Napp Pharmaceutical Holdings Limited v Dr Reddy's Laboratories (UK) Ltd & Anor

The claimant and appellant, Napp Pharmaceutical Holdings Ltd (“Napp”), is the proprietor of European patent (UK) No 2 305 194 (“the patent”) in respect of an invention of a transdermal patch for use in the treatment of pain. Napp contends that the first defendant and respondent Dr Reddy’s Laboratories (UK) Ltd (“DRL”) and the second defendant and respondent Sandoz Ltd (“Sandoz”) both threaten to infringe the patent by marketing their own transdermal patches. In his judgment dated 28 June 2016 ([2016] EWHC 1517 (Pat)), Arnold J concluded that there would be no infringement by either respondent if this marketing occurred and dismissed the various claims.

Napp’s claim was commenced on 19 February 2016 coupled with an application for an interim injunction against Sandoz made on the same day and followed by a similar application against DRL on 22 February 2016. On 16 March 2016 Arnold J ordered an expedited trial of the claim fixed for 7 and 8 June 2016. In consequence, each of the respondents gave an undertaking not to launch their product pending the determination of the claim. Following judgment, the judge continued interim relief against both respondents until 16 August 2016. He gave permission to appeal on grounds 1 and 3 of Napp’s draft grounds of appeal, but he refused it on ground 2. Napp lodged its notice of appeal on 30 June 2016. On 5 July 2016 I made an order for expedition of the appeal, and adjourned the application for permission to appeal on ground 2 to the full court with the appeal to follow on ground 2 if permission were to be granted.

On the appeal Mr James Abrahams QC argued the case for the appellant with Mr Henry Ward. The case for DRL was argued by Mr Michael Silverleaf QC with Mr Benet Brandreth and that for Sandoz by Mr Justin Turner QC with Ms Anna Edwards-Stuart.

At the conclusion of the appeal on 2 August 2016 we stated that we intended to dismiss the appeal, and give our reasons in writing at a later date. These are the reasons why I joined in the decision to dismiss the appeal. It is worth recording that the case thereby obtained a final decision at this instance in under six months from the issue of proceedings.

The judge explained the technical background to the patent at paragraphs 26 to 40 of his judgment in terms which are not challenged on this appeal - I will not repeat what he said here. There is also no challenge to what the judge said about the skilled team and the common general knowledge at paragraphs 77-88 and 103.

The patent is concerned with a transdermal patch for treating patients with the pain-relieving drug buprenorphine. Having made acknowledgements as to the existing state of the art, the patent states at paragraph [0010]:

“Despite these advances in the art, there remains a need for methods of treating patients with buprenorphine that provide effective analgesic levels of buprenorphine for prolonged periods of time while eliminating or minimizing dependence, tolerance, and side effects, thus providing a safe and effective method of pain management. Further, there remains a need for a transdermal formulation of an opioid analgesic, preferably, buprenorphine, that provides effective analgesic levels of buprenorphine for periods of time beyond that contemplated or practical in the prior art, while eliminating or minimizing dependence, tolerance, and side effects, thus providing a safe and effective method of pain management."

Accordingly, the specification states at [0011] that the object of the invention is to provide:

“a method and pharmaceutical formulation … which allows for reduced plasma concentrations of buprenorphine over a prolonged time period than possible according to prior art methods, while still providing effective pain management.”

The patent cross refers to International Patent Application No. WO 96/19975 “Hille” in a number of places. It is common ground that the skilled reader would read Hille together with the patent. Thus at [0073] the patent says:

"In a preferred embodiment, the transdermal delivery device is prepared in accordance with Example 1 appended hereto. In this example, the transdermal delivery device was prepared in accordance with the disclosure of [Hille]. In this device, the buprenorphine transdermal delivery device contains resorption-promoting auxiliary substances. The resorption-promoting auxiliary substance forms an undercooled mass. The delivery device contains 10% buprenorphine base, 10-15% acid (such as levulinic acid), about 10% softener (such as oleyloleate); 55-70% polyacrylate; and 0-10% polyvinylpyrollidone (PVP)."

At [0074] the specification explains that, in order to achieve particular nominal delivery rates, amounts of buprenorphine of “about 5 mg”, “about 10 mg”, “about 20 mg”, “about 30 mg” and “about 40 mg” are used.

The specification then explains at [0075] that the described delivery device, although designed to be adhered to the patient and provide analgesically effective doses for only about three days, can in fact be so adhered and provide effective doses for much longer times, e.g. from about 5 to 8 days. At [0084] the specification goes on to provide a theory of operation, through the creation of a depot or reservoir in the skin. The skin depot which is formed is then replenished with small amounts of buprenorphine, from whence it is released into the plasma.

Example 1 (so numbered even though it is the only example in the patent) is described at [0098]-[0101]. The description of this Example begins:

"[0098] A seven day pharmacokinetic/pharmacodynamic study was conducted on 24 healthy human patients. … In this study, the buprenorphine was administered via a transdermal patch which is described in [Hille].

[0099] The transdermal patch is prepared in accordance with the disclosure of [Hille] for Example 1 therein as follows:"

What follows is a detailed description of the method of making the patch which is taken verbatim from Example 1 in the text of Hille. The patent continues at [0100]:

"The formulation utilized for Example 1 is substantially the same as that described in Example 3 of [Hille] which is prepared in accordance with Example 1 and is stated therein to include 10% buprenorphine, 10% levulinic acid, 10% polyvinylpyrollidone, 10% oleyloleate, and 60% polyacrylate. "

The question of exactly what the skilled reader would understand the formulation of Example 1 to be is an issue in the appeal. Hille goes on to observe that certain penetration enhancers and other excipients are solids at room temperature. This means that it is difficult for them to diffuse out of the polymer matrix. The task of the invention is therefore “to improve the delivery of excipients, which are solid at room temperature, out of a matrix”. The invention solves the problem by using an excipient, such as dodecanol or levulinic acid, which forms as a supercooled melt. Such a melt will have a melting point which is above room temperature, but can be made to remain as a liquid when cooled to room temperature.

In the conventional way, Hille illustrates the invention by reference to a number of examples. “Example 1” at page 3 (which is the text reproduced in paragraph [0099] of the patent) is as follows:

"1.139

[kg] of a 47.83 w/% polyacrylate solution of a self-crossing acrylate copolymer of 2-ethylhexylacrylate, vinyl acetate, acrylic acid (solvent: ethyl acetate:heptane:iso-propanol:toluene:acetylacetonate in the ratio of 37:26:26:4:1), 100 g levulinic acid, 150 g oleyl oleate, 100 g polyvinylpyrrolidone, 150 g ethanol, 200 g ethyl acetate and 100 g buprenorphine base are homogenised. The mixture is stirred for about 2 hours and checked visually to see if all the solids are dissolved. The evaporation loss is controlled by weighing, and, if necessary, replenishing the loss of solvent with ethyl acetate.

The mixture is then spread onto a 420 mm wide transparent polyester film, so that the surface weight of the dried adhesive layer is 80 g per m². The polyester film, which can be removed again with a silicon treatment, serves as a protective layer.

The solvent is removed by drying with heated air which is passed over the damp track. The heat treatment does not only cause the solvents to evaporate, but also melts the levulinic acid. The adhesive film is subsequently covered with a 15 µ polyester film. A 16 cm2 area is punched out with appropriate cutting tools, and the edges remaining between the individual systems are removed."

It is common ground that the approximate composition of the drug adhesive blend (excluding the solvents) in Example 1 of Hille would be 10%-wt buprenorphine, 10%-wt levulinic acid, 15%-wt oleyl oleate, 55%-wt polyacrylate and 10%-wt polyvinylpyrollidone (PVP).

Hille then says that “Examples 2 to 5 were conducted in a manner analogous to Example 1” and that Table 1 shows “[T]he composition of Examples 2-5”. Hille then introduces Table 1 by saying: “The results of the in vitro penetration tests and the qualitative and quantitative compositions of buprenorphine TTS according to Example 1-5 are shown in Table 1.” Table 1 is shown below:

Cetiol is oleyl oleate. Somewhat oddly, Example 1 in Table 1 of Hille differs from Example 1 in the text of the document. Example 1 in Table 1 has 10% glutaric acid monomethyl ester, whereas the text specifies levulinic acid in an amount which can be calculated to be 10 %-wt of the inputs, and it has 10% oleyl oleate, whereas the text specifies oleyl oleate in an amount which can be calculated to be 15 %-wt. Finally Table 1 specifies 70% dried polyacrylate, whereas the text specifies adhesive in an amount which can be calculated as 55%-wt polyacrylate and 10%-wt polyvinylpyrollidone (PVP). For this reason, like the judge, I will refer to Example 1 in the text of Hille as "Example 1a", and that in Table 1 as “1b”.

The final column of the Table gives the penetration results. Example 3 has the highest figure for penetration and is therefore the best result. The Table does not give penetration results for Example 1a. On page 5 of the specification of Hille, it is argued that the differences in penetration are due to the presence of supercooled melts. This is achieved in Examples 1b, 3 and 5 by the use of glutaric acid, levulinic acid and dodecanol, but not with octanoic acid and undecenoic acid in Examples 2 and 4.

Hille therefore provides a “recipe”, including both method and proportions, in Example 1a which is then used as the template for the examples in Table, which are “prepared in a manner analogous to Example [1a]”. Thus it would be apparent to the skilled reader of Hille that Example 3, for instance, was prepared by adjusting the proportions (but not the method) of Example 1a.

It is common ground that it is only necessary to consider claim 1 of the patent, which is as follows:

"A buprenorphine transdermal delivery device comprising a polymer matrix layer containing buprenorphine or a pharmaceutically acceptable salt thereof, for use in treating pain in humans for a dosing interval of at least 7 days, wherein the transdermal delivery device comprises 10%-wt buprenorphine base, 10 to 15%-wt levulinic acid, about 10 %-wt oleyloleate, 55 to 70%-wt polyacrylate, and 0 to 10%-wt polyvinylpyrrolidone."

The dispute about how the skilled person would understand paragraph [0100] of the patent is as follows. Does the paragraph teach the skilled reader that the patch used in Example 1 of the patent is made in accordance with both the method and proportions of Example 1a of Hille (as Napp contend), or does it teach only that the method of Example 1a of Hille is used, the proportions being those of Example 3 of Hille (as the respondents contend)? The relevance of the dispute is that if the proportions of Example 1a are to be included then the patent is telling the skilled reader two things. Firstly, in a somewhat roundabout way, it is saying that the 15%-wt content of oleyloleate in Example 1 of Hille is “substantially the same” as the formulation of Example 3 of Hille, which, as the patent explains, is stated to have only 10%-wt oleyloleate. Secondly the only example in the patent, which one would assume to be within the scope of the claim, has 15% oleyloleate. Thus, when the claim says “about 10%-wt oleyloleate”, it includes 15%-wt oleyloleate.

The judge dealt with this issue at paragraphs 97 to 102 of his judgment, setting out the rival arguments. He concluded that the interpretation advanced by the respondents was the correct one.

The arguments which the judge accepted were in essence as follows. Firstly, the interpretation proffered by the respondents (in which Example 1 of Hille was merely referred to for its method, whilst Example 3 of Hille was referred to for the formulation) was the natural and least strained interpretation. By contrast, Napp’s interpretation required the skilled reader to calculate percentage weights of the components in Example 1a of Hille, and then conclude that paragraph [0100] of the patent was telling him that 10%-wt oleyloleate was substantially the same as 15%-wt oleyloleate, and, additionally, that 55%-wt polyacrylate was substantially the same as 60%-wt polyacrylate. Secondly, the skilled reader of Hille would note that Example 3 gave the best penetration results, providing a good reason for the patentee to have chosen to use the formulation of Example 3. Thirdly, on Napp’s interpretation, the reference to Example 3 of Hille served no apparent purpose.

There are two issues which divide the parties on the construction of the claim. The first is whether, when specifying the quantities of the various elements of the transdermal delivery device, the claim is referring to the “input” amounts used in formulation (i.e. they are ingredients in a recipe), or whether it is referring to the “output” amounts in the finished device.

The judge dealt with this issue at paragraphs 104 to 110 of his judgment and concluded that the claim was referring to the finished product. In that passage the judge rehearsed the arguments of both sides and concluded that the arguments of the respondents were correct.

The second issue of claim construction is the correct interpretation of the numerical features of the claim, and in particular what is meant by “10%-wt buprenorphine base”, “10 to 15%-wt levulinic acid”, “about 10%-wt oleyloleate.”

The judge dealt with the arguments about numerical limits at paragraphs 111 to 125 of his judgment. He considered that the buprenorphine limit was expressed to the nearest whole number, so extended from 9.5 to 10.5%-wt, and the levulinic acid range, for the same reason, extended from 9.5 to 15.5%-wt.

As to “about 10%-wt oleyloleate”, the judge considered that it allowed a margin of 1% around the figure of 10%-wt. This was because the word “about” would be understood to give a small degree of imprecision over and above that permitted by normal rounding. In so holding, the judge rejected the argument that the skilled reader would take from the cross-reference to Hille an understanding that the patentee regarded 15%-wt as within the expression “about 10%-wt”.

The details of the respondents’ formulations were treated as confidential before the judge and before us. It is sufficient to say that the judge’s findings on the two issues of construction were sufficient to dispose of the issue of threatened infringement. Further arguments concerned with infringement on the basis of the judge’s construction (i.e. that it is the output percentages which are relevant) are no longer pursued by Napp.

Ground 1 challenges the judge’s conclusions on the issues of construction which I have referred to above, contending that he should have held (1) that the %-wt figures refer to input components of the matrix layer of the device (excluding solvents removed during drying); (2) that “about 10%-wt oleyloleate” includes 15%-wt oleyloleate; (3) “10 to 15%-wt levulinic acid” refers to a range of %-wt levulinic acid to the nearest 5% (the range 7.5 to 17.5%-wt); and (4) “10%-wt buprenorphine” refers to the amount of buprenorphine to the nearest 5%-wt (the range 7.5 to 12.5%-wt); alternatively it should be construed as covering the range 10 ± 1 %-wt buprenoprphine. Ground 3 asserts, uncontroversially, that if the judge had so held, he would have found threatened infringement of claim 1 of the patent by each of Sandoz and DRL.

Ground 2, for which permission is sought, attacks certain of the judge’s conclusions on the evidence. The points taken are:

By a respondent’s notice, DRL contends that if the patent is construed in the manner contended for by Napp then claim 1 would be insufficient because it does not enable the reader to determine where the boundaries of the claim lie.

I have explained the relevance of this issue above. Mr Abrahams challenged the judge’s acceptance of the respondents’ interpretation of paragraph [0100] of the patent. He repeated the five points made before the judge, which the judge recorded at paragraph 99 of his judgment. He focused particularly on the fact that the patentee had repeated the text of Example 1a of Hille verbatim. Surely, Mr Abrahams submitted, if Example 1 in the patent was supposed to be the formulation of Example 3 of Hille, then what the patentee would have done is give the recipe which he had actually used. This would have involved editing Example 1 of Hille with the proportions necessary to achieve the formulation of Example 3. It would be bizarre to say “Here is a recipe for a patch which I am not telling you to make. Instead I am telling you how to make a patch to this final composition, without any detail of the quantities required to make it.” Mr Abrahams attacks the judge’s reasoning which I have set out at paragraph 23 above in the following way. Firstly he says that Napp’s reading of paragraph [0100] does not depend on the skilled person doing any calculations, at least at this stage of the argument. Instead the skilled person would just note that he is being told that Example 3 and Example 1 were substantially the same. It is only necessary for the skilled person to calculate the percentages in Example 1a of Hille when he comes to decide what “about 10%” means in the claim. Further he submits that it is logical for the patentee to refer to Example 3 precisely because it gives the best penetration results. What the reader is being told is that, although he must make the patch according to the precise teaching of Example 1a, it will produce results substantially the same as Example 3.

Like the judge, I prefer the respondents’ reading of paragraph [0100] of the patent. The patentee had to explain both the general method of making the supercooled melt which Hille had found to be successful and the quantities and proportions of the various ingredients. He does so by saying, in paragraph [0099], that the transdermal patch of the invention is “prepared in accordance with the disclosure of [Hille] for Example 1 therein”, and then goes on to say in paragraph [0100] that the formulation used for Example 1 of the patent is substantially the same as that described in Example 3 of Hille “which is prepared in accordance with Example 1 thereof and is stated to contain 10% buprenorphine, 10% levulinic acid, 10% polyvinylpyrollidone, 10% oleyloleate, and 60% polyacrylate”. This is to adopt the same approach as is used in Hille, which is that Example 1a is a template for the production of the patches in Table 1. Hence the skilled person would readily understand that he was to make an Example 3 patch, adapting the method of Hille’s Example 1a, just as he would have had to do if carrying out the teaching of Hille’s Example 3. I regard that as an entirely natural reading of the paragraph. The alternative reading, in which the skilled person is being told that the formulations of Example 1a and 3 of Hille are substantially the same, seems to me to be forced. The upshot would be that the patent does not expressly state even an approximate figure for the percentages of Example 1. They would have to be calculated. I consider that it is much more likely that the patentee is providing information as to the composition at this point, rather than making a comment on the similarity of the composition to that just described.

I acknowledge that there is some force in Mr Abrahams’ submission that, if Example 1 of the patent was indeed Hille’s Example 3 formulation, it ought to have been possible for the patentee to write out the recipe for Example 3 (which he should have been aware of), rather than copy out Hille’s Example 1a recipe. But it seems to me that it is rather more plausible that the patentee found it convenient to quote the recipe of Example 1a of Hille, so as to ensure he had given details of the method, leaving the reader to make the necessary adjustments to produce the Example 3 composition, which was the basis of his Example 1.

None of the other points takes the matter any further. Both sides were able to offer an explanation for the reference in the patent to Example 3 of Hille, and neither side’s interpretation was wholly free of linguistic anomaly. For example Mr Abrahams suggested that, on the respondents’ interpretation, it was not necessary to say that the formulation was substantially the same as that of Example 3 of Hille. I think this is just an example of cautious drafting, rather than a reason for rejecting the respondent’s interpretation.

Mr Abrahams developed his argument that the patent was referring to input and not output amounts by reference to the seven arguments which he advanced before the judge.

Firstly, Mr Abrahams submitted that, as a matter of language, there was no difficulty in describing a product by reference to its ingredients. He referred to Cephalon Inc v Orchid Europe Ltd [2011] EWHC 1591 in which I construed a claim to a pharmaceutical composition comprising particles of a certain size as referring to the particle size of the input API. He stressed the use of the word “formulation” in the body of the specification which he contended is more apt to cover the mix of ingredients than the final dried product.

Secondly, he submitted that paragraph [0099] of the patent, in setting out the method of making the patch, is plainly referring to input quantities, just as a cooking recipe would set out a list of weights of ingredients. Paragraph [0100] then switches to percentages by weight, but the reader would assume that the author was being consistent, and referring to inputs in both cases.

Thirdly, paragraph [0074] uses the word “about” to define the (absolute) amount of buprenorphine present in the patch, a recognition of the fact that the manufacturer would not be able to control the amount in the finished product with great precision. By contrast the manufacturer would be able to control the input percentage with greater precision.

Fourthly, Mr Abrahams submitted that the “standard” or “conventional” way of describing transdermal patches was to describe them by reference to their input ingredients, and that the skilled person would recognise that the patches in the patent, were being described in the same way. The judge was wrong in his rejection of the evidence of Dr Miller, which was to that effect. He had also been wrong to say that the evidence that the practice was “routine” was limited to the case where the skilled person would assume that there were no losses during manufacture.

Fifthly, the skilled person would assume he was not being set a challenge. However, on the respondents’ construction obtaining any particular target composition would take trial and error. The skilled person would have to make a best guess at what quantities to use, what drying conditions to use, make a batch of patches and test them, and then adjust the starting conditions until he achieved the composition which fell within the range he wanted. By contrast, on Napp’s construction there was no difficulty, he would simply use the specified quantities.

Sixthly, assaying the finished product would be challenging, particularly from a statistical perspective. No such problems arose if the claim is referring to input rather than output quantities. The judge had wrongly assumed that the evidence only established that it was straightforward to measure the amounts of API and penetration enhancer, levulinic acid, in the finished product. Other excipients were not tested in practice at all and so there were no accepted assays for them. Moreover, given the variation from patch to patch, and more so from batch to batch, a statistical analysis is necessary to determine whether and if so what percentage of patches complied with any given output specification.

Seventhly, transdermal delivery systems degrade over time, with the consequence that the amount of API will decrease with time. The input ingredients were a constant.

Before us Mr Abrahams made one additional point, based on Table 1 of Hille. He drew attention to the fact that the quantities for the ingredients were all in numbers divisible by 5, whereas other quantities, such as pKa values for the acids were, as one would expect, characteristic measured numbers, with no similar pattern. It was unrealistic to suppose that the patentee had carefully measured the finished composition, rather than simply relayed the percentages of his starting ingredients.

Whilst they were attractively presented, I am not impressed by these arguments. The starting point has to be the language of the claim, which requires a transdermal delivery device comprising a defined API in a specific percentage and excipients also in a defined percentage. It is a product claim, not a method claim. There is no sense in which the product, i.e. the device, is made at the stage of mixing the ingredients. The normal approach to such a claim would be to determine what the device contained, not to determine the percentages used in some earlier stage of the manufacturing process. That approach would, to my mind, be reinforced by the fact that the device is to be used in contact with the skin of the patient. It is therefore the proportions of the materials in the finished product which are important, not those elsewhere. The judge made an unchallenged finding at paragraph 103 of the judgment that the formation of a skin depot or reservoir, an important aspect of the way the invention works, was not simply a consequence of the presence of buprenorphine, but was affected by the properties of other excipients such as levulinic acid and oleyloleate.

Factual analogies with other cases are seldom useful. In the Cephalon case the evidence established that it required invention on the part of a skilled person to develop an assay which would be suitable for measuring particle size in the finished product. In such circumstances it is obviously possible that a skilled reader would assume that the patentee was treating the particle size of the ingredient as a proxy or measure for the particle size in the finished composition. Each case, however, must be decided on its own facts, and the facts here are very different.

The juxtaposition in both Hille and the patent of the absolute weights in the method of preparation and the percentage weights used subsequently (Napp’s second point) does not in my view lead to the conclusion that both are referring to input rather than output. On the contrary, as the respondents submitted, in both Hille and the patent the subsequent references are to the patches, not to the method of manufacture. The introduction to Table 1 in Hille makes it clear that the Examples are referring to the quantitative composition of the patches, not the mixes from which they are made.

There is nothing in Napp’s third point concerning the use of the word “about” in paragraph [0074]. This is an example of the sort of meticulous linguistic analysis which was deprecated by Lord Diplock in Catnic v Hill Smith [1981] F.S.R. 60; [1982] R.P.C. 183.

Napp’s fourth point seeks to establish a conventional way of describing transdermal delivery systems. The judge was entitled to hold that it had not been established that the conventional way to describe transdermal patches was by reference to the ingredients used in manufacture. Such a convention would give rise to a strong presumption that whenever the composition of a transdermal patch was described the established convention was being followed. Whilst Mr Abrahams was able to point to examples in a variety of contexts where patches had been described by reference to the ingredients used in their manufacture, isolated examples of this kind do not establish a convention.

Mr Abrahams relied on some evidence of Dr Miller that it was “routine” to identify products by reference to the starting formulation. I can well understand that, depending on the context, it may be clear that reported scientific work is identifying products by reference to the mix used in manufacture. In such a case the author is identifying the product he made as “the one where I used the following ingredients in the following proportions”. I can also understand that it may be routine in such work to adopt that nomenclature, even where losses are expected in manufacture. If the process is substantially the same, then comparison between various samples may remain valid, and the ingredients used may be a rational way of identifying the product. That, however, is a very long way from establishing a linguistic convention that all compositions of patches are to be taken as being described in this way.

Mr Abrahams says that the judge should not have rejected the evidence of Dr Miller on the basis that he did, namely that it was limited to the case where no losses were expected in manufacture. Even if the judge was wrong to say that the evidence of Dr Miller was limited to cases where there were no losses expected in manufacture, I still do not consider that this evidence advances Napp’s case. Once it is recognised that there is no standard or conventional way of describing these products, the correct interpretation of the claim must be context dependent. The fact that it is routine to describe a product in this way in some contexts does not mean that the patentee is doing so in the present case. This disposes of the second of Napp’s evidential points.

Napp’s fifth and sixth points are answered by the judge by saying that there was no evidence that reproducing the patch described in the patent would be difficult, or that carrying out assays would be difficult. On the contrary the evidence established that it was routine laboratory work. Mr Abrahams says that this does not really meet his point, which is that arriving at a particular targeted value of the ingredients in the finished product would require trial and error. I do not accept that this is a point which would sway the skilled reader on the issue of construction. He or she would realise that the issue of infringement turned on what the patch actually contained, and would carry out the routine laboratory work necessary to determine that. It does not therefore matter that the skilled person cannot predict the precise composition of patch which would result from any given mix of ingredients. This disposes of the third of Napp’s evidential points.

I do not accept that the judge overlooked the fact that the claimed patch included ingredients which would not normally have been tested for. Mr Abrahams relied particularly on the fact that polyacrylate was not normally tested for. This does not advance his case. The matter might have been different if the evidence had shown that once incorporated in the product there was no known way of determining the polyacrylate content. This disposes of the first of Napp’s evidential points.

Napp’s seventh point was that, because the API degrades over time, the output percentages were not fixed, but could change over time. I do not see this as a point of any real weight. The skilled reader would not assume the patentee to be referring to input percentages just because there is a potential for some degradation of the API in the final product.

Napp’s final point concerning the use of multiples of 5 in Table 1 of Hille, would have some force if the figures in the table were viewed in isolation. But the figures have to be read in conjunction with the introduction, which tells the reader that this is what the product contains. I am not persuaded that this consideration means that Hille was referring to percentages in the input formulation.

I would therefore reject Napp’s attack on the judge’s conclusion that the claim is referring to output percentages.

That would be sufficient to dispose of the appeal. However, as we were told that DRL have undertaken to work outside the numerical ranges in the claim, it remains material to consider whether the judge was right in his construction of those ranges.

The approach to the construction of numerical features and ranges in patent claims was recently considered in this court in Smith & Nephew plc v ConvaTec Technologies Inc [2015] EWCA Civ 607, [2015] RPC 32. At [38] Kitchin LJ (with whom Briggs and Christopher Clarke LJJ agreed) said:

“As I have said, the approach to be adopted to the interpretation of claims containing a numerical range is no different from that to be adopted in relation to any other claim. But certain points of particular relevance to claims of this kind do emerge from the authorities to which I have referred and which are worth emphasising. First, the scope of any such claim must be exactly the same whether one is considering infringement or validity. Secondly, there can be no justification for using rounding or any other kind of approximation to change the disclosure of the prior art or to modify the alleged infringement. Thirdly, the meaning and scope of a numerical range in a patent claim must be ascertained in light of the common general knowledge and in the context of the specification as a whole. Fourthly, it may be the case that, in light of the common general knowledge and the teaching of the specification, the skilled person would understand that the patentee has chosen to express the numerals in the claim to a particular but limited degree of precision and so intends the claim to include all values which fall within the claimed range when stated with the same degree of precision. Fifthly, whether that is so or not will depend upon all the circumstances including the number of decimal places or significant figures to which the numerals in the claim appear to have been expressed.”

Mr Abrahams submitted that this was a case where the patentee had chosen to express himself to the nearest 5%, at least in the case of buprenorphine and levulinic acid. He advanced five reasons why that was the case.

Firstly, Mr Abrahams submitted that the invention of the patent was not the precise formulation of the patch, but its use or application. What had been discovered or invented was the use of the patch for seven days. Accordingly the skilled person would not think that the patentee was being very precise about the formulation.

Secondly Mr Abrahams submitted that for a given formulation it was impossible to predict what would eventually come off the production line in terms of percentage composition. The product is subject to inherent variability.

Thirdly, he submitted that it was important to remember that the claim was expressed in percentages, not absolute amounts. A difference between 10%-wt and 11%-wt buprenorphine does not mean more buprenorphine. He accepted that a change in concentration of buprenorphine will change the flux rate into the skin depot, but that was not material to the way the invention worked.

Fourthly, he submitted that the patent and Hille described a patch in the early stages of development, where it was reasonable to expect the patentee to be describing his product in a “ballpark way”.

Fifthly, Mr Abrahams submitted that both the patent and Hille were working in increments of 5%.

As to the expression “about 10%-wt oleyloleate” Mr Abrahams relied on those four points and two further points. The first of these was that already discussed, namely that Example 1 of the patent was the same as Example 1a of Hille, and therefore used 15% oleyloleate, which the patentee considered to be substantially the same as 10%-wt. He relied on the fact that the Opposition Division of the European Patent Office had so read the patent in opposition proceedings involving companies associated with the respondents. I will call this the Example 1 argument.

The second point advanced under this head is based on paragraph [0073] of the patent, which is set out above at paragraph 8. The final sentence of that paragraph is an attempt to summarise the examples of Hille, and is clearly the source of the figures in claim 1. Mr Abrahams submitted that the skilled reader would appreciate that the patentee was trying to encompass all the examples of Hille, including Example 1a. Hence when he says in that paragraph “about 10% softener (such as oleyloleate)” he must have intended to include Example 1a, with its calculated 15%-wt oleyloleate. I will call this the paragraph [0073] argument.

I start with Napp’s first four general points which it says support its broad construction of the claims. The judge broadly accepted the first two and fourth of these, whilst reminding himself that numeric features, like any other features of a claim, cannot simply be ignored. It may be absolutely clear to a skilled reader that stepping outside the defined limits of the claim would have no material effect on the way the invention works: but the patentee may nevertheless have chosen to limit his monopoly to the figures or ranges claimed. As to the third point, the judge pointed out that a change in percentage weight was a change in concentration which would have an effect on flux, so he treated the point as irrelevant.

I think that the judge was right to reject these general points as pointing towards the construction of the claims for which Napp contends. Even though the skilled reader would know that a high degree of precision was not necessary to achieve success, he would still have to come to a conclusion as to what the patentee meant by the limits he had set in the claim.

The fifth of Napp’s points is the only one which could support treating the figures as being to the nearest 5%. It is true that the figures in the patent and in Hille are all multiples of 5%. But, even if one assumes that this is something which the reader will notice without being told, that fact alone tells one nothing about the degree of precision to which these numbers are expressed. It does not contradict the skilled person’s normal understanding that numbers written in this way would be treated as expressed to the nearest whole number.

A patent specification is not intended to be a puzzle game in which the skilled person must come up with his own theory as to what degree of precision was intended by the patentee. If the patentee had wished to claim, for example, 7.5-12.5%-wt buprenorphine, he was free to do so by making it express in the claim, or stating in the specification that the figures are expressed to the nearest 5%. He cannot expect the skilled reader to assume that, because he has chosen to write the claim in multiples of 5%, he is not concerned about deviations of as much as 2.5% on either side of that figure. The figures he chose may have been the very outer limits of what he considered would work. I therefore entirely agree with the judge that the figures for buprenorphine and levulinic acid are expressed to the nearest whole number.

Napp’s difficulties are exacerbated when one turns to what is meant by “about 10%-wt oleyloleate”. The use of the word “about” suggests that the patentee must have been intending something wider than would be the case if he had omitted that word and the claim was interpreted as being expressed to the nearest whole number. But how much wider did the patentee intend to go? The Example 1 argument, even if correct, would tell the reader that 15% was included. But if so, why not 16%, or 20%? At what point would one no longer have “about 10%-wt oleyloleate”? Likewise, what is the lower limit of the claim on this basis? Does it remain at 9.5%, or, is the same degree of precision to be adopted in the downward direction, so that it certainly covers 5% and may go down further? The paragraph [0073] argument leads to exactly the same conundrums.

I therefore do not think that either the Example 1 argument or the paragraph [0073] argument form a basis for supposing that the patentee was using “about 10%” as a way of expressing a “particular but limited degree of precision” (see Smith & Nephew cited in paragraph 59 above). Rather, I think that if the skilled person were to reason as Napp contends in those arguments, he or she would be wholly unclear as to where the limits of the claim actually lay.

The second difficulty with Napp’s arguments is that they both require the skilled reader to appreciate that Example 1a of Hille uses 15% olelyloleate. He could only do so by calculation, and I regard it as unrealistic to suppose that he would attempt to do so.

Accordingly, I think the judge was right to conclude that what lay behind the patentee’s use of “about” was not a desire to cover Example 1a of Hille. Rather the patentee was intending to claim a more generous degree of imprecision around the figure of 10% than would be accorded by the normal rounding convention. There are obvious difficulties with trying to prescribe precise outer limits to an integer which uses the word “about”, but I would not disagree with the judge’s conclusion that the claim extends no further than 9 to 11%.

In those circumstances there was no need to consider DRL’s respondent’s notice.

I agree.

I also agree.