Background

Napp Pharmaceuticals (Napp) is the proprietor of a European (UK) patent, EP2305194B (the Patent), which provides protection for transdermal buprenorphine patches. Napp have marketed buprenorphine patches under the trade name “BuTrans®” since 2005. Both Sandoz and Dr Reddy’s were preparing to launch generic transdermal buprenorphine patches in the UK, but Sandoz were the first to obtain a marketing authorisation for their product. When Sandoz obtained a marketing authorisation, Napp responded by launching an infringement action in the UK Patents Court against both parties and applying for a preliminary injunction against Sandoz. Sandoz and Dr Reddy’s defended themselves against the infringement action but neither defendant challenged the validity of the Patent.

Summary

The Napp BuTrans® product works by using penetration-enhancing excipients to help buprenorphine pass through the skin of a patient so that it can be adsorbed into the patient’s bloodstream.

Claim 1 of the Patent defines a transdermal patch comprising the following features: (A) 10 %-wt buprenorphine base; (B) 10 to 15 %-wt levulinic acid; (C) about 10 %-wt oleyloleate; (D) 55 to 70 %-wt polyacrylate; and (E) 0 to 10 %-wt polyvinylpyrrolidone.

Construction of the first three features (A)-(C) of claim 1 was key to determining whether the generic products would infringe claim 1 of the Patent.

Input or Output Values?

Firstly, the Court found it necessary to determine whether the percentage values recited in claim 1 of the Patent related to the percentages used as ingredients in the patches’ manufacture (“input values”) or the percentages found in the patches after manufacture (“output values”). Napp used the text of the application as filed, the text of a PCT application cross-referenced in the application as filed and evidence for conventional terminology for transdermal delivery systems (TDSs) to support their assertion that the percentages recited in the claims were input values. They also argued that it would be difficult to determine output values especially since TDSs degraded over time. Sandoz and Dr Reddy’s responded by explaining why the text of the application as filed and the text of the PCT application supported their assertion that the percentages recited in the claims were output values and argued that the evidence presented did not establish conventional terminology for TDSs. They also argued that it would not be difficult to determine output values and dismissed any reference to the storage of TDSs as irrelevant. After considering the arguments of the parties, the Judge (Arnold J) concluded that the claimed ranges specified output values.

What does “about 10%” mean?

Secondly, it was necessary for the Court to determine the correct interpretation of the numerical ranges recited in claim 1 of the Patent. To help him decide on this matter, the judge considered recent case law, in particular Smith & Nephew v ConvTec [2015] EWCA Civ 607. In relation to featured (A) to (C) of the claim, Napp argued that “10 %-wt” covered from ≥7.5 to <17.5 %-wt, and that “about 10 %- wt” meant up to 15 %-wt such that the range covered extended from ≥7.5 to <17.5 %-wt, Arnold J did not agree. However, in his Judgment, following the reasoning in Smith & Nephew v ConvTec and in keeping with Sandoz and Dr Reddy’s arguments, Arnold J adopted the usual rounding convention such that “10 %-wt” extended from ≥9.5 to <10.5 %-wt, that “10 to 15 %-wt” extended from ≥9.5 to <15.5 %-wt. This left the issue of what effect the word “about” had to the term “10%” in feature (C) of the claim. The Judge held that “about” should be “taken to connote a small degree of permitted imprecision over and above that implied by the usual rounding convention” and therefore found that “about 10 %-wt” extended from ≥9 to <11 %-wt.

Arnold J then had to consider the de minimis principle in the context of infringement following the provision of information from the defendants as to the likelihood of any of their products falling within the claimed rages – i.e. if only a very small proportion of the defendants’ products were to fall within the claims of the Patent then would infringement have occurred? Arnold J considered the difficulties associated with determining whether the de minimis principle applied in the present case, including errors inherent in analytical measurements and small sample sizes. He concluded that, even if the level of infringement could not be discounted as de minimis, an injunction would be both disproportionate and a barrier to legitimate trade; a financial remedy would be more appropriate.

However, Arnold J held that, in accordance with his interpretation of the ranges in claim 1 of the Patent and as a result of the statistical likelihood of any product falling within those ranges being greater than 1 in 10,000 (1 in 69 million), i.e. de minimis, neither Sandoz’s product nor Dr Reddy’s product infringed the claims.

Napp appealed the first instance decision of Arnold J on the grounds that the judge should have construed (i) the claimed numerical ranges to specify the input values; and (ii) the claimed numerical ranges more broadly. During the appeal Floyd LJ considered the arguments put forth by Napp, but rejected their attack on the judge’s construction of the claim to specify output values. In particular, Floyd LJ noted that the claim related to a “device” and that it is normal to interpret product claims by determining what a product contains, not by determining what was used in a product’s manufacture. Floyd LJ also considered the arguments put forth by Napp regarding numerical ranges and confirmed the judge’s application of the usual rounding convention. Since the Court of Appeal agreed with the judge from the first instance the appeal was dismissed.

As a consequence of the Appeal Court’s decision, Sandoz and Dr Reddy’s are free to manufacture and sell generic products in competition with BuTrans® patches in the UK.

Finally, it is encouraging that this case was resolved particularly quickly by the UK Courts. Napp’s original claim was filed at the UK High Court in February 2016, with a first instance decision given in June 2016. The Court of Appeal heard the appeal in August 2016 and gave its Judgment that same month.

Comments

Chemical compositions may be claimed in terms of the percentages used as ingredients in their manufacture, i.e. the input values, and hence ranges in a claim to a chemical composition may be construed as specifying the input values. However, claim 1 in this case related to a device and the claimed ranges were construed as specifying the output values. Applicants seeking to claim a medical device using ranges for some of the ingredients used when manufacturing the device, should give careful consideration as to how such ranges will be interpreted, providing appropriate definitions within the description when necessary.

Although both the judge at first instance and the Court of Appeal referred to the patent description when interpreting the numerical ranges in the claims, the information provided in the description was not sufficient to deviate from the usual rounding convention being applied to the claimed ranges. Therefore, if a broader interpretation of a claimed range is intended, Applicants should consider including explicit definitions in the description or broadening the language used in the claim itself.