Apotex Europe Ltd & Ors v Beecham Group Plc & Anor

LORD JUSTICE JACOB: This is an application for introduction of further evidence on appeal from a decision of Pumfrey J. The further evidence consists of some experimental work done by a Dr Sidhu some time ago. He was given a Canadian patent, asked to perform one of the examples and he did a number of experiments.

The issue to which this goes is the judge's finding of anticipation or obviousness over a British patent application called 407. This contains an example 1, which is directed at producing paroxetine anhydrate. What is said is that the instructions given are not sufficiently complete and that as a result one may not produce the anhydrate but produce instead the hemihydrate.

The evidence was not available at trial. In theory it might have been, because it was in the hands of some of the SKB personnel as a result of its being deployed by the defendants or their associated companies in Canadian proceedings. Mr Watson for Apotex accepted realistically that he cannot say that because of that deployment the British team for SKB could and should have put it in themselves.

What Mr Sidhu did was a whole series of experiments conveniently summarised in a table prepared by Mr Ward. What he actually achieved in the space of four days was the production of the anhydrate. However he started by attempting to follow the Canadian equivalent of example 1, using a comparatively small amount of isoproponol. He upped the quantities and sometimes took the precaution of keeping the atmosphere out by using a nitrogen or blanket, thereby keeping out the water.

It is submitted by Mr Turner that what was done by Dr Sidhu is inconsistent with what the principal witness for Apotex said he would have done when attempting to carry out 407 example 1. For Dr Cunningham did not start with a small quantity of isoproponol but did a calculation which suggested that for the purposes of getting a more satisfactory crystallisation he should start with 2 litres. He also said that even if he had started with a small amount he would expect a skilled man if necessary to change the amount. So, submits Mr Turner, the Sidhu experiment answers what Dr Cunningham said with hindsight he would have done: without hindsight, says Mr Turner, he would have started with a small amount as did Dr Sidhu, if that did not work then he would have abandoned the use of hydrochloric acid, go to HCL itself, the gas, and consequently avoided the water but ended up with the trouble of too much bound isoproponol solvates. So, says Mr Turner, the case would go the same as the previous litigation concerning this patent and the conclusion ought to be the same.

Further, says Mr Turner, in the circumstances this material never was privileged, although somebody seems to have thought that it was on the Apotex side, at least in this country. Perhaps also at least one SKB witness, thought there was some sort of restriction upon the disclosure. The material ought to have been disclosed on disclosure in this or to have been given to Dr Cunningham as open instructions, not material covered by the privilege. The whole matter should and ought to have been dealt with at trial.

There is I think some force in what Mr Turner says. It is a pity that it was not disclosed earlier; it was not dealt with expressly by Dr Cunningham. But it does not follow that it should be admitted now. There are practical reasons why it should not be admitted. For if it is admitted it seems almost inevitable that there would be a considerable expansion of the nature of this appeal. First, Dr Cunningham would have to comment on the material, although in practise a lot of comments which he would make on this material have been made in relation to other experiments; secondly, Dr Turner suggested that Dr Sidhu did not do his experiment without some prior assistance. He suggested either that Dr Sidhu had been given the patent in suit or its equivalent, or perhaps he was told of something by a Dr Horne or somebody else. I am bound to say that, looking at the affidavit, it would be odd if that were so because that was not the purpose for which he was doing the experiment. According to Dr Sidhu in his affidavit he was doing it to see if he could make the experiment work; if he was told those sorts of things first, what would be the point of his experiments at all? Why would he start with a small amount of material if he had been told it would not work? Or why would he start sometimes using nitrogen and sometimes not if told that he had to use nitrogen?

In short, on that evidence, there is considerable force in what Mr Watson says that this evidence is quite personal as to obviousness, namely, given the admittedly not entirely complete instructions in example 1, a little bit of reasonable persistence by an ordinary skilled man takes you into the claim.

Mr Turner says that was not the case that was advanced at trial; if this evidence comes in it is going to be a new case that is going to happen on the appeal, it is right other witnesses might be needed; Mr Sidhu might be called by one side or the other. In short, if this evidence is admitted the whole case is re-opened. I do not shrink from that. If I thought that the evidence would probably have an important influence on the result of the case I think it should be admitted. But I do not.

Dr Sidhu did not reason his way to using a substantial quantity of isoproponol straightaway, as did Dr Cunningham with hindsight, but on the other hand Dr Sidhu had the Canadian patent which did not contain the same information as was used by Dr Cunningham from 407. Nor does it seem likely that any inference should be drawn of the kind suggested by Mr Turner that Dr Sidhu was somehow helped in his experimental work. As I have said, I see no point in any such help. Certainly it is not clear that if he had patent in suit or its equivalent that that would have been the appropriate help.

There are these further considerations. There is in evidence already the work of some Japanese workers. They started with a small quantity then to a large quantity, which is what Dr Sidhu did. So it may be said: why would the introduction of his experiments make any difference? Further, the current position is that Dr Cunningham gave his evidence and gave reasons as to why he thought a significant quantity of isoproponol should be used. The judge evaluated that evidence against the evidence of Dr Lee for Smithkline Beecham. He preferred the evidence of Dr Cunningham. Although I have been shown the transcript as far as I could see there is no real evidence that Dr Cunningham's approach was not a reasonable approach. It is clear that Dr Lee thought an approach different from Dr Sidhu's would be better. That is not enough. We are here concerned with a recipe for making the product. Recipes are always subject to minor variations whether conducted by one worker or another, and so it is not surprising that one might have a different approach from another. Mr Turner says that Dr Sidhu did exactly what Dr Cunningham thought would be inappropriate; but that may well be because Dr Sidhu was not working with the same information, as I have said. In short, if this evidence were to be admitted a new case would be raised, and almost certainly would involve a new trial. On its face the evidence appears to confirm the case of obviousness in another way. It is not likely to have a material on the appeal.

The only other point I should mention is that some of the aspects relied upon by Mr Turner seem to be matters that plainly could have been dealt with at trial. Dr Cunningham's opinion that he would start with 2 litres appeared in his report a month before trial. Mr Turner wanted to use some of the material in the Sidhu experiments to say that unless the skilled man takes special precautions, like blanket filtering, plus having more isoproponol on that scale the experiment would not work. That is something that it seems to me could have been easily dealt with by an experiment in advance of trial. That is what experiments in patent actions are for. Mr Turner says that is hardly fair: there was only a month to go before trial and it was a speedy trial that had been ordered. I do not accept that explanation. If a company with the resources of Smithkline Beecham wanted to do those experiments they could have been done almost the same day as the report was received. They are not onerous experiments; they could easily have proved that Dr Cunningham's hypothesis as to what should be done was not enough to get the anhydrate - you also needed a nitrogen blanket.

For what it is worth I would also add this. We have the Sidhu notebooks. Nobody has actually inspected the experiments or seen a repetition of them. One does not know whether they are or are not repeatable. It was for that reason that rather similar work done by a Dr Wang was excluded from trial. Mr Turner says there is a difference. This is work done by an employee of the other side and therefore it somehow constitutes an admission against interest, or something like that. But I am by no means convinced that the work done in the laboratory done not for the purpose of litigation, as it apparently was not, should therefore somehow be treated as totally accurate and repeatable. If it should be so treated, why not Dr Wang's too?

For all these reasons I have come to the firm conclusion that this further evidence should not be admitted.