Hospira UK Ltd v Genentech Inc [2016] EWCA Civ 780

[2016] EWCA Civ 780Case No A3/2014/ 4302
IN THE COURT OF APPEAL (CIVIL DIVISION)
[2014] EWHC 3857 (Pat)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
PATENTS COURT
The Hon Mr Justice Birss
Venue Royal Courts of Justice, Strand, London, WC2A 2LLDate 27/07/2016
Before
LORD JUSTICE KITCHINLORD JUSTICE FLOYDLORD JUSTICE DAVID RICHARDS
Between
HOSPIRA UK LIMITEDClaimant/GENENTECH, INC.Defendant/
Michael Tappin QC and Mark Chacksfield (instructed by Marks & Clerk Solicitors) for AppellantRichard Meade QC, Tom Mitcheson QC and Jeremy Heald (instructed by Taylor Wessing) for RespondentHearing Hearing dates: 12 and 13 July 2016
JudgmentLord Justice Floyd
[1]This is an appeal from the decision of Birss J dated 21 November 2014 in an action by Hospira UK Limited (“Hospira”) for revocation of two patents, European patents (UK) Nos. 1 516 628 and 2 275 119. The judge held both patents invalid on the ground of lack of inventive step and added matter. Genentech Inc. (“Genentech”), whose patents these are, appeals to this court with permission granted by the judge himself.[2]Both patents are concerned with pharmaceutical formulations of the important breast cancer drug Herceptin, the active ingredient in which is the monoclonal antibody trastuzumab. There were two general approaches to formulating such proteins at the priority date. One was to produce a liquid formulation. The other was to produce a dry lyophilised (i.e. freeze-dried) formulation. Both approaches were common general knowledge. The liquid formulation would be directly injected. The dry, lyophilised formulation would be made up into a sterile solution for injection. Lyophilised material was likely to be stable for longer than a liquid since chemical reactions are faster in solution. On the other hand, steps needed to be taken to ensure that the lyophilisation process itself did not degrade the protein.[3]A lyoprotectant is a substance used to protect the protein in the lyophilisation process. Substances such as dextrose, glucose, lactose, and sucrose as well as some alcohols and glycols were in common use as lyoprotectants at the priority date.[4]There were other classes of excipients which would also be considered for inclusion in a lyophilised formulation. These included but were not limited to buffers and surfactants. Buffers were used to keep the pH at a level required for stability of the protein. Surfactants were used to minimise aggregation of proteins. Other excipients having defined roles, such as anti-oxidants and bulking agents, were also known and were used when required.[5]Although there are two patents, the disclosure of the two specifications is in all material respects identical. It is common ground that, at least for the objections of obviousness and added subject matter, it is sufficient to consider claim 2 of 1 516 628 which, in the proposed amended form considered by the judge, reads as follows:
“A formulation consisting of a lyophilized mixture of a lyoprotectant, a buffer, a surfactant and an antibody, wherein the lyoprotectant is trehalose, wherein the buffer is histidine, wherein the surfactant is polysorbate 20 and wherein the antibody is [trastuzumab], such that an amount of said lyophilized mixture containing 450mg of said antibody can be reconstituted with 20ml of BWFI (0.9 or 1.1% benzyl alcohol) to yield a concentrated protein solution containing 22 mg/ml of said antibody, 52 mM trehalose, 4 mM histidine pH 6.0, 0.009% polysorbate 20.”
[6]So the patent is about making a lyophilised formulation where you have chosen the excipients trehalose, histidine and polysorbate 20 as lyoprotectant, buffer and surfactant respectively. The third and fourth words of the claim “consisting of” are the draftsman’s way of saying that the ingredients are those specified and no others. The words of the claim following “such that” are a way of specifying the relative proportions of the excipients and the antibody. It is not suggested that there would be any invention in arriving at suitable proportions once the excipients had been identified.[7]Trastuzumab is a large protein molecule. The formulation of large protein molecules was a more difficult and complex task than the formulation of small molecules. The judge said that the task was a challenge and that finding a stable formulation was an essentially empirical exercise. At the priority date of the patent the biotechnology industry, although not yet involving the largest pharmaceutical companies, was a significant specialist industry. A number of protein products had been formulated for human use, including some in dry, lyophilised form.[8]The judge rejected allegations that the invention was obvious over the common general knowledge alone and over a document known as Draber. There is no appeal against those findings. He held, instead, that the invention was obvious over Carter, a pair of documents which disclose that trastuzumab, an anti-HER2 antibody, was in phase II clinical trials for breast cancer. That was because a skilled team consisting of a clinician and a formulator would, by the priority date, be motivated to produce a lyophilised version of the antibody, and the claimed combination was simply an example of the work product of the necessary screening programme to find a satisfactory combination of excipients. To arrive at such a combination would not have involved any inventive effort.

The law

[9]I start with the well-known passage from the judgment of Kitchin J (as he then was) in Generics (UK) Ltd v H. Lundbeck A/S [2007] EW HC 1040 (Pat), at [72], which has been approved at the highest level:
“The question of obviousness must be considered on the facts of each case. The court must consider the weight to be attached to any particular factor in the light of all the relevant circumstances. These may include such matters as the motive to find a solution to the problem the patent addresses, the number and extent of the possible avenues of research, the effort involved in pursuing them and the expectation of success.”
[10]It follows that the court is required to embark on a multi-factorial assessment. The approach of the appellate court to such questions is well-known: the decision is not open to independent evaluation in this court unless the judge has made an error of principle: see Biogen Inc v Medeva Plc [1997] RPC 1 at [45] per Lord Hoffmann.[11]It is also well-settled that “obvious to try” is not a substitute test for obviousness: see e.g. per Lewison LJ in Medimmune Ltd v Novartis Pharmaceuticals UK Ltd [2012] EWCA Civ 1234, [2013] RPC 27 at [181]. There is only one statutory question, namely whether the invention was obvious at the priority date. In the same case at [90] to [95] Kitchin LJ explained that whether the invention was obvious to try was merely one of many considerations which it may be appropriate for the court to take into account in addressing the statutory question. It must in any case be coupled with a reasonable or fair prospect of success. At [91] Kitchin LJ emphasised:
“Whether a route has a reasonable or fair prospect of success will depend upon all the circumstances including an ability rationally to predict a successful outcome, how long the project may take, the extent to which the field is unexplored, the complexity or otherwise of any necessary experiments, whether such experiments can be performed by routine means and whether the skilled person will have to make a series of correct decisions along the way.”
[12]Thus a judge’s assessment whether an approach has a reasonable or fair prospect of success is itself another multi-factorial assessment, where this court should again pay proper respect to the evaluation of a trial judge.[13]There is also no single standard of what amounts to a fair expectation of success. Thus Jacob LJ’s statement in Saint Gobain PAM SA v Fusion-Provida Ltd [2005] EWCA Civ 177 at [35] that it must be “more-or-less self-evident that what is being tested ought to work” is far from being a test of universal application. In Novartis AG v Generics (UK) Ltd (trading as Mylan) [2012] EWCA Civ 1623 Kitchin LJ (with whom Lewison and Munby LJJ agreed) warned against imposing a straitjacket on the law by adopting any form of words as a standard. He said at 55:
“What is a reasonable or fair expectation of success will again depend upon all the circumstances and will vary from case to case. Sometimes, as in Saint Gobain, it may be appropriate to consider whether it is more or less self-evident that what is being tested ought to work. So, as this court explained in that case, simply including something in a research project in the hope that something might turn up is unlikely to be enough. But I reject the submission that the court can only make a finding of obviousness where it is manifest that the test ought to work. That would be to impose a straitjacket on the assessment of obviousness which is not warranted by the statutory test and would, for example, preclude a finding of obviousness in the case where the results of an entirely routine test are unpredictable.”
[14]Indeed, so much was confirmed by Lord Hoffmann in Conor Medsystems v Angiotech Pharmaceuticals [2008] UKHL 49, [2008] RPC 28 at [42], approving Jacob LJ’s comprehensive review of the authorities in this area in the Court of Appeal in that case, including his view that “how much of an expectation depended upon the particular facts of the case.”[15]It is also often debated whether the correct question is whether the skilled person would, or whether he or she could, arrive at the claimed invention without inventive effort. A “would” test can be misleading, as it is liable to bring in irrelevant considerations, such as whether it would be worthwhile commercialising an otherwise technically obvious product, see per Oliver LJ in Windsurfing International Inc v Tabur Marine (Great Britain) Ltd. [1985] RPC 59 at 72. In Actavis UK Ltd v Novartis AG [2010] EWCA Civ 82; [2010] FSR 18, reliance was placed in argument on the EPO Guidelines for Examination which stress that the question is: “… whether there is any teaching in the prior art as a whole that would (not simply could, but would) have prompted the skilled person … to modify or adapt the closest prior art … thereby arriving at something falling within the terms of the claims, and thus achieving what the invention achieves”.[16]At [46] Jacob LJ (with whom Lloyd and Stanley Burnton LJJ agreed) rejected the suggestion that this passage went as far as to suggest that it had to be shown that the skilled person would go ahead and implement the idea commercially. I would however not accept (and I do not think that the court in Actavis v Novartis was accepting) that it must be established in every case that the skilled person would necessarily have arrived at the precise combination claimed. That would be to place another straitjacket on the law of obviousness. The skilled person may be faced with a range of obvious possibilities, making it statistically unlikely that he will settle on any one of them. They will all be obvious: see for example the well known discussion in Brugger v MedicAid [1996] RPC 635 at 661 lines 6-21, per Laddie J. In Hallen Co. v Brabantia (UK) Ltd. [1991] RPC 195 this court rejected a suggestion that a “would” test was always to be preferred. At page 212, Slade LJ, giving the judgment of the court, said:
“We accept that the “could” test is a minimum condition that must be satisfied on the facts before obviousness can be established in any given case. With this qualification, we agree with the judge’s comment … that the proper question depends on the facts of each case, though always bearing in mind that, under section 3 of the 1977 Act, the onus ultimately falls on the defendants to show that the alleged invention was obvious to a person skilled in the art, having regard to any matter which formed part of the state-of-the-art at the priority date.”
[17]Finally, the judgment of this court in Pozzoli v BDMO [2007] EWCA Civ 588 continues to provide a useful structured approach for judges and tribunals to approach the issue of obviousness.

The judgment of Birss J

[18]The judge analysed the issue of obviousness using the structured Pozzoli approach. He held that the difference between the disclosure of Carter and the claims of 628 was that Carter disclosed a liquid formulation of trastuzumab, whereas the claims related to a lyophilised formulation having the claimed characteristics.[19]The judge concluded that there was, at the priority date, a sufficient motivation to cause a skilled team to embark on the project of formulating trastuzumab. He also rejected a submission that there was no evidence that a liquid formulation would not be adequately stable, and that there was therefore no motivation to consider lyophilisation.[20]The judge then embarked on a consideration of the approach of the skilled team to investigating the lyophilisation of trastuzumab. He held that histidine was an obvious candidate, amongst others, as a buffer for the pH range at which trastuzumab was most stable. He next considered the surfactant and concluded that Tween 20 (polysorbate 20) was an obvious candidate. He then turned to the lyoprotectant. A sustained attack was mounted at trial as to whether trehalose would be amongst the lyoprotectants which the skilled formulator would consider. Notable amongst these was the fact that trehalose had not previously been used as a lyoprotectant in a drug formulation for humans, there were possible toxicity concerns given the route of administration and that these facts would mean that it was necessary to obtain regulatory approval for it. In the end the judge held trehalose was not simply one of the lyoprotectants worth considering, but was a very promising lyoprotectant, well worth testing. He was not persuaded that there was any concrete reason to be concerned by the parenteral route of administration. He concluded that concerns about toxicity would not put the skilled team off testing trehalose. As to regulatory approval, the team would know that regulatory approval would be needed if it turned out that trehalose was an excipient they wanted to use. That did not make testing trehalose inventive.[21]The judge ultimately concluded that if a skilled person were to draw up a list of eight lyoprotectants (two tiers of four) to test on a protein in 1996 then trehalose would be on that list.[22]As to the other excipients, the judge considered that histidine and Tween 20 might well be in the first-tier and that they would certainly be in a second tier. The judge then said this at paragraphs 225 and 226:
“225. The person skilled in the art is not a real person. The skilled person never sees what is inventive and never misses what is obvious. They represent part of the application of a legal standard to patents. It may be that some real skilled teams would find a working formulation which did not involve testing any of these ingredients. That does not alter the fact that all three ingredients are obvious agents to test. 226. In my judgment all of the differences between the claim and Carter are the result of nothing more than the application of routine screening techniques to common general knowledge excipients by a skilled team motivated in the way I have described already. There is no suggestion that any invention could be found to exist in the ratios or concentrations in the claim if the skilled team employed the relevant excipients. Accordingly there is a strong case that the claimed subject matter involves no inventive step over Carter.”
[23]The judge next dealt with and rejected Genentech’s submission that even if the evidence established that the skilled team could have arrived at the claimed combination that was not sufficient and that for a finding of obviousness, it needed to be shown that the skilled person would have arrived at it. The judge’s conclusions on this at [234] of his judgment were:
“It is not true to say that a real team would arrive at a formulation consisting of polysorbate 20, histidine and trehalose. It would be idle to pretend otherwise and Hospira do not do so. But what Hospira's submission is getting at is that the claimed result can be reached by the application of nothing other than routine approaches applied to excipients which were part of their common general knowledge. In my judgment on the facts of this case that is correct.”
[24]The judge next considered whether it could be said that the invention was just an arbitrary selection. He rejected that characterisation of the claim. The product made in accordance with the claim was stable and was therefore better than the prior art liquid formulation. It was also better than many lyophilised formulations. However it was not better than the Table 6 solution or the Table 4 or Table 7 solution. None of these solutions was arbitrary, but the fact that an invention produces a benefit does not necessarily mean it involves an inventive step. If the benefit is the product of the application of obvious steps to the common general knowledge starting from a public document, it did not help.[25]The judge also dealt with a suggestion that arriving at the claimed invention would require a “research programme”. He accepted that if the skilled team would only arrive at the claimed combination after a degree of persistence through unpromising results, then there could be a research programme sufficient to confer inventiveness. However he was not satisfied on the facts that this was such a case.[26]Finally the judge rejected the suggestion that Hospira’s notional path to the claimed combination involved any hindsight. This was not the case where each step in a chain of steps was necessary and it was only with hindsight that one could postulate the necessary series of steps to arrive at the invention. This was a case where, although a number of choices had to be made, the existence of those choices was not tainted with hindsight. In the present case the choices and tests fell to be made and carried out in parallel in a highly empirical field.[27]Accordingly the judge concluded that the invention was obvious over Carter.

Genentech’s submissions

[28]Mr Tappin submitted that the judge had made two errors in concluding that the skilled formulator would be motivated to make a lyophilised formulation of trastuzumab. The first error was to take the fact that trastuzumab was in a phase II trial as sufficient motivation. He accepted that the information in Carter made it at least credible that trastuzumab would be effective in the treatment of HER2+ breast cancer. However there was no evidence that the statement in Carter would have been sufficient for the formulator to have been asked by the clinician to formulate trastuzumab. The only basis on which Hospira’s clinician, Prof Leonard, had said that the formulator would be involved was knowledge of the phase II results and the existence of the phase III trial, neither of which formed part of the common general knowledge.[29]Secondly, Mr Tappin submitted that there was no evidence to support the judge’s finding that degradation in liquid formulations would be such that a lyophilised formulation would need to be investigated. He suggested that it was not until Genentech experienced problems in scaling-up their industrial process (when they started using stainless steel containers) that they realised that there was anything wrong with a liquid formulation from the point of view of stability. The skilled person would therefore have had no need to investigate a lyophilised formulation.[30]Mr Tappin’s major point, however, was that the judge had been wrong in principle to accept Hospira’s case that the claimed invention was obvious because it could be reached by the application of routine approaches using common general knowledge excipients. The task of formulating proteins was a difficult and unpredictable one and often encountered dead ends. There were no pointers in the common general knowledge or the pleaded prior art to the specific combination claimed, which, as the judge had held, had a beneficial effect and was not an arbitrary selection. In the absence of anything in the prior art or the common general knowledge to lead the skilled person to expect that the claimed combination would have a beneficial property, the necessary fair expectation of success was not present and it was not open to the judge to make a finding of obviousness.[31]Mr Tappin drew an analogy with the decision of this court in Teva UK Ltd v Leo Pharma A/S [2015] EWCA Civ 779; [2016] RPC 5. In that case the court held that the possible inclusion of a particular non-aqueous solvent in a formulation in a research programme in the hope that it might provide a stable medium for two otherwise incompatible active ingredients was not an adequate basis for a finding of obviousness. Similarly in this case, he submitted, the possibility that a stable formulation could be found by screening a variety of different excipients was not an adequate basis on which to find obviousness. It was the specific combination claimed which had to be obvious. The judge’s approach here was inconsistent with that adopted in Teva v Leo. On the other hand the case was very different from Ranbaxy (UK) Ltd v Warner-Lambert Co [2005] EWHC 2142 (Pat); [2006] FSR 14 where the prior art pointed to a limited number of salts, which would include the calcium salt of Atorvastatin, all of which would be included in a standard screen of salts, thereby leading the skilled person inevitably to the invention.[32]Mr Tappin also submitted that the judge had misapplied the distinction between “would” and “could”. The EPO jurisprudence showed that it was necessary for the court to decide whether the skilled person would carry out the necessary experiment on a formulation according to the claim. The judge had correctly concluded that a real skilled team would not arrive at a formulation as claimed. Having reached that conclusion, a finding of obviousness was not open to him. The possible inclusion of something in a program of testing, let alone a lengthy and unpredictable one, in order to see if it works but without any expectation of success did not establish lack of inventive step.[33]Mr Tappin submitted that the judge had also been right to conclude that the claimed invention was not an arbitrary selection but instead that it provided a technical benefit. Nevertheless, he had in effect treated the invention as an arbitrary selection by treating it as a case in which the invention would be obvious if the skilled person could have arrived at it.

Discussion

[34]I will deal first with Mr Tappin’s points about the motivation to be derived from the prior art and the common general knowledge. The prior art, Carter, is a pair of documents which it is accepted would be read together. The first is a book chapter entitled “Towards an immunotherapy for p185HER2 overexpressing tumors”. The only important sentence in the document is the penultimate one:
“One engineered molecule – our humanised anti-p185HER2 antibody 31 – is currently being evaluated in a phase II clinical trial for the treatment of p185HER2 – overexpressing breast cancer.”
[35]The molecule in question is trastuzumab. Reference 31 is to another Carter article which explains that trastuzumab is placed in phosphate buffered saline, i.e. a liquid solution. Thus the skilled person would know that trastuzumab was being developed and evaluated in phase II clinical trials for breast cancer.[36]It is true that the judge rejected Hospira’s case that the results of the phase II trials and the existence of the phase III trial were part of the common general knowledge. The judge did, however, find at [38] of his judgment, that it was common general knowledge that the HER2 receptor was a significant potential target for cancer therapies, and that skilled clinicians “would have been most interested to hear about drugs for that target being developed and would be interested in any information suggesting active steps had been taken towards developing a therapy or use in humans.” At [205] he addressed head-on Mr Tappin’s point that there was a lacuna in Hospira’s case on motivation because their evidence assumed the availability of phase II results and the existence of phase III trials. He concluded nevertheless that the real interest which the skilled person would have in any drugs targeted at the HER2 receptor, coupled with the knowledge of the earlier Slamon work, would amount to a sufficient motivation to cause a skilled team to think about formulating trastuzumab.[37]We were shown the evidence which underlay the judge’s conclusion. The evidence traversed the period from the 1980s to the priority date. The Slamon work was recognised to be high profile and made it clear that the HER2 receptor was a target which it was important to be able to hit. Professor Barrett-Lee, Genentech’s expert, made it clear that the statement in Carter would have been of real interest to clinicians because it indicated that an anti-HER2 antibody was now in phase II clinical trials. It would have been understood from the fact that phase II trials were under way that a significant amount of work would have been undertaken to get the antibody to this stage. It would also have been understood that the antibody must have shown efficacy or success in testing both in cell culture studies and in animal models.[38]The judge was well placed to form an impression of the degree of interest in trastuzumab by the time it was known that phase II clinical trials were under way. In my judgment, his conclusion on this aspect of the evidence was entirely open to him.[39]Mr Tappin’s second motivation point starts from the proposition that the skilled person would only start on a lyophilised formulation if he found that a liquid formulation was not sufficiently stable. Hospira’s expert, Professor Halbert, had been unable to point to any evidence indicating that a formulator would be unable to produce a liquid formulation of trastuzumab which was not sufficiently stable. The fact that Genentech subsequently changed to a lyophilised formulation was not a sufficient basis for the inference that they discovered it was unstable. There could have been other reasons. In fact there was material from which the judge could have concluded that Genentech only changed to a lyophilised formulation once they encountered scaling-up problems in large stainless steel containers.[40]This was not a point which featured largely at the trial and accordingly the judge dealt with it only very briefly. Genentech did not adduce any evidence from their own employees to mount a case that the liquid formulation showed itself to be stable until they encountered the problems in scaling up to large stainless steel containers. They relied solely on a single document which they used in cross-examination of Hospira’s expert. The only thing which was clear from the history was that Genentech started the phase III trial with a liquid but subsequently went on to develop a lyophilised formulation. The judge was perfectly entitled not to place reliance on the belated attempt to advance a more detailed theory of what was supposed to have happened at Genentech.[41]I think there is a lack of reality in this stability point. It was common ground that a drug company would prefer to make and sell a liquid formulation if one could be made stable enough. Genentech’s commercial formulation is not a liquid, but a lyophilised one. Moreover the patent itself explains at [0091], in a passage which the judge relied on, that trastuzumab undergoes degradation by deamidation in solution, so it is at least plausible that there would be a degradation problem which could not be solved in the liquid form. If Genentech had wanted to displace what I regard as the natural inference from these observations by evidence to the contrary, they could have done so, but they did not. The judge was, in my judgment, perfectly entitled to infer that trastuzumab could not be adequately stabilised in liquid form.[42]Having dealt with the judge’s findings on motivation, I turn to consider Mr Tappin’s main arguments as to how the judge erred in principle in his overall evaluation of obviousness. The first is, in essence, the argument that the judge was not entitled on the evidence available to him to find that there was a fair or reasonable expectation of success. The fact that the claimed excipients would each be on a list of eight or so excipients of each class which could be made the basis of screening experiments went nowhere towards establishing that there was the necessary expectation.[43]I am not at all persuaded by this argument. The judge’s factual findings went much further than the identification of the three excipients as common general knowledge candidates for testing. Firstly, the judge finds at paragraph [210] of his judgment that the results of using the standard screening methods described in Example 1 of the patent “are exactly what a skilled team would expect.” That is a finding that the skilled team embarking on the identified project would expect to find some good combinations and some less good. Later, when looking at the case in a slightly different way at paragraph 237 he said this:
“In my judgment in this case the skilled team who arrived at the Table 5 Solution would be pleased they produced a stable lyophilised formulation, and no doubt proud of their work, but not surprised. It is not a surprise is that a stable lyophilised formulation of trastuzumab can be made. Nor would the identity of the ingredients be a surprise. Although none of the excipients are the most commonly used, they are all part of the common general knowledge.”
[44]Mr Tappin attacked this paragraph of the judgment. In his submission it looked at matters the wrong way round. It was not right to consider the reaction of the skilled team once they had achieved success. Instead, it was necessary to ask what the skilled team’s expectation was at the outset of the project.[45]I do not agree. The judge’s findings at [210] and [237] need to be read together. At the beginning of the project the skilled team would expect to be able to produce promising candidates. If, at the end of the project, the skilled team would not be surprised by the result, that is further confirmation of what their expectation had been. Reading the judgment as a whole it is quite clear that the judge’s assessment was that the skilled team had an ex ante fair expectation of success.[46]The judge’s assessment was one he was plainly entitled to make. The skilled team knew that lyophilised formulations of proteins had been made, and made successfully, before. The three types of excipient are used for their known purposes, and all formed part of the formulator’s common general knowledge. There was nothing inventive in the screening approach adopted. There was no evidence that trastuzumab presented any special problems of its own. I can detect no error of principle in his conclusion on expectation of success.[47]I do not think Mr Tappin’s argument by analogy with Teva v Leo assists him. Teva v Leo did not state any new principle. The analogy breaks down when one sees what Sir Robin Jacob (with whom Kitchin and King LJJ agreed) thought about the evidence there:
“… it was far from established that any non-aqueous non-toxic solvent would produce a stable ointment. The judge found that “there is no proper evidence either way as to whether they [i.e. other such solvents] would work or not.”
Nor was it established that the skilled team would expect that any non-aqueous solvent would do. Keeping water wholly out was a known real problem.” (paragraph 19).[48]Later, at paragraph 25, Sir Robin Jacob says:
“it was not even proved that there was a good expectation that if you did try 20 [non-aqueous solvents] one of them would work.”
[49]Those facts are very different from the ones faced by Birss J in the present case. I am reminded of what Lopes LJ said in Savage v Harris & Sons (1896) 13 RPC 364 cited by Sir Donald Nicholls VC in Molnlycke AB and another v Procter & Gamble Ltd and others [1994] RPC 49 at 114:
“Cases, so far as regards the law, are most useful, but when they are applied to particular facts, they, as a rule, are of little service. Each case depends on its own particular facts and the facts of almost every case differ.”
[50]Next, I must deal with the could/would debate. I have already explained why I do not accept that it is necessary in every case for the court to conclude that the skilled person acting only on the basis of the prior art and his common general knowledge would arrive without invention at the precise combination claimed. Given that the screening methods were part of the common general knowledge, that the tests involved were routine, that the excipients were common general knowledge excipients and that there was no a priori reason why a successful lyophilised formulation could not be made, it seems to me that it was beyond argument that the claimed combination in this case was one that could be made by the skilled team. The question is whether this is the type of case where it is necessary to go further and ask whether the skilled person would necessarily have made the precise combination claimed.[51]In an empirical field it will be seldom be possible to predict in advance that any individual experiment will work. In many cases, the fact that a routine screening exercise could be carried out will be inadequate to establish obviousness. Nevertheless, on the facts of an individual case such as the present, the team may have a reasonable degree of confidence that a series of experiments will produce some which will work. To impose a requirement that the skilled team must be able to predict in advance which would be the successful combinations is wholly unrealistic. It would lead to the grant of patents for a whole variety of combinations which in fact involved no inventive effort.[52]It was inherent in Mr Tappin’s argument that it is only in the case of an invention which is an arbitrary selection that it is open to the court to find obviousness on the basis of a “could” test. As the judge had held that the combination here was not arbitrary, there could be no finding of obviousness. I do not agree. The case where the invention is simply a choice of one candidate from a field where all the individual candidates are the same is an extreme case. The present case is one where the skilled person would expect there to be a range of different results, some good and some bad, but that there was no invention in embarking on a screening process to pick out the good from the bad. The fact that one could not say in advance which the good ones would be does not, in every case, foreclose a finding of obviousness.[53]It is always necessary to remind oneself that it is not the function of this court to second-guess the judge’s finding of obviousness. The judge was evaluating a large number of inter-dependent factors. Despite Mr Tappin’s very clear and well sustained arguments, I do not think that the judge fell into any error of principle which would justify this court in undertaking its own evaluation.[54]For the reasons I have given, I would reject the grounds of appeal which relate to the issue of obviousness. In those circumstances the remaining issues do not arise, and I would dismiss the appeal. Lord Justice David Richards[55]I agree. Lord Justice Kitchin[56]I also agree.

Cited in 6 later judgments