Soundly Predicted, But Not Obvious: Validity of MS Drug Patent Upheld on Appeal

February 7, 2022
By Martin Brandsma and Nyrie Israelian

In patent law, if the basis for a sound prediction comes from the common general knowledge, will that same common general knowledge render an invention obvious? Not necessarily. A recent decision from the Federal Court of Appeal revisits the interplay between Canadian patent law concepts of sound prediction and obviousness. It reconfirms that obviousness is not merely the reverse of sound prediction: there may be “enough in the common general knowledge to support a sound prediction of utility of an invention, but not enough to find the invention obvious”.


In Pharmascience Inc v. Teva Canada Innovation, 2022 FCA 2, the Federal Court of Appeal (FCA) upheld the Federal Court’s (FC) decision regarding the validity of Teva’s Canadian Patent No. 2,760,802 (the “‘802 Patent”). Teva had brought two patent infringement actions pursuant to subsection 6(1) of the Patented Medicines (Notice of Compliance) Regulations involving Teva’s multiple sclerosis drug COPAXONE. Teva alleged that Pharmascience would infringe Canadian Patents Nos. 2,702,437 and the ‘802 Patent if it were to enter the market with its product GLATECT 40 mg—a generic version of Teva’s COPAXONE 40 mg. The FC found the ‘437 Patent invalid for obviousness, but that the ‘802 Patent—the subject of the appeal—was valid and infringed.

The ‘802 Patent claims the use of 40 mg of glatiramer acetate (GA) for treatment of a type of multiple sclerosis called relapsing-remitting multiple sclerosis (RRMS), via thrice-weekly injections with at least one day between each injection.

Standard of Review

The FCA began by setting out the applicable standards of review on appeal: utility being a question of mixed fact and law, sound prediction a question of fact, and obviousness a question of mixed fact and law. As issues of law are reviewed on a standard of correctness, and issues of fact or of mixed fact and law from which no legal error is extricable are reviewed on the standard of palpable and overriding error, findings of utility and obviousness are reviewed for palpable and overriding error, except when a question of law can be extricated)

Sound Prediction

The parties agreed that, as of the ‘802 Patent filing date, utility had not been demonstrated, requiring Teva to meet the requirements of sound prediction, namely: (i) a factual basis for the prediction, (ii) an articulable and “sound” line of reasoning from which the desired result can be inferred from the factual basis, and (iii) proper disclosure.

Pharmascience argued that the FC erred on the disclosure analysis by failing to recognize the heightened disclosure requirement applicable to inventions that are based on a sound prediction. Whether there is a heightened disclosure requirement for sound prediction was not disagreed by the parties and the FCA considered it “not necessary to comment on that question here”. The FCA found no error in the FC’s understanding of the disclosure requirement, as the lower court recognized a distinction between disclosure generally (ss. 27(3)) and for utility (s. 2).

Notably, the FCA was “hesitant to interfere” with the FC’s conclusion on utility as Pharmascience failed to adduce evidence from its own experts on the issue, for which it had the burden of proof. While Pharmascience relied on the evidence from Teva’s experts, the FCA noted that those reports discussed the issue of obviousness, not utility, and Teva’s experts were not instructed on the law concerning utility.


In the alternative, Pharmascience also argued that if the common general knowledge was in fact sufficient to permit a sound prediction that the invention of the ‘802 Patent would work, that same common general knowledge must make the invention obvious to try.

The FCA rejected Pharmascience’s argument. It noted that the legal tests for sound prediction and obviousness are distinct and different. There is no inconsistency between a finding, on one hand, that an idea is sufficiently described in a patent disclosure and the common general knowledge to support a sound prediction of utility, and, on the other, that an idea is not sufficiently known in the prior art to lead to a conclusion of obviousness. It was open to the FC to find “enough in the common general knowledge to support a sound prediction of utility of the invention, but not enough to find the invention obvious”.

Moreover, the FCA was not persuaded by Pharmascience’s argument that several foreign courts had invalidated patents corresponding to the ‘802 Patent for obviousness, and that a CIPO re-examination board concluded that all of the claims of the ‘802 Patent were likewise invalid for obviousness. The FCA emphasized that there are myriad reasons why the Court was not bound by these decisions: “As regards the foreign decisions, the law is different, the patents are likely different, and the evidence is surely different”. Further, the decision under appeal is that of the FC, not of CIPO’s re-examination board, and the analysis as an appellate court applying standards of review is necessarily different.

While the FCA ultimately upheld the finding of non-obviousness, it did clarify two points from the lower court’s analysis. First, the FC observed several times in respect of a prior art patent application that the POSITA would assess its teachings based on the fact that it “was simply an application that was not approved”. The FCA clarified that this reference should not be understood to mean that whether a prior art patent has issued is relevant to its citability on the issue of obviousness. Second, it was not correct for the FC to identify the fourth step of the Sanofi obviousness test (i.e., whether differences between the art and the invention are obvious steps) as the “obvious to try” test. The FCA noted that the fourth step of the Sanofi test does not necessarily involve an assessment of whether the invention was obvious to try. The FCA was satisfied, however, that an “obvious to try” test seemed appropriate in this case as it concerned “biological responses to certain therapeutic approaches”.


Subscribe to our newsletter

You can unsubscribe at any time. This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

This site is registered on as a development site.