In Napp Pharmaceutical Holdings Ltd v Dr Reddy’s Laboratories (UK) Ltd and Sandoz Limited, Arnold J refused to grant a final injunction against the defendants, who were proposing to launch generic seven-day analgesic transdermal patches in competition with a patented Napp product already on the market. The marketing authorisations sought by both defendants relied on the fact that their respective products are bioequivalent to Napp’s product. However the defendants claimed, and Arnold J agreed, that both products work around the patent. In reaching this decision, Arnold J examined the law on quia timet actions and on de minimis infringement in patent cases.
The case turned on two key issues: (i) construction of the patent – in particular: (a) construction of numerical ranges in the claims; and (b) whether those figures refer to ingredients in the recipe or the composition of the final product (i.e. input or output figures); and (ii) the risk of infringement presented by the degree of variability in each of the defendants’ products, coupled with inaccuracy in analytical techniques used to determine patch content.
The relative percentage content of three key chemicals held within the patch were specified in Napp’s patent, and it was around these numerical percentage composition ranges that the case hinged. Napp argued that the ranges in the patent should be construed widely – with the limit in the claim being expressed to the nearest 5%. Arnold J disagreed, and construed these numerical ranges in the claim as being expressed to the nearest whole number.
A small representative sample of the defendants’ products were tested and the chemical composition of all tested products fell outside of the ranges claimed in the patent (construed as being expressed to the nearest whole number). However, Napp also argued that the inherent variability in the patches is such that there is a threat by the defendants to sell an appreciable quantity of infringing products. This claim raised a number of legal issues including: (i) the application of the de minimis principle in the context of patent infringement; (ii) the test for infringement when the claim is made quia timet; (iii) the burden of proof; (iv) the standard of proof; and (v) remedies – including the availability of injunctive relief.
The court commented that application of the de minimis principle must be fact specific, but that where 0.01% of products fall within the claim, it is precisely the type of situation covered by that principle. The court found that if there is a threat to do something that would only involve infringement on a de minimis scale, that threat does not justify the commencement of proceedings. Arnold J decided it was for the claimant to prove, on the balance of probabilities, what proportion of patches would fall within the claim, and whether that proportion was more than de minimis. In considering the appropriate statistical test Arnold J held that Napp chose to rely on a small sample set, and so it was not entitled to take advantage of the uncertainties that flowed from its own choices.
Based on the statistical evidence, the court found the likelihood of infringement was de minimis and there was therefore no threat to infringe by the defendants. The court further approved of the testing regime and statistical protocol proposed by Dr Reddy’s to ensure that patches marketed in the UK do not fall within the claim other than to a de minimis extent.
As to injunctive relief, Arnold J commented (obiter, and applying the Enforcement Directive) that, even if the level of infringement is very low but cannot be discounted as de minimis, an injunction would be disproportionate and a barrier to legitimate trade. The practical effects of the injunction would be to require the defendant to operate even further outside of the boundaries of the claim, and effectively extend the scope of the patentee’s monopoly. Accordingly, in such a case the appropriate remedy would be a financial one.
Napp was granted permission to appeal in respect of issues (i)(a) and (b) (above), and sought an interim injunction against the defendants pending the appeal.
Notwithstanding the judge’s view that Napp’s prospects of success were “relatively weak“, Arnold J followed Novartis v Hospira  EWCA Civ 583, which stated that as long as the court was satisfied that there was a real prospect of success (which is a requirement for permission to appeal), it was usually not useful to attempt to form a view as to how much stronger the prospects of appeal are. Arnold J did consider the relative merits of the parties’ cases as per National Commercial Bank of Jamaica v Olint Corp  UKPC16, but found that in light of Novartis v Hospira the unlikely prospects of success by Napp were insufficient to outweigh the conclusion that he would otherwise reach. It is therefore currently difficult to see an English court deciding that the merits of an appeal are sufficiently low to justify applying Olint and refusing interim relief where a patentee is granted permission to appeal.
On the facts, Arnold J found that damages would not be an adequate remedy for any party. In deciding between the “lesser of two evils“, Arnold J favoured maintaining the status quo and granted the injunction, but only until 16 August, by which time the judge anticipated that the appeal would have been disposed of. One of the factors in Arnold J’s reasoning was that the defendants could have avoided the problem if they had cleared the path by revoking the patent or seeking a declaration of non-infringement in advance of their proposed launch. The tensions between “clearing the path” and the commercial realities of a highly competitive generic medicines marketplace where there is a clear desire not to advertise one’s intentions to your competitors therefore continue to make the UK a challenging legal environment within which to navigate generic pharmaceutical product launches.