Glaxosmithkline Biologicals SA v Novartis AG

I have to deal with the costs of two actions involving European Patent (UK) No. 1,401,489 (“the Patent”). The first in time is a claim by GlaxoSmithKline Biologicals SA for revocation. The second in time is a claim by Novartis AG, the patentee, against GlaxoSmithKline Biologicals SA and another company in the same group (together, “GSK”) for infringement.

The background is set out to a large extent in two previous judgments, the first being that of Kitchin J on 30 April 2009, [2009] EWHC 931 (Pat), refusing an application by Novartis for a stay of the proceedings for a period of 12 months. The second is a judgment of mine on 2 October 2009, [2009] EWHC 2422 (Pat), giving directions at a case management conference in the infringement proceedings. Concentrating on essentials, the position may be summarised as follows.

The Patent was granted on 5 April 2006. As granted, claim 1 was as follows:

“A process for purifying a bacterial capsular polysaccharide, comprising the steps of (a) precipitation of the polysaccharide using one or more cationic detergents, followed by (b) solubilisation of the precipitated polysaccharide using an alcohol.”

As can be seen, that is a fairly broad claim.

On 5 January 2007 GSK filed an opposition to the grant of the Patent in the European Patent Office. On 24 October 2008, the Opposition Division held that claim 1 as granted was not valid (indeed, its validity was not supported by Novartis before the Opposition Division), but nevertheless decided to maintain the Patent in amended form.

On 23 December 2008 GSK filed an appeal against that decision to the Technical Board of Appeal and on the same date it commenced the revocation proceedings in this country. In addition, GSK requested expedition of the appeal in a letter to the EPO dated 23 December 2008.

On 5 February 2009 Novartis' solicitors wrote to GSK's solicitors proposing that there should be a stay of the revocation proceedings for a period of 12 months in view of the pending EPO appeal and of parallel revocation proceedings which had been brought by GSK in Belgium.

For present purposes, it is relevant to note certain passages from that letter. The first reads:

"The suspensory effect of the pending appeal means that the claims presently in force in the UK, i.e. the claims as granted, differ from those presently considered allowable in the EPO. There is therefore already a discrepancy between the claims extant in the EPO and in these proceedings. However, given that your client has requested expedition of the appeal proceedings in the EPO, we anticipate that any delay should be minimised. We understand that expedited proceedings before Board 3.3.4 might be completed in around 2 years."

Pausing there, it was common ground before me that the letter dated 23 December 2008 did not in fact identify grounds which fell within the established criteria applied by the Boards of Appeal in the EPO for expediting appeals.

Returning to the letter of 5 February 2009, Novartis' solicitors also said:

"We consider that it would be most appropriate for the EPO to decide on the allowability of the claims and any amendments thereto. Not only is the EPO the most experienced forum for such matters, it would avoid a wholly unsatisfactory position of having multiple different versions and claims of the Patent being prosecuted simultaneously before the English Court and the EPO."

A little later, they said:

"In view of the above, we are instructed to propose that these proceedings be stayed for a period of 12 months. During that time, both the Belgian proceedings and the EPO Appeal are likely to have progressed significantly. At that stage, both parties will be in a better position to assess the situation having regard to all of the factors above. We consider that a stay would result in significant savings on costs and court time that would otherwise be wasted if the three sets of proceedings run in parallel."

GSK declined to agree to the stay proposed by Novartis in that letter. On 27 February 2009 Novartis served an application notice seeking an order for a stay. That application ultimately came before Kitchin J on 30 April 2009. It is important to note that, during the intervening period, an order was made on 3 March 2009 by a Belgian court for a saisie, which it appears was applied for by Novartis on 3 February 2009 That order authorised Novartis to:

“… collect proof of the alleged infringement of the Belgian, German, French and the UK parts of the Patent no EP 1401489 at the registered office of GSK Biologicals, rue de l'Institut 89, at 1330 Rixensart, and at any location where GSK has an operational seat, as well as at any affiliate of GSK in Belgium, where the processes and/or products suspected of infringing the Belgian, German, French and/or UK parts of EP '489 may be described, and where any information relating to the possible acts of infringement by GSK or any of its affiliated companies of the Belgian, German, French and English parts of EP '489 may be collected ....”

In his judgment refusing the stay, Kitchin J considered a series of factors which he considered relevant to the exercise of his discretion. In particular, there were nine factors which are set out and considered in paragraphs [32]-[41] of the judgment. For present purposes, the fifth factor discussed in the judgment is particularly relevant. Kitchin J said at [37]:

“Fifth, conversely, I accept that to allow these proceedings to continue will involve an element of duplication and it may prove to be the case that UK proceedings are entirely unnecessary if, in due course, the EPO decides to reverse the decision of the Opposition Division and to revoke the Patent. Against this, however, I must weigh in the balance that both parties to the present proceedings are substantial and the products in issue constitute an important part of the claimant's business. Moreover, the costs of the UK proceedings will only prove to have been wasted if the Board of Appeal does indeed revoke the Patent. If, on the other hand, it upholds the Patent, then the commencement and pursuit of the proceedings will prove to have been justified.”

Also relevant is part of what Kitchin J said in relation to the eighth factor at [40]:

“Eighth, the defendant also relies upon the fact that the period of the stay is short and initially only 12 months, by which time, it says, the EPO proceedings will be well advanced and may well have terminated and the Belgian first instance trial will have occurred. As for the EPO, I am far from persuaded that it is likely that within 12 months the appeal proceedings will be well advanced. Without acceleration, which the defendant has not yet requested, the appeal may not be heard until 2011. In the meantime, the claimant is faced with what I believe to be a significant threat to its UK business, which a first instance decision of this court would do much to diminish, assuming, of course, that it decides to revoke the Patent….”

As the latter part of that quotation indicates, a significant element of Kitchin J's reasoning in refusing the stay was that he considered, as a result of the saisie order in Belgium, that there was a significant threat posed by the UK designation of the Patent to GSK's vaccine business. Accordingly, he considered that GSK was justified in wanting to clear the path of that obstacle as quickly as possible.

It is also relevant to note that, at the time of the hearing before Kitchin J, Novartis’ position, as set out in a witness statement filed in advance of the hearing, was that it would support GSK’s application for an expedited hearing of the appeal. Despite this, neither party requested Kitchin J to write to the President of the Boards of Appeal of the EPO to request accelerated processing of the appeal.

On 18 May 2009 Novartis' European patent attorneys wrote to the EPO filing a cross-appeal against the decision of the Opposition Division and responding to GSK's letter of 23 December 2008 requesting accelerated processing of the appeal. In that letter, Novartis' attorneys referred to the established criteria for expedition and said:

"The first situation did not apply when the opponent made its request for accelerated processing in December 2008, and still no infringement proceedings have been brought. The patentee will notify the Board if this situation changes."

It is to be observed that, as the letter itself says, the first criterion is not limited to a situation where infringement proceedings have been brought, but extends to the situation where infringement proceeding are envisaged. The letter was silent as to the fact that Novartis had applied for, and been granted, a saisie in Belgium on the basis I have quoted earlier.

On 29 May 2009 GSK wrote to the EPO reiterating its request for expedition of the appeal. In that letter, it drew to the EPO's attention the saisie that had been granted in Belgium. It pointed out that the stated purpose of the saisie was to collect proof of the activities of GlaxoSmithKline Biologicals SA which allegedly infringed the Belgian, French, German and UK designations of the Patent. Accordingly, GSK argued that the saisie demonstrated that infringement proceedings were envisaged by Novartis against GlaxoSmithKline Biologicals SA in Belgium, France, Germany and the UK.

On 3 August 2009 Novartis brought the proceedings for infringement in this country. It also brought infringement proceedings in Belgium.

On 6 August 2009 GSK wrote once more to the EPO requesting expedition of the appeal, this time on the basis of the commencement of the infringement proceedings in this country.

On 11 August 2009 Novartis' European patent attorneys wrote to the EPO informing the EPO that Novartis had commenced infringement proceedings in both Belgium and the United Kingdom. The letter went on:

“The opponent has previously requested accelerated handling of the appeal proceedings. If the Board accepts its request then, of course, the patentee will co-operate, but this co-operation must not jeopardise the patentee's rights.”

A little later, the letter said this:

“The patentee does not resist the opponent's unilateral request for accelerated appeal proceedings, but merely asks for a reasonable procedure if the request is granted.”

The position as at the time of the hearing before me on 2 October 2009 was that the Technical Board of Appeal had indicated that it would give a decision on the request for expedition upon the expiry of the deadline for the filing of the parties' responses to the appeals that had been filed on 13 October 2009. During the course of the hearing, I suggested that I should write to the President of the Boards of Appeal requesting accelerated processing of the appeal. That suggestion was supported by GSK. Novartis, by contrast, expressed itself as being neutral on that proposal.

Immediately after the hearing, I did write to the President of the Boards of Appeal by letter dated 2 October 2009 requesting accelerated processing of the appeal. I am happy to acknowledge that the EPO reacted both promptly and favourably to that request. On 16 October 2009, the Technical Board of Appeal issued a summons to the parties for a two-day hearing of the appeal on 22 and 23 December 2009.

Not long after that, it was agreed between the parties that it made sense to stay both of the UK proceedings pending the decision of the Board of Appeal and, as a consequence, to vacate the trial of the revocation proceedings which had been fixed for 11 January 2010. In due course, I made a consent order to that effect on 11 November 2009.

The appeal was heard by the Board of Appeal over the course of 22 and 23 December 2009. As related in the written reasons subsequently given by the Board of Appeal dated 19 February 2010, during the course of the hearing, the main request filed by Novartis and four auxiliary requests filed by Novartis with a letter dated 13 October 2009 were debated. In addition, two additional auxiliary requests filed by Novartis during the course of the hearing were also considered. At the conclusion of the oral proceedings, Novartis withdrew all requests which it had filed except for two, that is to say, its main request and one of the two auxiliary requests filed during the course of the oral proceedings.

The decision of the Board of Appeal was to refuse the main request on the ground of lack of inventive step over an item of prior art referred to in the EPO proceedings as D27. So far as the auxiliary request is concerned, it is important to note that, as recorded in the decision, GSK (the opponent in the EPO proceedings) made no submissions in relation to the patentability of the claims the subject of that auxiliary request. That was because GSK took the view that it was not concerned by that set of claims as there was no prospect of it infringing them, or being alleged to have infringed them.

So far as the patentability of the claims the subject of that auxiliary request is concerned, the Board of Appeal held, in the absence of arguments to the contrary by GSK, that those claims satisfied the requirements of the European Patent Convention in that they were novel and did not lack an inventive step.

Claim 1 of the claims upheld by the Board of Appeal, is in the following terms:

"A process for conjugating a bacterial capsular polysaccharide to a carrier protein, comprising purifying the polysaccharide, comprising the steps of (a) precipitation of the polysaccharide using one or more cationic detergents, followed by (b) solubilisation of the precipitated polysaccharide using ethanol at a final concentration of between 75% and 95%, then (c) treating the solubilised polysaccharide obtained in step (b) to remove contaminants comprising size filtration and/or ultrafiltration, activation or functionalisation of the polysaccharide, and conjugation of the polysaccharide to a carrier protein, wherein the carrier protein is a CRM197 diphtheria toxoid, and wherein the bacterial capsular polysaccharide is from N.meningitidis and the cationic detergent is cetyltrimethylammonium bromide."

As can be seen, that claim is a significantly narrower claim than claim 1 as originally granted.

Following the decision of the Board of Appeal, both parties have recognised that the UK proceedings are academic. They are academic because, so far as the allegation of infringement is concerned, Novartis accepts that it has no case of infringement against GSK in respect of the amended claims upheld by the Board of Appeal. So far as validity is concerned, GSK, there being no threat of infringement, sees no point in pursuing any claim for revocation of the Patent in this country. Accordingly, both parties seek permission to discontinue their respective claims: GSK seeks permission to discontinue its revocation proceedings and Novartis seeks permission to discontinue its infringement proceedings. Each consents to the other's discontinuance.

The only issue between the parties is as to the costs of the proceedings. I say "the only issue", but that nevertheless is a significant issue because, as can be expected given the parties, the nature of the issues and the stage at which the proceedings had reached prior to 11 November 2009, significant costs have been expended on each side, of the order of £500,000 each.

In these circumstances, counsel for Novartis submits, and I accept, that the applicable provision in the CPR is rule 38.6(1), which provides:

"Unless the court orders otherwise, a claimant who discontinues is liable for the costs which a defendant against whom the claimant discontinues incurred on or before the date on which notice of discontinuance was served on the defendant."

My attention has been drawn to two authorities as to the interpretation and application of that rule. The first is the decision of the Court of Appeal in Walker v Walker [2005] EWCA Civ 247 and the second is the decision of Lightman J in RBG Resources plc v Rastogi [2005] EWHC 994 (Ch). In the latter case, Lightman J deduced the following principles from the judgment of the Court of Appeal at [48]:

“(i)

it is no part of the function of a court on an application to discontinue to attempt to reach a decision on whether or not the claim will succeed (para 12);

(ii)

the burden is upon the party who seeks to persuade the court that some other consequence should follow than that the claimant should bear the defendant's costs on discontinuance and the task of the court is to consider whether there is some good reason to depart from the normal order (para 24);

(iii)

the test to be applied is not the simple one of looking at the action as it is and seeing what is the fair and just thing to do at the moment in time (para 36);

(iv)

justice will normally lead to the conclusion that a defendant who defends himself at substantial expense against a claimant who changes his mind in the middle of the action for no good reason other than that he has re-evaluated the factors which have remained unchallenged should be compensated in costs (para 36); and

(v)

it is not the law that a claimant will only be required to pay a defendant's costs on discontinuance if he is in effect surrendering and acknowledging defeat (para 37).”

Counsel for GSK submitted, and I accept, that the factual circumstances in the present case are significantly different to those pertaining in either Walker v Walker or RBG. Furthermore, he submitted, and again I accept, that, particularly so far as the revocation claim is concerned, it would be wrong to place too much emphasis upon the fact that GSK is discontinuing. Rather, it is important to have regard to the reason why GSK is discontinuing, namely, the effect of the intervening decision of the Board of Appeal on 23 December 2009.

The respective positions of the parties on costs are diametrically opposed. GSK submits that it should have all of its costs of the proceedings to date. Novartis contends that it should have all of its costs of the proceedings to date, although by way of fallback position it says that it should have the costs of the revocation proceedings even if it not the infringement proceedings, or at any rate, it should have them since the date of the request for a stay.

If one views the matter in terms of a good reason to depart from the normal order under rule 38.6(1), the principal point made by counsel for GSK is that, in commercial terms, GSK is the winner of the litigation. GSK says that it is the winner because the net result of all the proceedings has been that Novartis is left with a patent which GSK is not even alleged to infringe. On that basis, GSK submits that costs should follow the event.

Novartis counters that, first, that is an overly simplistic view of who was the winner given that GSK attacked the validity of the Patent as a whole and was unsuccessful in doing so. Furthermore, the Patent as upheld by the Board of Appeal remains of significant commercial value to Novartis because it covers a product of Novartis called Menveo.

Secondly, and to my mind more importantly, Novartis points out that, in the events that have happened, its application for a stay was justified. If the stay had been granted, the costs of the UK proceedings incurred since then would have been avoided. Accordingly, Novartis contends that a significant quantity of costs on both sides has been wasted as a result of GSK's successful opposition to the stay.

Before proceeding further, I should observe that both counsel relied upon a number of subsidiary points in support of their clients' respective positions. To my mind, the subsidiary points on either side are of little weight by comparison to the main points. Furthermore, the arguments on both sides involved criticising the other side's conduct of the respective proceedings. During the course of the hearing on 2 October 2009, I observed that it appeared to me that both sides were jockeying for procedural position in a way that did neither of them any credit. Revisiting the matter nine months later, I find that impression confirmed, and indeed reinforced.

Returning to the main points which have been urged upon me in support of the parties' respective positions, it seems to me that, if it had not been for the stay application, GSK's argument would have been unanswerable. In those circumstances, GSK would, in commercial terms, have been the winner of the litigation and I would see that a good reason why they should recover all of their costs.

However, it seems to me that the stay application puts a different complexion on the matter. One of the main arguments advanced by Novartis in support of the application for a stay was that duplication of proceedings between the UK and the EPO would lead to a risk of wasted costs. GSK chose to resist the stay because of what it perceived to be the commercial importance of achieving certainty in the United Kingdom at the earliest possible date. As can be seen from the passages that I have quoted from Kitchin J's judgment, he accepted GSK's argument whilst recognising the validity of Novartis' point that it entailed a risk of wasted costs.

In the events that have occurred, it can now be seen that costs have indeed been wasted, which would not have been wasted if the stay had been granted. The question that arises in those circumstances is whether good reason has been shown for departing from the normal order under CPR rule 38.6(1).

It is convenient to begin with the infringement proceedings. So far as those are concerned, notwithstanding the point regarding the application for a stay, I do not see that there is any good reason for departing from the normal order. At the time those proceedings were brought, the stay application had already been refused. There was no obligation on Novartis to bring infringement proceedings at that point in time. It chose to do so in the face of its own recognition of the potential for wasted costs. It seems to me that, in those circumstances, given that the proceedings have ultimately failed, Novartis should pay the costs of the infringement proceedings.

Turning to the revocation proceedings, matters stand somewhat differently. As I have already observed, it seems to me that, with the benefit of hindsight, it can now be seen that Novartis was correct to prophesy that costs would be wasted if a stay was not granted. I do not consider, however, that that leads to the conclusion that Novartis should recover its costs since the request for a stay was made. To my mind, it is very telling that, once the Board of Appeal had agreed to expedite the hearing of the appeal, GSK agreed to a stay of the UK proceedings.

That demonstrates two things. First, it demonstrates a common recognition at that stage that a stay was prudent to avoid wasted costs; but secondly, it demonstrates that, as is so often the case in these situations, the relative timings of the two sets of proceedings are crucial. It is clear from the history that an important factor in GSK's resistance to a stay up to that point had been the fact that the EPO proceedings were, unless accelerated, likely to take up to two years to resolve. Once the Board of Appeal accepted the request for expedition and granted an early date for the hearing of the appeal, GSK's position understandably changed.

It seems to me to be clear that, if expedition had been granted at an earlier stage, likewise GSK would have consented to a stay at an earlier stage. An important reason why expedition was not granted at an earlier stage is because GSK's request for expedition was not supported by Novartis. That was despite the facts that grounds for seeking expedition existed as soon as the saisie order was made, and arguably as soon as it was applied for, and that Novartis’ position before Kitchin J was that it would support the request.

As I have related, even as late as 11 August 2009, the position adopted by Novartis through its European patent attorneys was no more than to say that it did not resist a request for accelerated appeal proceedings, but at the same time to request what was described as "a reasonable procedure". Nor did Novartis support my proposal to request expedition at the hearing on 2 October 2009.

Returning to the question of who bears the responsibility for the wasted costs of the UK revocation proceedings, I have come to the conclusion that both parties bear a measure of responsibility. GSK bear a measure of responsibility for having resisted a stay, but that responsibility is shared by Novartis for having resisted, or at any rate not supported, the application for expedition of the appeal for as long as they did. I consider that, if Novartis had supported the application, at least once the saisie had been ordered, it is likely that expedition would have been granted sooner by the EPO, particularly if the parties had asked Kitchin J to support the application. If expedition had been granted at an earlier date, then it is likely that the parties would have agreed to a stay earlier than they did.

My conclusion, therefore, is that there is good reason for departing from the normal order under rule 38.6, but not in the manner proposed by either party. I consider that, in the particular circumstances of this case, the right order is that Novartis should bear the costs of the revocation proceedings down to the date of its application notice seeking a stay. As from that date, each party should bear their own costs.