top of page

Is the Promise Doctrine No More?

Michelle Vinitsky Windsor Law J.D. Candidate, 2020

The promise doctrine has been criticized for its rigidity in requiring each and every promise made in a patent specification to be fulfilled at the relevant date. In AstraZeneca, the Supreme Court of Canada put the application of this doctrine to rest when it comes to assessing the “utility” requirement of patentability, while leaving the door open for some reformulation of the promise doctrine in other spheres of patent law. Recent lower court judgments have shown little appetite so far towards any reformulation of the promise doctrine.

What is Utility?

The Supreme Court of Canada’s 2017 ruling of AstraZeneca Canada Inc v Apotex Inc(“AstraZeneca”)[1]is a landmark judgement in patent law on the question of patent utility. According to section 2 of the Patent Act, an invention must be useful.[2]Usefulness has been described as carrying out some known objective.[3]It is not mere laboratory curiosity whose only possible claim to utility is as a starting material for further research.[4]Ultimately, the alleged invention must be “susceptible of fulfilling its purpose” in order for utility to be established.[5]

The Promise Doctrine

 In Eli Lilly Canada Inc v Novopharm Ltd,[6]the Federal Court of Appeal set out that where the specification of the patent application does not promise a specific result, no particular level of utility is required and “a mere scintilla of utility” will suffice. However, where the specification sets out a specific and explicit “promise,” utility will be measured against that promise.[7]In other words, the promise doctrine sets out that if an inventor promises a particular utility in the patent description, the patent will be held to that utility, and if that utility cannot be shown at the filing date, the patent will be rendered invalid for want of utility.

The Supreme Court of Canada’s Response to the Promise Doctrine

In AstraZeneca, the Supreme Court, in a unanimous judgement, rejected the promise doctrine by stating “the promise doctrine is not the correct method of determining whether the utility required under s. 2 of the PatentActis met”[8]and “…it is not good law.”[9]

The Court took the position that the promise doctrine was incongruent with the words and the scheme of the Patent Act.[10]More specifically, Justice Rowe set out that the doctrine was “excessively onerous” in that it determines the standard of utility by reference to the promises expressed in the patent description, and where there are multiple promises set out in the patent description, each one is required to be fulfilled for the patent to be valid.[11]The Supreme Court explained that there is a difference between the requirement for an invention to be useful (as per s. 2 of the Patent Act) and the requirement to disclose an invention’s “operation or use” (as per s. 27(3) of the Patent Act).[12]S. 2 is described as a condition precedent to an invention while s. 27(3) is described as a disclosure requirement that is different from the s. 2 requirement.[13]The promise doctrine is said to import s. 27(3) into s. 2 by requiring that any disclosed use be demonstrated or soundly predicted as of the filing date.[14]This function of the promise doctrine, Justice Rowe stated, is inappropriate.[15]Justice Rowe also explained that the promise doctrine creates the risk that a useful invention will be deprived of protection simply because not everypromised use was sufficiently demonstrated or soundly predicted by the filing date.[16]The Court described this reality as “punitive” and having “no basis in the Act.”[17]

Moreover, the Court set out alternatives to the promise doctrine in order to still address the issue of patentees making promises that go beyond those made in the patent specification. These can include the possibility of rendering a patent invalid for failing to fulfill s. 27(3) requirements, being overbroad, or overpromising to the extent that the omission or addition is “wilfully made for the purpose of misleading” as per s. 53 of the Patent Act.[18]

Furthermore, Justice Rowe rejected the Federal Court’s finding that the patent in question failed for lack of utility because it promised more than it could provide.[19]Justice Rowe stated that “promises are not the yardstick against which utility is to be measured.”[20]The Court found that the ‘653 patent was soundly predicted to be useful and contained sufficient utility in order to satisfy s. 2 of the Patent Act.[21]

As a result, it would appear as though courts will be less likely to deny a patent application due to unfulfilled promises at the relevant date. However, the Court may have left open the possibility that such patents, under certain circumstances, could still be invalidated for reasons of insufficiency, overbreadth or misleading statements.

Decisions Following AstraZeneca

Following AstraZeneca, parties have continued in their attempt to employ the promise doctrine in some form or another. In response to these attempts, lower courts have rejected these arguments, consistent with the Supreme Court’s findings.

In the 2017 judgment of Bristol-Myers Squibb Canada Co v Apotex(“BMS”), the Federal Court of Appeal rejected the Federal Court’s decision to invalidate BMS’ patent because it failed to establish the various promises of their patent by the relevant date.[22]It was outlined that courts must no longer determine whether a patent fulfils the promises it makes when determining whether the patent application has sufficient utility.[23]Rather, one must first determine the subject-matter of the claim; second, determine whether this subject-matter is useful—meaning is it capable of practical purpose (i.e. an actual result) and; third, determine whether that usefulness is established by demonstration or sound prediction as of the filing date.[24]In the case of BMS, the discovery of a substance that acts as an inhibitor of an enzyme involved in many diseases was found to be an important advancement that meets the minimal utility requirements following the AstraZenecadecision.[25]

In Apotex Inc v Abbott Laboratories Ltd,[26]Apotex sought to amend their pleadings which set out to invalidate Abbott’s patent. Apotex’s position was that the Supreme Court in AstraZenecamay have rejected the use of the promise doctrine under a utility analysis but argued that it was still viable under grounds of overbreadth and insufficiency (i.e. failure to comply with s. 27(3) of Patent Act) and “the making of wilfully misleading statements” in the patent specification (i.e. failure to comply with s. 53 of the Patent Act).[27]The Ontario Superior Court of Justice reasoned that it would be counterintuitive for the Supreme Court to intend that promise-based arguments simply be imported into claims of overbreadth or misrepresentation.[28]In Apotex Inc v Abbott Laboratories Ltd, it was stated that a party cannot change what was really a promise-based pleading into a pleading that addresses another head of invalidity even if the Supreme Court has recognized that overpromising could be addressed under ss. 27(3) or 53 of the Patent Act.[29]While the Ontario Superior Court of Justice accepted that a party can legitimately plead misrepresentation under s. 53, even if that misrepresentation happens to look like a promise, this did not describe the current case as the court found that Apotex’s fraud allegations were deficient.[30]In Apotex Inc v Abbott Laboratories Ltd, the court described the contested amendments as merely a “repackaging” of the promise doctrine allegations and Apotex’s request for leave to plead the contested amendments was dismissed.[31]

Assessment on the Effect of AstraZeneca

In my opinion, I agree with the Supreme Court’s decision in AstraZenecato reject the use of the promise doctrine when determining whether a patent is sufficiently useful. As mentioned in paragraph 37 of the decision, the promise doctrine appears to be excessively onerous in requiring each and every promise made in the patent specification to be fulfilled as of the filing date. Moreover, upholding the promise doctrine could create a chilling effect on patent applications. Inventors may be discouraged from disclosing their findings for fear that their patent will later be invalidated because one of the promises made was not sufficiently demonstrated or soundly predicted by the filing date. Furthermore, the rejection of the promise doctrine prevents generic drug manufacturers from capitalizing on the inventive efforts of others by pointing out unfulfilled promises and grounding a finding of patent invalidity. Moreover, while patentees who overpromise might face consequences under ss. 27(3) and 53 of the Patent Act, there must be legitimate grounds for such invalidation. Ultimately, it appears as though the promise doctrine is no more, and in instances subsequent to AstraZenecawhere parties have sought a “repackaging” of the doctrine through alternative grounds, they have failed.


[1]AstraZeneca Canada Inc v Apotex Inc, 2017 SCC 36 [AstraZeneca].

[2]Patent Act, RSC 1985, c P-4, s 2.

[3]Re Application of Abitibi Co(1982), 62 CPR (2d) 81, para 31, (Patent Appeal Board and Commissioner of Patents).


[5]Wandscheer et al v Sicard Limited, [1948] SCR 1, para 10, 8 CPR 35.

[6]Eli Lilly Canada Inc v Novopharm Ltd, 2010 FCA 197, at para 76, [2012] 1 FCR 349.


[8]AstraZenecasupranote 1 at para 24.

[9]Ibidat para 51.

[10]Ibidat para 36.

[11]Ibidat para 37.

[12]Ibidat para 43.

[13]Consolboard Inc v MacMillan Bloedel (Saskatchewan) Ltd, [1981] 1 SCR 504 at 527, 122 DLR (3d) 203.

[14]AstraZenecasupranote 1 at para 44.


[16]Ibidat para 50.

[17]Ibidat para 51.

[18]Ibidat para 46.

[19]Ibidat para 63.



[22]Bristol-Myers Squibb Canada Co v Apotex Inc, 2017 FCA 190, at para 34, 152 CPR (4th) 265 [BMS].

[23]Ibidat para 35.

[24]AstraZenecasupranote 1 at paras 54-55.

[25]Ibidat para 40.

[26]Apotex Inc v Abbott Laboratories Limited, 2018 ONSC 5199, 297 ACWS (3d) 161.

[27]Ibid at para 8.

[28]Ibidat para 27.

[29]Ibidat para 28.

[30]Ibidat paras 32, 41, 47.

[31]Ibidat paras 42, 48.

Wissam Aoun

Faculty Member

Wissam Aoun is is an Assistant Professor of Law at the University of Windsor and Director of the International Intellectual Property Law Clinic, a collaboration between Windsor Law and University of Detroit Mercy Law. His area of expertise is intellectual property law including patents, trademarks, copyright and commercialization of technology.

519-253-3000 ext. 4207


bottom of page