceutical preparations to the United Kingdom market. At the end of last year it notified the claimant that it intended to sell here a medicament in which alendronate was the active ingredient. That ingredient is to be made by company in India referred to during the trial as "CIPLA". The defendant supplied the claimant with a confidential description of the CIPLA process. That description concentrates only on those features which could have a bearing on the question of whether or not the process falls within the scope of the Patent. The defendant gave the claimant a limited time to concede non-infringement, failing which it said it intended to commence proceedings for a declaration to that effect. In the event, the claimant commenced the current proceedings before the expiry of the time limit given by the defendant. The action is based on the confidential process description which had been supplied by the defendant. At a case management conference on 14 July of this year, before the defence had been served, a timetable was put in place allowing for a rapid hearing of the issue of infringement. It was also ordered that if the defendant chose to challenge validity the determination of that issue was to be stayed until after this trial on infringement. A counterclaim for invalidity has been served by the defendant. For the above reasons, it is not a matter which arises for determination now. Indeed, because there is no dispute as to the nature of the alleged infringing process, the only issue I have to decide is one of construction of the Patent.
4. Because CIPLA considers that its process is confidential, not only was the process description confidential but so was some of the evidence. At the beginning of the trial, I was invited to sit in private. Rather than do that for the whole of the trial, a code was used to identify certain key steps, ingredients or parameters in the CIPLA process and, as a result, the opening of Mr David Kitchin QC, who appeared for the claimant, was conducted in open court. I will adopt the same approach in this judgment. Anything which is arguably confidential will be referred to by a code letter or otherwise in a manner which will hide its identity.

Essential Chemistry

5. The chemical reaction by means of which alendronic acid is made is the subject of US patent 4,407,761. The general procedure for making acids of this type involves reacting an aminocarboxylic acid with a phosphonating agent, and then hydrolysing (i.e. reacting with water) under the influence of a non-oxidising acid. When alendronic acid itself is to be made, the aminocarboxylic acid used is gamma-amino butyric acid ("GABA"), the phosphonylating mixture can consist of a mixture of phosphorous acid (H₃PO₃) and phosphorous trichloride (PCl₃) and the non-oxidising acid is hydrochloric acid (HCl). The GABA, phosphorous acid and phosphorous trichloride are mixed in a solvent, chlorobenzene (PhCl). The whole reaction can be depicted, semi-figuratively as follows:


In the latter diagram, the hydrolysis stage using hydrochloric acid, is represented by the notation "(ii) H₂O" (the chemical formula for water).

6. It is elementary chemistry that if a base reacts with an acid, the salt of the acid is produced. Thus if Sodium Hydroxide (NaOH), a base, is reacted with Hydrogen Chloride (HCl), an acid, the salt called Sodium Chloride (table salt) is produced. Water is a by-product:

7. At all relevant times it has been common general knowledge that when a substance ('solute') is dissolved in a solvent, it can be made to crystallise out in a number of ways. One of the most simple involves cooling the solution. As the temperature falls the amount of solute which the solvent can dissolve falls. If there is too much solute in the solvent for the latter to dissolve at the lower temperature, the excess comes out of solution. It crystallises out. A well known alternative method involves changing the nature of the solvent. This can be achieved by adding an agent called an anti-solvent to the solution. The solution now has a lowered ability to dissolve the solute and, again, the excess crystallises out.

The Patent Specification

8. The Patent is short and simple to understand. The specification starts with the following passage:

"This invention relates to 4-amino-1-hydroxybutylidene-1,1-bisphosphonic acid monosodium salt trihydrate and a process therefor, where the end product is obtained in particularly pure form and at high yields in a one-pot procedure, and compositions containing said salt and its use in their manufacture."

9. Although this passage says that the invention relates to alendronate and a process for making it, the claims now are limited to the process alone, all claims to the product having been abandoned by the claimant just before the commencement of another trial. For present purposes, therefore, it is only necessary to consider the process disclosed and claimed in the Patent.
10. The specification continues by referring to a number of prior art patents. In particular it refers to US Patent No. 4,407,761 and the latter's disclosure of a method of making alendronic acid. However it notes that there is a problem with that process which it describes as follows:

"Problems result from this reaction whereby it does not remain homogeneous and local solidification occurs. This solidification causes variable yields, which in part results from the exothermic nature of the reactions with development of hotspots."

11. The parties agreed that the notional skilled addressee of the Patent would also read the US patent. That does state that the reaction product is solid and it also states:

"If desired, inert organic diluents, which do not solublize the reaction product, particularly chlorinated hydrocarbons, such as chlorobenzene, tetrachloroethane, tetrachloroethylene, and trichloroethylene, and dioxane can be used in the reaction." (US 4,407,761 Column 3 lines 8 to 12).

12. In fact that piece of prior art does not state that the formation of solid is a problem and it does not refer to any defects caused thereby. It does not address the question of why the reaction product is solid, that is to say what the mechanism is which makes it solidify, nor does it address the question of why that product is insoluble in the inert organic diluents it identifies. Similarly, the Patent in suit does not address either of these questions. The Patent then sets out the advantages which its process claims to achieve:

"Our new process allows the reaction to remain fluid and homogeneous and makes manufacturing of [alendronate] possible. It also has the advantage of requiring only one process step and provides a yield of 85-90%."

13. The inventors' discovery is then explained as follows:

"It has been found that pure crystalline [alendronate] can surprisingly be obtained (in high yields) by the reaction of 4-aminobutyric acid with phosphonating reactants in the presence of methanesulfonic acid at a temperature of less than 85^oC to yield a reaction mixture containing [alendronic acid]; the crystalline monosodium salt trihydrate [i.e. alendronate] is crystallized directly from the reaction mixture in about 90% yield after quenching, hydrolysis, and pH adjustment to about 4.3 with no further purification necessary."

14. A number of features of this paragraph should be noted. The discovery is stated in wide terms. Although GABA is identified as one of the reagents, the phosphonating reactants are not specified. However it does specify that methanesulphonic acid (MSA) is present. What the latter ingredient does, or how it facilitates the reaction is not stated. Furthermore, there are many other features of the discovered process which are not identified. For example, the method and ingredients used in the quenching and hydrolysis stages are not expressly identified. Although it is said that the pH is adjusted to about 4.3, there is no limitation to any particular method of pH adjustment, in particular no reference to the use of any particular alkali to counter the acidity of the reaction product, nor any reference to the temperature at which the adjustment should be carried out. Nor is there any description of how the process is manipulated so as to cause the alendronate to crystallise out nor any description of any further purification steps.
15. As I have mentioned, there is nothing in the specification which states how the MSA achieves the beneficial results claimed. The only passage which might be said to throw any light on this is the following sentence in the Patent:

"It is not necessary to use a diluent when methanesulfonic acid is used in the reaction."

16. This, however, does not state that MSA is a diluent, only that diluents are rendered unnecessary when MSA is used. Thus, even this passage tells the reader nothing about MSA's modus operandi.
17. Although there is a small amount of other general teaching in the specification, neither party suggested that any of it was relevant to the issues to be determined in this trial. The only other part of the specification which should be referred to is the two examples. These are introduced by the usual statement that they are intended to be illustrative of the practice of the invention without being limiting in any way. The first example consists of a worked experiment producing alendronate. It is not necessary to go through all of it for the purpose of this judgment. However it should be noted that it contains details of how the experiment was carried out including timings, concentrations, temperatures and ingredients. In this example the phosphonating agents used are phosphorous acid (H₃PO₃) and phosphorous trichloride (PCl₃) and details are given of, amongst many other things, the pH adjustment stage including the alkali used (NaOH) and the temperature at which this is carried out, the crystallisation step (which consists of cooling the reaction product to between 0-5^oC) and a filtering, washing and drying stage following crystallisation. The latter stage includes washing with water and 95% ethyl alcohol (ethanol). It is stated that the yield was 90%.
18. Example 2 is really not another example at all. It is the result of an analysis of the product of Example 1. For present purposes it is only relevant to note that it records the purity of the alendronate (measured by a method called HPLC) to be 99.5%.

The Patent Claims.

19. Although claims 1 and 2 are in issue, the parties agreed that there is no material difference between them for the purpose of this action. Reference need only be made to claim 1 which reads as follows:

"A process for the preparation of crystalline 4-amino-1-hydroxybutylidene-1,1-bisphosphonic acid monosodium salt trihydrate which comprises:

(a) reacting 4-aminobutyric acid with a mixture of phosphorous acid and PCl₃ in the presence of methanesulfonic acid;

(b) treating with water;

(c) bringing the pH to 4.3 with sodium hydroxide solution at a temperature of 20-25ºC;

(d) cooling to 0-5ºC; and

(e) collecting the desired compound by filtration, washing with water and 95% ethanol and air drying."

20. It will be appreciated that the following integers are derived from Example 1 rather than from the general description in the specification; (i) the identity of the phosphonating agents, (ii) stage (b), (iii) the use of sodium hydroxide and the temperature range in stage (c), (iv) stage (d) and (v) stage (e). There are many other process steps in the Example which have not found their way into this claim. The function of the various stages in the process covered by the claim are as follows. Stages (a) and (b) contain the reactions by which alendronic acid is produced. The pH adjustment stage (c) involves adding the alkali (NaOH) to the alendronic acid. This is the stage where the salt, alendronate, is formed (see paragraph 6 above). The cooling in stage (d) is a means for forcing the alendronate out of solution (see paragraph 7 above). Finally stage (e) is a post-crystallisation process used to prepare a dry product.

Relevant features of the CIPLA process

21. It has been said on a number of occasions that the scope of a patent must be determined without regard to the features of the alleged infringing product or process. In this case it is useful to have in mind the relevant features of the CIPLA process if only to help to identify the areas of dispute between the parties. At least at first blush, the CIPLA process is very different to that covered by claim 1. In particular (i) it does not use MSA, (ii) nor the temperature range at stage (c), (iii) nor does it use simple cooling to crystallise at stage (d) so that, amongst other things, crystallisation takes place at a different temperature, (iv) nor does it involve either washing with water or the use of ethanol in stage (e). Furthermore it is followed by a purification stage, a detailed description of which is not necessary to include in this judgment. Mr Floyd QC, who appeared for the defendant, produced a convenient table which sets out the differences. Modified so as not to reveal the details of the CIPLA process, the table is set out below:

Claim 1	CIPLA process
(a) reacting GABA with a mixture of phosphorous acid, PCl3 in the presence of MSA at a temperature of less than 85oC	No use of MSA
(b) treat with water	Yes
(c) bring to pH 4.3 with NaOH at 20-25oC	No – different temperature range
(d) cool (reaction mixture) to 0-5oC	No – different temperature range. Different method of crystallisation used
(e) collect alendronate by filtration, wash with water, with 95% ethanol and air dry	No – neither water wash nor ethanol used
	Post-collection purification procedure

22. The omissions from the claim are so extensive that it is tempting to say that what is left is little more than the Cheshire Cat's smile. Mr Kitchin argued that the differences between the CIPLA process and the requirements of the claim are of no significance. As he put it in his skeleton argument, step (a) in the claim, that is to say the use of MSA, is the step in which the problems of solidification in the prior art are addressed and it is here that the patentee's contribution resides. It is therefore the most important integer. The remaining features are standard work up steps. He said that the CIPLA process takes the whole of the benefit of the Merck invention, while introducing trivial variants to it. CIPLA uses a sulphonic acid which, whilst being different to the one specifically mentioned in the Patent (i.e. MSA) produces the same effect and obviously so. This theme that the CIPLA process is an obvious variant of the process as defined in the claims, manifested itself in a number of ways. In his skeleton he argued that the notional skilled addressee would not expect the patentee to exclude obviously immaterial variants and, therefore, such variants are to be covered by the claim. Similarly, in his first expert report served on behalf of the claimant, Professor Scott addressed the use by CIPLA of a sulphonic acid other than MSA. He said:

"9.5 I believe that the CIPLA process takes advantage of the benefit of the invention in the 152 Patent, particularly because the use of a sulphonic acid … in the first stage of the reaction allows the reaction to remain fluid and homogeneous throughout. This was the fundamental problem with the earlier methods for production of alendronate and which rendered them impractical for production on a commercial scale. Prior to the 152 Patent there was no way around this problem. However, once the 152 patent disclosed that MSA could keep the reaction fluid, it did not require any great insight that other sulphonic acids should perform the same function.

12.5 Perhaps another way to consider the question is as follows. If the team had to do the reaction and did not have a supply of MSA available but had plenty of [the sulphonic acid which CIPLA uses], they would think that it would be very reasonable to try out the [CIPLA alternative] and would do so expecting it to work in the same way as the MSA. …

12.6 Another way to judge the level of expectation of the likely effects of [the CIPLA alternative] is to consider what would have happened if the designer of the CIPLA process had submitted a paper for publication in a chemical journal disclosing that he had found that [the CIPLA alternative acid] keeps the reaction homogeneous. In the light of the publication of the Merck patent (and the paper by Kieczykowski referred to above), I am certain that such a paper would not be accepted in any respectable journal.

12.7 Finally, if the team were able to witness the CIPLA process being performed and could see that during it the reaction remained fluid and homogeneous, I believe it would be readily apparent to them that the reason why this was occurring was because the [CIPLA alternative] was having the same effect as the MSA. They would presumably have familiarity with the published work prior to the 152 Patent and realise that no solvents other than sulphonic acids seem to work."

23. Before turning to consider how these arguments were advanced and the basis for them, it is useful to have in mind what the claimant says is the true scope of the monopoly secured by the Patent. Mr Kitchin was keen to focus on the CIPLA process and to say that, whatever the Patent's scope, that process fell within it. However, as noted above, the proper approach is to construe the Patent and its claims in the absence of the infringement and only when this has been done, to look at the infringement. One of the advantages of this is that it allows one to appreciate the full breadth of the monopoly asserted.
24. Approached this way, it appears that the claimant's starting point is the suggestion that the core to the invention is using a suitable acid to keep the reaction mixture fluid and homogeneous. Based upon the evidence of Professor Scott, this means that the monopoly extends or should extend beyond the use of MSA to the use of any strong polar organic acid which, when mixed with the reaction ingredients, is not solid at the temperature at which the reaction to form alendronate takes place. In practice this means that the reference to MSA in feature (a) of Claim 1 should be notionally replaced by a reference to any sulphonic acid which is liquid at the relevant temperature. As far as stage (c) is concerned, the claimant relied on evidence to the effect that it was obvious that neutralisation could be carried out at temperatures higher than 20 – 25^oC, as long as the NaOH was cold enough when added to the reactor and was also in suitable dilution and that sufficiently effective stirring was used – all being interrelated factors which would prevent the NaOH/acid reaction, which gives out heat, from becoming too hot. On the assumption that all these other factors were suitably controlled, it was said that any temperature below about 80^oC would do. I do not think Mr Kitchin addressed the question of whether the lower limit of 20^oC could be lowered further. Finally, all the remaining features of the claim, that is to say all the features in (d) and (e) could be ignored and replaced by a requirement that the alendronate should be collected by any suitable means of crystallisation and then dried. There is no dispute that if this construction is correct, the CIPLA process infringes.
25. Needless to say, a claim written with the intention of securing this scope of protection would look nothing like Claim 1. But, argued Mr Kitchin, the Protocol on Interpretation requires the court to be "fair" to the patentee and that involves giving him protection for his inventive contribution so as to prevent competitors taking advantage of the invention without his permission. Obviously immaterial departures from what is claimed should be ignored. To do otherwise would be to construe the claim literally and that is prohibited by the Protocol.
26. It may be instructive to see how the claimant arrives at this conclusion. The starting point is the threefold test in Improver v Remington [1990] FSR 181 namely; (a) Does the variant have a material effect on the way in which the invention works? (b) Would the fact that that the variant had no material effect have been obvious at the date of the publication of the patent to a reader skilled in the art? (c) Would the reader skilled in the art have nevertheless understood from the language of the claim that the patentee intended that strict compliance with the primary meaning was an essential requirement of the invention? Based on this, at the very outset, and well before any evidence was served by either party, the claimant carried out certain experiments designed, so it was said, to address the first two Improver questions, particularly the first one.
27. The claimant's case was that it met each of the three Improver questions. However, by the time of the trial the claimant's case had changed or developed. As Mr Kitchin argued, even if his client fails to overcome the third Improver question, that was not an end of the case. On the contrary, he relies on Pharmacia v Merck [2002] RPC 775 at paras 40-41 for the proposition that in many chemical cases, of which this is one, the Improver questions are difficult to apply. His primary case is that Improver could and should be set to one side and that the fairness which the Protocol on Interpretation requires, as explained above, that I should give to the claimant a monopoly co-extensive with the invention the patentee has made, once the true scope of that invention has been identified by the skilled addressee of the Patent.
28. I put to Mr Kitchin a hypothetical case in which the patentee has hit upon a wrong theory as to how his invention works and, based upon that error, he states in his patent that certain things should be excluded from the scope of the monopoly. Both his erroneous understanding and his erroneous exclusions are set out explicitly in the patent. Furthermore, I asked him to assume in such a case that any ordinary skilled addressee would realise both that the stated theory was incorrect as were the exclusions and that the benefits of the invention were to be found in the excluded area. I suggested that in such a case it was impossible, whether using the Improver analysis or by using the Protocol directly, for the monopoly to cover the expressly excluded matter. Mr Kitchin did not agree. Even in such a case it might be fair to give the patentee all the protection which was necessary to prevent his technical contribution to the art being used by others without his permission.
29. Mr Floyd disagreed with all of this. Whether one used the Improver structured approach or simply applied the Protocol, his client did not infringe. He suggested that what Mr Kitchin was engaged in was not interpretation but redrafting.
30. The differences between the approaches of the parties are profound. But before addressing them, there is one other matter which should be mentioned. I have noted already that at the very outset and before any expert evidence had been exchanged, the claimant carried out certain experiments. By the time of the trial, the claimant had served Notices of Experiments in relation to 3 experiments. The first and last of these were the subject of repetitions requested and witnessed by the defendant. During the trial, not only were these three experiments relied on, but another nine experiments carried out by Professor Scott's team in an attempt to select the best operating conditions to be used for the claimant's last Notice of Experiments, were in issue.
31. Nearly every part of these experiments, including the ingredients used, the conditions employed, the methods adopted to measure results and the conclusions to be drawn was the subject of dispute between Professor Scott and the defendant's expert, Dr Cunningham. This, and challenges to the experts' respective views of what would or would not have been obvious to the notional skilled man in the art reading the Patent resulted not only in the production and service of 3 expert reports by each expert, but over 2 days of cross-examination and nearly 2 days of submissions by Counsel.
32. When the case came on for trial, I was surprised at the profusion of points which the parties intended to canvas. After all, this was an action which in July I had ordered should come on as a speedy trial in just a few months. The reason I did that was because, even on the very limited material before me at that time, it appeared that this was a dispute of no complexity which ought to have been capable of rapid resolution. It never occurred to me at that time that a trial of this weight would result. By the third day of the trial, my surprise had turned to alarm. I asked the parties' respective solicitors to prepare brief schedules of costs so that I could see what this trial was costing.
33. Those schedules, served on the fourth and last morning of the trial, were astonishing. They disclosed that the total costs incurred by the parties up to and including the last day of the trial exceeded £850,000. Of that, the claimant's costs exceeded £0.5M. All of this was for a trial in which the only issue was one of construction of a very short and easily comprehensible patent dealing which comparatively simple chemistry.
34. This raises two issues. First is the question of whether, whatever the law on construction of patents, this action was run in a proportionate manner. That is something I will return to at the end of this judgment. The other is whether this indicates that the approach to determining construction adopted in this trial, particularly by the claimant, is wrong. In my view, it does. One can stand back and think how courts construe other documents, such as contracts, deeds and wills. It is doubtful that in any of them would it be appropriate to have expert evidence of the breadth and complexity of the evidence served here. A patent is a document written by the patentee for publication to the world at large and designed not only to set out clearly what the invention is but to describe the monopoly sought in unambiguous terms. It is supposed to be comprehensible to members of the relevant trade simply on reading. If our law has reached the stage where experiments and extensive expert evidence is admissible to aid in construing patents, then it suggests that something has gone wrong.
35. Whatever the judicial attempts to give guidance, the bedrock on which construction must be based consists of the statutory provisions which determine the form of a patent and the function of its parts. They are to be found in the Patents Act, 1977 ("the Act") and the European Patent Convention ("EPC"). The Act provides as follows:

"s 14 (1) Every application for a patent –

(a) shall be made in the prescribed form and shall be filed at the Patent Office in the prescribed manner; …

(2) Every application for a patent shall contain –

(a) a request for the grant of a patent

(b) a specification containing a description of the invention, a claim or claims and any drawing referred to in the description or any claim; …

(3) The specification of an application shall disclose the invention in a manner which is clear enough and complete enough for the invention to be performed by a person skilled in the art.

(5) The claim or claims shall –

(a) define the matter for which the applicant seeks protection;

(b) be clear and concise

(c) be supported by the description, …"

36. It is to be noted that in accordance with the provisions of s 130(7) of the Act, the provisions of s 14(3) (the function of the specification) and 14(5) (form and function of the claims) are to have, as nearly as practicable, the same effects as the corresponding provisions of the EPC. The relevant Articles of the latter are 78, 83 and 84. It is not necessary to set them out. They are in substantially the same terms as the equivalent provisions in the Act. The only other section in the Act which needs to be referred to is s. 125 which, insofar as material, provides:

"(1) For the purpose of this Act an invention for a patent for which an application has been made or for which a patent has been granted shall, unless the context otherwise requires, be taken to be that specified in a claim of the specification of the application or patent, as the case may be, as interpreted by the description and any drawings contained in that specification, and the extent of the protection conferred by a patent or application for a patent shall be determined accordingly.

(3) The Protocol on the Interpretation of Article 69 of the European Patent Convention (which Article contains a provision corresponding to subsection (1) above) shall, as for the time being in force, apply for the purposes of subsection (1) above as it applies for the purposes of that Article."

37. Once again, s 130(7) stipulates that these provisions are framed so as to have the same effect as the corresponding provisions of the EPC. The relevant provisions of the latter are Article 69(1) and the Protocol on Interpretation ("the Protocol") which read respectively as follows:

"Article 69

Extent of protection

(1) The extent of the protection conferred by a European patent or a European patent application shall be determined by the terms of the claims. Nevertheless, the description and drawings shall be used to interpret the claims.

Protocol on the Interpretation of Article 69 of the Convention

Article 69 should not be interpreted in the sense that the extent of the protection conferred by a European patent is to be understood as that defined by the strict, literal meaning of the wording used in the claims, the description and drawings being employed only for the purpose of resolving an ambiguity found in the claims. Neither should it be interpreted in the sense that the claims serve only as a guideline and that the actual protection conferred may extend to what, from a consideration of the description and drawings by a person skilled in the art, the patentee has contemplated. On the contrary, it is to be interpreted as defining a position between these extremes which combines a fair protection for the patentee with a reasonable degree of certainty for third parties."

38. The purpose of a patent is to convey to the public what the patentee considers to be his invention and what monopoly he has chosen to obtain. These are not necessarily the same. The former is primarily to be found in the specification and the latter is primarily to be found in the claims. Although he is not deemed to be a patent lawyer, the patentee should be taken to be aware of the primary, and rather different, purposes of the specification and the claims when drafting his patent. So, the patentee must be taken to know the framework of form and purpose when he drafts his patent. It is his duty to communicate his invention and his assertion of monopoly to the public in language it will understand. He is warned by the Protocol, that his exclusive rights will not necessarily extend to everything which, from a reading of the specification, it can be seen that he contemplated. Furthermore the drafting of the specification and claims has to be considered against the background that no one is forced to apply for a patent or to seek as wide protection as possible. The patentee can be taken to be aware of the fact that there is always a balance to be achieved between width of protection and validity. It is up to the patentee to choose the level of risk he wishes to run.
39. Once the patent is drafted it is to be construed by the notional skilled reader. In doing that, he is to assume that the patentee is also a man skilled in the art. Mr Kitchin doubted the latter point, but that point seems to have been dealt with by Lord Diplock in Catnic Components Limited v Hill & Smith Limited [1982] RPC 183, 243 line 21. It follows that a patent is to be taken as written by a man skilled in the art for reading by a man skilled in the art and using the language of that art. It is because of the latter point that courts accept expert evidence concerning the meaning of technical terms. It appears to me that there is only one qualification to the presumed expertise of the patentee. He must be deemed to match the notional skilled addressee save to the extent that what he writes in the patent shows him to have a greater or lesser familiarity with or knowledge and understanding of the art.
40. For the patent system to work fairly to the public, the monopoly sought by the patentee must be discernible to the reader when he reads the patent. He must be able to understand the limits of the prohibited field before he sets out to make a competing product or install a competing process. It cannot be right that he needs to carry out experiments to determine the width of the monopoly. It seems to me that that accords with common sense and also with the decision of the Court of Appeal in Sara Lee Household & Body Care Limited v Johnson Wax Limited [2001] EWCA Civ 1609 at para 36.
41. In drafting the specification and claims, the patentee can choose language as wide and narrow as he likes. This is particularly easy when scientific language is employed. To take the example of the acid + base reaction referred to in paragraph 6 above, he can simply refer to reacting an acid (meaning any acid) with a base (meaning any base) and he can leave out all process conditions such as the quantities reacted, the duration of the reaction and the temperature at which the reaction is to be conducted. He can limit himself to a group of acids or bases, the size of the group being defined by him, or to a single acid or base – such as HCl and NaOH. If he chooses to refer to the temperature at which the reaction is carried out, he may simply say it should be above or below a particular temperature, or between particular temperatures or at a specific temperature. All of these linguistic tools have been used in the Patent in suit. In addition to this, the patentee can choose to explain the mechanism or theory underlying his invention assuming, of course, that he believes he understands why it works. All these drafting possibilities are available to the patentee and that they are so will be appreciated by the notional addressee.
42. Notwithstanding the adaptability of scientific language, the patentee is not expected to be omniscient or to exhibit super human thoroughness in drafting. He may not be able to foresee future developments which will be useable with his invention but which make no material difference to the way it works. For example, the invention may relate to a mechanical device in which two parts are connected together. A new method of connection might not be covered by the exact words of the claim, but the notional reader would be reasonably confident that the patentee would have intended to cover the new method as well. Furthermore the patentee may choose a form of language which emphasises which features of an invention are important and which are not. For example it is common to find claims which start with general description followed by "characterised in" followed by a list of features. The addressee would appreciate that the latter features are particularly important but the features before the words "characterised in" are less so. If there is a variant to the latter which obviously does not affect the way in which the invention works, the notional reader may be reasonably confident that the inventor wanted to cover this variant as well. In these types of cases, the monopoly is likely to extend to the new variant.
43. In my judgment, the same principles underpin the Protocol. The aim is always to determine objectively from the words used particularly, but not necessarily exclusively, in the claim to what the patentee wanted to claim exclusive rights. The Protocol is directed at allowing protection for the discernible intention of the patentee, to be derived from the words used to express that intention.
44. Thus the first sentence in the Protocol is clearly directed to prohibiting acontextual construction of the claims. It says expressly that the context may be looked at even when no question of ambiguity of the claim arises. What it does not say, or suggest, is that by looking at the claims in the context of the specification the exclusive rights can be extended beyond that which the patentee wanted to cover.
45. The second sentence in the Protocol emphasises the primacy of the claim. The notional addressee is not expected to find the patentee's presumed intention from the specification. In particular, as noted already, this sentence makes clear that just because a reading of the specification suggests that the patentee "contemplated" a wider protection than that set out in the claims does not mean that he obtains that wider protection. The patentee is taken to know the rules. If he wants a monopoly which covers all embodiments which he contemplates may make use of his technical contribution to the art, he should use language which conveys that intention to the notional reader and that language should primarily be found in the claims. It seems to me that this part of the Protocol conveys the same message as is to be found in the speech of Lord Cairns in Dudgeon v Thomson (1877) 3 App Cas 34 when he said that there is no such thing as infringing the equity of a patent.
46. If there is any difficulty it resides in the last sentence of the Protocol which reads:

"On the contrary, it is to be interpreted as defining a position between these extremes which combines a fair protection for the patentee with a reasonable degree of certainty for third parties."

47. As I have stated above, Mr Kitchin argued that the requirement for fairness means that the court should give as much protection as it feels is justified having regard to the inventive contribution made by the patentee even if that includes embodiments which the patentee intended to exclude, for example because he misunderstood how his own invention works. I reject that submission. The courts are not a branch of social services whose job it is to help the infirm or the unwise and the Protocol does not require them to be so. There is no cannon of construction which would justify the courts in granting a patentee more protection than that which, objectively assessed, he indicated he wanted. Indeed to do so would not be "fair" to the patentee. It could expose him to a greater risk of invalidity than he was prepared to shoulder (see American Home Products Corporation v Novartis Pharmaceuticals UK Ltd [2001] RPC 157 para 30). So, when it talks of "fair" protection for the patentee, the Protocol can only be referring to protection which the patentee intended to obtain, such intention being objectively assessed. To the extent that such fair protection exceeds what is explicitly set out in the claims, it will not be given save to the extent that it gives reasonable certainty to the public.
48. It seems to me that what the Protocol requires is that the monopoly should cover all embodiments, whether explicitly mentioned in the claims or not, which the notional skilled reader would conclude, with reasonable confidence, the inventor wanted to cover. Where it is clear that the patentee did not intend to obtain protection for particular variants, it is not open to the court to extend the monopoly to cover them. Similarly, if a notional skilled addressee cannot conclude with reasonable confidence that the inventor wanted to obtain protection for a particular embodiment, it must follow that the patent conveys the message that the patentee might well have intended to exclude that embodiment. To give protection in such circumstances would run the risk of going against the intention of the patentee, thereby being unfair to him, and would not be giving third parties a reasonable degree of certainty as required by the Protocol.
49. Determining whether a skilled reader would conclude with reasonable confidence that a particular embodiment was one the patentee wanted to cover involves assessing all the facts of the case. The wording of the claims is the most important one, but is not necessarily determinative. Matters such as the way the inventor describes his inventive contribution and his explanation, if any, of how the invention achieves its claimed results are matters to be taken into account. The factors, and how they interrelate to each other, will vary from case to case.

The Improver questions

50. How does this approach marry up with the Improver questions? As the Court of Appeal said in Pharmacia, the Improver questions are "normally useful tools" but they may not be easy or appropriate to apply in every case. In the end one must return to look at the Protocol itself. There is nothing to suggest that Hoffmann J intended to put the three questions forward as a rigid checklist with three boxes, each of which needs to be ticked appropriately for there to be infringement. The Improver questions may well provide a useful structured approach to the issue of infringement, just as the Windsurfing test does in relation to obviousness. They are not a substitute for the Protocol.
51. It should be recalled that in Improver, Hoffmann J was applying the approach to purposive construction explained by Lord Diplock in Catnic (supra). It will be recalled that there it was said that the question whether limitations in the claim were to be ignored;

"Is to be answered in the negative only when it would be apparent to any reader skilled in the art that a particular descriptive word or phrase used in a claim cannot have been intended by a patentee, who was also skilled in the art, to exclude minor variants which, to the knowledge of both him and the readers to whom the patent was addressed, could have no material effect upon the way in which the invention worked." ([1982] RPC 183, 243)

52. It seems to me that the approach of reasonable confidence referred to above is consistent with this passage in Lord Diplock's speech. Unless it is apparent that a particular limitation in the claim "cannot have been intended by the patentee", the limitation is effective. If the notional reader cannot determine from a reading of the patent whether the limitation was intended or not, it cannot be apparent that the patentee intended it not to be so. In such a case the limitation cannot be ignored.
53. It is the first two Improver questions, and particularly the first, which were said to justify the carrying out of experiments in this case. As Hoffmann J said, the answers to these two questions;

"… are used to provide the factual background against which the specification must be construed." ([1990] FSR 181, 190)

54. That, however, does not mean that it is legitimate to carry out experiments to prove these matters. As the judge said on the same page of the report;

"It is the third question which raises the question of construction and Lord Diplock's formulation [in Catnic] makes it clear that on this question the answer to the first two questions are not conclusive."

55. The first Improver question does not create a factual dispute which needs to be resolved in order to determine whether an embodiment outside the acontextual scope of the claim is within the monopoly created by the patent. It is only question (3) which determines construction. In most cases the answer to the first question will not be in dispute. If the variant does not work in the same way as the invention, it cannot be within the scope of the patent. The patentee could not have intended otherwise. Thus a negative answer to this question will inevitably lead to the third and crucial question being answered in a way which excludes the variant from protection. But what if it is unclear to those skilled in the art whether the variant works in the same way as the invention, either because the way the invention works is not clear or the way the variant works is not clear, or both? In such a case it is impossible for it to be apparent to the reader that the limitations in the claim were not intended by the patentee. In other words, adopting the purposive construction set out in Catnic, the issue of construction must be answered so as to exclude the variant from protection. Much the same analysis applies to the second Improver question. Where it is not obvious that the variant works in the same way;

"… the reader is entitled to assume that the patentee thought at the time of the specification that he had good reason for limiting his monopoly so strictly and intended to do so …" (per Lord Diplock in Catnic p. 243

56. Neither of these questions justifies the carrying out of experiments. At the most the answer to these questions provide shortcuts to the only important question, namely question (3).
57. For the above reasons, if one is adopting the structured approach, it is only Improver question (3) which needs to be answered. However, it appears to me that there is one respect in which that question needs to be approached with caution. In formulating the questions, Hoffmann J said he was applying the guidance in Catnic. There is no doubt that the binding authority on construction in this jurisdiction is the latter case. That was confirmed by the Court of Appeal in Kastner v Rizla [1995] RPC 585. Yet there appears to be a potential difference between the way the issue of construction was put in Catnic to the way it is explained in Improver. Imagine the case where the notional skilled reader does not understand why the patentee put a limitation in the claim. According to Catnic, in such a case the limitation is effective because it is not "apparent" that the limitations "cannot have been intended by the patentee". The variants will be excluded from the monopoly. By contrast the same facts would be answered differently if Improver question (3) is applied rigidly. Where the reason for introducing the limitation is unclear, the skilled reader could not say that he understood "that the patentee intended to confine his claim". Thus Improver question (3) would be answered in the negative and the claim will be construed to include the variants. I think that this difference is more imagined than real. At p. 197 of Improver it appears to have been accepted that if the notional skilled addressee would speculate that the patentee had good reason for including the limitation in the claims, the limitation is effective. In other words the Catnic approach was adopted.

The scope of protection in this case

58. The scope of protection advanced by Mr Kitchin is referred to at paragraph 24 above. The wording of the specification and the claims are clear. There is no suggestion that either contain any ambiguities. There is no dispute as to the meaning of technical terms nor is it suggested that there is anything particularly difficult or esoteric about the chemistry referred to in the specification. In those circumstances, I cannot see why the construction of this patent requires any expert evidence. If it is appropriate, as I think it is, to construe the patent by asking whether the notional reader would be reasonably confident that the patentee would have intended to cover the expanded process argued for by Mr Kitchin, no evidence is needed.
59. Adopting this approach, it is convenient to consider the scope of protection in two parts. First is the requirement in Claim 1 that the reaction be conducted in the presence of MSA. It is central to the claimant's case of infringement that the protection extends well beyond the use of MSA to the use of any sulphonic acid which is liquid when in mixture with the other ingredients and at the temperature of the reaction. It is unclear where the requirement that the acid be liquid at the reaction temperature draws the dividing line between included and excluded acids, because according to the claim, the temperature at which the reaction is to be carried can vary. Only the upper limit, 85^oC, is set.
60. Even ignoring this, I am not persuaded that the patent should be construed so as to give protection for the wider family of sulphonic acids. The claimant is caught on the horns of a dilemma. If, as the claimant asserts, it is obvious that other sulphonic acids besides MSA would work in the same way, then there are only two possibilities. Either the patentee, sharing the knowledge and skills of the notional addressee, must have been aware of this or he was not. In the former case, the decision to refer only to MSA in the specification and claims is likely to have involved a decision on his part not to seek protection for all the other sulphonic acids. In the latter case, he was not aware of the possibility of using any other sulphonic acid and in this respect the claim accurately reflects what he wanted a monopoly for. It is not necessary to consider which of these two possibilities is the most likely, but it is noticeable that the patentee not only does not refer to any other sulphonic acid or even hint that any other acid might work, but he ventures no opinion as to why MSA works and why the prior art failed. That lacuna is not filled by looking at the prior art referred to expressly in the patent. In particular, as noted above, US Patent No 4,407,761, which uses chlorobenzene as the solvent and in which the product goes solid, does not refer to this as a defect, does not address the question of why it solidifies nor does it touch upon the reason why the solvent does not work (assuming, of course, that solution of the end product was something which that inventor was trying to achieve – which is not what the US patent suggests). In my view the notional reader could not be reasonably confident that the inventor wanted to cover sulphonic acids other than MSA. To adopt the words of Lord Diplock in Catnic, the reader is entitled to assume that the patentee thought at the time of the specification that he had good reason for limiting his monopoly to MSA and intended to do so. In those circumstances, to widen out the protection is not an act of fairness to the patentee, nor does it give reasonable certainty to third parties.
61. The second part of the scope of protection in relation to which the parties disagreed relates to the detailed processing steps set out in stages (c) to (d) of the claim. Mr Kitchin's argument was that all of the specifics in this part of the claim can be ignored. The crucial step of keeping the reaction mixture fluid and homogeneous has been achieved by the use of the sulphonic acid. What remains are unimportant and easily changeable collection and washing steps. It would be obvious to the skilled addressee that not only the precise details of stages (c) to (d) could be changed within wide limits, but some of these stages could be replaced by equivalent and known alternatives. I do not accept this submission either.
62. The wording of the patent is consistent with the patentee understanding the available ways in which he could choose the width or narrowness of the integers to be put in his claims. Most of the numerical limits in claim 1 have been derived from Example 1 yet there are many other processing details in that example which have not been transposed into the claim. The patentee clearly has selected which processing conditions to use to limit the scope of the claim. Furthermore, within the claim itself the patentee has chosen to impose different degrees of specificity. Thus feature (a) only requires the use of a reaction temperature below 85^oC. No lower limit is imposed. By contrast the temperatures to be employed at stages (c) and (d) are both much narrower and circumscribed by precise upper and lower limits. Similarly, although there are no quantities or concentrations given for the GABA, phosphorous acid, PCl₃ and MSA in stage (a) or in relation to the sodium hydroxide in stage (e), there is a precise concentration of the ethanol given in stage (e) and, as with the MSA, no reference to any other alcohol or other agent besides ethanol.
63. In addition, reading the specification reinforces the view that the patentee, may well have intended to give the process claims a comparatively narrow scope. The stated objectives of the patent include obtaining high purity and yield, capable of being delivered in a one-pot reaction. The worked example shows that these objectives can be achieved. However there is nothing to suggest that the patentee has considered, let alone determined, how altering the operating procedures or replacing them with alternatives affects purity or yield and by how much. For example, changing temperature of the neutralisation stage (c) to be higher or lower than the 20 – 25^oC range specified might affect either of these objectives or, more precisely, the patentee might have feared that to be so. Similarly there is nothing to suggest that he contemplated using anything other than 95% ethanol in stage (e). According to the specification the patentee carried out a single experiment. It would have been impossible to draw any conclusions from that as to how variations of operating procedures could affect purity and yield and by how much. Finally, there is nothing in the way in which the specification or claims are written which suggests that the patentee considered stages (c) to (d) to be in any way less important than, say, stage (a). This is not a case where the patentee has used any of the well known drafting techniques which can be employed to highlight the essential part of the invention. The patent reads as if the patentee believes he has found a single "packet" of interacting ingredients and processing conditions which together give him, in high yield, the high quality product he wants. It follows that the notional reader cannot have reasonable confidence that the patentee wanted to cover processes in which the detailed parameters in stages (c) to (d) were changed. For these reasons, in relation to this part of the claim as well, there is no basis for saying that expansion of the protection is necessary out of fairness to the patentee nor would any such expansion produce reasonable certainty to third parties. I should only add in relation to the last point that it is quite impossible to determine from a reading of the patent how far it is possible to change the features of stages (c) to (d). It would be impossible for a notional addressee, working on the material in the patent and matters of common general knowledge, to know how far from the clear limits in the claim it is possible to go. To replace clear limits by something which is hopelessly vague would not be to produce reasonable certainty for third parties as required by the Protocol.
64. For the above reasons, the importation of products made to the CIPLA process do not infringe the Patent.

Applying the Improver questions in this case

65. Since this case may go higher, it may be helpful to analyse the issues in accordance with the structured approach in Improver insofar as that is possible. In doing this it will be necessary to have regard to the evidence.
66. I should start by setting out my assessment of the two experts. In my view both were admirably clear and fair. They were both obviously highly knowledgeable. Mr Kitchin suggested that Dr Cunningham was "gloomy" which accounted for his reluctance to see the obviousness of the widened applicability of the process in the Patent. I do not think this is fair. Dr Cunningham struck me as thorough and cautious, but not unreasonably so. Professor Scott was also impressive. He is a natural enthusiast and that shone through his evidence. He was more prepared to express opinions about the likely course of experiments he had not conducted and the probable impact of changing reaction ingredients or conditions but I do not think that this was because he was engaged in reckless speculation. Rather it was because he considered that he had a clear understanding of how the process in the Patent worked and this enabled him to be confident as to how it could be modified to achieve comparable results. I doubt whether the notional skilled addressee would have had quite as much confidence.
67. As Hoffmann J said in relation to Improver question (1), whether use of a product or process outside the acontextual scope of a claim has a material effect on the way the invention works depends on the level of generality at which one describes the way the invention works. If, in this case, the inventions consists simply of running a process for making alendronate in which the reaction is maintained in a fluid and homogeneous state, it is not disputed that using other sulphonic acids (as long as, when mixed with the other reagents, they are molten at the reaction temperature) will work the same way. Furthermore it is also accepted that the processing and crystallisation features of claim 1 can be modified within fairly wide limits without preventing the production of alendronate, at least in moderate quality and at moderate yield. However, if the invention consists of running a process for making alendronate in which the reaction is maintained in a fluid and homogeneous state so as to secure high purity and a yield of at least 85 – 90%, it is by no means clear that making the alterations advocated by Mr Kitchin will result in a process which works the same way.
68. This issue was the subject of the claimant's Notice of Experiments for experiment No. 3. The second experiment contained in the first Notice of Experiments was designed to show that replacing MSA by CIPLA's sulphonic acid resulted in a homogeneous reaction mixture producing alendronate. It did that, but at a yield of about 40% and with comparatively low purity. Unsurprisingly, the defendant did not ask for this experiment to be repeated. As a consequence the claimant relied on its experiment No. 3 which was designed so as to optimise the yield when CIPLA's sulphonic acid is used. This experiment was repeated under inspection by the defendant. The yield in the Notice of Experiments was 77%. On the repeat it was 75%. In fact it transpired that there was a series of 9 experiments conducted by Professor Scott's team before experiment No. 3 was included in the second Notice of Experiments. Those experiments were designed to work out how the yield could be increased without significantly compromising the quality of the alendronate produced. One experiment managed to achieve a yield of 84.6% but the product was much more impure. The last in the series became the one contained in the second Notice of Experiments.
69. Professor Scott was confident that, given time, he could increase yield. He put it as follows in his first report:

"I have no doubt that further small modifications could increase the yield even further. I would be very surprised if, having increased the yield from 43% to 77% by the changes referred to above, it were not possible to increase it any further. There does not appear to be any fundamental reason why the yield would not go any higher - for example we have not detected the products of any competing side reaction. If I were asked to further optimise this process I would have several avenues for investigation, such as those below. I would be disappointed if we could not improve the yield by a further 5-10% by application of … simple modifications." (para 11.20)

70. He suggested three modifications. In order of decreasing importance they were; (i) use of a better grade of the sulphonic acid, (ii) increasing the level of the sulphonic acid slightly and (iii) increasing the temperature. Dr Cunningham did not share Professor Scott's confidence. This dispute resulted in a significant amount of cross-examination. As far as modification (i) was concerned, Professor Scott said in his first report that the sulphonic acid used in the experiments was technical grade. This would contain water which "may be itself affecting the yield" (First report para 11.20(a)). Under cross examination it transpired that he was not sure how much water there was in the sulphonic acid he used nor did he express an opinion as to how much the control of this impurity could be expected to increase yield. On the question of increasing the sulphonic acid level, it appears from some of the experiments that this is likely to increase the impurity levels also. In the end, the Professor said that he did not know whether the amount of acid would or would not increase yield. As far as item (iii) is concerned, the idea that the reaction temperature could be raised ran into the difficulty that this was contrary to the teaching of the specification to keep the reaction temperature down and, again, the extent to which, if at all, this might raise yield was not clear. I have come to the conclusion that the Professor's confidence would not be matched by the man skilled in the art. Of more importance to the point now under consideration, none of this evidence demonstrates on a balance of probabilities that CIPLA's sulphonic acid could be substituted for MSA while still achieving the yield and high purity which the patented process promises. It follows that if, as I believe, the invention works so as to produce alendronate at high purity with a yield of at least 85 – 90% yield, then the use of CIPLA's sulphonic acid is not shown to work in the same way. There is no evidence before me which would allow me to come to a different conclusion in relation to any other sulphonic acid other than MSA. This means that the claimant fails at the first hurdle of Improver.
71. I should just add that the nature of the CIPLA process itself is consistent with these doubts as to the ability of other sulphonic acids to perform as well as MSA. As the Patent points out, one of the advantages of the patented process is that it achieves high yield and purity in a one-pot reaction. In other words, it produces such high yields and purity that it is not necessary to have a distinct purification step. The whole process can be pushed through to completion in one go. It appears that in the CIPLA process the same high yield at high purity was not achievable without the removal of the collection stage (stage (e) in the claim) and its replacement by a more sophisticated and separate purification process.
72. On the assumption that I am wrong in relation to Improver question (1), I can turn to question (2). Would it have been obvious to a man skilled in the art that the variants asserted by the claimant would work in the same way as the process expressly covered by claim 1?
73. Professor Scott says it is obvious what is going wrong with the prior art and how the claimant's MSA-based process overcomes the problems. Based on that, he says that it is obvious that MSA can be replaced by certain other sulphonic acids and that the rest of the process steps can be replaced by other equivalent or alternative steps without materially altering the way the process works. As noted already, neither the prior art nor the Patent touch on these matters, save that the latter points to the fact that the prior art processes go solid and are inhomogeneous. The prior art does not acknowledge that a problem exists. The Patent offers no explanation for why solidification occurs, nor how it is solved. In the circumstances it is not surprising that it does not touch on how its specific solution can be extended to the use of alternatives to MSA and other processing steps. There is no suggestion that it can be so extended. Dr Cunningham did not accept that the mode of operation of MSA or the extensions suggested by the claimant were obvious or would obviously work in materially the same way. In relation to the latter point it was not obvious that MSA could be replaced by another sulphonic acid without, at the same time, running a significant risk that the purity and high yields promised by the Patent would be compromised.
74. The disagreement between the experts on this issue led to extensive cross examination. An outline of the claimant's argument is as follows. The first step is to understand that it is the lack of homogeneity in the prior art processes which result in low yields and purity. This is caused by the difficulty in stirring a solid reaction mass. This is referred to in the Patent. The solution is to ensure that the reaction mass remains in a fluid and stirrable condition. The chlorobenzene solvent referred to in the prior art fails to keep the system fluid. The explanation for the creation of a solid is as follows. The structural formula of alendronic acid given at paragraph 1 above, shows that it has a positive charge at one end and a negative charge at the other – such a molecule is sometimes referred to as a zwitterion. This allows an alendronic acid molecule to react with another and so on to form large molecules consisting of chains of alendronic acid units connected together. This is a polymer. Because polymers are large molecules, they are less soluble than the individual building blocks from which they are formed. Thus the product of the reaction is a solid. In the prior art, the polymer is eventually broken down by boiling with an acid which severs the bonds holding the building blocks together (a hydrolysis reaction) to release alendronic acid.
75. It is then said that, once one has read the description in the Patent, it is obvious that the MSA is acting as follows. The polymerisation is being prevented. The basic (-NH₃+) end of the alendronic acid molecules are prevented from reacting with the acidic (-PO₃H^-) part of other alendronic acid molecules. Instead they react with the very strong acid, MSA. The salt created by the reaction of MSA with alendronic acid is, in turn, soluble in MSA. Thus, as the alendronic acid molecules are produced by the reaction between GABA and the other starting materials, instead of reacting amongst themselves and producing a solid mass which interferes with the rest of the reaction to produce more alendronic acid, a salt of the alendronic acid is formed which is kept in solution and it is possible to maintain homogeneity. It would be obvious that this is achieved by the activity of the sulphonic group in the MSA. It would therefore be obvious and expected that substantially the same effect could be achieved by using other sulphonic acids. Since very large sulphonic acids have high melting points, if they are introduced into the reaction mixture they may make it go solid because the sulphonic acid itself is solid at the temperature at which the reaction is being conducted. Thus, other sulphonic acids, including the one used by CIPLA, can be used to achieve substantially the same result save that the list of alternatives will not include those sulphonic acids which have too high melting points. Furthermore, since it is obvious that it is the use of the sulphonic group in the MSA which is responsible for the high yield and high purity, the remaining process steps would be regarded as non-critical and capable of being replaced by readily available alternatives to secure similar yield and purity.
76. This is very much a précis of the thought processes which, the claimant argued, leads one to answer Improver question (2) in the affirmative. The full sequence of analysis thinking is rather more complex. Much of it was set out in a document entitled "Merck's note on evidence" which Mr Kitchin helpfully produced to support his closing submissions. To avoid cluttering this judgment, a version of it, modified so as to hide the identity of the reagents used by CIPLA is annexed hereto.
77. The cross examination of Dr Cunningham followed the familiar step by step course referred to by Lord Diplock in Technograph Printed Circuits Ltd v Mills & Rockley (Electronics) Ltd [1972] RPC 346, 362. Mr Kitchin conducted it, as one would expect, with great skill. Although Mr Kitchin managed to get Dr Cunningham to agree with him on some of the steps, he failed to shake him from his view that, overall, the man in the art would not consider the substitution of MSA by other sulphonic acids to be obvious and certainly not if what was wanted was a process which produced equivalent purity and yield to that described in the Patent. It was because of that that Mr Kitchin described Dr Cunningham as gloomy.
78. I was unconvinced by this cross examination. Furthermore, I think it is important to appreciate that the test of obviousness in Improver question (2) cannot be the same test as used to invalidate a patent over published prior art. In the latter case it is enough to demonstrate that the reader of the prior art found the prospects of achieving the desired result sufficiently encouraging to warrant trying it out, even if there are commercial reasons why going down that route is unattractive (see Brugger v Medic-Aid [1996] RPC 635, 661). Where, however, one is attempting to broaden the patent monopoly to cover variants which are not within the acontextual meaning of the claims, a higher degree of confidence of success must be involved. The reader must have little or no doubt that the variant will, not may, work in the same way to produce the same results. In my view, the distinction between the obviousness test for invalidity and the obviousness required for Improver question (2) is apparent from AHP paragraphs 28 and 29. The claimant gets nowhere near demonstrating the relevant level of confidence in this case. For example, for Professor Scott to say that he would be disappointed if he could not improve the yield and purity were another sulphonic acid to be used instead of MSA does not mean that it is obvious that such improvement would be achieved.
79. In my view it is not proved that the notional skilled reader would have even thought of substituting another sulphonic acid for MSA. The hesitation of Dr Cunningham represented more closely the approach of such a reader. Furthermore, it is not proved that any skilled worker would have considered it obvious that replacement of the MSA by another sulphonic acid would be so effective that it would achieve the same, or substantially the same, level of purity and yield promised by the Patent. He would not have been confident that such substitution would produce such a yield and purity that the production of alendronate could be achieved in a one-pot process as promised by the patentee, rather than with a separate extraction and purification step (as in the CIPLA process). Improver question 2 is answered in the negative.
80. It is not necessary to consider Improver question 3 on the assumption that my conclusions on the first two questions are wrong, since that has been dealt with in substance earlier in this judgment.
81. It follows that, whether one follows the direct route or the structured Improver route, the same conclusion is arrived at. The case of infringement fails.

The cost of this action.

82. There is one core issue in this case. Do products made by the CIPLA process infringe claim 1 notwithstanding the fact that that process fails to use the major ingredient and does not use four of the five identified stages referred to in the claim? This apparently simple issue has been allowed to grow into a four day trial costing over £850,000. That cannot be an appropriate way of conducting litigation.
83. There is a school of thought that it is no part of the court's function to tell parties how to conduct their litigation and, in particular, to impose its will on the parties when it comes to costs. If parties wish to spend large sums on litigation, the choice is theirs. That belief is based on an assumption, which I doubt is always justified, that the parties have a free and informed choice when it comes to spending money on litigation. Furthermore it does not pay regard to the fact that sometimes one party may want the costs to be large so as to drain or deter its opponent. In such cases, costs can become an instrument of war. I am not suggesting that this latter consideration applies in this case. But, these qualifications aside, I accept that a judge is ill advised to make comments about charging rates or fees in the absence of well accepted standards. Fees reflect what the market will bear. Competition, not the judges, should act as the restraint.
84. The cost of patent litigation has been a source of concern here for many years. It is worth remembering what Lord Esher said in Ungar v Sugg (1892) 9 RPC 113:

"Well, then, the moment there is a patent case one can see it before the case is opened, or called in the list. How can we see it? We can see it by a pile of books as high as this [holding up the papers] invariably, one set for each Counsel, one set for each Judge, of course, and by the voluminous shorthand notes: we know 'Here is a patent case.'

Now, what is the result of all this? Why that a man had better have his patent infringed, or have anything happen to him in this world, short of losing all his family by influenza, than have a dispute about a patent. His patent is swallowed up, and he is ruined. Whose fault is it? It is really not the fault of the law; it is the fault of the mode of conducting the law in a patent case. That is what causes all this mischief." (p 116)

85. The fact that it is not the court's task to intervene in relation to the question of fees does not relieve it of its duty to control the way in which litigation is conducted. Lord Esher was right over a hundred years ago when he said that what is wrong is the manner in which patent litigation is conducted. If our system allows parties to leave no stone unturned in the course of litigation, it is not really a surprise that they do so, particularly when the stakes are high, and even if some of the stones which they turn are no bigger than grains of sand.
86. Whatever the position may have been in the past, it is no longer the case that the parties and their lawyers have a free hand to conduct an action as they like. Under CPR Part 1.2 the court "must" seek to give effect to the overriding objective when it exercises any power given to it by the Rules or when it interprets rules. It should be remembered that the overriding objective is stated as follows:

"1.1 – (1) These Rules are a new procedural code with the over-riding objective of enabling the court to deal with cases justly.

(2) Dealing with a case justly includes, so far as is practicable –

(a) ensuring that the parties are on an equal footing

(b) saving expense;

(c) dealing with the case in ways which are proportionate –

(i) to the amount of money involved;

(ii) to the importance of the case;

(iii) to the complexity of the issues; and

(iv) to the financial position of each party;

(d) ensuring that it is dealt with expeditiously and fairly;

(e) allotting to it an appropriate share of the court's resources, while taking into account the need to allot resources to other cases."

87. In my view, the fact that the court must apply the rules with a view to saving expense, dealing with a case in a manner which is proportionate, ensuring that cases are dealt with expeditiously and in a manner which takes into account the need to allocate resources to other cases makes it clear that the court should not allow the parties to conduct the litigation in a bubble where the only considerations are what the parties and their lawyers want. Anyone with experience of litigation knows that clients' (and sometimes their lawyers') horns can become so locked with their opponents that their desire to win blinds them to what is at stake. While the parties must continue to have a major say in the way the litigation is run, this is subject to the court's case management responsibilities under the CPR. For a case which could have been determined well within in a day to run for four suggests that there has been a failure to properly implement the restraints inherent in the overriding objective. The question to be answered is what steps should be taken to avoid a repetition.
88. Some of the costs incurred and time expended in this case are attributable to the conduct of experiments directed to an issue of construction. I have referred to this already. At the case management conference the claimant indicated that it might carry out some experiments. Mr Speck, for the defendant expressed some misgivings but, because the proceedings were at an early stage and the claimant had not disclosed precisely what experiments it intended to conduct, he was not in a position to develop his concerns. I also expressed misgivings. Unfortunately, directions were given which included the usual provisions as to the conduct of experiments. In my view no such automatic provision should have been included. Perhaps in this case it would have been better if, in July, I had asked the claimant for what experiments it had in mind. In the future, no experiments should be conducted going to Catnic-type questions of construction unless the court has given informed permission for them in advance. I have little doubt that if I had been told what experiments were planned, I would have refused permission for them to be conducted.
89. In addition to this, the court has a number of powers which allow it to control how litigation is conducted. These are conveniently referred to in the Chancery Guide. Paragraph 3.1 points out that the court's case management powers include a power to fix timetables, control costs and give directions to ensure the trial proceeds quickly and efficiently. Paragraph 8.1 states that;

"The court may, either at the outset of the trial or at any time thereafter, fix time limits for oral submissions, speeches and the examination and cross-examination of witnesses."

90. Furthermore paragraph 8.16 confirms that;

"The trial judge may disallow expert evidence which either is not relevant for any reason, or which he or she regards as excessive and disproportionate in all the circumstances, even though permission for the evidence has been given."

91. These provision should be read in conjunction with CPR Part 32.1 which gives the court power to control what evidence is adduced and how, and to limit cross-examination and CPR Part 31.5.2 which gives the power of the court to dispense with or limit standard disclosure. All of these powers can and should be used to prevent a case being conducted in a disproportionate manner.
92. These general powers lay behind the Patents Court Streamlined Procedure which was introduced in the Patent Court Guide of August of this year (unfortunately after the case management conference in this case). The procedure is defined as follows:

"Streamlined Procedure

(a) Nature of a streamlined procedure.

A streamlined procedure is one in which, save and to the extent that it is otherwise ordered:

i all factual and expert evidence is in writing;

ii there is no requirement to give disclosure of documents;

iii there are no experiments,

iv cross-examination is only permitted on any topic or topics where it is necessary and is confined to those topics;

v the total duration of the trial fixed is and will normally be not more than one day;

vi the date for trial will be fixed when the Order for a streamlined trial is made and will normally be about six months thereafter.

A streamlined procedure also includes minor variants of the above (e.g. disclosure confined to a limited issue)."

93. Two things should be noted about this procedure. First, although it may be sought by either party, the court has the power to impose it of its own motion. The court will take this power seriously. In some cases that may mean that the court will, at an early stage, require the parties to give it a fuller understanding than normal of what the issues are so that it is better placed to decide whether to impose the streamlined procedure. Second, the Guide makes it clear at paragraph 10 (e) that the parties' legal advisers are obliged to draw their clients' attention to the availability of this procedure. It must be expected that the court will take that requirement seriously and will monitor compliance with it.
ANNEX

In this note, CIPLA's sulphonic acid will be referred to as "Acid X"

94. The skilled person would know the following:
95. A variety of approaches had been taken to keep the reaction fluid in the prior art. These had failed. In particular the chlorinated hydrocarbons did not work. The skilled person would therefore appreciate that polar aprotic solvents were ineffective.
96. In contrast, MSA does work and is not itself phosphonylated.
97. The skilled team would, upon seeing the Patent, ask what was it about MSA that allows it to keep the reaction fluid.
98. It is clear that it is the sulphonic acid that is making the difference. It is the engine room of the molecule. That is so in relation to both the chemical and physical properties of the molecule.
99. The skilled team knows of the following characteristics of MSA: (a) its melting point. It is a liquid which melts at or just below room temperature (20°C), (b) it is a very polar molecule, (c) it is a very strong acid and (d) it is the sulphonic acid group that is responsible for the polarity and acidity of the molecule.
100. Without the use of MSA in the process, a solid is formed which, on hydrolysis, is converted to alendronic acid. Therefore, the solid must have been an oligomer (dimers, trimers etc) formed by condensation reactions.
101. The skilled team would also know that use of MSA prevents solidification.
102. In relation to its chemical reactivity consider GABA and alendronic acid. Both are amino acids and contain a basic group, NH₂. It is basic chemistry (GCSE level) that an acid reacts with a base to produce a salt.
103. If the GABA or alendronic acid are not in the form of zwitterions, the amine will be protonated by the MSA to give NH₃. The counterion will be MSA.
104. If they are in the form of zwitterions, the position is as explained by Professor Scott in his drawing Xl and the same compound as in 10 will be produced by protonation of the negatively charged part of the molecule. Again the counterion will be MSA.
105. The skilled team will know that the methylsulphonate salt will be soluble in MSA. Both are highly polar and "like dissolves like".
106. Similarly, the zwitterions are in any event polar.
107. Also they know that PC1₃ and Phosphorous acid are soluble in MSA.
108. In relation to CIPLA's Acid X, the skilled team would know the following:
109. It has the same chemistry as MSA — they are both very strong acids with similar pKas.
110. It has the same physical characteristics as MSA:
111. Both are very polar, sulphonic acids, strong acids, monofunctional, liquids at reaction temperature.
112. Acid X melts at a different temperature to MSA but will have melted at the reaction temperature Although this is close to the melting point of Acid X, the reaction temperature is still well above it. In any event, this fact is not relevant when chemical reactivity (ie acid/base behaviour) is considered. Acid/base reactions are very fast and are diffusion controlled.
113. For all these reasons, the skilled person would expect that Acid X would keep the system fluid and homogeneous and consequently allow the production of a pure product in a high yield.
114. The class of chemicals which the skilled person would understand would give rise to success is not large. On the contrary it is quite limited.
115. Concerning the likely effect of Acid X on the crystallisation and isolation steps, the following would be obvious. When the reaction mass in the Patent is neutralised with NaOH, a number of salts will be formed. Alendronate sulphate is one of them. The other will be the sodium salt of MSA, i.e. sodium methane sulphonate (SMS). This is not spelled out in the specification, but would be immediately apparent. SMS must therefore be sufficiently different in its solubility in the reaction mixture from alendronate to allow the alendronate to crystallise out in this simple fashion in such a pure form.
116. If Acid X is used, at the neutralisation stage sodium X sulphonate (SXS) will be formed.
117. Both SMS and SXS are freely soluble in water. This is to be contrasted with sodium alendronate, which is very insoluble.
118. The purity and yield obtained in the Patent is a result of harnessing the difference between the solubilities in water of alendronate and SMS. It makes them easy to separate. Similar differences in solubility could be expected if other sodium sulphonates (such as SXS) are present. Alternative standard ways exist of using that difference in crystallinity. For example, the one deployed in the patent could be modified by changing some of the detailed parameters set out or it could be replaced by an alternative system using, for example, a separate stage of recrystallisation employing an anti-solvent.