In Novartis v Teva & Ors, the Patents Court confirmed its ability to order an injunction against the infringement of patent applications prior to grant, but gave a stern warning as to when they will exercise the power and their concern to avoid evergreening (strategies used to extend the lifetime of a patent). In this case, Roth J. refused to grant the injunction because there was no loss that could not be adequately remedied in damages once the patent was granted (if that came to pass). Birss L.J. refused permission to appeal Roth J.’s decision to the Court of Appeal.
Novartis, brought a claim for a preliminary injunction against Teva and a number of other generics manufacturers to keep generic versions of Novartis’ prescription-only multiple sclerosis treatment, ‘fingolimod’ (sold under the brand name ‘Gilenya’) off the market. Novartis based the application on a pending European Patent Application with no. 2,959,894 (‘EP’894’). In November 2020, after the EPO refused to grant the patent, Novartis appealed to the Technical Board of Appeal (‘TBA’). The TBA did not hear the appeal until February 2022. The TBA considered EP’894 with regard to sufficiency and inventive step and remitted the application back to the Examination Division three days later. However, whilst awaiting the decision on grant in March 2022, Novartis filed pre-emptive proceedings, seeking a preliminary injunction against five generics (owned by Dr. Reddy’s, Glenmark, Teva, Tillomed, and Zentiva) attempting to enter the market. The generics had each already received marketing authorisation.
The Court’s Jurisdiction to Grant Injunctions
Given that the time (and at the time of writing), the application EP’894 had not yet been granted, the Court had to decide whether it had jurisdiction to injunct against generic market entry. The generics (in particular Dr. Reddy’s and Zentiva) argued that the Court had no such power. However, Roth J. found that, although a prospective patentee can only bring infringement proceeding post-grant, there is nothing in the statute preventing the Court from using its general authority to issue injunctions in such a case. The judge further found that, given that the grant of EP’894 was held back for merely administrative reasons, Novartis could seek interim relief for losses suffered between the generic entry and the date of grant. However, Roth J. warned that this was an exceptional case on it facts. In most cases, patentees attempting to seek interim relief on the basis that a patent is very likely to be granted (but has yet to be so) will be rejected by the Patents Court.
The Injunction Application
Having decided that he had the power issue injunctions in these circumstances, Roth J. then went on to consider whether one was justified in this case. To do so he applied the test from American Cyanamid:
1. Is there is a reasonable question to be tried?
This was a non-issue, as all parties agreed there was a reasonable question to be tried.
2. Are damages an adequate remedy for the Claimant (Novartis)?
Fingolimod (the medicine in dispute), unlike most other prescription medication, is prescribed in secondary care rather than by GPs. This means that the price is not determined by the NHS drug tariff. In this circumstance, the NHS puts out a tender in anticipation of generics entering the market. Prior to such tender, although generic entry would depress the cost of Fingolimod, there would not be the ‘downward price spiral’ often referred to in pharmaceutical interim injunction cases.
Whether a price spiral will occur in the time between the interim injunction and the final trial is extremely fact sensitive. Relevant factors in this case included that the time period before the tender was very short, and that the generics would keep a good record of their sales, which could then easily be quantified for damages if Novartis was successful at trial.
Roth J. found that there was no basis for arguing that future sales of Gilenya would be materially affected by the temporary sale of the generics – rather the monopoly prices would be restored when it returned to claim a monopoly. The judge was not persuaded that Novartis would suffer reputational damage due to the refusal of the injunction. He therefore found that damages would be an adequate remedy for Novartis and therefore no interim relief should be granted.
3. Would damages be an adequate remedy for the Defendants (the generics)?
Roth J. considered whether a cross-undertaking in damages would be an adequate remedy for the generics, in case he was wrong about the damages being an adequate remedy for Novartis. He considered that the forecasts for volume and price of generic sales were relevant but said that would be hard for a Court to adequately quantify damages for each of the generics, both individuals and collectively. Roth J. found that damages could not be an adequate remedy for the Defendants, as it would be hard to quantify such damages on an adequate basis. This is because such quantification would rely on a counterfactual world where there was generic competition, but in this world that competition did not happen and so accurately benchmarking the losses to the counterfactual could not adequately take place.
4. Where is the balance of convenience?
While it is not usually necessary for the Court to consider the balance of convenience unless damages would not be an adequate remedy for either party, in this case after finding that damages would not be an adequate remedy for the generics, Roth J. went on to consider, obiter, the arguments on balance of convenience. The judge would not make any decision as to the validity of the EP’894, as the position at that time was unclear. He also warned against evergreening and said that the balance of convenience would generally not allow potential patentees to stop generics clearing the way via preliminary injunction by extending the period until grant and thereby increasing the life of the patent.
Permission to Appeal
Permission to appeal was rejected by Roth J. himself and by Birss L.J. in the Court of Appeal. The key finding from Birss L.J. was that Roth J. applied the correct standard when he found that Novartis would be able to restore the monopoly prices for Gilenya should the patent be granted.
The substantive High Court trial, for which only the validity of the patent is in dispute, has been fixed for October 2022, prior to which it is extremely likely that EP’894 will have been granted by the EPO.
The generics had argued that the finding of jurisdiction to be able to issue the injunction in general would open the floodgates, but Roth J. rejected this based on the unusual circumstances of this case (i.e. that there was near certainty that a patent would be granted and to its scope). The unanswered question is how often would a pre-grant injunction be granted by the courts and in what circumstances and on that we have little guidance.
Roth J. thought that in the future, the Patents Court would be extremely wary of attempts to seek interim belief merely on the basis that the grant of a patent was ‘very likely,’ because that would go against the public policy of allowing Parliament to make decisions, and Parliament had decided that a claimant should not receive a remedy until the scope of the patent is finally determined by the IPO/EPO (i.e. at the point of grant).