840 patent – allegations of invalidity were not justified
736 patent – allegations of invalidity were justified

On June 8 2016 the Federal Court issued an order under the Patented Medicines (Notice of Compliance) Regulations prohibiting the minister of health from issuing a notice of compliance to Teva for atazanavir sulfate (Bristol-Myers Squibb's (BMS) Reyataz) until expiry of Novartis AG's Patent 2,250,840 (the '840 patent).(1) The court dismissed the prohibition application relating to BMS's Patent 2,317,736 (the '736 patent').

840 patent – allegations of invalidity were not justified

The 840 patent is a compound patent, claiming atazanavir.

Admissibility of prior art
A preliminary dispute arose regarding admissible prior art. Teva asserted that Australian Patent Application 9352479 (AU 479), which was filed and published before the relevant date, disclosed a compound which would have been the "starting point" in an obviousness analysis. However, on the face of the specification, the impugned pages were stamped with the date August 12 1996, which was after the relevant date in this case. Neither party provided evidence from an Australian patent law expert, but the court drew the "logical inference" that the pages were added after filing. The specific claim on those pages disclosing a relevant compound therefore did not form part of the prior art.

Teva also relied on a letter from the patentee to the European Patent Office which allegedly disclosed some testing data on the effectiveness of the prior art compound. The court found that there was no evidence to establish that the letter was available to the public at the relevant date. The court also rejected the letter as prior art on the basis that it was in German and no English translation was provided, and because Teva did not cite the letter as prior art in its notice of allegation and was thus not entitled to rely on it.

Inventive concept
In the first stage of the obviousness analysis, the court construed the inventive concept to be "a particular protease inhibitor (atazanavir) that is useful in the inhibition of HIV infections, and which has four advantageous properties":

  • a high degree of cellular activity better than saquinavir, indinavir and ritonavir;
  • high oral bioavailability, blood levels and plasma concentrations;
  • an advantageous resistance profile; and
  • high selectivity against HIV protease.

The court also commented on the expert evidence regarding inventive concept. Since Teva's expert witness did not describe in his affidavit the instructions that he had received on how to identify the inventive concept and could not recall those instructions on cross-examination, instead providing a "faulty understanding of the requirements of Canadian patent law", the evidence of BMS's expert was preferred by the court.

Starting point
The court rejected that a specific compound asserted by Teva to be claimed in the AU 479 application was the appropriate starting point, based on the above factual findings. Moreover, the inventors' course of conduct supported the argument that the asserted example was not the obvious starting point, as they tried many other approaches. However, the court accepted that the azapeptide "backbone" was "one, but by no means the only, logical place to start".

Obvious to try
Teva argued that the inventors' modifications to the starting backbone were obvious to try. The court rejected the submissions that the substitutions the inventors made were obvious to try, as they were not "more or less self evident" to the skilled person. In particular, it was not possible for the skilled person to predict the impact of structural changes on pharmaceutical properties and bioavailability.

A threshold issue arose as to whether atazanavir fell within the scope of a prior art genus patent. The court held that it did not, preferring BMS's expert evidence on the meaning of 'arylalkyl'. This was dispositive of the issue and the court held Teva's anticipation allegations to be unjustified.

However, the court went on to consider the allegations of anticipation on the merits, and found that the asserted genus patent did not anticipate the compound. On the disclosure requirement, the court found that the genus patent did not disclose advantageous properties of atazanavir, which were part of the construed inventive concept. On the enablement requirement, the court found that there was "no clear direction to get to atazanavir" from the prior art, and thus no enablement for the claimed compound. As such Teva's anticipation allegations were also unjustified for those reasons.

736 patent – allegations of invalidity were justified

The 736 patent claimed the bisulphate salt of atazanavir.

Claim construction
The primary point of contention between the parties was whether the claim to "a bisulphate salt of" atazanavir and a dependent claim included both Type I and Type II polymorphic forms of atazanavir bisulphate (as asserted by Teva), or only Type I (as asserted by BMS). After reviewing extensive expert evidence on construction and the patent specification, the court concluded that the claim, which provided a specific chemical structure for the Type I form, encompassed only Type I.

The court found that the work in making the bisulphate salt was part of a "standard salt screen", and that it was arrived at without difficulty – in this case on the "first day" of the drug development project. With respect to characterisation of the salts and their properties (eg, solubility, stability and hygroscopicity), the court did not consider the work to have been either prolonged or arduous. Thus, the court concluded that Teva's allegations of obviousness were justified. Bristol-Myers Squibb has appealed.

For further information on this topic please contact Kevin Siu at Smart & Biggar/Fetherstonhaugh by telephone (+1 416 593 5514) or email ([email protected]). The Smart & Biggar/Fetherstonhaugh website can be accessed at www.smart-biggar.ca.


(1) Bristol-Myers Squibb Canada Co v Teva Canada Limited, 2016 FC 580.