The validity of salt selection patents has been suspect in Canadian courts for many years, on the basis that the work performed in arriving at the invention (a salt form of a known compound) is routine work that is routinely done by every pharmaceutical company1.

Relatively recently, a patent to a crystal form of a salt was upheld on the basis that, though the empirical work leading to the invention was routine work routinely done, the amount of work and the unpredictability of the result were such that it could not be said that the invention was "more or less self-evident"2. On April 11, the Federal Court released its decision in Merck Sharpe & Dohme Corp. and Merck Canada Inc. v Pharmascience Inc.3, extending the trend that there may be inventiveness in salt selection and polymorph cases. Ultimately, the courts are increasingly saying that each case must be considered on its own facts.

What you need to know

  • There is no overriding principle that salt screen or polymorph patents are obvious just because they are matters of routine testing routinely performed in the pharmaceutical industry. The obviousness of every patent is a question of fact and must be determined on the facts, evidence and arguments presented in every case.
  • Where the amount of experimentation required is significant and the predictability of the patented outcome is uncertain, there will be a greater chance that a court will find the invention to be unobvious.
  • Testimony from the inventors, directly or indirectly, is important. This case underscores the need to support the inventiveness of salt selection and polymorph patents with evidence of the invention story and to have this information "at the ready" to be told at trial when called upon to do so.
  • Patent challengers alleging insufficiency of process claims bear the burden of showing that a skilled person, on the Canadian filing date, did not have sufficient information to make the claimed crystal forms. Still, patentees should ensure that they can demonstrate that the claimed process can be worked if they have evidence to that effect.


In this case, Pharmascience (PMS) challenged the validity of Canadian Patent No. 2,529,400 (400 Patent) on the basis of obviousness and insufficiency. The 400 Patent, related to Merck's innovative drug product, JANUVIA®, used to treat type 2 diabetes, contains patent claims to the medicine sitagliptin phosphate monohydrate (a salt), including claims to particular crystal form and a process to make it.

As is often the case, a prior art patent existed that disclosed a genus of compounds that included sitagliptin (and pharmaceutically acceptable salts) and specifically exemplified sitagliptin as both a free base and a hydrochloride salt, but as one of many examples in the prior art patent. Against this backdrop, the question before the Court was whether, in light of this prior art, the identification of sitagliptin as the lead compound, the selection of another salt, and the identification of the most stable crystal form of that salt were inventive.

The Honourable Justice Angela Furlanetto, a former patent litigator prior to her appointment to the Bench, rejected the notion that salt and polymorph patents were obvious per se. Rather, the Court noted that all prior cases, even Amlodipine and Atazanavir, supported the view that each case had to turn on its own facts.

Noting that "obviousness is a difficult test to satisfy because it necessitates showing that a [person of ordinary skill in the art] would have come directly and without difficulty to the invention, without the benefit of hindsight", the Court applied the four-step approach to obviousness suggested by the Supreme Court of Canada4 as well as its "obvious to try" analysis. The Court found that the closest prior art was the prior patent disclosing sitagliptin hydrochloride and its free base, and accepted that the inventive concept of the claims before her was the identification of the patented compound with its enhanced chemical and physical properties over prior art compounds. The Court parsed the invention story and compared and contrasted the experience of the inventors and collaborators with the expectations of a person of ordinary skill, as well as other "obvious to try" factors such as motivation:

"where a specific problem was not identified [in the prior art], it is difficult to see how a course of action would have been obvious to try without there being known and expected advantages from taking further steps. Just because there were known methods does not mean that a person would necessarily apply them unless they were more or less self-evident to try ."

The Court reviewed the detailed invention story, told through a single witness/inventor from Merck, and found it to be "not straight forward or predictable". Rather, Justice Furlanetto found that, even under an accelerated timeline, Merck's multidisciplinary team took over a year to arrive at the monohydrate salt, and only found it by chance. While the work to get to the crystal form was straightforward, the Court found this amount of work and effort to be "significant . without predictable results".

Much of the decision is taken up with objections to the testimony of Merck's sole inventor, as well as criticisms made by PMS that other inventors should have testified. Although the Court rejected these objections, this decision underscores the point that serious objections will be forcefully made. Patentees should be prepared to support the invention story with witnesses and evidence, as Merck did here, as well as a compelling narrative.

As to sufficiency, PMS argued that the claimed patented process could not be worked to make the patented crystal form (a test assessed at the Canadian patent application filing date). This argument triggered a debate as to whether some of Merck's experiments made something other than the claimed polymorph. This appears to have been a "side show" as, ultimately, the Court found that PMS, as the patent challenger, had not proven that the crystalline form of sitagliptin monohydrate could not be made following the methods set out in the patent.


1 See, for example, Pfizer Limited v Ratiopharm Inc., 2010 FCA 204 (Amlodipine) and Bristol-Myers Squibb Canada Co v Teva Canada Limited, 2017 FCA FCA 76 (Atazanavir).

2 Teva Canada Limited v Pfizer Canada Inc., 2019 FCA 15.

3 Merck Sharpe & Dohme Corp. and Merck Canada Inc. v Pharmascience Inc., 2022 FC 417: An appeal from this decision may be made to the Federal Court of Appeal, within 30 days of April 11, 2022—i.e., May 11, 2022.

4 Apotex Inc. v Sanofi-Synthelabo Canada Inc., 2008 SCC 61.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.