On March 14, 2017, the Federal Court of Appeal (FCA) dismissed Teva’s appeal of Locke J.’s Order prohibiting the Minister of Health from issuing a notice of compliance (NOC) to Teva in respect of its calcipotriol and betamethasone dipropionate ointment (Leo Pharma’s DOVOBET) until the expiry of Patent No. 2,370,565: Teva Canada Limited v Leo Pharma Inc, 2017 FCA 50. The patent covers a three-component, non-aqueous ointment used in the treatment of psoriasis. The ointment is comprised of vitamin D or a vitamin D analogue (Component A), a corticosteroid (Component B), and a solvent (Component C). Teva’s appeal related to sound prediction and sufficiency.

Locke J. rejected Teva’s allegation relating to sound prediction: each element of the test was satisfied. Although no one at Leo understood exactly why the claimed combination worked, and although it was not expressly mentioned in the disclosure, the skilled person would have understood that the inventors’ line of reasoning was that the alternatives for Components A and B share a chemical scaffold, and would therefore behave similarly.

Locke J. rejected Teva’s allegation of insufficiency on the basis that Teva had failed to raise the allegation in the Notice of Allegation. He nonetheless evaluated the allegation on its merits, finding that it was not necessary for the disclosure to indicate that calcipotriol (Component A) had to be dissolved in the solvent (Component C). The skilled person would be able to make that determination based on his or her own knowledge.

Sound prediction

Teva argued that the Federal Court had erred in law by reformulating the test for sound prediction to permit reliance on a line of reasoning advanced by Leo’s expert, rather than by Leo or the inventors, to establish that the utility of Claim 17 was soundly predicted.

The FCA viewed this as a factual question to be reviewed for palpable and overriding error, reformulating the issues on appeal as: (1) whether Locke J. had evidence of the factual basis on which the inventors relied to claim the compositions covered by Claim 17; and (2) whether the Federal Court had evidence relating to the actual logic or reasoning of the inventors. In particular, the Court rejected Teva’s contention that one cannot make a sound prediction without knowing exactly why a particular combination works: the application of the doctrine turns on the nature of the invention, and in this case the invention was quite simple and its ingredients well-known. It was open to the Court to infer that the inventors’ logic and line of reasoning were, as a matter of fact, conveyed by the patent, and it was proper to do so based on expert evidence. No express statement in the patent as to the line of reasoning was required.


Teva argued that Locke J. had erred in law by failing to allow Teva to advance its allegations of insufficiency and by failing to find insufficiency. The Court only analyzed Teva’s substantive argument of insufficiency, noting that in any event it agreed with the Court below that the issue was not properly raised in the Notice of Allegation.

Teva alleged that the disclosure was insufficient on the basis that the order in which the components of the invention should be combined was an essential element of the invention that should have been disclosed. The FCA again viewed Teva’s argument as factual and therefore subject to a palpable and overriding error standard of review. The FCA concluded that the excerpt of the cross-examination of Leo’s expert on which Teva relied in support of its argument was far from conclusive, and that Teva’s own expert evidence suggested that no information regarding process was required. The relevant processes were well-known.

The FCA further rejected Teva’s argument that the Federal Court had erred in law by suggesting that trial and error may be permitted to determine the order in which the components should be combined. The Federal Court had only raised trial and error experimentation as a possibility, rather than a necessity, and it had not concluded that a minor research project was required, unlike in Teva Canada Inc v Pfizer Canada Inc, 2012 SCC 60. Further, the FCA confirmed that some non-inventive trial and error may be required to put a properly disclosed invention into practice, and the Supreme Court in Teva had not changed the law on this point.