How much disclosure is required to support utility of a patented drug?

In a rather lengthy 271-page judgment, the Federal Court of Canada has held that Sanofi-Aventis Patent 1,336,777 (the “’777 patent”) for clopidogrel bisulfate (Plavix) was infringed by Apotex, but was invalid on the basis of lack of utility (Sanofi-Synthelabo Canada Inc v Apotex Inc, 2011 FC 1486, appeal pending). Noticeably, the same patent had previously been the subject of a complex litigation under the Patented Medicines (Notice of Compliance) Regulations, and was found to be both infringed and valid in a landmark decision by the Supreme Court of Canada (Apotex Inc v Sanofi-Synthelabo Canada Inc, 2008 SCC 61). After its loss in the Supreme Court, Apotex brought an impeachment action against Sanofi, after which Sanofi filed an infringement action. The decision rendered by the Federal Court on 6th December 2011 relates to both actions. The key issue in the determination of invalidty of the ‘777 patent in the present case was the lack of demonstrated or soundly predicted utility, an issue which was not addressed in the notice of compliance proceedings.

The ‘777 patent discloses and claims clopidogrel, its bisulfate salt and processes for its manufacture. The compound is the single enantiomer of a previously disclosed racemate used as an anticoagulant drug in humans. The ‘777 patent describes various tests on clopidogrel, which demonstrated that the compound had a better therapeutic index than the racemic mixture and a better tolerability, including ex vivo measurements of inhibition of platelet aggregation and in vivo measurements of toxicity and tolerability conducted in rats. Although not mentioned in the patent, human clinical trials were ongoing at the time when the patent application was filed.

In its analysis, the court ascertained the promise of the patent and opined that the ‘777 patent made an explicit promise for use of the compound in humans. Because the court was not persuaded that the utility in humans had been demonstrated, the court analysed whether the promise for use in humans was soundly predicted.

After considering the science and contextual history of the development and testing of the patented compound, the Federal Court acknowledged that Sanofi had a factual basis and a sound line of reasoning for predicting that clopidogrel bisulfate could be used in humans. However, the court concluded that the underlying factual basis and line of reasoning that grounded the inventor’s alleged prediction were not disclosed and, as a result, the patent failed the third branch of the sound prediction test.

This is one of a series of at least four fairly recent decisions in which patents for drugs were found invalid for lack of utility (Eli Lilly Canada v Novapharm Ltd, 2011 FC 1288, relating to the drug Olanzaprine (Ziprexa); Sanofi-Aventis Canada Inc v Apotex Inc, 2011 FCA 300, relating to the drug ramipril (Altace); Apotex Inc v Pfizer Canada Inc, 2011 FCA 236, relating to the drug latanoprost (Zalatan)). This calls into question the validity of many pharmaceutical patents, given the very high thresholds for utility in Canada compared to other jurisdictions. Indeed, the recent decisions of the Federal Court and the Court of Appeal seem to suggest that clinical trial data may be required for new drugs directed to humans. The Supreme Court of Canada may address this high threshold in Teva v Pfizer, which will be heard during Spring 2012.

These recent decisions highlight the importance for applicants to include in their patent applications any supporting data that they may have on hand to ensure that the utility requirement is met. The language used in the application should also be considered carefully to ensure the patent is not bound to a higher promised utility than what may be supported.